diff options
| author | Maxime Ripard <maxime.ripard@bootlin.com> | 2019-01-11 16:32:10 +0100 |
|---|---|---|
| committer | Maxime Ripard <maxime.ripard@bootlin.com> | 2019-01-11 16:32:10 +0100 |
| commit | 23d19ba06b9c5614d6457f5fed349ec8f6d4dac9 (patch) | |
| tree | 39f0b657e5b1b5b958780cae4ae6360f69548d50 /drivers/gpu | |
| parent | drm/atomic: Add missing () to function ref in kerneldoc (diff) | |
| parent | Merge tag 'tilcdc-4.22' of https://github.com/jsarha/linux into drm-next (diff) | |
| download | linux-23d19ba06b9c5614d6457f5fed349ec8f6d4dac9.tar.xz linux-23d19ba06b9c5614d6457f5fed349ec8f6d4dac9.zip | |
Merge drm/drm-next into drm-misc-next
drm-next has been forwarded to 5.0-rc1, and we need it to apply the damage
helper for dirtyfb series from Noralf Trønnes.
Signed-off-by: Maxime Ripard <maxime.ripard@bootlin.com>
Diffstat (limited to 'drivers/gpu')
614 files changed, 22290 insertions, 12494 deletions
diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile index 1fafc2f8e8f9..ce8d1d384319 100644 --- a/drivers/gpu/drm/Makefile +++ b/drivers/gpu/drm/Makefile @@ -32,12 +32,12 @@ drm-$(CONFIG_AGP) += drm_agpsupport.o drm-$(CONFIG_DEBUG_FS) += drm_debugfs.o drm_debugfs_crc.o drm-$(CONFIG_DRM_LOAD_EDID_FIRMWARE) += drm_edid_load.o -drm_kms_helper-y := drm_crtc_helper.o drm_dp_helper.o drm_probe_helper.o \ +drm_kms_helper-y := drm_crtc_helper.o drm_dp_helper.o drm_dsc.o drm_probe_helper.o \ drm_plane_helper.o drm_dp_mst_topology.o drm_atomic_helper.o \ drm_kms_helper_common.o drm_dp_dual_mode_helper.o \ drm_simple_kms_helper.o drm_modeset_helper.o \ drm_scdc_helper.o drm_gem_framebuffer_helper.o \ - drm_atomic_state_helper.o + drm_atomic_state_helper.o drm_damage_helper.o drm_kms_helper-$(CONFIG_DRM_PANEL_BRIDGE) += bridge/panel.o drm_kms_helper-$(CONFIG_DRM_FBDEV_EMULATION) += drm_fb_helper.o diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 42f882c633ee..bcef6ea4bcf9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -81,6 +81,8 @@ #include "amdgpu_job.h" #include "amdgpu_bo_list.h" #include "amdgpu_gem.h" +#include "amdgpu_doorbell.h" +#include "amdgpu_amdkfd.h" #define MAX_GPU_INSTANCE 16 @@ -162,6 +164,7 @@ extern int amdgpu_si_support; extern int amdgpu_cik_support; #endif +#define AMDGPU_VM_MAX_NUM_CTX 4096 #define AMDGPU_SG_THRESHOLD (256*1024*1024) #define AMDGPU_DEFAULT_GTT_SIZE_MB 3072ULL /* 3GB by default */ #define AMDGPU_WAIT_IDLE_TIMEOUT_IN_MS 3000 @@ -234,7 +237,7 @@ enum amdgpu_kiq_irq { #define MAX_KIQ_REG_WAIT 5000 /* in usecs, 5ms */ #define MAX_KIQ_REG_BAILOUT_INTERVAL 5 /* in msecs, 5ms */ -#define MAX_KIQ_REG_TRY 20 +#define MAX_KIQ_REG_TRY 80 /* 20 -> 80 */ int amdgpu_device_ip_set_clockgating_state(void *dev, enum amd_ip_block_type block_type, @@ -361,123 +364,6 @@ int amdgpu_fence_slab_init(void); void amdgpu_fence_slab_fini(void); /* - * GPU doorbell structures, functions & helpers - */ -typedef enum _AMDGPU_DOORBELL_ASSIGNMENT -{ - AMDGPU_DOORBELL_KIQ = 0x000, - AMDGPU_DOORBELL_HIQ = 0x001, - AMDGPU_DOORBELL_DIQ = 0x002, - AMDGPU_DOORBELL_MEC_RING0 = 0x010, - AMDGPU_DOORBELL_MEC_RING1 = 0x011, - AMDGPU_DOORBELL_MEC_RING2 = 0x012, - AMDGPU_DOORBELL_MEC_RING3 = 0x013, - AMDGPU_DOORBELL_MEC_RING4 = 0x014, - AMDGPU_DOORBELL_MEC_RING5 = 0x015, - AMDGPU_DOORBELL_MEC_RING6 = 0x016, - AMDGPU_DOORBELL_MEC_RING7 = 0x017, - AMDGPU_DOORBELL_GFX_RING0 = 0x020, - AMDGPU_DOORBELL_sDMA_ENGINE0 = 0x1E0, - AMDGPU_DOORBELL_sDMA_ENGINE1 = 0x1E1, - AMDGPU_DOORBELL_IH = 0x1E8, - AMDGPU_DOORBELL_MAX_ASSIGNMENT = 0x3FF, - AMDGPU_DOORBELL_INVALID = 0xFFFF -} AMDGPU_DOORBELL_ASSIGNMENT; - -struct amdgpu_doorbell { - /* doorbell mmio */ - resource_size_t base; - resource_size_t size; - u32 __iomem *ptr; - u32 num_doorbells; /* Number of doorbells actually reserved for amdgpu. */ -}; - -/* - * 64bit doorbell, offset are in QWORD, occupy 2KB doorbell space - */ -typedef enum _AMDGPU_DOORBELL64_ASSIGNMENT -{ - /* - * All compute related doorbells: kiq, hiq, diq, traditional compute queue, user queue, should locate in - * a continues range so that programming CP_MEC_DOORBELL_RANGE_LOWER/UPPER can cover this range. - * Compute related doorbells are allocated from 0x00 to 0x8a - */ - - - /* kernel scheduling */ - AMDGPU_DOORBELL64_KIQ = 0x00, - - /* HSA interface queue and debug queue */ - AMDGPU_DOORBELL64_HIQ = 0x01, - AMDGPU_DOORBELL64_DIQ = 0x02, - - /* Compute engines */ - AMDGPU_DOORBELL64_MEC_RING0 = 0x03, - AMDGPU_DOORBELL64_MEC_RING1 = 0x04, - AMDGPU_DOORBELL64_MEC_RING2 = 0x05, - AMDGPU_DOORBELL64_MEC_RING3 = 0x06, - AMDGPU_DOORBELL64_MEC_RING4 = 0x07, - AMDGPU_DOORBELL64_MEC_RING5 = 0x08, - AMDGPU_DOORBELL64_MEC_RING6 = 0x09, - AMDGPU_DOORBELL64_MEC_RING7 = 0x0a, - - /* User queue doorbell range (128 doorbells) */ - AMDGPU_DOORBELL64_USERQUEUE_START = 0x0b, - AMDGPU_DOORBELL64_USERQUEUE_END = 0x8a, - - /* Graphics engine */ - AMDGPU_DOORBELL64_GFX_RING0 = 0x8b, - - /* - * Other graphics doorbells can be allocated here: from 0x8c to 0xdf - * Graphics voltage island aperture 1 - * default non-graphics QWORD index is 0xe0 - 0xFF inclusive - */ - - /* sDMA engines reserved from 0xe0 -0xef */ - AMDGPU_DOORBELL64_sDMA_ENGINE0 = 0xE0, - AMDGPU_DOORBELL64_sDMA_HI_PRI_ENGINE0 = 0xE1, - AMDGPU_DOORBELL64_sDMA_ENGINE1 = 0xE8, - AMDGPU_DOORBELL64_sDMA_HI_PRI_ENGINE1 = 0xE9, - - /* For vega10 sriov, the sdma doorbell must be fixed as follow - * to keep the same setting with host driver, or it will - * happen conflicts - */ - AMDGPU_VEGA10_DOORBELL64_sDMA_ENGINE0 = 0xF0, - AMDGPU_VEGA10_DOORBELL64_sDMA_HI_PRI_ENGINE0 = 0xF1, - AMDGPU_VEGA10_DOORBELL64_sDMA_ENGINE1 = 0xF2, - AMDGPU_VEGA10_DOORBELL64_sDMA_HI_PRI_ENGINE1 = 0xF3, - - /* Interrupt handler */ - AMDGPU_DOORBELL64_IH = 0xF4, /* For legacy interrupt ring buffer */ - AMDGPU_DOORBELL64_IH_RING1 = 0xF5, /* For page migration request log */ - AMDGPU_DOORBELL64_IH_RING2 = 0xF6, /* For page migration translation/invalidation log */ - - /* VCN engine use 32 bits doorbell */ - AMDGPU_DOORBELL64_VCN0_1 = 0xF8, /* lower 32 bits for VNC0 and upper 32 bits for VNC1 */ - AMDGPU_DOORBELL64_VCN2_3 = 0xF9, - AMDGPU_DOORBELL64_VCN4_5 = 0xFA, - AMDGPU_DOORBELL64_VCN6_7 = 0xFB, - - /* overlap the doorbell assignment with VCN as they are mutually exclusive - * VCE engine's doorbell is 32 bit and two VCE ring share one QWORD - */ - AMDGPU_DOORBELL64_UVD_RING0_1 = 0xF8, - AMDGPU_DOORBELL64_UVD_RING2_3 = 0xF9, - AMDGPU_DOORBELL64_UVD_RING4_5 = 0xFA, - AMDGPU_DOORBELL64_UVD_RING6_7 = 0xFB, - - AMDGPU_DOORBELL64_VCE_RING0_1 = 0xFC, - AMDGPU_DOORBELL64_VCE_RING2_3 = 0xFD, - AMDGPU_DOORBELL64_VCE_RING4_5 = 0xFE, - AMDGPU_DOORBELL64_VCE_RING6_7 = 0xFF, - - AMDGPU_DOORBELL64_MAX_ASSIGNMENT = 0xFF, - AMDGPU_DOORBELL64_INVALID = 0xFFFF -} AMDGPU_DOORBELL64_ASSIGNMENT; - -/* * IRQS. */ @@ -654,6 +540,8 @@ struct amdgpu_asic_funcs { struct amdgpu_ring *ring); /* check if the asic needs a full reset of if soft reset will work */ bool (*need_full_reset)(struct amdgpu_device *adev); + /* initialize doorbell layout for specific asic*/ + void (*init_doorbell_index)(struct amdgpu_device *adev); }; /* @@ -976,6 +864,9 @@ struct amdgpu_device { /* GDS */ struct amdgpu_gds gds; + /* KFD */ + struct amdgpu_kfd_dev kfd; + /* display related functionality */ struct amdgpu_display_manager dm; @@ -989,9 +880,6 @@ struct amdgpu_device { atomic64_t visible_pin_size; atomic64_t gart_pin_size; - /* amdkfd interface */ - struct kfd_dev *kfd; - /* soc15 register offset based on ip, instance and segment */ uint32_t *reg_offset[MAX_HWIP][HWIP_MAX_INSTANCE]; @@ -1023,6 +911,10 @@ struct amdgpu_device { unsigned long last_mm_index; bool in_gpu_reset; struct mutex lock_reset; + struct amdgpu_doorbell_index doorbell_index; + + int asic_reset_res; + struct work_struct xgmi_reset_work; }; static inline struct amdgpu_device *amdgpu_ttm_adev(struct ttm_bo_device *bdev) @@ -1047,11 +939,6 @@ uint8_t amdgpu_mm_rreg8(struct amdgpu_device *adev, uint32_t offset); u32 amdgpu_io_rreg(struct amdgpu_device *adev, u32 reg); void amdgpu_io_wreg(struct amdgpu_device *adev, u32 reg, u32 v); -u32 amdgpu_mm_rdoorbell(struct amdgpu_device *adev, u32 index); -void amdgpu_mm_wdoorbell(struct amdgpu_device *adev, u32 index, u32 v); -u64 amdgpu_mm_rdoorbell64(struct amdgpu_device *adev, u32 index); -void amdgpu_mm_wdoorbell64(struct amdgpu_device *adev, u32 index, u64 v); - bool amdgpu_device_asic_has_dc_support(enum amd_asic_type asic_type); bool amdgpu_device_has_dc_support(struct amdgpu_device *adev); @@ -1113,11 +1000,6 @@ int emu_soc_asic_init(struct amdgpu_device *adev); #define RREG32_IO(reg) amdgpu_io_rreg(adev, (reg)) #define WREG32_IO(reg, v) amdgpu_io_wreg(adev, (reg), (v)) -#define RDOORBELL32(index) amdgpu_mm_rdoorbell(adev, (index)) -#define WDOORBELL32(index, v) amdgpu_mm_wdoorbell(adev, (index), (v)) -#define RDOORBELL64(index) amdgpu_mm_rdoorbell64(adev, (index)) -#define WDOORBELL64(index, v) amdgpu_mm_wdoorbell64(adev, (index), (v)) - #define REG_FIELD_SHIFT(reg, field) reg##__##field##__SHIFT #define REG_FIELD_MASK(reg, field) reg##__##field##_MASK @@ -1159,6 +1041,7 @@ int emu_soc_asic_init(struct amdgpu_device *adev); #define amdgpu_asic_flush_hdp(adev, r) (adev)->asic_funcs->flush_hdp((adev), (r)) #define amdgpu_asic_invalidate_hdp(adev, r) (adev)->asic_funcs->invalidate_hdp((adev), (r)) #define amdgpu_asic_need_full_reset(adev) (adev)->asic_funcs->need_full_reset((adev)) +#define amdgpu_asic_init_doorbell_index(adev) (adev)->asic_funcs->init_doorbell_index((adev)) /* Common functions */ bool amdgpu_device_should_recover_gpu(struct amdgpu_device *adev); @@ -1219,12 +1102,6 @@ void amdgpu_disable_vblank_kms(struct drm_device *dev, unsigned int pipe); long amdgpu_kms_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg); - -/* - * functions used by amdgpu_xgmi.c - */ -int amdgpu_xgmi_add_device(struct amdgpu_device *adev); - /* * functions used by amdgpu_encoder.c */ @@ -1252,6 +1129,9 @@ bool amdgpu_acpi_is_pcie_performance_request_supported(struct amdgpu_device *ade int amdgpu_acpi_pcie_performance_request(struct amdgpu_device *adev, u8 perf_req, bool advertise); int amdgpu_acpi_pcie_notify_device_ready(struct amdgpu_device *adev); + +void amdgpu_acpi_get_backlight_caps(struct amdgpu_device *adev, + struct amdgpu_dm_backlight_caps *caps); #else static inline int amdgpu_acpi_init(struct amdgpu_device *adev) { return 0; } static inline void amdgpu_acpi_fini(struct amdgpu_device *adev) { } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c index 7f0afc526419..4376b17ca594 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c @@ -41,28 +41,21 @@ struct amdgpu_atif_notification_cfg { }; struct amdgpu_atif_notifications { - bool display_switch; - bool expansion_mode_change; bool thermal_state; bool forced_power_state; bool system_power_state; - bool display_conf_change; - bool px_gfx_switch; bool brightness_change; bool dgpu_display_event; + bool gpu_package_power_limit; }; struct amdgpu_atif_functions { bool system_params; bool sbios_requests; - bool select_active_disp; - bool lid_state; - bool get_tv_standard; - bool set_tv_standard; - bool get_panel_expansion_mode; - bool set_panel_expansion_mode; bool temperature_change; - bool graphics_device_types; + bool query_backlight_transfer_characteristics; + bool ready_to_undock; + bool external_gpu_information; }; struct amdgpu_atif { @@ -72,6 +65,7 @@ struct amdgpu_atif { struct amdgpu_atif_functions functions; struct amdgpu_atif_notification_cfg notification_cfg; struct amdgpu_encoder *encoder_for_bl; + struct amdgpu_dm_backlight_caps backlight_caps; }; /* Call the ATIF method @@ -137,15 +131,12 @@ static union acpi_object *amdgpu_atif_call(struct amdgpu_atif *atif, */ static void amdgpu_atif_parse_notification(struct amdgpu_atif_notifications *n, u32 mask) { - n->display_switch = mask & ATIF_DISPLAY_SWITCH_REQUEST_SUPPORTED; - n->expansion_mode_change = mask & ATIF_EXPANSION_MODE_CHANGE_REQUEST_SUPPORTED; n->thermal_state = mask & ATIF_THERMAL_STATE_CHANGE_REQUEST_SUPPORTED; n->forced_power_state = mask & ATIF_FORCED_POWER_STATE_CHANGE_REQUEST_SUPPORTED; n->system_power_state = mask & ATIF_SYSTEM_POWER_SOURCE_CHANGE_REQUEST_SUPPORTED; - n->display_conf_change = mask & ATIF_DISPLAY_CONF_CHANGE_REQUEST_SUPPORTED; - n->px_gfx_switch = mask & ATIF_PX_GFX_SWITCH_REQUEST_SUPPORTED; n->brightness_change = mask & ATIF_PANEL_BRIGHTNESS_CHANGE_REQUEST_SUPPORTED; n->dgpu_display_event = mask & ATIF_DGPU_DISPLAY_EVENT_SUPPORTED; + n->gpu_package_power_limit = mask & ATIF_GPU_PACKAGE_POWER_LIMIT_REQUEST_SUPPORTED; } /** @@ -162,14 +153,11 @@ static void amdgpu_atif_parse_functions(struct amdgpu_atif_functions *f, u32 mas { f->system_params = mask & ATIF_GET_SYSTEM_PARAMETERS_SUPPORTED; f->sbios_requests = mask & ATIF_GET_SYSTEM_BIOS_REQUESTS_SUPPORTED; - f->select_active_disp = mask & ATIF_SELECT_ACTIVE_DISPLAYS_SUPPORTED; - f->lid_state = mask & ATIF_GET_LID_STATE_SUPPORTED; - f->get_tv_standard = mask & ATIF_GET_TV_STANDARD_FROM_CMOS_SUPPORTED; - f->set_tv_standard = mask & ATIF_SET_TV_STANDARD_IN_CMOS_SUPPORTED; - f->get_panel_expansion_mode = mask & ATIF_GET_PANEL_EXPANSION_MODE_FROM_CMOS_SUPPORTED; - f->set_panel_expansion_mode = mask & ATIF_SET_PANEL_EXPANSION_MODE_IN_CMOS_SUPPORTED; f->temperature_change = mask & ATIF_TEMPERATURE_CHANGE_NOTIFICATION_SUPPORTED; - f->graphics_device_types = mask & ATIF_GET_GRAPHICS_DEVICE_TYPES_SUPPORTED; + f->query_backlight_transfer_characteristics = + mask & ATIF_QUERY_BACKLIGHT_TRANSFER_CHARACTERISTICS_SUPPORTED; + f->ready_to_undock = mask & ATIF_READY_TO_UNDOCK_NOTIFICATION_SUPPORTED; + f->external_gpu_information = mask & ATIF_GET_EXTERNAL_GPU_INFORMATION_SUPPORTED; } /** @@ -311,6 +299,65 @@ out: } /** + * amdgpu_atif_query_backlight_caps - get min and max backlight input signal + * + * @handle: acpi handle + * + * Execute the QUERY_BRIGHTNESS_TRANSFER_CHARACTERISTICS ATIF function + * to determine the acceptable range of backlight values + * + * Backlight_caps.caps_valid will be set to true if the query is successful + * + * The input signals are in range 0-255 + * + * This function assumes the display with backlight is the first LCD + * + * Returns 0 on success, error on failure. + */ +static int amdgpu_atif_query_backlight_caps(struct amdgpu_atif *atif) +{ + union acpi_object *info; + struct atif_qbtc_output characteristics; + struct atif_qbtc_arguments arguments; + struct acpi_buffer params; + size_t size; + int err = 0; + + arguments.size = sizeof(arguments); + arguments.requested_display = ATIF_QBTC_REQUEST_LCD1; + + params.length = sizeof(arguments); + params.pointer = (void *)&arguments; + + info = amdgpu_atif_call(atif, + ATIF_FUNCTION_QUERY_BRIGHTNESS_TRANSFER_CHARACTERISTICS, + ¶ms); + if (!info) { + err = -EIO; + goto out; + } + + size = *(u16 *) info->buffer.pointer; + if (size < 10) { + err = -EINVAL; + goto out; + } + + memset(&characteristics, 0, sizeof(characteristics)); + size = min(sizeof(characteristics), size); + memcpy(&characteristics, info->buffer.pointer, size); + + atif->backlight_caps.caps_valid = true; + atif->backlight_caps.min_input_signal = + characteristics.min_input_signal; + atif->backlight_caps.max_input_signal = + characteristics.max_input_signal; +out: + kfree(info); + return err; +} + +/** * amdgpu_atif_get_sbios_requests - get requested sbios event * * @handle: acpi handle @@ -799,6 +846,17 @@ int amdgpu_acpi_init(struct amdgpu_device *adev) } } + if (atif->functions.query_backlight_transfer_characteristics) { + ret = amdgpu_atif_query_backlight_caps(atif); + if (ret) { + DRM_DEBUG_DRIVER("Call to QUERY_BACKLIGHT_TRANSFER_CHARACTERISTICS failed: %d\n", + ret); + atif->backlight_caps.caps_valid = false; + } + } else { + atif->backlight_caps.caps_valid = false; + } + out: adev->acpi_nb.notifier_call = amdgpu_acpi_event; register_acpi_notifier(&adev->acpi_nb); @@ -806,6 +864,18 @@ out: return ret; } +void amdgpu_acpi_get_backlight_caps(struct amdgpu_device *adev, + struct amdgpu_dm_backlight_caps *caps) +{ + if (!adev->atif) { + caps->caps_valid = false; + return; + } + caps->caps_valid = adev->atif->backlight_caps.caps_valid; + caps->min_input_signal = adev->atif->backlight_caps.min_input_signal; + caps->max_input_signal = adev->atif->backlight_caps.max_input_signal; +} + /** * amdgpu_acpi_fini - tear down driver acpi support * @@ -816,6 +886,5 @@ out: void amdgpu_acpi_fini(struct amdgpu_device *adev) { unregister_acpi_notifier(&adev->acpi_nb); - if (adev->atif) - kfree(adev->atif); + kfree(adev->atif); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c index bcf1666fb31d..2dfaf158ef07 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c @@ -26,15 +26,26 @@ #include "amdgpu.h" #include "amdgpu_gfx.h" #include <linux/module.h> +#include <linux/dma-buf.h> const struct kgd2kfd_calls *kgd2kfd; static const unsigned int compute_vmid_bitmap = 0xFF00; +/* Total memory size in system memory and all GPU VRAM. Used to + * estimate worst case amount of memory to reserve for page tables + */ +uint64_t amdgpu_amdkfd_total_mem_size; + int amdgpu_amdkfd_init(void) { + struct sysinfo si; int ret; + si_meminfo(&si); + amdgpu_amdkfd_total_mem_size = si.totalram - si.totalhigh; + amdgpu_amdkfd_total_mem_size *= si.mem_unit; + #ifdef CONFIG_HSA_AMD ret = kgd2kfd_init(KFD_INTERFACE_VERSION, &kgd2kfd); if (ret) @@ -73,9 +84,11 @@ void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev) case CHIP_FIJI: case CHIP_POLARIS10: case CHIP_POLARIS11: + case CHIP_POLARIS12: kfd2kgd = amdgpu_amdkfd_gfx_8_0_get_functions(); break; case CHIP_VEGA10: + case CHIP_VEGA12: case CHIP_VEGA20: case CHIP_RAVEN: kfd2kgd = amdgpu_amdkfd_gfx_9_0_get_functions(); @@ -85,8 +98,11 @@ void amdgpu_amdkfd_device_probe(struct amdgpu_device *adev) return; } - adev->kfd = kgd2kfd->probe((struct kgd_dev *)adev, - adev->pdev, kfd2kgd); + adev->kfd.dev = kgd2kfd->probe((struct kgd_dev *)adev, + adev->pdev, kfd2kgd); + + if (adev->kfd.dev) + amdgpu_amdkfd_total_mem_size += adev->gmc.real_vram_size; } /** @@ -126,7 +142,8 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev) { int i, n; int last_valid_bit; - if (adev->kfd) { + + if (adev->kfd.dev) { struct kgd2kfd_shared_resources gpu_resources = { .compute_vmid_bitmap = compute_vmid_bitmap, .num_pipe_per_mec = adev->gfx.mec.num_pipe_per_mec, @@ -165,7 +182,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev) &gpu_resources.doorbell_start_offset); if (adev->asic_type < CHIP_VEGA10) { - kgd2kfd->device_init(adev->kfd, &gpu_resources); + kgd2kfd->device_init(adev->kfd.dev, &gpu_resources); return; } @@ -179,25 +196,14 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev) * process in case of 64-bit doorbells so we * can use each doorbell assignment twice. */ - if (adev->asic_type == CHIP_VEGA10) { - gpu_resources.sdma_doorbell[0][i] = - AMDGPU_VEGA10_DOORBELL64_sDMA_ENGINE0 + (i >> 1); - gpu_resources.sdma_doorbell[0][i+1] = - AMDGPU_VEGA10_DOORBELL64_sDMA_ENGINE0 + 0x200 + (i >> 1); - gpu_resources.sdma_doorbell[1][i] = - AMDGPU_VEGA10_DOORBELL64_sDMA_ENGINE1 + (i >> 1); - gpu_resources.sdma_doorbell[1][i+1] = - AMDGPU_VEGA10_DOORBELL64_sDMA_ENGINE1 + 0x200 + (i >> 1); - } else { - gpu_resources.sdma_doorbell[0][i] = - AMDGPU_DOORBELL64_sDMA_ENGINE0 + (i >> 1); - gpu_resources.sdma_doorbell[0][i+1] = - AMDGPU_DOORBELL64_sDMA_ENGINE0 + 0x200 + (i >> 1); - gpu_resources.sdma_doorbell[1][i] = - AMDGPU_DOORBELL64_sDMA_ENGINE1 + (i >> 1); - gpu_resources.sdma_doorbell[1][i+1] = - AMDGPU_DOORBELL64_sDMA_ENGINE1 + 0x200 + (i >> 1); - } + gpu_resources.sdma_doorbell[0][i] = + adev->doorbell_index.sdma_engine0 + (i >> 1); + gpu_resources.sdma_doorbell[0][i+1] = + adev->doorbell_index.sdma_engine0 + 0x200 + (i >> 1); + gpu_resources.sdma_doorbell[1][i] = + adev->doorbell_index.sdma_engine1 + (i >> 1); + gpu_resources.sdma_doorbell[1][i+1] = + adev->doorbell_index.sdma_engine1 + 0x200 + (i >> 1); } /* Doorbells 0x0e0-0ff and 0x2e0-2ff are reserved for * SDMA, IH and VCN. So don't use them for the CP. @@ -205,37 +211,37 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev) gpu_resources.reserved_doorbell_mask = 0x1e0; gpu_resources.reserved_doorbell_val = 0x0e0; - kgd2kfd->device_init(adev->kfd, &gpu_resources); + kgd2kfd->device_init(adev->kfd.dev, &gpu_resources); } } void amdgpu_amdkfd_device_fini(struct amdgpu_device *adev) { - if (adev->kfd) { - kgd2kfd->device_exit(adev->kfd); - adev->kfd = NULL; + if (adev->kfd.dev) { + kgd2kfd->device_exit(adev->kfd.dev); + adev->kfd.dev = NULL; } } void amdgpu_amdkfd_interrupt(struct amdgpu_device *adev, const void *ih_ring_entry) { - if (adev->kfd) - kgd2kfd->interrupt(adev->kfd, ih_ring_entry); + if (adev->kfd.dev) + kgd2kfd->interrupt(adev->kfd.dev, ih_ring_entry); } void amdgpu_amdkfd_suspend(struct amdgpu_device *adev) { - if (adev->kfd) - kgd2kfd->suspend(adev->kfd); + if (adev->kfd.dev) + kgd2kfd->suspend(adev->kfd.dev); } int amdgpu_amdkfd_resume(struct amdgpu_device *adev) { int r = 0; - if (adev->kfd) - r = kgd2kfd->resume(adev->kfd); + if (adev->kfd.dev) + r = kgd2kfd->resume(adev->kfd.dev); return r; } @@ -244,8 +250,8 @@ int amdgpu_amdkfd_pre_reset(struct amdgpu_device *adev) { int r = 0; - if (adev->kfd) - r = kgd2kfd->pre_reset(adev->kfd); + if (adev->kfd.dev) + r = kgd2kfd->pre_reset(adev->kfd.dev); return r; } @@ -254,8 +260,8 @@ int amdgpu_amdkfd_post_reset(struct amdgpu_device *adev) { int r = 0; - if (adev->kfd) - r = kgd2kfd->post_reset(adev->kfd); + if (adev->kfd.dev) + r = kgd2kfd->post_reset(adev->kfd.dev); return r; } @@ -428,6 +434,62 @@ void amdgpu_amdkfd_get_cu_info(struct kgd_dev *kgd, struct kfd_cu_info *cu_info) cu_info->lds_size = acu_info.lds_size; } +int amdgpu_amdkfd_get_dmabuf_info(struct kgd_dev *kgd, int dma_buf_fd, + struct kgd_dev **dma_buf_kgd, + uint64_t *bo_size, void *metadata_buffer, + size_t buffer_size, uint32_t *metadata_size, + uint32_t *flags) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)kgd; + struct dma_buf *dma_buf; + struct drm_gem_object *obj; + struct amdgpu_bo *bo; + uint64_t metadata_flags; + int r = -EINVAL; + + dma_buf = dma_buf_get(dma_buf_fd); + if (IS_ERR(dma_buf)) + return PTR_ERR(dma_buf); + + if (dma_buf->ops != &amdgpu_dmabuf_ops) + /* Can't handle non-graphics buffers */ + goto out_put; + + obj = dma_buf->priv; + if (obj->dev->driver != adev->ddev->driver) + /* Can't handle buffers from different drivers */ + goto out_put; + + adev = obj->dev->dev_private; + bo = gem_to_amdgpu_bo(obj); + if (!(bo->preferred_domains & (AMDGPU_GEM_DOMAIN_VRAM | + AMDGPU_GEM_DOMAIN_GTT))) + /* Only VRAM and GTT BOs are supported */ + goto out_put; + + r = 0; + if (dma_buf_kgd) + *dma_buf_kgd = (struct kgd_dev *)adev; + if (bo_size) + *bo_size = amdgpu_bo_size(bo); + if (metadata_size) + *metadata_size = bo->metadata_size; + if (metadata_buffer) + r = amdgpu_bo_get_metadata(bo, metadata_buffer, buffer_size, + metadata_size, &metadata_flags); + if (flags) { + *flags = (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM) ? + ALLOC_MEM_FLAGS_VRAM : ALLOC_MEM_FLAGS_GTT; + + if (bo->flags & AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED) + *flags |= ALLOC_MEM_FLAGS_PUBLIC; + } + +out_put: + dma_buf_put(dma_buf); + return r; +} + uint64_t amdgpu_amdkfd_get_vram_usage(struct kgd_dev *kgd) { struct amdgpu_device *adev = (struct amdgpu_device *)kgd; @@ -510,7 +572,7 @@ void amdgpu_amdkfd_set_compute_idle(struct kgd_dev *kgd, bool idle) bool amdgpu_amdkfd_is_kfd_vmid(struct amdgpu_device *adev, u32 vmid) { - if (adev->kfd) { + if (adev->kfd.dev) { if ((1 << vmid) & compute_vmid_bitmap) return true; } @@ -524,7 +586,7 @@ bool amdkfd_fence_check_mm(struct dma_fence *f, struct mm_struct *mm) return false; } -void amdgpu_amdkfd_unreserve_system_memory_limit(struct amdgpu_bo *bo) +void amdgpu_amdkfd_unreserve_memory_limit(struct amdgpu_bo *bo) { } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h index bcf587b4ba98..70429f7aa9a8 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h @@ -27,7 +27,6 @@ #include <linux/types.h> #include <linux/mm.h> -#include <linux/mmu_context.h> #include <linux/workqueue.h> #include <kgd_kfd_interface.h> #include <drm/ttm/ttm_execbuf_util.h> @@ -35,6 +34,7 @@ #include "amdgpu_vm.h" extern const struct kgd2kfd_calls *kgd2kfd; +extern uint64_t amdgpu_amdkfd_total_mem_size; struct amdgpu_device; @@ -77,6 +77,11 @@ struct amdgpu_amdkfd_fence { char timeline_name[TASK_COMM_LEN]; }; +struct amdgpu_kfd_dev { + struct kfd_dev *dev; + uint64_t vram_used; +}; + struct amdgpu_amdkfd_fence *amdgpu_amdkfd_fence_create(u64 context, struct mm_struct *mm); bool amdkfd_fence_check_mm(struct dma_fence *f, struct mm_struct *mm); @@ -144,6 +149,11 @@ uint64_t amdgpu_amdkfd_get_gpu_clock_counter(struct kgd_dev *kgd); uint32_t amdgpu_amdkfd_get_max_engine_clock_in_mhz(struct kgd_dev *kgd); void amdgpu_amdkfd_get_cu_info(struct kgd_dev *kgd, struct kfd_cu_info *cu_info); +int amdgpu_amdkfd_get_dmabuf_info(struct kgd_dev *kgd, int dma_buf_fd, + struct kgd_dev **dmabuf_kgd, + uint64_t *bo_size, void *metadata_buffer, + size_t buffer_size, uint32_t *metadata_size, + uint32_t *flags); uint64_t amdgpu_amdkfd_get_vram_usage(struct kgd_dev *kgd); uint64_t amdgpu_amdkfd_get_hive_id(struct kgd_dev *kgd); @@ -195,7 +205,13 @@ int amdgpu_amdkfd_gpuvm_restore_process_bos(void *process_info, int amdgpu_amdkfd_gpuvm_get_vm_fault_info(struct kgd_dev *kgd, struct kfd_vm_fault_info *info); +int amdgpu_amdkfd_gpuvm_import_dmabuf(struct kgd_dev *kgd, + struct dma_buf *dmabuf, + uint64_t va, void *vm, + struct kgd_mem **mem, uint64_t *size, + uint64_t *mmap_offset); + void amdgpu_amdkfd_gpuvm_init_mem_limits(void); -void amdgpu_amdkfd_unreserve_system_memory_limit(struct amdgpu_bo *bo); +void amdgpu_amdkfd_unreserve_memory_limit(struct amdgpu_bo *bo); #endif /* AMDGPU_AMDKFD_H_INCLUDED */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c index 72a357dae070..ff7fac7df34b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c @@ -23,6 +23,7 @@ #include <linux/fdtable.h> #include <linux/uaccess.h> #include <linux/firmware.h> +#include <linux/mmu_context.h> #include <drm/drmP.h> #include "amdgpu.h" #include "amdgpu_amdkfd.h" diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c index 0e2a56b6a9b6..56ea929f524b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c @@ -24,6 +24,7 @@ #include <linux/fdtable.h> #include <linux/uaccess.h> #include <linux/firmware.h> +#include <linux/mmu_context.h> #include <drm/drmP.h> #include "amdgpu.h" #include "amdgpu_amdkfd.h" diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c index 03b604c96d94..5c51d4910650 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c @@ -26,6 +26,7 @@ #include <linux/fdtable.h> #include <linux/uaccess.h> #include <linux/firmware.h> +#include <linux/mmu_context.h> #include <drm/drmP.h> #include "amdgpu.h" #include "amdgpu_amdkfd.h" diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index df0a059565f9..be1ab43473c6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -25,6 +25,7 @@ #include <linux/list.h> #include <linux/pagemap.h> #include <linux/sched/mm.h> +#include <linux/dma-buf.h> #include <drm/drmP.h> #include "amdgpu_object.h" #include "amdgpu_vm.h" @@ -46,9 +47,9 @@ /* Impose limit on how much memory KFD can use */ static struct { uint64_t max_system_mem_limit; - uint64_t max_userptr_mem_limit; + uint64_t max_ttm_mem_limit; int64_t system_mem_used; - int64_t userptr_mem_used; + int64_t ttm_mem_used; spinlock_t mem_limit_lock; } kfd_mem_limit; @@ -90,8 +91,8 @@ static bool check_if_add_bo_to_vm(struct amdgpu_vm *avm, } /* Set memory usage limits. Current, limits are - * System (kernel) memory - 3/8th System RAM - * Userptr memory - 3/4th System RAM + * System (TTM + userptr) memory - 3/4th System RAM + * TTM memory - 3/8th System RAM */ void amdgpu_amdkfd_gpuvm_init_mem_limits(void) { @@ -103,48 +104,61 @@ void amdgpu_amdkfd_gpuvm_init_mem_limits(void) mem *= si.mem_unit; spin_lock_init(&kfd_mem_limit.mem_limit_lock); - kfd_mem_limit.max_system_mem_limit = (mem >> 1) - (mem >> 3); - kfd_mem_limit.max_userptr_mem_limit = mem - (mem >> 2); - pr_debug("Kernel memory limit %lluM, userptr limit %lluM\n", + kfd_mem_limit.max_system_mem_limit = (mem >> 1) + (mem >> 2); + kfd_mem_limit.max_ttm_mem_limit = (mem >> 1) - (mem >> 3); + pr_debug("Kernel memory limit %lluM, TTM limit %lluM\n", (kfd_mem_limit.max_system_mem_limit >> 20), - (kfd_mem_limit.max_userptr_mem_limit >> 20)); + (kfd_mem_limit.max_ttm_mem_limit >> 20)); } -static int amdgpu_amdkfd_reserve_system_mem_limit(struct amdgpu_device *adev, - uint64_t size, u32 domain) +static int amdgpu_amdkfd_reserve_mem_limit(struct amdgpu_device *adev, + uint64_t size, u32 domain, bool sg) { - size_t acc_size; + size_t acc_size, system_mem_needed, ttm_mem_needed, vram_needed; + uint64_t reserved_for_pt = amdgpu_amdkfd_total_mem_size >> 9; int ret = 0; acc_size = ttm_bo_dma_acc_size(&adev->mman.bdev, size, sizeof(struct amdgpu_bo)); - spin_lock(&kfd_mem_limit.mem_limit_lock); + vram_needed = 0; if (domain == AMDGPU_GEM_DOMAIN_GTT) { - if (kfd_mem_limit.system_mem_used + (acc_size + size) > - kfd_mem_limit.max_system_mem_limit) { - ret = -ENOMEM; - goto err_no_mem; - } - kfd_mem_limit.system_mem_used += (acc_size + size); - } else if (domain == AMDGPU_GEM_DOMAIN_CPU) { - if ((kfd_mem_limit.system_mem_used + acc_size > - kfd_mem_limit.max_system_mem_limit) || - (kfd_mem_limit.userptr_mem_used + (size + acc_size) > - kfd_mem_limit.max_userptr_mem_limit)) { - ret = -ENOMEM; - goto err_no_mem; - } - kfd_mem_limit.system_mem_used += acc_size; - kfd_mem_limit.userptr_mem_used += size; + /* TTM GTT memory */ + system_mem_needed = acc_size + size; + ttm_mem_needed = acc_size + size; + } else if (domain == AMDGPU_GEM_DOMAIN_CPU && !sg) { + /* Userptr */ + system_mem_needed = acc_size + size; + ttm_mem_needed = acc_size; + } else { + /* VRAM and SG */ + system_mem_needed = acc_size; + ttm_mem_needed = acc_size; + if (domain == AMDGPU_GEM_DOMAIN_VRAM) + vram_needed = size; + } + + spin_lock(&kfd_mem_limit.mem_limit_lock); + + if ((kfd_mem_limit.system_mem_used + system_mem_needed > + kfd_mem_limit.max_system_mem_limit) || + (kfd_mem_limit.ttm_mem_used + ttm_mem_needed > + kfd_mem_limit.max_ttm_mem_limit) || + (adev->kfd.vram_used + vram_needed > + adev->gmc.real_vram_size - reserved_for_pt)) { + ret = -ENOMEM; + } else { + kfd_mem_limit.system_mem_used += system_mem_needed; + kfd_mem_limit.ttm_mem_used += ttm_mem_needed; + adev->kfd.vram_used += vram_needed; } -err_no_mem: + spin_unlock(&kfd_mem_limit.mem_limit_lock); return ret; } -static void unreserve_system_mem_limit(struct amdgpu_device *adev, - uint64_t size, u32 domain) +static void unreserve_mem_limit(struct amdgpu_device *adev, + uint64_t size, u32 domain, bool sg) { size_t acc_size; @@ -154,35 +168,39 @@ static void unreserve_system_mem_limit(struct amdgpu_device *adev, spin_lock(&kfd_mem_limit.mem_limit_lock); if (domain == AMDGPU_GEM_DOMAIN_GTT) { kfd_mem_limit.system_mem_used -= (acc_size + size); - } else if (domain == AMDGPU_GEM_DOMAIN_CPU) { + kfd_mem_limit.ttm_mem_used -= (acc_size + size); + } else if (domain == AMDGPU_GEM_DOMAIN_CPU && !sg) { + kfd_mem_limit.system_mem_used -= (acc_size + size); + kfd_mem_limit.ttm_mem_used -= acc_size; + } else { kfd_mem_limit.system_mem_used -= acc_size; - kfd_mem_limit.userptr_mem_used -= size; + kfd_mem_limit.ttm_mem_used -= acc_size; + if (domain == AMDGPU_GEM_DOMAIN_VRAM) { + adev->kfd.vram_used -= size; + WARN_ONCE(adev->kfd.vram_used < 0, + "kfd VRAM memory accounting unbalanced"); + } } WARN_ONCE(kfd_mem_limit.system_mem_used < 0, "kfd system memory accounting unbalanced"); - WARN_ONCE(kfd_mem_limit.userptr_mem_used < 0, - "kfd userptr memory accounting unbalanced"); + WARN_ONCE(kfd_mem_limit.ttm_mem_used < 0, + "kfd TTM memory accounting unbalanced"); spin_unlock(&kfd_mem_limit.mem_limit_lock); } -void amdgpu_amdkfd_unreserve_system_memory_limit(struct amdgpu_bo *bo) +void amdgpu_amdkfd_unreserve_memory_limit(struct amdgpu_bo *bo) { - spin_lock(&kfd_mem_limit.mem_limit_lock); + struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev); + u32 domain = bo->preferred_domains; + bool sg = (bo->preferred_domains == AMDGPU_GEM_DOMAIN_CPU); if (bo->flags & AMDGPU_AMDKFD_USERPTR_BO) { - kfd_mem_limit.system_mem_used -= bo->tbo.acc_size; - kfd_mem_limit.userptr_mem_used -= amdgpu_bo_size(bo); - } else if (bo->preferred_domains == AMDGPU_GEM_DOMAIN_GTT) { - kfd_mem_limit.system_mem_used -= - (bo->tbo.acc_size + amdgpu_bo_size(bo)); + domain = AMDGPU_GEM_DOMAIN_CPU; + sg = false; } - WARN_ONCE(kfd_mem_limit.system_mem_used < 0, - "kfd system memory accounting unbalanced"); - WARN_ONCE(kfd_mem_limit.userptr_mem_used < 0, - "kfd userptr memory accounting unbalanced"); - spin_unlock(&kfd_mem_limit.mem_limit_lock); + unreserve_mem_limit(adev, amdgpu_bo_size(bo), domain, sg); } @@ -395,23 +413,6 @@ static int vm_validate_pt_pd_bos(struct amdgpu_vm *vm) return 0; } -static int sync_vm_fence(struct amdgpu_device *adev, struct amdgpu_sync *sync, - struct dma_fence *f) -{ - int ret = amdgpu_sync_fence(adev, sync, f, false); - - /* Sync objects can't handle multiple GPUs (contexts) updating - * sync->last_vm_update. Fortunately we don't need it for - * KFD's purposes, so we can just drop that fence. - */ - if (sync->last_vm_update) { - dma_fence_put(sync->last_vm_update); - sync->last_vm_update = NULL; - } - - return ret; -} - static int vm_update_pds(struct amdgpu_vm *vm, struct amdgpu_sync *sync) { struct amdgpu_bo *pd = vm->root.base.bo; @@ -422,7 +423,7 @@ static int vm_update_pds(struct amdgpu_vm *vm, struct amdgpu_sync *sync) if (ret) return ret; - return sync_vm_fence(adev, sync, vm->last_update); + return amdgpu_sync_fence(NULL, sync, vm->last_update, false); } /* add_bo_to_vm - Add a BO to a VM @@ -536,7 +537,7 @@ static void add_kgd_mem_to_kfd_bo_list(struct kgd_mem *mem, struct amdgpu_bo *bo = mem->bo; INIT_LIST_HEAD(&entry->head); - entry->shared = true; + entry->num_shared = 1; entry->bo = &bo->tbo; mutex_lock(&process_info->lock); if (userptr) @@ -677,7 +678,7 @@ static int reserve_bo_and_vm(struct kgd_mem *mem, ctx->kfd_bo.priority = 0; ctx->kfd_bo.tv.bo = &bo->tbo; - ctx->kfd_bo.tv.shared = true; + ctx->kfd_bo.tv.num_shared = 1; ctx->kfd_bo.user_pages = NULL; list_add(&ctx->kfd_bo.tv.head, &ctx->list); @@ -741,7 +742,7 @@ static int reserve_bo_and_cond_vms(struct kgd_mem *mem, ctx->kfd_bo.priority = 0; ctx->kfd_bo.tv.bo = &bo->tbo; - ctx->kfd_bo.tv.shared = true; + ctx->kfd_bo.tv.num_shared = 1; ctx->kfd_bo.user_pages = NULL; list_add(&ctx->kfd_bo.tv.head, &ctx->list); @@ -826,7 +827,7 @@ static int unmap_bo_from_gpuvm(struct amdgpu_device *adev, /* Add the eviction fence back */ amdgpu_bo_fence(pd, &vm->process_info->eviction_fence->base, true); - sync_vm_fence(adev, sync, bo_va->last_pt_update); + amdgpu_sync_fence(NULL, sync, bo_va->last_pt_update, false); return 0; } @@ -851,7 +852,7 @@ static int update_gpuvm_pte(struct amdgpu_device *adev, return ret; } - return sync_vm_fence(adev, sync, bo_va->last_pt_update); + return amdgpu_sync_fence(NULL, sync, bo_va->last_pt_update, false); } static int map_bo_to_gpuvm(struct amdgpu_device *adev, @@ -886,6 +887,24 @@ update_gpuvm_pte_failed: return ret; } +static struct sg_table *create_doorbell_sg(uint64_t addr, uint32_t size) +{ + struct sg_table *sg = kmalloc(sizeof(*sg), GFP_KERNEL); + + if (!sg) + return NULL; + if (sg_alloc_table(sg, 1, GFP_KERNEL)) { + kfree(sg); + return NULL; + } + sg->sgl->dma_address = addr; + sg->sgl->length = size; +#ifdef CONFIG_NEED_SG_DMA_LENGTH + sg->sgl->dma_length = size; +#endif + return sg; +} + static int process_validate_vms(struct amdkfd_process_info *process_info) { struct amdgpu_vm *peer_vm; @@ -901,6 +920,26 @@ static int process_validate_vms(struct amdkfd_process_info *process_info) return 0; } +static int process_sync_pds_resv(struct amdkfd_process_info *process_info, + struct amdgpu_sync *sync) +{ + struct amdgpu_vm *peer_vm; + int ret; + + list_for_each_entry(peer_vm, &process_info->vm_list_head, + vm_list_node) { + struct amdgpu_bo *pd = peer_vm->root.base.bo; + + ret = amdgpu_sync_resv(NULL, + sync, pd->tbo.resv, + AMDGPU_FENCE_OWNER_UNDEFINED, false); + if (ret) + return ret; + } + + return 0; +} + static int process_update_pds(struct amdkfd_process_info *process_info, struct amdgpu_sync *sync) { @@ -1149,6 +1188,8 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( { struct amdgpu_device *adev = get_amdgpu_device(kgd); struct amdgpu_vm *avm = (struct amdgpu_vm *)vm; + enum ttm_bo_type bo_type = ttm_bo_type_device; + struct sg_table *sg = NULL; uint64_t user_addr = 0; struct amdgpu_bo *bo; struct amdgpu_bo_param bp; @@ -1177,13 +1218,25 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( if (!offset || !*offset) return -EINVAL; user_addr = *offset; + } else if (flags & ALLOC_MEM_FLAGS_DOORBELL) { + domain = AMDGPU_GEM_DOMAIN_GTT; + alloc_domain = AMDGPU_GEM_DOMAIN_CPU; + bo_type = ttm_bo_type_sg; + alloc_flags = 0; + if (size > UINT_MAX) + return -EINVAL; + sg = create_doorbell_sg(*offset, size); + if (!sg) + return -ENOMEM; } else { return -EINVAL; } *mem = kzalloc(sizeof(struct kgd_mem), GFP_KERNEL); - if (!*mem) - return -ENOMEM; + if (!*mem) { + ret = -ENOMEM; + goto err; + } INIT_LIST_HEAD(&(*mem)->bo_va_list); mutex_init(&(*mem)->lock); (*mem)->aql_queue = !!(flags & ALLOC_MEM_FLAGS_AQL_QUEUE_MEM); @@ -1199,7 +1252,8 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( byte_align = (adev->family == AMDGPU_FAMILY_VI && adev->asic_type != CHIP_FIJI && adev->asic_type != CHIP_POLARIS10 && - adev->asic_type != CHIP_POLARIS11) ? + adev->asic_type != CHIP_POLARIS11 && + adev->asic_type != CHIP_POLARIS12) ? VI_BO_SIZE_ALIGN : 1; mapping_flags = AMDGPU_VM_PAGE_READABLE; @@ -1215,10 +1269,10 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( amdgpu_sync_create(&(*mem)->sync); - ret = amdgpu_amdkfd_reserve_system_mem_limit(adev, size, alloc_domain); + ret = amdgpu_amdkfd_reserve_mem_limit(adev, size, alloc_domain, !!sg); if (ret) { pr_debug("Insufficient system memory\n"); - goto err_reserve_system_mem; + goto err_reserve_limit; } pr_debug("\tcreate BO VA 0x%llx size 0x%llx domain %s\n", @@ -1229,7 +1283,7 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( bp.byte_align = byte_align; bp.domain = alloc_domain; bp.flags = alloc_flags; - bp.type = ttm_bo_type_device; + bp.type = bo_type; bp.resv = NULL; ret = amdgpu_bo_create(adev, &bp, &bo); if (ret) { @@ -1237,6 +1291,10 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( domain_string(alloc_domain), ret); goto err_bo_create; } + if (bo_type == ttm_bo_type_sg) { + bo->tbo.sg = sg; + bo->tbo.ttm->sg = sg; + } bo->kfd_bo = *mem; (*mem)->bo = bo; if (user_addr) @@ -1266,12 +1324,17 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( allocate_init_user_pages_failed: amdgpu_bo_unref(&bo); /* Don't unreserve system mem limit twice */ - goto err_reserve_system_mem; + goto err_reserve_limit; err_bo_create: - unreserve_system_mem_limit(adev, size, alloc_domain); -err_reserve_system_mem: + unreserve_mem_limit(adev, size, alloc_domain, !!sg); +err_reserve_limit: mutex_destroy(&(*mem)->lock); kfree(*mem); +err: + if (sg) { + sg_free_table(sg); + kfree(sg); + } return ret; } @@ -1341,6 +1404,14 @@ int amdgpu_amdkfd_gpuvm_free_memory_of_gpu( /* Free the sync object */ amdgpu_sync_free(&mem->sync); + /* If the SG is not NULL, it's one we created for a doorbell + * BO. We need to free it. + */ + if (mem->bo->tbo.sg) { + sg_free_table(mem->bo->tbo.sg); + kfree(mem->bo->tbo.sg); + } + /* Free the BO*/ amdgpu_bo_unref(&mem->bo); mutex_destroy(&mem->lock); @@ -1405,7 +1476,8 @@ int amdgpu_amdkfd_gpuvm_map_memory_to_gpu( * the queues are still stopped and we can leave mapping for * the next restore worker */ - if (bo->tbo.mem.mem_type == TTM_PL_SYSTEM) + if (amdgpu_ttm_tt_get_usermm(bo->tbo.ttm) && + bo->tbo.mem.mem_type == TTM_PL_SYSTEM) is_invalid_userptr = true; if (check_if_add_bo_to_vm(avm, mem)) { @@ -1642,6 +1714,60 @@ int amdgpu_amdkfd_gpuvm_get_vm_fault_info(struct kgd_dev *kgd, return 0; } +int amdgpu_amdkfd_gpuvm_import_dmabuf(struct kgd_dev *kgd, + struct dma_buf *dma_buf, + uint64_t va, void *vm, + struct kgd_mem **mem, uint64_t *size, + uint64_t *mmap_offset) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)kgd; + struct drm_gem_object *obj; + struct amdgpu_bo *bo; + struct amdgpu_vm *avm = (struct amdgpu_vm *)vm; + + if (dma_buf->ops != &amdgpu_dmabuf_ops) + /* Can't handle non-graphics buffers */ + return -EINVAL; + + obj = dma_buf->priv; + if (obj->dev->dev_private != adev) + /* Can't handle buffers from other devices */ + return -EINVAL; + + bo = gem_to_amdgpu_bo(obj); + if (!(bo->preferred_domains & (AMDGPU_GEM_DOMAIN_VRAM | + AMDGPU_GEM_DOMAIN_GTT))) + /* Only VRAM and GTT BOs are supported */ + return -EINVAL; + + *mem = kzalloc(sizeof(struct kgd_mem), GFP_KERNEL); + if (!*mem) + return -ENOMEM; + + if (size) + *size = amdgpu_bo_size(bo); + + if (mmap_offset) + *mmap_offset = amdgpu_bo_mmap_offset(bo); + + INIT_LIST_HEAD(&(*mem)->bo_va_list); + mutex_init(&(*mem)->lock); + (*mem)->mapping_flags = + AMDGPU_VM_PAGE_READABLE | AMDGPU_VM_PAGE_WRITEABLE | + AMDGPU_VM_PAGE_EXECUTABLE | AMDGPU_VM_MTYPE_NC; + + (*mem)->bo = amdgpu_bo_ref(bo); + (*mem)->va = va; + (*mem)->domain = (bo->preferred_domains & AMDGPU_GEM_DOMAIN_VRAM) ? + AMDGPU_GEM_DOMAIN_VRAM : AMDGPU_GEM_DOMAIN_GTT; + (*mem)->mapped_to_gpu_memory = 0; + (*mem)->process_info = avm->process_info; + add_kgd_mem_to_kfd_bo_list(*mem, avm->process_info, false); + amdgpu_sync_create(&(*mem)->sync); + + return 0; +} + /* Evict a userptr BO by stopping the queues if necessary * * Runs in MMU notifier, may be in RECLAIM_FS context. This means it @@ -1808,7 +1934,7 @@ static int validate_invalid_user_pages(struct amdkfd_process_info *process_info) validate_list.head) { list_add_tail(&mem->resv_list.head, &resv_list); mem->resv_list.bo = mem->validate_list.bo; - mem->resv_list.shared = mem->validate_list.shared; + mem->resv_list.num_shared = mem->validate_list.num_shared; } /* Reserve all BOs and page tables for validation */ @@ -2027,7 +2153,7 @@ int amdgpu_amdkfd_gpuvm_restore_process_bos(void *info, struct dma_fence **ef) list_add_tail(&mem->resv_list.head, &ctx.list); mem->resv_list.bo = mem->validate_list.bo; - mem->resv_list.shared = mem->validate_list.shared; + mem->resv_list.num_shared = mem->validate_list.num_shared; } ret = ttm_eu_reserve_buffers(&ctx.ticket, &ctx.list, @@ -2044,13 +2170,10 @@ int amdgpu_amdkfd_gpuvm_restore_process_bos(void *info, struct dma_fence **ef) if (ret) goto validate_map_fail; - /* Wait for PD/PTs validate to finish */ - /* FIXME: I think this isn't needed */ - list_for_each_entry(peer_vm, &process_info->vm_list_head, - vm_list_node) { - struct amdgpu_bo *bo = peer_vm->root.base.bo; - - ttm_bo_wait(&bo->tbo, false, false); + ret = process_sync_pds_resv(process_info, &sync_obj); + if (ret) { + pr_debug("Memory eviction: Failed to sync to PD BO moving fence. Try again\n"); + goto validate_map_fail; } /* Validate BOs and map them to GPUVM (update VM page tables). */ @@ -2066,7 +2189,11 @@ int amdgpu_amdkfd_gpuvm_restore_process_bos(void *info, struct dma_fence **ef) pr_debug("Memory eviction: Validate BOs failed. Try again\n"); goto validate_map_fail; } - + ret = amdgpu_sync_fence(NULL, &sync_obj, bo->tbo.moving, false); + if (ret) { + pr_debug("Memory eviction: Sync BO fence failed. Try again\n"); + goto validate_map_fail; + } list_for_each_entry(bo_va_entry, &mem->bo_va_list, bo_list) { ret = update_gpuvm_pte((struct amdgpu_device *) @@ -2087,6 +2214,7 @@ int amdgpu_amdkfd_gpuvm_restore_process_bos(void *info, struct dma_fence **ef) goto validate_map_fail; } + /* Wait for validate and PT updates to finish */ amdgpu_sync_wait(&sync_obj, false); /* Release old eviction fence and create new one, because fence only @@ -2105,10 +2233,7 @@ int amdgpu_amdkfd_gpuvm_restore_process_bos(void *info, struct dma_fence **ef) process_info->eviction_fence = new_fence; *ef = dma_fence_get(&new_fence->base); - /* Wait for validate to finish and attach new eviction fence */ - list_for_each_entry(mem, &process_info->kfd_bo_list, - validate_list.head) - ttm_bo_wait(&mem->bo->tbo, false, false); + /* Attach new eviction fence to all BOs */ list_for_each_entry(mem, &process_info->kfd_bo_list, validate_list.head) amdgpu_bo_fence(mem->bo, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c index 14d2982a47cc..5c79da8e1150 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c @@ -118,7 +118,6 @@ int amdgpu_bo_list_create(struct amdgpu_device *adev, struct drm_file *filp, entry->priority = min(info[i].bo_priority, AMDGPU_BO_LIST_MAX_PRIORITY); entry->tv.bo = &bo->tbo; - entry->tv.shared = !bo->prime_shared_count; if (bo->preferred_domains == AMDGPU_GEM_DOMAIN_GDS) list->gds_obj = bo; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c index ceadeeadfa56..387f1cf1dc20 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c @@ -381,7 +381,8 @@ static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device, (adev->pdev->revision == 0xe7) || (adev->pdev->revision == 0xef))) || ((adev->pdev->device == 0x6fdf) && - (adev->pdev->revision == 0xef))) { + ((adev->pdev->revision == 0xef) || + (adev->pdev->revision == 0xff)))) { info->is_kicker = true; strcpy(fw_name, "amdgpu/polaris10_k_smc.bin"); } else if ((adev->pdev->device == 0x67df) && diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c index dc54e9efd910..1c49b8266d69 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c @@ -50,7 +50,8 @@ static int amdgpu_cs_user_fence_chunk(struct amdgpu_cs_parser *p, bo = amdgpu_bo_ref(gem_to_amdgpu_bo(gobj)); p->uf_entry.priority = 0; p->uf_entry.tv.bo = &bo->tbo; - p->uf_entry.tv.shared = true; + /* One for TTM and one for the CS job */ + p->uf_entry.tv.num_shared = 2; p->uf_entry.user_pages = NULL; drm_gem_object_put_unlocked(gobj); @@ -124,14 +125,14 @@ static int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, union drm_amdgpu_cs goto free_chunk; } + mutex_lock(&p->ctx->lock); + /* skip guilty context job */ if (atomic_read(&p->ctx->guilty) == 1) { ret = -ECANCELED; goto free_chunk; } - mutex_lock(&p->ctx->lock); - /* get chunks */ chunk_array_user = u64_to_user_ptr(cs->in.chunks); if (copy_from_user(chunk_array, chunk_array_user, @@ -598,6 +599,10 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p, return r; } + /* One for TTM and one for the CS job */ + amdgpu_bo_list_for_each_entry(e, p->bo_list) + e->tv.num_shared = 2; + amdgpu_bo_list_get_list(p->bo_list, &p->validated); if (p->bo_list->first_userptr != p->bo_list->num_entries) p->mn = amdgpu_mn_get(p->adev, AMDGPU_MN_TYPE_GFX); @@ -717,8 +722,14 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p, gws = p->bo_list->gws_obj; oa = p->bo_list->oa_obj; - amdgpu_bo_list_for_each_entry(e, p->bo_list) - e->bo_va = amdgpu_vm_bo_find(vm, ttm_to_amdgpu_bo(e->tv.bo)); + amdgpu_bo_list_for_each_entry(e, p->bo_list) { + struct amdgpu_bo *bo = ttm_to_amdgpu_bo(e->tv.bo); + + /* Make sure we use the exclusive slot for shared BOs */ + if (bo->prime_shared_count) + e->tv.num_shared = 0; + e->bo_va = amdgpu_vm_bo_find(vm, bo); + } if (gds) { p->job->gds_base = amdgpu_bo_gpu_offset(gds) >> PAGE_SHIFT; @@ -955,10 +966,6 @@ static int amdgpu_cs_vm_handling(struct amdgpu_cs_parser *p) if (r) return r; - r = reservation_object_reserve_shared(vm->root.base.bo->tbo.resv, 1); - if (r) - return r; - p->job->vm_pd_addr = amdgpu_gmc_pd_addr(vm->root.base.bo); if (amdgpu_vm_debug) { @@ -1421,6 +1428,9 @@ int amdgpu_cs_fence_to_handle_ioctl(struct drm_device *dev, void *data, if (IS_ERR(fence)) return PTR_ERR(fence); + if (!fence) + fence = dma_fence_get_stub(); + switch (info->in.what) { case AMDGPU_FENCE_TO_HANDLE_GET_SYNCOBJ: r = drm_syncobj_create(&syncobj, 0, fence); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c index 0c590ddf250a..7e22be7ca68a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c @@ -43,7 +43,7 @@ int amdgpu_allocate_static_csa(struct amdgpu_device *adev, struct amdgpu_bo **bo r = amdgpu_bo_create_kernel(adev, size, PAGE_SIZE, domain, bo, NULL, &ptr); - if (!bo) + if (!*bo) return -ENOMEM; memset(ptr, 0, size); @@ -74,7 +74,7 @@ int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm, INIT_LIST_HEAD(&list); INIT_LIST_HEAD(&csa_tv.head); csa_tv.bo = &bo->tbo; - csa_tv.shared = true; + csa_tv.num_shared = 1; list_add(&csa_tv.head, &list); amdgpu_vm_get_pd_bo(vm, &list, &pd); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c index f9b54236102d..d85184b5b35c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c @@ -39,6 +39,7 @@ const unsigned int amdgpu_ctx_num_entities[AMDGPU_HW_IP_NUM] = { [AMDGPU_HW_IP_UVD_ENC] = 1, [AMDGPU_HW_IP_VCN_DEC] = 1, [AMDGPU_HW_IP_VCN_ENC] = 1, + [AMDGPU_HW_IP_VCN_JPEG] = 1, }; static int amdgput_ctx_total_num_entities(void) @@ -247,7 +248,7 @@ static int amdgpu_ctx_alloc(struct amdgpu_device *adev, return -ENOMEM; mutex_lock(&mgr->lock); - r = idr_alloc(&mgr->ctx_handles, ctx, 1, 0, GFP_KERNEL); + r = idr_alloc(&mgr->ctx_handles, ctx, 1, AMDGPU_VM_MAX_NUM_CTX, GFP_KERNEL); if (r < 0) { mutex_unlock(&mgr->lock); kfree(ctx); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 590588a82471..8a078f4ae73d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -59,6 +59,8 @@ #include "amdgpu_amdkfd.h" #include "amdgpu_pm.h" +#include "amdgpu_xgmi.h" + MODULE_FIRMWARE("amdgpu/vega10_gpu_info.bin"); MODULE_FIRMWARE("amdgpu/vega12_gpu_info.bin"); MODULE_FIRMWARE("amdgpu/raven_gpu_info.bin"); @@ -513,6 +515,7 @@ void amdgpu_device_pci_config_reset(struct amdgpu_device *adev) */ static int amdgpu_device_doorbell_init(struct amdgpu_device *adev) { + /* No doorbell on SI hardware generation */ if (adev->asic_type < CHIP_BONAIRE) { adev->doorbell.base = 0; @@ -525,15 +528,26 @@ static int amdgpu_device_doorbell_init(struct amdgpu_device *adev) if (pci_resource_flags(adev->pdev, 2) & IORESOURCE_UNSET) return -EINVAL; + amdgpu_asic_init_doorbell_index(adev); + /* doorbell bar mapping */ adev->doorbell.base = pci_resource_start(adev->pdev, 2); adev->doorbell.size = pci_resource_len(adev->pdev, 2); adev->doorbell.num_doorbells = min_t(u32, adev->doorbell.size / sizeof(u32), - AMDGPU_DOORBELL_MAX_ASSIGNMENT+1); + adev->doorbell_index.max_assignment+1); if (adev->doorbell.num_doorbells == 0) return -EINVAL; + /* For Vega, reserve and map two pages on doorbell BAR since SDMA + * paging queue doorbell use the second page. The + * AMDGPU_DOORBELL64_MAX_ASSIGNMENT definition assumes all the + * doorbells are in the first page. So with paging queue enabled, + * the max num_doorbells should + 1 page (0x400 in dword) + */ + if (adev->asic_type >= CHIP_VEGA10) + adev->doorbell.num_doorbells += 0x400; + adev->doorbell.ptr = ioremap(adev->doorbell.base, adev->doorbell.num_doorbells * sizeof(u32)); @@ -1851,6 +1865,9 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev) { int i, r; + if (adev->gmc.xgmi.num_physical_nodes > 1) + amdgpu_xgmi_remove_device(adev); + amdgpu_amdkfd_device_fini(adev); amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE); @@ -2340,6 +2357,19 @@ bool amdgpu_device_has_dc_support(struct amdgpu_device *adev) return amdgpu_device_asic_has_dc_support(adev->asic_type); } + +static void amdgpu_device_xgmi_reset_func(struct work_struct *__work) +{ + struct amdgpu_device *adev = + container_of(__work, struct amdgpu_device, xgmi_reset_work); + + adev->asic_reset_res = amdgpu_asic_reset(adev); + if (adev->asic_reset_res) + DRM_WARN("ASIC reset failed with err r, %d for drm dev, %s", + adev->asic_reset_res, adev->ddev->unique); +} + + /** * amdgpu_device_init - initialize the driver * @@ -2438,6 +2468,8 @@ int amdgpu_device_init(struct amdgpu_device *adev, INIT_DELAYED_WORK(&adev->gfx.gfx_off_delay_work, amdgpu_device_delay_enable_gfx_off); + INIT_WORK(&adev->xgmi_reset_work, amdgpu_device_xgmi_reset_func); + adev->gfx.gfx_off_req_count = 1; adev->pm.ac_power = power_supply_is_system_supplied() > 0 ? true : false; @@ -2458,9 +2490,6 @@ int amdgpu_device_init(struct amdgpu_device *adev, DRM_INFO("register mmio base: 0x%08X\n", (uint32_t)adev->rmmio_base); DRM_INFO("register mmio size: %u\n", (unsigned)adev->rmmio_size); - /* doorbell bar mapping */ - amdgpu_device_doorbell_init(adev); - /* io port mapping */ for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) { if (pci_resource_flags(adev->pdev, i) & IORESOURCE_IO) { @@ -2479,6 +2508,9 @@ int amdgpu_device_init(struct amdgpu_device *adev, if (r) return r; + /* doorbell bar mapping and doorbell index init*/ + amdgpu_device_doorbell_init(adev); + /* if we have > 1 VGA cards, then disable the amdgpu VGA resources */ /* this will fail for cards that aren't VGA class devices, just * ignore it */ @@ -3151,86 +3183,6 @@ static int amdgpu_device_recover_vram(struct amdgpu_device *adev) return 0; } -/** - * amdgpu_device_reset - reset ASIC/GPU for bare-metal or passthrough - * - * @adev: amdgpu device pointer - * - * attempt to do soft-reset or full-reset and reinitialize Asic - * return 0 means succeeded otherwise failed - */ -static int amdgpu_device_reset(struct amdgpu_device *adev) -{ - bool need_full_reset, vram_lost = 0; - int r; - - need_full_reset = amdgpu_device_ip_need_full_reset(adev); - - if (!need_full_reset) { - amdgpu_device_ip_pre_soft_reset(adev); - r = amdgpu_device_ip_soft_reset(adev); - amdgpu_device_ip_post_soft_reset(adev); - if (r || amdgpu_device_ip_check_soft_reset(adev)) { - DRM_INFO("soft reset failed, will fallback to full reset!\n"); - need_full_reset = true; - } - } - - if (need_full_reset) { - r = amdgpu_device_ip_suspend(adev); - -retry: - r = amdgpu_asic_reset(adev); - /* post card */ - amdgpu_atom_asic_init(adev->mode_info.atom_context); - - if (!r) { - dev_info(adev->dev, "GPU reset succeeded, trying to resume\n"); - r = amdgpu_device_ip_resume_phase1(adev); - if (r) - goto out; - - vram_lost = amdgpu_device_check_vram_lost(adev); - if (vram_lost) { - DRM_ERROR("VRAM is lost!\n"); - atomic_inc(&adev->vram_lost_counter); - } - - r = amdgpu_gtt_mgr_recover( - &adev->mman.bdev.man[TTM_PL_TT]); - if (r) - goto out; - - r = amdgpu_device_fw_loading(adev); - if (r) - return r; - - r = amdgpu_device_ip_resume_phase2(adev); - if (r) - goto out; - - if (vram_lost) - amdgpu_device_fill_reset_magic(adev); - } - } - -out: - if (!r) { - amdgpu_irq_gpu_reset_resume_helper(adev); - r = amdgpu_ib_ring_tests(adev); - if (r) { - dev_err(adev->dev, "ib ring test failed (%d).\n", r); - r = amdgpu_device_ip_suspend(adev); - need_full_reset = true; - goto retry; - } - } - - if (!r) - r = amdgpu_device_recover_vram(adev); - - return r; -} /** * amdgpu_device_reset_sriov - reset ASIC for SR-IOV vf @@ -3306,6 +3258,8 @@ bool amdgpu_device_should_recover_gpu(struct amdgpu_device *adev) if (amdgpu_gpu_recovery == -1) { switch (adev->asic_type) { + case CHIP_BONAIRE: + case CHIP_HAWAII: case CHIP_TOPAZ: case CHIP_TONGA: case CHIP_FIJI: @@ -3329,31 +3283,13 @@ disabled: return false; } -/** - * amdgpu_device_gpu_recover - reset the asic and recover scheduler - * - * @adev: amdgpu device pointer - * @job: which job trigger hang - * - * Attempt to reset the GPU if it has hung (all asics). - * Returns 0 for success or an error on failure. - */ -int amdgpu_device_gpu_recover(struct amdgpu_device *adev, - struct amdgpu_job *job) -{ - int i, r, resched; - - dev_info(adev->dev, "GPU reset begin!\n"); - - mutex_lock(&adev->lock_reset); - atomic_inc(&adev->gpu_reset_counter); - adev->in_gpu_reset = 1; - - /* Block kfd */ - amdgpu_amdkfd_pre_reset(adev); - /* block TTM */ - resched = ttm_bo_lock_delayed_workqueue(&adev->mman.bdev); +static int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev, + struct amdgpu_job *job, + bool *need_full_reset_arg) +{ + int i, r = 0; + bool need_full_reset = *need_full_reset_arg; /* block all schedulers and reset given job's ring */ for (i = 0; i < AMDGPU_MAX_RINGS; ++i) { @@ -3373,10 +3309,144 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, amdgpu_fence_driver_force_completion(ring); } - if (amdgpu_sriov_vf(adev)) - r = amdgpu_device_reset_sriov(adev, job ? false : true); - else - r = amdgpu_device_reset(adev); + + + if (!amdgpu_sriov_vf(adev)) { + + if (!need_full_reset) + need_full_reset = amdgpu_device_ip_need_full_reset(adev); + + if (!need_full_reset) { + amdgpu_device_ip_pre_soft_reset(adev); + r = amdgpu_device_ip_soft_reset(adev); + amdgpu_device_ip_post_soft_reset(adev); + if (r || amdgpu_device_ip_check_soft_reset(adev)) { + DRM_INFO("soft reset failed, will fallback to full reset!\n"); + need_full_reset = true; + } + } + + if (need_full_reset) + r = amdgpu_device_ip_suspend(adev); + + *need_full_reset_arg = need_full_reset; + } + + return r; +} + +static int amdgpu_do_asic_reset(struct amdgpu_hive_info *hive, + struct list_head *device_list_handle, + bool *need_full_reset_arg) +{ + struct amdgpu_device *tmp_adev = NULL; + bool need_full_reset = *need_full_reset_arg, vram_lost = false; + int r = 0; + + /* + * ASIC reset has to be done on all HGMI hive nodes ASAP + * to allow proper links negotiation in FW (within 1 sec) + */ + if (need_full_reset) { + list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) { + /* For XGMI run all resets in parallel to speed up the process */ + if (tmp_adev->gmc.xgmi.num_physical_nodes > 1) { + if (!queue_work(system_highpri_wq, &tmp_adev->xgmi_reset_work)) + r = -EALREADY; + } else + r = amdgpu_asic_reset(tmp_adev); + + if (r) { + DRM_ERROR("ASIC reset failed with err r, %d for drm dev, %s", + r, tmp_adev->ddev->unique); + break; + } + } + + /* For XGMI wait for all PSP resets to complete before proceed */ + if (!r) { + list_for_each_entry(tmp_adev, device_list_handle, + gmc.xgmi.head) { + if (tmp_adev->gmc.xgmi.num_physical_nodes > 1) { + flush_work(&tmp_adev->xgmi_reset_work); + r = tmp_adev->asic_reset_res; + if (r) + break; + } + } + } + } + + + list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) { + if (need_full_reset) { + /* post card */ + if (amdgpu_atom_asic_init(tmp_adev->mode_info.atom_context)) + DRM_WARN("asic atom init failed!"); + + if (!r) { + dev_info(tmp_adev->dev, "GPU reset succeeded, trying to resume\n"); + r = amdgpu_device_ip_resume_phase1(tmp_adev); + if (r) + goto out; + + vram_lost = amdgpu_device_check_vram_lost(tmp_adev); + if (vram_lost) { + DRM_ERROR("VRAM is lost!\n"); + atomic_inc(&tmp_adev->vram_lost_counter); + } + + r = amdgpu_gtt_mgr_recover( + &tmp_adev->mman.bdev.man[TTM_PL_TT]); + if (r) + goto out; + + r = amdgpu_device_fw_loading(tmp_adev); + if (r) + return r; + + r = amdgpu_device_ip_resume_phase2(tmp_adev); + if (r) + goto out; + + if (vram_lost) + amdgpu_device_fill_reset_magic(tmp_adev); + + /* Update PSP FW topology after reset */ + if (hive && tmp_adev->gmc.xgmi.num_physical_nodes > 1) + r = amdgpu_xgmi_update_topology(hive, tmp_adev); + } + } + + +out: + if (!r) { + amdgpu_irq_gpu_reset_resume_helper(tmp_adev); + r = amdgpu_ib_ring_tests(tmp_adev); + if (r) { + dev_err(tmp_adev->dev, "ib ring test failed (%d).\n", r); + r = amdgpu_device_ip_suspend(tmp_adev); + need_full_reset = true; + r = -EAGAIN; + goto end; + } + } + + if (!r) + r = amdgpu_device_recover_vram(tmp_adev); + else + tmp_adev->asic_reset_res = r; + } + +end: + *need_full_reset_arg = need_full_reset; + return r; +} + +static void amdgpu_device_post_asic_reset(struct amdgpu_device *adev, + struct amdgpu_job *job) +{ + int i; for (i = 0; i < AMDGPU_MAX_RINGS; ++i) { struct amdgpu_ring *ring = adev->rings[i]; @@ -3388,7 +3458,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, * or all rings (in the case @job is NULL) * after above amdgpu_reset accomplished */ - if ((!job || job->base.sched == &ring->sched) && !r) + if ((!job || job->base.sched == &ring->sched) && !adev->asic_reset_res) drm_sched_job_recovery(&ring->sched); kthread_unpark(ring->sched.thread); @@ -3398,21 +3468,144 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, drm_helper_resume_force_mode(adev->ddev); } - ttm_bo_unlock_delayed_workqueue(&adev->mman.bdev, resched); + adev->asic_reset_res = 0; +} - if (r) { - /* bad news, how to tell it to userspace ? */ - dev_info(adev->dev, "GPU reset(%d) failed\n", atomic_read(&adev->gpu_reset_counter)); - amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_GPU_RESET_FAIL, 0, r); - } else { - dev_info(adev->dev, "GPU reset(%d) succeeded!\n",atomic_read(&adev->gpu_reset_counter)); - } +static void amdgpu_device_lock_adev(struct amdgpu_device *adev) +{ + mutex_lock(&adev->lock_reset); + atomic_inc(&adev->gpu_reset_counter); + adev->in_gpu_reset = 1; + /* Block kfd: SRIOV would do it separately */ + if (!amdgpu_sriov_vf(adev)) + amdgpu_amdkfd_pre_reset(adev); +} - /*unlock kfd */ - amdgpu_amdkfd_post_reset(adev); +static void amdgpu_device_unlock_adev(struct amdgpu_device *adev) +{ + /*unlock kfd: SRIOV would do it separately */ + if (!amdgpu_sriov_vf(adev)) + amdgpu_amdkfd_post_reset(adev); amdgpu_vf_error_trans_all(adev); adev->in_gpu_reset = 0; mutex_unlock(&adev->lock_reset); +} + + +/** + * amdgpu_device_gpu_recover - reset the asic and recover scheduler + * + * @adev: amdgpu device pointer + * @job: which job trigger hang + * + * Attempt to reset the GPU if it has hung (all asics). + * Attempt to do soft-reset or full-reset and reinitialize Asic + * Returns 0 for success or an error on failure. + */ + +int amdgpu_device_gpu_recover(struct amdgpu_device *adev, + struct amdgpu_job *job) +{ + int r; + struct amdgpu_hive_info *hive = NULL; + bool need_full_reset = false; + struct amdgpu_device *tmp_adev = NULL; + struct list_head device_list, *device_list_handle = NULL; + + INIT_LIST_HEAD(&device_list); + + dev_info(adev->dev, "GPU reset begin!\n"); + + /* + * In case of XGMI hive disallow concurrent resets to be triggered + * by different nodes. No point also since the one node already executing + * reset will also reset all the other nodes in the hive. + */ + hive = amdgpu_get_xgmi_hive(adev); + if (hive && adev->gmc.xgmi.num_physical_nodes > 1 && + !mutex_trylock(&hive->hive_lock)) + return 0; + + /* Start with adev pre asic reset first for soft reset check.*/ + amdgpu_device_lock_adev(adev); + r = amdgpu_device_pre_asic_reset(adev, + job, + &need_full_reset); + if (r) { + /*TODO Should we stop ?*/ + DRM_ERROR("GPU pre asic reset failed with err, %d for drm dev, %s ", + r, adev->ddev->unique); + adev->asic_reset_res = r; + } + + /* Build list of devices to reset */ + if (need_full_reset && adev->gmc.xgmi.num_physical_nodes > 1) { + if (!hive) { + amdgpu_device_unlock_adev(adev); + return -ENODEV; + } + + /* + * In case we are in XGMI hive mode device reset is done for all the + * nodes in the hive to retrain all XGMI links and hence the reset + * sequence is executed in loop on all nodes. + */ + device_list_handle = &hive->device_list; + } else { + list_add_tail(&adev->gmc.xgmi.head, &device_list); + device_list_handle = &device_list; + } + +retry: /* Rest of adevs pre asic reset from XGMI hive. */ + list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) { + + if (tmp_adev == adev) + continue; + + amdgpu_device_lock_adev(tmp_adev); + r = amdgpu_device_pre_asic_reset(tmp_adev, + NULL, + &need_full_reset); + /*TODO Should we stop ?*/ + if (r) { + DRM_ERROR("GPU pre asic reset failed with err, %d for drm dev, %s ", + r, tmp_adev->ddev->unique); + tmp_adev->asic_reset_res = r; + } + } + + /* Actual ASIC resets if needed.*/ + /* TODO Implement XGMI hive reset logic for SRIOV */ + if (amdgpu_sriov_vf(adev)) { + r = amdgpu_device_reset_sriov(adev, job ? false : true); + if (r) + adev->asic_reset_res = r; + } else { + r = amdgpu_do_asic_reset(hive, device_list_handle, &need_full_reset); + if (r && r == -EAGAIN) + goto retry; + } + + /* Post ASIC reset for all devs .*/ + list_for_each_entry(tmp_adev, device_list_handle, gmc.xgmi.head) { + amdgpu_device_post_asic_reset(tmp_adev, tmp_adev == adev ? job : NULL); + + if (r) { + /* bad news, how to tell it to userspace ? */ + dev_info(tmp_adev->dev, "GPU reset(%d) failed\n", atomic_read(&adev->gpu_reset_counter)); + amdgpu_vf_error_put(tmp_adev, AMDGIM_ERROR_VF_GPU_RESET_FAIL, 0, r); + } else { + dev_info(tmp_adev->dev, "GPU reset(%d) succeeded!\n", atomic_read(&adev->gpu_reset_counter)); + } + + amdgpu_device_unlock_adev(tmp_adev); + } + + if (hive && adev->gmc.xgmi.num_physical_nodes > 1) + mutex_unlock(&hive->hive_lock); + + if (r) + dev_info(adev->dev, "GPU reset end with ret = %d\n", r); return r; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c index 686a26de50f9..15ce7e681d67 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c @@ -631,6 +631,11 @@ int amdgpu_display_modeset_create_props(struct amdgpu_device *adev) drm_property_create_range(adev->ddev, 0, "max bpc", 8, 16); if (!adev->mode_info.max_bpc_property) return -ENOMEM; + adev->mode_info.abm_level_property = + drm_property_create_range(adev->ddev, 0, + "abm level", 0, 4); + if (!adev->mode_info.abm_level_property) + return -ENOMEM; } return 0; @@ -857,7 +862,12 @@ int amdgpu_display_get_crtc_scanoutpos(struct drm_device *dev, /* Inside "upper part" of vblank area? Apply corrective offset if so: */ if (in_vbl && (*vpos >= vbl_start)) { vtotal = mode->crtc_vtotal; - *vpos = *vpos - vtotal; + + /* With variable refresh rate displays the vpos can exceed + * the vtotal value. Clamp to 0 to return -vbl_end instead + * of guessing the remaining number of lines until scanout. + */ + *vpos = (*vpos < vtotal) ? (*vpos - vtotal) : 0; } /* Correct for shifted end of vbl at vbl_end. */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h new file mode 100644 index 000000000000..be620b29f4aa --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_doorbell.h @@ -0,0 +1,243 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +/* + * GPU doorbell structures, functions & helpers + */ +struct amdgpu_doorbell { + /* doorbell mmio */ + resource_size_t base; + resource_size_t size; + u32 __iomem *ptr; + u32 num_doorbells; /* Number of doorbells actually reserved for amdgpu. */ +}; + +/* Reserved doorbells for amdgpu (including multimedia). + * KFD can use all the rest in the 2M doorbell bar. + * For asic before vega10, doorbell is 32-bit, so the + * index/offset is in dword. For vega10 and after, doorbell + * can be 64-bit, so the index defined is in qword. + */ +struct amdgpu_doorbell_index { + uint32_t kiq; + uint32_t mec_ring0; + uint32_t mec_ring1; + uint32_t mec_ring2; + uint32_t mec_ring3; + uint32_t mec_ring4; + uint32_t mec_ring5; + uint32_t mec_ring6; + uint32_t mec_ring7; + uint32_t userqueue_start; + uint32_t userqueue_end; + uint32_t gfx_ring0; + uint32_t sdma_engine0; + uint32_t sdma_engine1; + uint32_t sdma_engine2; + uint32_t sdma_engine3; + uint32_t sdma_engine4; + uint32_t sdma_engine5; + uint32_t sdma_engine6; + uint32_t sdma_engine7; + uint32_t ih; + union { + struct { + uint32_t vcn_ring0_1; + uint32_t vcn_ring2_3; + uint32_t vcn_ring4_5; + uint32_t vcn_ring6_7; + } vcn; + struct { + uint32_t uvd_ring0_1; + uint32_t uvd_ring2_3; + uint32_t uvd_ring4_5; + uint32_t uvd_ring6_7; + uint32_t vce_ring0_1; + uint32_t vce_ring2_3; + uint32_t vce_ring4_5; + uint32_t vce_ring6_7; + } uvd_vce; + }; + uint32_t max_assignment; +}; + +typedef enum _AMDGPU_DOORBELL_ASSIGNMENT +{ + AMDGPU_DOORBELL_KIQ = 0x000, + AMDGPU_DOORBELL_HIQ = 0x001, + AMDGPU_DOORBELL_DIQ = 0x002, + AMDGPU_DOORBELL_MEC_RING0 = 0x010, + AMDGPU_DOORBELL_MEC_RING1 = 0x011, + AMDGPU_DOORBELL_MEC_RING2 = 0x012, + AMDGPU_DOORBELL_MEC_RING3 = 0x013, + AMDGPU_DOORBELL_MEC_RING4 = 0x014, + AMDGPU_DOORBELL_MEC_RING5 = 0x015, + AMDGPU_DOORBELL_MEC_RING6 = 0x016, + AMDGPU_DOORBELL_MEC_RING7 = 0x017, + AMDGPU_DOORBELL_GFX_RING0 = 0x020, + AMDGPU_DOORBELL_sDMA_ENGINE0 = 0x1E0, + AMDGPU_DOORBELL_sDMA_ENGINE1 = 0x1E1, + AMDGPU_DOORBELL_IH = 0x1E8, + AMDGPU_DOORBELL_MAX_ASSIGNMENT = 0x3FF, + AMDGPU_DOORBELL_INVALID = 0xFFFF +} AMDGPU_DOORBELL_ASSIGNMENT; + +typedef enum _AMDGPU_VEGA20_DOORBELL_ASSIGNMENT +{ + /* Compute + GFX: 0~255 */ + AMDGPU_VEGA20_DOORBELL_KIQ = 0x000, + AMDGPU_VEGA20_DOORBELL_HIQ = 0x001, + AMDGPU_VEGA20_DOORBELL_DIQ = 0x002, + AMDGPU_VEGA20_DOORBELL_MEC_RING0 = 0x003, + AMDGPU_VEGA20_DOORBELL_MEC_RING1 = 0x004, + AMDGPU_VEGA20_DOORBELL_MEC_RING2 = 0x005, + AMDGPU_VEGA20_DOORBELL_MEC_RING3 = 0x006, + AMDGPU_VEGA20_DOORBELL_MEC_RING4 = 0x007, + AMDGPU_VEGA20_DOORBELL_MEC_RING5 = 0x008, + AMDGPU_VEGA20_DOORBELL_MEC_RING6 = 0x009, + AMDGPU_VEGA20_DOORBELL_MEC_RING7 = 0x00A, + AMDGPU_VEGA20_DOORBELL_USERQUEUE_START = 0x00B, + AMDGPU_VEGA20_DOORBELL_USERQUEUE_END = 0x08A, + AMDGPU_VEGA20_DOORBELL_GFX_RING0 = 0x08B, + /* SDMA:256~335*/ + AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE0 = 0x100, + AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE1 = 0x10A, + AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE2 = 0x114, + AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE3 = 0x11E, + AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE4 = 0x128, + AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE5 = 0x132, + AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE6 = 0x13C, + AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE7 = 0x146, + /* IH: 376~391 */ + AMDGPU_VEGA20_DOORBELL_IH = 0x178, + /* MMSCH: 392~407 + * overlap the doorbell assignment with VCN as they are mutually exclusive + * VCE engine's doorbell is 32 bit and two VCE ring share one QWORD + */ + AMDGPU_VEGA20_DOORBELL64_VCN0_1 = 0x188, /* lower 32 bits for VNC0 and upper 32 bits for VNC1 */ + AMDGPU_VEGA20_DOORBELL64_VCN2_3 = 0x189, + AMDGPU_VEGA20_DOORBELL64_VCN4_5 = 0x18A, + AMDGPU_VEGA20_DOORBELL64_VCN6_7 = 0x18B, + + AMDGPU_VEGA20_DOORBELL64_UVD_RING0_1 = 0x188, + AMDGPU_VEGA20_DOORBELL64_UVD_RING2_3 = 0x189, + AMDGPU_VEGA20_DOORBELL64_UVD_RING4_5 = 0x18A, + AMDGPU_VEGA20_DOORBELL64_UVD_RING6_7 = 0x18B, + + AMDGPU_VEGA20_DOORBELL64_VCE_RING0_1 = 0x18C, + AMDGPU_VEGA20_DOORBELL64_VCE_RING2_3 = 0x18D, + AMDGPU_VEGA20_DOORBELL64_VCE_RING4_5 = 0x18E, + AMDGPU_VEGA20_DOORBELL64_VCE_RING6_7 = 0x18F, + AMDGPU_VEGA20_DOORBELL_MAX_ASSIGNMENT = 0x18F, + AMDGPU_VEGA20_DOORBELL_INVALID = 0xFFFF +} AMDGPU_VEGA20_DOORBELL_ASSIGNMENT; + +/* + * 64bit doorbell, offset are in QWORD, occupy 2KB doorbell space + */ +typedef enum _AMDGPU_DOORBELL64_ASSIGNMENT +{ + /* + * All compute related doorbells: kiq, hiq, diq, traditional compute queue, user queue, should locate in + * a continues range so that programming CP_MEC_DOORBELL_RANGE_LOWER/UPPER can cover this range. + * Compute related doorbells are allocated from 0x00 to 0x8a + */ + + + /* kernel scheduling */ + AMDGPU_DOORBELL64_KIQ = 0x00, + + /* HSA interface queue and debug queue */ + AMDGPU_DOORBELL64_HIQ = 0x01, + AMDGPU_DOORBELL64_DIQ = 0x02, + + /* Compute engines */ + AMDGPU_DOORBELL64_MEC_RING0 = 0x03, + AMDGPU_DOORBELL64_MEC_RING1 = 0x04, + AMDGPU_DOORBELL64_MEC_RING2 = 0x05, + AMDGPU_DOORBELL64_MEC_RING3 = 0x06, + AMDGPU_DOORBELL64_MEC_RING4 = 0x07, + AMDGPU_DOORBELL64_MEC_RING5 = 0x08, + AMDGPU_DOORBELL64_MEC_RING6 = 0x09, + AMDGPU_DOORBELL64_MEC_RING7 = 0x0a, + + /* User queue doorbell range (128 doorbells) */ + AMDGPU_DOORBELL64_USERQUEUE_START = 0x0b, + AMDGPU_DOORBELL64_USERQUEUE_END = 0x8a, + + /* Graphics engine */ + AMDGPU_DOORBELL64_GFX_RING0 = 0x8b, + + /* + * Other graphics doorbells can be allocated here: from 0x8c to 0xdf + * Graphics voltage island aperture 1 + * default non-graphics QWORD index is 0xe0 - 0xFF inclusive + */ + + /* For vega10 sriov, the sdma doorbell must be fixed as follow + * to keep the same setting with host driver, or it will + * happen conflicts + */ + AMDGPU_DOORBELL64_sDMA_ENGINE0 = 0xF0, + AMDGPU_DOORBELL64_sDMA_HI_PRI_ENGINE0 = 0xF1, + AMDGPU_DOORBELL64_sDMA_ENGINE1 = 0xF2, + AMDGPU_DOORBELL64_sDMA_HI_PRI_ENGINE1 = 0xF3, + + /* Interrupt handler */ + AMDGPU_DOORBELL64_IH = 0xF4, /* For legacy interrupt ring buffer */ + AMDGPU_DOORBELL64_IH_RING1 = 0xF5, /* For page migration request log */ + AMDGPU_DOORBELL64_IH_RING2 = 0xF6, /* For page migration translation/invalidation log */ + + /* VCN engine use 32 bits doorbell */ + AMDGPU_DOORBELL64_VCN0_1 = 0xF8, /* lower 32 bits for VNC0 and upper 32 bits for VNC1 */ + AMDGPU_DOORBELL64_VCN2_3 = 0xF9, + AMDGPU_DOORBELL64_VCN4_5 = 0xFA, + AMDGPU_DOORBELL64_VCN6_7 = 0xFB, + + /* overlap the doorbell assignment with VCN as they are mutually exclusive + * VCE engine's doorbell is 32 bit and two VCE ring share one QWORD + */ + AMDGPU_DOORBELL64_UVD_RING0_1 = 0xF8, + AMDGPU_DOORBELL64_UVD_RING2_3 = 0xF9, + AMDGPU_DOORBELL64_UVD_RING4_5 = 0xFA, + AMDGPU_DOORBELL64_UVD_RING6_7 = 0xFB, + + AMDGPU_DOORBELL64_VCE_RING0_1 = 0xFC, + AMDGPU_DOORBELL64_VCE_RING2_3 = 0xFD, + AMDGPU_DOORBELL64_VCE_RING4_5 = 0xFE, + AMDGPU_DOORBELL64_VCE_RING6_7 = 0xFF, + + AMDGPU_DOORBELL64_MAX_ASSIGNMENT = 0xFF, + AMDGPU_DOORBELL64_INVALID = 0xFFFF +} AMDGPU_DOORBELL64_ASSIGNMENT; + +u32 amdgpu_mm_rdoorbell(struct amdgpu_device *adev, u32 index); +void amdgpu_mm_wdoorbell(struct amdgpu_device *adev, u32 index, u32 v); +u64 amdgpu_mm_rdoorbell64(struct amdgpu_device *adev, u32 index); +void amdgpu_mm_wdoorbell64(struct amdgpu_device *adev, u32 index, u64 v); + +#define RDOORBELL32(index) amdgpu_mm_rdoorbell(adev, (index)) +#define WDOORBELL32(index, v) amdgpu_mm_wdoorbell(adev, (index), (v)) +#define RDOORBELL64(index) amdgpu_mm_rdoorbell64(adev, (index)) +#define WDOORBELL64(index, v) amdgpu_mm_wdoorbell64(adev, (index), (v)) + diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 8de55f7f1a3a..c806f984bcc5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -454,9 +454,10 @@ module_param_named(cntl_sb_buf_per_se, amdgpu_cntl_sb_buf_per_se, int, 0444); /** * DOC: param_buf_per_se (int) - * Override the size of Off-Chip Pramater Cache per Shader Engine in Byte. The default is 0 (depending on gfx). + * Override the size of Off-Chip Parameter Cache per Shader Engine in Byte. + * The default is 0 (depending on gfx). */ -MODULE_PARM_DESC(param_buf_per_se, "the size of Off-Chip Pramater Cache per Shader Engine (default depending on gfx)"); +MODULE_PARM_DESC(param_buf_per_se, "the size of Off-Chip Parameter Cache per Shader Engine (default depending on gfx)"); module_param_named(param_buf_per_se, amdgpu_param_buf_per_se, int, 0444); /** @@ -864,6 +865,7 @@ static const struct pci_device_id pciidlist[] = { /* VEGAM */ {0x1002, 0x694C, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGAM}, {0x1002, 0x694E, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGAM}, + {0x1002, 0x694F, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGAM}, /* Vega 10 */ {0x1002, 0x6860, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, {0x1002, 0x6861, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, @@ -872,7 +874,13 @@ static const struct pci_device_id pciidlist[] = { {0x1002, 0x6864, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, {0x1002, 0x6867, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, {0x1002, 0x6868, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, + {0x1002, 0x6869, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, + {0x1002, 0x686a, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, + {0x1002, 0x686b, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, {0x1002, 0x686c, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, + {0x1002, 0x686d, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, + {0x1002, 0x686e, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, + {0x1002, 0x686f, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, {0x1002, 0x687f, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA10}, /* Vega 12 */ {0x1002, 0x69A0, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA12}, @@ -885,6 +893,7 @@ static const struct pci_device_id pciidlist[] = { {0x1002, 0x66A1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20}, {0x1002, 0x66A2, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20}, {0x1002, 0x66A3, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20}, + {0x1002, 0x66A4, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20}, {0x1002, 0x66A7, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20}, {0x1002, 0x66AF, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_VEGA20}, /* Raven */ @@ -1220,9 +1229,6 @@ static struct drm_driver kms_driver = { .patchlevel = KMS_DRIVER_PATCHLEVEL, }; -static struct drm_driver *driver; -static struct pci_driver *pdriver; - static struct pci_driver amdgpu_kms_pci_driver = { .name = DRIVER_NAME, .id_table = pciidlist, @@ -1252,16 +1258,14 @@ static int __init amdgpu_init(void) goto error_fence; DRM_INFO("amdgpu kernel modesetting enabled.\n"); - driver = &kms_driver; - pdriver = &amdgpu_kms_pci_driver; - driver->num_ioctls = amdgpu_max_kms_ioctl; + kms_driver.num_ioctls = amdgpu_max_kms_ioctl; amdgpu_register_atpx_handler(); /* Ignore KFD init failures. Normal when CONFIG_HSA_AMD is not set. */ amdgpu_amdkfd_init(); /* let modprobe override vga console setting */ - return pci_register_driver(pdriver); + return pci_register_driver(&amdgpu_kms_pci_driver); error_fence: amdgpu_sync_fini(); @@ -1273,7 +1277,7 @@ error_sync: static void __exit amdgpu_exit(void) { amdgpu_amdkfd_fini(); - pci_unregister_driver(pdriver); + pci_unregister_driver(&amdgpu_kms_pci_driver); amdgpu_unregister_atpx_handler(); amdgpu_sync_fini(); amdgpu_fence_slab_fini(); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 7b3d1ebda9df..f4f00217546e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -169,7 +169,7 @@ void amdgpu_gem_object_close(struct drm_gem_object *obj, INIT_LIST_HEAD(&duplicates); tv.bo = &bo->tbo; - tv.shared = true; + tv.num_shared = 1; list_add(&tv.head, &list); amdgpu_vm_get_pd_bo(vm, &list, &vm_pd); @@ -604,7 +604,10 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data, return -ENOENT; abo = gem_to_amdgpu_bo(gobj); tv.bo = &abo->tbo; - tv.shared = !!(abo->flags & AMDGPU_GEM_CREATE_VM_ALWAYS_VALID); + if (abo->flags & AMDGPU_GEM_CREATE_VM_ALWAYS_VALID) + tv.num_shared = 1; + else + tv.num_shared = 0; list_add(&tv.head, &list); } else { gobj = NULL; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h index d63daba9b17c..f1ddfc50bcc7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.h @@ -54,6 +54,8 @@ void *amdgpu_gem_prime_vmap(struct drm_gem_object *obj); void amdgpu_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr); int amdgpu_gem_prime_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma); +extern const struct dma_buf_ops amdgpu_dmabuf_ops; + /* * GEM objects. */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c index 6a70c0b7105f..97a60da62004 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c @@ -250,7 +250,7 @@ int amdgpu_gfx_kiq_init_ring(struct amdgpu_device *adev, ring->adev = NULL; ring->ring_obj = NULL; ring->use_doorbell = true; - ring->doorbell_index = AMDGPU_DOORBELL_KIQ; + ring->doorbell_index = adev->doorbell_index.kiq; r = amdgpu_gfx_kiq_acquire(adev, ring); if (r) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h index 8c57924c075f..81e6070d255b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h @@ -99,6 +99,7 @@ struct amdgpu_xgmi { unsigned num_physical_nodes; /* gpu list in the same hive */ struct list_head head; + bool supported; }; struct amdgpu_gmc { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h index 9ce8c93ec19b..f877bb78d10a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h @@ -51,14 +51,12 @@ struct amdgpu_ih_ring { struct amdgpu_ih_funcs { /* ring read/write ptr handling, called from interrupt context */ u32 (*get_wptr)(struct amdgpu_device *adev); - bool (*prescreen_iv)(struct amdgpu_device *adev); void (*decode_iv)(struct amdgpu_device *adev, struct amdgpu_iv_entry *entry); void (*set_rptr)(struct amdgpu_device *adev); }; #define amdgpu_ih_get_wptr(adev) (adev)->irq.ih_funcs->get_wptr((adev)) -#define amdgpu_ih_prescreen_iv(adev) (adev)->irq.ih_funcs->prescreen_iv((adev)) #define amdgpu_ih_decode_iv(adev, iv) (adev)->irq.ih_funcs->decode_iv((adev), (iv)) #define amdgpu_ih_set_rptr(adev) (adev)->irq.ih_funcs->set_rptr((adev)) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c index 6b6524f04ce0..b7968f426862 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c @@ -145,13 +145,6 @@ static void amdgpu_irq_callback(struct amdgpu_device *adev, u32 ring_index = ih->rptr >> 2; struct amdgpu_iv_entry entry; - /* Prescreening of high-frequency interrupts */ - if (!amdgpu_ih_prescreen_iv(adev)) - return; - - /* Before dispatching irq to IP blocks, send it to amdkfd */ - amdgpu_amdkfd_interrupt(adev, (const void *) &ih->ring[ring_index]); - entry.iv_entry = (const uint32_t *)&ih->ring[ring_index]; amdgpu_ih_decode_iv(adev, &entry); @@ -371,39 +364,38 @@ void amdgpu_irq_dispatch(struct amdgpu_device *adev, unsigned client_id = entry->client_id; unsigned src_id = entry->src_id; struct amdgpu_irq_src *src; + bool handled = false; int r; trace_amdgpu_iv(entry); if (client_id >= AMDGPU_IRQ_CLIENTID_MAX) { DRM_DEBUG("Invalid client_id in IV: %d\n", client_id); - return; - } - if (src_id >= AMDGPU_MAX_IRQ_SRC_ID) { + } else if (src_id >= AMDGPU_MAX_IRQ_SRC_ID) { DRM_DEBUG("Invalid src_id in IV: %d\n", src_id); - return; - } - if (adev->irq.virq[src_id]) { + } else if (adev->irq.virq[src_id]) { generic_handle_irq(irq_find_mapping(adev->irq.domain, src_id)); - } else { - if (!adev->irq.client[client_id].sources) { - DRM_DEBUG("Unregistered interrupt client_id: %d src_id: %d\n", - client_id, src_id); - return; - } - src = adev->irq.client[client_id].sources[src_id]; - if (!src) { - DRM_DEBUG("Unhandled interrupt src_id: %d\n", src_id); - return; - } + } else if (!adev->irq.client[client_id].sources) { + DRM_DEBUG("Unregistered interrupt client_id: %d src_id: %d\n", + client_id, src_id); + } else if ((src = adev->irq.client[client_id].sources[src_id])) { r = src->funcs->process(adev, src, entry); - if (r) + if (r < 0) DRM_ERROR("error processing interrupt (%d)\n", r); + else if (r) + handled = true; + + } else { + DRM_DEBUG("Unhandled interrupt src_id: %d\n", src_id); } + + /* Send it to amdkfd as well if it isn't already handled */ + if (!handled) + amdgpu_amdkfd_interrupt(adev, entry->iv_entry); } /** diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c index e0af44fd6a0c..0a17fb1af204 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c @@ -32,6 +32,9 @@ static void amdgpu_job_timedout(struct drm_sched_job *s_job) { struct amdgpu_ring *ring = to_amdgpu_ring(s_job->sched); struct amdgpu_job *job = to_amdgpu_job(s_job); + struct amdgpu_task_info ti; + + memset(&ti, 0, sizeof(struct amdgpu_task_info)); if (amdgpu_ring_soft_recovery(ring, job->vmid, s_job->s_fence->parent)) { DRM_ERROR("ring %s timeout, but soft recovered\n", @@ -39,9 +42,12 @@ static void amdgpu_job_timedout(struct drm_sched_job *s_job) return; } + amdgpu_vm_get_task_info(ring->adev, job->pasid, &ti); DRM_ERROR("ring %s timeout, signaled seq=%u, emitted seq=%u\n", job->base.sched->name, atomic_read(&ring->fence_drv.last_seq), ring->fence_drv.sync_seq); + DRM_ERROR("Process information: process %s pid %d thread %s pid %d\n", + ti.process_name, ti.tgid, ti.task_name, ti.pid); if (amdgpu_device_should_recover_gpu(ring->adev)) amdgpu_device_gpu_recover(ring->adev, job); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c index 9b3164c0f861..bc62bf41b7e9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c @@ -467,9 +467,6 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file if (!info->return_size || !info->return_pointer) return -EINVAL; - /* Ensure IB tests are run on ring */ - flush_delayed_work(&adev->late_init_work); - switch (info->query) { case AMDGPU_INFO_ACCEL_WORKING: ui32 = adev->accel_working; @@ -950,6 +947,9 @@ int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv) struct amdgpu_fpriv *fpriv; int r, pasid; + /* Ensure IB tests are run on ring */ + flush_delayed_work(&adev->late_init_work); + file_priv->driver_priv = NULL; r = pm_runtime_get_sync(dev->dev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c index e55508b39496..3e6823fdd939 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c @@ -238,44 +238,40 @@ static void amdgpu_mn_invalidate_node(struct amdgpu_mn_node *node, * amdgpu_mn_invalidate_range_start_gfx - callback to notify about mm change * * @mn: our notifier - * @mm: the mm this callback is about - * @start: start of updated range - * @end: end of updated range + * @range: mmu notifier context * * Block for operations on BOs to finish and mark pages as accessed and * potentially dirty. */ static int amdgpu_mn_invalidate_range_start_gfx(struct mmu_notifier *mn, - struct mm_struct *mm, - unsigned long start, - unsigned long end, - bool blockable) + const struct mmu_notifier_range *range) { struct amdgpu_mn *amn = container_of(mn, struct amdgpu_mn, mn); struct interval_tree_node *it; + unsigned long end; /* notification is exclusive, but interval is inclusive */ - end -= 1; + end = range->end - 1; /* TODO we should be able to split locking for interval tree and * amdgpu_mn_invalidate_node */ - if (amdgpu_mn_read_lock(amn, blockable)) + if (amdgpu_mn_read_lock(amn, range->blockable)) return -EAGAIN; - it = interval_tree_iter_first(&amn->objects, start, end); + it = interval_tree_iter_first(&amn->objects, range->start, end); while (it) { struct amdgpu_mn_node *node; - if (!blockable) { + if (!range->blockable) { amdgpu_mn_read_unlock(amn); return -EAGAIN; } node = container_of(it, struct amdgpu_mn_node, it); - it = interval_tree_iter_next(it, start, end); + it = interval_tree_iter_next(it, range->start, end); - amdgpu_mn_invalidate_node(node, start, end); + amdgpu_mn_invalidate_node(node, range->start, end); } return 0; @@ -294,39 +290,38 @@ static int amdgpu_mn_invalidate_range_start_gfx(struct mmu_notifier *mn, * are restorted in amdgpu_mn_invalidate_range_end_hsa. */ static int amdgpu_mn_invalidate_range_start_hsa(struct mmu_notifier *mn, - struct mm_struct *mm, - unsigned long start, - unsigned long end, - bool blockable) + const struct mmu_notifier_range *range) { struct amdgpu_mn *amn = container_of(mn, struct amdgpu_mn, mn); struct interval_tree_node *it; + unsigned long end; /* notification is exclusive, but interval is inclusive */ - end -= 1; + end = range->end - 1; - if (amdgpu_mn_read_lock(amn, blockable)) + if (amdgpu_mn_read_lock(amn, range->blockable)) return -EAGAIN; - it = interval_tree_iter_first(&amn->objects, start, end); + it = interval_tree_iter_first(&amn->objects, range->start, end); while (it) { struct amdgpu_mn_node *node; struct amdgpu_bo *bo; - if (!blockable) { + if (!range->blockable) { amdgpu_mn_read_unlock(amn); return -EAGAIN; } node = container_of(it, struct amdgpu_mn_node, it); - it = interval_tree_iter_next(it, start, end); + it = interval_tree_iter_next(it, range->start, end); list_for_each_entry(bo, &node->bos, mn_list) { struct kgd_mem *mem = bo->kfd_bo; if (amdgpu_ttm_tt_affect_userptr(bo->tbo.ttm, - start, end)) - amdgpu_amdkfd_evict_userptr(mem, mm); + range->start, + end)) + amdgpu_amdkfd_evict_userptr(mem, range->mm); } } @@ -344,9 +339,7 @@ static int amdgpu_mn_invalidate_range_start_hsa(struct mmu_notifier *mn, * Release the lock again to allow new command submissions. */ static void amdgpu_mn_invalidate_range_end(struct mmu_notifier *mn, - struct mm_struct *mm, - unsigned long start, - unsigned long end) + const struct mmu_notifier_range *range) { struct amdgpu_mn *amn = container_of(mn, struct amdgpu_mn, mn); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h index 0dc2c5c57015..aadd0fa42e43 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h @@ -38,7 +38,6 @@ #include <drm/drm_crtc_helper.h> #include <drm/drm_fb_helper.h> #include <drm/drm_plane_helper.h> -#include <drm/drm_fb_helper.h> #include <linux/i2c.h> #include <linux/i2c-algo-bit.h> #include <linux/hrtimer.h> @@ -294,13 +293,6 @@ struct amdgpu_display_funcs { uint16_t connector_object_id, struct amdgpu_hpd *hpd, struct amdgpu_router *router); - /* it is used to enter or exit into free sync mode */ - int (*notify_freesync)(struct drm_device *dev, void *data, - struct drm_file *filp); - /* it is used to allow enablement of freesync mode */ - int (*set_freesync_property)(struct drm_connector *connector, - struct drm_property *property, - uint64_t val); }; @@ -340,6 +332,8 @@ struct amdgpu_mode_info { struct drm_property *dither_property; /* maximum number of bits per channel for monitor color */ struct drm_property *max_bpc_property; + /* Adaptive Backlight Modulation (power feature) */ + struct drm_property *abm_level_property; /* hardcoded DFP edid from BIOS */ struct edid *bios_hardcoded_edid; int bios_hardcoded_edid_size; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c index cf768acb51dc..728e15e5d68a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c @@ -81,7 +81,7 @@ static void amdgpu_bo_destroy(struct ttm_buffer_object *tbo) amdgpu_bo_subtract_pin_size(bo); if (bo->kfd_bo) - amdgpu_amdkfd_unreserve_system_memory_limit(bo); + amdgpu_amdkfd_unreserve_memory_limit(bo); amdgpu_bo_kunmap(bo); @@ -608,53 +608,6 @@ int amdgpu_bo_create(struct amdgpu_device *adev, } /** - * amdgpu_bo_backup_to_shadow - Backs up an &amdgpu_bo buffer object - * @adev: amdgpu device object - * @ring: amdgpu_ring for the engine handling the buffer operations - * @bo: &amdgpu_bo buffer to be backed up - * @resv: reservation object with embedded fence - * @fence: dma_fence associated with the operation - * @direct: whether to submit the job directly - * - * Copies an &amdgpu_bo buffer object to its shadow object. - * Not used for now. - * - * Returns: - * 0 for success or a negative error code on failure. - */ -int amdgpu_bo_backup_to_shadow(struct amdgpu_device *adev, - struct amdgpu_ring *ring, - struct amdgpu_bo *bo, - struct reservation_object *resv, - struct dma_fence **fence, - bool direct) - -{ - struct amdgpu_bo *shadow = bo->shadow; - uint64_t bo_addr, shadow_addr; - int r; - - if (!shadow) - return -EINVAL; - - bo_addr = amdgpu_bo_gpu_offset(bo); - shadow_addr = amdgpu_bo_gpu_offset(bo->shadow); - - r = reservation_object_reserve_shared(bo->tbo.resv, 1); - if (r) - goto err; - - r = amdgpu_copy_buffer(ring, bo_addr, shadow_addr, - amdgpu_bo_size(bo), resv, fence, - direct, false); - if (!r) - amdgpu_bo_fence(bo, *fence, true); - -err: - return r; -} - -/** * amdgpu_bo_validate - validate an &amdgpu_bo buffer object * @bo: pointer to the buffer object * @@ -959,7 +912,7 @@ int amdgpu_bo_unpin(struct amdgpu_bo *bo) struct ttm_operation_ctx ctx = { false, false }; int r, i; - if (!bo->pin_count) { + if (WARN_ON_ONCE(!bo->pin_count)) { dev_warn(adev->dev, "%p unpin not necessary\n", bo); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h index 7d3312d0da11..9291c2f837e9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.h @@ -267,11 +267,6 @@ int amdgpu_bo_fault_reserve_notify(struct ttm_buffer_object *bo); void amdgpu_bo_fence(struct amdgpu_bo *bo, struct dma_fence *fence, bool shared); u64 amdgpu_bo_gpu_offset(struct amdgpu_bo *bo); -int amdgpu_bo_backup_to_shadow(struct amdgpu_device *adev, - struct amdgpu_ring *ring, - struct amdgpu_bo *bo, - struct reservation_object *resv, - struct dma_fence **fence, bool direct); int amdgpu_bo_validate(struct amdgpu_bo *bo); int amdgpu_bo_restore_shadow(struct amdgpu_bo *shadow, struct dma_fence **fence); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c index 7235cd0b0fa9..1f61ed95727c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c @@ -33,6 +33,8 @@ #include <linux/hwmon.h> #include <linux/hwmon-sysfs.h> #include <linux/nospec.h> +#include "hwmgr.h" +#define WIDTH_4K 3840 static int amdgpu_debugfs_pm_init(struct amdgpu_device *adev); @@ -1642,6 +1644,19 @@ static umode_t hwmon_attributes_visible(struct kobject *kobj, attr == &sensor_dev_attr_fan1_enable.dev_attr.attr)) return 0; + /* Skip fan attributes on APU */ + if ((adev->flags & AMD_IS_APU) && + (attr == &sensor_dev_attr_pwm1.dev_attr.attr || + attr == &sensor_dev_attr_pwm1_enable.dev_attr.attr || + attr == &sensor_dev_attr_pwm1_max.dev_attr.attr || + attr == &sensor_dev_attr_pwm1_min.dev_attr.attr || + attr == &sensor_dev_attr_fan1_input.dev_attr.attr || + attr == &sensor_dev_attr_fan1_min.dev_attr.attr || + attr == &sensor_dev_attr_fan1_max.dev_attr.attr || + attr == &sensor_dev_attr_fan1_target.dev_attr.attr || + attr == &sensor_dev_attr_fan1_enable.dev_attr.attr)) + return 0; + /* Skip limit attributes if DPM is not enabled */ if (!adev->pm.dpm_enabled && (attr == &sensor_dev_attr_temp1_crit.dev_attr.attr || @@ -1956,6 +1971,17 @@ void amdgpu_dpm_enable_uvd(struct amdgpu_device *adev, bool enable) amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_UVD, !enable); mutex_unlock(&adev->pm.mutex); } + /* enable/disable Low Memory PState for UVD (4k videos) */ + if (adev->asic_type == CHIP_STONEY && + adev->uvd.decode_image_width >= WIDTH_4K) { + struct pp_hwmgr *hwmgr = adev->powerplay.pp_handle; + + if (hwmgr && hwmgr->hwmgr_func && + hwmgr->hwmgr_func->update_nbdpm_pstate) + hwmgr->hwmgr_func->update_nbdpm_pstate(hwmgr, + !enable, + true); + } } void amdgpu_dpm_enable_vce(struct amdgpu_device *adev, bool enable) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c index 3e44d889f7af..71913a18d142 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_prime.c @@ -39,8 +39,6 @@ #include <drm/amdgpu_drm.h> #include <linux/dma-buf.h> -static const struct dma_buf_ops amdgpu_dmabuf_ops; - /** * amdgpu_gem_prime_get_sg_table - &drm_driver.gem_prime_get_sg_table * implementation @@ -332,7 +330,7 @@ static int amdgpu_gem_begin_cpu_access(struct dma_buf *dma_buf, return ret; } -static const struct dma_buf_ops amdgpu_dmabuf_ops = { +const struct dma_buf_ops amdgpu_dmabuf_ops = { .attach = amdgpu_gem_map_attach, .detach = amdgpu_gem_map_detach, .map_dma_buf = drm_gem_map_dma_buf, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index e05dc66b1090..8fab0d637ee5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -155,10 +155,14 @@ psp_cmd_submit_buf(struct psp_context *psp, return ret; } -static void psp_prep_tmr_cmd_buf(struct psp_gfx_cmd_resp *cmd, +static void psp_prep_tmr_cmd_buf(struct psp_context *psp, + struct psp_gfx_cmd_resp *cmd, uint64_t tmr_mc, uint32_t size) { - cmd->cmd_id = GFX_CMD_ID_SETUP_TMR; + if (psp_support_vmr_ring(psp)) + cmd->cmd_id = GFX_CMD_ID_SETUP_VMR; + else + cmd->cmd_id = GFX_CMD_ID_SETUP_TMR; cmd->cmd.cmd_setup_tmr.buf_phy_addr_lo = lower_32_bits(tmr_mc); cmd->cmd.cmd_setup_tmr.buf_phy_addr_hi = upper_32_bits(tmr_mc); cmd->cmd.cmd_setup_tmr.buf_size = size; @@ -192,7 +196,7 @@ static int psp_tmr_load(struct psp_context *psp) if (!cmd) return -ENOMEM; - psp_prep_tmr_cmd_buf(cmd, psp->tmr_mc_addr, PSP_TMR_SIZE); + psp_prep_tmr_cmd_buf(psp, cmd, psp->tmr_mc_addr, PSP_TMR_SIZE); DRM_INFO("reserve 0x%x from 0x%llx for PSP TMR SIZE\n", PSP_TMR_SIZE, psp->tmr_mc_addr); @@ -536,8 +540,10 @@ static int psp_load_fw(struct amdgpu_device *adev) int ret; struct psp_context *psp = &adev->psp; - if (amdgpu_sriov_vf(adev) && adev->in_gpu_reset != 0) + if (amdgpu_sriov_vf(adev) && adev->in_gpu_reset) { + psp_ring_destroy(psp, PSP_RING_TYPE__KM); goto skip_memalloc; + } psp->cmd = kzalloc(sizeof(struct psp_gfx_cmd_resp), GFP_KERNEL); if (!psp->cmd) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h index 9ec5d1a666a6..3ee573b4016e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h @@ -83,12 +83,13 @@ struct psp_funcs enum AMDGPU_UCODE_ID ucode_type); bool (*smu_reload_quirk)(struct psp_context *psp); int (*mode1_reset)(struct psp_context *psp); - uint64_t (*xgmi_get_node_id)(struct psp_context *psp); - uint64_t (*xgmi_get_hive_id)(struct psp_context *psp); + int (*xgmi_get_node_id)(struct psp_context *psp, uint64_t *node_id); + int (*xgmi_get_hive_id)(struct psp_context *psp, uint64_t *hive_id); int (*xgmi_get_topology_info)(struct psp_context *psp, int number_devices, struct psp_xgmi_topology_info *topology); int (*xgmi_set_topology_info)(struct psp_context *psp, int number_devices, struct psp_xgmi_topology_info *topology); + bool (*support_vmr_ring)(struct psp_context *psp); }; struct psp_xgmi_context { @@ -192,12 +193,14 @@ struct psp_xgmi_topology_info { ((psp)->funcs->bootloader_load_sos ? (psp)->funcs->bootloader_load_sos((psp)) : 0) #define psp_smu_reload_quirk(psp) \ ((psp)->funcs->smu_reload_quirk ? (psp)->funcs->smu_reload_quirk((psp)) : false) +#define psp_support_vmr_ring(psp) \ + ((psp)->funcs->support_vmr_ring ? (psp)->funcs->support_vmr_ring((psp)) : false) #define psp_mode1_reset(psp) \ ((psp)->funcs->mode1_reset ? (psp)->funcs->mode1_reset((psp)) : false) -#define psp_xgmi_get_node_id(psp) \ - ((psp)->funcs->xgmi_get_node_id ? (psp)->funcs->xgmi_get_node_id((psp)) : 0) -#define psp_xgmi_get_hive_id(psp) \ - ((psp)->funcs->xgmi_get_hive_id ? (psp)->funcs->xgmi_get_hive_id((psp)) : 0) +#define psp_xgmi_get_node_id(psp, node_id) \ + ((psp)->funcs->xgmi_get_node_id ? (psp)->funcs->xgmi_get_node_id((psp), (node_id)) : -EINVAL) +#define psp_xgmi_get_hive_id(psp, hive_id) \ + ((psp)->funcs->xgmi_get_hive_id ? (psp)->funcs->xgmi_get_hive_id((psp), (hive_id)) : -EINVAL) #define psp_xgmi_get_topology_info(psp, num_device, topology) \ ((psp)->funcs->xgmi_get_topology_info ? \ (psp)->funcs->xgmi_get_topology_info((psp), (num_device), (topology)) : -EINVAL) @@ -217,7 +220,6 @@ extern const struct amdgpu_ip_block_version psp_v10_0_ip_block; int psp_gpu_reset(struct amdgpu_device *adev); int psp_xgmi_invoke(struct psp_context *psp, uint32_t ta_cmd_id); - extern const struct amdgpu_ip_block_version psp_v11_0_ip_block; #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c index 5b75bdc8dc28..335a0edf114b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c @@ -397,7 +397,7 @@ bool amdgpu_ring_soft_recovery(struct amdgpu_ring *ring, unsigned int vmid, { ktime_t deadline = ktime_add_us(ktime_get(), 10000); - if (!ring->funcs->soft_recovery) + if (!ring->funcs->soft_recovery || !fence) return false; atomic_inc(&ring->adev->gpu_reset_counter); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h index 0beb01fef83f..d87e828a084b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h @@ -29,7 +29,7 @@ #include <drm/drm_print.h> /* max number of rings */ -#define AMDGPU_MAX_RINGS 21 +#define AMDGPU_MAX_RINGS 23 #define AMDGPU_MAX_GFX_RINGS 1 #define AMDGPU_MAX_COMPUTE_RINGS 8 #define AMDGPU_MAX_VCE_RINGS 3 diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c index 69896f451e8a..4e5d13e41f6a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c @@ -692,6 +692,8 @@ static int amdgpu_uvd_cs_msg_decode(struct amdgpu_device *adev, uint32_t *msg, buf_sizes[0x1] = dpb_size; buf_sizes[0x2] = image_size; buf_sizes[0x4] = min_ctx_size; + /* store image width to adjust nb memory pstate */ + adev->uvd.decode_image_width = width; return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.h index a3ab1a41060f..5eb63288d157 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.h @@ -65,6 +65,8 @@ struct amdgpu_uvd { struct drm_sched_entity entity; struct delayed_work idle_work; unsigned harvest_config; + /* store image width to adjust nb memory state */ + unsigned decode_image_width; }; int amdgpu_uvd_sw_init(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c index e2e42e3fbcf3..ecf6f96df2ad 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c @@ -262,7 +262,7 @@ static int amdgpu_vcn_pause_dpg_mode(struct amdgpu_device *adev, ring = &adev->vcn.ring_dec; WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_WPTR, - RREG32_SOC15(UVD, 0, mmUVD_SCRATCH2)); + RREG32_SOC15(UVD, 0, mmUVD_SCRATCH2) & 0x7FFFFFFF); SOC15_WAIT_ON_RREG(UVD, 0, mmUVD_POWER_STATUS, UVD_PGFSM_CONFIG__UVDM_UVDU_PWR_ON, UVD_POWER_STATUS__UVD_POWER_STATUS_MASK, ret_code); @@ -322,7 +322,7 @@ static int amdgpu_vcn_pause_dpg_mode(struct amdgpu_device *adev, ring = &adev->vcn.ring_dec; WREG32_SOC15(UVD, 0, mmUVD_RBC_RB_WPTR, - RREG32_SOC15(UVD, 0, mmUVD_SCRATCH2)); + RREG32_SOC15(UVD, 0, mmUVD_SCRATCH2) & 0x7FFFFFFF); SOC15_WAIT_ON_RREG(UVD, 0, mmUVD_POWER_STATUS, UVD_PGFSM_CONFIG__UVDM_UVDU_PWR_ON, UVD_POWER_STATUS__UVD_POWER_STATUS_MASK, ret_code); @@ -396,16 +396,26 @@ void amdgpu_vcn_ring_begin_use(struct amdgpu_ring *ring) if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) { struct dpg_pause_state new_state; + unsigned int fences = 0; + unsigned int i; - if (ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC) + for (i = 0; i < adev->vcn.num_enc_rings; ++i) { + fences += amdgpu_fence_count_emitted(&adev->vcn.ring_enc[i]); + } + if (fences) new_state.fw_based = VCN_DPG_STATE__PAUSE; else - new_state.fw_based = adev->vcn.pause_state.fw_based; + new_state.fw_based = VCN_DPG_STATE__UNPAUSE; - if (ring->funcs->type == AMDGPU_RING_TYPE_VCN_JPEG) + if (amdgpu_fence_count_emitted(&adev->vcn.ring_jpeg)) new_state.jpeg = VCN_DPG_STATE__PAUSE; else - new_state.jpeg = adev->vcn.pause_state.jpeg; + new_state.jpeg = VCN_DPG_STATE__UNPAUSE; + + if (ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC) + new_state.fw_based = VCN_DPG_STATE__PAUSE; + else if (ring->funcs->type == AMDGPU_RING_TYPE_VCN_JPEG) + new_state.jpeg = VCN_DPG_STATE__PAUSE; amdgpu_vcn_pause_dpg_mode(adev, &new_state); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c index cfee74732edb..462a04e0f5e6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c @@ -334,7 +334,7 @@ void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev) if (adev->fw_vram_usage.va != NULL) { adev->virt.fw_reserve.p_pf2vf = - (struct amdgim_pf2vf_info_header *)( + (struct amd_sriov_msg_pf2vf_info_header *)( adev->fw_vram_usage.va + AMDGIM_DATAEXCHANGE_OFFSET); AMDGPU_FW_VRAM_PF2VF_READ(adev, header.size, &pf2vf_size); AMDGPU_FW_VRAM_PF2VF_READ(adev, checksum, &checksum); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h index 0728fbc9a692..722deefc0a7e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h @@ -63,8 +63,8 @@ struct amdgpu_virt_ops { * Firmware Reserve Frame buffer */ struct amdgpu_virt_fw_reserve { - struct amdgim_pf2vf_info_header *p_pf2vf; - struct amdgim_vf2pf_info_header *p_vf2pf; + struct amd_sriov_msg_pf2vf_info_header *p_pf2vf; + struct amd_sriov_msg_vf2pf_info_header *p_vf2pf; unsigned int checksum_key; }; /* @@ -85,15 +85,17 @@ enum AMDGIM_FEATURE_FLAG { AMDGIM_FEATURE_GIM_FLR_VRAMLOST = 0x4, }; -struct amdgim_pf2vf_info_header { +struct amd_sriov_msg_pf2vf_info_header { /* the total structure size in byte. */ uint32_t size; /* version of this structure, written by the GIM */ uint32_t version; + /* reserved */ + uint32_t reserved[2]; } __aligned(4); struct amdgim_pf2vf_info_v1 { /* header contains size and version */ - struct amdgim_pf2vf_info_header header; + struct amd_sriov_msg_pf2vf_info_header header; /* max_width * max_height */ unsigned int uvd_enc_max_pixels_count; /* 16x16 pixels/sec, codec independent */ @@ -112,7 +114,7 @@ struct amdgim_pf2vf_info_v1 { struct amdgim_pf2vf_info_v2 { /* header contains size and version */ - struct amdgim_pf2vf_info_header header; + struct amd_sriov_msg_pf2vf_info_header header; /* use private key from mailbox 2 to create chueksum */ uint32_t checksum; /* The features flags of the GIM driver supports. */ @@ -137,20 +139,22 @@ struct amdgim_pf2vf_info_v2 { uint64_t vcefw_kboffset; /* VCE FW size in KB */ uint32_t vcefw_ksize; - uint32_t reserved[AMDGIM_GET_STRUCTURE_RESERVED_SIZE(256, 0, 0, (9 + sizeof(struct amdgim_pf2vf_info_header)/sizeof(uint32_t)), 3)]; + uint32_t reserved[AMDGIM_GET_STRUCTURE_RESERVED_SIZE(256, 0, 0, (9 + sizeof(struct amd_sriov_msg_pf2vf_info_header)/sizeof(uint32_t)), 3)]; } __aligned(4); -struct amdgim_vf2pf_info_header { +struct amd_sriov_msg_vf2pf_info_header { /* the total structure size in byte. */ uint32_t size; /*version of this structure, written by the guest */ uint32_t version; + /* reserved */ + uint32_t reserved[2]; } __aligned(4); struct amdgim_vf2pf_info_v1 { /* header contains size and version */ - struct amdgim_vf2pf_info_header header; + struct amd_sriov_msg_vf2pf_info_header header; /* driver version */ char driver_version[64]; /* driver certification, 1=WHQL, 0=None */ @@ -180,7 +184,7 @@ struct amdgim_vf2pf_info_v1 { struct amdgim_vf2pf_info_v2 { /* header contains size and version */ - struct amdgim_vf2pf_info_header header; + struct amd_sriov_msg_vf2pf_info_header header; uint32_t checksum; /* driver version */ uint8_t driver_version[64]; @@ -206,7 +210,7 @@ struct amdgim_vf2pf_info_v2 { uint32_t uvd_enc_usage; /* guest uvd engine usage percentage. 0xffff means N/A. */ uint32_t uvd_enc_health; - uint32_t reserved[AMDGIM_GET_STRUCTURE_RESERVED_SIZE(256, 64, 0, (12 + sizeof(struct amdgim_vf2pf_info_header)/sizeof(uint32_t)), 0)]; + uint32_t reserved[AMDGIM_GET_STRUCTURE_RESERVED_SIZE(256, 64, 0, (12 + sizeof(struct amd_sriov_msg_vf2pf_info_header)/sizeof(uint32_t)), 0)]; } __aligned(4); #define AMDGPU_FW_VRAM_VF2PF_VER 2 diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index 58a2363040dd..e73d152659a2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -181,7 +181,7 @@ static unsigned amdgpu_vm_num_entries(struct amdgpu_device *adev, if (level == adev->vm_manager.root_level) /* For the root directory */ - return round_up(adev->vm_manager.max_pfn, 1 << shift) >> shift; + return round_up(adev->vm_manager.max_pfn, 1ULL << shift) >> shift; else if (level != AMDGPU_VM_PTB) /* Everything in between */ return 512; @@ -617,7 +617,8 @@ void amdgpu_vm_get_pd_bo(struct amdgpu_vm *vm, { entry->priority = 0; entry->tv.bo = &vm->root.base.bo->tbo; - entry->tv.shared = true; + /* One for the VM updates, one for TTM and one for the CS job */ + entry->tv.num_shared = 3; entry->user_pages = NULL; list_add(&entry->tv.head, validated); } @@ -773,10 +774,6 @@ static int amdgpu_vm_clear_bo(struct amdgpu_device *adev, ring = container_of(vm->entity.rq->sched, struct amdgpu_ring, sched); - r = reservation_object_reserve_shared(bo->tbo.resv, 1); - if (r) - return r; - r = ttm_bo_validate(&bo->tbo, &bo->placement, &ctx); if (r) goto error; @@ -1656,9 +1653,11 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params, if (!amdgpu_vm_pt_descendant(adev, &cursor)) return -ENOENT; continue; - } else if (frag >= parent_shift) { + } else if (frag >= parent_shift && + cursor.level - 1 != adev->vm_manager.root_level) { /* If the fragment size is even larger than the parent - * shift we should go up one level and check it again. + * shift we should go up one level and check it again + * unless one level up is the root level. */ if (!amdgpu_vm_pt_ancestor(&cursor)) return -ENOENT; @@ -1666,10 +1665,10 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params, } /* Looks good so far, calculate parameters for the update */ - incr = AMDGPU_GPU_PAGE_SIZE << shift; + incr = (uint64_t)AMDGPU_GPU_PAGE_SIZE << shift; mask = amdgpu_vm_entries_mask(adev, cursor.level); pe_start = ((cursor.pfn >> shift) & mask) * 8; - entry_end = (mask + 1) << shift; + entry_end = (uint64_t)(mask + 1) << shift; entry_end += cursor.pfn & ~(entry_end - 1); entry_end = min(entry_end, end); @@ -1682,7 +1681,7 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params, flags | AMDGPU_PTE_FRAG(frag)); pe_start += nptes * 8; - dst += nptes * AMDGPU_GPU_PAGE_SIZE << shift; + dst += (uint64_t)nptes * AMDGPU_GPU_PAGE_SIZE << shift; frag_start = upd_end; if (frag_start >= frag_end) { @@ -1842,10 +1841,6 @@ static int amdgpu_vm_bo_update_mapping(struct amdgpu_device *adev, if (r) goto error_free; - r = reservation_object_reserve_shared(vm->root.base.bo->tbo.resv, 1); - if (r) - goto error_free; - r = amdgpu_vm_update_ptes(¶ms, start, last + 1, addr, flags); if (r) goto error_free; @@ -3026,6 +3021,10 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, if (r) goto error_free_root; + r = reservation_object_reserve_shared(root->tbo.resv, 1); + if (r) + goto error_unreserve; + r = amdgpu_vm_clear_bo(adev, vm, root, adev->vm_manager.root_level, vm->pte_support_ats); @@ -3055,7 +3054,6 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, } INIT_KFIFO(vm->faults); - vm->fault_credit = 16; return 0; @@ -3268,42 +3266,6 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm) } /** - * amdgpu_vm_pasid_fault_credit - Check fault credit for given PASID - * - * @adev: amdgpu_device pointer - * @pasid: PASID do identify the VM - * - * This function is expected to be called in interrupt context. - * - * Returns: - * True if there was fault credit, false otherwise - */ -bool amdgpu_vm_pasid_fault_credit(struct amdgpu_device *adev, - unsigned int pasid) -{ - struct amdgpu_vm *vm; - - spin_lock(&adev->vm_manager.pasid_lock); - vm = idr_find(&adev->vm_manager.pasid_idr, pasid); - if (!vm) { - /* VM not found, can't track fault credit */ - spin_unlock(&adev->vm_manager.pasid_lock); - return true; - } - - /* No lock needed. only accessed by IRQ handler */ - if (!vm->fault_credit) { - /* Too many faults in this VM */ - spin_unlock(&adev->vm_manager.pasid_lock); - return false; - } - - vm->fault_credit--; - spin_unlock(&adev->vm_manager.pasid_lock); - return true; -} - -/** * amdgpu_vm_manager_init - init the VM manager * * @adev: amdgpu_device pointer diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index 2a8898d19c8b..e8dcfd59fc93 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -229,9 +229,6 @@ struct amdgpu_vm { /* Up to 128 pending retry page faults */ DECLARE_KFIFO(faults, u64, 128); - /* Limit non-retry fault storms */ - unsigned int fault_credit; - /* Points to the KFD process VM info */ struct amdkfd_process_info *process_info; @@ -299,8 +296,6 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm, unsigned int pasid); void amdgpu_vm_release_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm); void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm); -bool amdgpu_vm_pasid_fault_credit(struct amdgpu_device *adev, - unsigned int pasid); void amdgpu_vm_get_pd_bo(struct amdgpu_vm *vm, struct list_head *validated, struct amdgpu_bo_list_entry *entry); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index 909216a9b447..8a8bc60cb6b4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -23,7 +23,7 @@ */ #include <linux/list.h> #include "amdgpu.h" -#include "amdgpu_psp.h" +#include "amdgpu_xgmi.h" static DEFINE_MUTEX(xgmi_mutex); @@ -31,15 +31,16 @@ static DEFINE_MUTEX(xgmi_mutex); #define AMDGPU_MAX_XGMI_HIVE 8 #define AMDGPU_MAX_XGMI_DEVICE_PER_HIVE 4 -struct amdgpu_hive_info { - uint64_t hive_id; - struct list_head device_list; -}; - static struct amdgpu_hive_info xgmi_hives[AMDGPU_MAX_XGMI_HIVE]; static unsigned hive_count = 0; -static struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev) + +void *amdgpu_xgmi_hive_try_lock(struct amdgpu_hive_info *hive) +{ + return &hive->device_list; +} + +struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev) { int i; struct amdgpu_hive_info *tmp; @@ -58,39 +59,73 @@ static struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev) tmp = &xgmi_hives[hive_count++]; tmp->hive_id = adev->gmc.xgmi.hive_id; INIT_LIST_HEAD(&tmp->device_list); + mutex_init(&tmp->hive_lock); + return tmp; } +int amdgpu_xgmi_update_topology(struct amdgpu_hive_info *hive, struct amdgpu_device *adev) +{ + int ret = -EINVAL; + + /* Each psp need to set the latest topology */ + ret = psp_xgmi_set_topology_info(&adev->psp, + hive->number_devices, + &hive->topology_info); + if (ret) + dev_err(adev->dev, + "XGMI: Set topology failure on device %llx, hive %llx, ret %d", + adev->gmc.xgmi.node_id, + adev->gmc.xgmi.hive_id, ret); + else + dev_info(adev->dev, "XGMI: Set topology for node %d, hive 0x%llx.\n", + adev->gmc.xgmi.physical_node_id, + adev->gmc.xgmi.hive_id); + + return ret; +} + int amdgpu_xgmi_add_device(struct amdgpu_device *adev) { - struct psp_xgmi_topology_info *tmp_topology; + struct psp_xgmi_topology_info *hive_topology; struct amdgpu_hive_info *hive; struct amdgpu_xgmi *entry; - struct amdgpu_device *tmp_adev; + struct amdgpu_device *tmp_adev = NULL; int count = 0, ret = -EINVAL; - if ((adev->asic_type < CHIP_VEGA20) || - (adev->flags & AMD_IS_APU) ) + if (!adev->gmc.xgmi.supported) return 0; - adev->gmc.xgmi.node_id = psp_xgmi_get_node_id(&adev->psp); - adev->gmc.xgmi.hive_id = psp_xgmi_get_hive_id(&adev->psp); - tmp_topology = kzalloc(sizeof(struct psp_xgmi_topology_info), GFP_KERNEL); - if (!tmp_topology) - return -ENOMEM; + ret = psp_xgmi_get_node_id(&adev->psp, &adev->gmc.xgmi.node_id); + if (ret) { + dev_err(adev->dev, + "XGMI: Failed to get node id\n"); + return ret; + } + + ret = psp_xgmi_get_hive_id(&adev->psp, &adev->gmc.xgmi.hive_id); + if (ret) { + dev_err(adev->dev, + "XGMI: Failed to get hive id\n"); + return ret; + } + mutex_lock(&xgmi_mutex); hive = amdgpu_get_xgmi_hive(adev); if (!hive) goto exit; + hive_topology = &hive->topology_info; + list_add_tail(&adev->gmc.xgmi.head, &hive->device_list); list_for_each_entry(entry, &hive->device_list, head) - tmp_topology->nodes[count++].node_id = entry->node_id; + hive_topology->nodes[count++].node_id = entry->node_id; + hive->number_devices = count; /* Each psp need to get the latest topology */ list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) { - ret = psp_xgmi_get_topology_info(&tmp_adev->psp, count, tmp_topology); + ret = psp_xgmi_get_topology_info(&tmp_adev->psp, count, hive_topology); if (ret) { dev_err(tmp_adev->dev, "XGMI: Get topology failure on device %llx, hive %llx, ret %d", @@ -101,25 +136,33 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev) } } - /* Each psp need to set the latest topology */ list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) { - ret = psp_xgmi_set_topology_info(&tmp_adev->psp, count, tmp_topology); - if (ret) { - dev_err(tmp_adev->dev, - "XGMI: Set topology failure on device %llx, hive %llx, ret %d", - tmp_adev->gmc.xgmi.node_id, - tmp_adev->gmc.xgmi.hive_id, ret); - /* To do : continue with some node failed or disable the whole hive */ + ret = amdgpu_xgmi_update_topology(hive, tmp_adev); + if (ret) break; - } } - if (!ret) - dev_info(adev->dev, "XGMI: Add node %d to hive 0x%llx.\n", - adev->gmc.xgmi.physical_node_id, - adev->gmc.xgmi.hive_id); exit: mutex_unlock(&xgmi_mutex); - kfree(tmp_topology); return ret; } + +void amdgpu_xgmi_remove_device(struct amdgpu_device *adev) +{ + struct amdgpu_hive_info *hive; + + if (!adev->gmc.xgmi.supported) + return; + + mutex_lock(&xgmi_mutex); + + hive = amdgpu_get_xgmi_hive(adev); + if (!hive) + goto exit; + + if (!(hive->number_devices--)) + mutex_destroy(&hive->hive_lock); + +exit: + mutex_unlock(&xgmi_mutex); +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h new file mode 100644 index 000000000000..6151eb9c8ad3 --- /dev/null +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.h @@ -0,0 +1,40 @@ +/* + * Copyright 2016 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#ifndef __AMDGPU_XGMI_H__ +#define __AMDGPU_XGMI_H__ + +#include "amdgpu_psp.h" + +struct amdgpu_hive_info { + uint64_t hive_id; + struct list_head device_list; + struct psp_xgmi_topology_info topology_info; + int number_devices; + struct mutex hive_lock; +}; + +struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev); +int amdgpu_xgmi_update_topology(struct amdgpu_hive_info *hive, struct amdgpu_device *adev); +int amdgpu_xgmi_add_device(struct amdgpu_device *adev); +void amdgpu_xgmi_remove_device(struct amdgpu_device *adev); + +#endif diff --git a/drivers/gpu/drm/amd/amdgpu/cik.c b/drivers/gpu/drm/amd/amdgpu/cik.c index f41f5f57e9f3..71c50d8900e3 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik.c +++ b/drivers/gpu/drm/amd/amdgpu/cik.c @@ -1755,6 +1755,7 @@ static const struct amdgpu_asic_funcs cik_asic_funcs = .flush_hdp = &cik_flush_hdp, .invalidate_hdp = &cik_invalidate_hdp, .need_full_reset = &cik_need_full_reset, + .init_doorbell_index = &legacy_doorbell_index_init, }; static int cik_common_early_init(void *handle) diff --git a/drivers/gpu/drm/amd/amdgpu/cik.h b/drivers/gpu/drm/amd/amdgpu/cik.h index e49c6f15a0a0..54c625a2e570 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik.h +++ b/drivers/gpu/drm/amd/amdgpu/cik.h @@ -30,4 +30,5 @@ void cik_srbm_select(struct amdgpu_device *adev, u32 me, u32 pipe, u32 queue, u32 vmid); int cik_set_ip_blocks(struct amdgpu_device *adev); +void legacy_doorbell_index_init(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/cik_ih.c b/drivers/gpu/drm/amd/amdgpu/cik_ih.c index b5775c6a857b..8a8b4967a101 100644 --- a/drivers/gpu/drm/amd/amdgpu/cik_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/cik_ih.c @@ -228,34 +228,6 @@ static u32 cik_ih_get_wptr(struct amdgpu_device *adev) * [127:96] - reserved */ -/** - * cik_ih_prescreen_iv - prescreen an interrupt vector - * - * @adev: amdgpu_device pointer - * - * Returns true if the interrupt vector should be further processed. - */ -static bool cik_ih_prescreen_iv(struct amdgpu_device *adev) -{ - u32 ring_index = adev->irq.ih.rptr >> 2; - u16 pasid; - - switch (le32_to_cpu(adev->irq.ih.ring[ring_index]) & 0xff) { - case 146: - case 147: - pasid = le32_to_cpu(adev->irq.ih.ring[ring_index + 2]) >> 16; - if (!pasid || amdgpu_vm_pasid_fault_credit(adev, pasid)) - return true; - break; - default: - /* Not a VM fault */ - return true; - } - - adev->irq.ih.rptr += 16; - return false; -} - /** * cik_ih_decode_iv - decode an interrupt vector * @@ -461,7 +433,6 @@ static const struct amd_ip_funcs cik_ih_ip_funcs = { static const struct amdgpu_ih_funcs cik_ih_funcs = { .get_wptr = cik_ih_get_wptr, - .prescreen_iv = cik_ih_prescreen_iv, .decode_iv = cik_ih_decode_iv, .set_rptr = cik_ih_set_rptr }; diff --git a/drivers/gpu/drm/amd/amdgpu/cz_ih.c b/drivers/gpu/drm/amd/amdgpu/cz_ih.c index df5ac4d85a00..9d3ea298e116 100644 --- a/drivers/gpu/drm/amd/amdgpu/cz_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/cz_ih.c @@ -208,34 +208,6 @@ static u32 cz_ih_get_wptr(struct amdgpu_device *adev) } /** - * cz_ih_prescreen_iv - prescreen an interrupt vector - * - * @adev: amdgpu_device pointer - * - * Returns true if the interrupt vector should be further processed. - */ -static bool cz_ih_prescreen_iv(struct amdgpu_device *adev) -{ - u32 ring_index = adev->irq.ih.rptr >> 2; - u16 pasid; - - switch (le32_to_cpu(adev->irq.ih.ring[ring_index]) & 0xff) { - case 146: - case 147: - pasid = le32_to_cpu(adev->irq.ih.ring[ring_index + 2]) >> 16; - if (!pasid || amdgpu_vm_pasid_fault_credit(adev, pasid)) - return true; - break; - default: - /* Not a VM fault */ - return true; - } - - adev->irq.ih.rptr += 16; - return false; -} - -/** * cz_ih_decode_iv - decode an interrupt vector * * @adev: amdgpu_device pointer @@ -442,7 +414,6 @@ static const struct amd_ip_funcs cz_ih_ip_funcs = { static const struct amdgpu_ih_funcs cz_ih_funcs = { .get_wptr = cz_ih_get_wptr, - .prescreen_iv = cz_ih_prescreen_iv, .decode_iv = cz_ih_decode_iv, .set_rptr = cz_ih_set_rptr }; diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c index f467b9bd090d..3a9fb6018c16 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c @@ -4363,7 +4363,7 @@ static int gfx_v7_0_compute_ring_init(struct amdgpu_device *adev, int ring_id, ring->ring_obj = NULL; ring->use_doorbell = true; - ring->doorbell_index = AMDGPU_DOORBELL_MEC_RING0 + ring_id; + ring->doorbell_index = adev->doorbell_index.mec_ring0 + ring_id; sprintf(ring->name, "comp_%d.%d.%d", ring->me, ring->pipe, ring->queue); irq_type = AMDGPU_CP_IRQ_COMPUTE_MEC1_PIPE0_EOP diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c index cb066a8dccd7..381f593b0cda 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c @@ -44,7 +44,6 @@ #include "gca/gfx_8_0_d.h" #include "gca/gfx_8_0_enum.h" #include "gca/gfx_8_0_sh_mask.h" -#include "gca/gfx_8_0_enum.h" #include "dce/dce_10_0_d.h" #include "dce/dce_10_0_sh_mask.h" @@ -1891,7 +1890,7 @@ static int gfx_v8_0_compute_ring_init(struct amdgpu_device *adev, int ring_id, ring->ring_obj = NULL; ring->use_doorbell = true; - ring->doorbell_index = AMDGPU_DOORBELL_MEC_RING0 + ring_id; + ring->doorbell_index = adev->doorbell_index.mec_ring0 + ring_id; ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr + (ring_id * GFX8_MEC_HPD_SIZE); sprintf(ring->name, "comp_%d.%d.%d", ring->me, ring->pipe, ring->queue); @@ -2002,7 +2001,7 @@ static int gfx_v8_0_sw_init(void *handle) /* no gfx doorbells on iceland */ if (adev->asic_type != CHIP_TOPAZ) { ring->use_doorbell = true; - ring->doorbell_index = AMDGPU_DOORBELL_GFX_RING0; + ring->doorbell_index = adev->doorbell_index.gfx_ring0; } r = amdgpu_ring_init(adev, ring, 1024, &adev->gfx.eop_irq, @@ -4069,6 +4068,11 @@ static void gfx_v8_0_rlc_start(struct amdgpu_device *adev) static int gfx_v8_0_rlc_resume(struct amdgpu_device *adev) { + if (amdgpu_sriov_vf(adev)) { + gfx_v8_0_init_csb(adev); + return 0; + } + adev->gfx.rlc.funcs->stop(adev); adev->gfx.rlc.funcs->reset(adev); gfx_v8_0_init_pg(adev); @@ -4216,7 +4220,7 @@ static void gfx_v8_0_set_cpg_door_bell(struct amdgpu_device *adev, struct amdgpu tmp = REG_SET_FIELD(0, CP_RB_DOORBELL_RANGE_LOWER, DOORBELL_RANGE_LOWER, - AMDGPU_DOORBELL_GFX_RING0); + adev->doorbell_index.gfx_ring0); WREG32(mmCP_RB_DOORBELL_RANGE_LOWER, tmp); WREG32(mmCP_RB_DOORBELL_RANGE_UPPER, @@ -4645,8 +4649,8 @@ static int gfx_v8_0_kcq_init_queue(struct amdgpu_ring *ring) static void gfx_v8_0_set_mec_doorbell_range(struct amdgpu_device *adev) { if (adev->asic_type > CHIP_TONGA) { - WREG32(mmCP_MEC_DOORBELL_RANGE_LOWER, AMDGPU_DOORBELL_KIQ << 2); - WREG32(mmCP_MEC_DOORBELL_RANGE_UPPER, AMDGPU_DOORBELL_MEC_RING7 << 2); + WREG32(mmCP_MEC_DOORBELL_RANGE_LOWER, adev->doorbell_index.kiq << 2); + WREG32(mmCP_MEC_DOORBELL_RANGE_UPPER, adev->doorbell_index.mec_ring7 << 2); } /* enable doorbells */ WREG32_FIELD(CP_PQ_STATUS, DOORBELL_ENABLE, 1); @@ -4948,14 +4952,13 @@ static bool gfx_v8_0_check_soft_reset(void *handle) static int gfx_v8_0_pre_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - u32 grbm_soft_reset = 0, srbm_soft_reset = 0; + u32 grbm_soft_reset = 0; if ((!adev->gfx.grbm_soft_reset) && (!adev->gfx.srbm_soft_reset)) return 0; grbm_soft_reset = adev->gfx.grbm_soft_reset; - srbm_soft_reset = adev->gfx.srbm_soft_reset; /* stop the rlc */ adev->gfx.rlc.funcs->stop(adev); @@ -5052,14 +5055,13 @@ static int gfx_v8_0_soft_reset(void *handle) static int gfx_v8_0_post_soft_reset(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; - u32 grbm_soft_reset = 0, srbm_soft_reset = 0; + u32 grbm_soft_reset = 0; if ((!adev->gfx.grbm_soft_reset) && (!adev->gfx.srbm_soft_reset)) return 0; grbm_soft_reset = adev->gfx.grbm_soft_reset; - srbm_soft_reset = adev->gfx.srbm_soft_reset; if (REG_GET_FIELD(grbm_soft_reset, GRBM_SOFT_RESET, SOFT_RESET_CP) || REG_GET_FIELD(grbm_soft_reset, GRBM_SOFT_RESET, SOFT_RESET_CPF) || diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index c27caa144c57..7556716038d3 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -86,6 +86,7 @@ MODULE_FIRMWARE("amdgpu/picasso_me.bin"); MODULE_FIRMWARE("amdgpu/picasso_mec.bin"); MODULE_FIRMWARE("amdgpu/picasso_mec2.bin"); MODULE_FIRMWARE("amdgpu/picasso_rlc.bin"); +MODULE_FIRMWARE("amdgpu/picasso_rlc_am4.bin"); MODULE_FIRMWARE("amdgpu/raven2_ce.bin"); MODULE_FIRMWARE("amdgpu/raven2_pfp.bin"); @@ -645,7 +646,20 @@ static int gfx_v9_0_init_microcode(struct amdgpu_device *adev) adev->gfx.ce_fw_version = le32_to_cpu(cp_hdr->header.ucode_version); adev->gfx.ce_feature_version = le32_to_cpu(cp_hdr->ucode_feature_version); - snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_rlc.bin", chip_name); + /* + * For Picasso && AM4 SOCKET board, we use picasso_rlc_am4.bin + * instead of picasso_rlc.bin. + * Judgment method: + * PCO AM4: revision >= 0xC8 && revision <= 0xCF + * or revision >= 0xD8 && revision <= 0xDF + * otherwise is PCO FP5 + */ + if (!strcmp(chip_name, "picasso") && + (((adev->pdev->revision >= 0xC8) && (adev->pdev->revision <= 0xCF)) || + ((adev->pdev->revision >= 0xD8) && (adev->pdev->revision <= 0xDF)))) + snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_rlc_am4.bin", chip_name); + else + snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_rlc.bin", chip_name); err = request_firmware(&adev->gfx.rlc_fw, fw_name, adev->dev); if (err) goto out; @@ -1566,7 +1580,7 @@ static int gfx_v9_0_compute_ring_init(struct amdgpu_device *adev, int ring_id, ring->ring_obj = NULL; ring->use_doorbell = true; - ring->doorbell_index = (AMDGPU_DOORBELL_MEC_RING0 + ring_id) << 1; + ring->doorbell_index = (adev->doorbell_index.mec_ring0 + ring_id) << 1; ring->eop_gpu_addr = adev->gfx.mec.hpd_eop_gpu_addr + (ring_id * GFX9_MEC_HPD_SIZE); sprintf(ring->name, "comp_%d.%d.%d", ring->me, ring->pipe, ring->queue); @@ -1655,7 +1669,7 @@ static int gfx_v9_0_sw_init(void *handle) else sprintf(ring->name, "gfx_%d", i); ring->use_doorbell = true; - ring->doorbell_index = AMDGPU_DOORBELL64_GFX_RING0 << 1; + ring->doorbell_index = adev->doorbell_index.gfx_ring0 << 1; r = amdgpu_ring_init(adev, ring, 1024, &adev->gfx.eop_irq, AMDGPU_CP_IRQ_GFX_EOP); if (r) @@ -2326,12 +2340,13 @@ static void gfx_v9_0_rlc_start(struct amdgpu_device *adev) #endif WREG32_FIELD15(GC, 0, RLC_CNTL, RLC_ENABLE_F32, 1); + udelay(50); /* carrizo do enable cp interrupt after cp inited */ - if (!(adev->flags & AMD_IS_APU)) + if (!(adev->flags & AMD_IS_APU)) { gfx_v9_0_enable_gui_idle_interrupt(adev, true); - - udelay(50); + udelay(50); + } #ifdef AMDGPU_RLC_DEBUG_RETRY /* RLC_GPM_GENERAL_6 : RLC Ucode version */ @@ -2981,9 +2996,9 @@ static int gfx_v9_0_kiq_init_register(struct amdgpu_ring *ring) /* enable the doorbell if requested */ if (ring->use_doorbell) { WREG32_SOC15(GC, 0, mmCP_MEC_DOORBELL_RANGE_LOWER, - (AMDGPU_DOORBELL64_KIQ *2) << 2); + (adev->doorbell_index.kiq * 2) << 2); WREG32_SOC15(GC, 0, mmCP_MEC_DOORBELL_RANGE_UPPER, - (AMDGPU_DOORBELL64_USERQUEUE_END * 2) << 2); + (adev->doorbell_index.userqueue_end * 2) << 2); } WREG32_SOC15(GC, 0, mmCP_HQD_PQ_DOORBELL_CONTROL, diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index 531aaf377592..1ad7e6b8ed1d 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -56,6 +56,9 @@ MODULE_FIRMWARE("amdgpu/tonga_mc.bin"); MODULE_FIRMWARE("amdgpu/polaris11_mc.bin"); MODULE_FIRMWARE("amdgpu/polaris10_mc.bin"); MODULE_FIRMWARE("amdgpu/polaris12_mc.bin"); +MODULE_FIRMWARE("amdgpu/polaris11_k_mc.bin"); +MODULE_FIRMWARE("amdgpu/polaris10_k_mc.bin"); +MODULE_FIRMWARE("amdgpu/polaris12_k_mc.bin"); static const u32 golden_settings_tonga_a11[] = { @@ -224,13 +227,39 @@ static int gmc_v8_0_init_microcode(struct amdgpu_device *adev) chip_name = "tonga"; break; case CHIP_POLARIS11: - chip_name = "polaris11"; + if (((adev->pdev->device == 0x67ef) && + ((adev->pdev->revision == 0xe0) || + (adev->pdev->revision == 0xe5))) || + ((adev->pdev->device == 0x67ff) && + ((adev->pdev->revision == 0xcf) || + (adev->pdev->revision == 0xef) || + (adev->pdev->revision == 0xff)))) + chip_name = "polaris11_k"; + else if ((adev->pdev->device == 0x67ef) && + (adev->pdev->revision == 0xe2)) + chip_name = "polaris11_k"; + else + chip_name = "polaris11"; break; case CHIP_POLARIS10: - chip_name = "polaris10"; + if ((adev->pdev->device == 0x67df) && + ((adev->pdev->revision == 0xe1) || + (adev->pdev->revision == 0xf7))) + chip_name = "polaris10_k"; + else + chip_name = "polaris10"; break; case CHIP_POLARIS12: - chip_name = "polaris12"; + if (((adev->pdev->device == 0x6987) && + ((adev->pdev->revision == 0xc0) || + (adev->pdev->revision == 0xc3))) || + ((adev->pdev->device == 0x6981) && + ((adev->pdev->revision == 0x00) || + (adev->pdev->revision == 0x01) || + (adev->pdev->revision == 0x10)))) + chip_name = "polaris12_k"; + else + chip_name = "polaris12"; break; case CHIP_FIJI: case CHIP_CARRIZO: @@ -337,7 +366,7 @@ static int gmc_v8_0_polaris_mc_load_microcode(struct amdgpu_device *adev) const struct mc_firmware_header_v1_0 *hdr; const __le32 *fw_data = NULL; const __le32 *io_mc_regs = NULL; - u32 data, vbios_version; + u32 data; int i, ucode_size, regs_size; /* Skip MC ucode loading on SR-IOV capable boards. @@ -348,13 +377,6 @@ static int gmc_v8_0_polaris_mc_load_microcode(struct amdgpu_device *adev) if (amdgpu_sriov_bios(adev)) return 0; - WREG32(mmMC_SEQ_IO_DEBUG_INDEX, 0x9F); - data = RREG32(mmMC_SEQ_IO_DEBUG_DATA); - vbios_version = data & 0xf; - - if (vbios_version == 0) - return 0; - if (!adev->gmc.fw) return -EINVAL; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index 811231e4ec53..bacdaef77b6c 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -244,6 +244,62 @@ static int gmc_v9_0_vm_fault_interrupt_state(struct amdgpu_device *adev, return 0; } +/** + * vega10_ih_prescreen_iv - prescreen an interrupt vector + * + * @adev: amdgpu_device pointer + * + * Returns true if the interrupt vector should be further processed. + */ +static bool gmc_v9_0_prescreen_iv(struct amdgpu_device *adev, + struct amdgpu_iv_entry *entry, + uint64_t addr) +{ + struct amdgpu_vm *vm; + u64 key; + int r; + + /* No PASID, can't identify faulting process */ + if (!entry->pasid) + return true; + + /* Not a retry fault */ + if (!(entry->src_data[1] & 0x80)) + return true; + + /* Track retry faults in per-VM fault FIFO. */ + spin_lock(&adev->vm_manager.pasid_lock); + vm = idr_find(&adev->vm_manager.pasid_idr, entry->pasid); + if (!vm) { + /* VM not found, process it normally */ + spin_unlock(&adev->vm_manager.pasid_lock); + return true; + } + + key = AMDGPU_VM_FAULT(entry->pasid, addr); + r = amdgpu_vm_add_fault(vm->fault_hash, key); + + /* Hash table is full or the fault is already being processed, + * ignore further page faults + */ + if (r != 0) { + spin_unlock(&adev->vm_manager.pasid_lock); + return false; + } + /* No locking required with single writer and single reader */ + r = kfifo_put(&vm->faults, key); + if (!r) { + /* FIFO is full. Ignore it until there is space */ + amdgpu_vm_clear_fault(vm->fault_hash, key); + spin_unlock(&adev->vm_manager.pasid_lock); + return false; + } + + spin_unlock(&adev->vm_manager.pasid_lock); + /* It's the first fault for this address, process it normally */ + return true; +} + static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, struct amdgpu_irq_src *source, struct amdgpu_iv_entry *entry) @@ -255,6 +311,9 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, addr = (u64)entry->src_data[0] << 12; addr |= ((u64)entry->src_data[1] & 0xf) << 44; + if (!gmc_v9_0_prescreen_iv(adev, entry, addr)) + return 1; /* This also prevents sending it to KFD */ + if (!amdgpu_sriov_vf(adev)) { status = RREG32(hub->vm_l2_pro_fault_status); WREG32_P(hub->vm_l2_pro_fault_cntl, 1, ~1); @@ -338,9 +397,12 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, struct amdgpu_vmhub *hub = &adev->vmhub[i]; u32 tmp = gmc_v9_0_get_invalidate_req(vmid, flush_type); - if (i == AMDGPU_GFXHUB && !adev->in_gpu_reset && - adev->gfx.kiq.ring.sched.ready && - (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev))) { + /* This is necessary for a HW workaround under SRIOV as well + * as GFXOFF under bare metal + */ + if (adev->gfx.kiq.ring.sched.ready && + (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev)) && + !adev->in_gpu_reset) { uint32_t req = hub->vm_inv_eng0_req + eng; uint32_t ack = hub->vm_inv_eng0_ack + eng; @@ -656,37 +718,46 @@ static bool gmc_v9_0_keep_stolen_memory(struct amdgpu_device *adev) } } -static int gmc_v9_0_late_init(void *handle) +static int gmc_v9_0_allocate_vm_inv_eng(struct amdgpu_device *adev) { - struct amdgpu_device *adev = (struct amdgpu_device *)handle; - /* - * The latest engine allocation on gfx9 is: - * Engine 0, 1: idle - * Engine 2, 3: firmware - * Engine 4~13: amdgpu ring, subject to change when ring number changes - * Engine 14~15: idle - * Engine 16: kfd tlb invalidation - * Engine 17: Gart flushes - */ - unsigned vm_inv_eng[AMDGPU_MAX_VMHUBS] = { 4, 4 }; + struct amdgpu_ring *ring; + unsigned vm_inv_engs[AMDGPU_MAX_VMHUBS] = + {GFXHUB_FREE_VM_INV_ENGS_BITMAP, MMHUB_FREE_VM_INV_ENGS_BITMAP}; unsigned i; - int r; + unsigned vmhub, inv_eng; - if (!gmc_v9_0_keep_stolen_memory(adev)) - amdgpu_bo_late_init(adev); + for (i = 0; i < adev->num_rings; ++i) { + ring = adev->rings[i]; + vmhub = ring->funcs->vmhub; + + inv_eng = ffs(vm_inv_engs[vmhub]); + if (!inv_eng) { + dev_err(adev->dev, "no VM inv eng for ring %s\n", + ring->name); + return -EINVAL; + } - for(i = 0; i < adev->num_rings; ++i) { - struct amdgpu_ring *ring = adev->rings[i]; - unsigned vmhub = ring->funcs->vmhub; + ring->vm_inv_eng = inv_eng - 1; + change_bit(inv_eng - 1, (unsigned long *)(&vm_inv_engs[vmhub])); - ring->vm_inv_eng = vm_inv_eng[vmhub]++; dev_info(adev->dev, "ring %s uses VM inv eng %u on hub %u\n", ring->name, ring->vm_inv_eng, ring->funcs->vmhub); } - /* Engine 16 is used for KFD and 17 for GART flushes */ - for(i = 0; i < AMDGPU_MAX_VMHUBS; ++i) - BUG_ON(vm_inv_eng[i] > 16); + return 0; +} + +static int gmc_v9_0_late_init(void *handle) +{ + struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int r; + + if (!gmc_v9_0_keep_stolen_memory(adev)) + amdgpu_bo_late_init(adev); + + r = gmc_v9_0_allocate_vm_inv_eng(adev); + if (r) + return r; if (adev->asic_type == CHIP_VEGA10 && !amdgpu_sriov_vf(adev)) { r = gmc_v9_0_ecc_available(adev); @@ -899,6 +970,9 @@ static int gmc_v9_0_sw_init(void *handle) /* This interrupt is VMC page fault.*/ r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_VMC, VMC_1_0__SRCID__VM_FAULT, &adev->gmc.vm_fault); + if (r) + return r; + r = amdgpu_irq_add_id(adev, SOC15_IH_CLIENTID_UTCL2, UTCL2_1_0__SRCID__FAULT, &adev->gmc.vm_fault); @@ -931,7 +1005,7 @@ static int gmc_v9_0_sw_init(void *handle) } adev->need_swiotlb = drm_get_max_iomem() > ((u64)1 << dma_bits); - if (adev->asic_type == CHIP_VEGA20) { + if (adev->gmc.xgmi.supported) { r = gfxhub_v1_1_get_xgmi_info(adev); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.h b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.h index b030ca5ea107..5c8deac65580 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.h +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.h @@ -24,6 +24,16 @@ #ifndef __GMC_V9_0_H__ #define __GMC_V9_0_H__ + /* + * The latest engine allocation on gfx9 is: + * Engine 2, 3: firmware + * Engine 0, 1, 4~16: amdgpu ring, + * subject to change when ring number changes + * Engine 17: Gart flushes + */ +#define GFXHUB_FREE_VM_INV_ENGS_BITMAP 0x1FFF3 +#define MMHUB_FREE_VM_INV_ENGS_BITMAP 0x1FFF3 + extern const struct amd_ip_funcs gmc_v9_0_ip_funcs; extern const struct amdgpu_ip_block_version gmc_v9_0_ip_block; diff --git a/drivers/gpu/drm/amd/amdgpu/iceland_ih.c b/drivers/gpu/drm/amd/amdgpu/iceland_ih.c index cf0fc61aebe6..a3984d10b604 100644 --- a/drivers/gpu/drm/amd/amdgpu/iceland_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/iceland_ih.c @@ -208,34 +208,6 @@ static u32 iceland_ih_get_wptr(struct amdgpu_device *adev) } /** - * iceland_ih_prescreen_iv - prescreen an interrupt vector - * - * @adev: amdgpu_device pointer - * - * Returns true if the interrupt vector should be further processed. - */ -static bool iceland_ih_prescreen_iv(struct amdgpu_device *adev) -{ - u32 ring_index = adev->irq.ih.rptr >> 2; - u16 pasid; - - switch (le32_to_cpu(adev->irq.ih.ring[ring_index]) & 0xff) { - case 146: - case 147: - pasid = le32_to_cpu(adev->irq.ih.ring[ring_index + 2]) >> 16; - if (!pasid || amdgpu_vm_pasid_fault_credit(adev, pasid)) - return true; - break; - default: - /* Not a VM fault */ - return true; - } - - adev->irq.ih.rptr += 16; - return false; -} - -/** * iceland_ih_decode_iv - decode an interrupt vector * * @adev: amdgpu_device pointer @@ -440,7 +412,6 @@ static const struct amd_ip_funcs iceland_ih_ip_funcs = { static const struct amdgpu_ih_funcs iceland_ih_funcs = { .get_wptr = iceland_ih_get_wptr, - .prescreen_iv = iceland_ih_prescreen_iv, .decode_iv = iceland_ih_decode_iv, .set_rptr = iceland_ih_set_rptr }; diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c index 64e875d528dd..6a0fcd67662a 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c @@ -37,7 +37,6 @@ #include "gmc/gmc_8_2_sh_mask.h" #include "oss/oss_3_0_d.h" #include "oss/oss_3_0_sh_mask.h" -#include "gca/gfx_8_0_sh_mask.h" #include "dce/dce_10_0_d.h" #include "dce/dce_10_0_sh_mask.h" #include "smu/smu_7_1_3_d.h" diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c b/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c index 6f9c54978cc1..accdedd63c98 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v6_1.c @@ -32,6 +32,7 @@ #define smnCPM_CONTROL 0x11180460 #define smnPCIE_CNTL2 0x11180070 #define smnPCIE_CONFIG_CNTL 0x11180044 +#define smnPCIE_CI_CNTL 0x11180080 static u32 nbio_v6_1_get_rev_id(struct amdgpu_device *adev) { @@ -270,6 +271,12 @@ static void nbio_v6_1_init_registers(struct amdgpu_device *adev) if (def != data) WREG32_PCIE(smnPCIE_CONFIG_CNTL, data); + + def = data = RREG32_PCIE(smnPCIE_CI_CNTL); + data = REG_SET_FIELD(data, PCIE_CI_CNTL, CI_SLV_ORDERING_DIS, 1); + + if (def != data) + WREG32_PCIE(smnPCIE_CI_CNTL, data); } const struct amdgpu_nbio_funcs nbio_v6_1_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c index f8cee95d61cc..4cd31a276dcd 100644 --- a/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c +++ b/drivers/gpu/drm/amd/amdgpu/nbio_v7_4.c @@ -31,6 +31,7 @@ #define smnCPM_CONTROL 0x11180460 #define smnPCIE_CNTL2 0x11180070 +#define smnPCIE_CI_CNTL 0x11180080 static u32 nbio_v7_4_get_rev_id(struct amdgpu_device *adev) { @@ -222,7 +223,13 @@ static void nbio_v7_4_detect_hw_virt(struct amdgpu_device *adev) static void nbio_v7_4_init_registers(struct amdgpu_device *adev) { + uint32_t def, data; + + def = data = RREG32_PCIE(smnPCIE_CI_CNTL); + data = REG_SET_FIELD(data, PCIE_CI_CNTL, CI_SLV_ORDERING_DIS, 1); + if (def != data) + WREG32_PCIE(smnPCIE_CI_CNTL, data); } const struct amdgpu_nbio_funcs nbio_v7_4_funcs = { diff --git a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h index 882bd83a28c4..0de00fbe9233 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h +++ b/drivers/gpu/drm/amd/amdgpu/psp_gfx_if.h @@ -43,6 +43,8 @@ enum psp_gfx_crtl_cmd_id GFX_CTRL_CMD_ID_ENABLE_INT = 0x00050000, /* enable PSP-to-Gfx interrupt */ GFX_CTRL_CMD_ID_DISABLE_INT = 0x00060000, /* disable PSP-to-Gfx interrupt */ GFX_CTRL_CMD_ID_MODE1_RST = 0x00070000, /* trigger the Mode 1 reset */ + GFX_CTRL_CMD_ID_CONSUME_CMD = 0x000A0000, /* send interrupt to psp for updating write pointer of vf */ + GFX_CTRL_CMD_ID_DESTROY_GPCOM_RING = 0x000C0000, /* destroy GPCOM ring */ GFX_CTRL_CMD_ID_MAX = 0x000F0000, /* max command ID */ }; @@ -89,7 +91,8 @@ enum psp_gfx_cmd_id GFX_CMD_ID_LOAD_IP_FW = 0x00000006, /* load HW IP FW */ GFX_CMD_ID_DESTROY_TMR = 0x00000007, /* destroy TMR region */ GFX_CMD_ID_SAVE_RESTORE = 0x00000008, /* save/restore HW IP FW */ - + GFX_CMD_ID_SETUP_VMR = 0x00000009, /* setup VMR region */ + GFX_CMD_ID_DESTROY_VMR = 0x0000000A, /* destroy VMR region */ }; diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c index 295c2205485a..d78b4306a36f 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v10_0.c @@ -240,12 +240,9 @@ static int psp_v10_0_ring_stop(struct psp_context *psp, enum psp_ring_type ring_type) { int ret = 0; - struct psp_ring *ring; unsigned int psp_ring_reg = 0; struct amdgpu_device *adev = psp->adev; - ring = &psp->km_ring; - /* Write the ring destroy command to C2PMSG_64 */ psp_ring_reg = 3 << 16; WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, psp_ring_reg); diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c index e5dd052d9e06..0c6e7f9b143f 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c @@ -34,6 +34,7 @@ #include "nbio/nbio_7_4_offset.h" MODULE_FIRMWARE("amdgpu/vega20_sos.bin"); +MODULE_FIRMWARE("amdgpu/vega20_asd.bin"); MODULE_FIRMWARE("amdgpu/vega20_ta.bin"); /* address block */ @@ -100,6 +101,7 @@ static int psp_v11_0_init_microcode(struct psp_context *psp) char fw_name[30]; int err = 0; const struct psp_firmware_header_v1_0 *sos_hdr; + const struct psp_firmware_header_v1_0 *asd_hdr; const struct ta_firmware_header_v1_0 *ta_hdr; DRM_DEBUG("\n"); @@ -132,14 +134,30 @@ static int psp_v11_0_init_microcode(struct psp_context *psp) adev->psp.sos_start_addr = (uint8_t *)adev->psp.sys_start_addr + le32_to_cpu(sos_hdr->sos_offset_bytes); + snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_asd.bin", chip_name); + err = request_firmware(&adev->psp.asd_fw, fw_name, adev->dev); + if (err) + goto out1; + + err = amdgpu_ucode_validate(adev->psp.asd_fw); + if (err) + goto out1; + + asd_hdr = (const struct psp_firmware_header_v1_0 *)adev->psp.asd_fw->data; + adev->psp.asd_fw_version = le32_to_cpu(asd_hdr->header.ucode_version); + adev->psp.asd_feature_version = le32_to_cpu(asd_hdr->ucode_feature_version); + adev->psp.asd_ucode_size = le32_to_cpu(asd_hdr->header.ucode_size_bytes); + adev->psp.asd_start_addr = (uint8_t *)asd_hdr + + le32_to_cpu(asd_hdr->header.ucode_array_offset_bytes); + snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_ta.bin", chip_name); err = request_firmware(&adev->psp.ta_fw, fw_name, adev->dev); if (err) - goto out; + goto out2; err = amdgpu_ucode_validate(adev->psp.ta_fw); if (err) - goto out; + goto out2; ta_hdr = (const struct ta_firmware_header_v1_0 *)adev->psp.ta_fw->data; adev->psp.ta_xgmi_ucode_version = le32_to_cpu(ta_hdr->ta_xgmi_ucode_version); @@ -148,14 +166,18 @@ static int psp_v11_0_init_microcode(struct psp_context *psp) le32_to_cpu(ta_hdr->header.ucode_array_offset_bytes); return 0; + +out2: + release_firmware(adev->psp.ta_fw); + adev->psp.ta_fw = NULL; +out1: + release_firmware(adev->psp.asd_fw); + adev->psp.asd_fw = NULL; out: - if (err) { - dev_err(adev->dev, - "psp v11.0: Failed to load firmware \"%s\"\n", - fw_name); - release_firmware(adev->psp.sos_fw); - adev->psp.sos_fw = NULL; - } + dev_err(adev->dev, + "psp v11.0: Failed to load firmware \"%s\"\n", fw_name); + release_firmware(adev->psp.sos_fw); + adev->psp.sos_fw = NULL; return err; } @@ -171,8 +193,11 @@ static int psp_v11_0_bootloader_load_sysdrv(struct psp_context *psp) * are already been loaded. */ sol_reg = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81); - if (sol_reg) + if (sol_reg) { + psp->sos_fw_version = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_58); + printk("sos fw version = 0x%x.\n", psp->sos_fw_version); return 0; + } /* Wait for bootloader to signify that is ready having bit 31 of C2PMSG_35 set to 1 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), @@ -288,6 +313,13 @@ static int psp_v11_0_ring_init(struct psp_context *psp, return 0; } +static bool psp_v11_0_support_vmr_ring(struct psp_context *psp) +{ + if (amdgpu_sriov_vf(psp->adev) && psp->sos_fw_version > 0x80045) + return true; + return false; +} + static int psp_v11_0_ring_create(struct psp_context *psp, enum psp_ring_type ring_type) { @@ -296,26 +328,47 @@ static int psp_v11_0_ring_create(struct psp_context *psp, struct psp_ring *ring = &psp->km_ring; struct amdgpu_device *adev = psp->adev; - /* Write low address of the ring to C2PMSG_69 */ - psp_ring_reg = lower_32_bits(ring->ring_mem_mc_addr); - WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_69, psp_ring_reg); - /* Write high address of the ring to C2PMSG_70 */ - psp_ring_reg = upper_32_bits(ring->ring_mem_mc_addr); - WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_70, psp_ring_reg); - /* Write size of ring to C2PMSG_71 */ - psp_ring_reg = ring->ring_size; - WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_71, psp_ring_reg); - /* Write the ring initialization command to C2PMSG_64 */ - psp_ring_reg = ring_type; - psp_ring_reg = psp_ring_reg << 16; - WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, psp_ring_reg); - - /* there might be handshake issue with hardware which needs delay */ - mdelay(20); - - /* Wait for response flag (bit 31) in C2PMSG_64 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x8000FFFF, false); + if (psp_v11_0_support_vmr_ring(psp)) { + /* Write low address of the ring to C2PMSG_102 */ + psp_ring_reg = lower_32_bits(ring->ring_mem_mc_addr); + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_102, psp_ring_reg); + /* Write high address of the ring to C2PMSG_103 */ + psp_ring_reg = upper_32_bits(ring->ring_mem_mc_addr); + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_103, psp_ring_reg); + + /* Write the ring initialization command to C2PMSG_101 */ + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_101, + GFX_CTRL_CMD_ID_INIT_GPCOM_RING); + + /* there might be handshake issue with hardware which needs delay */ + mdelay(20); + + /* Wait for response flag (bit 31) in C2PMSG_101 */ + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + 0x80000000, 0x8000FFFF, false); + + } else { + /* Write low address of the ring to C2PMSG_69 */ + psp_ring_reg = lower_32_bits(ring->ring_mem_mc_addr); + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_69, psp_ring_reg); + /* Write high address of the ring to C2PMSG_70 */ + psp_ring_reg = upper_32_bits(ring->ring_mem_mc_addr); + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_70, psp_ring_reg); + /* Write size of ring to C2PMSG_71 */ + psp_ring_reg = ring->ring_size; + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_71, psp_ring_reg); + /* Write the ring initialization command to C2PMSG_64 */ + psp_ring_reg = ring_type; + psp_ring_reg = psp_ring_reg << 16; + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, psp_ring_reg); + + /* there might be handshake issue with hardware which needs delay */ + mdelay(20); + + /* Wait for response flag (bit 31) in C2PMSG_64 */ + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + 0x80000000, 0x8000FFFF, false); + } return ret; } @@ -326,15 +379,24 @@ static int psp_v11_0_ring_stop(struct psp_context *psp, int ret = 0; struct amdgpu_device *adev = psp->adev; - /* Write the ring destroy command to C2PMSG_64 */ - WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, GFX_CTRL_CMD_ID_DESTROY_RINGS); + /* Write the ring destroy command*/ + if (psp_v11_0_support_vmr_ring(psp)) + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_101, + GFX_CTRL_CMD_ID_DESTROY_GPCOM_RING); + else + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, + GFX_CTRL_CMD_ID_DESTROY_RINGS); /* there might be handshake issue with hardware which needs delay */ mdelay(20); - /* Wait for response flag (bit 31) in C2PMSG_64 */ - ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), - 0x80000000, 0x80000000, false); + /* Wait for response flag (bit 31) */ + if (psp_v11_0_support_vmr_ring(psp)) + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_101), + 0x80000000, 0x80000000, false); + else + ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_64), + 0x80000000, 0x80000000, false); return ret; } @@ -373,7 +435,10 @@ static int psp_v11_0_cmd_submit(struct psp_context *psp, uint32_t rb_frame_size_dw = sizeof(struct psp_gfx_rb_frame) / 4; /* KM (GPCOM) prepare write pointer */ - psp_write_ptr_reg = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_67); + if (psp_v11_0_support_vmr_ring(psp)) + psp_write_ptr_reg = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_102); + else + psp_write_ptr_reg = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_67); /* Update KM RB frame pointer to new frame */ /* write_frame ptr increments by size of rb_frame in bytes */ @@ -402,7 +467,11 @@ static int psp_v11_0_cmd_submit(struct psp_context *psp, /* Update the write Pointer in DWORDs */ psp_write_ptr_reg = (psp_write_ptr_reg + rb_frame_size_dw) % ring_size_dw; - WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_67, psp_write_ptr_reg); + if (psp_v11_0_support_vmr_ring(psp)) { + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_102, psp_write_ptr_reg); + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_101, GFX_CTRL_CMD_ID_CONSUME_CMD); + } else + WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_67, psp_write_ptr_reg); return 0; } @@ -547,7 +616,7 @@ static int psp_v11_0_mode1_reset(struct psp_context *psp) /*send the mode 1 reset command*/ WREG32(offset, GFX_CTRL_CMD_ID_MODE1_RST); - mdelay(1000); + msleep(500); offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_33); @@ -640,7 +709,7 @@ static int psp_v11_0_xgmi_set_topology_info(struct psp_context *psp, return psp_xgmi_invoke(psp, TA_COMMAND_XGMI__SET_TOPOLOGY_INFO); } -static u64 psp_v11_0_xgmi_get_hive_id(struct psp_context *psp) +static int psp_v11_0_xgmi_get_hive_id(struct psp_context *psp, uint64_t *hive_id) { struct ta_xgmi_shared_memory *xgmi_cmd; int ret; @@ -653,12 +722,14 @@ static u64 psp_v11_0_xgmi_get_hive_id(struct psp_context *psp) /* Invoke xgmi ta to get hive id */ ret = psp_xgmi_invoke(psp, xgmi_cmd->cmd_id); if (ret) - return 0; - else - return xgmi_cmd->xgmi_out_message.get_hive_id.hive_id; + return ret; + + *hive_id = xgmi_cmd->xgmi_out_message.get_hive_id.hive_id; + + return 0; } -static u64 psp_v11_0_xgmi_get_node_id(struct psp_context *psp) +static int psp_v11_0_xgmi_get_node_id(struct psp_context *psp, uint64_t *node_id) { struct ta_xgmi_shared_memory *xgmi_cmd; int ret; @@ -671,9 +742,11 @@ static u64 psp_v11_0_xgmi_get_node_id(struct psp_context *psp) /* Invoke xgmi ta to get the node id */ ret = psp_xgmi_invoke(psp, xgmi_cmd->cmd_id); if (ret) - return 0; - else - return xgmi_cmd->xgmi_out_message.get_node_id.node_id; + return ret; + + *node_id = xgmi_cmd->xgmi_out_message.get_node_id.node_id; + + return 0; } static const struct psp_funcs psp_v11_0_funcs = { @@ -692,6 +765,7 @@ static const struct psp_funcs psp_v11_0_funcs = { .xgmi_set_topology_info = psp_v11_0_xgmi_set_topology_info, .xgmi_get_hive_id = psp_v11_0_xgmi_get_hive_id, .xgmi_get_node_id = psp_v11_0_xgmi_get_node_id, + .support_vmr_ring = psp_v11_0_support_vmr_ring, }; void psp_v11_0_set_psp_funcs(struct psp_context *psp) diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c index 9cea0bbe4525..79694ff16969 100644 --- a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c +++ b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c @@ -240,8 +240,11 @@ static int psp_v3_1_bootloader_load_sos(struct psp_context *psp) * are already been loaded. */ sol_reg = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_81); - if (sol_reg) + if (sol_reg) { + psp->sos_fw_version = RREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_58); + printk("sos fw version = 0x%x.\n", psp->sos_fw_version); return 0; + } /* Wait for bootloader to signify that is ready having bit 31 of C2PMSG_35 set to 1 */ ret = psp_wait_for(psp, SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_35), @@ -356,12 +359,9 @@ static int psp_v3_1_ring_stop(struct psp_context *psp, enum psp_ring_type ring_type) { int ret = 0; - struct psp_ring *ring; unsigned int psp_ring_reg = 0; struct amdgpu_device *adev = psp->adev; - ring = &psp->km_ring; - /* Write the ring destroy command to C2PMSG_64 */ psp_ring_reg = 3 << 16; WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_64, psp_ring_reg); @@ -593,9 +593,9 @@ static int psp_v3_1_mode1_reset(struct psp_context *psp) } /*send the mode 1 reset command*/ - WREG32(offset, 0x70000); + WREG32(offset, GFX_CTRL_CMD_ID_MODE1_RST); - mdelay(1000); + msleep(500); offset = SOC15_REG_OFFSET(MP0, 0, mmMP0_SMN_C2PMSG_33); diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c index b6a25f92d566..1bccc5fe2d9d 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c @@ -1146,7 +1146,7 @@ static int sdma_v3_0_sw_init(void *handle) if (!amdgpu_sriov_vf(adev)) { ring->use_doorbell = true; ring->doorbell_index = (i == 0) ? - AMDGPU_DOORBELL_sDMA_ENGINE0 : AMDGPU_DOORBELL_sDMA_ENGINE1; + adev->doorbell_index.sdma_engine0 : adev->doorbell_index.sdma_engine1; } else { ring->use_pollmem = true; } diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c index f4490cdd9804..fd0bfe140ee0 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c @@ -925,11 +925,9 @@ static void sdma_v4_0_page_resume(struct amdgpu_device *adev, unsigned int i) OFFSET, ring->doorbell_index); WREG32_SDMA(i, mmSDMA0_PAGE_DOORBELL, doorbell); WREG32_SDMA(i, mmSDMA0_PAGE_DOORBELL_OFFSET, doorbell_offset); - /* TODO: enable doorbell support */ - /*adev->nbio_funcs->sdma_doorbell_range(adev, i, ring->use_doorbell, - ring->doorbell_index);*/ - sdma_v4_0_ring_set_wptr(ring); + /* paging queue doorbell range is setup at sdma_v4_0_gfx_resume */ + sdma_v4_0_page_ring_set_wptr(ring); /* set minor_ptr_update to 0 after wptr programed */ WREG32_SDMA(i, mmSDMA0_PAGE_MINOR_PTR_UPDATE, 0); @@ -1449,23 +1447,45 @@ static void sdma_v4_0_ring_emit_reg_wait(struct amdgpu_ring *ring, uint32_t reg, sdma_v4_0_wait_reg_mem(ring, 0, 0, reg, 0, val, mask, 10); } +static bool sdma_v4_0_fw_support_paging_queue(struct amdgpu_device *adev) +{ + uint fw_version = adev->sdma.instance[0].fw_version; + + switch (adev->asic_type) { + case CHIP_VEGA10: + return fw_version >= 430; + case CHIP_VEGA12: + /*return fw_version >= 31;*/ + return false; + case CHIP_VEGA20: + return fw_version >= 123; + default: + return false; + } +} + static int sdma_v4_0_early_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + int r; - if (adev->asic_type == CHIP_RAVEN) { + if (adev->asic_type == CHIP_RAVEN) adev->sdma.num_instances = 1; - adev->sdma.has_page_queue = false; - } else { + else adev->sdma.num_instances = 2; - /* TODO: Page queue breaks driver reload under SRIOV */ - if ((adev->asic_type == CHIP_VEGA10) && amdgpu_sriov_vf((adev))) - adev->sdma.has_page_queue = false; - else if (adev->asic_type != CHIP_VEGA20 && - adev->asic_type != CHIP_VEGA12) - adev->sdma.has_page_queue = true; + + r = sdma_v4_0_init_microcode(adev); + if (r) { + DRM_ERROR("Failed to load sdma firmware!\n"); + return r; } + /* TODO: Page queue breaks driver reload under SRIOV */ + if ((adev->asic_type == CHIP_VEGA10) && amdgpu_sriov_vf((adev))) + adev->sdma.has_page_queue = false; + else if (sdma_v4_0_fw_support_paging_queue(adev)) + adev->sdma.has_page_queue = true; + sdma_v4_0_set_ring_funcs(adev); sdma_v4_0_set_buffer_funcs(adev); sdma_v4_0_set_vm_pte_funcs(adev); @@ -1474,7 +1494,6 @@ static int sdma_v4_0_early_init(void *handle) return 0; } - static int sdma_v4_0_sw_init(void *handle) { struct amdgpu_ring *ring; @@ -1493,12 +1512,6 @@ static int sdma_v4_0_sw_init(void *handle) if (r) return r; - r = sdma_v4_0_init_microcode(adev); - if (r) { - DRM_ERROR("Failed to load sdma firmware!\n"); - return r; - } - for (i = 0; i < adev->sdma.num_instances; i++) { ring = &adev->sdma.instance[i].ring; ring->ring_obj = NULL; @@ -1507,15 +1520,10 @@ static int sdma_v4_0_sw_init(void *handle) DRM_INFO("use_doorbell being set to: [%s]\n", ring->use_doorbell?"true":"false"); - if (adev->asic_type == CHIP_VEGA10) - ring->doorbell_index = (i == 0) ? - (AMDGPU_VEGA10_DOORBELL64_sDMA_ENGINE0 << 1) //get DWORD offset - : (AMDGPU_VEGA10_DOORBELL64_sDMA_ENGINE1 << 1); // get DWORD offset - else - ring->doorbell_index = (i == 0) ? - (AMDGPU_DOORBELL64_sDMA_ENGINE0 << 1) //get DWORD offset - : (AMDGPU_DOORBELL64_sDMA_ENGINE1 << 1); // get DWORD offset - + /* doorbell size is 2 dwords, get DWORD offset */ + ring->doorbell_index = (i == 0) ? + (adev->doorbell_index.sdma_engine0 << 1) + : (adev->doorbell_index.sdma_engine1 << 1); sprintf(ring->name, "sdma%d", i); r = amdgpu_ring_init(adev, ring, 1024, @@ -1529,7 +1537,15 @@ static int sdma_v4_0_sw_init(void *handle) if (adev->sdma.has_page_queue) { ring = &adev->sdma.instance[i].page; ring->ring_obj = NULL; - ring->use_doorbell = false; + ring->use_doorbell = true; + + /* paging queue use same doorbell index/routing as gfx queue + * with 0x400 (4096 dwords) offset on second doorbell page + */ + ring->doorbell_index = (i == 0) ? + (adev->doorbell_index.sdma_engine0 << 1) + : (adev->doorbell_index.sdma_engine1 << 1); + ring->doorbell_index += 0x400; sprintf(ring->name, "page%d", i); r = amdgpu_ring_init(adev, ring, 1024, @@ -1689,13 +1705,15 @@ static int sdma_v4_0_process_trap_irq(struct amdgpu_device *adev, amdgpu_fence_process(&adev->sdma.instance[instance].ring); break; case 1: - /* XXX compute */ + if (adev->asic_type == CHIP_VEGA20) + amdgpu_fence_process(&adev->sdma.instance[instance].page); break; case 2: /* XXX compute */ break; case 3: - amdgpu_fence_process(&adev->sdma.instance[instance].page); + if (adev->asic_type != CHIP_VEGA20) + amdgpu_fence_process(&adev->sdma.instance[instance].page); break; } return 0; diff --git a/drivers/gpu/drm/amd/amdgpu/si_ih.c b/drivers/gpu/drm/amd/amdgpu/si_ih.c index b3d7d9f83202..2938fb9f17cc 100644 --- a/drivers/gpu/drm/amd/amdgpu/si_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/si_ih.c @@ -118,19 +118,6 @@ static u32 si_ih_get_wptr(struct amdgpu_device *adev) return (wptr & adev->irq.ih.ptr_mask); } -/** - * si_ih_prescreen_iv - prescreen an interrupt vector - * - * @adev: amdgpu_device pointer - * - * Returns true if the interrupt vector should be further processed. - */ -static bool si_ih_prescreen_iv(struct amdgpu_device *adev) -{ - /* Process all interrupts */ - return true; -} - static void si_ih_decode_iv(struct amdgpu_device *adev, struct amdgpu_iv_entry *entry) { @@ -301,7 +288,6 @@ static const struct amd_ip_funcs si_ih_ip_funcs = { static const struct amdgpu_ih_funcs si_ih_funcs = { .get_wptr = si_ih_get_wptr, - .prescreen_iv = si_ih_prescreen_iv, .decode_iv = si_ih_decode_iv, .set_rptr = si_ih_set_rptr }; diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index 4cc0dcb1a187..8849b74078d6 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -507,6 +507,9 @@ int soc15_set_ip_blocks(struct amdgpu_device *adev) return -EINVAL; } + if (adev->asic_type == CHIP_VEGA20) + adev->gmc.xgmi.supported = true; + if (adev->flags & AMD_IS_APU) adev->nbio_funcs = &nbio_v7_0_funcs; else if (adev->asic_type == CHIP_VEGA20) @@ -613,6 +616,24 @@ static const struct amdgpu_asic_funcs soc15_asic_funcs = .flush_hdp = &soc15_flush_hdp, .invalidate_hdp = &soc15_invalidate_hdp, .need_full_reset = &soc15_need_full_reset, + .init_doorbell_index = &vega10_doorbell_index_init, +}; + +static const struct amdgpu_asic_funcs vega20_asic_funcs = +{ + .read_disabled_bios = &soc15_read_disabled_bios, + .read_bios_from_rom = &soc15_read_bios_from_rom, + .read_register = &soc15_read_register, + .reset = &soc15_asic_reset, + .set_vga_state = &soc15_vga_set_state, + .get_xclk = &soc15_get_xclk, + .set_uvd_clocks = &soc15_set_uvd_clocks, + .set_vce_clocks = &soc15_set_vce_clocks, + .get_config_memsize = &soc15_get_config_memsize, + .flush_hdp = &soc15_flush_hdp, + .invalidate_hdp = &soc15_invalidate_hdp, + .need_full_reset = &soc15_need_full_reset, + .init_doorbell_index = &vega20_doorbell_index_init, }; static int soc15_common_early_init(void *handle) @@ -632,11 +653,11 @@ static int soc15_common_early_init(void *handle) adev->se_cac_rreg = &soc15_se_cac_rreg; adev->se_cac_wreg = &soc15_se_cac_wreg; - adev->asic_funcs = &soc15_asic_funcs; adev->external_rev_id = 0xFF; switch (adev->asic_type) { case CHIP_VEGA10: + adev->asic_funcs = &soc15_asic_funcs; adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_MGLS | AMD_CG_SUPPORT_GFX_RLC_LS | @@ -660,6 +681,7 @@ static int soc15_common_early_init(void *handle) adev->external_rev_id = 0x1; break; case CHIP_VEGA12: + adev->asic_funcs = &soc15_asic_funcs; adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_MGLS | AMD_CG_SUPPORT_GFX_CGCG | @@ -682,6 +704,7 @@ static int soc15_common_early_init(void *handle) adev->external_rev_id = adev->rev_id + 0x14; break; case CHIP_VEGA20: + adev->asic_funcs = &vega20_asic_funcs; adev->cg_flags = AMD_CG_SUPPORT_GFX_MGCG | AMD_CG_SUPPORT_GFX_MGLS | AMD_CG_SUPPORT_GFX_CGCG | @@ -704,6 +727,7 @@ static int soc15_common_early_init(void *handle) adev->external_rev_id = adev->rev_id + 0x28; break; case CHIP_RAVEN: + adev->asic_funcs = &soc15_asic_funcs; if (adev->rev_id >= 0x8) adev->external_rev_id = adev->rev_id + 0x81; else if (adev->pdev->device == 0x15d8) diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.h b/drivers/gpu/drm/amd/amdgpu/soc15.h index f8ad7804dc40..a66c8bfbbaa6 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15.h @@ -58,4 +58,6 @@ void soc15_program_register_sequence(struct amdgpu_device *adev, int vega10_reg_base_init(struct amdgpu_device *adev); int vega20_reg_base_init(struct amdgpu_device *adev); +void vega10_doorbell_index_init(struct amdgpu_device *adev); +void vega20_doorbell_index_init(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/soc15_common.h b/drivers/gpu/drm/amd/amdgpu/soc15_common.h index 958b10a57073..49c262540940 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15_common.h +++ b/drivers/gpu/drm/amd/amdgpu/soc15_common.h @@ -49,14 +49,19 @@ #define SOC15_WAIT_ON_RREG(ip, inst, reg, expected_value, mask, ret) \ do { \ + uint32_t old_ = 0; \ uint32_t tmp_ = RREG32(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg); \ uint32_t loop = adev->usec_timeout; \ while ((tmp_ & (mask)) != (expected_value)) { \ - udelay(2); \ + if (old_ != tmp_) { \ + loop = adev->usec_timeout; \ + old_ = tmp_; \ + } else \ + udelay(1); \ tmp_ = RREG32(adev->reg_offset[ip##_HWIP][inst][reg##_BASE_IDX] + reg); \ loop--; \ if (!loop) { \ - DRM_ERROR("Register(%d) [%s] failed to reach value 0x%08x != 0x%08x\n", \ + DRM_WARN("Register(%d) [%s] failed to reach value 0x%08x != 0x%08x\n", \ inst, #reg, (unsigned)expected_value, (unsigned)(tmp_ & (mask))); \ ret = -ETIMEDOUT; \ break; \ diff --git a/drivers/gpu/drm/amd/amdgpu/tonga_ih.c b/drivers/gpu/drm/amd/amdgpu/tonga_ih.c index 3abffd06b5c7..15da06ddeb75 100644 --- a/drivers/gpu/drm/amd/amdgpu/tonga_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/tonga_ih.c @@ -219,34 +219,6 @@ static u32 tonga_ih_get_wptr(struct amdgpu_device *adev) } /** - * tonga_ih_prescreen_iv - prescreen an interrupt vector - * - * @adev: amdgpu_device pointer - * - * Returns true if the interrupt vector should be further processed. - */ -static bool tonga_ih_prescreen_iv(struct amdgpu_device *adev) -{ - u32 ring_index = adev->irq.ih.rptr >> 2; - u16 pasid; - - switch (le32_to_cpu(adev->irq.ih.ring[ring_index]) & 0xff) { - case 146: - case 147: - pasid = le32_to_cpu(adev->irq.ih.ring[ring_index + 2]) >> 16; - if (!pasid || amdgpu_vm_pasid_fault_credit(adev, pasid)) - return true; - break; - default: - /* Not a VM fault */ - return true; - } - - adev->irq.ih.rptr += 16; - return false; -} - -/** * tonga_ih_decode_iv - decode an interrupt vector * * @adev: amdgpu_device pointer @@ -322,7 +294,7 @@ static int tonga_ih_sw_init(void *handle) return r; adev->irq.ih.use_doorbell = true; - adev->irq.ih.doorbell_index = AMDGPU_DOORBELL_IH; + adev->irq.ih.doorbell_index = adev->doorbell_index.ih; r = amdgpu_irq_init(adev); @@ -506,7 +478,6 @@ static const struct amd_ip_funcs tonga_ih_ip_funcs = { static const struct amdgpu_ih_funcs tonga_ih_funcs = { .get_wptr = tonga_ih_get_wptr, - .prescreen_iv = tonga_ih_prescreen_iv, .decode_iv = tonga_ih_decode_iv, .set_rptr = tonga_ih_set_rptr }; diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c index 90bbcee00f28..d69c8f6daaf8 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c @@ -116,16 +116,16 @@ static int uvd_v4_2_sw_init(void *handle) if (r) return r; - r = amdgpu_uvd_resume(adev); - if (r) - return r; - ring = &adev->uvd.inst->ring; sprintf(ring->name, "uvd"); r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst->irq, 0); if (r) return r; + r = amdgpu_uvd_resume(adev); + if (r) + return r; + r = amdgpu_uvd_entity_init(adev); return r; diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c index 1c5e12703103..ee8cd06ddc38 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c @@ -113,16 +113,16 @@ static int uvd_v5_0_sw_init(void *handle) if (r) return r; - r = amdgpu_uvd_resume(adev); - if (r) - return r; - ring = &adev->uvd.inst->ring; sprintf(ring->name, "uvd"); r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst->irq, 0); if (r) return r; + r = amdgpu_uvd_resume(adev); + if (r) + return r; + r = amdgpu_uvd_entity_init(adev); return r; diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c index f184842ef2a2..d4f4a66f8324 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c @@ -400,16 +400,16 @@ static int uvd_v6_0_sw_init(void *handle) DRM_INFO("UVD ENC is disabled\n"); } - r = amdgpu_uvd_resume(adev); - if (r) - return r; - ring = &adev->uvd.inst->ring; sprintf(ring->name, "uvd"); r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst->irq, 0); if (r) return r; + r = amdgpu_uvd_resume(adev); + if (r) + return r; + if (uvd_v6_0_enc_support(adev)) { for (i = 0; i < adev->uvd.num_enc_rings; ++i) { ring = &adev->uvd.inst->ring_enc[i]; diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c index 8a4595968d98..aef924026a28 100644 --- a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c @@ -430,16 +430,12 @@ static int uvd_v7_0_sw_init(void *handle) DRM_INFO("PSP loading UVD firmware\n"); } - r = amdgpu_uvd_resume(adev); - if (r) - return r; - for (j = 0; j < adev->uvd.num_uvd_inst; j++) { if (adev->uvd.harvest_config & (1 << j)) continue; if (!amdgpu_sriov_vf(adev)) { ring = &adev->uvd.inst[j].ring; - sprintf(ring->name, "uvd<%d>", j); + sprintf(ring->name, "uvd_%d", ring->me); r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst[j].irq, 0); if (r) return r; @@ -447,7 +443,7 @@ static int uvd_v7_0_sw_init(void *handle) for (i = 0; i < adev->uvd.num_enc_rings; ++i) { ring = &adev->uvd.inst[j].ring_enc[i]; - sprintf(ring->name, "uvd_enc%d<%d>", i, j); + sprintf(ring->name, "uvd_enc_%d.%d", ring->me, i); if (amdgpu_sriov_vf(adev)) { ring->use_doorbell = true; @@ -455,9 +451,9 @@ static int uvd_v7_0_sw_init(void *handle) * sriov, so set unused location for other unused rings. */ if (i == 0) - ring->doorbell_index = AMDGPU_DOORBELL64_UVD_RING0_1 * 2; + ring->doorbell_index = adev->doorbell_index.uvd_vce.uvd_ring0_1 * 2; else - ring->doorbell_index = AMDGPU_DOORBELL64_UVD_RING2_3 * 2 + 1; + ring->doorbell_index = adev->doorbell_index.uvd_vce.uvd_ring2_3 * 2 + 1; } r = amdgpu_ring_init(adev, ring, 512, &adev->uvd.inst[j].irq, 0); if (r) @@ -465,6 +461,10 @@ static int uvd_v7_0_sw_init(void *handle) } } + r = amdgpu_uvd_resume(adev); + if (r) + return r; + r = amdgpu_uvd_entity_init(adev); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c index 3e84840859a7..2668effadd27 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c @@ -37,7 +37,6 @@ #include "gca/gfx_8_0_d.h" #include "smu/smu_7_1_2_d.h" #include "smu/smu_7_1_2_sh_mask.h" -#include "gca/gfx_8_0_d.h" #include "gca/gfx_8_0_sh_mask.h" #include "ivsrcid/ivsrcid_vislands30.h" diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c index 0054ba1b9a68..9fb34b7d8e03 100644 --- a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c @@ -466,9 +466,9 @@ static int vce_v4_0_sw_init(void *handle) * so set unused location for other unused rings. */ if (i == 0) - ring->doorbell_index = AMDGPU_DOORBELL64_VCE_RING0_1 * 2; + ring->doorbell_index = adev->doorbell_index.uvd_vce.vce_ring0_1 * 2; else - ring->doorbell_index = AMDGPU_DOORBELL64_VCE_RING2_3 * 2 + 1; + ring->doorbell_index = adev->doorbell_index.uvd_vce.vce_ring2_3 * 2 + 1; } r = amdgpu_ring_init(adev, ring, 512, &adev->vce.irq, 0); if (r) diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c index c1a03505f956..89bb2fef90eb 100644 --- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c @@ -48,6 +48,7 @@ static void vcn_v1_0_set_enc_ring_funcs(struct amdgpu_device *adev); static void vcn_v1_0_set_jpeg_ring_funcs(struct amdgpu_device *adev); static void vcn_v1_0_set_irq_funcs(struct amdgpu_device *adev); static void vcn_v1_0_jpeg_ring_set_patch_ring(struct amdgpu_ring *ring, uint32_t ptr); +static int vcn_v1_0_set_powergating_state(void *handle, enum amd_powergating_state state); /** * vcn_v1_0_early_init - set function pointers @@ -213,8 +214,9 @@ static int vcn_v1_0_hw_fini(void *handle) struct amdgpu_device *adev = (struct amdgpu_device *)handle; struct amdgpu_ring *ring = &adev->vcn.ring_dec; - if (RREG32_SOC15(VCN, 0, mmUVD_STATUS)) - vcn_v1_0_stop(adev); + if ((adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) || + RREG32_SOC15(VCN, 0, mmUVD_STATUS)) + vcn_v1_0_set_powergating_state(adev, AMD_PG_STATE_GATE); ring->sched.ready = false; @@ -1086,7 +1088,8 @@ static int vcn_v1_0_start_dpg_mode(struct amdgpu_device *adev) WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_RBC_RB_CNTL), 0, ~UVD_RBC_RB_CNTL__RB_NO_FETCH_MASK); - /* initialize wptr */ + /* initialize JPEG wptr */ + ring = &adev->vcn.ring_jpeg; ring->wptr = RREG32_SOC15(UVD, 0, mmUVD_JRBC_RB_WPTR); /* copy patch commands to the jpeg ring */ @@ -1158,21 +1161,29 @@ static int vcn_v1_0_stop_spg_mode(struct amdgpu_device *adev) static int vcn_v1_0_stop_dpg_mode(struct amdgpu_device *adev) { int ret_code = 0; + uint32_t tmp; /* Wait for power status to be UVD_POWER_STATUS__UVD_POWER_STATUS_TILES_OFF */ SOC15_WAIT_ON_RREG(UVD, 0, mmUVD_POWER_STATUS, UVD_POWER_STATUS__UVD_POWER_STATUS_TILES_OFF, UVD_POWER_STATUS__UVD_POWER_STATUS_MASK, ret_code); - if (!ret_code) { - int tmp = RREG32_SOC15(UVD, 0, mmUVD_RBC_RB_WPTR) & 0x7FFFFFFF; - /* wait for read ptr to be equal to write ptr */ - SOC15_WAIT_ON_RREG(UVD, 0, mmUVD_RBC_RB_RPTR, tmp, 0xFFFFFFFF, ret_code); + /* wait for read ptr to be equal to write ptr */ + tmp = RREG32_SOC15(UVD, 0, mmUVD_RB_WPTR); + SOC15_WAIT_ON_RREG(UVD, 0, mmUVD_RB_RPTR, tmp, 0xFFFFFFFF, ret_code); - SOC15_WAIT_ON_RREG(UVD, 0, mmUVD_POWER_STATUS, - UVD_POWER_STATUS__UVD_POWER_STATUS_TILES_OFF, - UVD_POWER_STATUS__UVD_POWER_STATUS_MASK, ret_code); - } + tmp = RREG32_SOC15(UVD, 0, mmUVD_RB_WPTR2); + SOC15_WAIT_ON_RREG(UVD, 0, mmUVD_RB_RPTR2, tmp, 0xFFFFFFFF, ret_code); + + tmp = RREG32_SOC15(UVD, 0, mmUVD_JRBC_RB_WPTR); + SOC15_WAIT_ON_RREG(UVD, 0, mmUVD_JRBC_RB_RPTR, tmp, 0xFFFFFFFF, ret_code); + + tmp = RREG32_SOC15(UVD, 0, mmUVD_RBC_RB_WPTR) & 0x7FFFFFFF; + SOC15_WAIT_ON_RREG(UVD, 0, mmUVD_RBC_RB_RPTR, tmp, 0xFFFFFFFF, ret_code); + + SOC15_WAIT_ON_RREG(UVD, 0, mmUVD_POWER_STATUS, + UVD_POWER_STATUS__UVD_POWER_STATUS_TILES_OFF, + UVD_POWER_STATUS__UVD_POWER_STATUS_MASK, ret_code); /* disable dynamic power gating mode */ WREG32_P(SOC15_REG_OFFSET(UVD, 0, mmUVD_POWER_STATUS), 0, diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c index a0fda6f9252a..2c250b01a903 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c @@ -220,90 +220,6 @@ static u32 vega10_ih_get_wptr(struct amdgpu_device *adev) } /** - * vega10_ih_prescreen_iv - prescreen an interrupt vector - * - * @adev: amdgpu_device pointer - * - * Returns true if the interrupt vector should be further processed. - */ -static bool vega10_ih_prescreen_iv(struct amdgpu_device *adev) -{ - u32 ring_index = adev->irq.ih.rptr >> 2; - u32 dw0, dw3, dw4, dw5; - u16 pasid; - u64 addr, key; - struct amdgpu_vm *vm; - int r; - - dw0 = le32_to_cpu(adev->irq.ih.ring[ring_index + 0]); - dw3 = le32_to_cpu(adev->irq.ih.ring[ring_index + 3]); - dw4 = le32_to_cpu(adev->irq.ih.ring[ring_index + 4]); - dw5 = le32_to_cpu(adev->irq.ih.ring[ring_index + 5]); - - /* Filter retry page faults, let only the first one pass. If - * there are too many outstanding faults, ignore them until - * some faults get cleared. - */ - switch (dw0 & 0xff) { - case SOC15_IH_CLIENTID_VMC: - case SOC15_IH_CLIENTID_UTCL2: - break; - default: - /* Not a VM fault */ - return true; - } - - pasid = dw3 & 0xffff; - /* No PASID, can't identify faulting process */ - if (!pasid) - return true; - - /* Not a retry fault, check fault credit */ - if (!(dw5 & 0x80)) { - if (!amdgpu_vm_pasid_fault_credit(adev, pasid)) - goto ignore_iv; - return true; - } - - /* Track retry faults in per-VM fault FIFO. */ - spin_lock(&adev->vm_manager.pasid_lock); - vm = idr_find(&adev->vm_manager.pasid_idr, pasid); - addr = ((u64)(dw5 & 0xf) << 44) | ((u64)dw4 << 12); - key = AMDGPU_VM_FAULT(pasid, addr); - if (!vm) { - /* VM not found, process it normally */ - spin_unlock(&adev->vm_manager.pasid_lock); - return true; - } else { - r = amdgpu_vm_add_fault(vm->fault_hash, key); - - /* Hash table is full or the fault is already being processed, - * ignore further page faults - */ - if (r != 0) { - spin_unlock(&adev->vm_manager.pasid_lock); - goto ignore_iv; - } - } - /* No locking required with single writer and single reader */ - r = kfifo_put(&vm->faults, key); - if (!r) { - /* FIFO is full. Ignore it until there is space */ - amdgpu_vm_clear_fault(vm->fault_hash, key); - spin_unlock(&adev->vm_manager.pasid_lock); - goto ignore_iv; - } - - spin_unlock(&adev->vm_manager.pasid_lock); - /* It's the first fault for this address, process it normally */ - return true; - -ignore_iv: - adev->irq.ih.rptr += 32; - return false; -} - -/** * vega10_ih_decode_iv - decode an interrupt vector * * @adev: amdgpu_device pointer @@ -385,7 +301,7 @@ static int vega10_ih_sw_init(void *handle) return r; adev->irq.ih.use_doorbell = true; - adev->irq.ih.doorbell_index = AMDGPU_DOORBELL64_IH << 1; + adev->irq.ih.doorbell_index = adev->doorbell_index.ih << 1; r = amdgpu_irq_init(adev); @@ -487,7 +403,6 @@ const struct amd_ip_funcs vega10_ih_ip_funcs = { static const struct amdgpu_ih_funcs vega10_ih_funcs = { .get_wptr = vega10_ih_get_wptr, - .prescreen_iv = vega10_ih_prescreen_iv, .decode_iv = vega10_ih_decode_iv, .set_rptr = vega10_ih_set_rptr }; diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c b/drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c index c5c9b2bc190d..422674bb3cdf 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c +++ b/drivers/gpu/drm/amd/amdgpu/vega10_reg_init.c @@ -56,4 +56,32 @@ int vega10_reg_base_init(struct amdgpu_device *adev) return 0; } +void vega10_doorbell_index_init(struct amdgpu_device *adev) +{ + adev->doorbell_index.kiq = AMDGPU_DOORBELL64_KIQ; + adev->doorbell_index.mec_ring0 = AMDGPU_DOORBELL64_MEC_RING0; + adev->doorbell_index.mec_ring1 = AMDGPU_DOORBELL64_MEC_RING1; + adev->doorbell_index.mec_ring2 = AMDGPU_DOORBELL64_MEC_RING2; + adev->doorbell_index.mec_ring3 = AMDGPU_DOORBELL64_MEC_RING3; + adev->doorbell_index.mec_ring4 = AMDGPU_DOORBELL64_MEC_RING4; + adev->doorbell_index.mec_ring5 = AMDGPU_DOORBELL64_MEC_RING5; + adev->doorbell_index.mec_ring6 = AMDGPU_DOORBELL64_MEC_RING6; + adev->doorbell_index.mec_ring7 = AMDGPU_DOORBELL64_MEC_RING7; + adev->doorbell_index.userqueue_start = AMDGPU_DOORBELL64_USERQUEUE_START; + adev->doorbell_index.userqueue_end = AMDGPU_DOORBELL64_USERQUEUE_END; + adev->doorbell_index.gfx_ring0 = AMDGPU_DOORBELL64_GFX_RING0; + adev->doorbell_index.sdma_engine0 = AMDGPU_DOORBELL64_sDMA_ENGINE0; + adev->doorbell_index.sdma_engine1 = AMDGPU_DOORBELL64_sDMA_ENGINE1; + adev->doorbell_index.ih = AMDGPU_DOORBELL64_IH; + adev->doorbell_index.uvd_vce.uvd_ring0_1 = AMDGPU_DOORBELL64_UVD_RING0_1; + adev->doorbell_index.uvd_vce.uvd_ring2_3 = AMDGPU_DOORBELL64_UVD_RING2_3; + adev->doorbell_index.uvd_vce.uvd_ring4_5 = AMDGPU_DOORBELL64_UVD_RING4_5; + adev->doorbell_index.uvd_vce.uvd_ring6_7 = AMDGPU_DOORBELL64_UVD_RING6_7; + adev->doorbell_index.uvd_vce.vce_ring0_1 = AMDGPU_DOORBELL64_VCE_RING0_1; + adev->doorbell_index.uvd_vce.vce_ring2_3 = AMDGPU_DOORBELL64_VCE_RING2_3; + adev->doorbell_index.uvd_vce.vce_ring4_5 = AMDGPU_DOORBELL64_VCE_RING4_5; + adev->doorbell_index.uvd_vce.vce_ring6_7 = AMDGPU_DOORBELL64_VCE_RING6_7; + /* In unit of dword doorbell */ + adev->doorbell_index.max_assignment = AMDGPU_DOORBELL64_MAX_ASSIGNMENT << 1; +} diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c b/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c index d13fc4fcb517..edce413fda9a 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c +++ b/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c @@ -54,4 +54,37 @@ int vega20_reg_base_init(struct amdgpu_device *adev) return 0; } +void vega20_doorbell_index_init(struct amdgpu_device *adev) +{ + adev->doorbell_index.kiq = AMDGPU_VEGA20_DOORBELL_KIQ; + adev->doorbell_index.mec_ring0 = AMDGPU_VEGA20_DOORBELL_MEC_RING0; + adev->doorbell_index.mec_ring1 = AMDGPU_VEGA20_DOORBELL_MEC_RING1; + adev->doorbell_index.mec_ring2 = AMDGPU_VEGA20_DOORBELL_MEC_RING2; + adev->doorbell_index.mec_ring3 = AMDGPU_VEGA20_DOORBELL_MEC_RING3; + adev->doorbell_index.mec_ring4 = AMDGPU_VEGA20_DOORBELL_MEC_RING4; + adev->doorbell_index.mec_ring5 = AMDGPU_VEGA20_DOORBELL_MEC_RING5; + adev->doorbell_index.mec_ring6 = AMDGPU_VEGA20_DOORBELL_MEC_RING6; + adev->doorbell_index.mec_ring7 = AMDGPU_VEGA20_DOORBELL_MEC_RING7; + adev->doorbell_index.userqueue_start = AMDGPU_VEGA20_DOORBELL_USERQUEUE_START; + adev->doorbell_index.userqueue_end = AMDGPU_VEGA20_DOORBELL_USERQUEUE_END; + adev->doorbell_index.gfx_ring0 = AMDGPU_VEGA20_DOORBELL_GFX_RING0; + adev->doorbell_index.sdma_engine0 = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE0; + adev->doorbell_index.sdma_engine1 = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE1; + adev->doorbell_index.sdma_engine2 = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE2; + adev->doorbell_index.sdma_engine3 = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE3; + adev->doorbell_index.sdma_engine4 = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE4; + adev->doorbell_index.sdma_engine5 = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE5; + adev->doorbell_index.sdma_engine6 = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE6; + adev->doorbell_index.sdma_engine7 = AMDGPU_VEGA20_DOORBELL_sDMA_ENGINE7; + adev->doorbell_index.ih = AMDGPU_VEGA20_DOORBELL_IH; + adev->doorbell_index.uvd_vce.uvd_ring0_1 = AMDGPU_VEGA20_DOORBELL64_UVD_RING0_1; + adev->doorbell_index.uvd_vce.uvd_ring2_3 = AMDGPU_VEGA20_DOORBELL64_UVD_RING2_3; + adev->doorbell_index.uvd_vce.uvd_ring4_5 = AMDGPU_VEGA20_DOORBELL64_UVD_RING4_5; + adev->doorbell_index.uvd_vce.uvd_ring6_7 = AMDGPU_VEGA20_DOORBELL64_UVD_RING6_7; + adev->doorbell_index.uvd_vce.vce_ring0_1 = AMDGPU_VEGA20_DOORBELL64_VCE_RING0_1; + adev->doorbell_index.uvd_vce.vce_ring2_3 = AMDGPU_VEGA20_DOORBELL64_VCE_RING2_3; + adev->doorbell_index.uvd_vce.vce_ring4_5 = AMDGPU_VEGA20_DOORBELL64_VCE_RING4_5; + adev->doorbell_index.uvd_vce.vce_ring6_7 = AMDGPU_VEGA20_DOORBELL64_VCE_RING6_7; + adev->doorbell_index.max_assignment = AMDGPU_VEGA20_DOORBELL_MAX_ASSIGNMENT << 1; +} diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c index 07880d35e9de..77e367459101 100644 --- a/drivers/gpu/drm/amd/amdgpu/vi.c +++ b/drivers/gpu/drm/amd/amdgpu/vi.c @@ -87,9 +87,9 @@ static u32 vi_pcie_rreg(struct amdgpu_device *adev, u32 reg) u32 r; spin_lock_irqsave(&adev->pcie_idx_lock, flags); - WREG32(mmPCIE_INDEX, reg); - (void)RREG32(mmPCIE_INDEX); - r = RREG32(mmPCIE_DATA); + WREG32_NO_KIQ(mmPCIE_INDEX, reg); + (void)RREG32_NO_KIQ(mmPCIE_INDEX); + r = RREG32_NO_KIQ(mmPCIE_DATA); spin_unlock_irqrestore(&adev->pcie_idx_lock, flags); return r; } @@ -99,10 +99,10 @@ static void vi_pcie_wreg(struct amdgpu_device *adev, u32 reg, u32 v) unsigned long flags; spin_lock_irqsave(&adev->pcie_idx_lock, flags); - WREG32(mmPCIE_INDEX, reg); - (void)RREG32(mmPCIE_INDEX); - WREG32(mmPCIE_DATA, v); - (void)RREG32(mmPCIE_DATA); + WREG32_NO_KIQ(mmPCIE_INDEX, reg); + (void)RREG32_NO_KIQ(mmPCIE_INDEX); + WREG32_NO_KIQ(mmPCIE_DATA, v); + (void)RREG32_NO_KIQ(mmPCIE_DATA); spin_unlock_irqrestore(&adev->pcie_idx_lock, flags); } @@ -123,8 +123,8 @@ static void vi_smc_wreg(struct amdgpu_device *adev, u32 reg, u32 v) unsigned long flags; spin_lock_irqsave(&adev->smc_idx_lock, flags); - WREG32(mmSMC_IND_INDEX_11, (reg)); - WREG32(mmSMC_IND_DATA_11, (v)); + WREG32_NO_KIQ(mmSMC_IND_INDEX_11, (reg)); + WREG32_NO_KIQ(mmSMC_IND_DATA_11, (v)); spin_unlock_irqrestore(&adev->smc_idx_lock, flags); } @@ -955,6 +955,7 @@ static const struct amdgpu_asic_funcs vi_asic_funcs = .flush_hdp = &vi_flush_hdp, .invalidate_hdp = &vi_invalidate_hdp, .need_full_reset = &vi_need_full_reset, + .init_doorbell_index = &legacy_doorbell_index_init, }; #define CZ_REV_BRISTOL(rev) \ @@ -1712,3 +1713,21 @@ int vi_set_ip_blocks(struct amdgpu_device *adev) return 0; } + +void legacy_doorbell_index_init(struct amdgpu_device *adev) +{ + adev->doorbell_index.kiq = AMDGPU_DOORBELL_KIQ; + adev->doorbell_index.mec_ring0 = AMDGPU_DOORBELL_MEC_RING0; + adev->doorbell_index.mec_ring1 = AMDGPU_DOORBELL_MEC_RING1; + adev->doorbell_index.mec_ring2 = AMDGPU_DOORBELL_MEC_RING2; + adev->doorbell_index.mec_ring3 = AMDGPU_DOORBELL_MEC_RING3; + adev->doorbell_index.mec_ring4 = AMDGPU_DOORBELL_MEC_RING4; + adev->doorbell_index.mec_ring5 = AMDGPU_DOORBELL_MEC_RING5; + adev->doorbell_index.mec_ring6 = AMDGPU_DOORBELL_MEC_RING6; + adev->doorbell_index.mec_ring7 = AMDGPU_DOORBELL_MEC_RING7; + adev->doorbell_index.gfx_ring0 = AMDGPU_DOORBELL_GFX_RING0; + adev->doorbell_index.sdma_engine0 = AMDGPU_DOORBELL_sDMA_ENGINE0; + adev->doorbell_index.sdma_engine1 = AMDGPU_DOORBELL_sDMA_ENGINE1; + adev->doorbell_index.ih = AMDGPU_DOORBELL_IH; + adev->doorbell_index.max_assignment = AMDGPU_DOORBELL_MAX_ASSIGNMENT; +} diff --git a/drivers/gpu/drm/amd/amdgpu/vi.h b/drivers/gpu/drm/amd/amdgpu/vi.h index 0429fe332269..8de0772f986c 100644 --- a/drivers/gpu/drm/amd/amdgpu/vi.h +++ b/drivers/gpu/drm/amd/amdgpu/vi.h @@ -30,4 +30,5 @@ void vi_srbm_select(struct amdgpu_device *adev, u32 me, u32 pipe, u32 queue, u32 vmid); int vi_set_ip_blocks(struct amdgpu_device *adev); +void legacy_doorbell_index_init(struct amdgpu_device *adev); #endif diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index 5f4062b41add..083bd8114db1 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -33,6 +33,7 @@ #include <linux/time.h> #include <linux/mm.h> #include <linux/mman.h> +#include <linux/dma-buf.h> #include <asm/processor.h> #include "kfd_priv.h" #include "kfd_device_queue_manager.h" @@ -157,8 +158,7 @@ static int set_queue_properties_from_user(struct queue_properties *q_properties, } if ((args->ring_base_address) && - (!access_ok(VERIFY_WRITE, - (const void __user *) args->ring_base_address, + (!access_ok((const void __user *) args->ring_base_address, sizeof(uint64_t)))) { pr_err("Can't access ring base address\n"); return -EFAULT; @@ -169,31 +169,27 @@ static int set_queue_properties_from_user(struct queue_properties *q_properties, return -EINVAL; } - if (!access_ok(VERIFY_WRITE, - (const void __user *) args->read_pointer_address, + if (!access_ok((const void __user *) args->read_pointer_address, sizeof(uint32_t))) { pr_err("Can't access read pointer\n"); return -EFAULT; } - if (!access_ok(VERIFY_WRITE, - (const void __user *) args->write_pointer_address, + if (!access_ok((const void __user *) args->write_pointer_address, sizeof(uint32_t))) { pr_err("Can't access write pointer\n"); return -EFAULT; } if (args->eop_buffer_address && - !access_ok(VERIFY_WRITE, - (const void __user *) args->eop_buffer_address, + !access_ok((const void __user *) args->eop_buffer_address, sizeof(uint32_t))) { pr_debug("Can't access eop buffer"); return -EFAULT; } if (args->ctx_save_restore_address && - !access_ok(VERIFY_WRITE, - (const void __user *) args->ctx_save_restore_address, + !access_ok((const void __user *) args->ctx_save_restore_address, sizeof(uint32_t))) { pr_debug("Can't access ctx save restore buffer"); return -EFAULT; @@ -364,8 +360,7 @@ static int kfd_ioctl_update_queue(struct file *filp, struct kfd_process *p, } if ((args->ring_base_address) && - (!access_ok(VERIFY_WRITE, - (const void __user *) args->ring_base_address, + (!access_ok((const void __user *) args->ring_base_address, sizeof(uint64_t)))) { pr_err("Can't access ring base address\n"); return -EFAULT; @@ -1273,6 +1268,12 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep, return -EINVAL; } + if (flags & KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL) { + if (args->size != kfd_doorbell_process_slice(dev)) + return -EINVAL; + offset = kfd_get_process_doorbells(dev, p); + } + mutex_lock(&p->mutex); pdd = kfd_bind_process_to_device(dev, p); @@ -1550,6 +1551,115 @@ copy_from_user_failed: return err; } +static int kfd_ioctl_get_dmabuf_info(struct file *filep, + struct kfd_process *p, void *data) +{ + struct kfd_ioctl_get_dmabuf_info_args *args = data; + struct kfd_dev *dev = NULL; + struct kgd_dev *dma_buf_kgd; + void *metadata_buffer = NULL; + uint32_t flags; + unsigned int i; + int r; + + /* Find a KFD GPU device that supports the get_dmabuf_info query */ + for (i = 0; kfd_topology_enum_kfd_devices(i, &dev) == 0; i++) + if (dev) + break; + if (!dev) + return -EINVAL; + + if (args->metadata_ptr) { + metadata_buffer = kzalloc(args->metadata_size, GFP_KERNEL); + if (!metadata_buffer) + return -ENOMEM; + } + + /* Get dmabuf info from KGD */ + r = amdgpu_amdkfd_get_dmabuf_info(dev->kgd, args->dmabuf_fd, + &dma_buf_kgd, &args->size, + metadata_buffer, args->metadata_size, + &args->metadata_size, &flags); + if (r) + goto exit; + + /* Reverse-lookup gpu_id from kgd pointer */ + dev = kfd_device_by_kgd(dma_buf_kgd); + if (!dev) { + r = -EINVAL; + goto exit; + } + args->gpu_id = dev->id; + args->flags = flags; + + /* Copy metadata buffer to user mode */ + if (metadata_buffer) { + r = copy_to_user((void __user *)args->metadata_ptr, + metadata_buffer, args->metadata_size); + if (r != 0) + r = -EFAULT; + } + +exit: + kfree(metadata_buffer); + + return r; +} + +static int kfd_ioctl_import_dmabuf(struct file *filep, + struct kfd_process *p, void *data) +{ + struct kfd_ioctl_import_dmabuf_args *args = data; + struct kfd_process_device *pdd; + struct dma_buf *dmabuf; + struct kfd_dev *dev; + int idr_handle; + uint64_t size; + void *mem; + int r; + + dev = kfd_device_by_id(args->gpu_id); + if (!dev) + return -EINVAL; + + dmabuf = dma_buf_get(args->dmabuf_fd); + if (IS_ERR(dmabuf)) + return PTR_ERR(dmabuf); + + mutex_lock(&p->mutex); + + pdd = kfd_bind_process_to_device(dev, p); + if (IS_ERR(pdd)) { + r = PTR_ERR(pdd); + goto err_unlock; + } + + r = amdgpu_amdkfd_gpuvm_import_dmabuf(dev->kgd, dmabuf, + args->va_addr, pdd->vm, + (struct kgd_mem **)&mem, &size, + NULL); + if (r) + goto err_unlock; + + idr_handle = kfd_process_device_create_obj_handle(pdd, mem); + if (idr_handle < 0) { + r = -EFAULT; + goto err_free; + } + + mutex_unlock(&p->mutex); + + args->handle = MAKE_HANDLE(args->gpu_id, idr_handle); + + return 0; + +err_free: + amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->kgd, (struct kgd_mem *)mem); +err_unlock: + mutex_unlock(&p->mutex); + return r; +} + #define AMDKFD_IOCTL_DEF(ioctl, _func, _flags) \ [_IOC_NR(ioctl)] = {.cmd = ioctl, .func = _func, .flags = _flags, \ .cmd_drv = 0, .name = #ioctl} @@ -1635,7 +1745,13 @@ static const struct amdkfd_ioctl_desc amdkfd_ioctls[] = { kfd_ioctl_set_cu_mask, 0), AMDKFD_IOCTL_DEF(AMDKFD_IOC_GET_QUEUE_WAVE_STATE, - kfd_ioctl_get_queue_wave_state, 0) + kfd_ioctl_get_queue_wave_state, 0), + + AMDKFD_IOCTL_DEF(AMDKFD_IOC_GET_DMABUF_INFO, + kfd_ioctl_get_dmabuf_info, 0), + + AMDKFD_IOCTL_DEF(AMDKFD_IOC_IMPORT_DMABUF, + kfd_ioctl_import_dmabuf, 0), }; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c index 3783d122f283..b7bc7d7d048f 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c @@ -133,6 +133,7 @@ static struct kfd_gpu_cache_info carrizo_cache_info[] = { #define fiji_cache_info carrizo_cache_info #define polaris10_cache_info carrizo_cache_info #define polaris11_cache_info carrizo_cache_info +#define polaris12_cache_info carrizo_cache_info /* TODO - check & update Vega10 cache details */ #define vega10_cache_info carrizo_cache_info #define raven_cache_info carrizo_cache_info @@ -647,7 +648,12 @@ static int kfd_fill_gpu_cache_info(struct kfd_dev *kdev, pcache_info = polaris11_cache_info; num_of_cache_types = ARRAY_SIZE(polaris11_cache_info); break; + case CHIP_POLARIS12: + pcache_info = polaris12_cache_info; + num_of_cache_types = ARRAY_SIZE(polaris12_cache_info); + break; case CHIP_VEGA10: + case CHIP_VEGA12: case CHIP_VEGA20: pcache_info = vega10_cache_info; num_of_cache_types = ARRAY_SIZE(vega10_cache_info); @@ -847,7 +853,7 @@ static int kfd_fill_mem_info_for_cpu(int numa_node_id, int *avail_size, */ pgdat = NODE_DATA(numa_node_id); for (zone_type = 0; zone_type < MAX_NR_ZONES; zone_type++) - mem_in_bytes += pgdat->node_zones[zone_type].managed_pages; + mem_in_bytes += zone_managed_pages(&pgdat->node_zones[zone_type]); mem_in_bytes <<= PAGE_SHIFT; sub_type_hdr->length_low = lower_32_bits(mem_in_bytes); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c index c004647c8cb4..8be9677c0c07 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c @@ -205,6 +205,22 @@ static const struct kfd_device_info polaris11_device_info = { .num_sdma_queues_per_engine = 2, }; +static const struct kfd_device_info polaris12_device_info = { + .asic_family = CHIP_POLARIS12, + .max_pasid_bits = 16, + .max_no_of_hqd = 24, + .doorbell_size = 4, + .ih_ring_entry_size = 4 * sizeof(uint32_t), + .event_interrupt_class = &event_interrupt_class_cik, + .num_of_watch_points = 4, + .mqd_size_aligned = MQD_SIZE_ALIGNED, + .supports_cwsr = true, + .needs_iommu_device = false, + .needs_pci_atomics = true, + .num_sdma_engines = 2, + .num_sdma_queues_per_engine = 2, +}; + static const struct kfd_device_info vega10_device_info = { .asic_family = CHIP_VEGA10, .max_pasid_bits = 16, @@ -237,6 +253,22 @@ static const struct kfd_device_info vega10_vf_device_info = { .num_sdma_queues_per_engine = 2, }; +static const struct kfd_device_info vega12_device_info = { + .asic_family = CHIP_VEGA12, + .max_pasid_bits = 16, + .max_no_of_hqd = 24, + .doorbell_size = 8, + .ih_ring_entry_size = 8 * sizeof(uint32_t), + .event_interrupt_class = &event_interrupt_class_v9, + .num_of_watch_points = 4, + .mqd_size_aligned = MQD_SIZE_ALIGNED, + .supports_cwsr = true, + .needs_iommu_device = false, + .needs_pci_atomics = false, + .num_sdma_engines = 2, + .num_sdma_queues_per_engine = 2, +}; + static const struct kfd_device_info vega20_device_info = { .asic_family = CHIP_VEGA20, .max_pasid_bits = 16, @@ -331,6 +363,14 @@ static const struct kfd_deviceid supported_devices[] = { { 0x67EB, &polaris11_device_info }, /* Polaris11 */ { 0x67EF, &polaris11_device_info }, /* Polaris11 */ { 0x67FF, &polaris11_device_info }, /* Polaris11 */ + { 0x6980, &polaris12_device_info }, /* Polaris12 */ + { 0x6981, &polaris12_device_info }, /* Polaris12 */ + { 0x6985, &polaris12_device_info }, /* Polaris12 */ + { 0x6986, &polaris12_device_info }, /* Polaris12 */ + { 0x6987, &polaris12_device_info }, /* Polaris12 */ + { 0x6995, &polaris12_device_info }, /* Polaris12 */ + { 0x6997, &polaris12_device_info }, /* Polaris12 */ + { 0x699F, &polaris12_device_info }, /* Polaris12 */ { 0x6860, &vega10_device_info }, /* Vega10 */ { 0x6861, &vega10_device_info }, /* Vega10 */ { 0x6862, &vega10_device_info }, /* Vega10 */ @@ -338,12 +378,24 @@ static const struct kfd_deviceid supported_devices[] = { { 0x6864, &vega10_device_info }, /* Vega10 */ { 0x6867, &vega10_device_info }, /* Vega10 */ { 0x6868, &vega10_device_info }, /* Vega10 */ + { 0x6869, &vega10_device_info }, /* Vega10 */ + { 0x686A, &vega10_device_info }, /* Vega10 */ + { 0x686B, &vega10_device_info }, /* Vega10 */ { 0x686C, &vega10_vf_device_info }, /* Vega10 vf*/ + { 0x686D, &vega10_device_info }, /* Vega10 */ + { 0x686E, &vega10_device_info }, /* Vega10 */ + { 0x686F, &vega10_device_info }, /* Vega10 */ { 0x687F, &vega10_device_info }, /* Vega10 */ + { 0x69A0, &vega12_device_info }, /* Vega12 */ + { 0x69A1, &vega12_device_info }, /* Vega12 */ + { 0x69A2, &vega12_device_info }, /* Vega12 */ + { 0x69A3, &vega12_device_info }, /* Vega12 */ + { 0x69AF, &vega12_device_info }, /* Vega12 */ { 0x66a0, &vega20_device_info }, /* Vega20 */ { 0x66a1, &vega20_device_info }, /* Vega20 */ { 0x66a2, &vega20_device_info }, /* Vega20 */ { 0x66a3, &vega20_device_info }, /* Vega20 */ + { 0x66a4, &vega20_device_info }, /* Vega20 */ { 0x66a7, &vega20_device_info }, /* Vega20 */ { 0x66af, &vega20_device_info } /* Vega20 */ }; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index fb9d66ea13b7..8372556b52eb 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -1547,7 +1547,7 @@ static int get_wave_state(struct device_queue_manager *dqm, u32 *ctl_stack_used_size, u32 *save_area_used_size) { - struct mqd_manager *mqd; + struct mqd_manager *mqd_mgr; int r; dqm_lock(dqm); @@ -1558,19 +1558,19 @@ static int get_wave_state(struct device_queue_manager *dqm, goto dqm_unlock; } - mqd = dqm->ops.get_mqd_manager(dqm, KFD_MQD_TYPE_COMPUTE); - if (!mqd) { + mqd_mgr = dqm->ops.get_mqd_manager(dqm, KFD_MQD_TYPE_COMPUTE); + if (!mqd_mgr) { r = -ENOMEM; goto dqm_unlock; } - if (!mqd->get_wave_state) { + if (!mqd_mgr->get_wave_state) { r = -EINVAL; goto dqm_unlock; } - r = mqd->get_wave_state(mqd, q->mqd, ctl_stack, ctl_stack_used_size, - save_area_used_size); + r = mqd_mgr->get_wave_state(mqd_mgr, q->mqd, ctl_stack, + ctl_stack_used_size, save_area_used_size); dqm_unlock: dqm_unlock(dqm); @@ -1741,10 +1741,12 @@ struct device_queue_manager *device_queue_manager_init(struct kfd_dev *dev) case CHIP_FIJI: case CHIP_POLARIS10: case CHIP_POLARIS11: + case CHIP_POLARIS12: device_queue_manager_init_vi_tonga(&dqm->asic_ops); break; case CHIP_VEGA10: + case CHIP_VEGA12: case CHIP_VEGA20: case CHIP_RAVEN: device_queue_manager_init_v9(&dqm->asic_ops); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c index fd60a116be37..c3a5dcfe877a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_vi.c @@ -24,7 +24,6 @@ #include "kfd_device_queue_manager.h" #include "gca/gfx_8_0_enum.h" #include "gca/gfx_8_0_sh_mask.h" -#include "gca/gfx_8_0_enum.h" #include "oss/oss_3_0_sh_mask.h" static bool set_cache_memory_policy_vi(struct device_queue_manager *dqm, diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c index 3d66cec414af..213ea5454d11 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_flat_memory.c @@ -397,9 +397,11 @@ int kfd_init_apertures(struct kfd_process *process) case CHIP_FIJI: case CHIP_POLARIS10: case CHIP_POLARIS11: + case CHIP_POLARIS12: kfd_init_apertures_vi(pdd, id); break; case CHIP_VEGA10: + case CHIP_VEGA12: case CHIP_VEGA20: case CHIP_RAVEN: kfd_init_apertures_v9(pdd, id); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c index f836897bbf58..a85904ad0d5f 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c @@ -23,7 +23,7 @@ #include "kfd_priv.h" #include "kfd_events.h" #include "soc15_int.h" - +#include "kfd_device_queue_manager.h" static bool event_interrupt_isr_v9(struct kfd_dev *dev, const uint32_t *ih_ring_entry, @@ -39,20 +39,39 @@ static bool event_interrupt_isr_v9(struct kfd_dev *dev, vmid > dev->vm_info.last_vmid_kfd) return 0; - /* If there is no valid PASID, it's likely a firmware bug */ - pasid = SOC15_PASID_FROM_IH_ENTRY(ih_ring_entry); - if (WARN_ONCE(pasid == 0, "FW bug: No PASID in KFD interrupt")) - return 0; - source_id = SOC15_SOURCE_ID_FROM_IH_ENTRY(ih_ring_entry); client_id = SOC15_CLIENT_ID_FROM_IH_ENTRY(ih_ring_entry); + pasid = SOC15_PASID_FROM_IH_ENTRY(ih_ring_entry); + + /* This is a known issue for gfx9. Under non HWS, pasid is not set + * in the interrupt payload, so we need to find out the pasid on our + * own. + */ + if (!pasid && dev->dqm->sched_policy == KFD_SCHED_POLICY_NO_HWS) { + const uint32_t pasid_mask = 0xffff; - pr_debug("client id 0x%x, source id %d, pasid 0x%x. raw data:\n", - client_id, source_id, pasid); + *patched_flag = true; + memcpy(patched_ihre, ih_ring_entry, + dev->device_info->ih_ring_entry_size); + + pasid = dev->kfd2kgd->get_atc_vmid_pasid_mapping_pasid( + dev->kgd, vmid); + + /* Patch the pasid field */ + patched_ihre[3] = cpu_to_le32((le32_to_cpu(patched_ihre[3]) + & ~pasid_mask) | pasid); + } + + pr_debug("client id 0x%x, source id %d, vmid %d, pasid 0x%x. raw data:\n", + client_id, source_id, vmid, pasid); pr_debug("%8X, %8X, %8X, %8X, %8X, %8X, %8X, %8X.\n", data[0], data[1], data[2], data[3], data[4], data[5], data[6], data[7]); + /* If there is no valid PASID, it's likely a bug */ + if (WARN_ONCE(pasid == 0, "Bug: No PASID in KFD interrupt")) + return 0; + /* Interrupt types we care about: various signals and faults. * They will be forwarded to a work queue (see below). */ diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c index 6c31f7370193..f1596881f20a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c @@ -313,6 +313,7 @@ struct kernel_queue *kernel_queue_init(struct kfd_dev *dev, case CHIP_FIJI: case CHIP_POLARIS10: case CHIP_POLARIS11: + case CHIP_POLARIS12: kernel_queue_init_vi(&kq->ops_asic_specific); break; @@ -322,6 +323,7 @@ struct kernel_queue *kernel_queue_init(struct kfd_dev *dev, break; case CHIP_VEGA10: + case CHIP_VEGA12: case CHIP_VEGA20: case CHIP_RAVEN: kernel_queue_init_v9(&kq->ops_asic_specific); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c index 6910028010d6..aed9b9b82213 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c @@ -38,8 +38,10 @@ struct mqd_manager *mqd_manager_init(enum KFD_MQD_TYPE type, case CHIP_FIJI: case CHIP_POLARIS10: case CHIP_POLARIS11: + case CHIP_POLARIS12: return mqd_manager_init_vi_tonga(type, dev); case CHIP_VEGA10: + case CHIP_VEGA12: case CHIP_VEGA20: case CHIP_RAVEN: return mqd_manager_init_v9(type, dev); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c index c6080ed3b6a7..045a229436a0 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager.c @@ -226,9 +226,11 @@ int pm_init(struct packet_manager *pm, struct device_queue_manager *dqm) case CHIP_FIJI: case CHIP_POLARIS10: case CHIP_POLARIS11: + case CHIP_POLARIS12: pm->pmf = &kfd_vi_pm_funcs; break; case CHIP_VEGA10: + case CHIP_VEGA12: case CHIP_VEGA20: case CHIP_RAVEN: pm->pmf = &kfd_v9_pm_funcs; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index dec8e64f36bd..0689d4ccbbc0 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -793,6 +793,7 @@ struct kfd_topology_device *kfd_topology_device_by_proximity_domain( struct kfd_topology_device *kfd_topology_device_by_id(uint32_t gpu_id); struct kfd_dev *kfd_device_by_id(uint32_t gpu_id); struct kfd_dev *kfd_device_by_pci_dev(const struct pci_dev *pdev); +struct kfd_dev *kfd_device_by_kgd(const struct kgd_dev *kgd); int kfd_topology_enum_kfd_devices(uint8_t idx, struct kfd_dev **kdev); int kfd_numa_node_to_apic_id(int numa_node_id); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index c73b4ff61f99..5f5b2acedbac 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -101,7 +101,25 @@ struct kfd_dev *kfd_device_by_pci_dev(const struct pci_dev *pdev) down_read(&topology_lock); list_for_each_entry(top_dev, &topology_device_list, list) - if (top_dev->gpu->pdev == pdev) { + if (top_dev->gpu && top_dev->gpu->pdev == pdev) { + device = top_dev->gpu; + break; + } + + up_read(&topology_lock); + + return device; +} + +struct kfd_dev *kfd_device_by_kgd(const struct kgd_dev *kgd) +{ + struct kfd_topology_device *top_dev; + struct kfd_dev *device = NULL; + + down_read(&topology_lock); + + list_for_each_entry(top_dev, &topology_device_list, list) + if (top_dev->gpu && top_dev->gpu->kgd == kgd) { device = top_dev->gpu; break; } @@ -1272,12 +1290,14 @@ int kfd_topology_add_device(struct kfd_dev *gpu) case CHIP_FIJI: case CHIP_POLARIS10: case CHIP_POLARIS11: + case CHIP_POLARIS12: pr_debug("Adding doorbell packet type capability\n"); dev->node_props.capability |= ((HSA_CAP_DOORBELL_TYPE_1_0 << HSA_CAP_DOORBELL_TYPE_TOTALBITS_SHIFT) & HSA_CAP_DOORBELL_TYPE_TOTALBITS_MASK); break; case CHIP_VEGA10: + case CHIP_VEGA12: case CHIP_VEGA20: case CHIP_RAVEN: dev->node_props.capability |= ((HSA_CAP_DOORBELL_TYPE_2_0 << diff --git a/drivers/gpu/drm/amd/display/Makefile b/drivers/gpu/drm/amd/display/Makefile index c97dc9613325..cfde1568c79a 100644 --- a/drivers/gpu/drm/amd/display/Makefile +++ b/drivers/gpu/drm/amd/display/Makefile @@ -32,11 +32,12 @@ subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/modules/inc subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/modules/freesync subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/modules/color subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/modules/info_packet +subdir-ccflags-y += -I$(FULL_AMD_DISPLAY_PATH)/modules/power #TODO: remove when Timing Sync feature is complete subdir-ccflags-y += -DBUILD_FEATURE_TIMING_SYNC=0 -DAL_LIBS = amdgpu_dm dc modules/freesync modules/color modules/info_packet +DAL_LIBS = amdgpu_dm dc modules/freesync modules/color modules/info_packet modules/power AMD_DAL = $(addsuffix /Makefile, $(addprefix $(FULL_AMD_DISPLAY_PATH)/,$(DAL_LIBS))) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index d8d0b206a79c..8a626d16e8e3 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -23,6 +23,9 @@ * */ +/* The caprices of the preprocessor require that this be declared right here */ +#define CREATE_TRACE_POINTS + #include "dm_services_types.h" #include "dc.h" #include "dc/inc/core_types.h" @@ -38,7 +41,6 @@ #include "amd_shared.h" #include "amdgpu_dm_irq.h" #include "dm_helpers.h" -#include "dm_services_types.h" #include "amdgpu_dm_mst_types.h" #if defined(CONFIG_DEBUG_FS) #include "amdgpu_dm_debugfs.h" @@ -55,6 +57,7 @@ #include <drm/drmP.h> #include <drm/drm_atomic.h> +#include <drm/drm_atomic_uapi.h> #include <drm/drm_atomic_helper.h> #include <drm/drm_dp_mst_helper.h> #include <drm/drm_fb_helper.h> @@ -72,6 +75,8 @@ #endif #include "modules/inc/mod_freesync.h" +#include "modules/power/power_helpers.h" +#include "modules/inc/mod_info_packet.h" #define FIRMWARE_RAVEN_DMCU "amdgpu/raven_dmcu.bin" MODULE_FIRMWARE(FIRMWARE_RAVEN_DMCU); @@ -129,6 +134,8 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state); static int amdgpu_dm_atomic_check(struct drm_device *dev, struct drm_atomic_state *state); +static void handle_cursor_update(struct drm_plane *plane, + struct drm_plane_state *old_plane_state); @@ -324,12 +331,29 @@ static void dm_crtc_high_irq(void *interrupt_params) struct common_irq_params *irq_params = interrupt_params; struct amdgpu_device *adev = irq_params->adev; struct amdgpu_crtc *acrtc; + struct dm_crtc_state *acrtc_state; acrtc = get_crtc_by_otg_inst(adev, irq_params->irq_src - IRQ_TYPE_VBLANK); if (acrtc) { drm_crtc_handle_vblank(&acrtc->base); amdgpu_dm_crtc_handle_crc_irq(&acrtc->base); + + acrtc_state = to_dm_crtc_state(acrtc->base.state); + + if (acrtc_state->stream && + acrtc_state->vrr_params.supported && + acrtc_state->freesync_config.state == VRR_STATE_ACTIVE_VARIABLE) { + mod_freesync_handle_v_update( + adev->dm.freesync_module, + acrtc_state->stream, + &acrtc_state->vrr_params); + + dc_stream_adjust_vmin_vmax( + adev->dm.dc, + acrtc_state->stream, + &acrtc_state->vrr_params.adjust); + } } } @@ -398,6 +422,8 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) /* Zero all the fields */ memset(&init_data, 0, sizeof(init_data)); + mutex_init(&adev->dm.dc_lock); + if(amdgpu_dm_irq_init(adev)) { DRM_ERROR("amdgpu: failed to initialize DM IRQ support.\n"); goto error; @@ -512,6 +538,9 @@ static void amdgpu_dm_fini(struct amdgpu_device *adev) /* DC Destroy TODO: Replace destroy DAL */ if (adev->dm.dc) dc_destroy(&adev->dm.dc); + + mutex_destroy(&adev->dm.dc_lock); + return; } @@ -643,6 +672,26 @@ static int dm_late_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; + struct dmcu_iram_parameters params; + unsigned int linear_lut[16]; + int i; + struct dmcu *dmcu = adev->dm.dc->res_pool->dmcu; + bool ret; + + for (i = 0; i < 16; i++) + linear_lut[i] = 0xFFFF * i / 15; + + params.set = 0; + params.backlight_ramping_start = 0xCCCC; + params.backlight_ramping_reduction = 0xCCCCCCCC; + params.backlight_lut_array_size = 16; + params.backlight_lut_array = linear_lut; + + ret = dmcu_load_iram(dmcu, params); + + if (!ret) + return -EINVAL; + return detect_mst_link_for_all_connectors(adev->ddev); } @@ -969,45 +1018,6 @@ const struct amdgpu_ip_block_version dm_ip_block = }; -static struct drm_atomic_state * -dm_atomic_state_alloc(struct drm_device *dev) -{ - struct dm_atomic_state *state = kzalloc(sizeof(*state), GFP_KERNEL); - - if (!state) - return NULL; - - if (drm_atomic_state_init(dev, &state->base) < 0) - goto fail; - - return &state->base; - -fail: - kfree(state); - return NULL; -} - -static void -dm_atomic_state_clear(struct drm_atomic_state *state) -{ - struct dm_atomic_state *dm_state = to_dm_atomic_state(state); - - if (dm_state->context) { - dc_release_state(dm_state->context); - dm_state->context = NULL; - } - - drm_atomic_state_default_clear(state); -} - -static void -dm_atomic_state_alloc_free(struct drm_atomic_state *state) -{ - struct dm_atomic_state *dm_state = to_dm_atomic_state(state); - drm_atomic_state_default_release(state); - kfree(dm_state); -} - /** * DOC: atomic * @@ -1019,9 +1029,6 @@ static const struct drm_mode_config_funcs amdgpu_dm_mode_funcs = { .output_poll_changed = drm_fb_helper_output_poll_changed, .atomic_check = amdgpu_dm_atomic_check, .atomic_commit = amdgpu_dm_atomic_commit, - .atomic_state_alloc = dm_atomic_state_alloc, - .atomic_state_clear = dm_atomic_state_clear, - .atomic_state_free = dm_atomic_state_alloc_free }; static struct drm_mode_config_helper_funcs amdgpu_dm_mode_config_helperfuncs = { @@ -1543,8 +1550,117 @@ static int dcn10_register_irq_handlers(struct amdgpu_device *adev) } #endif +/* + * Acquires the lock for the atomic state object and returns + * the new atomic state. + * + * This should only be called during atomic check. + */ +static int dm_atomic_get_state(struct drm_atomic_state *state, + struct dm_atomic_state **dm_state) +{ + struct drm_device *dev = state->dev; + struct amdgpu_device *adev = dev->dev_private; + struct amdgpu_display_manager *dm = &adev->dm; + struct drm_private_state *priv_state; + int ret; + + if (*dm_state) + return 0; + + ret = drm_modeset_lock(&dm->atomic_obj_lock, state->acquire_ctx); + if (ret) + return ret; + + priv_state = drm_atomic_get_private_obj_state(state, &dm->atomic_obj); + if (IS_ERR(priv_state)) + return PTR_ERR(priv_state); + + *dm_state = to_dm_atomic_state(priv_state); + + return 0; +} + +struct dm_atomic_state * +dm_atomic_get_new_state(struct drm_atomic_state *state) +{ + struct drm_device *dev = state->dev; + struct amdgpu_device *adev = dev->dev_private; + struct amdgpu_display_manager *dm = &adev->dm; + struct drm_private_obj *obj; + struct drm_private_state *new_obj_state; + int i; + + for_each_new_private_obj_in_state(state, obj, new_obj_state, i) { + if (obj->funcs == dm->atomic_obj.funcs) + return to_dm_atomic_state(new_obj_state); + } + + return NULL; +} + +struct dm_atomic_state * +dm_atomic_get_old_state(struct drm_atomic_state *state) +{ + struct drm_device *dev = state->dev; + struct amdgpu_device *adev = dev->dev_private; + struct amdgpu_display_manager *dm = &adev->dm; + struct drm_private_obj *obj; + struct drm_private_state *old_obj_state; + int i; + + for_each_old_private_obj_in_state(state, obj, old_obj_state, i) { + if (obj->funcs == dm->atomic_obj.funcs) + return to_dm_atomic_state(old_obj_state); + } + + return NULL; +} + +static struct drm_private_state * +dm_atomic_duplicate_state(struct drm_private_obj *obj) +{ + struct dm_atomic_state *old_state, *new_state; + + new_state = kzalloc(sizeof(*new_state), GFP_KERNEL); + if (!new_state) + return NULL; + + __drm_atomic_helper_private_obj_duplicate_state(obj, &new_state->base); + + new_state->context = dc_create_state(); + if (!new_state->context) { + kfree(new_state); + return NULL; + } + + old_state = to_dm_atomic_state(obj->state); + if (old_state && old_state->context) + dc_resource_state_copy_construct(old_state->context, + new_state->context); + + return &new_state->base; +} + +static void dm_atomic_destroy_state(struct drm_private_obj *obj, + struct drm_private_state *state) +{ + struct dm_atomic_state *dm_state = to_dm_atomic_state(state); + + if (dm_state && dm_state->context) + dc_release_state(dm_state->context); + + kfree(dm_state); +} + +static struct drm_private_state_funcs dm_atomic_state_funcs = { + .atomic_duplicate_state = dm_atomic_duplicate_state, + .atomic_destroy_state = dm_atomic_destroy_state, +}; + static int amdgpu_dm_mode_config_init(struct amdgpu_device *adev) { + struct dm_atomic_state *state; int r; adev->mode_info.mode_config_initialized = true; @@ -1562,6 +1678,25 @@ static int amdgpu_dm_mode_config_init(struct amdgpu_device *adev) adev->ddev->mode_config.fb_base = adev->gmc.aper_base; + drm_modeset_lock_init(&adev->dm.atomic_obj_lock); + + state = kzalloc(sizeof(*state), GFP_KERNEL); + if (!state) + return -ENOMEM; + + state->context = dc_create_state(); + if (!state->context) { + kfree(state); + return -ENOMEM; + } + + dc_resource_state_copy_construct_current(adev->dm.dc, state->context); + + drm_atomic_private_obj_init(adev->ddev, + &adev->dm.atomic_obj, + &state->base, + &dm_atomic_state_funcs); + r = amdgpu_display_modeset_create_props(adev); if (r) return r; @@ -1569,27 +1704,60 @@ static int amdgpu_dm_mode_config_init(struct amdgpu_device *adev) return 0; } +#define AMDGPU_DM_DEFAULT_MIN_BACKLIGHT 12 +#define AMDGPU_DM_DEFAULT_MAX_BACKLIGHT 255 + #if defined(CONFIG_BACKLIGHT_CLASS_DEVICE) ||\ defined(CONFIG_BACKLIGHT_CLASS_DEVICE_MODULE) +static void amdgpu_dm_update_backlight_caps(struct amdgpu_display_manager *dm) +{ +#if defined(CONFIG_ACPI) + struct amdgpu_dm_backlight_caps caps; + + if (dm->backlight_caps.caps_valid) + return; + + amdgpu_acpi_get_backlight_caps(dm->adev, &caps); + if (caps.caps_valid) { + dm->backlight_caps.min_input_signal = caps.min_input_signal; + dm->backlight_caps.max_input_signal = caps.max_input_signal; + dm->backlight_caps.caps_valid = true; + } else { + dm->backlight_caps.min_input_signal = + AMDGPU_DM_DEFAULT_MIN_BACKLIGHT; + dm->backlight_caps.max_input_signal = + AMDGPU_DM_DEFAULT_MAX_BACKLIGHT; + } +#else + dm->backlight_caps.min_input_signal = AMDGPU_DM_DEFAULT_MIN_BACKLIGHT; + dm->backlight_caps.max_input_signal = AMDGPU_DM_DEFAULT_MAX_BACKLIGHT; +#endif +} + static int amdgpu_dm_backlight_update_status(struct backlight_device *bd) { struct amdgpu_display_manager *dm = bl_get_data(bd); + struct amdgpu_dm_backlight_caps caps; + uint32_t brightness = bd->props.brightness; - /* backlight_pwm_u16_16 parameter is in unsigned 32 bit, 16 bit integer - * and 16 bit fractional, where 1.0 is max backlight value. - * bd->props.brightness is 8 bit format and needs to be converted by - * scaling via copy lower byte to upper byte of 16 bit value. - */ - uint32_t brightness = bd->props.brightness * 0x101; - + amdgpu_dm_update_backlight_caps(dm); + caps = dm->backlight_caps; /* - * PWM interperts 0 as 100% rather than 0% because of HW - * limitation for level 0. So limiting minimum brightness level - * to 1. + * The brightness input is in the range 0-255 + * It needs to be rescaled to be between the + * requested min and max input signal + * + * It also needs to be scaled up by 0x101 to + * match the DC interface which has a range of + * 0 to 0xffff */ - if (bd->props.brightness < 1) - brightness = 0x101; + brightness = + brightness + * 0x101 + * (caps.max_input_signal - caps.min_input_signal) + / AMDGPU_MAX_BL_LEVEL + + caps.min_input_signal * 0x101; if (dc_link_set_backlight_level(dm->backlight_link, brightness, 0, 0)) @@ -1619,6 +1787,8 @@ amdgpu_dm_register_backlight_device(struct amdgpu_display_manager *dm) char bl_name[16]; struct backlight_properties props = { 0 }; + amdgpu_dm_update_backlight_caps(dm); + props.max_brightness = AMDGPU_MAX_BL_LEVEL; props.brightness = AMDGPU_MAX_BL_LEVEL; props.type = BACKLIGHT_RAW; @@ -1850,6 +2020,7 @@ fail: static void amdgpu_dm_destroy_drm_device(struct amdgpu_display_manager *dm) { drm_mode_config_cleanup(dm->ddev); + drm_atomic_private_obj_fini(&dm->atomic_obj); return; } @@ -1869,73 +2040,6 @@ static void dm_bandwidth_update(struct amdgpu_device *adev) /* TODO: implement later */ } -static int amdgpu_notify_freesync(struct drm_device *dev, void *data, - struct drm_file *filp) -{ - struct drm_atomic_state *state; - struct drm_modeset_acquire_ctx ctx; - struct drm_crtc *crtc; - struct drm_connector *connector; - struct drm_connector_state *old_con_state, *new_con_state; - int ret = 0; - uint8_t i; - bool enable = false; - - drm_modeset_acquire_init(&ctx, 0); - - state = drm_atomic_state_alloc(dev); - if (!state) { - ret = -ENOMEM; - goto out; - } - state->acquire_ctx = &ctx; - -retry: - drm_for_each_crtc(crtc, dev) { - ret = drm_atomic_add_affected_connectors(state, crtc); - if (ret) - goto fail; - - /* TODO rework amdgpu_dm_commit_planes so we don't need this */ - ret = drm_atomic_add_affected_planes(state, crtc); - if (ret) - goto fail; - } - - for_each_oldnew_connector_in_state(state, connector, old_con_state, new_con_state, i) { - struct dm_connector_state *dm_new_con_state = to_dm_connector_state(new_con_state); - struct drm_crtc_state *new_crtc_state; - struct amdgpu_crtc *acrtc = to_amdgpu_crtc(dm_new_con_state->base.crtc); - struct dm_crtc_state *dm_new_crtc_state; - - if (!acrtc) { - ASSERT(0); - continue; - } - - new_crtc_state = drm_atomic_get_new_crtc_state(state, &acrtc->base); - dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); - - dm_new_crtc_state->freesync_enabled = enable; - } - - ret = drm_atomic_commit(state); - -fail: - if (ret == -EDEADLK) { - drm_atomic_state_clear(state); - drm_modeset_backoff(&ctx); - goto retry; - } - - drm_atomic_state_put(state); - -out: - drm_modeset_drop_locks(&ctx); - drm_modeset_acquire_fini(&ctx); - return ret; -} - static const struct amdgpu_display_funcs dm_display_funcs = { .bandwidth_update = dm_bandwidth_update, /* called unconditionally */ .vblank_get_counter = dm_vblank_get_counter,/* called unconditionally */ @@ -1948,8 +2052,6 @@ static const struct amdgpu_display_funcs dm_display_funcs = { dm_crtc_get_scanoutpos,/* called unconditionally */ .add_encoder = NULL, /* VBIOS parsing. DAL does it. */ .add_connector = NULL, /* VBIOS parsing. DAL does it. */ - .notify_freesync = amdgpu_notify_freesync, - }; #if defined(CONFIG_DEBUG_KERNEL_DC) @@ -2550,7 +2652,8 @@ static void adjust_colour_depth_from_display_info(struct dc_crtc_timing *timing_ static void fill_stream_properties_from_drm_display_mode(struct dc_stream_state *stream, const struct drm_display_mode *mode_in, - const struct drm_connector *connector) + const struct drm_connector *connector, + const struct dc_stream_state *old_stream) { struct dc_crtc_timing *timing_out = &stream->timing; const struct drm_display_info *info = &connector->display_info; @@ -2576,7 +2679,18 @@ fill_stream_properties_from_drm_display_mode(struct dc_stream_state *stream, connector); timing_out->scan_type = SCANNING_TYPE_NODATA; timing_out->hdmi_vic = 0; - timing_out->vic = drm_match_cea_mode(mode_in); + + if(old_stream) { + timing_out->vic = old_stream->timing.vic; + timing_out->flags.HSYNC_POSITIVE_POLARITY = old_stream->timing.flags.HSYNC_POSITIVE_POLARITY; + timing_out->flags.VSYNC_POSITIVE_POLARITY = old_stream->timing.flags.VSYNC_POSITIVE_POLARITY; + } else { + timing_out->vic = drm_match_cea_mode(mode_in); + if (mode_in->flags & DRM_MODE_FLAG_PHSYNC) + timing_out->flags.HSYNC_POSITIVE_POLARITY = 1; + if (mode_in->flags & DRM_MODE_FLAG_PVSYNC) + timing_out->flags.VSYNC_POSITIVE_POLARITY = 1; + } timing_out->h_addressable = mode_in->crtc_hdisplay; timing_out->h_total = mode_in->crtc_htotal; @@ -2592,10 +2706,6 @@ fill_stream_properties_from_drm_display_mode(struct dc_stream_state *stream, mode_in->crtc_vsync_end - mode_in->crtc_vsync_start; timing_out->pix_clk_khz = mode_in->crtc_clock; timing_out->aspect_ratio = get_aspect_ratio(mode_in); - if (mode_in->flags & DRM_MODE_FLAG_PHSYNC) - timing_out->flags.HSYNC_POSITIVE_POLARITY = 1; - if (mode_in->flags & DRM_MODE_FLAG_PVSYNC) - timing_out->flags.VSYNC_POSITIVE_POLARITY = 1; stream->output_color_space = get_output_color_space(timing_out); @@ -2618,9 +2728,9 @@ static void fill_audio_info(struct audio_info *audio_info, cea_revision = drm_connector->display_info.cea_rev; - strncpy(audio_info->display_name, + strscpy(audio_info->display_name, edid_caps->display_name, - AUDIO_INFO_DISPLAY_NAME_SIZE_IN_CHARS - 1); + AUDIO_INFO_DISPLAY_NAME_SIZE_IN_CHARS); if (cea_revision >= 3) { audio_info->mode_count = edid_caps->audio_mode_count; @@ -2758,13 +2868,18 @@ static void dm_enable_per_frame_crtc_master_sync(struct dc_state *context) static struct dc_stream_state * create_stream_for_sink(struct amdgpu_dm_connector *aconnector, const struct drm_display_mode *drm_mode, - const struct dm_connector_state *dm_state) + const struct dm_connector_state *dm_state, + const struct dc_stream_state *old_stream) { struct drm_display_mode *preferred_mode = NULL; struct drm_connector *drm_connector; struct dc_stream_state *stream = NULL; struct drm_display_mode mode = *drm_mode; bool native_mode_found = false; + bool scale = dm_state ? (dm_state->scaling != RMX_OFF) : false; + int mode_refresh; + int preferred_refresh = 0; + struct dc_sink *sink = NULL; if (aconnector == NULL) { DRM_ERROR("aconnector is NULL!\n"); @@ -2803,6 +2918,8 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, struct drm_display_mode, head); + mode_refresh = drm_mode_vrefresh(&mode); + if (preferred_mode == NULL) { /* * This may not be an error, the use case is when we have no @@ -2815,13 +2932,23 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, decide_crtc_timing_for_drm_display_mode( &mode, preferred_mode, dm_state ? (dm_state->scaling != RMX_OFF) : false); + preferred_refresh = drm_mode_vrefresh(preferred_mode); } if (!dm_state) drm_mode_set_crtcinfo(&mode, 0); - fill_stream_properties_from_drm_display_mode(stream, - &mode, &aconnector->base); + /* + * If scaling is enabled and refresh rate didn't change + * we copy the vic and polarities of the old timings + */ + if (!scale || mode_refresh != preferred_refresh) + fill_stream_properties_from_drm_display_mode(stream, + &mode, &aconnector->base, NULL); + else + fill_stream_properties_from_drm_display_mode(stream, + &mode, &aconnector->base, old_stream); + update_stream_scaling_settings(&mode, dm_state, stream); fill_audio_info( @@ -2833,6 +2960,7 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector, if (dm_state && dm_state->freesync_capable) stream->ignore_msa_timing_param = true; + finish: if (sink && sink->sink_signal == SIGNAL_TYPE_VIRTUAL && aconnector->base.force != DRM_FORCE_ON) dc_sink_release(sink); @@ -2899,9 +3027,12 @@ dm_crtc_duplicate_state(struct drm_crtc *crtc) dc_stream_retain(state->stream); } - state->adjust = cur->adjust; + state->vrr_params = cur->vrr_params; state->vrr_infopacket = cur->vrr_infopacket; - state->freesync_enabled = cur->freesync_enabled; + state->abm_level = cur->abm_level; + state->vrr_supported = cur->vrr_supported; + state->freesync_config = cur->freesync_config; + state->crc_enabled = cur->crc_enabled; /* TODO Duplicate dc_stream after objects are stream object is flattened */ @@ -3017,6 +3148,9 @@ int amdgpu_dm_connector_atomic_set_property(struct drm_connector *connector, } else if (property == adev->mode_info.max_bpc_property) { dm_new_state->max_bpc = val; ret = 0; + } else if (property == adev->mode_info.abm_level_property) { + dm_new_state->abm_level = val; + ret = 0; } return ret; @@ -3062,7 +3196,11 @@ int amdgpu_dm_connector_atomic_get_property(struct drm_connector *connector, } else if (property == adev->mode_info.max_bpc_property) { *val = dm_state->max_bpc; ret = 0; + } else if (property == adev->mode_info.abm_level_property) { + *val = dm_state->abm_level; + ret = 0; } + return ret; } @@ -3106,6 +3244,7 @@ void amdgpu_dm_connector_funcs_reset(struct drm_connector *connector) state->underscan_enable = false; state->underscan_hborder = 0; state->underscan_vborder = 0; + state->max_bpc = 8; __drm_atomic_helper_connector_reset(connector, &state->base); } @@ -3126,7 +3265,12 @@ amdgpu_dm_connector_atomic_duplicate_state(struct drm_connector *connector) __drm_atomic_helper_connector_duplicate_state(connector, &new_state->base); new_state->freesync_capable = state->freesync_capable; - new_state->freesync_enable = state->freesync_enable; + new_state->abm_level = state->abm_level; + new_state->scaling = state->scaling; + new_state->underscan_enable = state->underscan_enable; + new_state->underscan_hborder = state->underscan_hborder; + new_state->underscan_vborder = state->underscan_vborder; + new_state->max_bpc = state->max_bpc; return &new_state->base; } @@ -3228,7 +3372,7 @@ enum drm_mode_status amdgpu_dm_connector_mode_valid(struct drm_connector *connec goto fail; } - stream = create_stream_for_sink(aconnector, mode, NULL); + stream = create_stream_for_sink(aconnector, mode, NULL, NULL); if (stream == NULL) { DRM_ERROR("Failed to create stream for sink!\n"); goto fail; @@ -3499,10 +3643,53 @@ static int dm_plane_atomic_check(struct drm_plane *plane, return -EINVAL; } +static int dm_plane_atomic_async_check(struct drm_plane *plane, + struct drm_plane_state *new_plane_state) +{ + struct drm_plane_state *old_plane_state = + drm_atomic_get_old_plane_state(new_plane_state->state, plane); + + /* Only support async updates on cursor planes. */ + if (plane->type != DRM_PLANE_TYPE_CURSOR) + return -EINVAL; + + /* + * DRM calls prepare_fb and cleanup_fb on new_plane_state for + * async commits so don't allow fb changes. + */ + if (old_plane_state->fb != new_plane_state->fb) + return -EINVAL; + + return 0; +} + +static void dm_plane_atomic_async_update(struct drm_plane *plane, + struct drm_plane_state *new_state) +{ + struct drm_plane_state *old_state = + drm_atomic_get_old_plane_state(new_state->state, plane); + + if (plane->state->fb != new_state->fb) + drm_atomic_set_fb_for_plane(plane->state, new_state->fb); + + plane->state->src_x = new_state->src_x; + plane->state->src_y = new_state->src_y; + plane->state->src_w = new_state->src_w; + plane->state->src_h = new_state->src_h; + plane->state->crtc_x = new_state->crtc_x; + plane->state->crtc_y = new_state->crtc_y; + plane->state->crtc_w = new_state->crtc_w; + plane->state->crtc_h = new_state->crtc_h; + + handle_cursor_update(plane, old_state); +} + static const struct drm_plane_helper_funcs dm_plane_helper_funcs = { .prepare_fb = dm_plane_helper_prepare_fb, .cleanup_fb = dm_plane_helper_cleanup_fb, .atomic_check = dm_plane_atomic_check, + .atomic_async_check = dm_plane_atomic_async_check, + .atomic_async_update = dm_plane_atomic_async_update }; /* @@ -3716,7 +3903,7 @@ amdgpu_dm_create_common_mode(struct drm_encoder *encoder, mode->hdisplay = hdisplay; mode->vdisplay = vdisplay; mode->type &= ~DRM_MODE_TYPE_PREFERRED; - strncpy(mode->name, name, DRM_DISPLAY_MODE_LEN); + strscpy(mode->name, name, DRM_DISPLAY_MODE_LEN); return mode; @@ -3876,6 +4063,17 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm, adev->mode_info.max_bpc_property, 0); + if (connector_type == DRM_MODE_CONNECTOR_eDP && + dc_is_dmcu_initialized(adev->dm.dc)) { + drm_object_attach_property(&aconnector->base.base, + adev->mode_info.abm_level_property, 0); + } + + if (connector_type == DRM_MODE_CONNECTOR_HDMIA || + connector_type == DRM_MODE_CONNECTOR_DisplayPort) { + drm_connector_attach_vrr_capable_property( + &aconnector->base); + } } static int amdgpu_dm_i2c_xfer(struct i2c_adapter *i2c_adap, @@ -4180,6 +4378,7 @@ static int get_cursor_position(struct drm_plane *plane, struct drm_crtc *crtc, static void handle_cursor_update(struct drm_plane *plane, struct drm_plane_state *old_plane_state) { + struct amdgpu_device *adev = plane->dev->dev_private; struct amdgpu_framebuffer *afb = to_amdgpu_framebuffer(plane->state->fb); struct drm_crtc *crtc = afb ? plane->state->crtc : old_plane_state->crtc; struct dm_crtc_state *crtc_state = crtc ? to_dm_crtc_state(crtc->state) : NULL; @@ -4204,9 +4403,12 @@ static void handle_cursor_update(struct drm_plane *plane, if (!position.enable) { /* turn off cursor */ - if (crtc_state && crtc_state->stream) + if (crtc_state && crtc_state->stream) { + mutex_lock(&adev->dm.dc_lock); dc_stream_set_cursor_position(crtc_state->stream, &position); + mutex_unlock(&adev->dm.dc_lock); + } return; } @@ -4224,6 +4426,7 @@ static void handle_cursor_update(struct drm_plane *plane, attributes.pitch = attributes.width; if (crtc_state->stream) { + mutex_lock(&adev->dm.dc_lock); if (!dc_stream_set_cursor_attributes(crtc_state->stream, &attributes)) DRM_ERROR("DC failed to set cursor attributes\n"); @@ -4231,6 +4434,7 @@ static void handle_cursor_update(struct drm_plane *plane, if (!dc_stream_set_cursor_position(crtc_state->stream, &position)) DRM_ERROR("DC failed to set cursor position\n"); + mutex_unlock(&adev->dm.dc_lock); } } @@ -4252,6 +4456,102 @@ static void prepare_flip_isr(struct amdgpu_crtc *acrtc) acrtc->crtc_id); } +struct dc_stream_status *dc_state_get_stream_status( + struct dc_state *state, + struct dc_stream_state *stream) +{ + uint8_t i; + + for (i = 0; i < state->stream_count; i++) { + if (stream == state->streams[i]) + return &state->stream_status[i]; + } + + return NULL; +} + +static void update_freesync_state_on_stream( + struct amdgpu_display_manager *dm, + struct dm_crtc_state *new_crtc_state, + struct dc_stream_state *new_stream, + struct dc_plane_state *surface, + u32 flip_timestamp_in_us) +{ + struct mod_vrr_params vrr_params = new_crtc_state->vrr_params; + struct dc_info_packet vrr_infopacket = {0}; + struct mod_freesync_config config = new_crtc_state->freesync_config; + + if (!new_stream) + return; + + /* + * TODO: Determine why min/max totals and vrefresh can be 0 here. + * For now it's sufficient to just guard against these conditions. + */ + + if (!new_stream->timing.h_total || !new_stream->timing.v_total) + return; + + if (new_crtc_state->vrr_supported && + config.min_refresh_in_uhz && + config.max_refresh_in_uhz) { + config.state = new_crtc_state->base.vrr_enabled ? + VRR_STATE_ACTIVE_VARIABLE : + VRR_STATE_INACTIVE; + } else { + config.state = VRR_STATE_UNSUPPORTED; + } + + mod_freesync_build_vrr_params(dm->freesync_module, + new_stream, + &config, &vrr_params); + + if (surface) { + mod_freesync_handle_preflip( + dm->freesync_module, + surface, + new_stream, + flip_timestamp_in_us, + &vrr_params); + } + + mod_freesync_build_vrr_infopacket( + dm->freesync_module, + new_stream, + &vrr_params, + PACKET_TYPE_VRR, + TRANSFER_FUNC_UNKNOWN, + &vrr_infopacket); + + new_crtc_state->freesync_timing_changed = + (memcmp(&new_crtc_state->vrr_params.adjust, + &vrr_params.adjust, + sizeof(vrr_params.adjust)) != 0); + + new_crtc_state->freesync_vrr_info_changed = + (memcmp(&new_crtc_state->vrr_infopacket, + &vrr_infopacket, + sizeof(vrr_infopacket)) != 0); + + new_crtc_state->vrr_params = vrr_params; + new_crtc_state->vrr_infopacket = vrr_infopacket; + + new_stream->adjust = new_crtc_state->vrr_params.adjust; + new_stream->vrr_infopacket = vrr_infopacket; + + if (new_crtc_state->freesync_vrr_info_changed) + DRM_DEBUG_KMS("VRR packet update: crtc=%u enabled=%d state=%d", + new_crtc_state->base.crtc->base.id, + (int)new_crtc_state->base.vrr_enabled, + (int)vrr_params.state); + + if (new_crtc_state->freesync_timing_changed) + DRM_DEBUG_KMS("VRR timing update: crtc=%u min=%u max=%u\n", + new_crtc_state->base.crtc->base.id, + vrr_params.adjust.v_total_min, + vrr_params.adjust.v_total_max); +} + /* * Executes flip * @@ -4263,6 +4563,7 @@ static void amdgpu_dm_do_flip(struct drm_crtc *crtc, struct dc_state *state) { unsigned long flags; + uint64_t timestamp_ns; uint32_t target_vblank; int r, vpos, hpos; struct amdgpu_crtc *acrtc = to_amdgpu_crtc(crtc); @@ -4273,8 +4574,10 @@ static void amdgpu_dm_do_flip(struct drm_crtc *crtc, struct dc_flip_addrs addr = { {0} }; /* TODO eliminate or rename surface_update */ struct dc_surface_update surface_updates[1] = { {0} }; + struct dc_stream_update stream_update = {0}; struct dm_crtc_state *acrtc_state = to_dm_crtc_state(crtc->state); struct dc_stream_status *stream_status; + struct dc_plane_state *surface; /* Prepare wait for target vblank early - before the fence-waits */ @@ -4324,6 +4627,9 @@ static void amdgpu_dm_do_flip(struct drm_crtc *crtc, addr.address.grph.addr.high_part = upper_32_bits(afb->address); addr.flip_immediate = async_flip; + timestamp_ns = ktime_get_ns(); + addr.flip_timestamp_in_us = div_u64(timestamp_ns, 1000); + if (acrtc->base.state->event) prepare_flip_isr(acrtc); @@ -4337,21 +4643,51 @@ static void amdgpu_dm_do_flip(struct drm_crtc *crtc, return; } - surface_updates->surface = stream_status->plane_states[0]; - if (!surface_updates->surface) { + surface = stream_status->plane_states[0]; + surface_updates->surface = surface; + + if (!surface) { DRM_ERROR("No surface for CRTC: id=%d\n", acrtc->crtc_id); return; } surface_updates->flip_addr = &addr; + if (acrtc_state->stream) { + update_freesync_state_on_stream( + &adev->dm, + acrtc_state, + acrtc_state->stream, + surface, + addr.flip_timestamp_in_us); + + if (acrtc_state->freesync_timing_changed) + stream_update.adjust = + &acrtc_state->stream->adjust; + + if (acrtc_state->freesync_vrr_info_changed) + stream_update.vrr_infopacket = + &acrtc_state->stream->vrr_infopacket; + } + + /* Update surface timing information. */ + surface->time.time_elapsed_in_us[surface->time.index] = + addr.flip_timestamp_in_us - surface->time.prev_update_time_in_us; + surface->time.prev_update_time_in_us = addr.flip_timestamp_in_us; + surface->time.index++; + if (surface->time.index >= DC_PLANE_UPDATE_TIMES_MAX) + surface->time.index = 0; + + mutex_lock(&adev->dm.dc_lock); + dc_commit_updates_for_stream(adev->dm.dc, surface_updates, 1, acrtc_state->stream, - NULL, + &stream_update, &surface_updates->surface, state); + mutex_unlock(&adev->dm.dc_lock); DRM_DEBUG_DRIVER("%s Flipping to hi: 0x%x, low: 0x%x \n", __func__, @@ -4366,6 +4702,7 @@ static void amdgpu_dm_do_flip(struct drm_crtc *crtc, * with a dc_plane_state and follow the atomic model a bit more closely here. */ static bool commit_planes_to_stream( + struct amdgpu_display_manager *dm, struct dc *dc, struct dc_plane_state **plane_states, uint8_t new_plane_count, @@ -4382,6 +4719,7 @@ static bool commit_planes_to_stream( struct dc_stream_state *dc_stream = dm_new_crtc_state->stream; struct dc_stream_update *stream_update = kzalloc(sizeof(struct dc_stream_update), GFP_KERNEL); + unsigned int abm_level; if (!stream_update) { BREAK_TO_DEBUGGER(); @@ -4409,9 +4747,9 @@ static bool commit_planes_to_stream( stream_update->dst = dc_stream->dst; stream_update->out_transfer_func = dc_stream->out_transfer_func; - if (dm_new_crtc_state->freesync_enabled != dm_old_crtc_state->freesync_enabled) { - stream_update->vrr_infopacket = &dc_stream->vrr_infopacket; - stream_update->adjust = &dc_stream->adjust; + if (dm_new_crtc_state->abm_level != dm_old_crtc_state->abm_level) { + abm_level = dm_new_crtc_state->abm_level; + stream_update->abm_level = &abm_level; } for (i = 0; i < new_plane_count; i++) { @@ -4441,11 +4779,13 @@ static bool commit_planes_to_stream( updates[i].scaling_info = &scaling_info[i]; } + mutex_lock(&dm->dc_lock); dc_commit_updates_for_stream( dc, updates, new_plane_count, dc_stream, stream_update, plane_states, state); + mutex_unlock(&dm->dc_lock); kfree(flip_addr); kfree(plane_info); @@ -4455,6 +4795,7 @@ static bool commit_planes_to_stream( } static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, + struct dc_state *dc_state, struct drm_device *dev, struct amdgpu_display_manager *dm, struct drm_crtc *pcrtc, @@ -4471,7 +4812,6 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, struct dm_crtc_state *acrtc_state = to_dm_crtc_state(new_pcrtc_state); struct dm_crtc_state *dm_old_crtc_state = to_dm_crtc_state(drm_atomic_get_old_crtc_state(state, pcrtc)); - struct dm_atomic_state *dm_state = to_dm_atomic_state(state); int planes_count = 0; unsigned long flags; @@ -4532,7 +4872,7 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, crtc, fb, (uint32_t)drm_crtc_vblank_count(crtc) + *wait_for_vblank, - dm_state->context); + dc_state); } } @@ -4549,15 +4889,15 @@ static void amdgpu_dm_commit_planes(struct drm_atomic_state *state, spin_unlock_irqrestore(&pcrtc->dev->event_lock, flags); } - dc_stream_attach->adjust = acrtc_state->adjust; - dc_stream_attach->vrr_infopacket = acrtc_state->vrr_infopacket; + dc_stream_attach->abm_level = acrtc_state->abm_level; - if (false == commit_planes_to_stream(dm->dc, + if (false == commit_planes_to_stream(dm, + dm->dc, plane_states_constructed, planes_count, acrtc_state, dm_old_crtc_state, - dm_state->context)) + dc_state)) dm_error("%s: Failed to attach plane!\n", __func__); } else { /*TODO BUG Here should go disable planes on CRTC. */ @@ -4625,6 +4965,7 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) struct amdgpu_device *adev = dev->dev_private; struct amdgpu_display_manager *dm = &adev->dm; struct dm_atomic_state *dm_state; + struct dc_state *dc_state = NULL, *dc_state_temp = NULL; uint32_t i, j; struct drm_crtc *crtc; struct drm_crtc_state *old_crtc_state, *new_crtc_state; @@ -4637,7 +4978,16 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) drm_atomic_helper_update_legacy_modeset_state(dev, state); - dm_state = to_dm_atomic_state(state); + dm_state = dm_atomic_get_new_state(state); + if (dm_state && dm_state->context) { + dc_state = dm_state->context; + } else { + /* No state changes, retain current state. */ + dc_state_temp = dc_create_state(); + ASSERT(dc_state_temp); + dc_state = dc_state_temp; + dc_resource_state_copy_construct_current(dm->dc, dc_state); + } /* update changed items */ for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { @@ -4710,9 +5060,11 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) } } /* for_each_crtc_in_state() */ - if (dm_state->context) { - dm_enable_per_frame_crtc_master_sync(dm_state->context); - WARN_ON(!dc_commit_state(dm->dc, dm_state->context)); + if (dc_state) { + dm_enable_per_frame_crtc_master_sync(dc_state); + mutex_lock(&dm->dc_lock); + WARN_ON(!dc_commit_state(dm->dc, dc_state)); + mutex_unlock(&dm->dc_lock); } for_each_new_crtc_in_state(state, crtc, new_crtc_state, i) { @@ -4725,13 +5077,17 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) dc_stream_get_status(dm_new_crtc_state->stream); if (!status) + status = dc_state_get_stream_status(dc_state, + dm_new_crtc_state->stream); + + if (!status) DC_ERR("got no status for stream %p on acrtc%p\n", dm_new_crtc_state->stream, acrtc); else acrtc->otg_inst = status->primary_otg_inst; } } - /* Handle scaling and underscan changes*/ + /* Handle scaling, underscan, and abm changes*/ for_each_oldnew_connector_in_state(state, connector, old_con_state, new_con_state, i) { struct dm_connector_state *dm_new_con_state = to_dm_connector_state(new_con_state); struct dm_connector_state *dm_old_con_state = to_dm_connector_state(old_con_state); @@ -4747,11 +5103,14 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) if (!acrtc || drm_atomic_crtc_needs_modeset(new_crtc_state)) continue; - /* Skip anything that is not scaling or underscan changes */ - if (!is_scaling_state_different(dm_new_con_state, dm_old_con_state)) - continue; dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); + dm_old_crtc_state = to_dm_crtc_state(old_crtc_state); + + /* Skip anything that is not scaling or underscan changes */ + if (!is_scaling_state_different(dm_new_con_state, dm_old_con_state) && + (dm_new_crtc_state->abm_level == dm_old_crtc_state->abm_level)) + continue; update_stream_scaling_settings(&dm_new_con_state->base.crtc->mode, dm_new_con_state, (struct dc_stream_state *)dm_new_crtc_state->stream); @@ -4763,17 +5122,17 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) WARN_ON(!status); WARN_ON(!status->plane_count); - dm_new_crtc_state->stream->adjust = dm_new_crtc_state->adjust; - dm_new_crtc_state->stream->vrr_infopacket = dm_new_crtc_state->vrr_infopacket; + dm_new_crtc_state->stream->abm_level = dm_new_crtc_state->abm_level; /*TODO How it works with MPO ?*/ if (!commit_planes_to_stream( + dm, dm->dc, status->plane_states, status->plane_count, dm_new_crtc_state, to_dm_crtc_state(old_crtc_state), - dm_state->context)) + dc_state)) dm_error("%s: Failed to update stream scaling!\n", __func__); } @@ -4806,7 +5165,8 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); if (dm_new_crtc_state->stream) - amdgpu_dm_commit_planes(state, dev, dm, crtc, &wait_for_vblank); + amdgpu_dm_commit_planes(state, dc_state, dev, + dm, crtc, &wait_for_vblank); } @@ -4846,6 +5206,9 @@ static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state) for (i = 0; i < crtc_disable_count; i++) pm_runtime_put_autosuspend(dev->dev); pm_runtime_mark_last_busy(dev->dev); + + if (dc_state_temp) + dc_release_state(dc_state_temp); } @@ -4989,20 +5352,18 @@ static int do_aquire_global_lock(struct drm_device *dev, return ret < 0 ? ret : 0; } -void set_freesync_on_stream(struct amdgpu_display_manager *dm, - struct dm_crtc_state *new_crtc_state, - struct dm_connector_state *new_con_state, - struct dc_stream_state *new_stream) +static void get_freesync_config_for_crtc( + struct dm_crtc_state *new_crtc_state, + struct dm_connector_state *new_con_state) { struct mod_freesync_config config = {0}; - struct mod_vrr_params vrr = {0}; - struct dc_info_packet vrr_infopacket = {0}; struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(new_con_state->base.connector); - if (new_con_state->freesync_capable && - new_con_state->freesync_enable) { - config.state = new_crtc_state->freesync_enabled ? + new_crtc_state->vrr_supported = new_con_state->freesync_capable; + + if (new_con_state->freesync_capable) { + config.state = new_crtc_state->base.vrr_enabled ? VRR_STATE_ACTIVE_VARIABLE : VRR_STATE_INACTIVE; config.min_refresh_in_uhz = @@ -5010,21 +5371,21 @@ void set_freesync_on_stream(struct amdgpu_display_manager *dm, config.max_refresh_in_uhz = aconnector->max_vfreq * 1000000; config.vsif_supported = true; + config.btr = true; } - mod_freesync_build_vrr_params(dm->freesync_module, - new_stream, - &config, &vrr); + new_crtc_state->freesync_config = config; +} - mod_freesync_build_vrr_infopacket(dm->freesync_module, - new_stream, - &vrr, - packet_type_fs1, - NULL, - &vrr_infopacket); +static void reset_freesync_config_for_crtc( + struct dm_crtc_state *new_crtc_state) +{ + new_crtc_state->vrr_supported = false; - new_crtc_state->adjust = vrr.adjust; - new_crtc_state->vrr_infopacket = vrr_infopacket; + memset(&new_crtc_state->vrr_params, 0, + sizeof(new_crtc_state->vrr_params)); + memset(&new_crtc_state->vrr_infopacket, 0, + sizeof(new_crtc_state->vrr_infopacket)); } static int dm_update_crtcs_state(struct amdgpu_display_manager *dm, @@ -5032,11 +5393,11 @@ static int dm_update_crtcs_state(struct amdgpu_display_manager *dm, bool enable, bool *lock_and_validation_needed) { + struct dm_atomic_state *dm_state = NULL; struct drm_crtc *crtc; struct drm_crtc_state *old_crtc_state, *new_crtc_state; int i; struct dm_crtc_state *dm_old_crtc_state, *dm_new_crtc_state; - struct dm_atomic_state *dm_state = to_dm_atomic_state(state); struct dc_stream_state *new_stream; int ret = 0; @@ -5084,7 +5445,8 @@ static int dm_update_crtcs_state(struct amdgpu_display_manager *dm, new_stream = create_stream_for_sink(aconnector, &new_crtc_state->mode, - dm_new_conn_state); + dm_new_conn_state, + dm_old_crtc_state->stream); /* * we can have no stream on ACTION_SET if a display @@ -5099,8 +5461,7 @@ static int dm_update_crtcs_state(struct amdgpu_display_manager *dm, break; } - set_freesync_on_stream(dm, dm_new_crtc_state, - dm_new_conn_state, new_stream); + dm_new_crtc_state->abm_level = dm_new_conn_state->abm_level; if (dc_is_stream_unchanged(new_stream, dm_old_crtc_state->stream) && dc_is_stream_scaling_unchanged(new_stream, dm_old_crtc_state->stream)) { @@ -5110,9 +5471,6 @@ static int dm_update_crtcs_state(struct amdgpu_display_manager *dm, } } - if (dm_old_crtc_state->freesync_enabled != dm_new_crtc_state->freesync_enabled) - new_crtc_state->mode_changed = true; - if (!drm_atomic_crtc_needs_modeset(new_crtc_state)) goto next_crtc; @@ -5134,6 +5492,10 @@ static int dm_update_crtcs_state(struct amdgpu_display_manager *dm, if (!dm_old_crtc_state->stream) goto next_crtc; + ret = dm_atomic_get_state(state, &dm_state); + if (ret) + goto fail; + DRM_DEBUG_DRIVER("Disabling DRM crtc: %d\n", crtc->base.id); @@ -5149,6 +5511,8 @@ static int dm_update_crtcs_state(struct amdgpu_display_manager *dm, dc_stream_release(dm_old_crtc_state->stream); dm_new_crtc_state->stream = NULL; + reset_freesync_config_for_crtc(dm_new_crtc_state); + *lock_and_validation_needed = true; } else {/* Add stream for any updated/enabled CRTC */ @@ -5168,6 +5532,10 @@ static int dm_update_crtcs_state(struct amdgpu_display_manager *dm, WARN_ON(dm_new_crtc_state->stream); + ret = dm_atomic_get_state(state, &dm_state); + if (ret) + goto fail; + dm_new_crtc_state->stream = new_stream; dc_stream_retain(new_stream); @@ -5226,7 +5594,9 @@ next_crtc: amdgpu_dm_set_ctm(dm_new_crtc_state); } - + /* Update Freesync settings. */ + get_freesync_config_for_crtc(dm_new_crtc_state, + dm_new_conn_state); } return ret; @@ -5242,12 +5612,13 @@ static int dm_update_planes_state(struct dc *dc, bool enable, bool *lock_and_validation_needed) { + + struct dm_atomic_state *dm_state = NULL; struct drm_crtc *new_plane_crtc, *old_plane_crtc; struct drm_crtc_state *old_crtc_state, *new_crtc_state; struct drm_plane *plane; struct drm_plane_state *old_plane_state, *new_plane_state; struct dm_crtc_state *dm_new_crtc_state, *dm_old_crtc_state; - struct dm_atomic_state *dm_state = to_dm_atomic_state(state); struct dm_plane_state *dm_new_plane_state, *dm_old_plane_state; int i ; /* TODO return page_flip_needed() function */ @@ -5285,6 +5656,10 @@ static int dm_update_planes_state(struct dc *dc, DRM_DEBUG_ATOMIC("Disabling DRM plane: %d on DRM crtc %d\n", plane->base.id, old_plane_crtc->base.id); + ret = dm_atomic_get_state(state, &dm_state); + if (ret) + return ret; + if (!dc_remove_plane_from_context( dc, dm_old_crtc_state->stream, @@ -5339,6 +5714,12 @@ static int dm_update_planes_state(struct dc *dc, return ret; } + ret = dm_atomic_get_state(state, &dm_state); + if (ret) { + dc_plane_state_release(dc_new_plane_state); + return ret; + } + /* * Any atomic check errors that occur after this will * not need a release. The plane state will be attached @@ -5370,11 +5751,14 @@ static int dm_update_planes_state(struct dc *dc, return ret; } -enum surface_update_type dm_determine_update_type_for_commit(struct dc *dc, struct drm_atomic_state *state) -{ - - int i, j, num_plane; +static int +dm_determine_update_type_for_commit(struct dc *dc, + struct drm_atomic_state *state, + enum surface_update_type *out_type) +{ + struct dm_atomic_state *dm_state = NULL, *old_dm_state = NULL; + int i, j, num_plane, ret = 0; struct drm_plane_state *old_plane_state, *new_plane_state; struct dm_plane_state *new_dm_plane_state, *old_dm_plane_state; struct drm_crtc *new_plane_crtc, *old_plane_crtc; @@ -5394,7 +5778,7 @@ enum surface_update_type dm_determine_update_type_for_commit(struct dc *dc, stru DRM_ERROR("Plane or surface update failed to allocate"); /* Set type to FULL to avoid crashing in DC*/ update_type = UPDATE_TYPE_FULL; - goto ret; + goto cleanup; } for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { @@ -5448,27 +5832,40 @@ enum surface_update_type dm_determine_update_type_for_commit(struct dc *dc, stru } if (num_plane > 0) { - status = dc_stream_get_status(new_dm_crtc_state->stream); + ret = dm_atomic_get_state(state, &dm_state); + if (ret) + goto cleanup; + + old_dm_state = dm_atomic_get_old_state(state); + if (!old_dm_state) { + ret = -EINVAL; + goto cleanup; + } + + status = dc_state_get_stream_status(old_dm_state->context, + new_dm_crtc_state->stream); + update_type = dc_check_update_surfaces_for_stream(dc, updates, num_plane, &stream_update, status); if (update_type > UPDATE_TYPE_MED) { update_type = UPDATE_TYPE_FULL; - goto ret; + goto cleanup; } } } else if (!new_dm_crtc_state->stream && old_dm_crtc_state->stream) { update_type = UPDATE_TYPE_FULL; - goto ret; + goto cleanup; } } -ret: +cleanup: kfree(updates); kfree(surface); - return update_type; + *out_type = update_type; + return ret; } /** @@ -5500,8 +5897,8 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, struct drm_atomic_state *state) { struct amdgpu_device *adev = dev->dev_private; + struct dm_atomic_state *dm_state = NULL; struct dc *dc = adev->dm.dc; - struct dm_atomic_state *dm_state = to_dm_atomic_state(state); struct drm_connector *connector; struct drm_connector_state *old_con_state, *new_con_state; struct drm_crtc *crtc; @@ -5522,12 +5919,9 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, goto fail; for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) { - struct dm_crtc_state *dm_new_crtc_state = to_dm_crtc_state(new_crtc_state); - struct dm_crtc_state *dm_old_crtc_state = to_dm_crtc_state(old_crtc_state); - if (!drm_atomic_crtc_needs_modeset(new_crtc_state) && !new_crtc_state->color_mgmt_changed && - (dm_old_crtc_state->freesync_enabled == dm_new_crtc_state->freesync_enabled)) + !new_crtc_state->vrr_enabled) continue; if (!new_crtc_state->enable) @@ -5542,10 +5936,6 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, goto fail; } - dm_state->context = dc_create_state(); - ASSERT(dm_state->context); - dc_resource_state_copy_construct_current(dc, dm_state->context); - /* Remove exiting planes if they are modified */ ret = dm_update_planes_state(dc, state, false, &lock_and_validation_needed); if (ret) { @@ -5598,7 +5988,9 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, lock_and_validation_needed = true; } - update_type = dm_determine_update_type_for_commit(dc, state); + ret = dm_determine_update_type_for_commit(dc, state, &update_type); + if (ret) + goto fail; if (overall_update_type < update_type) overall_update_type = update_type; @@ -5616,6 +6008,9 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, if (overall_update_type > UPDATE_TYPE_FAST) { + ret = dm_atomic_get_state(state, &dm_state); + if (ret) + goto fail; ret = do_aquire_global_lock(dev, state); if (ret) @@ -5625,6 +6020,13 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev, ret = -EINVAL; goto fail; } + } else if (state->legacy_cursor_update) { + /* + * This is a fast cursor update coming from the plane update + * helper, check if it can be done asynchronously for better + * performance. + */ + state->async_update = !drm_atomic_helper_async_check(dev, state); } /* Must be success */ @@ -5670,14 +6072,15 @@ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector, struct detailed_data_monitor_range *range; struct amdgpu_dm_connector *amdgpu_dm_connector = to_amdgpu_dm_connector(connector); - struct dm_connector_state *dm_con_state; + struct dm_connector_state *dm_con_state = NULL; struct drm_device *dev = connector->dev; struct amdgpu_device *adev = dev->dev_private; + bool freesync_capable = false; if (!connector->state) { DRM_ERROR("%s - Connector has no state", __func__); - return; + goto update; } if (!edid) { @@ -5687,9 +6090,7 @@ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector, amdgpu_dm_connector->max_vfreq = 0; amdgpu_dm_connector->pixel_clock_mhz = 0; - dm_con_state->freesync_capable = false; - dm_con_state->freesync_enable = false; - return; + goto update; } dm_con_state = to_dm_connector_state(connector->state); @@ -5697,10 +6098,10 @@ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector, edid_check_required = false; if (!amdgpu_dm_connector->dc_sink) { DRM_ERROR("dc_sink NULL, could not add free_sync module.\n"); - return; + goto update; } if (!adev->dm.freesync_module) - return; + goto update; /* * if edid non zero restrict freesync only for dp and edp */ @@ -5712,7 +6113,6 @@ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector, amdgpu_dm_connector); } } - dm_con_state->freesync_capable = false; if (edid_check_required == true && (edid->version > 1 || (edid->version == 1 && edid->revision > 1))) { for (i = 0; i < 4; i++) { @@ -5744,8 +6144,16 @@ void amdgpu_dm_update_freesync_caps(struct drm_connector *connector, if (amdgpu_dm_connector->max_vfreq - amdgpu_dm_connector->min_vfreq > 10) { - dm_con_state->freesync_capable = true; + freesync_capable = true; } } + +update: + if (dm_con_state) + dm_con_state->freesync_capable = freesync_capable; + + if (connector->vrr_capable_property) + drm_connector_set_vrr_capable_property(connector, + freesync_capable); } diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index 607c3cdd7d0c..fbd161ddc3f4 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -84,6 +84,18 @@ struct dm_comressor_info { }; /** + * struct amdgpu_dm_backlight_caps - Usable range of backlight values from ACPI + * @min_input_signal: minimum possible input in range 0-255 + * @max_input_signal: maximum possible input in range 0-255 + * @caps_valid: true if these values are from the ACPI interface + */ +struct amdgpu_dm_backlight_caps { + int min_input_signal; + int max_input_signal; + bool caps_valid; +}; + +/** * struct amdgpu_display_manager - Central amdgpu display manager device * * @dc: Display Core control structure @@ -112,6 +124,25 @@ struct amdgpu_display_manager { u16 display_indexes_num; /** + * @atomic_obj + * + * In combination with &dm_atomic_state it helps manage + * global atomic state that doesn't map cleanly into existing + * drm resources, like &dc_context. + */ + struct drm_private_obj atomic_obj; + + struct drm_modeset_lock atomic_obj_lock; + + /** + * @dc_lock: + * + * Guards access to DC functions that can issue register write + * sequences. + */ + struct mutex dc_lock; + + /** * @irq_handler_list_low_tab: * * Low priority IRQ handler table. @@ -158,6 +189,7 @@ struct amdgpu_display_manager { struct backlight_device *backlight_dev; const struct dc_link *backlight_link; + struct amdgpu_dm_backlight_caps backlight_caps; struct mod_freesync *freesync_module; @@ -231,15 +263,21 @@ struct dm_crtc_state { int crc_skip_count; bool crc_enabled; - bool freesync_enabled; - struct dc_crtc_timing_adjust adjust; + bool freesync_timing_changed; + bool freesync_vrr_info_changed; + + bool vrr_supported; + struct mod_freesync_config freesync_config; + struct mod_vrr_params vrr_params; struct dc_info_packet vrr_infopacket; + + int abm_level; }; #define to_dm_crtc_state(x) container_of(x, struct dm_crtc_state, base) struct dm_atomic_state { - struct drm_atomic_state base; + struct drm_private_state base; struct dc_state *context; }; @@ -254,8 +292,8 @@ struct dm_connector_state { uint8_t underscan_hborder; uint8_t max_bpc; bool underscan_enable; - bool freesync_enable; bool freesync_capable; + uint8_t abm_level; }; #define to_dm_connector_state(x)\ diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c index 24665795b1ad..24632727e127 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c @@ -343,10 +343,9 @@ dm_dp_add_mst_connector(struct drm_dp_mst_topology_mgr *mgr, master->connector_id); aconnector->mst_encoder = dm_dp_create_fake_mst_encoder(master); + drm_connector_attach_encoder(&aconnector->base, + &aconnector->mst_encoder->base); - /* - * TODO: understand why this one is needed - */ drm_object_attach_property( &connector->base, dev->mode_config.path_property, diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h new file mode 100644 index 000000000000..d898981684d5 --- /dev/null +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_trace.h @@ -0,0 +1,104 @@ +/* + * Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM amdgpu_dm + +#if !defined(_AMDGPU_DM_TRACE_H) || defined(TRACE_HEADER_MULTI_READ) +#define _AMDGPU_DM_TRACE_H_ + +#include <linux/tracepoint.h> + +TRACE_EVENT(amdgpu_dc_rreg, + TP_PROTO(unsigned long *read_count, uint32_t reg, uint32_t value), + TP_ARGS(read_count, reg, value), + TP_STRUCT__entry( + __field(uint32_t, reg) + __field(uint32_t, value) + ), + TP_fast_assign( + __entry->reg = reg; + __entry->value = value; + *read_count = *read_count + 1; + ), + TP_printk("reg=0x%08lx, value=0x%08lx", + (unsigned long)__entry->reg, + (unsigned long)__entry->value) +); + +TRACE_EVENT(amdgpu_dc_wreg, + TP_PROTO(unsigned long *write_count, uint32_t reg, uint32_t value), + TP_ARGS(write_count, reg, value), + TP_STRUCT__entry( + __field(uint32_t, reg) + __field(uint32_t, value) + ), + TP_fast_assign( + __entry->reg = reg; + __entry->value = value; + *write_count = *write_count + 1; + ), + TP_printk("reg=0x%08lx, value=0x%08lx", + (unsigned long)__entry->reg, + (unsigned long)__entry->value) +); + + +TRACE_EVENT(amdgpu_dc_performance, + TP_PROTO(unsigned long read_count, unsigned long write_count, + unsigned long *last_read, unsigned long *last_write, + const char *func, unsigned int line), + TP_ARGS(read_count, write_count, last_read, last_write, func, line), + TP_STRUCT__entry( + __field(uint32_t, reads) + __field(uint32_t, writes) + __field(uint32_t, read_delta) + __field(uint32_t, write_delta) + __string(func, func) + __field(uint32_t, line) + ), + TP_fast_assign( + __entry->reads = read_count; + __entry->writes = write_count; + __entry->read_delta = read_count - *last_read; + __entry->write_delta = write_count - *last_write; + __assign_str(func, func); + __entry->line = line; + *last_read = read_count; + *last_write = write_count; + ), + TP_printk("%s:%d reads=%08ld (%08ld total), writes=%08ld (%08ld total)", + __get_str(func), __entry->line, + (unsigned long)__entry->read_delta, + (unsigned long)__entry->reads, + (unsigned long)__entry->write_delta, + (unsigned long)__entry->writes) +); +#endif /* _AMDGPU_DM_TRACE_H_ */ + +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#define TRACE_INCLUDE_FILE amdgpu_dm_trace +#include <trace/define_trace.h> diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c index 751bb614fc0e..c513ab6f3843 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c @@ -638,6 +638,7 @@ static enum bp_result get_ss_info_v4_1( { enum bp_result result = BP_RESULT_OK; struct atom_display_controller_info_v4_1 *disp_cntl_tbl = NULL; + struct atom_smu_info_v3_3 *smu_info = NULL; if (!ss_info) return BP_RESULT_BADINPUT; @@ -650,6 +651,7 @@ static enum bp_result get_ss_info_v4_1( if (!disp_cntl_tbl) return BP_RESULT_BADBIOSTABLE; + ss_info->type.STEP_AND_DELAY_INFO = false; ss_info->spread_percentage_divider = 1000; /* BIOS no longer uses target clock. Always enable for now */ @@ -688,6 +690,19 @@ static enum bp_result get_ss_info_v4_1( */ result = BP_RESULT_UNSUPPORTED; break; + case AS_SIGNAL_TYPE_XGMI: + smu_info = GET_IMAGE(struct atom_smu_info_v3_3, + DATA_TABLES(smu_info)); + if (!smu_info) + return BP_RESULT_BADBIOSTABLE; + + ss_info->spread_spectrum_percentage = + smu_info->waflclk_ss_percentage; + ss_info->spread_spectrum_range = + smu_info->gpuclk_ss_rate_10hz * 10; + if (smu_info->waflclk_ss_mode & ATOM_SS_CENTRE_SPREAD_MODE) + ss_info->type.CENTER_MODE = true; + break; default: result = BP_RESULT_UNSUPPORTED; } diff --git a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c index 65b006ad372e..8196f3bb10c7 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/command_table_helper2.c @@ -67,6 +67,7 @@ bool dal_bios_parser_init_cmd_tbl_helper2( return true; #endif case DCE_VERSION_12_0: + case DCE_VERSION_12_1: *h = dal_cmd_tbl_helper_dce112_get_table2(); return true; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 3279e26c3440..5fd52094d459 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -62,6 +62,55 @@ const static char DC_BUILD_ID[] = "production-build"; +/** + * DOC: Overview + * + * DC is the OS-agnostic component of the amdgpu DC driver. + * + * DC maintains and validates a set of structs representing the state of the + * driver and writes that state to AMD hardware + * + * Main DC HW structs: + * + * struct dc - The central struct. One per driver. Created on driver load, + * destroyed on driver unload. + * + * struct dc_context - One per driver. + * Used as a backpointer by most other structs in dc. + * + * struct dc_link - One per connector (the physical DP, HDMI, miniDP, or eDP + * plugpoints). Created on driver load, destroyed on driver unload. + * + * struct dc_sink - One per display. Created on boot or hotplug. + * Destroyed on shutdown or hotunplug. A dc_link can have a local sink + * (the display directly attached). It may also have one or more remote + * sinks (in the Multi-Stream Transport case) + * + * struct resource_pool - One per driver. Represents the hw blocks not in the + * main pipeline. Not directly accessible by dm. + * + * Main dc state structs: + * + * These structs can be created and destroyed as needed. There is a full set of + * these structs in dc->current_state representing the currently programmed state. + * + * struct dc_state - The global DC state to track global state information, + * such as bandwidth values. + * + * struct dc_stream_state - Represents the hw configuration for the pipeline from + * a framebuffer to a display. Maps one-to-one with dc_sink. + * + * struct dc_plane_state - Represents a framebuffer. Each stream has at least one, + * and may have more in the Multi-Plane Overlay case. + * + * struct resource_context - Represents the programmable state of everything in + * the resource_pool. Not directly accessible by dm. + * + * struct pipe_ctx - A member of struct resource_context. Represents the + * internal hardware pipeline components. Each dc_plane_state has either + * one or two (in the pipe-split case). + */ + /******************************************************************************* * Private functions ******************************************************************************/ @@ -102,10 +151,6 @@ static bool create_links( return false; } - if (connectors_num == 0 && num_virtual_links == 0) { - dm_error("DC: Number of connectors is zero!\n"); - } - dm_output_to_console( "DC: %s: connectors_num: physical:%d, virtual:%d\n", __func__, @@ -175,6 +220,17 @@ failed_alloc: return false; } +static struct dc_perf_trace *dc_perf_trace_create(void) +{ + return kzalloc(sizeof(struct dc_perf_trace), GFP_KERNEL); +} + +static void dc_perf_trace_destroy(struct dc_perf_trace **perf_trace) +{ + kfree(*perf_trace); + *perf_trace = NULL; +} + /** ***************************************************************************** * Function: dc_stream_adjust_vmin_vmax @@ -240,7 +296,7 @@ bool dc_stream_get_crtc_position(struct dc *dc, } /** - * dc_stream_configure_crc: Configure CRC capture for the given stream. + * dc_stream_configure_crc() - Configure CRC capture for the given stream. * @dc: DC Object * @stream: The stream to configure CRC on. * @enable: Enable CRC if true, disable otherwise. @@ -292,7 +348,7 @@ bool dc_stream_configure_crc(struct dc *dc, struct dc_stream_state *stream, } /** - * dc_stream_get_crc: Get CRC values for the given stream. + * dc_stream_get_crc() - Get CRC values for the given stream. * @dc: DC object * @stream: The DC stream state of the stream to get CRCs from. * @r_cr, g_y, b_cb: CRC values for the three channels are stored here. @@ -328,7 +384,7 @@ void dc_stream_set_dither_option(struct dc_stream_state *stream, enum dc_dither_option option) { struct bit_depth_reduction_params params; - struct dc_link *link = stream->status.link; + struct dc_link *link = stream->sink->link; struct pipe_ctx *pipes = NULL; int i; @@ -536,6 +592,8 @@ static void destruct(struct dc *dc) if (dc->ctx->created_bios) dal_bios_parser_destroy(&dc->ctx->dc_bios); + dc_perf_trace_destroy(&dc->ctx->perf_trace); + kfree(dc->ctx); dc->ctx = NULL; @@ -659,6 +717,12 @@ static bool construct(struct dc *dc, goto fail; } + dc_ctx->perf_trace = dc_perf_trace_create(); + if (!dc_ctx->perf_trace) { + ASSERT_CRITICAL(false); + goto fail; + } + /* Create GPIO service */ dc_ctx->gpio_service = dal_gpio_service_create( dc_version, @@ -1329,6 +1393,11 @@ static enum surface_update_type check_update_surfaces_for_stream( return overall_type; } +/** + * dc_check_update_surfaces_for_stream() - Determine update type (fast, med, or full) + * + * See :c:type:`enum surface_update_type <surface_update_type>` for explanation of update types + */ enum surface_update_type dc_check_update_surfaces_for_stream( struct dc *dc, struct dc_surface_update *updates, @@ -1398,7 +1467,8 @@ static void commit_planes_do_stream_update(struct dc *dc, if ((stream_update->hdr_static_metadata && !stream->use_dynamic_meta) || stream_update->vrr_infopacket || - stream_update->vsc_infopacket) { + stream_update->vsc_infopacket || + stream_update->vsp_infopacket) { resource_build_info_frame(pipe_ctx); dc->hwss.update_info_frame(pipe_ctx); } @@ -1409,6 +1479,14 @@ static void commit_planes_do_stream_update(struct dc *dc, if (stream_update->output_csc_transform) dc_stream_program_csc_matrix(dc, stream); + if (stream_update->dither_option) { + resource_build_bit_depth_reduction_params(pipe_ctx->stream, + &pipe_ctx->stream->bit_depth_params); + pipe_ctx->stream_res.opp->funcs->opp_program_fmt(pipe_ctx->stream_res.opp, + &stream->bit_depth_params, + &stream->clamping); + } + /* Full fe update*/ if (update_type == UPDATE_TYPE_FAST) continue; @@ -1492,9 +1570,6 @@ static void commit_planes_for_stream(struct dc *dc, } } - if (update_type == UPDATE_TYPE_FULL) - context_timing_trace(dc, &context->res_ctx); - // Update Type FAST, Surface updates if (update_type == UPDATE_TYPE_FAST) { /* Lock the top pipe while updating plane addrs, since freesync requires @@ -1631,6 +1706,9 @@ enum dc_irq_source dc_interrupt_to_irq_source( return dal_irq_service_to_irq_source(dc->res_pool->irqs, src_id, ext_id); } +/** + * dc_interrupt_set() - Enable/disable an AMD hw interrupt source + */ bool dc_interrupt_set(struct dc *dc, enum dc_irq_source src, bool enable) { @@ -1686,6 +1764,15 @@ void dc_resume(struct dc *dc) core_link_resume(dc->links[i]); } +bool dc_is_dmcu_initialized(struct dc *dc) +{ + struct dmcu *dmcu = dc->res_pool->dmcu; + + if (dmcu) + return dmcu->funcs->is_dmcu_initialized(dmcu); + return false; +} + bool dc_submit_i2c( struct dc *dc, uint32_t link_index, @@ -1715,6 +1802,11 @@ static bool link_add_remote_sink_helper(struct dc_link *dc_link, struct dc_sink return true; } +/** + * dc_link_add_remote_sink() - Create a sink and attach it to an existing link + * + * EDID length is in bytes + */ struct dc_sink *dc_link_add_remote_sink( struct dc_link *link, const uint8_t *edid, @@ -1773,6 +1865,12 @@ fail_add_sink: return NULL; } +/** + * dc_link_remove_remote_sink() - Remove a remote sink from a dc_link + * + * Note that this just removes the struct dc_sink - it doesn't + * program hardware or alter other members of dc_link + */ void dc_link_remove_remote_sink(struct dc_link *link, struct dc_sink *sink) { int i; @@ -1810,4 +1908,4 @@ void get_clock_requirements_for_state(struct dc_state *state, struct AsicStateEx info->dcfClockDeepSleep = (unsigned int)state->bw.dcn.clk.dcfclk_deep_sleep_khz; info->fClock = (unsigned int)state->bw.dcn.clk.fclk_khz; info->phyClock = (unsigned int)state->bw.dcn.clk.phyclk_khz; -}
\ No newline at end of file +} diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c index 7ee9c033acbd..52deacf39841 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c @@ -198,6 +198,13 @@ static bool program_hpd_filter( return result; } +/** + * dc_link_detect_sink() - Determine if there is a sink connected + * + * @type: Returned connection type + * Does not detect downstream devices, such as MST sinks + * or display connected through active dongles + */ bool dc_link_detect_sink(struct dc_link *link, enum dc_connection_type *type) { uint32_t is_hpd_high = 0; @@ -208,6 +215,9 @@ bool dc_link_detect_sink(struct dc_link *link, enum dc_connection_type *type) return true; } + if (link->connector_signal == SIGNAL_TYPE_EDP) + link->dc->hwss.edp_wait_for_hpd_ready(link, true); + /* todo: may need to lock gpio access */ hpd_pin = get_hpd_gpio(link->ctx->dc_bios, link->link_id, link->ctx->gpio_service); if (hpd_pin == NULL) @@ -324,15 +334,15 @@ static enum signal_type get_basic_signal_type( return SIGNAL_TYPE_NONE; } -/* - * @brief - * Check whether there is a dongle on DP connector +/** + * dc_link_is_dp_sink_present() - Check if there is a native DP + * or passive DP-HDMI dongle connected */ bool dc_link_is_dp_sink_present(struct dc_link *link) { enum gpio_result gpio_result; uint32_t clock_pin = 0; - + uint8_t retry = 0; struct ddc *ddc; enum connector_id connector_id = @@ -361,11 +371,22 @@ bool dc_link_is_dp_sink_present(struct dc_link *link) return present; } - /* Read GPIO: DP sink is present if both clock and data pins are zero */ - /* [anaumov] in DAL2, there was no check for GPIO failure */ - - gpio_result = dal_gpio_get_value(ddc->pin_clock, &clock_pin); - ASSERT(gpio_result == GPIO_RESULT_OK); + /* + * Read GPIO: DP sink is present if both clock and data pins are zero + * + * [W/A] plug-unplug DP cable, sometimes customer board has + * one short pulse on clk_pin(1V, < 1ms). DP will be config to HDMI/DVI + * then monitor can't br light up. Add retry 3 times + * But in real passive dongle, it need additional 3ms to detect + */ + do { + gpio_result = dal_gpio_get_value(ddc->pin_clock, &clock_pin); + ASSERT(gpio_result == GPIO_RESULT_OK); + if (clock_pin) + udelay(1000); + else + break; + } while (retry++ < 3); present = (gpio_result == GPIO_RESULT_OK) && !clock_pin; @@ -593,6 +614,14 @@ static bool is_same_edid(struct dc_edid *old_edid, struct dc_edid *new_edid) return (memcmp(old_edid->raw_edid, new_edid->raw_edid, new_edid->length) == 0); } +/** + * dc_link_detect() - Detect if a sink is attached to a given link + * + * link->local_sink is created or destroyed as needed. + * + * This does not create remote sinks but will trigger DM + * to start MST detection if a branch is detected. + */ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) { struct dc_sink_init_data sink_init_data = { 0 }; @@ -688,12 +717,26 @@ bool dc_link_detect(struct dc_link *link, enum dc_detect_reason reason) if (memcmp(&link->dpcd_caps, &prev_dpcd_caps, sizeof(struct dpcd_caps))) same_dpcd = false; } - /* Active dongle downstream unplug */ + /* Active dongle plug in without display or downstream unplug*/ if (link->type == dc_connection_active_dongle && link->dpcd_caps.sink_count. bits.SINK_COUNT == 0) { - if (prev_sink != NULL) + if (prev_sink != NULL) { + /* Downstream unplug */ dc_sink_release(prev_sink); + } else { + /* Empty dongle plug in */ + for (i = 0; i < LINK_TRAINING_MAX_VERIFY_RETRY; i++) { + int fail_count = 0; + + dp_verify_link_cap(link, + &link->reported_link_cap, + &fail_count); + + if (fail_count == 0) + break; + } + } return true; } @@ -1396,8 +1439,6 @@ static enum dc_status enable_link_dp( else status = DC_FAIL_DP_LINK_TRAINING; - enable_stream_features(pipe_ctx); - return status; } @@ -2175,11 +2216,11 @@ bool dc_link_set_backlight_level(const struct dc_link *link, backlight_pwm_u16_16, backlight_pwm_u16_16); if (dc_is_embedded_signal(link->connector_signal)) { - if (stream != NULL) { - for (i = 0; i < MAX_PIPES; i++) { + for (i = 0; i < MAX_PIPES; i++) { + if (core_dc->current_state->res_ctx.pipe_ctx[i].stream) { if (core_dc->current_state->res_ctx. - pipe_ctx[i].stream - == stream) + pipe_ctx[i].stream->sink->link + == link) /* DMCU -1 for all controller id values, * therefore +1 here */ @@ -2218,7 +2259,7 @@ bool dc_link_set_psr_enable(const struct dc_link *link, bool enable, bool wait) struct dc *core_dc = link->ctx->dc; struct dmcu *dmcu = core_dc->res_pool->dmcu; - if (dmcu != NULL && link->psr_enabled) + if ((dmcu != NULL && dmcu->funcs->is_dmcu_initialized(dmcu)) && link->psr_enabled) dmcu->funcs->set_psr_enable(dmcu, enable, wait); return true; @@ -2594,6 +2635,9 @@ void core_link_enable_stream( core_dc->hwss.unblank_stream(pipe_ctx, &pipe_ctx->stream->sink->link->cur_link_settings); + if (dc_is_dp_signal(pipe_ctx->stream->signal)) + enable_stream_features(pipe_ctx); + dc_link_set_backlight_level(pipe_ctx->stream->sink->link, pipe_ctx->stream->bl_pwm_level, 0, @@ -2606,11 +2650,11 @@ void core_link_disable_stream(struct pipe_ctx *pipe_ctx, int option) { struct dc *core_dc = pipe_ctx->stream->ctx->dc; + core_dc->hwss.blank_stream(pipe_ctx); + if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) deallocate_mst_payload(pipe_ctx); - core_dc->hwss.blank_stream(pipe_ctx); - core_dc->hwss.disable_stream(pipe_ctx, option); disable_link(pipe_ctx->stream->sink->link, pipe_ctx->stream->signal); diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c index d91df5ef0cb3..0caacb60b02f 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c @@ -1089,6 +1089,121 @@ static struct dc_link_settings get_max_link_cap(struct dc_link *link) return max_link_cap; } +static enum dc_status read_hpd_rx_irq_data( + struct dc_link *link, + union hpd_irq_data *irq_data) +{ + static enum dc_status retval; + + /* The HW reads 16 bytes from 200h on HPD, + * but if we get an AUX_DEFER, the HW cannot retry + * and this causes the CTS tests 4.3.2.1 - 3.2.4 to + * fail, so we now explicitly read 6 bytes which is + * the req from the above mentioned test cases. + * + * For DP 1.4 we need to read those from 2002h range. + */ + if (link->dpcd_caps.dpcd_rev.raw < DPCD_REV_14) + retval = core_link_read_dpcd( + link, + DP_SINK_COUNT, + irq_data->raw, + sizeof(union hpd_irq_data)); + else { + /* Read 14 bytes in a single read and then copy only the required fields. + * This is more efficient than doing it in two separate AUX reads. */ + + uint8_t tmp[DP_SINK_STATUS_ESI - DP_SINK_COUNT_ESI + 1]; + + retval = core_link_read_dpcd( + link, + DP_SINK_COUNT_ESI, + tmp, + sizeof(tmp)); + + if (retval != DC_OK) + return retval; + + irq_data->bytes.sink_cnt.raw = tmp[DP_SINK_COUNT_ESI - DP_SINK_COUNT_ESI]; + irq_data->bytes.device_service_irq.raw = tmp[DP_DEVICE_SERVICE_IRQ_VECTOR_ESI0 - DP_SINK_COUNT_ESI]; + irq_data->bytes.lane01_status.raw = tmp[DP_LANE0_1_STATUS_ESI - DP_SINK_COUNT_ESI]; + irq_data->bytes.lane23_status.raw = tmp[DP_LANE2_3_STATUS_ESI - DP_SINK_COUNT_ESI]; + irq_data->bytes.lane_status_updated.raw = tmp[DP_LANE_ALIGN_STATUS_UPDATED_ESI - DP_SINK_COUNT_ESI]; + irq_data->bytes.sink_status.raw = tmp[DP_SINK_STATUS_ESI - DP_SINK_COUNT_ESI]; + } + + return retval; +} + +static bool hpd_rx_irq_check_link_loss_status( + struct dc_link *link, + union hpd_irq_data *hpd_irq_dpcd_data) +{ + uint8_t irq_reg_rx_power_state = 0; + enum dc_status dpcd_result = DC_ERROR_UNEXPECTED; + union lane_status lane_status; + uint32_t lane; + bool sink_status_changed; + bool return_code; + + sink_status_changed = false; + return_code = false; + + if (link->cur_link_settings.lane_count == 0) + return return_code; + + /*1. Check that Link Status changed, before re-training.*/ + + /*parse lane status*/ + for (lane = 0; lane < link->cur_link_settings.lane_count; lane++) { + /* check status of lanes 0,1 + * changed DpcdAddress_Lane01Status (0x202) + */ + lane_status.raw = get_nibble_at_index( + &hpd_irq_dpcd_data->bytes.lane01_status.raw, + lane); + + if (!lane_status.bits.CHANNEL_EQ_DONE_0 || + !lane_status.bits.CR_DONE_0 || + !lane_status.bits.SYMBOL_LOCKED_0) { + /* if one of the channel equalization, clock + * recovery or symbol lock is dropped + * consider it as (link has been + * dropped) dp sink status has changed + */ + sink_status_changed = true; + break; + } + } + + /* Check interlane align.*/ + if (sink_status_changed || + !hpd_irq_dpcd_data->bytes.lane_status_updated.bits.INTERLANE_ALIGN_DONE) { + + DC_LOG_HW_HPD_IRQ("%s: Link Status changed.\n", __func__); + + return_code = true; + + /*2. Check that we can handle interrupt: Not in FS DOS, + * Not in "Display Timeout" state, Link is trained. + */ + dpcd_result = core_link_read_dpcd(link, + DP_SET_POWER, + &irq_reg_rx_power_state, + sizeof(irq_reg_rx_power_state)); + + if (dpcd_result != DC_OK) { + DC_LOG_HW_HPD_IRQ("%s: DPCD read failed to obtain power state.\n", + __func__); + } else { + if (irq_reg_rx_power_state != DP_SET_POWER_D0) + return_code = false; + } + } + + return return_code; +} + bool dp_verify_link_cap( struct dc_link *link, struct dc_link_settings *known_limit_link_setting, @@ -1104,12 +1219,14 @@ bool dp_verify_link_cap( struct clock_source *dp_cs; enum clock_source_id dp_cs_id = CLOCK_SOURCE_ID_EXTERNAL; enum link_training_result status; + union hpd_irq_data irq_data; if (link->dc->debug.skip_detection_link_training) { link->verified_link_cap = *known_limit_link_setting; return true; } + memset(&irq_data, 0, sizeof(irq_data)); success = false; skip_link_training = false; @@ -1168,9 +1285,15 @@ bool dp_verify_link_cap( (*fail_count)++; } - if (success) + if (success) { link->verified_link_cap = *cur; - + udelay(1000); + if (read_hpd_rx_irq_data(link, &irq_data) == DC_OK) + if (hpd_rx_irq_check_link_loss_status( + link, + &irq_data)) + (*fail_count)++; + } /* always disable the link before trying another * setting or before returning we'll enable it later * based on the actual mode we're driving @@ -1572,122 +1695,6 @@ void decide_link_settings(struct dc_stream_state *stream, } /*************************Short Pulse IRQ***************************/ - -static bool hpd_rx_irq_check_link_loss_status( - struct dc_link *link, - union hpd_irq_data *hpd_irq_dpcd_data) -{ - uint8_t irq_reg_rx_power_state = 0; - enum dc_status dpcd_result = DC_ERROR_UNEXPECTED; - union lane_status lane_status; - uint32_t lane; - bool sink_status_changed; - bool return_code; - - sink_status_changed = false; - return_code = false; - - if (link->cur_link_settings.lane_count == 0) - return return_code; - - /*1. Check that Link Status changed, before re-training.*/ - - /*parse lane status*/ - for (lane = 0; lane < link->cur_link_settings.lane_count; lane++) { - /* check status of lanes 0,1 - * changed DpcdAddress_Lane01Status (0x202) - */ - lane_status.raw = get_nibble_at_index( - &hpd_irq_dpcd_data->bytes.lane01_status.raw, - lane); - - if (!lane_status.bits.CHANNEL_EQ_DONE_0 || - !lane_status.bits.CR_DONE_0 || - !lane_status.bits.SYMBOL_LOCKED_0) { - /* if one of the channel equalization, clock - * recovery or symbol lock is dropped - * consider it as (link has been - * dropped) dp sink status has changed - */ - sink_status_changed = true; - break; - } - } - - /* Check interlane align.*/ - if (sink_status_changed || - !hpd_irq_dpcd_data->bytes.lane_status_updated.bits.INTERLANE_ALIGN_DONE) { - - DC_LOG_HW_HPD_IRQ("%s: Link Status changed.\n", __func__); - - return_code = true; - - /*2. Check that we can handle interrupt: Not in FS DOS, - * Not in "Display Timeout" state, Link is trained. - */ - dpcd_result = core_link_read_dpcd(link, - DP_SET_POWER, - &irq_reg_rx_power_state, - sizeof(irq_reg_rx_power_state)); - - if (dpcd_result != DC_OK) { - DC_LOG_HW_HPD_IRQ("%s: DPCD read failed to obtain power state.\n", - __func__); - } else { - if (irq_reg_rx_power_state != DP_SET_POWER_D0) - return_code = false; - } - } - - return return_code; -} - -static enum dc_status read_hpd_rx_irq_data( - struct dc_link *link, - union hpd_irq_data *irq_data) -{ - static enum dc_status retval; - - /* The HW reads 16 bytes from 200h on HPD, - * but if we get an AUX_DEFER, the HW cannot retry - * and this causes the CTS tests 4.3.2.1 - 3.2.4 to - * fail, so we now explicitly read 6 bytes which is - * the req from the above mentioned test cases. - * - * For DP 1.4 we need to read those from 2002h range. - */ - if (link->dpcd_caps.dpcd_rev.raw < DPCD_REV_14) - retval = core_link_read_dpcd( - link, - DP_SINK_COUNT, - irq_data->raw, - sizeof(union hpd_irq_data)); - else { - /* Read 14 bytes in a single read and then copy only the required fields. - * This is more efficient than doing it in two separate AUX reads. */ - - uint8_t tmp[DP_SINK_STATUS_ESI - DP_SINK_COUNT_ESI + 1]; - - retval = core_link_read_dpcd( - link, - DP_SINK_COUNT_ESI, - tmp, - sizeof(tmp)); - - if (retval != DC_OK) - return retval; - - irq_data->bytes.sink_cnt.raw = tmp[DP_SINK_COUNT_ESI - DP_SINK_COUNT_ESI]; - irq_data->bytes.device_service_irq.raw = tmp[DP_DEVICE_SERVICE_IRQ_VECTOR_ESI0 - DP_SINK_COUNT_ESI]; - irq_data->bytes.lane01_status.raw = tmp[DP_LANE0_1_STATUS_ESI - DP_SINK_COUNT_ESI]; - irq_data->bytes.lane23_status.raw = tmp[DP_LANE2_3_STATUS_ESI - DP_SINK_COUNT_ESI]; - irq_data->bytes.lane_status_updated.raw = tmp[DP_LANE_ALIGN_STATUS_UPDATED_ESI - DP_SINK_COUNT_ESI]; - irq_data->bytes.sink_status.raw = tmp[DP_SINK_STATUS_ESI - DP_SINK_COUNT_ESI]; - } - - return retval; -} - static bool allow_hpd_rx_irq(const struct dc_link *link) { /* @@ -2196,7 +2203,7 @@ static void get_active_converter_info( } if (link->dpcd_caps.dpcd_rev.raw >= DPCD_REV_11) { - uint8_t det_caps[4]; + uint8_t det_caps[16]; /* CTS 4.2.2.7 expects source to read Detailed Capabilities Info : 00080h-0008F.*/ union dwnstream_port_caps_byte0 *port_caps = (union dwnstream_port_caps_byte0 *)det_caps; core_link_read_dpcd(link, DP_DOWNSTREAM_PORT_0, @@ -2240,7 +2247,8 @@ static void get_active_converter_info( translate_dpcd_max_bpc( hdmi_color_caps.bits.MAX_BITS_PER_COLOR_COMPONENT); - link->dpcd_caps.dongle_caps.extendedCapValid = true; + if (link->dpcd_caps.dongle_caps.dp_hdmi_max_pixel_clk != 0) + link->dpcd_caps.dongle_caps.extendedCapValid = true; } break; @@ -2371,11 +2379,22 @@ static bool retrieve_link_cap(struct dc_link *link) dpcd_data[DP_TRAINING_AUX_RD_INTERVAL]; if (aux_rd_interval.bits.EXT_RECIEVER_CAP_FIELD_PRESENT == 1) { - core_link_read_dpcd( + uint8_t ext_cap_data[16]; + + memset(ext_cap_data, '\0', sizeof(ext_cap_data)); + for (i = 0; i < read_dpcd_retry_cnt; i++) { + status = core_link_read_dpcd( link, DP_DP13_DPCD_REV, - dpcd_data, - sizeof(dpcd_data)); + ext_cap_data, + sizeof(ext_cap_data)); + if (status == DC_OK) { + memcpy(dpcd_data, ext_cap_data, sizeof(dpcd_data)); + break; + } + } + if (status != DC_OK) + dm_error("%s: Read extend caps data failed, use cap from dpcd 0.\n", __func__); } } diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c index 82cd1d6e6e59..0065ec7d5330 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_hwss.c @@ -96,6 +96,7 @@ void dp_enable_link_phy( link_settings, clock_source); } + link->cur_link_settings = *link_settings; dp_receiver_power_ctrl(link, true); } @@ -307,6 +308,7 @@ void dp_retrain_link_dp_test(struct dc_link *link, link->link_enc, link_setting, pipes[i].clock_source->id); + link->cur_link_settings = *link_setting; dp_receiver_power_ctrl(link, true); @@ -316,7 +318,6 @@ void dp_retrain_link_dp_test(struct dc_link *link, skip_video_pattern, LINK_TRAINING_ATTEMPTS); - link->cur_link_settings = *link_setting; link->dc->hwss.enable_stream(&pipes[i]); diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index fc65b0055167..76137df74a53 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -83,7 +83,10 @@ enum dce_version resource_parse_asic_id(struct hw_asic_id asic_id) dc_version = DCE_VERSION_11_22; break; case FAMILY_AI: - dc_version = DCE_VERSION_12_0; + if (ASICREV_IS_VEGA20_P(asic_id.hw_internal_rev)) + dc_version = DCE_VERSION_12_1; + else + dc_version = DCE_VERSION_12_0; break; #if defined(CONFIG_DRM_AMD_DC_DCN1_0) case FAMILY_RV: @@ -136,6 +139,7 @@ struct resource_pool *dc_create_resource_pool( num_virtual_links, dc); break; case DCE_VERSION_12_0: + case DCE_VERSION_12_1: res_pool = dce120_create_resource_pool( num_virtual_links, dc); break; @@ -478,10 +482,29 @@ static enum pixel_format convert_pixel_format_to_dalsurface( return dal_pixel_format; } -static void rect_swap_helper(struct rect *rect) +static inline void get_vp_scan_direction( + enum dc_rotation_angle rotation, + bool horizontal_mirror, + bool *orthogonal_rotation, + bool *flip_vert_scan_dir, + bool *flip_horz_scan_dir) { - swap(rect->height, rect->width); - swap(rect->x, rect->y); + *orthogonal_rotation = false; + *flip_vert_scan_dir = false; + *flip_horz_scan_dir = false; + if (rotation == ROTATION_ANGLE_180) { + *flip_vert_scan_dir = true; + *flip_horz_scan_dir = true; + } else if (rotation == ROTATION_ANGLE_90) { + *orthogonal_rotation = true; + *flip_horz_scan_dir = true; + } else if (rotation == ROTATION_ANGLE_270) { + *orthogonal_rotation = true; + *flip_vert_scan_dir = true; + } + + if (horizontal_mirror) + *flip_horz_scan_dir = !*flip_horz_scan_dir; } static void calculate_viewport(struct pipe_ctx *pipe_ctx) @@ -490,33 +513,14 @@ static void calculate_viewport(struct pipe_ctx *pipe_ctx) const struct dc_stream_state *stream = pipe_ctx->stream; struct scaler_data *data = &pipe_ctx->plane_res.scl_data; struct rect surf_src = plane_state->src_rect; - struct rect clip = { 0 }; + struct rect clip, dest; int vpc_div = (data->format == PIXEL_FORMAT_420BPP8 || data->format == PIXEL_FORMAT_420BPP10) ? 2 : 1; bool pri_split = pipe_ctx->bottom_pipe && pipe_ctx->bottom_pipe->plane_state == pipe_ctx->plane_state; bool sec_split = pipe_ctx->top_pipe && pipe_ctx->top_pipe->plane_state == pipe_ctx->plane_state; - bool flip_vert_scan_dir = false, flip_horz_scan_dir = false; - - - /* - * We need take horizontal mirror into account. On an unrotated surface this means - * that the viewport offset is actually the offset from the other side of source - * image so we have to subtract the right edge of the viewport from the right edge of - * the source window. Similar to mirror we need to take into account how offset is - * affected for 270/180 rotations - */ - if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_180) { - flip_vert_scan_dir = true; - flip_horz_scan_dir = true; - } else if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_90) - flip_vert_scan_dir = true; - else if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_270) - flip_horz_scan_dir = true; - - if (pipe_ctx->plane_state->horizontal_mirror) - flip_horz_scan_dir = !flip_horz_scan_dir; + bool orthogonal_rotation, flip_y_start, flip_x_start; if (stream->view_format == VIEW_3D_FORMAT_SIDE_BY_SIDE || stream->view_format == VIEW_3D_FORMAT_TOP_AND_BOTTOM) { @@ -524,13 +528,10 @@ static void calculate_viewport(struct pipe_ctx *pipe_ctx) sec_split = false; } - if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_90 || - pipe_ctx->plane_state->rotation == ROTATION_ANGLE_270) - rect_swap_helper(&surf_src); - /* The actual clip is an intersection between stream * source and surface clip */ + dest = plane_state->dst_rect; clip.x = stream->src.x > plane_state->clip_rect.x ? stream->src.x : plane_state->clip_rect.x; @@ -547,66 +548,77 @@ static void calculate_viewport(struct pipe_ctx *pipe_ctx) stream->src.y + stream->src.height - clip.y : plane_state->clip_rect.y + plane_state->clip_rect.height - clip.y ; - /* offset = surf_src.ofs + (clip.ofs - surface->dst_rect.ofs) * scl_ratio - * note: surf_src.ofs should be added after rotation/mirror offset direction - * adjustment since it is already in viewport space - * num_pixels = clip.num_pix * scl_ratio + /* + * Need to calculate how scan origin is shifted in vp space + * to correctly rotate clip and dst */ - data->viewport.x = (clip.x - plane_state->dst_rect.x) * - surf_src.width / plane_state->dst_rect.width; - data->viewport.width = clip.width * - surf_src.width / plane_state->dst_rect.width; - - data->viewport.y = (clip.y - plane_state->dst_rect.y) * - surf_src.height / plane_state->dst_rect.height; - data->viewport.height = clip.height * - surf_src.height / plane_state->dst_rect.height; + get_vp_scan_direction( + plane_state->rotation, + plane_state->horizontal_mirror, + &orthogonal_rotation, + &flip_y_start, + &flip_x_start); - if (flip_vert_scan_dir) - data->viewport.y = surf_src.height - data->viewport.y - data->viewport.height; - if (flip_horz_scan_dir) - data->viewport.x = surf_src.width - data->viewport.x - data->viewport.width; + if (orthogonal_rotation) { + swap(clip.x, clip.y); + swap(clip.width, clip.height); + swap(dest.x, dest.y); + swap(dest.width, dest.height); + } + if (flip_x_start) { + clip.x = dest.x + dest.width - clip.x - clip.width; + dest.x = 0; + } + if (flip_y_start) { + clip.y = dest.y + dest.height - clip.y - clip.height; + dest.y = 0; + } - data->viewport.x += surf_src.x; - data->viewport.y += surf_src.y; + /* offset = surf_src.ofs + (clip.ofs - surface->dst_rect.ofs) * scl_ratio + * num_pixels = clip.num_pix * scl_ratio + */ + data->viewport.x = surf_src.x + (clip.x - dest.x) * surf_src.width / dest.width; + data->viewport.width = clip.width * surf_src.width / dest.width; + + data->viewport.y = surf_src.y + (clip.y - dest.y) * surf_src.height / dest.height; + data->viewport.height = clip.height * surf_src.height / dest.height; + + /* Handle split */ + if (pri_split || sec_split) { + if (orthogonal_rotation) { + if (flip_y_start != pri_split) + data->viewport.height /= 2; + else { + data->viewport.y += data->viewport.height / 2; + /* Ceil offset pipe */ + data->viewport.height = (data->viewport.height + 1) / 2; + } + } else { + if (flip_x_start != pri_split) + data->viewport.width /= 2; + else { + data->viewport.x += data->viewport.width / 2; + /* Ceil offset pipe */ + data->viewport.width = (data->viewport.width + 1) / 2; + } + } + } /* Round down, compensate in init */ data->viewport_c.x = data->viewport.x / vpc_div; data->viewport_c.y = data->viewport.y / vpc_div; - data->inits.h_c = (data->viewport.x % vpc_div) != 0 ? - dc_fixpt_half : dc_fixpt_zero; - data->inits.v_c = (data->viewport.y % vpc_div) != 0 ? - dc_fixpt_half : dc_fixpt_zero; + data->inits.h_c = (data->viewport.x % vpc_div) != 0 ? dc_fixpt_half : dc_fixpt_zero; + data->inits.v_c = (data->viewport.y % vpc_div) != 0 ? dc_fixpt_half : dc_fixpt_zero; + /* Round up, assume original video size always even dimensions */ data->viewport_c.width = (data->viewport.width + vpc_div - 1) / vpc_div; data->viewport_c.height = (data->viewport.height + vpc_div - 1) / vpc_div; - - /* Handle hsplit */ - if (sec_split) { - data->viewport.x += data->viewport.width / 2; - data->viewport_c.x += data->viewport_c.width / 2; - /* Ceil offset pipe */ - data->viewport.width = (data->viewport.width + 1) / 2; - data->viewport_c.width = (data->viewport_c.width + 1) / 2; - } else if (pri_split) { - if (data->viewport.width > 1) - data->viewport.width /= 2; - if (data->viewport_c.width > 1) - data->viewport_c.width /= 2; - } - - if (plane_state->rotation == ROTATION_ANGLE_90 || - plane_state->rotation == ROTATION_ANGLE_270) { - rect_swap_helper(&data->viewport_c); - rect_swap_helper(&data->viewport); - } } -static void calculate_recout(struct pipe_ctx *pipe_ctx, struct rect *recout_full) +static void calculate_recout(struct pipe_ctx *pipe_ctx) { const struct dc_plane_state *plane_state = pipe_ctx->plane_state; const struct dc_stream_state *stream = pipe_ctx->stream; - struct rect surf_src = plane_state->src_rect; struct rect surf_clip = plane_state->clip_rect; bool pri_split = pipe_ctx->bottom_pipe && pipe_ctx->bottom_pipe->plane_state == pipe_ctx->plane_state; @@ -614,10 +626,6 @@ static void calculate_recout(struct pipe_ctx *pipe_ctx, struct rect *recout_full pipe_ctx->top_pipe->plane_state == pipe_ctx->plane_state; bool top_bottom_split = stream->view_format == VIEW_3D_FORMAT_TOP_AND_BOTTOM; - if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_90 || - pipe_ctx->plane_state->rotation == ROTATION_ANGLE_270) - rect_swap_helper(&surf_src); - pipe_ctx->plane_res.scl_data.recout.x = stream->dst.x; if (stream->src.x < surf_clip.x) pipe_ctx->plane_res.scl_data.recout.x += (surf_clip.x @@ -646,7 +654,7 @@ static void calculate_recout(struct pipe_ctx *pipe_ctx, struct rect *recout_full stream->dst.y + stream->dst.height - pipe_ctx->plane_res.scl_data.recout.y; - /* Handle h & vsplit */ + /* Handle h & v split, handle rotation using viewport */ if (sec_split && top_bottom_split) { pipe_ctx->plane_res.scl_data.recout.y += pipe_ctx->plane_res.scl_data.recout.height / 2; @@ -655,44 +663,14 @@ static void calculate_recout(struct pipe_ctx *pipe_ctx, struct rect *recout_full (pipe_ctx->plane_res.scl_data.recout.height + 1) / 2; } else if (pri_split && top_bottom_split) pipe_ctx->plane_res.scl_data.recout.height /= 2; - else if (pri_split || sec_split) { - /* HMirror XOR Secondary_pipe XOR Rotation_180 */ - bool right_view = (sec_split != plane_state->horizontal_mirror) != - (plane_state->rotation == ROTATION_ANGLE_180); - - if (plane_state->rotation == ROTATION_ANGLE_90 - || plane_state->rotation == ROTATION_ANGLE_270) - /* Secondary_pipe XOR Rotation_270 */ - right_view = (plane_state->rotation == ROTATION_ANGLE_270) != sec_split; - - if (right_view) { - pipe_ctx->plane_res.scl_data.recout.x += - pipe_ctx->plane_res.scl_data.recout.width / 2; - /* Ceil offset pipe */ - pipe_ctx->plane_res.scl_data.recout.width = - (pipe_ctx->plane_res.scl_data.recout.width + 1) / 2; - } else { - if (pipe_ctx->plane_res.scl_data.recout.width > 1) - pipe_ctx->plane_res.scl_data.recout.width /= 2; - } - } - /* Unclipped recout offset = stream dst offset + ((surf dst offset - stream surf_src offset) - * * 1/ stream scaling ratio) - (surf surf_src offset * 1/ full scl - * ratio) - */ - recout_full->x = stream->dst.x + (plane_state->dst_rect.x - stream->src.x) - * stream->dst.width / stream->src.width - - surf_src.x * plane_state->dst_rect.width / surf_src.width - * stream->dst.width / stream->src.width; - recout_full->y = stream->dst.y + (plane_state->dst_rect.y - stream->src.y) - * stream->dst.height / stream->src.height - - surf_src.y * plane_state->dst_rect.height / surf_src.height - * stream->dst.height / stream->src.height; - - recout_full->width = plane_state->dst_rect.width - * stream->dst.width / stream->src.width; - recout_full->height = plane_state->dst_rect.height - * stream->dst.height / stream->src.height; + else if (sec_split) { + pipe_ctx->plane_res.scl_data.recout.x += + pipe_ctx->plane_res.scl_data.recout.width / 2; + /* Ceil offset pipe */ + pipe_ctx->plane_res.scl_data.recout.width = + (pipe_ctx->plane_res.scl_data.recout.width + 1) / 2; + } else if (pri_split) + pipe_ctx->plane_res.scl_data.recout.width /= 2; } static void calculate_scaling_ratios(struct pipe_ctx *pipe_ctx) @@ -705,9 +683,10 @@ static void calculate_scaling_ratios(struct pipe_ctx *pipe_ctx) const int out_w = stream->dst.width; const int out_h = stream->dst.height; + /*Swap surf_src height and width since scaling ratios are in recout rotation*/ if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_90 || pipe_ctx->plane_state->rotation == ROTATION_ANGLE_270) - rect_swap_helper(&surf_src); + swap(surf_src.height, surf_src.width); pipe_ctx->plane_res.scl_data.ratios.horz = dc_fixpt_from_fraction( surf_src.width, @@ -744,351 +723,202 @@ static void calculate_scaling_ratios(struct pipe_ctx *pipe_ctx) pipe_ctx->plane_res.scl_data.ratios.vert_c, 19); } -static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx, struct rect *recout_full) +static inline void adjust_vp_and_init_for_seamless_clip( + bool flip_scan_dir, + int recout_skip, + int src_size, + int taps, + struct fixed31_32 ratio, + struct fixed31_32 *init, + int *vp_offset, + int *vp_size) { - struct scaler_data *data = &pipe_ctx->plane_res.scl_data; - struct rect src = pipe_ctx->plane_state->src_rect; - int vpc_div = (data->format == PIXEL_FORMAT_420BPP8 - || data->format == PIXEL_FORMAT_420BPP10) ? 2 : 1; - bool flip_vert_scan_dir = false, flip_horz_scan_dir = false; - - /* - * Need to calculate the scan direction for viewport to make adjustments - */ - if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_180) { - flip_vert_scan_dir = true; - flip_horz_scan_dir = true; - } else if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_90) - flip_vert_scan_dir = true; - else if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_270) - flip_horz_scan_dir = true; - - if (pipe_ctx->plane_state->horizontal_mirror) - flip_horz_scan_dir = !flip_horz_scan_dir; - - if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_90 || - pipe_ctx->plane_state->rotation == ROTATION_ANGLE_270) { - rect_swap_helper(&src); - rect_swap_helper(&data->viewport_c); - rect_swap_helper(&data->viewport); - } - - /* - * Init calculated according to formula: - * init = (scaling_ratio + number_of_taps + 1) / 2 - * init_bot = init + scaling_ratio - * init_c = init + truncated_vp_c_offset(from calculate viewport) - */ - data->inits.h = dc_fixpt_truncate(dc_fixpt_div_int( - dc_fixpt_add_int(data->ratios.horz, data->taps.h_taps + 1), 2), 19); - - data->inits.h_c = dc_fixpt_truncate(dc_fixpt_add(data->inits.h_c, dc_fixpt_div_int( - dc_fixpt_add_int(data->ratios.horz_c, data->taps.h_taps_c + 1), 2)), 19); - - data->inits.v = dc_fixpt_truncate(dc_fixpt_div_int( - dc_fixpt_add_int(data->ratios.vert, data->taps.v_taps + 1), 2), 19); - - data->inits.v_c = dc_fixpt_truncate(dc_fixpt_add(data->inits.v_c, dc_fixpt_div_int( - dc_fixpt_add_int(data->ratios.vert_c, data->taps.v_taps_c + 1), 2)), 19); - - if (!flip_horz_scan_dir) { + if (!flip_scan_dir) { /* Adjust for viewport end clip-off */ - if ((data->viewport.x + data->viewport.width) < (src.x + src.width)) { - int vp_clip = src.x + src.width - data->viewport.width - data->viewport.x; - int int_part = dc_fixpt_floor( - dc_fixpt_sub(data->inits.h, data->ratios.horz)); + if ((*vp_offset + *vp_size) < src_size) { + int vp_clip = src_size - *vp_size - *vp_offset; + int int_part = dc_fixpt_floor(dc_fixpt_sub(*init, ratio)); int_part = int_part > 0 ? int_part : 0; - data->viewport.width += int_part < vp_clip ? int_part : vp_clip; - } - if ((data->viewport_c.x + data->viewport_c.width) < (src.x + src.width) / vpc_div) { - int vp_clip = (src.x + src.width) / vpc_div - - data->viewport_c.width - data->viewport_c.x; - int int_part = dc_fixpt_floor( - dc_fixpt_sub(data->inits.h_c, data->ratios.horz_c)); - - int_part = int_part > 0 ? int_part : 0; - data->viewport_c.width += int_part < vp_clip ? int_part : vp_clip; + *vp_size += int_part < vp_clip ? int_part : vp_clip; } /* Adjust for non-0 viewport offset */ - if (data->viewport.x) { + if (*vp_offset) { int int_part; - data->inits.h = dc_fixpt_add(data->inits.h, dc_fixpt_mul_int( - data->ratios.horz, data->recout.x - recout_full->x)); - int_part = dc_fixpt_floor(data->inits.h) - data->viewport.x; - if (int_part < data->taps.h_taps) { - int int_adj = data->viewport.x >= (data->taps.h_taps - int_part) ? - (data->taps.h_taps - int_part) : data->viewport.x; - data->viewport.x -= int_adj; - data->viewport.width += int_adj; + *init = dc_fixpt_add(*init, dc_fixpt_mul_int(ratio, recout_skip)); + int_part = dc_fixpt_floor(*init) - *vp_offset; + if (int_part < taps) { + int int_adj = *vp_offset >= (taps - int_part) ? + (taps - int_part) : *vp_offset; + *vp_offset -= int_adj; + *vp_size += int_adj; int_part += int_adj; - } else if (int_part > data->taps.h_taps) { - data->viewport.x += int_part - data->taps.h_taps; - data->viewport.width -= int_part - data->taps.h_taps; - int_part = data->taps.h_taps; + } else if (int_part > taps) { + *vp_offset += int_part - taps; + *vp_size -= int_part - taps; + int_part = taps; } - data->inits.h.value &= 0xffffffff; - data->inits.h = dc_fixpt_add_int(data->inits.h, int_part); - } - - if (data->viewport_c.x) { - int int_part; - - data->inits.h_c = dc_fixpt_add(data->inits.h_c, dc_fixpt_mul_int( - data->ratios.horz_c, data->recout.x - recout_full->x)); - int_part = dc_fixpt_floor(data->inits.h_c) - data->viewport_c.x; - if (int_part < data->taps.h_taps_c) { - int int_adj = data->viewport_c.x >= (data->taps.h_taps_c - int_part) ? - (data->taps.h_taps_c - int_part) : data->viewport_c.x; - data->viewport_c.x -= int_adj; - data->viewport_c.width += int_adj; - int_part += int_adj; - } else if (int_part > data->taps.h_taps_c) { - data->viewport_c.x += int_part - data->taps.h_taps_c; - data->viewport_c.width -= int_part - data->taps.h_taps_c; - int_part = data->taps.h_taps_c; - } - data->inits.h_c.value &= 0xffffffff; - data->inits.h_c = dc_fixpt_add_int(data->inits.h_c, int_part); + init->value &= 0xffffffff; + *init = dc_fixpt_add_int(*init, int_part); } } else { /* Adjust for non-0 viewport offset */ - if (data->viewport.x) { - int int_part = dc_fixpt_floor( - dc_fixpt_sub(data->inits.h, data->ratios.horz)); + if (*vp_offset) { + int int_part = dc_fixpt_floor(dc_fixpt_sub(*init, ratio)); int_part = int_part > 0 ? int_part : 0; - data->viewport.width += int_part < data->viewport.x ? int_part : data->viewport.x; - data->viewport.x -= int_part < data->viewport.x ? int_part : data->viewport.x; - } - if (data->viewport_c.x) { - int int_part = dc_fixpt_floor( - dc_fixpt_sub(data->inits.h_c, data->ratios.horz_c)); - - int_part = int_part > 0 ? int_part : 0; - data->viewport_c.width += int_part < data->viewport_c.x ? int_part : data->viewport_c.x; - data->viewport_c.x -= int_part < data->viewport_c.x ? int_part : data->viewport_c.x; + *vp_size += int_part < *vp_offset ? int_part : *vp_offset; + *vp_offset -= int_part < *vp_offset ? int_part : *vp_offset; } /* Adjust for viewport end clip-off */ - if ((data->viewport.x + data->viewport.width) < (src.x + src.width)) { - int int_part; - int end_offset = src.x + src.width - - data->viewport.x - data->viewport.width; - - /* - * this is init if vp had no offset, keep in mind this is from the - * right side of vp due to scan direction - */ - data->inits.h = dc_fixpt_add(data->inits.h, dc_fixpt_mul_int( - data->ratios.horz, data->recout.x - recout_full->x)); - /* - * this is the difference between first pixel of viewport available to read - * and init position, takning into account scan direction - */ - int_part = dc_fixpt_floor(data->inits.h) - end_offset; - if (int_part < data->taps.h_taps) { - int int_adj = end_offset >= (data->taps.h_taps - int_part) ? - (data->taps.h_taps - int_part) : end_offset; - data->viewport.width += int_adj; - int_part += int_adj; - } else if (int_part > data->taps.h_taps) { - data->viewport.width += int_part - data->taps.h_taps; - int_part = data->taps.h_taps; - } - data->inits.h.value &= 0xffffffff; - data->inits.h = dc_fixpt_add_int(data->inits.h, int_part); - } - - if ((data->viewport_c.x + data->viewport_c.width) < (src.x + src.width) / vpc_div) { + if ((*vp_offset + *vp_size) < src_size) { int int_part; - int end_offset = (src.x + src.width) / vpc_div - - data->viewport_c.x - data->viewport_c.width; + int end_offset = src_size - *vp_offset - *vp_size; /* * this is init if vp had no offset, keep in mind this is from the * right side of vp due to scan direction */ - data->inits.h_c = dc_fixpt_add(data->inits.h_c, dc_fixpt_mul_int( - data->ratios.horz_c, data->recout.x - recout_full->x)); + *init = dc_fixpt_add(*init, dc_fixpt_mul_int(ratio, recout_skip)); /* * this is the difference between first pixel of viewport available to read * and init position, takning into account scan direction */ - int_part = dc_fixpt_floor(data->inits.h_c) - end_offset; - if (int_part < data->taps.h_taps_c) { - int int_adj = end_offset >= (data->taps.h_taps_c - int_part) ? - (data->taps.h_taps_c - int_part) : end_offset; - data->viewport_c.width += int_adj; + int_part = dc_fixpt_floor(*init) - end_offset; + if (int_part < taps) { + int int_adj = end_offset >= (taps - int_part) ? + (taps - int_part) : end_offset; + *vp_size += int_adj; int_part += int_adj; - } else if (int_part > data->taps.h_taps_c) { - data->viewport_c.width += int_part - data->taps.h_taps_c; - int_part = data->taps.h_taps_c; + } else if (int_part > taps) { + *vp_size += int_part - taps; + int_part = taps; } - data->inits.h_c.value &= 0xffffffff; - data->inits.h_c = dc_fixpt_add_int(data->inits.h_c, int_part); + init->value &= 0xffffffff; + *init = dc_fixpt_add_int(*init, int_part); } - } - if (!flip_vert_scan_dir) { - /* Adjust for viewport end clip-off */ - if ((data->viewport.y + data->viewport.height) < (src.y + src.height)) { - int vp_clip = src.y + src.height - data->viewport.height - data->viewport.y; - int int_part = dc_fixpt_floor( - dc_fixpt_sub(data->inits.v, data->ratios.vert)); - - int_part = int_part > 0 ? int_part : 0; - data->viewport.height += int_part < vp_clip ? int_part : vp_clip; - } - if ((data->viewport_c.y + data->viewport_c.height) < (src.y + src.height) / vpc_div) { - int vp_clip = (src.y + src.height) / vpc_div - - data->viewport_c.height - data->viewport_c.y; - int int_part = dc_fixpt_floor( - dc_fixpt_sub(data->inits.v_c, data->ratios.vert_c)); - - int_part = int_part > 0 ? int_part : 0; - data->viewport_c.height += int_part < vp_clip ? int_part : vp_clip; - } - - /* Adjust for non-0 viewport offset */ - if (data->viewport.y) { - int int_part; - - data->inits.v = dc_fixpt_add(data->inits.v, dc_fixpt_mul_int( - data->ratios.vert, data->recout.y - recout_full->y)); - int_part = dc_fixpt_floor(data->inits.v) - data->viewport.y; - if (int_part < data->taps.v_taps) { - int int_adj = data->viewport.y >= (data->taps.v_taps - int_part) ? - (data->taps.v_taps - int_part) : data->viewport.y; - data->viewport.y -= int_adj; - data->viewport.height += int_adj; - int_part += int_adj; - } else if (int_part > data->taps.v_taps) { - data->viewport.y += int_part - data->taps.v_taps; - data->viewport.height -= int_part - data->taps.v_taps; - int_part = data->taps.v_taps; - } - data->inits.v.value &= 0xffffffff; - data->inits.v = dc_fixpt_add_int(data->inits.v, int_part); - } - - if (data->viewport_c.y) { - int int_part; +} - data->inits.v_c = dc_fixpt_add(data->inits.v_c, dc_fixpt_mul_int( - data->ratios.vert_c, data->recout.y - recout_full->y)); - int_part = dc_fixpt_floor(data->inits.v_c) - data->viewport_c.y; - if (int_part < data->taps.v_taps_c) { - int int_adj = data->viewport_c.y >= (data->taps.v_taps_c - int_part) ? - (data->taps.v_taps_c - int_part) : data->viewport_c.y; - data->viewport_c.y -= int_adj; - data->viewport_c.height += int_adj; - int_part += int_adj; - } else if (int_part > data->taps.v_taps_c) { - data->viewport_c.y += int_part - data->taps.v_taps_c; - data->viewport_c.height -= int_part - data->taps.v_taps_c; - int_part = data->taps.v_taps_c; - } - data->inits.v_c.value &= 0xffffffff; - data->inits.v_c = dc_fixpt_add_int(data->inits.v_c, int_part); - } - } else { - /* Adjust for non-0 viewport offset */ - if (data->viewport.y) { - int int_part = dc_fixpt_floor( - dc_fixpt_sub(data->inits.v, data->ratios.vert)); +static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx) +{ + const struct dc_plane_state *plane_state = pipe_ctx->plane_state; + const struct dc_stream_state *stream = pipe_ctx->stream; + struct scaler_data *data = &pipe_ctx->plane_res.scl_data; + struct rect src = pipe_ctx->plane_state->src_rect; + int recout_skip_h, recout_skip_v, surf_size_h, surf_size_v; + int vpc_div = (data->format == PIXEL_FORMAT_420BPP8 + || data->format == PIXEL_FORMAT_420BPP10) ? 2 : 1; + bool orthogonal_rotation, flip_vert_scan_dir, flip_horz_scan_dir; - int_part = int_part > 0 ? int_part : 0; - data->viewport.height += int_part < data->viewport.y ? int_part : data->viewport.y; - data->viewport.y -= int_part < data->viewport.y ? int_part : data->viewport.y; - } - if (data->viewport_c.y) { - int int_part = dc_fixpt_floor( - dc_fixpt_sub(data->inits.v_c, data->ratios.vert_c)); + /* + * Need to calculate the scan direction for viewport to make adjustments + */ + get_vp_scan_direction( + plane_state->rotation, + plane_state->horizontal_mirror, + &orthogonal_rotation, + &flip_vert_scan_dir, + &flip_horz_scan_dir); + + /* Calculate src rect rotation adjusted to recout space */ + surf_size_h = src.x + src.width; + surf_size_v = src.y + src.height; + if (flip_horz_scan_dir) + src.x = 0; + if (flip_vert_scan_dir) + src.y = 0; + if (orthogonal_rotation) { + swap(src.x, src.y); + swap(src.width, src.height); + } - int_part = int_part > 0 ? int_part : 0; - data->viewport_c.height += int_part < data->viewport_c.y ? int_part : data->viewport_c.y; - data->viewport_c.y -= int_part < data->viewport_c.y ? int_part : data->viewport_c.y; - } + /* Recout matching initial vp offset = recout_offset - (stream dst offset + + * ((surf dst offset - stream src offset) * 1/ stream scaling ratio) + * - (surf surf_src offset * 1/ full scl ratio)) + */ + recout_skip_h = data->recout.x - (stream->dst.x + (plane_state->dst_rect.x - stream->src.x) + * stream->dst.width / stream->src.width - + src.x * plane_state->dst_rect.width / src.width + * stream->dst.width / stream->src.width); + recout_skip_v = data->recout.y - (stream->dst.y + (plane_state->dst_rect.y - stream->src.y) + * stream->dst.height / stream->src.height - + src.y * plane_state->dst_rect.height / src.height + * stream->dst.height / stream->src.height); + if (orthogonal_rotation) + swap(recout_skip_h, recout_skip_v); + /* + * Init calculated according to formula: + * init = (scaling_ratio + number_of_taps + 1) / 2 + * init_bot = init + scaling_ratio + * init_c = init + truncated_vp_c_offset(from calculate viewport) + */ + data->inits.h = dc_fixpt_truncate(dc_fixpt_div_int( + dc_fixpt_add_int(data->ratios.horz, data->taps.h_taps + 1), 2), 19); - /* Adjust for viewport end clip-off */ - if ((data->viewport.y + data->viewport.height) < (src.y + src.height)) { - int int_part; - int end_offset = src.y + src.height - - data->viewport.y - data->viewport.height; + data->inits.h_c = dc_fixpt_truncate(dc_fixpt_add(data->inits.h_c, dc_fixpt_div_int( + dc_fixpt_add_int(data->ratios.horz_c, data->taps.h_taps_c + 1), 2)), 19); - /* - * this is init if vp had no offset, keep in mind this is from the - * right side of vp due to scan direction - */ - data->inits.v = dc_fixpt_add(data->inits.v, dc_fixpt_mul_int( - data->ratios.vert, data->recout.y - recout_full->y)); - /* - * this is the difference between first pixel of viewport available to read - * and init position, taking into account scan direction - */ - int_part = dc_fixpt_floor(data->inits.v) - end_offset; - if (int_part < data->taps.v_taps) { - int int_adj = end_offset >= (data->taps.v_taps - int_part) ? - (data->taps.v_taps - int_part) : end_offset; - data->viewport.height += int_adj; - int_part += int_adj; - } else if (int_part > data->taps.v_taps) { - data->viewport.height += int_part - data->taps.v_taps; - int_part = data->taps.v_taps; - } - data->inits.v.value &= 0xffffffff; - data->inits.v = dc_fixpt_add_int(data->inits.v, int_part); - } + data->inits.v = dc_fixpt_truncate(dc_fixpt_div_int( + dc_fixpt_add_int(data->ratios.vert, data->taps.v_taps + 1), 2), 19); - if ((data->viewport_c.y + data->viewport_c.height) < (src.y + src.height) / vpc_div) { - int int_part; - int end_offset = (src.y + src.height) / vpc_div - - data->viewport_c.y - data->viewport_c.height; + data->inits.v_c = dc_fixpt_truncate(dc_fixpt_add(data->inits.v_c, dc_fixpt_div_int( + dc_fixpt_add_int(data->ratios.vert_c, data->taps.v_taps_c + 1), 2)), 19); - /* - * this is init if vp had no offset, keep in mind this is from the - * right side of vp due to scan direction - */ - data->inits.v_c = dc_fixpt_add(data->inits.v_c, dc_fixpt_mul_int( - data->ratios.vert_c, data->recout.y - recout_full->y)); - /* - * this is the difference between first pixel of viewport available to read - * and init position, taking into account scan direction - */ - int_part = dc_fixpt_floor(data->inits.v_c) - end_offset; - if (int_part < data->taps.v_taps_c) { - int int_adj = end_offset >= (data->taps.v_taps_c - int_part) ? - (data->taps.v_taps_c - int_part) : end_offset; - data->viewport_c.height += int_adj; - int_part += int_adj; - } else if (int_part > data->taps.v_taps_c) { - data->viewport_c.height += int_part - data->taps.v_taps_c; - int_part = data->taps.v_taps_c; - } - data->inits.v_c.value &= 0xffffffff; - data->inits.v_c = dc_fixpt_add_int(data->inits.v_c, int_part); - } - } + /* + * Taps, inits and scaling ratios are in recout space need to rotate + * to viewport rotation before adjustment + */ + adjust_vp_and_init_for_seamless_clip( + flip_horz_scan_dir, + recout_skip_h, + surf_size_h, + orthogonal_rotation ? data->taps.v_taps : data->taps.h_taps, + orthogonal_rotation ? data->ratios.vert : data->ratios.horz, + orthogonal_rotation ? &data->inits.v : &data->inits.h, + &data->viewport.x, + &data->viewport.width); + adjust_vp_and_init_for_seamless_clip( + flip_horz_scan_dir, + recout_skip_h, + surf_size_h / vpc_div, + orthogonal_rotation ? data->taps.v_taps_c : data->taps.h_taps_c, + orthogonal_rotation ? data->ratios.vert_c : data->ratios.horz_c, + orthogonal_rotation ? &data->inits.v_c : &data->inits.h_c, + &data->viewport_c.x, + &data->viewport_c.width); + adjust_vp_and_init_for_seamless_clip( + flip_vert_scan_dir, + recout_skip_v, + surf_size_v, + orthogonal_rotation ? data->taps.h_taps : data->taps.v_taps, + orthogonal_rotation ? data->ratios.horz : data->ratios.vert, + orthogonal_rotation ? &data->inits.h : &data->inits.v, + &data->viewport.y, + &data->viewport.height); + adjust_vp_and_init_for_seamless_clip( + flip_vert_scan_dir, + recout_skip_v, + surf_size_v / vpc_div, + orthogonal_rotation ? data->taps.h_taps_c : data->taps.v_taps_c, + orthogonal_rotation ? data->ratios.horz_c : data->ratios.vert_c, + orthogonal_rotation ? &data->inits.h_c : &data->inits.v_c, + &data->viewport_c.y, + &data->viewport_c.height); /* Interlaced inits based on final vert inits */ data->inits.v_bot = dc_fixpt_add(data->inits.v, data->ratios.vert); data->inits.v_c_bot = dc_fixpt_add(data->inits.v_c, data->ratios.vert_c); - if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_90 || - pipe_ctx->plane_state->rotation == ROTATION_ANGLE_270) { - rect_swap_helper(&data->viewport_c); - rect_swap_helper(&data->viewport); - } } bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) { const struct dc_plane_state *plane_state = pipe_ctx->plane_state; struct dc_crtc_timing *timing = &pipe_ctx->stream->timing; - struct rect recout_full = { 0 }; bool res = false; DC_LOGGER_INIT(pipe_ctx->stream->ctx->logger); /* Important: scaling ratio calculation requires pixel format, @@ -1105,7 +935,7 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) if (pipe_ctx->plane_res.scl_data.viewport.height < 16 || pipe_ctx->plane_res.scl_data.viewport.width < 16) return false; - calculate_recout(pipe_ctx, &recout_full); + calculate_recout(pipe_ctx); /** * Setting line buffer pixel depth to 24bpp yields banding @@ -1146,7 +976,7 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) if (res) /* May need to re-check lb size after this in some obscure scenario */ - calculate_inits_and_adj_vp(pipe_ctx, &recout_full); + calculate_inits_and_adj_vp(pipe_ctx); DC_LOG_SCALER( "%s: Viewport:\nheight:%d width:%d x:%d " @@ -1356,6 +1186,9 @@ bool dc_add_plane_to_context( return false; } + tail_pipe = resource_get_tail_pipe_for_stream(&context->res_ctx, stream); + ASSERT(tail_pipe); + free_pipe = acquire_free_pipe_for_stream(context, pool, stream); #if defined(CONFIG_DRM_AMD_DC_DCN1_0) @@ -1373,10 +1206,6 @@ bool dc_add_plane_to_context( free_pipe->plane_state = plane_state; if (head_pipe != free_pipe) { - - tail_pipe = resource_get_tail_pipe_for_stream(&context->res_ctx, stream); - ASSERT(tail_pipe); - free_pipe->stream_res.tg = tail_pipe->stream_res.tg; free_pipe->stream_res.abm = tail_pipe->stream_res.abm; free_pipe->stream_res.opp = tail_pipe->stream_res.opp; @@ -1622,6 +1451,14 @@ static bool are_stream_backends_same( return true; } +/** + * dc_is_stream_unchanged() - Compare two stream states for equivalence. + * + * Checks if there a difference between the two states + * that would require a mode change. + * + * Does not compare cursor position or attributes. + */ bool dc_is_stream_unchanged( struct dc_stream_state *old_stream, struct dc_stream_state *stream) { @@ -1632,6 +1469,9 @@ bool dc_is_stream_unchanged( return true; } +/** + * dc_is_stream_scaling_unchanged() - Compare scaling rectangles of two streams. + */ bool dc_is_stream_scaling_unchanged( struct dc_stream_state *old_stream, struct dc_stream_state *stream) { @@ -1791,16 +1631,19 @@ bool resource_is_stream_unchanged( return false; } +/** + * dc_add_stream_to_ctx() - Add a new dc_stream_state to a dc_state. + */ enum dc_status dc_add_stream_to_ctx( struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *stream) { - struct dc_context *dc_ctx = dc->ctx; enum dc_status res; + DC_LOGGER_INIT(dc->ctx->logger); if (new_ctx->stream_count >= dc->res_pool->timing_generator_count) { - DC_ERROR("Max streams reached, can't add stream %p !\n", stream); + DC_LOG_WARNING("Max streams reached, can't add stream %p !\n", stream); return DC_ERROR_UNEXPECTED; } @@ -1810,11 +1653,14 @@ enum dc_status dc_add_stream_to_ctx( res = dc->res_pool->funcs->add_stream_to_ctx(dc, new_ctx, stream); if (res != DC_OK) - DC_ERROR("Adding stream %p to context failed with err %d!\n", stream, res); + DC_LOG_WARNING("Adding stream %p to context failed with err %d!\n", stream, res); return res; } +/** + * dc_remove_stream_from_ctx() - Remove a stream from a dc_state. + */ enum dc_status dc_remove_stream_from_ctx( struct dc *dc, struct dc_state *new_ctx, @@ -1976,6 +1822,8 @@ enum dc_status resource_map_pool_resources( } */ + calculate_phy_pix_clks(stream); + /* acquire new resources */ pipe_idx = acquire_first_free_pipe(&context->res_ctx, pool, stream); @@ -2033,6 +1881,12 @@ enum dc_status resource_map_pool_resources( return DC_ERROR_UNEXPECTED; } +/** + * dc_resource_state_copy_construct_current() - Creates a new dc_state from existing state + * Is a shallow copy. Increments refcounts on existing streams and planes. + * @dc: copy out of dc->current_state + * @dst_ctx: copy into this + */ void dc_resource_state_copy_construct_current( const struct dc *dc, struct dc_state *dst_ctx) @@ -2048,6 +1902,14 @@ void dc_resource_state_construct( dst_ctx->dccg = dc->res_pool->clk_mgr; } +/** + * dc_validate_global_state() - Determine if HW can support a given state + * Checks HW resource availability and bandwidth requirement. + * @dc: dc struct for this driver + * @new_ctx: state to be validated + * + * Return: DC_OK if the result can be programmed. Otherwise, an error code. + */ enum dc_status dc_validate_global_state( struct dc *dc, struct dc_state *new_ctx) @@ -2375,113 +2237,15 @@ static void set_vendor_info_packet( struct dc_info_packet *info_packet, struct dc_stream_state *stream) { - uint32_t length = 0; - bool hdmi_vic_mode = false; - uint8_t checksum = 0; - uint32_t i = 0; - enum dc_timing_3d_format format; - // Can be different depending on packet content /*todo*/ - // unsigned int length = pPathMode->dolbyVision ? 24 : 5; - - info_packet->valid = false; - - format = stream->timing.timing_3d_format; - if (stream->view_format == VIEW_3D_FORMAT_NONE) - format = TIMING_3D_FORMAT_NONE; - - /* Can be different depending on packet content */ - length = 5; - - if (stream->timing.hdmi_vic != 0 - && stream->timing.h_total >= 3840 - && stream->timing.v_total >= 2160) - hdmi_vic_mode = true; - - /* According to HDMI 1.4a CTS, VSIF should be sent - * for both 3D stereo and HDMI VIC modes. - * For all other modes, there is no VSIF sent. */ + /* SPD info packet for FreeSync */ - if (format == TIMING_3D_FORMAT_NONE && !hdmi_vic_mode) + /* Check if Freesync is supported. Return if false. If true, + * set the corresponding bit in the info packet + */ + if (!stream->vsp_infopacket.valid) return; - /* 24bit IEEE Registration identifier (0x000c03). LSB first. */ - info_packet->sb[1] = 0x03; - info_packet->sb[2] = 0x0C; - info_packet->sb[3] = 0x00; - - /*PB4: 5 lower bytes = 0 (reserved). 3 higher bits = HDMI_Video_Format. - * The value for HDMI_Video_Format are: - * 0x0 (0b000) - No additional HDMI video format is presented in this - * packet - * 0x1 (0b001) - Extended resolution format present. 1 byte of HDMI_VIC - * parameter follows - * 0x2 (0b010) - 3D format indication present. 3D_Structure and - * potentially 3D_Ext_Data follows - * 0x3..0x7 (0b011..0b111) - reserved for future use */ - if (format != TIMING_3D_FORMAT_NONE) - info_packet->sb[4] = (2 << 5); - else if (hdmi_vic_mode) - info_packet->sb[4] = (1 << 5); - - /* PB5: If PB4 claims 3D timing (HDMI_Video_Format = 0x2): - * 4 lower bites = 0 (reserved). 4 higher bits = 3D_Structure. - * The value for 3D_Structure are: - * 0x0 - Frame Packing - * 0x1 - Field Alternative - * 0x2 - Line Alternative - * 0x3 - Side-by-Side (full) - * 0x4 - L + depth - * 0x5 - L + depth + graphics + graphics-depth - * 0x6 - Top-and-Bottom - * 0x7 - Reserved for future use - * 0x8 - Side-by-Side (Half) - * 0x9..0xE - Reserved for future use - * 0xF - Not used */ - switch (format) { - case TIMING_3D_FORMAT_HW_FRAME_PACKING: - case TIMING_3D_FORMAT_SW_FRAME_PACKING: - info_packet->sb[5] = (0x0 << 4); - break; - - case TIMING_3D_FORMAT_SIDE_BY_SIDE: - case TIMING_3D_FORMAT_SBS_SW_PACKED: - info_packet->sb[5] = (0x8 << 4); - length = 6; - break; - - case TIMING_3D_FORMAT_TOP_AND_BOTTOM: - case TIMING_3D_FORMAT_TB_SW_PACKED: - info_packet->sb[5] = (0x6 << 4); - break; - - default: - break; - } - - /*PB5: If PB4 is set to 0x1 (extended resolution format) - * fill PB5 with the correct HDMI VIC code */ - if (hdmi_vic_mode) - info_packet->sb[5] = stream->timing.hdmi_vic; - - /* Header */ - info_packet->hb0 = HDMI_INFOFRAME_TYPE_VENDOR; /* VSIF packet type. */ - info_packet->hb1 = 0x01; /* Version */ - - /* 4 lower bits = Length, 4 higher bits = 0 (reserved) */ - info_packet->hb2 = (uint8_t) (length); - - /* Calculate checksum */ - checksum = 0; - checksum += info_packet->hb0; - checksum += info_packet->hb1; - checksum += info_packet->hb2; - - for (i = 1; i <= length; i++) - checksum += info_packet->sb[i]; - - info_packet->sb[0] = (uint8_t) (0x100 - checksum); - - info_packet->valid = true; + *info_packet = stream->vsp_infopacket; } static void set_spd_info_packet( @@ -2537,10 +2301,6 @@ void dc_resource_state_destruct(struct dc_state *context) } } -/* - * Copy src_ctx into dst_ctx and retain all surfaces and streams referenced - * by the src_ctx - */ void dc_resource_state_copy_construct( const struct dc_state *src_ctx, struct dc_state *dst_ctx) diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c index e113439aaa86..66e5c4623a49 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c @@ -100,8 +100,6 @@ static void construct(struct dc_stream_state *stream, /* EDID CAP translation for HDMI 2.0 */ stream->timing.flags.LTE_340MCSC_SCRAMBLE = dc_sink_data->edid_caps.lte_340mcsc_scramble; - stream->status.link = stream->sink->link; - update_stream_signal(stream); stream->out_transfer_func = dc_create_transfer_func(); @@ -172,7 +170,7 @@ struct dc_stream_status *dc_stream_get_status( } /** - * Update the cursor attributes and set cursor surface address + * dc_stream_set_cursor_attributes() - Update cursor attributes and set cursor surface address */ bool dc_stream_set_cursor_attributes( struct dc_stream_state *stream, diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index d16a20c84792..4b5bbb13ce7f 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -36,9 +36,10 @@ #include "inc/hw_sequencer.h" #include "inc/compressor.h" +#include "inc/hw/dmcu.h" #include "dml/display_mode_lib.h" -#define DC_VER "3.2.04" +#define DC_VER "3.2.08" #define MAX_SURFACES 3 #define MAX_STREAMS 6 @@ -47,13 +48,6 @@ /******************************************************************************* * Display Core Interfaces ******************************************************************************/ -struct dmcu_version { - unsigned int date; - unsigned int month; - unsigned int year; - unsigned int interface_version; -}; - struct dc_versions { const char *dc_ver; struct dmcu_version dmcu_version; @@ -748,5 +742,6 @@ void dc_set_power_state( struct dc *dc, enum dc_acpi_cm_power_state power_state); void dc_resume(struct dc *dc); +bool dc_is_dmcu_initialized(struct dc *dc); #endif /* DC_INTERFACE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_helper.c b/drivers/gpu/drm/amd/display/dc/dc_helper.c index fcfd50b5dba0..4842d2378bbf 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_helper.c +++ b/drivers/gpu/drm/amd/display/dc/dc_helper.c @@ -234,14 +234,14 @@ uint32_t generic_reg_wait(const struct dc_context *ctx, if (field_value == condition_value) { if (i * delay_between_poll_us > 1000 && !IS_FPGA_MAXIMUS_DC(ctx->dce_environment)) - dm_output_to_console("REG_WAIT taking a while: %dms in %s line:%d\n", + DC_LOG_DC("REG_WAIT taking a while: %dms in %s line:%d\n", delay_between_poll_us * i / 1000, func_name, line); return reg_val; } } - dm_error("REG_WAIT timeout %dus * %d tries - %s line:%d\n", + DC_LOG_WARNING("REG_WAIT timeout %dus * %d tries - %s line:%d\n", delay_between_poll_us, time_out_num_tries, func_name, line); diff --git a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h index 7825e4b5e97c..e72fce4eca65 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h @@ -192,7 +192,6 @@ enum surface_pixel_format { /*swaped & float*/ SURFACE_PIXEL_FORMAT_GRPH_ABGR16161616F, /*grow graphics here if necessary */ - SURFACE_PIXEL_FORMAT_VIDEO_AYCrCb8888, SURFACE_PIXEL_FORMAT_VIDEO_BEGIN, SURFACE_PIXEL_FORMAT_VIDEO_420_YCbCr = SURFACE_PIXEL_FORMAT_VIDEO_BEGIN, @@ -200,6 +199,7 @@ enum surface_pixel_format { SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCbCr, SURFACE_PIXEL_FORMAT_VIDEO_420_10bpc_YCrCb, SURFACE_PIXEL_FORMAT_SUBSAMPLE_END, + SURFACE_PIXEL_FORMAT_VIDEO_AYCrCb8888, SURFACE_PIXEL_FORMAT_INVALID /*grow 444 video here if necessary */ @@ -358,15 +358,16 @@ union dc_tiling_info { } gfx8; struct { + enum swizzle_mode_values swizzle; unsigned int num_pipes; - unsigned int num_banks; + unsigned int max_compressed_frags; unsigned int pipe_interleave; + + unsigned int num_banks; unsigned int num_shader_engines; unsigned int num_rb_per_se; - unsigned int max_compressed_frags; bool shaderEnable; - enum swizzle_mode_values swizzle; bool meta_linear; bool rb_aligned; bool pipe_aligned; diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h index 8738f27a8708..29f19d57ff7a 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_link.h +++ b/drivers/gpu/drm/amd/display/dc/dc_link.h @@ -128,8 +128,10 @@ struct dc_link { const struct dc_link_status *dc_link_get_status(const struct dc_link *dc_link); -/* - * Return an enumerated dc_link. dc_link order is constant and determined at +/** + * dc_get_link_at_index() - Return an enumerated dc_link. + * + * dc_link order is constant and determined at * boot time. They cannot be created or destroyed. * Use dc_get_caps() to get number of links. */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_stream.h b/drivers/gpu/drm/amd/display/dc/dc_stream.h index c5bd1fbb6982..be34d638e15d 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_stream.h +++ b/drivers/gpu/drm/amd/display/dc/dc_stream.h @@ -56,6 +56,7 @@ struct dc_stream_state { struct dc_crtc_timing_adjust adjust; struct dc_info_packet vrr_infopacket; struct dc_info_packet vsc_infopacket; + struct dc_info_packet vsp_infopacket; struct rect src; /* composition area */ struct rect dst; /* stream addressable area */ @@ -104,8 +105,6 @@ struct dc_stream_state { bool dpms_off; bool apply_edp_fast_boot_optimization; - struct dc_stream_status status; - struct dc_cursor_attributes cursor_attributes; struct dc_cursor_position cursor_position; uint32_t sdr_white_level; // for boosting (SDR) cursor in HDR mode @@ -131,11 +130,13 @@ struct dc_stream_update { struct dc_crtc_timing_adjust *adjust; struct dc_info_packet *vrr_infopacket; struct dc_info_packet *vsc_infopacket; + struct dc_info_packet *vsp_infopacket; bool *dpms_off; struct colorspace_transform *gamut_remap; enum dc_color_space *output_color_space; + enum dc_dither_option *dither_option; struct dc_csc_transform *output_csc_transform; diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index 6e12d640d020..0b20ae23f169 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -73,10 +73,18 @@ struct hw_asic_id { void *atombios_base_address; }; +struct dc_perf_trace { + unsigned long read_count; + unsigned long write_count; + unsigned long last_entry_read; + unsigned long last_entry_write; +}; + struct dc_context { struct dc *dc; void *driver_context; /* e.g. amdgpu_device */ + struct dc_perf_trace *perf_trace; void *cgs_device; enum dce_environment dce_environment; @@ -191,7 +199,6 @@ union display_content_support { }; struct dc_panel_patch { - unsigned int disconnect_delay; unsigned int dppowerup_delay; unsigned int extra_t12_ms; }; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c index 9a28a04417d1..afd287f08bc9 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c @@ -94,7 +94,7 @@ static const struct state_dependent_clocks dce120_max_clks_by_state[] = { /*ClocksStatePerformance*/ { .display_clk_khz = 1133000, .pixel_clk_khz = 600000 } }; -static int dentist_get_divider_from_did(int did) +int dentist_get_divider_from_did(int did) { if (did < DENTIST_BASE_DID_1) did = DENTIST_BASE_DID_1; @@ -277,7 +277,8 @@ static int dce_set_clock( if (requested_clk_khz == 0) clk_mgr_dce->cur_min_clks_state = DM_PP_CLOCKS_STATE_NOMINAL; - dmcu->funcs->set_psr_wait_loop(dmcu, actual_clock / 1000 / 7); + if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) + dmcu->funcs->set_psr_wait_loop(dmcu, actual_clock / 1000 / 7); return actual_clock; } @@ -324,9 +325,11 @@ int dce112_set_clock(struct clk_mgr *clk_mgr, int requested_clk_khz) bp->funcs->set_dce_clock(bp, &dce_clk_params); if (!IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) { - if (clk_mgr_dce->dfs_bypass_disp_clk != actual_clock) - dmcu->funcs->set_psr_wait_loop(dmcu, - actual_clock / 1000 / 7); + if (dmcu && dmcu->funcs->is_dmcu_initialized(dmcu)) { + if (clk_mgr_dce->dfs_bypass_disp_clk != actual_clock) + dmcu->funcs->set_psr_wait_loop(dmcu, + actual_clock / 1000 / 7); + } } clk_mgr_dce->dfs_bypass_disp_clk = actual_clock; @@ -588,6 +591,8 @@ static void dce11_pplib_apply_display_requirements( dc, context->bw.dce.sclk_khz); + pp_display_cfg->min_dcfclock_khz = pp_display_cfg->min_engine_clock_khz; + pp_display_cfg->min_engine_clock_deep_sleep_khz = context->bw.dce.sclk_deep_sleep_khz; @@ -671,6 +676,11 @@ static void dce112_update_clocks(struct clk_mgr *clk_mgr, { struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr); struct dm_pp_power_level_change_request level_change_req; + int unpatched_disp_clk = context->bw.dce.dispclk_khz; + + /*TODO: W/A for dal3 linux, investigate why this works */ + if (!clk_mgr_dce->dfs_bypass_active) + context->bw.dce.dispclk_khz = context->bw.dce.dispclk_khz * 115 / 100; level_change_req.power_level = dce_get_required_clocks_state(clk_mgr, context); /* get max clock state from PPLIB */ @@ -685,6 +695,8 @@ static void dce112_update_clocks(struct clk_mgr *clk_mgr, clk_mgr->clks.dispclk_khz = context->bw.dce.dispclk_khz; } dce11_pplib_apply_display_requirements(clk_mgr->ctx->dc, context); + + context->bw.dce.dispclk_khz = unpatched_disp_clk; } static void dce12_update_clocks(struct clk_mgr *clk_mgr, diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.h index 046077797416..3bceb31d910d 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.h @@ -165,4 +165,6 @@ struct clk_mgr *dce120_clk_mgr_create(struct dc_context *ctx); void dce_clk_mgr_destroy(struct clk_mgr **clk_mgr); +int dentist_get_divider_from_did(int did); + #endif /* _DCE_CLK_MGR_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c index c47c81883d3c..cce0d18f91da 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_stream_encoder.c @@ -908,7 +908,6 @@ static void dce110_stream_encoder_dp_blank( struct stream_encoder *enc) { struct dce110_stream_encoder *enc110 = DCE110STRENC_FROM_STRENC(enc); - uint32_t retries = 0; uint32_t reg1 = 0; uint32_t max_retries = DP_BLANK_MAX_RETRY * 10; @@ -926,30 +925,28 @@ static void dce110_stream_encoder_dp_blank( * (2 = start of the next vertical blank) */ REG_UPDATE(DP_VID_STREAM_CNTL, DP_VID_STREAM_DIS_DEFER, 2); /* Larger delay to wait until VBLANK - use max retry of - * 10us*3000=30ms. This covers 16.6ms of typical 60 Hz mode + - * a little more because we may not trust delay accuracy. - */ + * 10us*3000=30ms. This covers 16.6ms of typical 60 Hz mode + + * a little more because we may not trust delay accuracy. + */ max_retries = DP_BLANK_MAX_RETRY * 150; /* disable DP stream */ REG_UPDATE(DP_VID_STREAM_CNTL, DP_VID_STREAM_ENABLE, 0); /* the encoder stops sending the video stream - * at the start of the vertical blanking. - * Poll for DP_VID_STREAM_STATUS == 0 - */ + * at the start of the vertical blanking. + * Poll for DP_VID_STREAM_STATUS == 0 + */ REG_WAIT(DP_VID_STREAM_CNTL, DP_VID_STREAM_STATUS, 0, 10, max_retries); - ASSERT(retries <= max_retries); - /* Tell the DP encoder to ignore timing from CRTC, must be done after - * the polling. If we set DP_STEER_FIFO_RESET before DP stream blank is - * complete, stream status will be stuck in video stream enabled state, - * i.e. DP_VID_STREAM_STATUS stuck at 1. - */ + * the polling. If we set DP_STEER_FIFO_RESET before DP stream blank is + * complete, stream status will be stuck in video stream enabled state, + * i.e. DP_VID_STREAM_STATUS stuck at 1. + */ REG_UPDATE(DP_STEER_FIFO, DP_STEER_FIFO_RESET, true); } diff --git a/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.c index bc50a8e25f4f..87771676acac 100644 --- a/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.c @@ -117,6 +117,18 @@ void dce100_prepare_bandwidth( false); } +void dce100_optimize_bandwidth( + struct dc *dc, + struct dc_state *context) +{ + dce110_set_safe_displaymarks(&context->res_ctx, dc->res_pool); + + dc->res_pool->clk_mgr->funcs->update_clocks( + dc->res_pool->clk_mgr, + context, + true); +} + /**************************************************************************/ void dce100_hw_sequencer_construct(struct dc *dc) @@ -125,6 +137,6 @@ void dce100_hw_sequencer_construct(struct dc *dc) dc->hwss.enable_display_power_gating = dce100_enable_display_power_gating; dc->hwss.prepare_bandwidth = dce100_prepare_bandwidth; - dc->hwss.optimize_bandwidth = dce100_prepare_bandwidth; + dc->hwss.optimize_bandwidth = dce100_optimize_bandwidth; } diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_compressor.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_compressor.c index 1f7f25013217..52d50e24a995 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_compressor.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_compressor.c @@ -64,65 +64,37 @@ static const struct dce110_compressor_reg_offsets reg_offsets[] = { static const uint32_t dce11_one_lpt_channel_max_resolution = 2560 * 1600; -enum fbc_idle_force { - /* Bit 0 - Display registers updated */ - FBC_IDLE_FORCE_DISPLAY_REGISTER_UPDATE = 0x00000001, - - /* Bit 2 - FBC_GRPH_COMP_EN register updated */ - FBC_IDLE_FORCE_GRPH_COMP_EN = 0x00000002, - /* Bit 3 - FBC_SRC_SEL register updated */ - FBC_IDLE_FORCE_SRC_SEL_CHANGE = 0x00000004, - /* Bit 4 - FBC_MIN_COMPRESSION register updated */ - FBC_IDLE_FORCE_MIN_COMPRESSION_CHANGE = 0x00000008, - /* Bit 5 - FBC_ALPHA_COMP_EN register updated */ - FBC_IDLE_FORCE_ALPHA_COMP_EN = 0x00000010, - /* Bit 6 - FBC_ZERO_ALPHA_CHUNK_SKIP_EN register updated */ - FBC_IDLE_FORCE_ZERO_ALPHA_CHUNK_SKIP_EN = 0x00000020, - /* Bit 7 - FBC_FORCE_COPY_TO_COMP_BUF register updated */ - FBC_IDLE_FORCE_FORCE_COPY_TO_COMP_BUF = 0x00000040, - - /* Bit 24 - Memory write to region 0 defined by MC registers. */ - FBC_IDLE_FORCE_MEMORY_WRITE_TO_REGION0 = 0x01000000, - /* Bit 25 - Memory write to region 1 defined by MC registers */ - FBC_IDLE_FORCE_MEMORY_WRITE_TO_REGION1 = 0x02000000, - /* Bit 26 - Memory write to region 2 defined by MC registers */ - FBC_IDLE_FORCE_MEMORY_WRITE_TO_REGION2 = 0x04000000, - /* Bit 27 - Memory write to region 3 defined by MC registers. */ - FBC_IDLE_FORCE_MEMORY_WRITE_TO_REGION3 = 0x08000000, - - /* Bit 28 - Memory write from any client other than MCIF */ - FBC_IDLE_FORCE_MEMORY_WRITE_OTHER_THAN_MCIF = 0x10000000, - /* Bit 29 - CG statics screen signal is inactive */ - FBC_IDLE_FORCE_CG_STATIC_SCREEN_IS_INACTIVE = 0x20000000, -}; - - static uint32_t align_to_chunks_number_per_line(uint32_t pixels) { return 256 * ((pixels + 255) / 256); } -static void reset_lb_on_vblank(struct dc_context *ctx) +static void reset_lb_on_vblank(struct compressor *compressor, uint32_t crtc_inst) { - uint32_t value, frame_count; + uint32_t value; + uint32_t frame_count; + uint32_t status_pos; uint32_t retry = 0; - uint32_t status_pos = - dm_read_reg(ctx, mmCRTC_STATUS_POSITION); + struct dce110_compressor *cp110 = TO_DCE110_COMPRESSOR(compressor); + + cp110->offsets = reg_offsets[crtc_inst]; + + status_pos = dm_read_reg(compressor->ctx, DCP_REG(mmCRTC_STATUS_POSITION)); /* Only if CRTC is enabled and counter is moving we wait for one frame. */ - if (status_pos != dm_read_reg(ctx, mmCRTC_STATUS_POSITION)) { + if (status_pos != dm_read_reg(compressor->ctx, DCP_REG(mmCRTC_STATUS_POSITION))) { /* Resetting LB on VBlank */ - value = dm_read_reg(ctx, mmLB_SYNC_RESET_SEL); + value = dm_read_reg(compressor->ctx, DCP_REG(mmLB_SYNC_RESET_SEL)); set_reg_field_value(value, 3, LB_SYNC_RESET_SEL, LB_SYNC_RESET_SEL); set_reg_field_value(value, 1, LB_SYNC_RESET_SEL, LB_SYNC_RESET_SEL2); - dm_write_reg(ctx, mmLB_SYNC_RESET_SEL, value); + dm_write_reg(compressor->ctx, DCP_REG(mmLB_SYNC_RESET_SEL), value); - frame_count = dm_read_reg(ctx, mmCRTC_STATUS_FRAME_COUNT); + frame_count = dm_read_reg(compressor->ctx, DCP_REG(mmCRTC_STATUS_FRAME_COUNT)); for (retry = 10000; retry > 0; retry--) { - if (frame_count != dm_read_reg(ctx, mmCRTC_STATUS_FRAME_COUNT)) + if (frame_count != dm_read_reg(compressor->ctx, DCP_REG(mmCRTC_STATUS_FRAME_COUNT))) break; udelay(10); } @@ -130,13 +102,11 @@ static void reset_lb_on_vblank(struct dc_context *ctx) dm_error("Frame count did not increase for 100ms.\n"); /* Resetting LB on VBlank */ - value = dm_read_reg(ctx, mmLB_SYNC_RESET_SEL); + value = dm_read_reg(compressor->ctx, DCP_REG(mmLB_SYNC_RESET_SEL)); set_reg_field_value(value, 2, LB_SYNC_RESET_SEL, LB_SYNC_RESET_SEL); set_reg_field_value(value, 0, LB_SYNC_RESET_SEL, LB_SYNC_RESET_SEL2); - dm_write_reg(ctx, mmLB_SYNC_RESET_SEL, value); - + dm_write_reg(compressor->ctx, DCP_REG(mmLB_SYNC_RESET_SEL), value); } - } static void wait_for_fbc_state_changed( @@ -226,10 +196,10 @@ void dce110_compressor_enable_fbc( uint32_t addr; uint32_t value, misc_value; - addr = mmFBC_CNTL; value = dm_read_reg(compressor->ctx, addr); set_reg_field_value(value, 1, FBC_CNTL, FBC_GRPH_COMP_EN); + /* params->inst is valid HW CRTC instance start from 0 */ set_reg_field_value( value, params->inst, @@ -238,8 +208,10 @@ void dce110_compressor_enable_fbc( /* Keep track of enum controller_id FBC is attached to */ compressor->is_enabled = true; - compressor->attached_inst = params->inst; - cp110->offsets = reg_offsets[params->inst]; + /* attached_inst is SW CRTC instance start from 1 + * 0 = CONTROLLER_ID_UNDEFINED means not attached crtc + */ + compressor->attached_inst = params->inst + CONTROLLER_ID_D0; /* Toggle it as there is bug in HW */ set_reg_field_value(value, 0, FBC_CNTL, FBC_GRPH_COMP_EN); @@ -268,9 +240,10 @@ void dce110_compressor_enable_fbc( void dce110_compressor_disable_fbc(struct compressor *compressor) { struct dce110_compressor *cp110 = TO_DCE110_COMPRESSOR(compressor); + uint32_t crtc_inst = 0; if (compressor->options.bits.FBC_SUPPORT) { - if (dce110_compressor_is_fbc_enabled_in_hw(compressor, NULL)) { + if (dce110_compressor_is_fbc_enabled_in_hw(compressor, &crtc_inst)) { uint32_t reg_data; /* Turn off compression */ reg_data = dm_read_reg(compressor->ctx, mmFBC_CNTL); @@ -284,8 +257,10 @@ void dce110_compressor_disable_fbc(struct compressor *compressor) wait_for_fbc_state_changed(cp110, false); } - /* Sync line buffer - dce100/110 only*/ - reset_lb_on_vblank(compressor->ctx); + /* Sync line buffer which fbc was attached to dce100/110 only */ + if (crtc_inst > CONTROLLER_ID_UNDEFINED && crtc_inst < CONTROLLER_ID_D3) + reset_lb_on_vblank(compressor, + crtc_inst - CONTROLLER_ID_D0); } } @@ -328,6 +303,8 @@ void dce110_compressor_program_compressed_surface_address_and_pitch( uint32_t compressed_surf_address_low_part = compressor->compr_surface_address.addr.low_part; + cp110->offsets = reg_offsets[params->inst]; + /* Clear content first. */ dm_write_reg( compressor->ctx, @@ -410,13 +387,7 @@ void dce110_compressor_set_fbc_invalidation_triggers( value = dm_read_reg(compressor->ctx, addr); set_reg_field_value( value, - fbc_trigger | - FBC_IDLE_FORCE_GRPH_COMP_EN | - FBC_IDLE_FORCE_SRC_SEL_CHANGE | - FBC_IDLE_FORCE_MIN_COMPRESSION_CHANGE | - FBC_IDLE_FORCE_ALPHA_COMP_EN | - FBC_IDLE_FORCE_ZERO_ALPHA_CHUNK_SKIP_EN | - FBC_IDLE_FORCE_FORCE_COPY_TO_COMP_BUF, + fbc_trigger, FBC_IDLE_FORCE_CLEAR_MASK, FBC_IDLE_FORCE_CLEAR_MASK); dm_write_reg(compressor->ctx, addr, value); @@ -549,7 +520,7 @@ void dce110_compressor_construct(struct dce110_compressor *compressor, compressor->base.channel_interleave_size = 0; compressor->base.dram_channels_num = 0; compressor->base.lpt_channels_num = 0; - compressor->base.attached_inst = 0; + compressor->base.attached_inst = CONTROLLER_ID_UNDEFINED; compressor->base.is_enabled = false; compressor->base.funcs = &dce110_compressor_funcs; diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c index 9724a17e352b..4bf24758217f 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c @@ -1267,10 +1267,19 @@ static void program_scaler(const struct dc *dc, pipe_ctx->plane_res.scl_data.lb_params.depth, &pipe_ctx->stream->bit_depth_params); - if (pipe_ctx->stream_res.tg->funcs->set_overscan_blank_color) + if (pipe_ctx->stream_res.tg->funcs->set_overscan_blank_color) { + /* + * The way 420 is packed, 2 channels carry Y component, 1 channel + * alternate between Cb and Cr, so both channels need the pixel + * value for Y + */ + if (pipe_ctx->stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR420) + color.color_r_cr = color.color_g_y; + pipe_ctx->stream_res.tg->funcs->set_overscan_blank_color( pipe_ctx->stream_res.tg, &color); + } pipe_ctx->plane_res.xfm->funcs->transform_set_scaler(pipe_ctx->plane_res.xfm, &pipe_ctx->plane_res.scl_data); @@ -1766,12 +1775,13 @@ static void set_static_screen_control(struct pipe_ctx **pipe_ctx, * Check if FBC can be enabled */ static bool should_enable_fbc(struct dc *dc, - struct dc_state *context, - uint32_t *pipe_idx) + struct dc_state *context, + uint32_t *pipe_idx) { uint32_t i; struct pipe_ctx *pipe_ctx = NULL; struct resource_context *res_ctx = &context->res_ctx; + unsigned int underlay_idx = dc->res_pool->underlay_pipe_index; ASSERT(dc->fbc_compressor); @@ -1786,14 +1796,28 @@ static bool should_enable_fbc(struct dc *dc, for (i = 0; i < dc->res_pool->pipe_count; i++) { if (res_ctx->pipe_ctx[i].stream) { + pipe_ctx = &res_ctx->pipe_ctx[i]; - *pipe_idx = i; - break; + + if (!pipe_ctx) + continue; + + /* fbc not applicable on underlay pipe */ + if (pipe_ctx->pipe_idx != underlay_idx) { + *pipe_idx = i; + break; + } } } - /* Pipe context should be found */ - ASSERT(pipe_ctx); + if (i == dc->res_pool->pipe_count) + return false; + + if (!pipe_ctx->stream->sink) + return false; + + if (!pipe_ctx->stream->sink->link) + return false; /* Only supports eDP */ if (pipe_ctx->stream->sink->link->connector_signal != SIGNAL_TYPE_EDP) @@ -1817,8 +1841,9 @@ static bool should_enable_fbc(struct dc *dc, /* * Enable FBC */ -static void enable_fbc(struct dc *dc, - struct dc_state *context) +static void enable_fbc( + struct dc *dc, + struct dc_state *context) { uint32_t pipe_idx = 0; @@ -1828,10 +1853,9 @@ static void enable_fbc(struct dc *dc, struct compressor *compr = dc->fbc_compressor; struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[pipe_idx]; - params.source_view_width = pipe_ctx->stream->timing.h_addressable; params.source_view_height = pipe_ctx->stream->timing.v_addressable; - + params.inst = pipe_ctx->stream_res.tg->inst; compr->compr_surface_address.quad_part = dc->ctx->fbc_gpu_addr; compr->funcs->surface_address_and_pitch(compr, ¶ms); @@ -2046,10 +2070,10 @@ enum dc_status dce110_apply_ctx_to_hw( return status; } - dcb->funcs->set_scratch_critical_state(dcb, false); - if (dc->fbc_compressor) - enable_fbc(dc, context); + enable_fbc(dc, dc->current_state); + + dcb->funcs->set_scratch_critical_state(dcb, false); return DC_OK; } @@ -2282,7 +2306,7 @@ static void dce110_enable_per_frame_crtc_position_reset( int i; gsl_params.gsl_group = 0; - gsl_params.gsl_master = grouped_pipes[0]->stream->triggered_crtc_reset.event_source->status.primary_otg_inst; + gsl_params.gsl_master = 0; for (i = 0; i < group_size; i++) grouped_pipes[i]->stream_res.tg->funcs->setup_global_swap_lock( @@ -2408,7 +2432,6 @@ static void dce110_program_front_end_for_pipe( struct dc_plane_state *plane_state = pipe_ctx->plane_state; struct xfm_grph_csc_adjustment adjust; struct out_csc_color_matrix tbl_entry; - unsigned int underlay_idx = dc->res_pool->underlay_pipe_index; unsigned int i; DC_LOGGER_INIT(); memset(&tbl_entry, 0, sizeof(tbl_entry)); @@ -2449,15 +2472,6 @@ static void dce110_program_front_end_for_pipe( program_scaler(dc, pipe_ctx); - /* fbc not applicable on Underlay pipe */ - if (dc->fbc_compressor && old_pipe->stream && - pipe_ctx->pipe_idx != underlay_idx) { - if (plane_state->tiling_info.gfx8.array_mode == DC_ARRAY_LINEAR_GENERAL) - dc->fbc_compressor->funcs->disable_fbc(dc->fbc_compressor); - else - enable_fbc(dc, dc->current_state); - } - mi->funcs->mem_input_program_surface_config( mi, plane_state->format, @@ -2534,6 +2548,9 @@ static void dce110_apply_ctx_for_surface( if (num_planes == 0) return; + if (dc->fbc_compressor) + dc->fbc_compressor->funcs->disable_fbc(dc->fbc_compressor); + for (i = 0; i < dc->res_pool->pipe_count; i++) { struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; struct pipe_ctx *old_pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i]; @@ -2576,6 +2593,9 @@ static void dce110_apply_ctx_for_surface( (pipe_ctx->plane_state || old_pipe_ctx->plane_state)) dc->hwss.pipe_control_lock(dc, pipe_ctx, false); } + + if (dc->fbc_compressor) + enable_fbc(dc, dc->current_state); } static void dce110_power_down_fe(struct dc *dc, struct pipe_ctx *pipe_ctx) diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c index 6d40b3d54ac1..cdd1d6b7b9f2 100644 --- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c @@ -41,7 +41,6 @@ #include "dce/dce_mem_input.h" #include "dce/dce_link_encoder.h" #include "dce/dce_stream_encoder.h" -#include "dce/dce_mem_input.h" #include "dce/dce_ipp.h" #include "dce/dce_transform.h" #include "dce/dce_opp.h" diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c index 20f531d27e2b..54abedbf1b43 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c @@ -223,7 +223,7 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr, &dc->res_pool->pp_smu_req; struct pp_smu_display_requirement_rv smu_req = *smu_req_cur; struct pp_smu_funcs_rv *pp_smu = dc->res_pool->pp_smu; - struct dm_pp_clock_for_voltage_req clock_voltage_req = {0}; + uint32_t requested_dcf_clock_in_khz = 0; bool send_request_to_increase = false; bool send_request_to_lower = false; int display_count; @@ -263,8 +263,6 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr, // F Clock if (should_set_clock(safe_to_lower, new_clocks->fclk_khz, clk_mgr->clks.fclk_khz)) { clk_mgr->clks.fclk_khz = new_clocks->fclk_khz; - clock_voltage_req.clk_type = DM_PP_CLOCK_TYPE_FCLK; - clock_voltage_req.clocks_in_khz = new_clocks->fclk_khz; smu_req.hard_min_fclk_mhz = new_clocks->fclk_khz / 1000; notify_hard_min_fclk_to_smu(pp_smu, new_clocks->fclk_khz); @@ -293,10 +291,9 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr, */ if (send_request_to_increase) { /*use dcfclk to request voltage*/ - clock_voltage_req.clk_type = DM_PP_CLOCK_TYPE_DCFCLK; - clock_voltage_req.clocks_in_khz = dcn_find_dcfclk_suits_all(dc, new_clocks); + requested_dcf_clock_in_khz = dcn_find_dcfclk_suits_all(dc, new_clocks); - notify_hard_min_dcfclk_to_smu(pp_smu, clock_voltage_req.clocks_in_khz); + notify_hard_min_dcfclk_to_smu(pp_smu, requested_dcf_clock_in_khz); if (pp_smu->set_display_requirement) pp_smu->set_display_requirement(&pp_smu->pp_smu, &smu_req); @@ -317,10 +314,9 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr, if (!send_request_to_increase && send_request_to_lower) { /*use dcfclk to request voltage*/ - clock_voltage_req.clk_type = DM_PP_CLOCK_TYPE_DCFCLK; - clock_voltage_req.clocks_in_khz = dcn_find_dcfclk_suits_all(dc, new_clocks); + requested_dcf_clock_in_khz = dcn_find_dcfclk_suits_all(dc, new_clocks); - notify_hard_min_dcfclk_to_smu(pp_smu, clock_voltage_req.clocks_in_khz); + notify_hard_min_dcfclk_to_smu(pp_smu, requested_dcf_clock_in_khz); if (pp_smu->set_display_requirement) pp_smu->set_display_requirement(&pp_smu->pp_smu, &smu_req); @@ -332,12 +328,10 @@ static void dcn1_update_clocks(struct clk_mgr *clk_mgr, *smu_req_cur = smu_req; } - static const struct clk_mgr_funcs dcn1_funcs = { .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz, .update_clocks = dcn1_update_clocks }; - struct clk_mgr *dcn1_clk_mgr_create(struct dc_context *ctx) { struct dc_debug_options *debug = &ctx->dc->debug; @@ -377,3 +371,5 @@ struct clk_mgr *dcn1_clk_mgr_create(struct dc_context *ctx) return &clk_mgr_dce->base; } + + diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.h index 9dbaf6578006..a995eda443a3 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.h @@ -28,6 +28,12 @@ #include "../dce/dce_clk_mgr.h" +struct clk_bypass { + uint32_t dcfclk_bypass; + uint32_t dispclk_pypass; + uint32_t dprefclk_bypass; +}; + void dcn1_pplib_apply_display_requirements( struct dc *dc, struct dc_state *context); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c index 3eea44092a04..7469333a2c8a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c @@ -324,7 +324,7 @@ bool cm_helper_translate_curve_to_hw_format( if (output_tf == NULL || lut_params == NULL || output_tf->type == TF_TYPE_BYPASS) return false; - PERF_TRACE(); + PERF_TRACE_CTX(output_tf->ctx); corner_points = lut_params->corner_points; rgb_resulted = lut_params->rgb_resulted; @@ -513,7 +513,7 @@ bool cm_helper_translate_curve_to_degamma_hw_format( if (output_tf == NULL || lut_params == NULL || output_tf->type == TF_TYPE_BYPASS) return false; - PERF_TRACE(); + PERF_TRACE_CTX(output_tf->ctx); corner_points = lut_params->corner_points; rgb_resulted = lut_params->rgb_resulted; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c index 4254e7e1a509..c7d1e678ebf5 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubbub.c @@ -100,7 +100,7 @@ bool hububu1_is_allow_self_refresh_enabled(struct hubbub *hubbub) REG_GET(DCHUBBUB_ARB_DRAM_STATE_CNTL, DCHUBBUB_ARB_ALLOW_SELF_REFRESH_FORCE_ENABLE, &enable); - return true ? false : enable; + return enable ? true : false; } diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c index 74132a1f3046..345af015d061 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.c @@ -99,6 +99,14 @@ static unsigned int hubp1_get_underflow_status(struct hubp *hubp) return hubp_underflow; } + +void hubp1_clear_underflow(struct hubp *hubp) +{ + struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp); + + REG_UPDATE(DCHUBP_CNTL, HUBP_UNDERFLOW_CLEAR, 1); +} + static void hubp1_set_hubp_blank_en(struct hubp *hubp, bool blank) { struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp); @@ -565,19 +573,6 @@ void hubp1_program_deadline( REFCYC_X_AFTER_SCALER, dlg_attr->refcyc_x_after_scaler, DST_Y_AFTER_SCALER, dlg_attr->dst_y_after_scaler); - if (REG(PREFETCH_SETTINS)) - REG_SET_2(PREFETCH_SETTINS, 0, - DST_Y_PREFETCH, dlg_attr->dst_y_prefetch, - VRATIO_PREFETCH, dlg_attr->vratio_prefetch); - else - REG_SET_2(PREFETCH_SETTINGS, 0, - DST_Y_PREFETCH, dlg_attr->dst_y_prefetch, - VRATIO_PREFETCH, dlg_attr->vratio_prefetch); - - REG_SET_2(VBLANK_PARAMETERS_0, 0, - DST_Y_PER_VM_VBLANK, dlg_attr->dst_y_per_vm_vblank, - DST_Y_PER_ROW_VBLANK, dlg_attr->dst_y_per_row_vblank); - REG_SET(REF_FREQ_TO_PIX_FREQ, 0, REF_FREQ_TO_PIX_FREQ, dlg_attr->ref_freq_to_pix_freq); @@ -585,9 +580,6 @@ void hubp1_program_deadline( REG_SET(VBLANK_PARAMETERS_1, 0, REFCYC_PER_PTE_GROUP_VBLANK_L, dlg_attr->refcyc_per_pte_group_vblank_l); - REG_SET(VBLANK_PARAMETERS_3, 0, - REFCYC_PER_META_CHUNK_VBLANK_L, dlg_attr->refcyc_per_meta_chunk_vblank_l); - if (REG(NOM_PARAMETERS_0)) REG_SET(NOM_PARAMETERS_0, 0, DST_Y_PER_PTE_ROW_NOM_L, dlg_attr->dst_y_per_pte_row_nom_l); @@ -602,27 +594,13 @@ void hubp1_program_deadline( REG_SET(NOM_PARAMETERS_5, 0, REFCYC_PER_META_CHUNK_NOM_L, dlg_attr->refcyc_per_meta_chunk_nom_l); - REG_SET_2(PER_LINE_DELIVERY_PRE, 0, - REFCYC_PER_LINE_DELIVERY_PRE_L, dlg_attr->refcyc_per_line_delivery_pre_l, - REFCYC_PER_LINE_DELIVERY_PRE_C, dlg_attr->refcyc_per_line_delivery_pre_c); - REG_SET_2(PER_LINE_DELIVERY, 0, REFCYC_PER_LINE_DELIVERY_L, dlg_attr->refcyc_per_line_delivery_l, REFCYC_PER_LINE_DELIVERY_C, dlg_attr->refcyc_per_line_delivery_c); - if (REG(PREFETCH_SETTINS_C)) - REG_SET(PREFETCH_SETTINS_C, 0, - VRATIO_PREFETCH_C, dlg_attr->vratio_prefetch_c); - else - REG_SET(PREFETCH_SETTINGS_C, 0, - VRATIO_PREFETCH_C, dlg_attr->vratio_prefetch_c); - REG_SET(VBLANK_PARAMETERS_2, 0, REFCYC_PER_PTE_GROUP_VBLANK_C, dlg_attr->refcyc_per_pte_group_vblank_c); - REG_SET(VBLANK_PARAMETERS_4, 0, - REFCYC_PER_META_CHUNK_VBLANK_C, dlg_attr->refcyc_per_meta_chunk_vblank_c); - if (REG(NOM_PARAMETERS_2)) REG_SET(NOM_PARAMETERS_2, 0, DST_Y_PER_PTE_ROW_NOM_C, dlg_attr->dst_y_per_pte_row_nom_c); @@ -642,10 +620,6 @@ void hubp1_program_deadline( QoS_LEVEL_LOW_WM, ttu_attr->qos_level_low_wm, QoS_LEVEL_HIGH_WM, ttu_attr->qos_level_high_wm); - REG_SET_2(DCN_GLOBAL_TTU_CNTL, 0, - MIN_TTU_VBLANK, ttu_attr->min_ttu_vblank, - QoS_LEVEL_FLIP, ttu_attr->qos_level_flip); - /* TTU - per luma/chroma */ /* Assumed surf0 is luma and 1 is chroma */ @@ -654,25 +628,15 @@ void hubp1_program_deadline( QoS_LEVEL_FIXED, ttu_attr->qos_level_fixed_l, QoS_RAMP_DISABLE, ttu_attr->qos_ramp_disable_l); - REG_SET(DCN_SURF0_TTU_CNTL1, 0, - REFCYC_PER_REQ_DELIVERY_PRE, - ttu_attr->refcyc_per_req_delivery_pre_l); - REG_SET_3(DCN_SURF1_TTU_CNTL0, 0, REFCYC_PER_REQ_DELIVERY, ttu_attr->refcyc_per_req_delivery_c, QoS_LEVEL_FIXED, ttu_attr->qos_level_fixed_c, QoS_RAMP_DISABLE, ttu_attr->qos_ramp_disable_c); - REG_SET(DCN_SURF1_TTU_CNTL1, 0, - REFCYC_PER_REQ_DELIVERY_PRE, - ttu_attr->refcyc_per_req_delivery_pre_c); - REG_SET_3(DCN_CUR0_TTU_CNTL0, 0, REFCYC_PER_REQ_DELIVERY, ttu_attr->refcyc_per_req_delivery_cur0, QoS_LEVEL_FIXED, ttu_attr->qos_level_fixed_cur0, QoS_RAMP_DISABLE, ttu_attr->qos_ramp_disable_cur0); - REG_SET(DCN_CUR0_TTU_CNTL1, 0, - REFCYC_PER_REQ_DELIVERY_PRE, ttu_attr->refcyc_per_req_delivery_pre_cur0); } static void hubp1_setup( @@ -690,6 +654,48 @@ static void hubp1_setup( hubp1_vready_workaround(hubp, pipe_dest); } +static void hubp1_setup_interdependent( + struct hubp *hubp, + struct _vcs_dpi_display_dlg_regs_st *dlg_attr, + struct _vcs_dpi_display_ttu_regs_st *ttu_attr) +{ + struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp); + + REG_SET_2(PREFETCH_SETTINS, 0, + DST_Y_PREFETCH, dlg_attr->dst_y_prefetch, + VRATIO_PREFETCH, dlg_attr->vratio_prefetch); + + REG_SET(PREFETCH_SETTINS_C, 0, + VRATIO_PREFETCH_C, dlg_attr->vratio_prefetch_c); + + REG_SET_2(VBLANK_PARAMETERS_0, 0, + DST_Y_PER_VM_VBLANK, dlg_attr->dst_y_per_vm_vblank, + DST_Y_PER_ROW_VBLANK, dlg_attr->dst_y_per_row_vblank); + + REG_SET(VBLANK_PARAMETERS_3, 0, + REFCYC_PER_META_CHUNK_VBLANK_L, dlg_attr->refcyc_per_meta_chunk_vblank_l); + + REG_SET(VBLANK_PARAMETERS_4, 0, + REFCYC_PER_META_CHUNK_VBLANK_C, dlg_attr->refcyc_per_meta_chunk_vblank_c); + + REG_SET_2(PER_LINE_DELIVERY_PRE, 0, + REFCYC_PER_LINE_DELIVERY_PRE_L, dlg_attr->refcyc_per_line_delivery_pre_l, + REFCYC_PER_LINE_DELIVERY_PRE_C, dlg_attr->refcyc_per_line_delivery_pre_c); + + REG_SET(DCN_SURF0_TTU_CNTL1, 0, + REFCYC_PER_REQ_DELIVERY_PRE, + ttu_attr->refcyc_per_req_delivery_pre_l); + REG_SET(DCN_SURF1_TTU_CNTL1, 0, + REFCYC_PER_REQ_DELIVERY_PRE, + ttu_attr->refcyc_per_req_delivery_pre_c); + REG_SET(DCN_CUR0_TTU_CNTL1, 0, + REFCYC_PER_REQ_DELIVERY_PRE, ttu_attr->refcyc_per_req_delivery_pre_cur0); + + REG_SET_2(DCN_GLOBAL_TTU_CNTL, 0, + MIN_TTU_VBLANK, ttu_attr->min_ttu_vblank, + QoS_LEVEL_FLIP, ttu_attr->qos_level_flip); +} + bool hubp1_is_flip_pending(struct hubp *hubp) { uint32_t flip_pending = 0; @@ -1178,6 +1184,7 @@ static const struct hubp_funcs dcn10_hubp_funcs = { hubp1_program_surface_config, .hubp_is_flip_pending = hubp1_is_flip_pending, .hubp_setup = hubp1_setup, + .hubp_setup_interdependent = hubp1_setup_interdependent, .hubp_set_vm_system_aperture_settings = hubp1_set_vm_system_aperture_settings, .hubp_set_vm_context0_settings = hubp1_set_vm_context0_settings, .set_blank = hubp1_set_blank, @@ -1190,6 +1197,7 @@ static const struct hubp_funcs dcn10_hubp_funcs = { .hubp_clk_cntl = hubp1_clk_cntl, .hubp_vtg_sel = hubp1_vtg_sel, .hubp_read_state = hubp1_read_state, + .hubp_clear_underflow = hubp1_clear_underflow, .hubp_disable_control = hubp1_disable_control, .hubp_get_underflow_status = hubp1_get_underflow_status, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h index 4890273b632b..62d4232e7796 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hubp.h @@ -251,6 +251,7 @@ HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_BLANK_EN, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_TTU_DISABLE, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_UNDERFLOW_STATUS, mask_sh),\ + HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_UNDERFLOW_CLEAR, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_NO_OUTSTANDING_REQ, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_VTG_SEL, mask_sh),\ HUBP_SF(HUBP0_DCHUBP_CNTL, HUBP_DISABLE, mask_sh),\ @@ -435,6 +436,7 @@ type HUBP_NO_OUTSTANDING_REQ;\ type HUBP_VTG_SEL;\ type HUBP_UNDERFLOW_STATUS;\ + type HUBP_UNDERFLOW_CLEAR;\ type NUM_PIPES;\ type NUM_BANKS;\ type PIPE_INTERLEAVE;\ @@ -739,6 +741,7 @@ void dcn10_hubp_construct( const struct dcn_mi_mask *hubp_mask); void hubp1_read_state(struct hubp *hubp); +void hubp1_clear_underflow(struct hubp *hubp); enum cursor_pitch hubp1_get_cursor_pitch(unsigned int pitch); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c index 87495dea45ec..91e015e14355 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c @@ -1227,7 +1227,8 @@ static bool dcn10_set_input_transfer_func(struct pipe_ctx *pipe_ctx, tf = plane_state->in_transfer_func; if (plane_state->gamma_correction && - !plane_state->gamma_correction->is_identity + !dpp_base->ctx->dc->debug.always_use_regamma + && !plane_state->gamma_correction->is_identity && dce_use_lut(plane_state->format)) dpp_base->funcs->dpp_program_input_lut(dpp_base, plane_state->gamma_correction); @@ -1400,7 +1401,7 @@ static void dcn10_enable_per_frame_crtc_position_reset( if (grouped_pipes[i]->stream_res.tg->funcs->enable_crtc_reset) grouped_pipes[i]->stream_res.tg->funcs->enable_crtc_reset( grouped_pipes[i]->stream_res.tg, - grouped_pipes[i]->stream->triggered_crtc_reset.event_source->status.primary_otg_inst, + 0, &grouped_pipes[i]->stream->triggered_crtc_reset); DC_SYNC_INFO("Waiting for trigger\n"); @@ -1770,7 +1771,7 @@ bool is_rgb_cspace(enum dc_color_space output_color_space) } } -static void dcn10_get_surface_visual_confirm_color( +void dcn10_get_surface_visual_confirm_color( const struct pipe_ctx *pipe_ctx, struct tg_color *color) { @@ -1806,7 +1807,7 @@ static void dcn10_get_surface_visual_confirm_color( } } -static void dcn10_get_hdr_visual_confirm_color( +void dcn10_get_hdr_visual_confirm_color( struct pipe_ctx *pipe_ctx, struct tg_color *color) { @@ -2067,6 +2068,10 @@ void update_dchubp_dpp( &pipe_ctx->ttu_regs, &pipe_ctx->rq_regs, &pipe_ctx->pipe_dlg_param); + hubp->funcs->hubp_setup_interdependent( + hubp, + &pipe_ctx->dlg_regs, + &pipe_ctx->ttu_regs); } size.grph.surface_size = pipe_ctx->plane_res.scl_data.viewport; @@ -2154,6 +2159,15 @@ static void dcn10_blank_pixel_data( color_space = stream->output_color_space; color_space_to_black_color(dc, color_space, &black_color); + /* + * The way 420 is packed, 2 channels carry Y component, 1 channel + * alternate between Cb and Cr, so both channels need the pixel + * value for Y + */ + if (stream->timing.pixel_encoding == PIXEL_ENCODING_YCBCR420) + black_color.color_r_cr = black_color.color_g_y; + + if (stream_res->tg->funcs->set_blank_color) stream_res->tg->funcs->set_blank_color( stream_res->tg, @@ -2337,6 +2351,34 @@ static void dcn10_apply_ctx_for_surface( dcn10_pipe_control_lock(dc, top_pipe_to_program, false); + if (top_pipe_to_program->plane_state && + top_pipe_to_program->plane_state->update_flags.bits.full_update) + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + /* Skip inactive pipes and ones already updated */ + if (!pipe_ctx->stream || pipe_ctx->stream == stream + || !pipe_ctx->plane_state) + continue; + + pipe_ctx->stream_res.tg->funcs->lock(pipe_ctx->stream_res.tg); + + pipe_ctx->plane_res.hubp->funcs->hubp_setup_interdependent( + pipe_ctx->plane_res.hubp, + &pipe_ctx->dlg_regs, + &pipe_ctx->ttu_regs); + } + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + if (!pipe_ctx->stream || pipe_ctx->stream == stream + || !pipe_ctx->plane_state) + continue; + + dcn10_pipe_control_lock(dc, pipe_ctx, false); + } + if (num_planes == 0) false_optc_underflow_wa(dc, stream, tg); @@ -2710,6 +2752,7 @@ static const struct hw_sequencer_funcs dcn10_funcs = { .set_avmute = dce110_set_avmute, .log_hw_state = dcn10_log_hw_state, .get_hw_state = dcn10_get_hw_state, + .clear_status_bits = dcn10_clear_status_bits, .wait_for_mpcc_disconnect = dcn10_wait_for_mpcc_disconnect, .edp_backlight_control = hwss_edp_backlight_control, .edp_power_control = hwss_edp_power_control, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h index 5e5610c9e600..f8eea10e4c64 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h @@ -51,6 +51,8 @@ void dcn10_get_hw_state( char *pBuf, unsigned int bufSize, unsigned int mask); +void dcn10_clear_status_bits(struct dc *dc, unsigned int mask); + bool is_lower_pipe_tree_visible(struct pipe_ctx *pipe_ctx); bool is_upper_pipe_tree_visible(struct pipe_ctx *pipe_ctx); @@ -61,6 +63,14 @@ void dcn10_program_pte_vm(struct dce_hwseq *hws, struct hubp *hubp); void set_hdr_multiplier(struct pipe_ctx *pipe_ctx); +void dcn10_get_surface_visual_confirm_color( + const struct pipe_ctx *pipe_ctx, + struct tg_color *color); + +void dcn10_get_hdr_visual_confirm_color( + struct pipe_ctx *pipe_ctx, + struct tg_color *color); + void update_dchubp_dpp( struct dc *dc, struct pipe_ctx *pipe_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c index 64158900730f..cd469014baa3 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer_debug.c @@ -44,6 +44,7 @@ #include "dcn10_hubp.h" #include "dcn10_hubbub.h" #include "dcn10_cm_common.h" +#include "dcn10_clk_mgr.h" static unsigned int snprintf_count(char *pBuf, unsigned int bufSize, char *fmt, ...) { @@ -454,12 +455,6 @@ static unsigned int dcn10_get_otg_states(struct dc *dc, char *pBuf, unsigned int remaining_buffer -= chars_printed; pBuf += chars_printed; - - // Clear underflow for debug purposes - // We want to keep underflow sticky bit on for the longevity tests outside of test environment. - // This function is called only from Windows or Diags test environment, hence it's safe to clear - // it from here without affecting the original intent. - tg->funcs->clear_optc_underflow(tg); } } @@ -469,19 +464,75 @@ static unsigned int dcn10_get_otg_states(struct dc *dc, char *pBuf, unsigned int static unsigned int dcn10_get_clock_states(struct dc *dc, char *pBuf, unsigned int bufSize) { unsigned int chars_printed = 0; + unsigned int remaining_buffer = bufSize; - chars_printed = snprintf_count(pBuf, bufSize, "dcfclk_khz,dcfclk_deep_sleep_khz,dispclk_khz," - "dppclk_khz,max_supported_dppclk_khz,fclk_khz,socclk_khz\n" - "%d,%d,%d,%d,%d,%d,%d\n", + chars_printed = snprintf_count(pBuf, bufSize, "dcfclk,dcfclk_deep_sleep,dispclk," + "dppclk,fclk,socclk\n" + "%d,%d,%d,%d,%d,%d\n", dc->current_state->bw.dcn.clk.dcfclk_khz, dc->current_state->bw.dcn.clk.dcfclk_deep_sleep_khz, dc->current_state->bw.dcn.clk.dispclk_khz, dc->current_state->bw.dcn.clk.dppclk_khz, - dc->current_state->bw.dcn.clk.max_supported_dppclk_khz, dc->current_state->bw.dcn.clk.fclk_khz, dc->current_state->bw.dcn.clk.socclk_khz); - return chars_printed; + remaining_buffer -= chars_printed; + pBuf += chars_printed; + + return bufSize - remaining_buffer; +} + +static void dcn10_clear_otpc_underflow(struct dc *dc) +{ + struct resource_pool *pool = dc->res_pool; + int i; + + for (i = 0; i < pool->timing_generator_count; i++) { + struct timing_generator *tg = pool->timing_generators[i]; + struct dcn_otg_state s = {0}; + + optc1_read_otg_state(DCN10TG_FROM_TG(tg), &s); + + if (s.otg_enabled & 1) + tg->funcs->clear_optc_underflow(tg); + } +} + +static void dcn10_clear_hubp_underflow(struct dc *dc) +{ + struct resource_pool *pool = dc->res_pool; + int i; + + for (i = 0; i < pool->pipe_count; i++) { + struct hubp *hubp = pool->hubps[i]; + struct dcn_hubp_state *s = &(TO_DCN10_HUBP(hubp)->state); + + hubp->funcs->hubp_read_state(hubp); + + if (!s->blank_en) + hubp->funcs->hubp_clear_underflow(hubp); + } +} + +void dcn10_clear_status_bits(struct dc *dc, unsigned int mask) +{ + /* + * Mask Format + * Bit 0 - 31: Status bit to clear + * + * Mask = 0x0 means clear all status bits + */ + const unsigned int DC_HW_STATE_MASK_HUBP_UNDERFLOW = 0x1; + const unsigned int DC_HW_STATE_MASK_OTPC_UNDERFLOW = 0x2; + + if (mask == 0x0) + mask = 0xFFFFFFFF; + + if (mask & DC_HW_STATE_MASK_HUBP_UNDERFLOW) + dcn10_clear_hubp_underflow(dc); + + if (mask & DC_HW_STATE_MASK_OTPC_UNDERFLOW) + dcn10_clear_otpc_underflow(dc); } void dcn10_get_hw_state(struct dc *dc, char *pBuf, unsigned int bufSize, unsigned int mask) @@ -491,16 +542,16 @@ void dcn10_get_hw_state(struct dc *dc, char *pBuf, unsigned int bufSize, unsigne * Bit 0 - 15: Hardware block mask * Bit 15: 1 = Invariant Only, 0 = All */ - const unsigned int DC_HW_STATE_MASK_HUBBUB = 0x1; - const unsigned int DC_HW_STATE_MASK_HUBP = 0x2; - const unsigned int DC_HW_STATE_MASK_RQ = 0x4; - const unsigned int DC_HW_STATE_MASK_DLG = 0x8; - const unsigned int DC_HW_STATE_MASK_TTU = 0x10; - const unsigned int DC_HW_STATE_MASK_CM = 0x20; - const unsigned int DC_HW_STATE_MASK_MPCC = 0x40; - const unsigned int DC_HW_STATE_MASK_OTG = 0x80; - const unsigned int DC_HW_STATE_MASK_CLOCKS = 0x100; - const unsigned int DC_HW_STATE_INVAR_ONLY = 0x8000; + const unsigned int DC_HW_STATE_MASK_HUBBUB = 0x1; + const unsigned int DC_HW_STATE_MASK_HUBP = 0x2; + const unsigned int DC_HW_STATE_MASK_RQ = 0x4; + const unsigned int DC_HW_STATE_MASK_DLG = 0x8; + const unsigned int DC_HW_STATE_MASK_TTU = 0x10; + const unsigned int DC_HW_STATE_MASK_CM = 0x20; + const unsigned int DC_HW_STATE_MASK_MPCC = 0x40; + const unsigned int DC_HW_STATE_MASK_OTG = 0x80; + const unsigned int DC_HW_STATE_MASK_CLOCKS = 0x100; + const unsigned int DC_HW_STATE_INVAR_ONLY = 0x8000; unsigned int chars_printed = 0; unsigned int remaining_buf_size = bufSize; @@ -556,6 +607,9 @@ void dcn10_get_hw_state(struct dc *dc, char *pBuf, unsigned int bufSize, unsigne remaining_buf_size -= chars_printed; } - if ((mask & DC_HW_STATE_MASK_CLOCKS) && remaining_buf_size > 0) + if ((mask & DC_HW_STATE_MASK_CLOCKS) && remaining_buf_size > 0) { chars_printed = dcn10_get_clock_states(dc, pBuf, remaining_buf_size); + pBuf += chars_printed; + remaining_buf_size -= chars_printed; + } } diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c index 7d1f66797cb3..7c138615f17d 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c @@ -335,9 +335,8 @@ void optc1_program_timing( /* Enable stereo - only when we need to pack 3D frame. Other types * of stereo handled in explicit call */ - h_div_2 = (dc_crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) ? - 1 : 0; + h_div_2 = optc1_is_two_pixels_per_containter(&patched_crtc_timing); REG_UPDATE(OTG_H_TIMING_CNTL, OTG_H_TIMING_DIV_BY2, h_div_2); @@ -360,20 +359,19 @@ void optc1_set_blank_data_double_buffer(struct timing_generator *optc, bool enab static void optc1_unblank_crtc(struct timing_generator *optc) { struct optc *optc1 = DCN10TG_FROM_TG(optc); - uint32_t vertical_interrupt_enable = 0; - - REG_GET(OTG_VERTICAL_INTERRUPT2_CONTROL, - OTG_VERTICAL_INTERRUPT2_INT_ENABLE, &vertical_interrupt_enable); - - /* temporary work around for vertical interrupt, once vertical interrupt enabled, - * this check will be removed. - */ - if (vertical_interrupt_enable) - optc1_set_blank_data_double_buffer(optc, true); REG_UPDATE_2(OTG_BLANK_CONTROL, OTG_BLANK_DATA_EN, 0, OTG_BLANK_DE_MODE, 0); + + /* W/A for automated testing + * Automated testing will fail underflow test as there + * sporadic underflows which occur during the optc blank + * sequence. As a w/a, clear underflow on unblank. + * This prevents the failure, but will not mask actual + * underflow that affect real use cases. + */ + optc1_clear_optc_underflow(optc); } /** @@ -1422,3 +1420,9 @@ void dcn10_timing_generator_init(struct optc *optc1) optc1->min_h_sync_width = 8; optc1->min_v_sync_width = 1; } + +bool optc1_is_two_pixels_per_containter(const struct dc_crtc_timing *timing) +{ + return timing->pixel_encoding == PIXEL_ENCODING_YCBCR420; +} + diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h index c1b114209fe8..8bacf0b6e27e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.h @@ -565,4 +565,6 @@ bool optc1_configure_crc(struct timing_generator *optc, bool optc1_get_crc(struct timing_generator *optc, uint32_t *r_cr, uint32_t *g_y, uint32_t *b_cb); +bool optc1_is_two_pixels_per_containter(const struct dc_crtc_timing *timing); + #endif /* __DC_TIMING_GENERATOR_DCN10_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c index 47dbe4bb294a..5d4772dec0ba 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c @@ -202,7 +202,6 @@ enum dcn10_clk_src_array_id { #define MMHUB_SR(reg_name)\ .reg_name = MMHUB_BASE(mm ## reg_name ## _BASE_IDX) + \ mm ## reg_name - /* macros to expend register list macro defined in HW object header file * end *********************/ @@ -436,7 +435,6 @@ static const struct dcn_optc_mask tg_mask = { TG_COMMON_MASK_SH_LIST_DCN1_0(_MASK) }; - static const struct bios_registers bios_regs = { NBIO_SR(BIOS_SCRATCH_0), NBIO_SR(BIOS_SCRATCH_3), @@ -497,7 +495,6 @@ static const struct dce110_clk_src_mask cs_mask = { CS_COMMON_MASK_SH_LIST_DCN1_0(_MASK) }; - static const struct resource_caps res_cap = { .num_timing_generator = 4, .num_opp = 4, @@ -1277,7 +1274,6 @@ static bool construct( goto fail; } } - pool->base.clk_mgr = dcn1_clk_mgr_create(ctx); if (pool->base.clk_mgr == NULL) { dm_error("DC: failed to create display clock!\n"); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c index 6f9078f3c4d3..b8b5525a389a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_stream_encoder.c @@ -766,7 +766,6 @@ void enc1_stream_encoder_dp_blank( struct stream_encoder *enc) { struct dcn10_stream_encoder *enc1 = DCN10STRENC_FROM_STRENC(enc); - uint32_t retries = 0; uint32_t reg1 = 0; uint32_t max_retries = DP_BLANK_MAX_RETRY * 10; @@ -803,8 +802,6 @@ void enc1_stream_encoder_dp_blank( 0, 10, max_retries); - ASSERT(retries <= max_retries); - /* Tell the DP encoder to ignore timing from CRTC, must be done after * the polling. If we set DP_STEER_FIFO_RESET before DP stream blank is * complete, stream status will be stuck in video stream enabled state, diff --git a/drivers/gpu/drm/amd/display/dc/dm_event_log.h b/drivers/gpu/drm/amd/display/dc/dm_event_log.h index 34a701ca879e..65663f4d93e1 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_event_log.h +++ b/drivers/gpu/drm/amd/display/dc/dm_event_log.h @@ -33,6 +33,7 @@ #define EVENT_LOG_AUX_REQ(ddc, type, action, address, len, data) #define EVENT_LOG_AUX_REP(ddc, type, replyStatus, len, data) +#define EVENT_LOG_CUST_MSG(tag, a, ...) #endif diff --git a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h index beb08fd12b1d..0029a39efb1c 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h +++ b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h @@ -102,7 +102,7 @@ struct pp_smu_funcs_rv { */ void (*set_display_count)(struct pp_smu *pp, int count); - /* which SMU message? are reader and writer WM separate SMU msg? */ + /* reader and writer WM's are sent together as part of one table*/ /* * PPSMC_MSG_SetDriverDramAddrHigh * PPSMC_MSG_SetDriverDramAddrLow diff --git a/drivers/gpu/drm/amd/display/dc/dm_services.h b/drivers/gpu/drm/amd/display/dc/dm_services.h index 28128c02de00..1961cc6d9143 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_services.h +++ b/drivers/gpu/drm/amd/display/dc/dm_services.h @@ -31,6 +31,8 @@ #define __DM_SERVICES_H__ +#include "amdgpu_dm_trace.h" + /* TODO: remove when DC is complete. */ #include "dm_services_types.h" #include "logger_interface.h" @@ -70,6 +72,7 @@ static inline uint32_t dm_read_reg_func( } #endif value = cgs_read_register(ctx->cgs_device, address); + trace_amdgpu_dc_rreg(&ctx->perf_trace->read_count, address, value); return value; } @@ -90,6 +93,7 @@ static inline void dm_write_reg_func( } #endif cgs_write_register(ctx->cgs_device, address, value); + trace_amdgpu_dc_wreg(&ctx->perf_trace->write_count, address, value); } static inline uint32_t dm_read_index_reg( @@ -351,8 +355,12 @@ unsigned long long dm_get_elapse_time_in_ns(struct dc_context *ctx, /* * performance tracing */ -void dm_perf_trace_timestamp(const char *func_name, unsigned int line); -#define PERF_TRACE() dm_perf_trace_timestamp(__func__, __LINE__) +#define PERF_TRACE() trace_amdgpu_dc_performance(CTX->perf_trace->read_count,\ + CTX->perf_trace->write_count, &CTX->perf_trace->last_entry_read,\ + &CTX->perf_trace->last_entry_write, __func__, __LINE__) +#define PERF_TRACE_CTX(__CTX) trace_amdgpu_dc_performance(__CTX->perf_trace->read_count,\ + __CTX->perf_trace->write_count, &__CTX->perf_trace->last_entry_read,\ + &__CTX->perf_trace->last_entry_write, __func__, __LINE__) /* diff --git a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c index f20161c5706d..dada04296025 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.c @@ -56,7 +56,6 @@ struct gpio_service *dal_gpio_service_create( struct dc_context *ctx) { struct gpio_service *service; - uint32_t index_of_id; service = kzalloc(sizeof(struct gpio_service), GFP_KERNEL); @@ -78,44 +77,33 @@ struct gpio_service *dal_gpio_service_create( goto failure_1; } - /* allocate and initialize business storage */ + /* allocate and initialize busyness storage */ { - const uint32_t bits_per_uint = sizeof(uint32_t) << 3; - index_of_id = 0; service->ctx = ctx; do { uint32_t number_of_bits = service->factory.number_of_pins[index_of_id]; + uint32_t i = 0; - uint32_t number_of_uints = - (number_of_bits + bits_per_uint - 1) / - bits_per_uint; - - uint32_t *slot; - - if (number_of_bits) { - uint32_t index_of_uint = 0; + if (number_of_bits) { + service->busyness[index_of_id] = + kcalloc(number_of_bits, sizeof(char), + GFP_KERNEL); - slot = kcalloc(number_of_uints, - sizeof(uint32_t), - GFP_KERNEL); - - if (!slot) { + if (!service->busyness[index_of_id]) { BREAK_TO_DEBUGGER(); goto failure_2; } do { - slot[index_of_uint] = 0; - - ++index_of_uint; - } while (index_of_uint < number_of_uints); - } else - slot = NULL; - - service->busyness[index_of_id] = slot; + service->busyness[index_of_id][i] = 0; + ++i; + } while (i < number_of_bits); + } else { + service->busyness[index_of_id] = NULL; + } ++index_of_id; } while (index_of_id < GPIO_ID_COUNT); @@ -125,13 +113,8 @@ struct gpio_service *dal_gpio_service_create( failure_2: while (index_of_id) { - uint32_t *slot; - --index_of_id; - - slot = service->busyness[index_of_id]; - - kfree(slot); + kfree(service->busyness[index_of_id]); } failure_1: @@ -169,9 +152,7 @@ void dal_gpio_service_destroy( uint32_t index_of_id = 0; do { - uint32_t *slot = (*ptr)->busyness[index_of_id]; - - kfree(slot); + kfree((*ptr)->busyness[index_of_id]); ++index_of_id; } while (index_of_id < GPIO_ID_COUNT); @@ -192,11 +173,7 @@ static bool is_pin_busy( enum gpio_id id, uint32_t en) { - const uint32_t bits_per_uint = sizeof(uint32_t) << 3; - - const uint32_t *slot = service->busyness[id] + (en / bits_per_uint); - - return 0 != (*slot & (1 << (en % bits_per_uint))); + return service->busyness[id][en]; } static void set_pin_busy( @@ -204,10 +181,7 @@ static void set_pin_busy( enum gpio_id id, uint32_t en) { - const uint32_t bits_per_uint = sizeof(uint32_t) << 3; - - service->busyness[id][en / bits_per_uint] |= - (1 << (en % bits_per_uint)); + service->busyness[id][en] = true; } static void set_pin_free( @@ -215,10 +189,7 @@ static void set_pin_free( enum gpio_id id, uint32_t en) { - const uint32_t bits_per_uint = sizeof(uint32_t) << 3; - - service->busyness[id][en / bits_per_uint] &= - ~(1 << (en % bits_per_uint)); + service->busyness[id][en] = false; } enum gpio_result dal_gpio_service_open( diff --git a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.h b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.h index c7f3081f59cc..1d501a43d13b 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.h +++ b/drivers/gpu/drm/amd/display/dc/gpio/gpio_service.h @@ -36,10 +36,9 @@ struct gpio_service { /* * @brief * Business storage. - * For each member of 'enum gpio_id', - * store array of bits (packed into uint32_t slots), - * index individual bit by 'en' value */ - uint32_t *busyness[GPIO_ID_COUNT]; + * one byte For each member of 'enum gpio_id' + */ + char *busyness[GPIO_ID_COUNT]; }; enum gpio_result dal_gpio_service_open( diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c index a683f4102e65..c2028c4744a6 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_factory.c @@ -79,6 +79,7 @@ bool dal_hw_factory_init( dal_hw_factory_dce110_init(factory); return true; case DCE_VERSION_12_0: + case DCE_VERSION_12_1: dal_hw_factory_dce120_init(factory); return true; #if defined(CONFIG_DRM_AMD_DC_DCN1_0) diff --git a/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c b/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c index 096f45628630..236ca28784a9 100644 --- a/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c +++ b/drivers/gpu/drm/amd/display/dc/gpio/hw_translate.c @@ -76,6 +76,7 @@ bool dal_hw_translate_init( dal_hw_translate_dce110_init(translate); return true; case DCE_VERSION_12_0: + case DCE_VERSION_12_1: dal_hw_translate_dce120_init(translate); return true; #if defined(CONFIG_DRM_AMD_DC_DCN1_0) diff --git a/drivers/gpu/drm/amd/display/dc/i2caux/i2caux.c b/drivers/gpu/drm/amd/display/dc/i2caux/i2caux.c index e56093f26eed..1ad6e49102ff 100644 --- a/drivers/gpu/drm/amd/display/dc/i2caux/i2caux.c +++ b/drivers/gpu/drm/amd/display/dc/i2caux/i2caux.c @@ -90,6 +90,7 @@ struct i2caux *dal_i2caux_create( case DCE_VERSION_10_0: return dal_i2caux_dce100_create(ctx); case DCE_VERSION_12_0: + case DCE_VERSION_12_1: return dal_i2caux_dce120_create(ctx); #if defined(CONFIG_DRM_AMD_DC_DCN1_0) case DCN_VERSION_1_0: diff --git a/drivers/gpu/drm/amd/display/dc/inc/compressor.h b/drivers/gpu/drm/amd/display/dc/inc/compressor.h index bcb18f5e1e60..7a147a9762a0 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/compressor.h +++ b/drivers/gpu/drm/amd/display/dc/inc/compressor.h @@ -77,6 +77,7 @@ struct compressor_funcs { }; struct compressor { struct dc_context *ctx; + /* CONTROLLER_ID_D0 + instance, CONTROLLER_ID_UNDEFINED = 0 */ uint32_t attached_inst; bool is_enabled; const struct compressor_funcs *funcs; diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index e3ee96afa60e..b168a5e9dd9d 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -272,6 +272,17 @@ union bw_context { struct dce_bw_output dce; }; +/** + * struct dc_state - The full description of a state requested by a user + * + * @streams: Stream properties + * @stream_status: The planes on a given stream + * @res_ctx: Persistent state of resources + * @bw: The output from bandwidth and watermark calculations + * @pp_display_cfg: PowerPlay clocks and settings + * @dcn_bw_vars: non-stack memory to support bandwidth calculations + * + */ struct dc_state { struct dc_stream_state *streams[MAX_PIPES]; struct dc_stream_status stream_status[MAX_PIPES]; @@ -279,7 +290,6 @@ struct dc_state { struct resource_context res_ctx; - /* The output from BW and WM calculations. */ union bw_context bw; /* Note: these are big structures, do *not* put on stack! */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h index 4550747fb61c..cb85eaa9857f 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dmcu.h @@ -32,6 +32,13 @@ enum dmcu_state { DMCU_RUNNING = 1 }; +struct dmcu_version { + unsigned int date; + unsigned int month; + unsigned int year; + unsigned int interface_version; +}; + struct dmcu { struct dc_context *ctx; const struct dmcu_funcs *funcs; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h index 334c48cdafdc..04c6989aac58 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h @@ -63,6 +63,11 @@ struct hubp_funcs { struct _vcs_dpi_display_rq_regs_st *rq_regs, struct _vcs_dpi_display_pipe_dest_params_st *pipe_dest); + void (*hubp_setup_interdependent)( + struct hubp *hubp, + struct _vcs_dpi_display_dlg_regs_st *dlg_regs, + struct _vcs_dpi_display_ttu_regs_st *ttu_regs); + void (*dcc_control)(struct hubp *hubp, bool enable, bool independent_64b_blks); void (*mem_program_viewport)( @@ -121,6 +126,7 @@ struct hubp_funcs { void (*hubp_clk_cntl)(struct hubp *hubp, bool enable); void (*hubp_vtg_sel)(struct hubp *hubp, uint32_t otg_inst); void (*hubp_read_state)(struct hubp *hubp); + void (*hubp_clear_underflow)(struct hubp *hubp); void (*hubp_disable_control)(struct hubp *hubp, bool disable_hubp); unsigned int (*hubp_get_underflow_status)(struct hubp *hubp); diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h index e9b702ce02dd..d6a85f48b6d1 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h @@ -200,6 +200,7 @@ struct hw_sequencer_funcs { void (*log_hw_state)(struct dc *dc, struct dc_log_buffer_ctx *log_ctx); void (*get_hw_state)(struct dc *dc, char *pBuf, unsigned int bufSize, unsigned int mask); + void (*clear_status_bits)(struct dc *dc, unsigned int mask); void (*wait_for_mpcc_disconnect)(struct dc *dc, struct resource_pool *res_pool, diff --git a/drivers/gpu/drm/amd/display/include/bios_parser_types.h b/drivers/gpu/drm/amd/display/include/bios_parser_types.h index f8dbfa5b89f2..7fd78a696800 100644 --- a/drivers/gpu/drm/amd/display/include/bios_parser_types.h +++ b/drivers/gpu/drm/amd/display/include/bios_parser_types.h @@ -41,6 +41,7 @@ enum as_signal_type { AS_SIGNAL_TYPE_LVDS, AS_SIGNAL_TYPE_DISPLAY_PORT, AS_SIGNAL_TYPE_GPU_PLL, + AS_SIGNAL_TYPE_XGMI, AS_SIGNAL_TYPE_UNKNOWN }; diff --git a/drivers/gpu/drm/amd/display/include/dal_types.h b/drivers/gpu/drm/amd/display/include/dal_types.h index 89627133e188..f5bd869d4320 100644 --- a/drivers/gpu/drm/amd/display/include/dal_types.h +++ b/drivers/gpu/drm/amd/display/include/dal_types.h @@ -42,6 +42,7 @@ enum dce_version { DCE_VERSION_11_2, DCE_VERSION_11_22, DCE_VERSION_12_0, + DCE_VERSION_12_1, DCE_VERSION_MAX, DCN_VERSION_1_0, #if defined(CONFIG_DRM_AMD_DC_DCN1_01) diff --git a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c index 7480f072c375..479b77c2e89e 100644 --- a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c +++ b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c @@ -813,20 +813,26 @@ static bool build_freesync_hdr(struct pwl_float_data_ex *rgb_regamma, const struct hw_x_point *coord_x = coordinate_x; struct fixed31_32 scaledX = dc_fixpt_zero; struct fixed31_32 scaledX1 = dc_fixpt_zero; - struct fixed31_32 max_display = dc_fixpt_from_int(fs_params->max_display); - struct fixed31_32 min_display = dc_fixpt_from_fraction(fs_params->min_display, 10000); - struct fixed31_32 max_content = dc_fixpt_from_int(fs_params->max_content); - struct fixed31_32 min_content = dc_fixpt_from_fraction(fs_params->min_content, 10000); + struct fixed31_32 max_display; + struct fixed31_32 min_display; + struct fixed31_32 max_content; + struct fixed31_32 min_content; struct fixed31_32 clip = dc_fixpt_one; struct fixed31_32 output; bool use_eetf = false; bool is_clipped = false; - struct fixed31_32 sdr_white_level = dc_fixpt_from_int(fs_params->sdr_white_level); + struct fixed31_32 sdr_white_level; if (fs_params == NULL || fs_params->max_content == 0 || fs_params->max_display == 0) return false; + max_display = dc_fixpt_from_int(fs_params->max_display); + min_display = dc_fixpt_from_fraction(fs_params->min_display, 10000); + max_content = dc_fixpt_from_int(fs_params->max_content); + min_content = dc_fixpt_from_fraction(fs_params->min_content, 10000); + sdr_white_level = dc_fixpt_from_int(fs_params->sdr_white_level); + if (fs_params->min_display > 1000) // cap at 0.1 at the bottom min_display = dc_fixpt_from_fraction(1, 10); if (fs_params->max_display < 100) // cap at 100 at the top @@ -1755,7 +1761,7 @@ bool mod_color_calculate_degamma_params(struct dc_transfer_func *input_tf, struct pwl_float_data *rgb_user = NULL; struct pwl_float_data_ex *curve = NULL; - struct gamma_pixel *axix_x = NULL; + struct gamma_pixel *axis_x = NULL; struct pixel_gamma_point *coeff = NULL; enum dc_transfer_func_predefined tf = TRANSFER_FUNCTION_SRGB; bool ret = false; @@ -1781,10 +1787,10 @@ bool mod_color_calculate_degamma_params(struct dc_transfer_func *input_tf, GFP_KERNEL); if (!curve) goto curve_alloc_fail; - axix_x = kvcalloc(ramp->num_entries + _EXTRA_POINTS, sizeof(*axix_x), + axis_x = kvcalloc(ramp->num_entries + _EXTRA_POINTS, sizeof(*axis_x), GFP_KERNEL); - if (!axix_x) - goto axix_x_alloc_fail; + if (!axis_x) + goto axis_x_alloc_fail; coeff = kvcalloc(MAX_HW_POINTS + _EXTRA_POINTS, sizeof(*coeff), GFP_KERNEL); if (!coeff) @@ -1797,7 +1803,7 @@ bool mod_color_calculate_degamma_params(struct dc_transfer_func *input_tf, tf = input_tf->tf; build_evenly_distributed_points( - axix_x, + axis_x, ramp->num_entries, dividers); @@ -1822,7 +1828,7 @@ bool mod_color_calculate_degamma_params(struct dc_transfer_func *input_tf, tf_pts->x_point_at_y1_blue = 1; map_regamma_hw_to_x_user(ramp, coeff, rgb_user, - coordinates_x, axix_x, curve, + coordinates_x, axis_x, curve, MAX_HW_POINTS, tf_pts, mapUserRamp && ramp->type != GAMMA_CUSTOM); if (ramp->type == GAMMA_CUSTOM) @@ -1832,8 +1838,8 @@ bool mod_color_calculate_degamma_params(struct dc_transfer_func *input_tf, kvfree(coeff); coeff_alloc_fail: - kvfree(axix_x); -axix_x_alloc_fail: + kvfree(axis_x); +axis_x_alloc_fail: kvfree(curve); curve_alloc_fail: kvfree(rgb_user); diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c index 620a171620ee..1544ed3f1747 100644 --- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c +++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c @@ -608,12 +608,12 @@ static void build_vrr_infopacket_data(const struct mod_vrr_params *vrr, static void build_vrr_infopacket_fs2_data(enum color_transfer_func app_tf, struct dc_info_packet *infopacket) { - if (app_tf != transfer_func_unknown) { + if (app_tf != TRANSFER_FUNC_UNKNOWN) { infopacket->valid = true; infopacket->sb[6] |= 0x08; // PB6 = [Bit 3 = Native Color Active] - if (app_tf == transfer_func_gamma_22) { + if (app_tf == TRANSFER_FUNC_GAMMA_22) { infopacket->sb[9] |= 0x04; // PB6 = [Bit 2 = Gamma 2.2 EOTF Active] } } @@ -688,11 +688,11 @@ void mod_freesync_build_vrr_infopacket(struct mod_freesync *mod_freesync, return; switch (packet_type) { - case packet_type_fs2: + case PACKET_TYPE_FS2: build_vrr_infopacket_v2(stream->signal, vrr, app_tf, infopacket); break; - case packet_type_vrr: - case packet_type_fs1: + case PACKET_TYPE_VRR: + case PACKET_TYPE_FS1: default: build_vrr_infopacket_v1(stream->signal, vrr, infopacket); } diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h b/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h index 786b34380f85..5b1c9a4c7643 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_info_packet.h @@ -26,15 +26,13 @@ #ifndef MOD_INFO_PACKET_H_ #define MOD_INFO_PACKET_H_ -struct info_packet_inputs { - const struct dc_stream_state *pStream; -}; +#include "mod_shared.h" -struct info_packets { - struct dc_info_packet *pVscInfoPacket; -}; +//Forward Declarations +struct dc_stream_state; +struct dc_info_packet; -void mod_build_infopackets(struct info_packet_inputs *inputs, - struct info_packets *info_packets); +void mod_build_vsc_infopacket(const struct dc_stream_state *stream, + struct dc_info_packet *info_packet); #endif diff --git a/drivers/gpu/drm/amd/display/modules/inc/mod_shared.h b/drivers/gpu/drm/amd/display/modules/inc/mod_shared.h index 238c431ae483..1bd02c0ac30c 100644 --- a/drivers/gpu/drm/amd/display/modules/inc/mod_shared.h +++ b/drivers/gpu/drm/amd/display/modules/inc/mod_shared.h @@ -23,27 +23,26 @@ * */ - #ifndef MOD_SHARED_H_ #define MOD_SHARED_H_ enum color_transfer_func { - transfer_func_unknown, - transfer_func_srgb, - transfer_func_bt709, - transfer_func_pq2084, - transfer_func_pq2084_interim, - transfer_func_linear_0_1, - transfer_func_linear_0_125, - transfer_func_dolbyvision, - transfer_func_gamma_22, - transfer_func_gamma_26 + TRANSFER_FUNC_UNKNOWN, + TRANSFER_FUNC_SRGB, + TRANSFER_FUNC_BT709, + TRANSFER_FUNC_PQ2084, + TRANSFER_FUNC_PQ2084_INTERIM, + TRANSFER_FUNC_LINEAR_0_1, + TRANSFER_FUNC_LINEAR_0_125, + TRANSFER_FUNC_GAMMA_22, + TRANSFER_FUNC_GAMMA_26 }; enum vrr_packet_type { - packet_type_vrr, - packet_type_fs1, - packet_type_fs2 + PACKET_TYPE_VRR, + PACKET_TYPE_FS1, + PACKET_TYPE_FS2 }; + #endif /* MOD_SHARED_H_ */ diff --git a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c index ff8bfb9b43b0..db06fab2ad5c 100644 --- a/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c +++ b/drivers/gpu/drm/amd/display/modules/info_packet/info_packet.c @@ -25,6 +25,10 @@ #include "mod_info_packet.h" #include "core_types.h" +#include "dc_types.h" +#include "mod_shared.h" + +#define HDMI_INFOFRAME_TYPE_VENDOR 0x81 enum ColorimetryRGBDP { ColorimetryRGB_DP_sRGB = 0, @@ -41,7 +45,7 @@ enum ColorimetryYCCDP { ColorimetryYCC_DP_ITU2020YCbCr = 7, }; -static void mod_build_vsc_infopacket(const struct dc_stream_state *stream, +void mod_build_vsc_infopacket(const struct dc_stream_state *stream, struct dc_info_packet *info_packet) { unsigned int vscPacketRevision = 0; @@ -159,7 +163,7 @@ static void mod_build_vsc_infopacket(const struct dc_stream_state *stream, * DPCD register is exposed in the new Extended Receiver Capability field for DPCD Rev. 1.4 * (and higher). When MISC1. bit 6. is Set to 1, a Source device uses a VSC SDP to indicate * the Pixel Encoding/Colorimetry Format and that a Sink device must ignore MISC1, bit 7, and - * MISC0, bits 7:1 (MISC1, bit 7. and MISC0, bits 7:1 become “don’t care”).) + * MISC0, bits 7:1 (MISC1, bit 7. and MISC0, bits 7:1 become "don't care").) */ if (vscPacketRevision == 0x5) { /* Secondary-data Packet ID = 0 */ @@ -320,10 +324,3 @@ static void mod_build_vsc_infopacket(const struct dc_stream_state *stream, } -void mod_build_infopackets(struct info_packet_inputs *inputs, - struct info_packets *info_packets) -{ - if (info_packets->pVscInfoPacket != NULL) - mod_build_vsc_infopacket(inputs->pStream, info_packets->pVscInfoPacket); -} - diff --git a/drivers/gpu/drm/amd/display/modules/power/Makefile b/drivers/gpu/drm/amd/display/modules/power/Makefile new file mode 100644 index 000000000000..87851f892a52 --- /dev/null +++ b/drivers/gpu/drm/amd/display/modules/power/Makefile @@ -0,0 +1,31 @@ +# +# Copyright 2017 Advanced Micro Devices, Inc. +# +# Permission is hereby granted, free of charge, to any person obtaining a +# copy of this software and associated documentation files (the "Software"), +# to deal in the Software without restriction, including without limitation +# the rights to use, copy, modify, merge, publish, distribute, sublicense, +# and/or sell copies of the Software, and to permit persons to whom the +# Software is furnished to do so, subject to the following conditions: +# +# The above copyright notice and this permission notice shall be included in +# all copies or substantial portions of the Software. +# +# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR +# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, +# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL +# THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR +# OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, +# ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR +# OTHER DEALINGS IN THE SOFTWARE. +# +# +# Makefile for the 'power' sub-module of DAL. +# + +MOD_POWER = power_helpers.o + +AMD_DAL_MOD_POWER = $(addprefix $(AMDDALPATH)/modules/power/,$(MOD_POWER)) +#$(info ************ DAL POWER MODULE MAKEFILE ************) + +AMD_DISPLAY_FILES += $(AMD_DAL_MOD_POWER)
\ No newline at end of file diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.c b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c new file mode 100644 index 000000000000..00f63b7dd32f --- /dev/null +++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.c @@ -0,0 +1,326 @@ +/* Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#include "power_helpers.h" +#include "dc/inc/hw/dmcu.h" + +#define DIV_ROUNDUP(a, b) (((a)+((b)/2))/(b)) + +/* Possible Min Reduction config from least aggressive to most aggressive + * 0 1 2 3 4 5 6 7 8 9 10 11 12 + * 100 98.0 94.1 94.1 85.1 80.3 75.3 69.4 60.0 57.6 50.2 49.8 40.0 % + */ +static const unsigned char min_reduction_table[13] = { +0xff, 0xfa, 0xf0, 0xf0, 0xd9, 0xcd, 0xc0, 0xb1, 0x99, 0x93, 0x80, 0x82, 0x66}; + +/* Possible Max Reduction configs from least aggressive to most aggressive + * 0 1 2 3 4 5 6 7 8 9 10 11 12 + * 96.1 89.8 85.1 80.3 69.4 64.7 64.7 50.2 39.6 30.2 30.2 30.2 19.6 % + */ +static const unsigned char max_reduction_table[13] = { +0xf5, 0xe5, 0xd9, 0xcd, 0xb1, 0xa5, 0xa5, 0x80, 0x65, 0x4d, 0x4d, 0x4d, 0x32}; + +/* Predefined ABM configuration sets. We may have different configuration sets + * in order to satisfy different power/quality requirements. + */ +static const unsigned char abm_config[abm_defines_max_config][abm_defines_max_level] = { +/* ABM Level 1, ABM Level 2, ABM Level 3, ABM Level 4 */ +{ 2, 5, 7, 8 }, /* Default - Medium aggressiveness */ +{ 2, 5, 8, 11 }, /* Alt #1 - Increased aggressiveness */ +{ 0, 2, 4, 8 }, /* Alt #2 - Minimal aggressiveness */ +{ 3, 6, 10, 12 }, /* Alt #3 - Super aggressiveness */ +}; + +#define NUM_AMBI_LEVEL 5 +#define NUM_AGGR_LEVEL 4 +#define NUM_POWER_FN_SEGS 8 +#define NUM_BL_CURVE_SEGS 16 + +/* NOTE: iRAM is 256B in size */ +struct iram_table_v_2 { + /* flags */ + uint16_t flags; /* 0x00 U16 */ + + /* parameters for ABM2.0 algorithm */ + uint8_t min_reduction[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL]; /* 0x02 U0.8 */ + uint8_t max_reduction[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL]; /* 0x16 U0.8 */ + uint8_t bright_pos_gain[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL]; /* 0x2a U2.6 */ + uint8_t bright_neg_gain[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL]; /* 0x3e U2.6 */ + uint8_t dark_pos_gain[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL]; /* 0x52 U2.6 */ + uint8_t dark_neg_gain[NUM_AMBI_LEVEL][NUM_AGGR_LEVEL]; /* 0x66 U2.6 */ + uint8_t iir_curve[NUM_AMBI_LEVEL]; /* 0x7a U0.8 */ + uint8_t deviation_gain; /* 0x7f U0.8 */ + + /* parameters for crgb conversion */ + uint16_t crgb_thresh[NUM_POWER_FN_SEGS]; /* 0x80 U3.13 */ + uint16_t crgb_offset[NUM_POWER_FN_SEGS]; /* 0x90 U1.15 */ + uint16_t crgb_slope[NUM_POWER_FN_SEGS]; /* 0xa0 U4.12 */ + + /* parameters for custom curve */ + /* thresholds for brightness --> backlight */ + uint16_t backlight_thresholds[NUM_BL_CURVE_SEGS]; /* 0xb0 U16.0 */ + /* offsets for brightness --> backlight */ + uint16_t backlight_offsets[NUM_BL_CURVE_SEGS]; /* 0xd0 U16.0 */ + + /* For reading PSR State directly from IRAM */ + uint8_t psr_state; /* 0xf0 */ + uint8_t dmcu_interface_version; /* 0xf1 */ + uint8_t dmcu_date_version_year_b0; /* 0xf2 */ + uint8_t dmcu_date_version_year_b1; /* 0xf3 */ + uint8_t dmcu_date_version_month; /* 0xf4 */ + uint8_t dmcu_date_version_day; /* 0xf5 */ + uint8_t dmcu_state; /* 0xf6 */ + + uint16_t blRampReduction; /* 0xf7 */ + uint16_t blRampStart; /* 0xf9 */ + uint8_t dummy5; /* 0xfb */ + uint8_t dummy6; /* 0xfc */ + uint8_t dummy7; /* 0xfd */ + uint8_t dummy8; /* 0xfe */ + uint8_t dummy9; /* 0xff */ +}; + +static uint16_t backlight_8_to_16(unsigned int backlight_8bit) +{ + return (uint16_t)(backlight_8bit * 0x101); +} + +static void fill_backlight_transform_table(struct dmcu_iram_parameters params, + struct iram_table_v_2 *table) +{ + unsigned int i; + unsigned int num_entries = NUM_BL_CURVE_SEGS; + unsigned int query_input_8bit; + unsigned int query_output_8bit; + unsigned int lut_index; + + table->backlight_thresholds[0] = 0; + table->backlight_offsets[0] = params.backlight_lut_array[0]; + table->backlight_thresholds[num_entries-1] = 0xFFFF; + table->backlight_offsets[num_entries-1] = + params.backlight_lut_array[params.backlight_lut_array_size - 1]; + + /* Setup all brightness levels between 0% and 100% exclusive + * Fills brightness-to-backlight transform table. Backlight custom curve + * describes transform from brightness to backlight. It will be defined + * as set of thresholds and set of offsets, together, implying + * extrapolation of custom curve into 16 uniformly spanned linear + * segments. Each threshold/offset represented by 16 bit entry in + * format U4.10. + */ + for (i = 1; i+1 < num_entries; i++) { + query_input_8bit = DIV_ROUNDUP((i * 256), num_entries); + + lut_index = (params.backlight_lut_array_size - 1) * i / (num_entries - 1); + ASSERT(lut_index < params.backlight_lut_array_size); + query_output_8bit = params.backlight_lut_array[lut_index] >> 8; + + table->backlight_thresholds[i] = + backlight_8_to_16(query_input_8bit); + table->backlight_offsets[i] = + backlight_8_to_16(query_output_8bit); + } +} + +bool dmcu_load_iram(struct dmcu *dmcu, + struct dmcu_iram_parameters params) +{ + struct iram_table_v_2 ram_table; + unsigned int set = params.set; + + if (dmcu == NULL) + return false; + + if (!dmcu->funcs->is_dmcu_initialized(dmcu)) + return true; + + memset(&ram_table, 0, sizeof(ram_table)); + + ram_table.flags = 0x0; + ram_table.deviation_gain = 0xb3; + + ram_table.blRampReduction = + cpu_to_be16(params.backlight_ramping_reduction); + ram_table.blRampStart = + cpu_to_be16(params.backlight_ramping_start); + + ram_table.min_reduction[0][0] = min_reduction_table[abm_config[set][0]]; + ram_table.min_reduction[1][0] = min_reduction_table[abm_config[set][0]]; + ram_table.min_reduction[2][0] = min_reduction_table[abm_config[set][0]]; + ram_table.min_reduction[3][0] = min_reduction_table[abm_config[set][0]]; + ram_table.min_reduction[4][0] = min_reduction_table[abm_config[set][0]]; + ram_table.max_reduction[0][0] = max_reduction_table[abm_config[set][0]]; + ram_table.max_reduction[1][0] = max_reduction_table[abm_config[set][0]]; + ram_table.max_reduction[2][0] = max_reduction_table[abm_config[set][0]]; + ram_table.max_reduction[3][0] = max_reduction_table[abm_config[set][0]]; + ram_table.max_reduction[4][0] = max_reduction_table[abm_config[set][0]]; + + ram_table.min_reduction[0][1] = min_reduction_table[abm_config[set][1]]; + ram_table.min_reduction[1][1] = min_reduction_table[abm_config[set][1]]; + ram_table.min_reduction[2][1] = min_reduction_table[abm_config[set][1]]; + ram_table.min_reduction[3][1] = min_reduction_table[abm_config[set][1]]; + ram_table.min_reduction[4][1] = min_reduction_table[abm_config[set][1]]; + ram_table.max_reduction[0][1] = max_reduction_table[abm_config[set][1]]; + ram_table.max_reduction[1][1] = max_reduction_table[abm_config[set][1]]; + ram_table.max_reduction[2][1] = max_reduction_table[abm_config[set][1]]; + ram_table.max_reduction[3][1] = max_reduction_table[abm_config[set][1]]; + ram_table.max_reduction[4][1] = max_reduction_table[abm_config[set][1]]; + + ram_table.min_reduction[0][2] = min_reduction_table[abm_config[set][2]]; + ram_table.min_reduction[1][2] = min_reduction_table[abm_config[set][2]]; + ram_table.min_reduction[2][2] = min_reduction_table[abm_config[set][2]]; + ram_table.min_reduction[3][2] = min_reduction_table[abm_config[set][2]]; + ram_table.min_reduction[4][2] = min_reduction_table[abm_config[set][2]]; + ram_table.max_reduction[0][2] = max_reduction_table[abm_config[set][2]]; + ram_table.max_reduction[1][2] = max_reduction_table[abm_config[set][2]]; + ram_table.max_reduction[2][2] = max_reduction_table[abm_config[set][2]]; + ram_table.max_reduction[3][2] = max_reduction_table[abm_config[set][2]]; + ram_table.max_reduction[4][2] = max_reduction_table[abm_config[set][2]]; + + ram_table.min_reduction[0][3] = min_reduction_table[abm_config[set][3]]; + ram_table.min_reduction[1][3] = min_reduction_table[abm_config[set][3]]; + ram_table.min_reduction[2][3] = min_reduction_table[abm_config[set][3]]; + ram_table.min_reduction[3][3] = min_reduction_table[abm_config[set][3]]; + ram_table.min_reduction[4][3] = min_reduction_table[abm_config[set][3]]; + ram_table.max_reduction[0][3] = max_reduction_table[abm_config[set][3]]; + ram_table.max_reduction[1][3] = max_reduction_table[abm_config[set][3]]; + ram_table.max_reduction[2][3] = max_reduction_table[abm_config[set][3]]; + ram_table.max_reduction[3][3] = max_reduction_table[abm_config[set][3]]; + ram_table.max_reduction[4][3] = max_reduction_table[abm_config[set][3]]; + + ram_table.bright_pos_gain[0][0] = 0x20; + ram_table.bright_pos_gain[0][1] = 0x20; + ram_table.bright_pos_gain[0][2] = 0x20; + ram_table.bright_pos_gain[0][3] = 0x20; + ram_table.bright_pos_gain[1][0] = 0x20; + ram_table.bright_pos_gain[1][1] = 0x20; + ram_table.bright_pos_gain[1][2] = 0x20; + ram_table.bright_pos_gain[1][3] = 0x20; + ram_table.bright_pos_gain[2][0] = 0x20; + ram_table.bright_pos_gain[2][1] = 0x20; + ram_table.bright_pos_gain[2][2] = 0x20; + ram_table.bright_pos_gain[2][3] = 0x20; + ram_table.bright_pos_gain[3][0] = 0x20; + ram_table.bright_pos_gain[3][1] = 0x20; + ram_table.bright_pos_gain[3][2] = 0x20; + ram_table.bright_pos_gain[3][3] = 0x20; + ram_table.bright_pos_gain[4][0] = 0x20; + ram_table.bright_pos_gain[4][1] = 0x20; + ram_table.bright_pos_gain[4][2] = 0x20; + ram_table.bright_pos_gain[4][3] = 0x20; + ram_table.bright_neg_gain[0][1] = 0x00; + ram_table.bright_neg_gain[0][2] = 0x00; + ram_table.bright_neg_gain[0][3] = 0x00; + ram_table.bright_neg_gain[1][0] = 0x00; + ram_table.bright_neg_gain[1][1] = 0x00; + ram_table.bright_neg_gain[1][2] = 0x00; + ram_table.bright_neg_gain[1][3] = 0x00; + ram_table.bright_neg_gain[2][0] = 0x00; + ram_table.bright_neg_gain[2][1] = 0x00; + ram_table.bright_neg_gain[2][2] = 0x00; + ram_table.bright_neg_gain[2][3] = 0x00; + ram_table.bright_neg_gain[3][0] = 0x00; + ram_table.bright_neg_gain[3][1] = 0x00; + ram_table.bright_neg_gain[3][2] = 0x00; + ram_table.bright_neg_gain[3][3] = 0x00; + ram_table.bright_neg_gain[4][0] = 0x00; + ram_table.bright_neg_gain[4][1] = 0x00; + ram_table.bright_neg_gain[4][2] = 0x00; + ram_table.bright_neg_gain[4][3] = 0x00; + ram_table.dark_pos_gain[0][0] = 0x00; + ram_table.dark_pos_gain[0][1] = 0x00; + ram_table.dark_pos_gain[0][2] = 0x00; + ram_table.dark_pos_gain[0][3] = 0x00; + ram_table.dark_pos_gain[1][0] = 0x00; + ram_table.dark_pos_gain[1][1] = 0x00; + ram_table.dark_pos_gain[1][2] = 0x00; + ram_table.dark_pos_gain[1][3] = 0x00; + ram_table.dark_pos_gain[2][0] = 0x00; + ram_table.dark_pos_gain[2][1] = 0x00; + ram_table.dark_pos_gain[2][2] = 0x00; + ram_table.dark_pos_gain[2][3] = 0x00; + ram_table.dark_pos_gain[3][0] = 0x00; + ram_table.dark_pos_gain[3][1] = 0x00; + ram_table.dark_pos_gain[3][2] = 0x00; + ram_table.dark_pos_gain[3][3] = 0x00; + ram_table.dark_pos_gain[4][0] = 0x00; + ram_table.dark_pos_gain[4][1] = 0x00; + ram_table.dark_pos_gain[4][2] = 0x00; + ram_table.dark_pos_gain[4][3] = 0x00; + ram_table.dark_neg_gain[0][0] = 0x00; + ram_table.dark_neg_gain[0][1] = 0x00; + ram_table.dark_neg_gain[0][2] = 0x00; + ram_table.dark_neg_gain[0][3] = 0x00; + ram_table.dark_neg_gain[1][0] = 0x00; + ram_table.dark_neg_gain[1][1] = 0x00; + ram_table.dark_neg_gain[1][2] = 0x00; + ram_table.dark_neg_gain[1][3] = 0x00; + ram_table.dark_neg_gain[2][0] = 0x00; + ram_table.dark_neg_gain[2][1] = 0x00; + ram_table.dark_neg_gain[2][2] = 0x00; + ram_table.dark_neg_gain[2][3] = 0x00; + ram_table.dark_neg_gain[3][0] = 0x00; + ram_table.dark_neg_gain[3][1] = 0x00; + ram_table.dark_neg_gain[3][2] = 0x00; + ram_table.dark_neg_gain[3][3] = 0x00; + ram_table.dark_neg_gain[4][0] = 0x00; + ram_table.dark_neg_gain[4][1] = 0x00; + ram_table.dark_neg_gain[4][2] = 0x00; + ram_table.dark_neg_gain[4][3] = 0x00; + ram_table.iir_curve[0] = 0x65; + ram_table.iir_curve[1] = 0x65; + ram_table.iir_curve[2] = 0x65; + ram_table.iir_curve[3] = 0x65; + ram_table.iir_curve[4] = 0x65; + ram_table.crgb_thresh[0] = cpu_to_be16(0x13b6); + ram_table.crgb_thresh[1] = cpu_to_be16(0x1648); + ram_table.crgb_thresh[2] = cpu_to_be16(0x18e3); + ram_table.crgb_thresh[3] = cpu_to_be16(0x1b41); + ram_table.crgb_thresh[4] = cpu_to_be16(0x1d46); + ram_table.crgb_thresh[5] = cpu_to_be16(0x1f21); + ram_table.crgb_thresh[6] = cpu_to_be16(0x2167); + ram_table.crgb_thresh[7] = cpu_to_be16(0x2384); + ram_table.crgb_offset[0] = cpu_to_be16(0x2999); + ram_table.crgb_offset[1] = cpu_to_be16(0x3999); + ram_table.crgb_offset[2] = cpu_to_be16(0x4666); + ram_table.crgb_offset[3] = cpu_to_be16(0x5999); + ram_table.crgb_offset[4] = cpu_to_be16(0x6333); + ram_table.crgb_offset[5] = cpu_to_be16(0x7800); + ram_table.crgb_offset[6] = cpu_to_be16(0x8c00); + ram_table.crgb_offset[7] = cpu_to_be16(0xa000); + ram_table.crgb_slope[0] = cpu_to_be16(0x3147); + ram_table.crgb_slope[1] = cpu_to_be16(0x2978); + ram_table.crgb_slope[2] = cpu_to_be16(0x23a2); + ram_table.crgb_slope[3] = cpu_to_be16(0x1f55); + ram_table.crgb_slope[4] = cpu_to_be16(0x1c63); + ram_table.crgb_slope[5] = cpu_to_be16(0x1a0f); + ram_table.crgb_slope[6] = cpu_to_be16(0x178d); + ram_table.crgb_slope[7] = cpu_to_be16(0x15ab); + + fill_backlight_transform_table( + params, &ram_table); + + return dmcu->funcs->load_iram( + dmcu, 0, (char *)(&ram_table), sizeof(ram_table)); +} diff --git a/drivers/gpu/drm/amd/display/modules/power/power_helpers.h b/drivers/gpu/drm/amd/display/modules/power/power_helpers.h new file mode 100644 index 000000000000..da5df00fedce --- /dev/null +++ b/drivers/gpu/drm/amd/display/modules/power/power_helpers.h @@ -0,0 +1,47 @@ +/* Copyright 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: AMD + * + */ + +#ifndef MODULES_POWER_POWER_HELPERS_H_ +#define MODULES_POWER_POWER_HELPERS_H_ + +#include "dc/inc/hw/dmcu.h" + + +enum abm_defines { + abm_defines_max_level = 4, + abm_defines_max_config = 4, +}; + +struct dmcu_iram_parameters { + unsigned int *backlight_lut_array; + unsigned int backlight_lut_array_size; + unsigned int backlight_ramping_reduction; + unsigned int backlight_ramping_start; + unsigned int set; +}; + +bool dmcu_load_iram(struct dmcu *dmcu, + struct dmcu_iram_parameters params); + +#endif /* MODULES_POWER_POWER_HELPERS_H_ */ diff --git a/drivers/gpu/drm/amd/include/amd_acpi.h b/drivers/gpu/drm/amd/include/amd_acpi.h index 9b9699fc433f..c72cbfe8f684 100644 --- a/drivers/gpu/drm/amd/include/amd_acpi.h +++ b/drivers/gpu/drm/amd/include/amd_acpi.h @@ -52,6 +52,30 @@ struct atif_sbios_requests { u8 backlight_level; /* panel backlight level (0-255) */ } __packed; +struct atif_qbtc_arguments { + u16 size; /* structure size in bytes (includes size field) */ + u8 requested_display; /* which display is requested */ +} __packed; + +#define ATIF_QBTC_MAX_DATA_POINTS 99 + +struct atif_qbtc_data_point { + u8 luminance; /* luminance in percent */ + u8 ipnut_signal; /* input signal in range 0-255 */ +} __packed; + +struct atif_qbtc_output { + u16 size; /* structure size in bytes (includes size field) */ + u16 flags; /* all zeroes */ + u8 error_code; /* error code */ + u8 ac_level; /* default brightness on AC power */ + u8 dc_level; /* default brightness on DC power */ + u8 min_input_signal; /* max input signal in range 0-255 */ + u8 max_input_signal; /* min input signal in range 0-255 */ + u8 number_of_points; /* number of data points */ + struct atif_qbtc_data_point data_points[ATIF_QBTC_MAX_DATA_POINTS]; +} __packed; + #define ATIF_NOTIFY_MASK 0x3 #define ATIF_NOTIFY_NONE 0 #define ATIF_NOTIFY_81 1 @@ -126,26 +150,18 @@ struct atcs_pref_req_output { * DWORD - supported functions bit vector */ /* Notifications mask */ -# define ATIF_DISPLAY_SWITCH_REQUEST_SUPPORTED (1 << 0) -# define ATIF_EXPANSION_MODE_CHANGE_REQUEST_SUPPORTED (1 << 1) # define ATIF_THERMAL_STATE_CHANGE_REQUEST_SUPPORTED (1 << 2) # define ATIF_FORCED_POWER_STATE_CHANGE_REQUEST_SUPPORTED (1 << 3) # define ATIF_SYSTEM_POWER_SOURCE_CHANGE_REQUEST_SUPPORTED (1 << 4) -# define ATIF_DISPLAY_CONF_CHANGE_REQUEST_SUPPORTED (1 << 5) -# define ATIF_PX_GFX_SWITCH_REQUEST_SUPPORTED (1 << 6) # define ATIF_PANEL_BRIGHTNESS_CHANGE_REQUEST_SUPPORTED (1 << 7) # define ATIF_DGPU_DISPLAY_EVENT_SUPPORTED (1 << 8) +# define ATIF_GPU_PACKAGE_POWER_LIMIT_REQUEST_SUPPORTED (1 << 12) /* supported functions vector */ # define ATIF_GET_SYSTEM_PARAMETERS_SUPPORTED (1 << 0) # define ATIF_GET_SYSTEM_BIOS_REQUESTS_SUPPORTED (1 << 1) -# define ATIF_SELECT_ACTIVE_DISPLAYS_SUPPORTED (1 << 2) -# define ATIF_GET_LID_STATE_SUPPORTED (1 << 3) -# define ATIF_GET_TV_STANDARD_FROM_CMOS_SUPPORTED (1 << 4) -# define ATIF_SET_TV_STANDARD_IN_CMOS_SUPPORTED (1 << 5) -# define ATIF_GET_PANEL_EXPANSION_MODE_FROM_CMOS_SUPPORTED (1 << 6) -# define ATIF_SET_PANEL_EXPANSION_MODE_IN_CMOS_SUPPORTED (1 << 7) # define ATIF_TEMPERATURE_CHANGE_NOTIFICATION_SUPPORTED (1 << 12) -# define ATIF_GET_GRAPHICS_DEVICE_TYPES_SUPPORTED (1 << 14) +# define ATIF_QUERY_BACKLIGHT_TRANSFER_CHARACTERISTICS_SUPPORTED (1 << 15) +# define ATIF_READY_TO_UNDOCK_NOTIFICATION_SUPPORTED (1 << 16) # define ATIF_GET_EXTERNAL_GPU_INFORMATION_SUPPORTED (1 << 20) #define ATIF_FUNCTION_GET_SYSTEM_PARAMETERS 0x1 /* ARG0: ATIF_FUNCTION_GET_SYSTEM_PARAMETERS @@ -170,6 +186,10 @@ struct atcs_pref_req_output { * n (0xd0-0xd9) is specified in notify command code. * bit 2: * 1 - lid changes not reported though int10 + * bit 3: + * 1 - system bios controls overclocking + * bit 4: + * 1 - enable overclocking */ #define ATIF_FUNCTION_GET_SYSTEM_BIOS_REQUESTS 0x2 /* ARG0: ATIF_FUNCTION_GET_SYSTEM_BIOS_REQUESTS @@ -177,28 +197,23 @@ struct atcs_pref_req_output { * OUTPUT: * WORD - structure size in bytes (includes size field) * DWORD - pending sbios requests - * BYTE - panel expansion mode + * BYTE - reserved (all zeroes) * BYTE - thermal state: target gfx controller * BYTE - thermal state: state id (0: exit state, non-0: state) * BYTE - forced power state: target gfx controller - * BYTE - forced power state: state id + * BYTE - forced power state: state id (0: forced state, non-0: state) * BYTE - system power source * BYTE - panel backlight level (0-255) + * BYTE - GPU package power limit: target gfx controller + * DWORD - GPU package power limit: value (24:8 fractional format, Watts) */ /* pending sbios requests */ -# define ATIF_DISPLAY_SWITCH_REQUEST (1 << 0) -# define ATIF_EXPANSION_MODE_CHANGE_REQUEST (1 << 1) # define ATIF_THERMAL_STATE_CHANGE_REQUEST (1 << 2) # define ATIF_FORCED_POWER_STATE_CHANGE_REQUEST (1 << 3) # define ATIF_SYSTEM_POWER_SOURCE_CHANGE_REQUEST (1 << 4) -# define ATIF_DISPLAY_CONF_CHANGE_REQUEST (1 << 5) -# define ATIF_PX_GFX_SWITCH_REQUEST (1 << 6) # define ATIF_PANEL_BRIGHTNESS_CHANGE_REQUEST (1 << 7) # define ATIF_DGPU_DISPLAY_EVENT (1 << 8) -/* panel expansion mode */ -# define ATIF_PANEL_EXPANSION_DISABLE 0 -# define ATIF_PANEL_EXPANSION_FULL 1 -# define ATIF_PANEL_EXPANSION_ASPECT 2 +# define ATIF_GPU_PACKAGE_POWER_LIMIT_REQUEST (1 << 12) /* target gfx controller */ # define ATIF_TARGET_GFX_SINGLE 0 # define ATIF_TARGET_GFX_PX_IGPU 1 @@ -208,76 +223,6 @@ struct atcs_pref_req_output { # define ATIF_POWER_SOURCE_DC 2 # define ATIF_POWER_SOURCE_RESTRICTED_AC_1 3 # define ATIF_POWER_SOURCE_RESTRICTED_AC_2 4 -#define ATIF_FUNCTION_SELECT_ACTIVE_DISPLAYS 0x3 -/* ARG0: ATIF_FUNCTION_SELECT_ACTIVE_DISPLAYS - * ARG1: - * WORD - structure size in bytes (includes size field) - * WORD - selected displays - * WORD - connected displays - * OUTPUT: - * WORD - structure size in bytes (includes size field) - * WORD - selected displays - */ -# define ATIF_LCD1 (1 << 0) -# define ATIF_CRT1 (1 << 1) -# define ATIF_TV (1 << 2) -# define ATIF_DFP1 (1 << 3) -# define ATIF_CRT2 (1 << 4) -# define ATIF_LCD2 (1 << 5) -# define ATIF_DFP2 (1 << 7) -# define ATIF_CV (1 << 8) -# define ATIF_DFP3 (1 << 9) -# define ATIF_DFP4 (1 << 10) -# define ATIF_DFP5 (1 << 11) -# define ATIF_DFP6 (1 << 12) -#define ATIF_FUNCTION_GET_LID_STATE 0x4 -/* ARG0: ATIF_FUNCTION_GET_LID_STATE - * ARG1: none - * OUTPUT: - * WORD - structure size in bytes (includes size field) - * BYTE - lid state (0: open, 1: closed) - * - * GET_LID_STATE only works at boot and resume, for general lid - * status, use the kernel provided status - */ -#define ATIF_FUNCTION_GET_TV_STANDARD_FROM_CMOS 0x5 -/* ARG0: ATIF_FUNCTION_GET_TV_STANDARD_FROM_CMOS - * ARG1: none - * OUTPUT: - * WORD - structure size in bytes (includes size field) - * BYTE - 0 - * BYTE - TV standard - */ -# define ATIF_TV_STD_NTSC 0 -# define ATIF_TV_STD_PAL 1 -# define ATIF_TV_STD_PALM 2 -# define ATIF_TV_STD_PAL60 3 -# define ATIF_TV_STD_NTSCJ 4 -# define ATIF_TV_STD_PALCN 5 -# define ATIF_TV_STD_PALN 6 -# define ATIF_TV_STD_SCART_RGB 9 -#define ATIF_FUNCTION_SET_TV_STANDARD_IN_CMOS 0x6 -/* ARG0: ATIF_FUNCTION_SET_TV_STANDARD_IN_CMOS - * ARG1: - * WORD - structure size in bytes (includes size field) - * BYTE - 0 - * BYTE - TV standard - * OUTPUT: none - */ -#define ATIF_FUNCTION_GET_PANEL_EXPANSION_MODE_FROM_CMOS 0x7 -/* ARG0: ATIF_FUNCTION_GET_PANEL_EXPANSION_MODE_FROM_CMOS - * ARG1: none - * OUTPUT: - * WORD - structure size in bytes (includes size field) - * BYTE - panel expansion mode - */ -#define ATIF_FUNCTION_SET_PANEL_EXPANSION_MODE_IN_CMOS 0x8 -/* ARG0: ATIF_FUNCTION_SET_PANEL_EXPANSION_MODE_IN_CMOS - * ARG1: - * WORD - structure size in bytes (includes size field) - * BYTE - panel expansion mode - * OUTPUT: none - */ #define ATIF_FUNCTION_TEMPERATURE_CHANGE_NOTIFICATION 0xD /* ARG0: ATIF_FUNCTION_TEMPERATURE_CHANGE_NOTIFICATION * ARG1: @@ -286,21 +231,43 @@ struct atcs_pref_req_output { * BYTE - current temperature (degress Celsius) * OUTPUT: none */ -#define ATIF_FUNCTION_GET_GRAPHICS_DEVICE_TYPES 0xF -/* ARG0: ATIF_FUNCTION_GET_GRAPHICS_DEVICE_TYPES - * ARG1: none +#define ATIF_FUNCTION_QUERY_BRIGHTNESS_TRANSFER_CHARACTERISTICS 0x10 +/* ARG0: ATIF_FUNCTION_QUERY_BRIGHTNESS_TRANSFER_CHARACTERISTICS + * ARG1: + * WORD - structure size in bytes (includes size field) + * BYTE - requested display * OUTPUT: - * WORD - number of gfx devices - * WORD - device structure size in bytes (excludes device size field) - * DWORD - flags \ - * WORD - bus number } repeated structure - * WORD - device number / + * WORD - structure size in bytes (includes size field) + * WORD - flags (currently all 16 bits are reserved) + * BYTE - error code (on failure, disregard all below fields) + * BYTE - AC level (default brightness in percent when machine has full power) + * BYTE - DC level (default brightness in percent when machine is on battery) + * BYTE - min input signal, in range 0-255, corresponding to 0% backlight + * BYTE - max input signal, in range 0-255, corresponding to 100% backlight + * BYTE - number of reported data points + * BYTE - luminance level in percent \ repeated structure + * BYTE - input signal in range 0-255 / does not have entries for 0% and 100% + */ +/* requested display */ +# define ATIF_QBTC_REQUEST_LCD1 0 +# define ATIF_QBTC_REQUEST_CRT1 1 +# define ATIF_QBTC_REQUEST_DFP1 3 +# define ATIF_QBTC_REQUEST_CRT2 4 +# define ATIF_QBTC_REQUEST_LCD2 5 +# define ATIF_QBTC_REQUEST_DFP2 7 +# define ATIF_QBTC_REQUEST_DFP3 9 +# define ATIF_QBTC_REQUEST_DFP4 10 +# define ATIF_QBTC_REQUEST_DFP5 11 +# define ATIF_QBTC_REQUEST_DFP6 12 +/* error code */ +# define ATIF_QBTC_ERROR_CODE_SUCCESS 0 +# define ATIF_QBTC_ERROR_CODE_FAILURE 1 +# define ATIF_QBTC_ERROR_CODE_DEVICE_NOT_SUPPORTED 2 +#define ATIF_FUNCTION_READY_TO_UNDOCK_NOTIFICATION 0x11 +/* ARG0: ATIF_FUNCTION_READY_TO_UNDOCK_NOTIFICATION + * ARG1: none + * OUTPUT: none */ -/* flags */ -# define ATIF_PX_REMOVABLE_GRAPHICS_DEVICE (1 << 0) -# define ATIF_XGP_PORT (1 << 1) -# define ATIF_VGA_ENABLED_GRAPHICS_DEVICE (1 << 2) -# define ATIF_XGP_PORT_IN_DOCK (1 << 3) #define ATIF_FUNCTION_GET_EXTERNAL_GPU_INFORMATION 0x15 /* ARG0: ATIF_FUNCTION_GET_EXTERNAL_GPU_INFORMATION * ARG1: none diff --git a/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_9_4_0_offset.h b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_9_4_0_offset.h new file mode 100644 index 000000000000..8f515875a34d --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_9_4_0_offset.h @@ -0,0 +1,32 @@ +/* + * Copyright (C) 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included + * in all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS + * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN + * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + */ +#ifndef _mmhub_9_4_0_OFFSET_HEADER +#define _mmhub_9_4_0_OFFSET_HEADER + + +// addressBlock: mmhub_utcl2_vmsharedpfdec +// base address: 0x6a040 +#define mmMC_VM_XGMI_LFB_CNTL 0x0823 +#define mmMC_VM_XGMI_LFB_CNTL_BASE_IDX 0 +#define mmMC_VM_XGMI_LFB_SIZE 0x0824 +#define mmMC_VM_XGMI_LFB_SIZE_BASE_IDX 0 + +#endif diff --git a/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_9_4_0_sh_mask.h b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_9_4_0_sh_mask.h new file mode 100644 index 000000000000..0a6b072d191e --- /dev/null +++ b/drivers/gpu/drm/amd/include/asic_reg/mmhub/mmhub_9_4_0_sh_mask.h @@ -0,0 +1,35 @@ +/* + * Copyright (C) 2018 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included + * in all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS + * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN + * AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN + * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + */ +#ifndef _mmhub_9_4_0_SH_MASK_HEADER +#define _mmhub_9_4_0_SH_MASK_HEADER + + +// addressBlock: mmhub_utcl2_vmsharedpfdec +//MC_VM_XGMI_LFB_CNTL +#define MC_VM_XGMI_LFB_CNTL__PF_LFB_REGION__SHIFT 0x0 +#define MC_VM_XGMI_LFB_CNTL__PF_MAX_REGION__SHIFT 0x4 +#define MC_VM_XGMI_LFB_CNTL__PF_LFB_REGION_MASK 0x00000007L +#define MC_VM_XGMI_LFB_CNTL__PF_MAX_REGION_MASK 0x00000070L +//MC_VM_XGMI_LFB_SIZE +#define MC_VM_XGMI_LFB_SIZE__PF_LFB_SIZE__SHIFT 0x0 +#define MC_VM_XGMI_LFB_SIZE__PF_LFB_SIZE_MASK 0x0000FFFFL + +#endif diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h index 58ac0b90c310..8154d67388cc 100644 --- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h @@ -188,8 +188,8 @@ struct tile_config { */ #define ALLOC_MEM_FLAGS_VRAM (1 << 0) #define ALLOC_MEM_FLAGS_GTT (1 << 1) -#define ALLOC_MEM_FLAGS_USERPTR (1 << 2) /* TODO */ -#define ALLOC_MEM_FLAGS_DOORBELL (1 << 3) /* TODO */ +#define ALLOC_MEM_FLAGS_USERPTR (1 << 2) +#define ALLOC_MEM_FLAGS_DOORBELL (1 << 3) /* * Allocation flags attributes/access options. diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h index 980e696989b1..1479ea1dc3e7 100644 --- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h @@ -276,6 +276,10 @@ struct amd_pm_funcs { struct amd_pp_simple_clock_info *clocks); int (*notify_smu_enable_pwe)(void *handle); int (*enable_mgpu_fan_boost)(void *handle); + int (*set_active_display_count)(void *handle, uint32_t count); + int (*set_hard_min_dcefclk_by_freq)(void *handle, uint32_t clock); + int (*set_hard_min_fclk_by_freq)(void *handle, uint32_t clock); + int (*set_min_deep_sleep_dcefclk)(void *handle, uint32_t clock); }; #endif diff --git a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c index d6aa1d414320..9bc27f468d5b 100644 --- a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c +++ b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c @@ -300,7 +300,7 @@ static int pp_set_clockgating_by_smu(void *handle, uint32_t msg_id) return -EINVAL; if (hwmgr->hwmgr_func->update_clock_gatings == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } @@ -387,7 +387,7 @@ static uint32_t pp_dpm_get_sclk(void *handle, bool low) return 0; if (hwmgr->hwmgr_func->get_sclk == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } mutex_lock(&hwmgr->smu_lock); @@ -405,7 +405,7 @@ static uint32_t pp_dpm_get_mclk(void *handle, bool low) return 0; if (hwmgr->hwmgr_func->get_mclk == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } mutex_lock(&hwmgr->smu_lock); @@ -422,7 +422,7 @@ static void pp_dpm_powergate_vce(void *handle, bool gate) return; if (hwmgr->hwmgr_func->powergate_vce == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return; } mutex_lock(&hwmgr->smu_lock); @@ -438,7 +438,7 @@ static void pp_dpm_powergate_uvd(void *handle, bool gate) return; if (hwmgr->hwmgr_func->powergate_uvd == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return; } mutex_lock(&hwmgr->smu_lock); @@ -505,7 +505,7 @@ static void pp_dpm_set_fan_control_mode(void *handle, uint32_t mode) return; if (hwmgr->hwmgr_func->set_fan_control_mode == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return; } mutex_lock(&hwmgr->smu_lock); @@ -522,7 +522,7 @@ static uint32_t pp_dpm_get_fan_control_mode(void *handle) return 0; if (hwmgr->hwmgr_func->get_fan_control_mode == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } mutex_lock(&hwmgr->smu_lock); @@ -540,7 +540,7 @@ static int pp_dpm_set_fan_speed_percent(void *handle, uint32_t percent) return -EINVAL; if (hwmgr->hwmgr_func->set_fan_speed_percent == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } mutex_lock(&hwmgr->smu_lock); @@ -558,7 +558,7 @@ static int pp_dpm_get_fan_speed_percent(void *handle, uint32_t *speed) return -EINVAL; if (hwmgr->hwmgr_func->get_fan_speed_percent == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } @@ -594,7 +594,7 @@ static int pp_dpm_set_fan_speed_rpm(void *handle, uint32_t rpm) return -EINVAL; if (hwmgr->hwmgr_func->set_fan_speed_rpm == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } mutex_lock(&hwmgr->smu_lock); @@ -720,12 +720,12 @@ static int pp_dpm_force_clock_level(void *handle, return -EINVAL; if (hwmgr->hwmgr_func->force_clock_level == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } if (hwmgr->dpm_level != AMD_DPM_FORCED_LEVEL_MANUAL) { - pr_info("force clock level is for dpm manual mode only.\n"); + pr_debug("force clock level is for dpm manual mode only.\n"); return -EINVAL; } @@ -745,7 +745,7 @@ static int pp_dpm_print_clock_levels(void *handle, return -EINVAL; if (hwmgr->hwmgr_func->print_clock_levels == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } mutex_lock(&hwmgr->smu_lock); @@ -763,7 +763,7 @@ static int pp_dpm_get_sclk_od(void *handle) return -EINVAL; if (hwmgr->hwmgr_func->get_sclk_od == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } mutex_lock(&hwmgr->smu_lock); @@ -781,7 +781,7 @@ static int pp_dpm_set_sclk_od(void *handle, uint32_t value) return -EINVAL; if (hwmgr->hwmgr_func->set_sclk_od == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } @@ -800,7 +800,7 @@ static int pp_dpm_get_mclk_od(void *handle) return -EINVAL; if (hwmgr->hwmgr_func->get_mclk_od == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } mutex_lock(&hwmgr->smu_lock); @@ -818,7 +818,7 @@ static int pp_dpm_set_mclk_od(void *handle, uint32_t value) return -EINVAL; if (hwmgr->hwmgr_func->set_mclk_od == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } mutex_lock(&hwmgr->smu_lock); @@ -878,7 +878,7 @@ static int pp_get_power_profile_mode(void *handle, char *buf) return -EINVAL; if (hwmgr->hwmgr_func->get_power_profile_mode == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return snprintf(buf, PAGE_SIZE, "\n"); } @@ -894,12 +894,12 @@ static int pp_set_power_profile_mode(void *handle, long *input, uint32_t size) return ret; if (hwmgr->hwmgr_func->set_power_profile_mode == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return ret; } if (hwmgr->dpm_level != AMD_DPM_FORCED_LEVEL_MANUAL) { - pr_info("power profile setting is for manual dpm mode only.\n"); + pr_debug("power profile setting is for manual dpm mode only.\n"); return ret; } @@ -917,7 +917,7 @@ static int pp_odn_edit_dpm_table(void *handle, uint32_t type, long *input, uint3 return -EINVAL; if (hwmgr->hwmgr_func->odn_edit_dpm_table == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return -EINVAL; } @@ -935,7 +935,7 @@ static int pp_dpm_switch_power_profile(void *handle, return -EINVAL; if (hwmgr->hwmgr_func->set_power_profile_mode == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return -EINVAL; } @@ -972,7 +972,7 @@ static int pp_set_power_limit(void *handle, uint32_t limit) return -EINVAL; if (hwmgr->hwmgr_func->set_power_limit == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return -EINVAL; } @@ -1072,7 +1072,7 @@ static int pp_get_current_clocks(void *handle, &hw_clocks, PHM_PerformanceLevelDesignation_Activity); if (ret) { - pr_info("Error in phm_get_clock_info \n"); + pr_debug("Error in phm_get_clock_info \n"); mutex_unlock(&hwmgr->smu_lock); return -EINVAL; } @@ -1212,7 +1212,7 @@ static int pp_dpm_powergate_mmhub(void *handle) return -EINVAL; if (hwmgr->hwmgr_func->powergate_mmhub == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } @@ -1227,7 +1227,7 @@ static int pp_dpm_powergate_gfx(void *handle, bool gate) return 0; if (hwmgr->hwmgr_func->powergate_gfx == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return 0; } @@ -1242,7 +1242,7 @@ static void pp_dpm_powergate_acp(void *handle, bool gate) return; if (hwmgr->hwmgr_func->powergate_acp == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return; } @@ -1257,7 +1257,7 @@ static void pp_dpm_powergate_sdma(void *handle, bool gate) return; if (hwmgr->hwmgr_func->powergate_sdma == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return; } @@ -1303,7 +1303,7 @@ static int pp_notify_smu_enable_pwe(void *handle) return -EINVAL; if (hwmgr->hwmgr_func->smus_notify_pwe == NULL) { - pr_info("%s was not implemented.\n", __func__); + pr_info_ratelimited("%s was not implemented.\n", __func__); return -EINVAL;; } @@ -1332,6 +1332,78 @@ static int pp_enable_mgpu_fan_boost(void *handle) return 0; } +static int pp_set_min_deep_sleep_dcefclk(void *handle, uint32_t clock) +{ + struct pp_hwmgr *hwmgr = handle; + + if (!hwmgr || !hwmgr->pm_en) + return -EINVAL; + + if (hwmgr->hwmgr_func->set_min_deep_sleep_dcefclk == NULL) { + pr_debug("%s was not implemented.\n", __func__); + return -EINVAL;; + } + + mutex_lock(&hwmgr->smu_lock); + hwmgr->hwmgr_func->set_min_deep_sleep_dcefclk(hwmgr, clock); + mutex_unlock(&hwmgr->smu_lock); + + return 0; +} + +static int pp_set_hard_min_dcefclk_by_freq(void *handle, uint32_t clock) +{ + struct pp_hwmgr *hwmgr = handle; + + if (!hwmgr || !hwmgr->pm_en) + return -EINVAL; + + if (hwmgr->hwmgr_func->set_hard_min_dcefclk_by_freq == NULL) { + pr_debug("%s was not implemented.\n", __func__); + return -EINVAL;; + } + + mutex_lock(&hwmgr->smu_lock); + hwmgr->hwmgr_func->set_hard_min_dcefclk_by_freq(hwmgr, clock); + mutex_unlock(&hwmgr->smu_lock); + + return 0; +} + +static int pp_set_hard_min_fclk_by_freq(void *handle, uint32_t clock) +{ + struct pp_hwmgr *hwmgr = handle; + + if (!hwmgr || !hwmgr->pm_en) + return -EINVAL; + + if (hwmgr->hwmgr_func->set_hard_min_fclk_by_freq == NULL) { + pr_debug("%s was not implemented.\n", __func__); + return -EINVAL;; + } + + mutex_lock(&hwmgr->smu_lock); + hwmgr->hwmgr_func->set_hard_min_fclk_by_freq(hwmgr, clock); + mutex_unlock(&hwmgr->smu_lock); + + return 0; +} + +static int pp_set_active_display_count(void *handle, uint32_t count) +{ + struct pp_hwmgr *hwmgr = handle; + int ret = 0; + + if (!hwmgr || !hwmgr->pm_en) + return -EINVAL; + + mutex_lock(&hwmgr->smu_lock); + ret = phm_set_active_display_count(hwmgr, count); + mutex_unlock(&hwmgr->smu_lock); + + return ret; +} + static const struct amd_pm_funcs pp_dpm_funcs = { .load_firmware = pp_dpm_load_fw, .wait_for_fw_loading_complete = pp_dpm_fw_loading_complete, @@ -1378,4 +1450,8 @@ static const struct amd_pm_funcs pp_dpm_funcs = { .get_display_mode_validation_clocks = pp_get_display_mode_validation_clocks, .notify_smu_enable_pwe = pp_notify_smu_enable_pwe, .enable_mgpu_fan_boost = pp_enable_mgpu_fan_boost, + .set_active_display_count = pp_set_active_display_count, + .set_min_deep_sleep_dcefclk = pp_set_min_deep_sleep_dcefclk, + .set_hard_min_dcefclk_by_freq = pp_set_hard_min_dcefclk_by_freq, + .set_hard_min_fclk_by_freq = pp_set_hard_min_fclk_by_freq, }; diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c b/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c index 85119c2bdcc8..1f92a9f4c9e3 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/hardwaremanager.c @@ -80,7 +80,9 @@ int phm_enable_dynamic_state_management(struct pp_hwmgr *hwmgr) PHM_FUNC_CHECK(hwmgr); adev = hwmgr->adev; - if (smum_is_dpm_running(hwmgr) && !amdgpu_passthrough(adev)) { + /* Skip for suspend/resume case */ + if (smum_is_dpm_running(hwmgr) && !amdgpu_passthrough(adev) + && adev->in_suspend) { pr_info("dpm has been enabled\n"); return 0; } @@ -286,8 +288,8 @@ int phm_store_dal_configuration_data(struct pp_hwmgr *hwmgr, if (display_config == NULL) return -EINVAL; - if (NULL != hwmgr->hwmgr_func->set_deep_sleep_dcefclk) - hwmgr->hwmgr_func->set_deep_sleep_dcefclk(hwmgr, display_config->min_dcef_deep_sleep_set_clk); + if (NULL != hwmgr->hwmgr_func->set_min_deep_sleep_dcefclk) + hwmgr->hwmgr_func->set_min_deep_sleep_dcefclk(hwmgr, display_config->min_dcef_deep_sleep_set_clk); for (index = 0; index < display_config->num_path_including_non_display; index++) { if (display_config->displays[index].controller_id != 0) @@ -478,3 +480,44 @@ int phm_disable_smc_firmware_ctf(struct pp_hwmgr *hwmgr) return hwmgr->hwmgr_func->disable_smc_firmware_ctf(hwmgr); } + +int phm_set_active_display_count(struct pp_hwmgr *hwmgr, uint32_t count) +{ + PHM_FUNC_CHECK(hwmgr); + + if (!hwmgr->hwmgr_func->set_active_display_count) + return -EINVAL; + + return hwmgr->hwmgr_func->set_active_display_count(hwmgr, count); +} + +int phm_set_min_deep_sleep_dcefclk(struct pp_hwmgr *hwmgr, uint32_t clock) +{ + PHM_FUNC_CHECK(hwmgr); + + if (!hwmgr->hwmgr_func->set_min_deep_sleep_dcefclk) + return -EINVAL; + + return hwmgr->hwmgr_func->set_min_deep_sleep_dcefclk(hwmgr, clock); +} + +int phm_set_hard_min_dcefclk_by_freq(struct pp_hwmgr *hwmgr, uint32_t clock) +{ + PHM_FUNC_CHECK(hwmgr); + + if (!hwmgr->hwmgr_func->set_hard_min_dcefclk_by_freq) + return -EINVAL; + + return hwmgr->hwmgr_func->set_hard_min_dcefclk_by_freq(hwmgr, clock); +} + +int phm_set_hard_min_fclk_by_freq(struct pp_hwmgr *hwmgr, uint32_t clock) +{ + PHM_FUNC_CHECK(hwmgr); + + if (!hwmgr->hwmgr_func->set_hard_min_fclk_by_freq) + return -EINVAL; + + return hwmgr->hwmgr_func->set_hard_min_fclk_by_freq(hwmgr, clock); +} + diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c index 47ac92369739..0173d0480024 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/hwmgr.c @@ -352,6 +352,9 @@ int hwmgr_handle_task(struct pp_hwmgr *hwmgr, enum amd_pp_task task_id, switch (task_id) { case AMD_PP_TASK_DISPLAY_CONFIG_CHANGE: + ret = phm_pre_display_configuration_changed(hwmgr); + if (ret) + return ret; ret = phm_set_cpu_power_state(hwmgr); if (ret) return ret; diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c b/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c index 91ffb7bc4ee7..56437866d120 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/pp_psm.c @@ -265,8 +265,6 @@ int psm_adjust_power_state_dynamic(struct pp_hwmgr *hwmgr, bool skip, if (skip) return 0; - phm_pre_display_configuration_changed(hwmgr); - phm_display_configuration_changed(hwmgr); if (hwmgr->ps) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c index dd18cb710391..f95c5f50eb0f 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu10_hwmgr.c @@ -216,12 +216,12 @@ static inline uint32_t convert_10k_to_mhz(uint32_t clock) return (clock + 99) / 100; } -static int smu10_set_deep_sleep_dcefclk(struct pp_hwmgr *hwmgr, uint32_t clock) +static int smu10_set_min_deep_sleep_dcefclk(struct pp_hwmgr *hwmgr, uint32_t clock) { struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend); if (smu10_data->need_min_deep_sleep_dcefclk && - smu10_data->deep_sleep_dcefclk != convert_10k_to_mhz(clock)) { + smu10_data->deep_sleep_dcefclk != convert_10k_to_mhz(clock)) { smu10_data->deep_sleep_dcefclk = convert_10k_to_mhz(clock); smum_send_msg_to_smc_with_parameter(hwmgr, PPSMC_MSG_SetMinDeepSleepDcefclk, @@ -230,6 +230,34 @@ static int smu10_set_deep_sleep_dcefclk(struct pp_hwmgr *hwmgr, uint32_t clock) return 0; } +static int smu10_set_hard_min_dcefclk_by_freq(struct pp_hwmgr *hwmgr, uint32_t clock) +{ + struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend); + + if (smu10_data->dcf_actual_hard_min_freq && + smu10_data->dcf_actual_hard_min_freq != convert_10k_to_mhz(clock)) { + smu10_data->dcf_actual_hard_min_freq = convert_10k_to_mhz(clock); + smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_SetHardMinDcefclkByFreq, + smu10_data->dcf_actual_hard_min_freq); + } + return 0; +} + +static int smu10_set_hard_min_fclk_by_freq(struct pp_hwmgr *hwmgr, uint32_t clock) +{ + struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend); + + if (smu10_data->f_actual_hard_min_freq && + smu10_data->f_actual_hard_min_freq != convert_10k_to_mhz(clock)) { + smu10_data->f_actual_hard_min_freq = convert_10k_to_mhz(clock); + smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_SetHardMinFclkByFreq, + smu10_data->f_actual_hard_min_freq); + } + return 0; +} + static int smu10_set_active_display_count(struct pp_hwmgr *hwmgr, uint32_t count) { struct smu10_hwmgr *smu10_data = (struct smu10_hwmgr *)(hwmgr->backend); @@ -1206,7 +1234,7 @@ static const struct pp_hwmgr_func smu10_hwmgr_funcs = { .get_max_high_clocks = smu10_get_max_high_clocks, .read_sensor = smu10_read_sensor, .set_active_display_count = smu10_set_active_display_count, - .set_deep_sleep_dcefclk = smu10_set_deep_sleep_dcefclk, + .set_min_deep_sleep_dcefclk = smu10_set_min_deep_sleep_dcefclk, .dynamic_state_management_enable = smu10_enable_dpm_tasks, .power_off_asic = smu10_power_off_asic, .asic_setup = smu10_setup_asic_task, @@ -1217,6 +1245,8 @@ static const struct pp_hwmgr_func smu10_hwmgr_funcs = { .display_clock_voltage_request = smu10_display_clock_voltage_request, .powergate_gfx = smu10_gfx_off_control, .powergate_sdma = smu10_powergate_sdma, + .set_hard_min_dcefclk_by_freq = smu10_set_hard_min_dcefclk_by_freq, + .set_hard_min_fclk_by_freq = smu10_set_hard_min_fclk_by_freq, }; int smu10_init_function_pointers(struct pp_hwmgr *hwmgr) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c index 88f6b35ea6fe..d91390459326 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c @@ -269,7 +269,7 @@ static int smu7_construct_voltage_tables(struct pp_hwmgr *hwmgr) hwmgr->dyn_state.mvdd_dependency_on_mclk); PP_ASSERT_WITH_CODE((0 == result), - "Failed to retrieve SVI2 MVDD table from dependancy table.", + "Failed to retrieve SVI2 MVDD table from dependency table.", return result;); } @@ -288,7 +288,7 @@ static int smu7_construct_voltage_tables(struct pp_hwmgr *hwmgr) result = phm_get_svi2_voltage_table_v0(&(data->vddci_voltage_table), hwmgr->dyn_state.vddci_dependency_on_mclk); PP_ASSERT_WITH_CODE((0 == result), - "Failed to retrieve SVI2 VDDCI table from dependancy table.", + "Failed to retrieve SVI2 VDDCI table from dependency table.", return result); } @@ -317,7 +317,7 @@ static int smu7_construct_voltage_tables(struct pp_hwmgr *hwmgr) table_info->vddc_lookup_table); PP_ASSERT_WITH_CODE((0 == result), - "Failed to retrieve SVI2 VDDC table from dependancy table.", return result;); + "Failed to retrieve SVI2 VDDC table from dependency table.", return result;); } tmp = smum_get_mac_definition(hwmgr, SMU_MAX_LEVELS_VDDC); @@ -2859,7 +2859,10 @@ static int smu7_vblank_too_short(struct pp_hwmgr *hwmgr, case CHIP_POLARIS10: case CHIP_POLARIS11: case CHIP_POLARIS12: - switch_limit_us = data->is_memory_gddr5 ? 190 : 150; + if (hwmgr->is_kicker) + switch_limit_us = data->is_memory_gddr5 ? 450 : 150; + else + switch_limit_us = data->is_memory_gddr5 ? 190 : 150; break; case CHIP_VEGAM: switch_limit_us = 30; @@ -3589,8 +3592,10 @@ static int smu7_find_dpm_states_clocks_in_dpm_table(struct pp_hwmgr *hwmgr, cons } if (i >= sclk_table->count) { - data->need_update_smu7_dpm_table |= DPMTABLE_OD_UPDATE_SCLK; - sclk_table->dpm_levels[i-1].value = sclk; + if (sclk > sclk_table->dpm_levels[i-1].value) { + data->need_update_smu7_dpm_table |= DPMTABLE_OD_UPDATE_SCLK; + sclk_table->dpm_levels[i-1].value = sclk; + } } else { /* TODO: Check SCLK in DAL's minimum clocks * in case DeepSleep divider update is required. @@ -3607,8 +3612,10 @@ static int smu7_find_dpm_states_clocks_in_dpm_table(struct pp_hwmgr *hwmgr, cons } if (i >= mclk_table->count) { - data->need_update_smu7_dpm_table |= DPMTABLE_OD_UPDATE_MCLK; - mclk_table->dpm_levels[i-1].value = mclk; + if (mclk > mclk_table->dpm_levels[i-1].value) { + data->need_update_smu7_dpm_table |= DPMTABLE_OD_UPDATE_MCLK; + mclk_table->dpm_levels[i-1].value = mclk; + } } if (data->display_timing.num_existing_displays != hwmgr->display_config->num_display) @@ -4219,9 +4226,17 @@ static int smu7_check_mc_firmware(struct pp_hwmgr *hwmgr) if (tmp & (1 << 23)) { data->mem_latency_high = MEM_LATENCY_HIGH; data->mem_latency_low = MEM_LATENCY_LOW; + if ((hwmgr->chip_id == CHIP_POLARIS10) || + (hwmgr->chip_id == CHIP_POLARIS11) || + (hwmgr->chip_id == CHIP_POLARIS12)) + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableFFC); } else { data->mem_latency_high = 330; data->mem_latency_low = 330; + if ((hwmgr->chip_id == CHIP_POLARIS10) || + (hwmgr->chip_id == CHIP_POLARIS11) || + (hwmgr->chip_id == CHIP_POLARIS12)) + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_DisableFFC); } return 0; diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu8_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu8_hwmgr.c index fef111ddb736..553a203ac47c 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu8_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu8_hwmgr.c @@ -1228,17 +1228,14 @@ static int smu8_dpm_force_dpm_level(struct pp_hwmgr *hwmgr, static int smu8_dpm_powerdown_uvd(struct pp_hwmgr *hwmgr) { - if (PP_CAP(PHM_PlatformCaps_UVDPowerGating)) { - smu8_nbdpm_pstate_enable_disable(hwmgr, true, true); + if (PP_CAP(PHM_PlatformCaps_UVDPowerGating)) return smum_send_msg_to_smc(hwmgr, PPSMC_MSG_UVDPowerOFF); - } return 0; } static int smu8_dpm_powerup_uvd(struct pp_hwmgr *hwmgr) { if (PP_CAP(PHM_PlatformCaps_UVDPowerGating)) { - smu8_nbdpm_pstate_enable_disable(hwmgr, false, true); return smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_UVDPowerON, @@ -1995,6 +1992,7 @@ static const struct pp_hwmgr_func smu8_hwmgr_funcs = { .power_state_set = smu8_set_power_state_tasks, .dynamic_state_management_disable = smu8_disable_dpm_tasks, .notify_cac_buffer_info = smu8_notify_cac_buffer_info, + .update_nbdpm_pstate = smu8_nbdpm_pstate_enable_disable, .get_thermal_temperature_range = smu8_get_thermal_temperature_range, }; diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c index e2bc6e0c229f..79c86247d0ac 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_hwmgr.c @@ -3266,8 +3266,10 @@ static int vega10_find_dpm_states_clocks_in_dpm_table(struct pp_hwmgr *hwmgr, co } if (i >= sclk_table->count) { - data->need_update_dpm_table |= DPMTABLE_OD_UPDATE_SCLK; - sclk_table->dpm_levels[i-1].value = sclk; + if (sclk > sclk_table->dpm_levels[i-1].value) { + data->need_update_dpm_table |= DPMTABLE_OD_UPDATE_SCLK; + sclk_table->dpm_levels[i-1].value = sclk; + } } for (i = 0; i < mclk_table->count; i++) { @@ -3276,8 +3278,10 @@ static int vega10_find_dpm_states_clocks_in_dpm_table(struct pp_hwmgr *hwmgr, co } if (i >= mclk_table->count) { - data->need_update_dpm_table |= DPMTABLE_OD_UPDATE_MCLK; - mclk_table->dpm_levels[i-1].value = mclk; + if (mclk > mclk_table->dpm_levels[i-1].value) { + data->need_update_dpm_table |= DPMTABLE_OD_UPDATE_MCLK; + mclk_table->dpm_levels[i-1].value = mclk; + } } if (data->display_timing.num_existing_displays != hwmgr->display_config->num_display) diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c index 2679d1240fa1..26154f9b2178 100644 --- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c +++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c @@ -49,6 +49,10 @@ #include "soc15_common.h" #include "smuio/smuio_9_0_offset.h" #include "smuio/smuio_9_0_sh_mask.h" +#include "nbio/nbio_7_4_sh_mask.h" + +#define smnPCIE_LC_SPEED_CNTL 0x11140290 +#define smnPCIE_LC_LINK_WIDTH_CNTL 0x11140288 static void vega20_set_default_registry_data(struct pp_hwmgr *hwmgr) { @@ -130,7 +134,7 @@ static void vega20_set_default_registry_data(struct pp_hwmgr *hwmgr) data->registry_data.disable_auto_wattman = 1; data->registry_data.auto_wattman_debug = 0; data->registry_data.auto_wattman_sample_period = 100; - data->registry_data.fclk_gfxclk_ratio = 0x3F6CCCCD; + data->registry_data.fclk_gfxclk_ratio = 0; data->registry_data.auto_wattman_threshold = 50; data->registry_data.gfxoff_controlled_by_driver = 1; data->gfxoff_allowed = false; @@ -1660,14 +1664,15 @@ static uint32_t vega20_find_highest_dpm_level( return i; } -static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) +static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr, uint32_t feature_mask) { struct vega20_hwmgr *data = (struct vega20_hwmgr *)(hwmgr->backend); uint32_t min_freq; int ret = 0; - if (data->smu_features[GNLD_DPM_GFXCLK].enabled) { + if (data->smu_features[GNLD_DPM_GFXCLK].enabled && + (feature_mask & FEATURE_DPM_GFXCLK_MASK)) { min_freq = data->dpm_table.gfx_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, @@ -1676,7 +1681,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_UCLK].enabled) { + if (data->smu_features[GNLD_DPM_UCLK].enabled && + (feature_mask & FEATURE_DPM_UCLK_MASK)) { min_freq = data->dpm_table.mem_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( hwmgr, PPSMC_MSG_SetSoftMinByFreq, @@ -1692,7 +1698,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_UVD].enabled) { + if (data->smu_features[GNLD_DPM_UVD].enabled && + (feature_mask & FEATURE_DPM_UVD_MASK)) { min_freq = data->dpm_table.vclk_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1710,7 +1717,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_VCE].enabled) { + if (data->smu_features[GNLD_DPM_VCE].enabled && + (feature_mask & FEATURE_DPM_VCE_MASK)) { min_freq = data->dpm_table.eclk_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1720,7 +1728,8 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_SOCCLK].enabled) { + if (data->smu_features[GNLD_DPM_SOCCLK].enabled && + (feature_mask & FEATURE_DPM_SOCCLK_MASK)) { min_freq = data->dpm_table.soc_table.dpm_state.soft_min_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1733,14 +1742,15 @@ static int vega20_upload_dpm_min_level(struct pp_hwmgr *hwmgr) return ret; } -static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr) +static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr, uint32_t feature_mask) { struct vega20_hwmgr *data = (struct vega20_hwmgr *)(hwmgr->backend); uint32_t max_freq; int ret = 0; - if (data->smu_features[GNLD_DPM_GFXCLK].enabled) { + if (data->smu_features[GNLD_DPM_GFXCLK].enabled && + (feature_mask & FEATURE_DPM_GFXCLK_MASK)) { max_freq = data->dpm_table.gfx_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1750,7 +1760,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_UCLK].enabled) { + if (data->smu_features[GNLD_DPM_UCLK].enabled && + (feature_mask & FEATURE_DPM_UCLK_MASK)) { max_freq = data->dpm_table.mem_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1760,7 +1771,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_UVD].enabled) { + if (data->smu_features[GNLD_DPM_UVD].enabled && + (feature_mask & FEATURE_DPM_UVD_MASK)) { max_freq = data->dpm_table.vclk_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1777,7 +1789,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_VCE].enabled) { + if (data->smu_features[GNLD_DPM_VCE].enabled && + (feature_mask & FEATURE_DPM_VCE_MASK)) { max_freq = data->dpm_table.eclk_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -1787,7 +1800,8 @@ static int vega20_upload_dpm_max_level(struct pp_hwmgr *hwmgr) return ret); } - if (data->smu_features[GNLD_DPM_SOCCLK].enabled) { + if (data->smu_features[GNLD_DPM_SOCCLK].enabled && + (feature_mask & FEATURE_DPM_SOCCLK_MASK)) { max_freq = data->dpm_table.soc_table.dpm_state.soft_max_level; PP_ASSERT_WITH_CODE(!(ret = smum_send_msg_to_smc_with_parameter( @@ -2126,12 +2140,12 @@ static int vega20_force_dpm_highest(struct pp_hwmgr *hwmgr) data->dpm_table.mem_table.dpm_state.soft_max_level = data->dpm_table.mem_table.dpm_levels[soft_level].value; - ret = vega20_upload_dpm_min_level(hwmgr); + ret = vega20_upload_dpm_min_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload boot level to highest!", return ret); - ret = vega20_upload_dpm_max_level(hwmgr); + ret = vega20_upload_dpm_max_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload dpm max level to highest!", return ret); @@ -2158,12 +2172,12 @@ static int vega20_force_dpm_lowest(struct pp_hwmgr *hwmgr) data->dpm_table.mem_table.dpm_state.soft_max_level = data->dpm_table.mem_table.dpm_levels[soft_level].value; - ret = vega20_upload_dpm_min_level(hwmgr); + ret = vega20_upload_dpm_min_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload boot level to highest!", return ret); - ret = vega20_upload_dpm_max_level(hwmgr); + ret = vega20_upload_dpm_max_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload dpm max level to highest!", return ret); @@ -2176,12 +2190,12 @@ static int vega20_unforce_dpm_levels(struct pp_hwmgr *hwmgr) { int ret = 0; - ret = vega20_upload_dpm_min_level(hwmgr); + ret = vega20_upload_dpm_min_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload DPM Bootup Levels!", return ret); - ret = vega20_upload_dpm_max_level(hwmgr); + ret = vega20_upload_dpm_max_level(hwmgr, 0xFFFFFFFF); PP_ASSERT_WITH_CODE(!ret, "Failed to upload DPM Max Levels!", return ret); @@ -2239,12 +2253,12 @@ static int vega20_force_clock_level(struct pp_hwmgr *hwmgr, data->dpm_table.gfx_table.dpm_state.soft_max_level = data->dpm_table.gfx_table.dpm_levels[soft_max_level].value; - ret = vega20_upload_dpm_min_level(hwmgr); + ret = vega20_upload_dpm_min_level(hwmgr, FEATURE_DPM_GFXCLK_MASK); PP_ASSERT_WITH_CODE(!ret, "Failed to upload boot level to lowest!", return ret); - ret = vega20_upload_dpm_max_level(hwmgr); + ret = vega20_upload_dpm_max_level(hwmgr, FEATURE_DPM_GFXCLK_MASK); PP_ASSERT_WITH_CODE(!ret, "Failed to upload dpm max level to highest!", return ret); @@ -2259,12 +2273,12 @@ static int vega20_force_clock_level(struct pp_hwmgr *hwmgr, data->dpm_table.mem_table.dpm_state.soft_max_level = data->dpm_table.mem_table.dpm_levels[soft_max_level].value; - ret = vega20_upload_dpm_min_level(hwmgr); + ret = vega20_upload_dpm_min_level(hwmgr, FEATURE_DPM_UCLK_MASK); PP_ASSERT_WITH_CODE(!ret, "Failed to upload boot level to lowest!", return ret); - ret = vega20_upload_dpm_max_level(hwmgr); + ret = vega20_upload_dpm_max_level(hwmgr, FEATURE_DPM_UCLK_MASK); PP_ASSERT_WITH_CODE(!ret, "Failed to upload dpm max level to highest!", return ret); @@ -2272,6 +2286,18 @@ static int vega20_force_clock_level(struct pp_hwmgr *hwmgr, break; case PP_PCIE: + soft_min_level = mask ? (ffs(mask) - 1) : 0; + soft_max_level = mask ? (fls(mask) - 1) : 0; + if (soft_min_level >= NUM_LINK_LEVELS || + soft_max_level >= NUM_LINK_LEVELS) + return -EINVAL; + + ret = smum_send_msg_to_smc_with_parameter(hwmgr, + PPSMC_MSG_SetMinLinkDpmByIndex, soft_min_level); + PP_ASSERT_WITH_CODE(!ret, + "Failed to set min link dpm level!", + return ret); + break; default: @@ -2748,9 +2774,14 @@ static int vega20_print_clock_levels(struct pp_hwmgr *hwmgr, data->od8_settings.od8_settings_array; OverDriveTable_t *od_table = &(data->smc_state_table.overdrive_table); + struct phm_ppt_v3_information *pptable_information = + (struct phm_ppt_v3_information *)hwmgr->pptable; + PPTable_t *pptable = (PPTable_t *)pptable_information->smc_pptable; + struct amdgpu_device *adev = hwmgr->adev; struct pp_clock_levels_with_latency clocks; int i, now, size = 0; int ret = 0; + uint32_t gen_speed, lane_width; switch (type) { case PP_SCLK: @@ -2788,6 +2819,28 @@ static int vega20_print_clock_levels(struct pp_hwmgr *hwmgr, break; case PP_PCIE: + gen_speed = (RREG32_PCIE(smnPCIE_LC_SPEED_CNTL) & + PSWUSP0_PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE_MASK) + >> PSWUSP0_PCIE_LC_SPEED_CNTL__LC_CURRENT_DATA_RATE__SHIFT; + lane_width = (RREG32_PCIE(smnPCIE_LC_LINK_WIDTH_CNTL) & + PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD_MASK) + >> PCIE_LC_LINK_WIDTH_CNTL__LC_LINK_WIDTH_RD__SHIFT; + for (i = 0; i < NUM_LINK_LEVELS; i++) + size += sprintf(buf + size, "%d: %s %s %dMhz %s\n", i, + (pptable->PcieGenSpeed[i] == 0) ? "2.5GT/s," : + (pptable->PcieGenSpeed[i] == 1) ? "5.0GT/s," : + (pptable->PcieGenSpeed[i] == 2) ? "8.0GT/s," : + (pptable->PcieGenSpeed[i] == 3) ? "16.0GT/s," : "", + (pptable->PcieLaneCount[i] == 1) ? "x1" : + (pptable->PcieLaneCount[i] == 2) ? "x2" : + (pptable->PcieLaneCount[i] == 3) ? "x4" : + (pptable->PcieLaneCount[i] == 4) ? "x8" : + (pptable->PcieLaneCount[i] == 5) ? "x12" : + (pptable->PcieLaneCount[i] == 6) ? "x16" : "", + pptable->LclkFreq[i], + (gen_speed == pptable->PcieGenSpeed[i]) && + (lane_width == pptable->PcieLaneCount[i]) ? + "*" : ""); break; case OD_SCLK: diff --git a/drivers/gpu/drm/amd/powerplay/inc/hardwaremanager.h b/drivers/gpu/drm/amd/powerplay/inc/hardwaremanager.h index 54fd0125d9cf..f4dab979a3a1 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/hardwaremanager.h +++ b/drivers/gpu/drm/amd/powerplay/inc/hardwaremanager.h @@ -463,5 +463,8 @@ extern int phm_display_clock_voltage_request(struct pp_hwmgr *hwmgr, extern int phm_get_max_high_clocks(struct pp_hwmgr *hwmgr, struct amd_pp_simple_clock_info *clocks); extern int phm_disable_smc_firmware_ctf(struct pp_hwmgr *hwmgr); + +extern int phm_set_active_display_count(struct pp_hwmgr *hwmgr, uint32_t count); + #endif /* _HARDWARE_MANAGER_H_ */ diff --git a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h index 07d180ce4d18..0d298a0409f5 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h +++ b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h @@ -309,7 +309,7 @@ struct pp_hwmgr_func { int (*avfs_control)(struct pp_hwmgr *hwmgr, bool enable); int (*disable_smc_firmware_ctf)(struct pp_hwmgr *hwmgr); int (*set_active_display_count)(struct pp_hwmgr *hwmgr, uint32_t count); - int (*set_deep_sleep_dcefclk)(struct pp_hwmgr *hwmgr, uint32_t clock); + int (*set_min_deep_sleep_dcefclk)(struct pp_hwmgr *hwmgr, uint32_t clock); int (*start_thermal_controller)(struct pp_hwmgr *hwmgr, struct PP_TemperatureRange *range); int (*notify_cac_buffer_info)(struct pp_hwmgr *hwmgr, uint32_t virtual_addr_low, @@ -317,6 +317,9 @@ struct pp_hwmgr_func { uint32_t mc_addr_low, uint32_t mc_addr_hi, uint32_t size); + int (*update_nbdpm_pstate)(struct pp_hwmgr *hwmgr, + bool enable, + bool lock); int (*get_thermal_temperature_range)(struct pp_hwmgr *hwmgr, struct PP_TemperatureRange *range); int (*get_power_profile_mode)(struct pp_hwmgr *hwmgr, char *buf); @@ -329,6 +332,8 @@ struct pp_hwmgr_func { int (*smus_notify_pwe)(struct pp_hwmgr *hwmgr); int (*powergate_sdma)(struct pp_hwmgr *hwmgr, bool bgate); int (*enable_mgpu_fan_boost)(struct pp_hwmgr *hwmgr); + int (*set_hard_min_dcefclk_by_freq)(struct pp_hwmgr *hwmgr, uint32_t clock); + int (*set_hard_min_fclk_by_freq)(struct pp_hwmgr *hwmgr, uint32_t clock); }; struct pp_table_func { diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu7_ppsmc.h b/drivers/gpu/drm/amd/powerplay/inc/smu7_ppsmc.h index 62f36ba2435b..6e19f4c7cf8f 100644 --- a/drivers/gpu/drm/amd/powerplay/inc/smu7_ppsmc.h +++ b/drivers/gpu/drm/amd/powerplay/inc/smu7_ppsmc.h @@ -386,6 +386,8 @@ typedef uint16_t PPSMC_Result; #define PPSMC_MSG_AgmResetPsm ((uint16_t) 0x403) #define PPSMC_MSG_ReadVftCell ((uint16_t) 0x404) +#define PPSMC_MSG_ApplyAvfsCksOffVoltage ((uint16_t) 0x415) + #define PPSMC_MSG_GFX_CU_PG_ENABLE ((uint16_t) 0x280) #define PPSMC_MSG_GFX_CU_PG_DISABLE ((uint16_t) 0x281) #define PPSMC_MSG_GetCurrPkgPwr ((uint16_t) 0x282) @@ -395,6 +397,9 @@ typedef uint16_t PPSMC_Result; #define PPSMC_MSG_SetVBITimeout ((uint16_t) 0x306) +#define PPSMC_MSG_EnableFFC ((uint16_t) 0x307) +#define PPSMC_MSG_DisableFFC ((uint16_t) 0x308) + #define PPSMC_MSG_EnableDpmDidt ((uint16_t) 0x309) #define PPSMC_MSG_DisableDpmDidt ((uint16_t) 0x30A) diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c index 2b2c26616902..52abca065764 100644 --- a/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c +++ b/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c @@ -1528,8 +1528,21 @@ static int polaris10_populate_clock_stretcher_data_table(struct pp_hwmgr *hwmgr) efuse = efuse >> 24; if (hwmgr->chip_id == CHIP_POLARIS10) { - min = 1000; - max = 2300; + if (hwmgr->is_kicker) { + min = 1200; + max = 2500; + } else { + min = 1000; + max = 2300; + } + } else if (hwmgr->chip_id == CHIP_POLARIS11) { + if (hwmgr->is_kicker) { + min = 900; + max = 2100; + } else { + min = 1100; + max = 2100; + } } else { min = 1100; max = 2100; @@ -1626,6 +1639,7 @@ static int polaris10_populate_avfs_parameters(struct pp_hwmgr *hwmgr) { struct smu7_hwmgr *data = (struct smu7_hwmgr *)(hwmgr->backend); struct polaris10_smumgr *smu_data = (struct polaris10_smumgr *)(hwmgr->smu_backend); + struct amdgpu_device *adev = hwmgr->adev; SMU74_Discrete_DpmTable *table = &(smu_data->smc_state_table); int result = 0; @@ -1646,6 +1660,59 @@ static int polaris10_populate_avfs_parameters(struct pp_hwmgr *hwmgr) result = atomctrl_get_avfs_information(hwmgr, &avfs_params); if (0 == result) { + if (((adev->pdev->device == 0x67ef) && + ((adev->pdev->revision == 0xe0) || + (adev->pdev->revision == 0xe5))) || + ((adev->pdev->device == 0x67ff) && + ((adev->pdev->revision == 0xcf) || + (adev->pdev->revision == 0xef) || + (adev->pdev->revision == 0xff)))) { + avfs_params.ucEnableApplyAVFS_CKS_OFF_Voltage = 1; + if ((adev->pdev->device == 0x67ef && adev->pdev->revision == 0xe5) || + (adev->pdev->device == 0x67ff && adev->pdev->revision == 0xef)) { + if ((avfs_params.ulGB_VDROOP_TABLE_CKSOFF_a0 == 0xEA522DD3) && + (avfs_params.ulGB_VDROOP_TABLE_CKSOFF_a1 == 0x5645A) && + (avfs_params.ulGB_VDROOP_TABLE_CKSOFF_a2 == 0x33F9E) && + (avfs_params.ulAVFSGB_FUSE_TABLE_CKSOFF_m1 == 0xFFFFC5CC) && + (avfs_params.usAVFSGB_FUSE_TABLE_CKSOFF_m2 == 0x1B1A) && + (avfs_params.ulAVFSGB_FUSE_TABLE_CKSOFF_b == 0xFFFFFCED)) { + avfs_params.ulGB_VDROOP_TABLE_CKSOFF_a0 = 0xF718F1D4; + avfs_params.ulGB_VDROOP_TABLE_CKSOFF_a1 = 0x323FD; + avfs_params.ulGB_VDROOP_TABLE_CKSOFF_a2 = 0x1E455; + avfs_params.ulAVFSGB_FUSE_TABLE_CKSOFF_m1 = 0; + avfs_params.usAVFSGB_FUSE_TABLE_CKSOFF_m2 = 0; + avfs_params.ulAVFSGB_FUSE_TABLE_CKSOFF_b = 0x23; + } + } + } else if (hwmgr->chip_id == CHIP_POLARIS12 && !hwmgr->is_kicker) { + avfs_params.ucEnableApplyAVFS_CKS_OFF_Voltage = 1; + avfs_params.ulGB_VDROOP_TABLE_CKSOFF_a0 = 0xF6B024DD; + avfs_params.ulGB_VDROOP_TABLE_CKSOFF_a1 = 0x3005E; + avfs_params.ulGB_VDROOP_TABLE_CKSOFF_a2 = 0x18A5F; + avfs_params.ulAVFSGB_FUSE_TABLE_CKSOFF_m1 = 0x315; + avfs_params.usAVFSGB_FUSE_TABLE_CKSOFF_m2 = 0xFED1; + avfs_params.ulAVFSGB_FUSE_TABLE_CKSOFF_b = 0x3B; + } else if (((adev->pdev->device == 0x67df) && + ((adev->pdev->revision == 0xe0) || + (adev->pdev->revision == 0xe3) || + (adev->pdev->revision == 0xe4) || + (adev->pdev->revision == 0xe5) || + (adev->pdev->revision == 0xe7) || + (adev->pdev->revision == 0xef))) || + ((adev->pdev->device == 0x6fdf) && + ((adev->pdev->revision == 0xef) || + (adev->pdev->revision == 0xff)))) { + avfs_params.ucEnableApplyAVFS_CKS_OFF_Voltage = 1; + avfs_params.ulGB_VDROOP_TABLE_CKSOFF_a0 = 0xF843B66B; + avfs_params.ulGB_VDROOP_TABLE_CKSOFF_a1 = 0x59CB5; + avfs_params.ulGB_VDROOP_TABLE_CKSOFF_a2 = 0xFFFF287F; + avfs_params.ulAVFSGB_FUSE_TABLE_CKSOFF_m1 = 0; + avfs_params.usAVFSGB_FUSE_TABLE_CKSOFF_m2 = 0xFF23; + avfs_params.ulAVFSGB_FUSE_TABLE_CKSOFF_b = 0x58; + } + } + + if (0 == result) { table->BTCGB_VDROOP_TABLE[0].a0 = PP_HOST_TO_SMC_UL(avfs_params.ulGB_VDROOP_TABLE_CKSON_a0); table->BTCGB_VDROOP_TABLE[0].a1 = PP_HOST_TO_SMC_UL(avfs_params.ulGB_VDROOP_TABLE_CKSON_a1); table->BTCGB_VDROOP_TABLE[0].a2 = PP_HOST_TO_SMC_UL(avfs_params.ulGB_VDROOP_TABLE_CKSON_a2); @@ -1984,6 +2051,12 @@ int polaris10_thermal_avfs_enable(struct pp_hwmgr *hwmgr) smum_send_msg_to_smc(hwmgr, PPSMC_MSG_EnableAvfs); + /* Apply avfs cks-off voltages to avoid the overshoot + * when switching to the highest sclk frequency + */ + if (data->apply_avfs_cks_off_voltage) + smum_send_msg_to_smc(hwmgr, PPSMC_MSG_ApplyAvfsCksOffVoltage); + return 0; } diff --git a/drivers/gpu/drm/armada/armada_gem.c b/drivers/gpu/drm/armada/armada_gem.c index 892c1d9304bb..642d0e70d0f8 100644 --- a/drivers/gpu/drm/armada/armada_gem.c +++ b/drivers/gpu/drm/armada/armada_gem.c @@ -334,7 +334,7 @@ int armada_gem_pwrite_ioctl(struct drm_device *dev, void *data, ptr = (char __user *)(uintptr_t)args->ptr; - if (!access_ok(VERIFY_READ, ptr, args->size)) + if (!access_ok(ptr, args->size)) return -EFAULT; ret = fault_in_pages_readable(ptr, args->size); diff --git a/drivers/gpu/drm/ast/ast_fb.c b/drivers/gpu/drm/ast/ast_fb.c index a80bca1a857f..c2e41369adcf 100644 --- a/drivers/gpu/drm/ast/ast_fb.c +++ b/drivers/gpu/drm/ast/ast_fb.c @@ -261,6 +261,7 @@ static void ast_fbdev_destroy(struct drm_device *dev, { struct ast_framebuffer *afb = &afbdev->afb; + drm_crtc_force_disable_all(dev); drm_fb_helper_unregister_fbi(&afbdev->helper); if (afb->obj) { diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c index dac355812adc..373700c05a00 100644 --- a/drivers/gpu/drm/ast/ast_main.c +++ b/drivers/gpu/drm/ast/ast_main.c @@ -583,7 +583,8 @@ void ast_driver_unload(struct drm_device *dev) drm_mode_config_cleanup(dev); ast_mm_fini(ast); - pci_iounmap(dev->pdev, ast->ioregs); + if (ast->ioregs != ast->regs + AST_IO_MM_OFFSET) + pci_iounmap(dev->pdev, ast->ioregs); pci_iounmap(dev->pdev, ast->regs); kfree(ast); } diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index 7c6ac3cadb6b..8bb355d5d43d 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -973,9 +973,21 @@ static int get_clock(void *i2c_priv) { struct ast_i2c_chan *i2c = i2c_priv; struct ast_private *ast = i2c->dev->dev_private; - uint32_t val; + uint32_t val, val2, count, pass; + + count = 0; + pass = 0; + val = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x10) >> 4) & 0x01; + do { + val2 = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x10) >> 4) & 0x01; + if (val == val2) { + pass++; + } else { + pass = 0; + val = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x10) >> 4) & 0x01; + } + } while ((pass < 5) && (count++ < 0x10000)); - val = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x10) >> 4; return val & 1 ? 1 : 0; } @@ -983,9 +995,21 @@ static int get_data(void *i2c_priv) { struct ast_i2c_chan *i2c = i2c_priv; struct ast_private *ast = i2c->dev->dev_private; - uint32_t val; + uint32_t val, val2, count, pass; + + count = 0; + pass = 0; + val = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x20) >> 5) & 0x01; + do { + val2 = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x20) >> 5) & 0x01; + if (val == val2) { + pass++; + } else { + pass = 0; + val = (ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x20) >> 5) & 0x01; + } + } while ((pass < 5) && (count++ < 0x10000)); - val = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x20) >> 5; return val & 1 ? 1 : 0; } @@ -998,7 +1022,7 @@ static void set_clock(void *i2c_priv, int clock) for (i = 0; i < 0x10000; i++) { ujcrb7 = ((clock & 0x01) ? 0 : 1); - ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0xfe, ujcrb7); + ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0xf4, ujcrb7); jtemp = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x01); if (ujcrb7 == jtemp) break; @@ -1014,7 +1038,7 @@ static void set_data(void *i2c_priv, int data) for (i = 0; i < 0x10000; i++) { ujcrb7 = ((data & 0x01) ? 0 : 1) << 2; - ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0xfb, ujcrb7); + ast_set_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0xf1, ujcrb7); jtemp = ast_get_index_reg_mask(ast, AST_IO_CRTC_PORT, 0xb7, 0x04); if (ujcrb7 == jtemp) break; diff --git a/drivers/gpu/drm/bridge/ti-sn65dsi86.c b/drivers/gpu/drm/bridge/ti-sn65dsi86.c index 680566d97adc..10243965ee7c 100644 --- a/drivers/gpu/drm/bridge/ti-sn65dsi86.c +++ b/drivers/gpu/drm/bridge/ti-sn65dsi86.c @@ -54,7 +54,7 @@ #define SN_AUX_ADDR_7_0_REG 0x76 #define SN_AUX_LENGTH_REG 0x77 #define SN_AUX_CMD_REG 0x78 -#define AUX_CMD_SEND BIT(1) +#define AUX_CMD_SEND BIT(0) #define AUX_CMD_REQ(x) ((x) << 4) #define SN_AUX_RDATA_REG(x) (0x79 + (x)) #define SN_SSC_CONFIG_REG 0x93 diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c index 008224f376fe..5eb40130fafb 100644 --- a/drivers/gpu/drm/drm_atomic.c +++ b/drivers/gpu/drm/drm_atomic.c @@ -531,6 +531,8 @@ static int drm_atomic_plane_check(const struct drm_plane_state *old_plane_state, struct drm_crtc *crtc = new_plane_state->crtc; const struct drm_framebuffer *fb = new_plane_state->fb; unsigned int fb_width, fb_height; + struct drm_mode_rect *clips; + uint32_t num_clips; int ret; /* either *both* CRTC and FB must be set, or neither */ @@ -604,6 +606,26 @@ static int drm_atomic_plane_check(const struct drm_plane_state *old_plane_state, return -ENOSPC; } + clips = drm_plane_get_damage_clips(new_plane_state); + num_clips = drm_plane_get_damage_clips_count(new_plane_state); + + /* Make sure damage clips are valid and inside the fb. */ + while (num_clips > 0) { + if (clips->x1 >= clips->x2 || + clips->y1 >= clips->y2 || + clips->x1 < 0 || + clips->y1 < 0 || + clips->x2 > fb_width || + clips->y2 > fb_height) { + DRM_DEBUG_ATOMIC("[PLANE:%d:%s] invalid damage clip %d %d %d %d\n", + plane->base.id, plane->name, clips->x1, + clips->y1, clips->x2, clips->y2); + return -EINVAL; + } + clips++; + num_clips--; + } + if (plane_switching_crtc(old_plane_state, new_plane_state)) { DRM_DEBUG_ATOMIC("[PLANE:%d:%s] switching CRTC directly\n", plane->base.id, plane->name); diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c index 8150fa8387d5..54e2ae614dcc 100644 --- a/drivers/gpu/drm/drm_atomic_helper.c +++ b/drivers/gpu/drm/drm_atomic_helper.c @@ -32,6 +32,7 @@ #include <drm/drm_crtc_helper.h> #include <drm/drm_atomic_helper.h> #include <drm/drm_writeback.h> +#include <drm/drm_damage_helper.h> #include <linux/dma-fence.h> #include "drm_crtc_helper_internal.h" @@ -862,6 +863,8 @@ drm_atomic_helper_check_planes(struct drm_device *dev, drm_atomic_helper_plane_changed(state, old_plane_state, new_plane_state, plane); + drm_atomic_helper_check_plane_damage(state, new_plane_state); + if (!funcs || !funcs->atomic_check) continue; diff --git a/drivers/gpu/drm/drm_atomic_state_helper.c b/drivers/gpu/drm/drm_atomic_state_helper.c index 60bd7d708e35..4985384e51f6 100644 --- a/drivers/gpu/drm/drm_atomic_state_helper.c +++ b/drivers/gpu/drm/drm_atomic_state_helper.c @@ -241,6 +241,7 @@ void __drm_atomic_helper_plane_duplicate_state(struct drm_plane *plane, state->fence = NULL; state->commit = NULL; + state->fb_damage_clips = NULL; } EXPORT_SYMBOL(__drm_atomic_helper_plane_duplicate_state); @@ -285,6 +286,8 @@ void __drm_atomic_helper_plane_destroy_state(struct drm_plane_state *state) if (state->commit) drm_crtc_commit_put(state->commit); + + drm_property_blob_put(state->fb_damage_clips); } EXPORT_SYMBOL(__drm_atomic_helper_plane_destroy_state); diff --git a/drivers/gpu/drm/drm_atomic_uapi.c b/drivers/gpu/drm/drm_atomic_uapi.c index 86ac33922b09..c40889888a16 100644 --- a/drivers/gpu/drm/drm_atomic_uapi.c +++ b/drivers/gpu/drm/drm_atomic_uapi.c @@ -433,6 +433,8 @@ static int drm_atomic_crtc_set_property(struct drm_crtc *crtc, ret = drm_atomic_set_mode_prop_for_crtc(state, mode); drm_property_blob_put(mode); return ret; + } else if (property == config->prop_vrr_enabled) { + state->vrr_enabled = val; } else if (property == config->degamma_lut_property) { ret = drm_atomic_replace_property_blob_from_id(dev, &state->degamma_lut, @@ -491,6 +493,8 @@ drm_atomic_crtc_get_property(struct drm_crtc *crtc, *val = state->active; else if (property == config->prop_mode_id) *val = (state->mode_blob) ? state->mode_blob->base.id : 0; + else if (property == config->prop_vrr_enabled) + *val = state->vrr_enabled; else if (property == config->degamma_lut_property) *val = (state->degamma_lut) ? state->degamma_lut->base.id : 0; else if (property == config->ctm_property) @@ -513,6 +517,8 @@ static int drm_atomic_plane_set_property(struct drm_plane *plane, { struct drm_device *dev = plane->dev; struct drm_mode_config *config = &dev->mode_config; + bool replaced = false; + int ret; if (property == config->prop_fb_id) { struct drm_framebuffer *fb = drm_framebuffer_lookup(dev, NULL, val); @@ -566,6 +572,14 @@ static int drm_atomic_plane_set_property(struct drm_plane *plane, state->color_encoding = val; } else if (property == plane->color_range_property) { state->color_range = val; + } else if (property == config->prop_fb_damage_clips) { + ret = drm_atomic_replace_property_blob_from_id(dev, + &state->fb_damage_clips, + val, + -1, + sizeof(struct drm_rect), + &replaced); + return ret; } else if (plane->funcs->atomic_set_property) { return plane->funcs->atomic_set_property(plane, state, property, val); @@ -621,6 +635,9 @@ drm_atomic_plane_get_property(struct drm_plane *plane, *val = state->color_encoding; } else if (property == plane->color_range_property) { *val = state->color_range; + } else if (property == config->prop_fb_damage_clips) { + *val = (state->fb_damage_clips) ? + state->fb_damage_clips->base.id : 0; } else if (plane->funcs->atomic_get_property) { return plane->funcs->atomic_get_property(plane, state, property, val); } else { diff --git a/drivers/gpu/drm/drm_auth.c b/drivers/gpu/drm/drm_auth.c index d9c0f7573905..1669c42c40ed 100644 --- a/drivers/gpu/drm/drm_auth.c +++ b/drivers/gpu/drm/drm_auth.c @@ -142,6 +142,7 @@ static int drm_new_set_master(struct drm_device *dev, struct drm_file *fpriv) lockdep_assert_held_once(&dev->master_mutex); + WARN_ON(fpriv->is_master); old_master = fpriv->master; fpriv->master = drm_master_create(dev); if (!fpriv->master) { @@ -170,6 +171,7 @@ out_err: /* drop references and restore old master on failure */ drm_master_put(&fpriv->master); fpriv->master = old_master; + fpriv->is_master = 0; return ret; } diff --git a/drivers/gpu/drm/drm_connector.c b/drivers/gpu/drm/drm_connector.c index 66b2fd20369a..847539645558 100644 --- a/drivers/gpu/drm/drm_connector.c +++ b/drivers/gpu/drm/drm_connector.c @@ -1326,6 +1326,105 @@ int drm_mode_create_scaling_mode_property(struct drm_device *dev) EXPORT_SYMBOL(drm_mode_create_scaling_mode_property); /** + * DOC: Variable refresh properties + * + * Variable refresh rate capable displays can dynamically adjust their + * refresh rate by extending the duration of their vertical front porch + * until page flip or timeout occurs. This can reduce or remove stuttering + * and latency in scenarios where the page flip does not align with the + * vblank interval. + * + * An example scenario would be an application flipping at a constant rate + * of 48Hz on a 60Hz display. The page flip will frequently miss the vblank + * interval and the same contents will be displayed twice. This can be + * observed as stuttering for content with motion. + * + * If variable refresh rate was active on a display that supported a + * variable refresh range from 35Hz to 60Hz no stuttering would be observable + * for the example scenario. The minimum supported variable refresh rate of + * 35Hz is below the page flip frequency and the vertical front porch can + * be extended until the page flip occurs. The vblank interval will be + * directly aligned to the page flip rate. + * + * Not all userspace content is suitable for use with variable refresh rate. + * Large and frequent changes in vertical front porch duration may worsen + * perceived stuttering for input sensitive applications. + * + * Panel brightness will also vary with vertical front porch duration. Some + * panels may have noticeable differences in brightness between the minimum + * vertical front porch duration and the maximum vertical front porch duration. + * Large and frequent changes in vertical front porch duration may produce + * observable flickering for such panels. + * + * Userspace control for variable refresh rate is supported via properties + * on the &drm_connector and &drm_crtc objects. + * + * "vrr_capable": + * Optional &drm_connector boolean property that drivers should attach + * with drm_connector_attach_vrr_capable_property() on connectors that + * could support variable refresh rates. Drivers should update the + * property value by calling drm_connector_set_vrr_capable_property(). + * + * Absence of the property should indicate absence of support. + * + * "vrr_enabled": + * Default &drm_crtc boolean property that notifies the driver that the + * content on the CRTC is suitable for variable refresh rate presentation. + * The driver will take this property as a hint to enable variable + * refresh rate support if the receiver supports it, ie. if the + * "vrr_capable" property is true on the &drm_connector object. The + * vertical front porch duration will be extended until page-flip or + * timeout when enabled. + * + * The minimum vertical front porch duration is defined as the vertical + * front porch duration for the current mode. + * + * The maximum vertical front porch duration is greater than or equal to + * the minimum vertical front porch duration. The duration is derived + * from the minimum supported variable refresh rate for the connector. + * + * The driver may place further restrictions within these minimum + * and maximum bounds. + * + * The semantics for the vertical blank timestamp differ when + * variable refresh rate is active. The vertical blank timestamp + * is defined to be an estimate using the current mode's fixed + * refresh rate timings. The semantics for the page-flip event + * timestamp remain the same. + */ + +/** + * drm_connector_attach_vrr_capable_property - creates the + * vrr_capable property + * @connector: connector to create the vrr_capable property on. + * + * This is used by atomic drivers to add support for querying + * variable refresh rate capability for a connector. + * + * Returns: + * Zero on success, negative errono on failure. + */ +int drm_connector_attach_vrr_capable_property( + struct drm_connector *connector) +{ + struct drm_device *dev = connector->dev; + struct drm_property *prop; + + if (!connector->vrr_capable_property) { + prop = drm_property_create_bool(dev, DRM_MODE_PROP_IMMUTABLE, + "vrr_capable"); + if (!prop) + return -ENOMEM; + + connector->vrr_capable_property = prop; + drm_object_attach_property(&connector->base, prop, 0); + } + + return 0; +} +EXPORT_SYMBOL(drm_connector_attach_vrr_capable_property); + +/** * drm_connector_attach_scaling_mode_property - attach atomic scaling mode property * @connector: connector to attach scaling mode property on. * @scaling_mode_mask: or'ed mask of BIT(%DRM_MODE_SCALE_\*). @@ -1688,6 +1787,24 @@ int drm_connector_attach_max_bpc_property(struct drm_connector *connector, EXPORT_SYMBOL(drm_connector_attach_max_bpc_property); /** + * drm_connector_set_vrr_capable_property - sets the variable refresh rate + * capable property for a connector + * @connector: drm connector + * @capable: True if the connector is variable refresh rate capable + * + * Should be used by atomic drivers to update the indicated support for + * variable refresh rate over a connector. + */ +void drm_connector_set_vrr_capable_property( + struct drm_connector *connector, bool capable) +{ + drm_object_property_set_value(&connector->base, + connector->vrr_capable_property, + capable); +} +EXPORT_SYMBOL(drm_connector_set_vrr_capable_property); + +/** * drm_connector_init_panel_orientation_property - * initialize the connecters panel_orientation property * @connector: connector for which to init the panel-orientation property. diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c index e224b9b7d17a..f660819d406e 100644 --- a/drivers/gpu/drm/drm_crtc.c +++ b/drivers/gpu/drm/drm_crtc.c @@ -330,6 +330,8 @@ int drm_crtc_init_with_planes(struct drm_device *dev, struct drm_crtc *crtc, drm_object_attach_property(&crtc->base, config->prop_mode_id, 0); drm_object_attach_property(&crtc->base, config->prop_out_fence_ptr, 0); + drm_object_attach_property(&crtc->base, + config->prop_vrr_enabled, 0); } return 0; diff --git a/drivers/gpu/drm/drm_damage_helper.c b/drivers/gpu/drm/drm_damage_helper.c new file mode 100644 index 000000000000..31032407254d --- /dev/null +++ b/drivers/gpu/drm/drm_damage_helper.c @@ -0,0 +1,335 @@ +// SPDX-License-Identifier: GPL-2.0 OR MIT +/************************************************************************** + * + * Copyright (c) 2018 VMware, Inc., Palo Alto, CA., USA + * All Rights Reserved. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the + * "Software"), to deal in the Software without restriction, including + * without limitation the rights to use, copy, modify, merge, publish, + * distribute, sub license, and/or sell copies of the Software, and to + * permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice (including the + * next paragraph) shall be included in all copies or substantial portions + * of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM, + * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR + * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE + * USE OR OTHER DEALINGS IN THE SOFTWARE. + * + * Authors: + * Deepak Rawat <drawat@vmware.com> + * Rob Clark <robdclark@gmail.com> + * + **************************************************************************/ + +#include <drm/drm_atomic.h> +#include <drm/drm_damage_helper.h> + +/** + * DOC: overview + * + * FB_DAMAGE_CLIPS is an optional plane property which provides a means to + * specify a list of damage rectangles on a plane in framebuffer coordinates of + * the framebuffer attached to the plane. In current context damage is the area + * of plane framebuffer that has changed since last plane update (also called + * page-flip), irrespective of whether currently attached framebuffer is same as + * framebuffer attached during last plane update or not. + * + * FB_DAMAGE_CLIPS is a hint to kernel which could be helpful for some drivers + * to optimize internally especially for virtual devices where each framebuffer + * change needs to be transmitted over network, usb, etc. + * + * Since FB_DAMAGE_CLIPS is a hint so it is an optional property. User-space can + * ignore damage clips property and in that case driver will do a full plane + * update. In case damage clips are provided then it is guaranteed that the area + * inside damage clips will be updated to plane. For efficiency driver can do + * full update or can update more than specified in damage clips. Since driver + * is free to read more, user-space must always render the entire visible + * framebuffer. Otherwise there can be corruptions. Also, if a user-space + * provides damage clips which doesn't encompass the actual damage to + * framebuffer (since last plane update) can result in incorrect rendering. + * + * FB_DAMAGE_CLIPS is a blob property with the layout of blob data is simply an + * array of &drm_mode_rect. Unlike plane &drm_plane_state.src coordinates, + * damage clips are not in 16.16 fixed point. Similar to plane src in + * framebuffer, damage clips cannot be negative. In damage clip, x1/y1 are + * inclusive and x2/y2 are exclusive. While kernel does not error for overlapped + * damage clips, it is strongly discouraged. + * + * Drivers that are interested in damage interface for plane should enable + * FB_DAMAGE_CLIPS property by calling drm_plane_enable_fb_damage_clips(). + * Drivers implementing damage can use drm_atomic_helper_damage_iter_init() and + * drm_atomic_helper_damage_iter_next() helper iterator function to get damage + * rectangles clipped to &drm_plane_state.src. + */ + +static void convert_clip_rect_to_rect(const struct drm_clip_rect *src, + struct drm_mode_rect *dest, + uint32_t num_clips, uint32_t src_inc) +{ + while (num_clips > 0) { + dest->x1 = src->x1; + dest->y1 = src->y1; + dest->x2 = src->x2; + dest->y2 = src->y2; + src += src_inc; + dest++; + num_clips--; + } +} + +/** + * drm_plane_enable_fb_damage_clips - Enables plane fb damage clips property. + * @plane: Plane on which to enable damage clips property. + * + * This function lets driver to enable the damage clips property on a plane. + */ +void drm_plane_enable_fb_damage_clips(struct drm_plane *plane) +{ + struct drm_device *dev = plane->dev; + struct drm_mode_config *config = &dev->mode_config; + + drm_object_attach_property(&plane->base, config->prop_fb_damage_clips, + 0); +} +EXPORT_SYMBOL(drm_plane_enable_fb_damage_clips); + +/** + * drm_atomic_helper_check_plane_damage - Verify plane damage on atomic_check. + * @state: The driver state object. + * @plane_state: Plane state for which to verify damage. + * + * This helper function makes sure that damage from plane state is discarded + * for full modeset. If there are more reasons a driver would want to do a full + * plane update rather than processing individual damage regions, then those + * cases should be taken care of here. + * + * Note that &drm_plane_state.fb_damage_clips == NULL in plane state means that + * full plane update should happen. It also ensure helper iterator will return + * &drm_plane_state.src as damage. + */ +void drm_atomic_helper_check_plane_damage(struct drm_atomic_state *state, + struct drm_plane_state *plane_state) +{ + struct drm_crtc_state *crtc_state; + + if (plane_state->crtc) { + crtc_state = drm_atomic_get_new_crtc_state(state, + plane_state->crtc); + + if (WARN_ON(!crtc_state)) + return; + + if (drm_atomic_crtc_needs_modeset(crtc_state)) { + drm_property_blob_put(plane_state->fb_damage_clips); + plane_state->fb_damage_clips = NULL; + } + } +} +EXPORT_SYMBOL(drm_atomic_helper_check_plane_damage); + +/** + * drm_atomic_helper_dirtyfb - Helper for dirtyfb. + * @fb: DRM framebuffer. + * @file_priv: Drm file for the ioctl call. + * @flags: Dirty fb annotate flags. + * @color: Color for annotate fill. + * @clips: Dirty region. + * @num_clips: Count of clip in clips. + * + * A helper to implement &drm_framebuffer_funcs.dirty using damage interface + * during plane update. If num_clips is 0 then this helper will do a full plane + * update. This is the same behaviour expected by DIRTFB IOCTL. + * + * Note that this helper is blocking implementation. This is what current + * drivers and userspace expect in their DIRTYFB IOCTL implementation, as a way + * to rate-limit userspace and make sure its rendering doesn't get ahead of + * uploading new data too much. + * + * Return: Zero on success, negative errno on failure. + */ +int drm_atomic_helper_dirtyfb(struct drm_framebuffer *fb, + struct drm_file *file_priv, unsigned int flags, + unsigned int color, struct drm_clip_rect *clips, + unsigned int num_clips) +{ + struct drm_modeset_acquire_ctx ctx; + struct drm_property_blob *damage = NULL; + struct drm_mode_rect *rects = NULL; + struct drm_atomic_state *state; + struct drm_plane *plane; + int ret = 0; + + /* + * When called from ioctl, we are interruptable, but not when called + * internally (ie. defio worker) + */ + drm_modeset_acquire_init(&ctx, + file_priv ? DRM_MODESET_ACQUIRE_INTERRUPTIBLE : 0); + + state = drm_atomic_state_alloc(fb->dev); + if (!state) { + ret = -ENOMEM; + goto out_drop_locks; + } + state->acquire_ctx = &ctx; + + if (clips) { + uint32_t inc = 1; + + if (flags & DRM_MODE_FB_DIRTY_ANNOTATE_COPY) { + inc = 2; + num_clips /= 2; + } + + rects = kcalloc(num_clips, sizeof(*rects), GFP_KERNEL); + if (!rects) { + ret = -ENOMEM; + goto out; + } + + convert_clip_rect_to_rect(clips, rects, num_clips, inc); + damage = drm_property_create_blob(fb->dev, + num_clips * sizeof(*rects), + rects); + if (IS_ERR(damage)) { + ret = PTR_ERR(damage); + damage = NULL; + goto out; + } + } + +retry: + drm_for_each_plane(plane, fb->dev) { + struct drm_plane_state *plane_state; + + if (plane->state->fb != fb) + continue; + + plane_state = drm_atomic_get_plane_state(state, plane); + if (IS_ERR(plane_state)) { + ret = PTR_ERR(plane_state); + goto out; + } + + drm_property_replace_blob(&plane_state->fb_damage_clips, + damage); + } + + ret = drm_atomic_commit(state); + +out: + if (ret == -EDEADLK) { + drm_atomic_state_clear(state); + ret = drm_modeset_backoff(&ctx); + if (!ret) + goto retry; + } + + drm_property_blob_put(damage); + kfree(rects); + drm_atomic_state_put(state); + +out_drop_locks: + drm_modeset_drop_locks(&ctx); + drm_modeset_acquire_fini(&ctx); + + return ret; + +} +EXPORT_SYMBOL(drm_atomic_helper_dirtyfb); + +/** + * drm_atomic_helper_damage_iter_init - Initialize the damage iterator. + * @iter: The iterator to initialize. + * @old_state: Old plane state for validation. + * @state: Plane state from which to iterate the damage clips. + * + * Initialize an iterator, which clips plane damage + * &drm_plane_state.fb_damage_clips to plane &drm_plane_state.src. This iterator + * returns full plane src in case damage is not present because either + * user-space didn't sent or driver discarded it (it want to do full plane + * update). Currently this iterator returns full plane src in case plane src + * changed but that can be changed in future to return damage. + * + * For the case when plane is not visible or plane update should not happen the + * first call to iter_next will return false. Note that this helper use clipped + * &drm_plane_state.src, so driver calling this helper should have called + * drm_atomic_helper_check_plane_state() earlier. + */ +void +drm_atomic_helper_damage_iter_init(struct drm_atomic_helper_damage_iter *iter, + const struct drm_plane_state *old_state, + const struct drm_plane_state *state) +{ + memset(iter, 0, sizeof(*iter)); + + if (!state || !state->crtc || !state->fb || !state->visible) + return; + + iter->clips = drm_helper_get_plane_damage_clips(state); + iter->num_clips = drm_plane_get_damage_clips_count(state); + + /* Round down for x1/y1 and round up for x2/y2 to catch all pixels */ + iter->plane_src.x1 = state->src.x1 >> 16; + iter->plane_src.y1 = state->src.y1 >> 16; + iter->plane_src.x2 = (state->src.x2 >> 16) + !!(state->src.x2 & 0xFFFF); + iter->plane_src.y2 = (state->src.y2 >> 16) + !!(state->src.y2 & 0xFFFF); + + if (!iter->clips || !drm_rect_equals(&state->src, &old_state->src)) { + iter->clips = 0; + iter->num_clips = 0; + iter->full_update = true; + } +} +EXPORT_SYMBOL(drm_atomic_helper_damage_iter_init); + +/** + * drm_atomic_helper_damage_iter_next - Advance the damage iterator. + * @iter: The iterator to advance. + * @rect: Return a rectangle in fb coordinate clipped to plane src. + * + * Since plane src is in 16.16 fixed point and damage clips are whole number, + * this iterator round off clips that intersect with plane src. Round down for + * x1/y1 and round up for x2/y2 for the intersected coordinate. Similar rounding + * off for full plane src, in case it's returned as damage. This iterator will + * skip damage clips outside of plane src. + * + * Return: True if the output is valid, false if reached the end. + * + * If the first call to iterator next returns false then it means no need to + * update the plane. + */ +bool +drm_atomic_helper_damage_iter_next(struct drm_atomic_helper_damage_iter *iter, + struct drm_rect *rect) +{ + bool ret = false; + + if (iter->full_update) { + *rect = iter->plane_src; + iter->full_update = false; + return true; + } + + while (iter->curr_clip < iter->num_clips) { + *rect = iter->clips[iter->curr_clip]; + iter->curr_clip++; + + if (drm_rect_intersect(rect, &iter->plane_src)) { + ret = true; + break; + } + } + + return ret; +} +EXPORT_SYMBOL(drm_atomic_helper_damage_iter_next); diff --git a/drivers/gpu/drm/drm_dp_helper.c b/drivers/gpu/drm/drm_dp_helper.c index 6d483487f2b4..2d6c491a0542 100644 --- a/drivers/gpu/drm/drm_dp_helper.c +++ b/drivers/gpu/drm/drm_dp_helper.c @@ -1428,17 +1428,19 @@ u8 drm_dp_dsc_sink_line_buf_depth(const u8 dsc_dpcd[DP_DSC_RECEIVER_CAP_SIZE]) } EXPORT_SYMBOL(drm_dp_dsc_sink_line_buf_depth); -u8 drm_dp_dsc_sink_max_color_depth(const u8 dsc_dpcd[DP_DSC_RECEIVER_CAP_SIZE]) +int drm_dp_dsc_sink_supported_input_bpcs(const u8 dsc_dpcd[DP_DSC_RECEIVER_CAP_SIZE], + u8 dsc_bpc[3]) { + int num_bpc = 0; u8 color_depth = dsc_dpcd[DP_DSC_DEC_COLOR_DEPTH_CAP - DP_DSC_SUPPORT]; if (color_depth & DP_DSC_12_BPC) - return 12; + dsc_bpc[num_bpc++] = 12; if (color_depth & DP_DSC_10_BPC) - return 10; + dsc_bpc[num_bpc++] = 10; if (color_depth & DP_DSC_8_BPC) - return 8; + dsc_bpc[num_bpc++] = 8; - return 0; + return num_bpc; } -EXPORT_SYMBOL(drm_dp_dsc_sink_max_color_depth); +EXPORT_SYMBOL(drm_dp_dsc_sink_supported_input_bpcs); diff --git a/drivers/gpu/drm/drm_dsc.c b/drivers/gpu/drm/drm_dsc.c new file mode 100644 index 000000000000..bc2b23adb072 --- /dev/null +++ b/drivers/gpu/drm/drm_dsc.c @@ -0,0 +1,228 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2018 Intel Corp + * + * Author: + * Manasi Navare <manasi.d.navare@intel.com> + */ + +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/errno.h> +#include <linux/byteorder/generic.h> +#include <drm/drm_dp_helper.h> +#include <drm/drm_dsc.h> + +/** + * DOC: dsc helpers + * + * These functions contain some common logic and helpers to deal with VESA + * Display Stream Compression standard required for DSC on Display Port/eDP or + * MIPI display interfaces. + */ + +/** + * drm_dsc_dp_pps_header_init() - Initializes the PPS Header + * for DisplayPort as per the DP 1.4 spec. + * @pps_sdp: Secondary data packet for DSC Picture Parameter Set + */ +void drm_dsc_dp_pps_header_init(struct drm_dsc_pps_infoframe *pps_sdp) +{ + memset(&pps_sdp->pps_header, 0, sizeof(pps_sdp->pps_header)); + + pps_sdp->pps_header.HB1 = DP_SDP_PPS; + pps_sdp->pps_header.HB2 = DP_SDP_PPS_HEADER_PAYLOAD_BYTES_MINUS_1; +} +EXPORT_SYMBOL(drm_dsc_dp_pps_header_init); + +/** + * drm_dsc_pps_infoframe_pack() - Populates the DSC PPS infoframe + * using the DSC configuration parameters in the order expected + * by the DSC Display Sink device. For the DSC, the sink device + * expects the PPS payload in the big endian format for the fields + * that span more than 1 byte. + * + * @pps_sdp: + * Secondary data packet for DSC Picture Parameter Set + * @dsc_cfg: + * DSC Configuration data filled by driver + */ +void drm_dsc_pps_infoframe_pack(struct drm_dsc_pps_infoframe *pps_sdp, + const struct drm_dsc_config *dsc_cfg) +{ + int i; + + /* Protect against someone accidently changing struct size */ + BUILD_BUG_ON(sizeof(pps_sdp->pps_payload) != + DP_SDP_PPS_HEADER_PAYLOAD_BYTES_MINUS_1 + 1); + + memset(&pps_sdp->pps_payload, 0, sizeof(pps_sdp->pps_payload)); + + /* PPS 0 */ + pps_sdp->pps_payload.dsc_version = + dsc_cfg->dsc_version_minor | + dsc_cfg->dsc_version_major << DSC_PPS_VERSION_MAJOR_SHIFT; + + /* PPS 1, 2 is 0 */ + + /* PPS 3 */ + pps_sdp->pps_payload.pps_3 = + dsc_cfg->line_buf_depth | + dsc_cfg->bits_per_component << DSC_PPS_BPC_SHIFT; + + /* PPS 4 */ + pps_sdp->pps_payload.pps_4 = + ((dsc_cfg->bits_per_pixel & DSC_PPS_BPP_HIGH_MASK) >> + DSC_PPS_MSB_SHIFT) | + dsc_cfg->vbr_enable << DSC_PPS_VBR_EN_SHIFT | + dsc_cfg->enable422 << DSC_PPS_SIMPLE422_SHIFT | + dsc_cfg->convert_rgb << DSC_PPS_CONVERT_RGB_SHIFT | + dsc_cfg->block_pred_enable << DSC_PPS_BLOCK_PRED_EN_SHIFT; + + /* PPS 5 */ + pps_sdp->pps_payload.bits_per_pixel_low = + (dsc_cfg->bits_per_pixel & DSC_PPS_LSB_MASK); + + /* + * The DSC panel expects the PPS packet to have big endian format + * for data spanning 2 bytes. Use a macro cpu_to_be16() to convert + * to big endian format. If format is little endian, it will swap + * bytes to convert to Big endian else keep it unchanged. + */ + + /* PPS 6, 7 */ + pps_sdp->pps_payload.pic_height = cpu_to_be16(dsc_cfg->pic_height); + + /* PPS 8, 9 */ + pps_sdp->pps_payload.pic_width = cpu_to_be16(dsc_cfg->pic_width); + + /* PPS 10, 11 */ + pps_sdp->pps_payload.slice_height = cpu_to_be16(dsc_cfg->slice_height); + + /* PPS 12, 13 */ + pps_sdp->pps_payload.slice_width = cpu_to_be16(dsc_cfg->slice_width); + + /* PPS 14, 15 */ + pps_sdp->pps_payload.chunk_size = cpu_to_be16(dsc_cfg->slice_chunk_size); + + /* PPS 16 */ + pps_sdp->pps_payload.initial_xmit_delay_high = + ((dsc_cfg->initial_xmit_delay & + DSC_PPS_INIT_XMIT_DELAY_HIGH_MASK) >> + DSC_PPS_MSB_SHIFT); + + /* PPS 17 */ + pps_sdp->pps_payload.initial_xmit_delay_low = + (dsc_cfg->initial_xmit_delay & DSC_PPS_LSB_MASK); + + /* PPS 18, 19 */ + pps_sdp->pps_payload.initial_dec_delay = + cpu_to_be16(dsc_cfg->initial_dec_delay); + + /* PPS 20 is 0 */ + + /* PPS 21 */ + pps_sdp->pps_payload.initial_scale_value = + dsc_cfg->initial_scale_value; + + /* PPS 22, 23 */ + pps_sdp->pps_payload.scale_increment_interval = + cpu_to_be16(dsc_cfg->scale_increment_interval); + + /* PPS 24 */ + pps_sdp->pps_payload.scale_decrement_interval_high = + ((dsc_cfg->scale_decrement_interval & + DSC_PPS_SCALE_DEC_INT_HIGH_MASK) >> + DSC_PPS_MSB_SHIFT); + + /* PPS 25 */ + pps_sdp->pps_payload.scale_decrement_interval_low = + (dsc_cfg->scale_decrement_interval & DSC_PPS_LSB_MASK); + + /* PPS 26[7:0], PPS 27[7:5] RESERVED */ + + /* PPS 27 */ + pps_sdp->pps_payload.first_line_bpg_offset = + dsc_cfg->first_line_bpg_offset; + + /* PPS 28, 29 */ + pps_sdp->pps_payload.nfl_bpg_offset = + cpu_to_be16(dsc_cfg->nfl_bpg_offset); + + /* PPS 30, 31 */ + pps_sdp->pps_payload.slice_bpg_offset = + cpu_to_be16(dsc_cfg->slice_bpg_offset); + + /* PPS 32, 33 */ + pps_sdp->pps_payload.initial_offset = + cpu_to_be16(dsc_cfg->initial_offset); + + /* PPS 34, 35 */ + pps_sdp->pps_payload.final_offset = cpu_to_be16(dsc_cfg->final_offset); + + /* PPS 36 */ + pps_sdp->pps_payload.flatness_min_qp = dsc_cfg->flatness_min_qp; + + /* PPS 37 */ + pps_sdp->pps_payload.flatness_max_qp = dsc_cfg->flatness_max_qp; + + /* PPS 38, 39 */ + pps_sdp->pps_payload.rc_model_size = + cpu_to_be16(DSC_RC_MODEL_SIZE_CONST); + + /* PPS 40 */ + pps_sdp->pps_payload.rc_edge_factor = DSC_RC_EDGE_FACTOR_CONST; + + /* PPS 41 */ + pps_sdp->pps_payload.rc_quant_incr_limit0 = + dsc_cfg->rc_quant_incr_limit0; + + /* PPS 42 */ + pps_sdp->pps_payload.rc_quant_incr_limit1 = + dsc_cfg->rc_quant_incr_limit1; + + /* PPS 43 */ + pps_sdp->pps_payload.rc_tgt_offset = DSC_RC_TGT_OFFSET_LO_CONST | + DSC_RC_TGT_OFFSET_HI_CONST << DSC_PPS_RC_TGT_OFFSET_HI_SHIFT; + + /* PPS 44 - 57 */ + for (i = 0; i < DSC_NUM_BUF_RANGES - 1; i++) + pps_sdp->pps_payload.rc_buf_thresh[i] = + dsc_cfg->rc_buf_thresh[i]; + + /* PPS 58 - 87 */ + /* + * For DSC sink programming the RC Range parameter fields + * are as follows: Min_qp[15:11], max_qp[10:6], offset[5:0] + */ + for (i = 0; i < DSC_NUM_BUF_RANGES; i++) { + pps_sdp->pps_payload.rc_range_parameters[i] = + ((dsc_cfg->rc_range_params[i].range_min_qp << + DSC_PPS_RC_RANGE_MINQP_SHIFT) | + (dsc_cfg->rc_range_params[i].range_max_qp << + DSC_PPS_RC_RANGE_MAXQP_SHIFT) | + (dsc_cfg->rc_range_params[i].range_bpg_offset)); + pps_sdp->pps_payload.rc_range_parameters[i] = + cpu_to_be16(pps_sdp->pps_payload.rc_range_parameters[i]); + } + + /* PPS 88 */ + pps_sdp->pps_payload.native_422_420 = dsc_cfg->native_422 | + dsc_cfg->native_420 << DSC_PPS_NATIVE_420_SHIFT; + + /* PPS 89 */ + pps_sdp->pps_payload.second_line_bpg_offset = + dsc_cfg->second_line_bpg_offset; + + /* PPS 90, 91 */ + pps_sdp->pps_payload.nsl_bpg_offset = + cpu_to_be16(dsc_cfg->nsl_bpg_offset); + + /* PPS 92, 93 */ + pps_sdp->pps_payload.second_line_offset_adj = + cpu_to_be16(dsc_cfg->second_line_offset_adj); + + /* PPS 94 - 127 are O */ +} +EXPORT_SYMBOL(drm_dsc_pps_infoframe_pack); diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c index 883a0c44714c..ca706fb1d975 100644 --- a/drivers/gpu/drm/drm_fb_helper.c +++ b/drivers/gpu/drm/drm_fb_helper.c @@ -71,7 +71,7 @@ MODULE_PARM_DESC(drm_fbdev_overalloc, #if IS_ENABLED(CONFIG_DRM_FBDEV_LEAK_PHYS_SMEM) static bool drm_leak_fbdev_smem = false; module_param_unsafe(drm_leak_fbdev_smem, bool, 0600); -MODULE_PARM_DESC(fbdev_emulation, +MODULE_PARM_DESC(drm_leak_fbdev_smem, "Allow unsafe leaking fbdev physical smem address [default=false]"); #endif diff --git a/drivers/gpu/drm/drm_file.c b/drivers/gpu/drm/drm_file.c index ffa8dc35515f..46f48f245eb5 100644 --- a/drivers/gpu/drm/drm_file.c +++ b/drivers/gpu/drm/drm_file.c @@ -525,7 +525,7 @@ ssize_t drm_read(struct file *filp, char __user *buffer, struct drm_device *dev = file_priv->minor->dev; ssize_t ret; - if (!access_ok(VERIFY_WRITE, buffer, count)) + if (!access_ok(buffer, count)) return -EFAULT; ret = mutex_lock_interruptible(&file_priv->event_read_lock); diff --git a/drivers/gpu/drm/drm_internal.h b/drivers/gpu/drm/drm_internal.h index 331112b2ae88..251d67e04c2d 100644 --- a/drivers/gpu/drm/drm_internal.h +++ b/drivers/gpu/drm/drm_internal.h @@ -101,6 +101,8 @@ struct device *drm_sysfs_minor_alloc(struct drm_minor *minor); int drm_sysfs_connector_add(struct drm_connector *connector); void drm_sysfs_connector_remove(struct drm_connector *connector); +void drm_sysfs_lease_event(struct drm_device *dev); + /* drm_gem.c */ int drm_gem_init(struct drm_device *dev); void drm_gem_destroy(struct drm_device *dev); diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c index 94bd872d56c4..7e6746b2d704 100644 --- a/drivers/gpu/drm/drm_ioctl.c +++ b/drivers/gpu/drm/drm_ioctl.c @@ -37,6 +37,7 @@ #include <linux/pci.h> #include <linux/export.h> +#include <linux/nospec.h> /** * DOC: getunique and setversion story @@ -800,13 +801,17 @@ long drm_ioctl(struct file *filp, if (is_driver_ioctl) { /* driver ioctl */ - if (nr - DRM_COMMAND_BASE >= dev->driver->num_ioctls) + unsigned int index = nr - DRM_COMMAND_BASE; + + if (index >= dev->driver->num_ioctls) goto err_i1; - ioctl = &dev->driver->ioctls[nr - DRM_COMMAND_BASE]; + index = array_index_nospec(index, dev->driver->num_ioctls); + ioctl = &dev->driver->ioctls[index]; } else { /* core ioctl */ if (nr >= DRM_CORE_IOCTL_COUNT) goto err_i1; + nr = array_index_nospec(nr, DRM_CORE_IOCTL_COUNT); ioctl = &drm_ioctls[nr]; } @@ -888,6 +893,7 @@ bool drm_ioctl_flags(unsigned int nr, unsigned int *flags) if (nr >= DRM_CORE_IOCTL_COUNT) return false; + nr = array_index_nospec(nr, DRM_CORE_IOCTL_COUNT); *flags = drm_ioctls[nr].flags; return true; diff --git a/drivers/gpu/drm/drm_lease.c b/drivers/gpu/drm/drm_lease.c index d9b4d3ff06f6..b735704653cb 100644 --- a/drivers/gpu/drm/drm_lease.c +++ b/drivers/gpu/drm/drm_lease.c @@ -292,7 +292,7 @@ void drm_lease_destroy(struct drm_master *master) if (master->lessor) { /* Tell the master to check the lessee list */ - drm_sysfs_hotplug_event(dev); + drm_sysfs_lease_event(dev); drm_master_put(&master->lessor); } diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c index 037e243ec863..4a1c2023ccf0 100644 --- a/drivers/gpu/drm/drm_mode_config.c +++ b/drivers/gpu/drm/drm_mode_config.c @@ -297,6 +297,12 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) return -ENOMEM; dev->mode_config.prop_crtc_id = prop; + prop = drm_property_create(dev, DRM_MODE_PROP_BLOB, "FB_DAMAGE_CLIPS", + 0); + if (!prop) + return -ENOMEM; + dev->mode_config.prop_fb_damage_clips = prop; + prop = drm_property_create_bool(dev, DRM_MODE_PROP_ATOMIC, "ACTIVE"); if (!prop) @@ -310,6 +316,12 @@ static int drm_mode_create_standard_properties(struct drm_device *dev) return -ENOMEM; dev->mode_config.prop_mode_id = prop; + prop = drm_property_create_bool(dev, 0, + "VRR_ENABLED"); + if (!prop) + return -ENOMEM; + dev->mode_config.prop_vrr_enabled = prop; + prop = drm_property_create(dev, DRM_MODE_PROP_BLOB, "DEGAMMA_LUT", 0); diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c index b3c1daad1169..ecb7b33002bb 100644 --- a/drivers/gpu/drm/drm_sysfs.c +++ b/drivers/gpu/drm/drm_sysfs.c @@ -301,6 +301,16 @@ void drm_sysfs_connector_remove(struct drm_connector *connector) connector->kdev = NULL; } +void drm_sysfs_lease_event(struct drm_device *dev) +{ + char *event_string = "LEASE=1"; + char *envp[] = { event_string, NULL }; + + DRM_DEBUG("generating lease event\n"); + + kobject_uevent_env(&dev->primary->kdev->kobj, KOBJ_CHANGE, envp); +} + /** * drm_sysfs_hotplug_event - generate a DRM uevent * @dev: DRM device diff --git a/drivers/gpu/drm/etnaviv/etnaviv_buffer.c b/drivers/gpu/drm/etnaviv/etnaviv_buffer.c index 7fea74861a87..160ce3c060a5 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_buffer.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_buffer.c @@ -439,6 +439,4 @@ void etnaviv_buffer_queue(struct etnaviv_gpu *gpu, u32 exec_state, if (drm_debug & DRM_UT_DRIVER) etnaviv_buffer_dump(gpu, buffer, 0, 0x50); - - gpu->lastctx = cmdbuf->ctx; } diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.c b/drivers/gpu/drm/etnaviv/etnaviv_drv.c index 52802e6049e0..18c27f795cf6 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_drv.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.c @@ -72,14 +72,8 @@ static void etnaviv_postclose(struct drm_device *dev, struct drm_file *file) for (i = 0; i < ETNA_MAX_PIPES; i++) { struct etnaviv_gpu *gpu = priv->gpu[i]; - if (gpu) { - mutex_lock(&gpu->lock); - if (gpu->lastctx == ctx) - gpu->lastctx = NULL; - mutex_unlock(&gpu->lock); - + if (gpu) drm_sched_entity_destroy(&ctx->sched_entity[i]); - } } kfree(ctx); @@ -345,7 +339,6 @@ static int etnaviv_ioctl_gem_userptr(struct drm_device *dev, void *data, struct drm_file *file) { struct drm_etnaviv_gem_userptr *args = data; - int access; if (args->flags & ~(ETNA_USERPTR_READ|ETNA_USERPTR_WRITE) || args->flags == 0) @@ -357,12 +350,7 @@ static int etnaviv_ioctl_gem_userptr(struct drm_device *dev, void *data, args->user_ptr & ~PAGE_MASK) return -EINVAL; - if (args->flags & ETNA_USERPTR_WRITE) - access = VERIFY_WRITE; - else - access = VERIFY_READ; - - if (!access_ok(access, (void __user *)(unsigned long)args->user_ptr, + if (!access_ok((void __user *)(unsigned long)args->user_ptr, args->user_size)) return -EFAULT; @@ -523,7 +511,7 @@ static int etnaviv_bind(struct device *dev) if (!priv) { dev_err(dev, "failed to allocate private data\n"); ret = -ENOMEM; - goto out_unref; + goto out_put; } drm->dev_private = priv; @@ -549,7 +537,7 @@ out_register: component_unbind_all(dev, drm); out_bind: kfree(priv); -out_unref: +out_put: drm_dev_put(drm); return ret; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.h b/drivers/gpu/drm/etnaviv/etnaviv_drv.h index 8d02d1b7dcf5..4bf698de5996 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_drv.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.h @@ -107,17 +107,6 @@ static inline size_t size_vstruct(size_t nelem, size_t elem_size, size_t base) return base + nelem * elem_size; } -/* returns true if fence a comes after fence b */ -static inline bool fence_after(u32 a, u32 b) -{ - return (s32)(a - b) > 0; -} - -static inline bool fence_after_eq(u32 a, u32 b) -{ - return (s32)(a - b) >= 0; -} - /* * Etnaviv timeouts are specified wrt CLOCK_MONOTONIC, not jiffies. * We need to calculate the timeout in terms of number of jiffies diff --git a/drivers/gpu/drm/etnaviv/etnaviv_dump.c b/drivers/gpu/drm/etnaviv/etnaviv_dump.c index 9146e30e24a6..3fbb4855396c 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_dump.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_dump.c @@ -118,6 +118,7 @@ void etnaviv_core_dump(struct etnaviv_gpu *gpu) unsigned int n_obj, n_bomap_pages; size_t file_size, mmu_size; __le64 *bomap, *bomap_start; + unsigned long flags; /* Only catch the first event, or when manually re-armed */ if (!etnaviv_dump_core) @@ -134,13 +135,13 @@ void etnaviv_core_dump(struct etnaviv_gpu *gpu) mmu_size + gpu->buffer.size; /* Add in the active command buffers */ - spin_lock(&gpu->sched.job_list_lock); + spin_lock_irqsave(&gpu->sched.job_list_lock, flags); list_for_each_entry(s_job, &gpu->sched.ring_mirror_list, node) { submit = to_etnaviv_submit(s_job); file_size += submit->cmdbuf.size; n_obj++; } - spin_unlock(&gpu->sched.job_list_lock); + spin_unlock_irqrestore(&gpu->sched.job_list_lock, flags); /* Add in the active buffer objects */ list_for_each_entry(vram, &gpu->mmu->mappings, mmu_node) { @@ -182,14 +183,14 @@ void etnaviv_core_dump(struct etnaviv_gpu *gpu) gpu->buffer.size, etnaviv_cmdbuf_get_va(&gpu->buffer)); - spin_lock(&gpu->sched.job_list_lock); + spin_lock_irqsave(&gpu->sched.job_list_lock, flags); list_for_each_entry(s_job, &gpu->sched.ring_mirror_list, node) { submit = to_etnaviv_submit(s_job); etnaviv_core_dump_mem(&iter, ETDUMP_BUF_CMD, submit->cmdbuf.vaddr, submit->cmdbuf.size, etnaviv_cmdbuf_get_va(&submit->cmdbuf)); } - spin_unlock(&gpu->sched.job_list_lock); + spin_unlock_irqrestore(&gpu->sched.job_list_lock, flags); /* Reserve space for the bomap */ if (n_bomap_pages) { diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c index f225fbc6edd2..6904535475de 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c @@ -3,10 +3,12 @@ * Copyright (C) 2015-2018 Etnaviv Project */ +#include <linux/clk.h> #include <linux/component.h> #include <linux/dma-fence.h> #include <linux/moduleparam.h> #include <linux/of_device.h> +#include <linux/regulator/consumer.h> #include <linux/thermal.h> #include "etnaviv_cmdbuf.h" @@ -976,7 +978,6 @@ int etnaviv_gpu_debugfs(struct etnaviv_gpu *gpu, struct seq_file *m) void etnaviv_gpu_recover_hang(struct etnaviv_gpu *gpu) { - unsigned long flags; unsigned int i = 0; dev_err(gpu->dev, "recover hung GPU!\n"); @@ -989,15 +990,13 @@ void etnaviv_gpu_recover_hang(struct etnaviv_gpu *gpu) etnaviv_hw_reset(gpu); /* complete all events, the GPU won't do it after the reset */ - spin_lock_irqsave(&gpu->event_spinlock, flags); + spin_lock(&gpu->event_spinlock); for_each_set_bit_from(i, gpu->event_bitmap, ETNA_NR_EVENTS) complete(&gpu->event_free); bitmap_zero(gpu->event_bitmap, ETNA_NR_EVENTS); - spin_unlock_irqrestore(&gpu->event_spinlock, flags); - gpu->completed_fence = gpu->active_fence; + spin_unlock(&gpu->event_spinlock); etnaviv_gpu_hw_init(gpu); - gpu->lastctx = NULL; gpu->exec_state = -1; mutex_unlock(&gpu->lock); @@ -1032,7 +1031,7 @@ static bool etnaviv_fence_signaled(struct dma_fence *fence) { struct etnaviv_fence *f = to_etnaviv_fence(fence); - return fence_completed(f->gpu, f->base.seqno); + return (s32)(f->gpu->completed_fence - f->base.seqno) >= 0; } static void etnaviv_fence_release(struct dma_fence *fence) @@ -1071,6 +1070,12 @@ static struct dma_fence *etnaviv_gpu_fence_alloc(struct etnaviv_gpu *gpu) return &f->base; } +/* returns true if fence a comes after fence b */ +static inline bool fence_after(u32 a, u32 b) +{ + return (s32)(a - b) > 0; +} + /* * event management: */ @@ -1078,7 +1083,7 @@ static struct dma_fence *etnaviv_gpu_fence_alloc(struct etnaviv_gpu *gpu) static int event_alloc(struct etnaviv_gpu *gpu, unsigned nr_events, unsigned int *events) { - unsigned long flags, timeout = msecs_to_jiffies(10 * 10000); + unsigned long timeout = msecs_to_jiffies(10 * 10000); unsigned i, acquired = 0; for (i = 0; i < nr_events; i++) { @@ -1095,7 +1100,7 @@ static int event_alloc(struct etnaviv_gpu *gpu, unsigned nr_events, timeout = ret; } - spin_lock_irqsave(&gpu->event_spinlock, flags); + spin_lock(&gpu->event_spinlock); for (i = 0; i < nr_events; i++) { int event = find_first_zero_bit(gpu->event_bitmap, ETNA_NR_EVENTS); @@ -1105,7 +1110,7 @@ static int event_alloc(struct etnaviv_gpu *gpu, unsigned nr_events, set_bit(event, gpu->event_bitmap); } - spin_unlock_irqrestore(&gpu->event_spinlock, flags); + spin_unlock(&gpu->event_spinlock); return 0; @@ -1118,18 +1123,11 @@ out: static void event_free(struct etnaviv_gpu *gpu, unsigned int event) { - unsigned long flags; - - spin_lock_irqsave(&gpu->event_spinlock, flags); - if (!test_bit(event, gpu->event_bitmap)) { dev_warn(gpu->dev, "event %u is already marked as free", event); - spin_unlock_irqrestore(&gpu->event_spinlock, flags); } else { clear_bit(event, gpu->event_bitmap); - spin_unlock_irqrestore(&gpu->event_spinlock, flags); - complete(&gpu->event_free); } } @@ -1306,8 +1304,6 @@ struct dma_fence *etnaviv_gpu_submit(struct etnaviv_gem_submit *submit) goto out_unlock; } - gpu->active_fence = gpu_fence->seqno; - if (submit->nr_pmrs) { gpu->event[event[1]].sync_point = &sync_point_perfmon_sample_pre; kref_get(&submit->refcount); @@ -1549,7 +1545,6 @@ static int etnaviv_gpu_hw_resume(struct etnaviv_gpu *gpu) etnaviv_gpu_update_clock(gpu); etnaviv_gpu_hw_init(gpu); - gpu->lastctx = NULL; gpu->exec_state = -1; mutex_unlock(&gpu->lock); @@ -1806,8 +1801,8 @@ static int etnaviv_gpu_rpm_suspend(struct device *dev) struct etnaviv_gpu *gpu = dev_get_drvdata(dev); u32 idle, mask; - /* If we have outstanding fences, we're not idle */ - if (gpu->completed_fence != gpu->active_fence) + /* If there are any jobs in the HW queue, we're not idle */ + if (atomic_read(&gpu->sched.hw_rq_count)) return -EBUSY; /* Check whether the hardware (except FE) is idle */ diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.h b/drivers/gpu/drm/etnaviv/etnaviv_gpu.h index 9a75a6937268..9bcf151f706b 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.h @@ -6,9 +6,6 @@ #ifndef __ETNAVIV_GPU_H__ #define __ETNAVIV_GPU_H__ -#include <linux/clk.h> -#include <linux/regulator/consumer.h> - #include "etnaviv_cmdbuf.h" #include "etnaviv_drv.h" @@ -88,6 +85,8 @@ struct etnaviv_event { struct etnaviv_cmdbuf_suballoc; struct etnaviv_cmdbuf; +struct regulator; +struct clk; #define ETNA_NR_EVENTS 30 @@ -98,7 +97,6 @@ struct etnaviv_gpu { struct mutex lock; struct etnaviv_chip_identity identity; enum etnaviv_sec_mode sec_mode; - struct etnaviv_file_private *lastctx; struct workqueue_struct *wq; struct drm_gpu_scheduler sched; @@ -121,7 +119,6 @@ struct etnaviv_gpu { struct mutex fence_lock; struct idr fence_idr; u32 next_fence; - u32 active_fence; u32 completed_fence; wait_queue_head_t fence_event; u64 fence_context; @@ -161,11 +158,6 @@ static inline u32 gpu_read(struct etnaviv_gpu *gpu, u32 reg) return readl(gpu->mmio + reg); } -static inline bool fence_completed(struct etnaviv_gpu *gpu, u32 fence) -{ - return fence_after_eq(gpu->completed_fence, fence); -} - int etnaviv_gpu_get_param(struct etnaviv_gpu *gpu, u32 param, u64 *value); int etnaviv_gpu_init(struct etnaviv_gpu *gpu); diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig index 208bc27be3cc..3691a140c950 100644 --- a/drivers/gpu/drm/exynos/Kconfig +++ b/drivers/gpu/drm/exynos/Kconfig @@ -10,11 +10,6 @@ config DRM_EXYNOS if DRM_EXYNOS -config DRM_EXYNOS_IOMMU - bool - depends on EXYNOS_IOMMU - default y - comment "CRTCs" config DRM_EXYNOS_FIMD diff --git a/drivers/gpu/drm/exynos/Makefile b/drivers/gpu/drm/exynos/Makefile index 2ad146bbf4f5..2fd2f3ee4fcf 100644 --- a/drivers/gpu/drm/exynos/Makefile +++ b/drivers/gpu/drm/exynos/Makefile @@ -4,10 +4,9 @@ # Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher. exynosdrm-y := exynos_drm_drv.o exynos_drm_crtc.o exynos_drm_fb.o \ - exynos_drm_gem.o exynos_drm_plane.o + exynos_drm_gem.o exynos_drm_plane.o exynos_drm_dma.o exynosdrm-$(CONFIG_DRM_FBDEV_EMULATION) += exynos_drm_fbdev.o -exynosdrm-$(CONFIG_DRM_EXYNOS_IOMMU) += exynos_drm_iommu.o exynosdrm-$(CONFIG_DRM_EXYNOS_FIMD) += exynos_drm_fimd.o exynosdrm-$(CONFIG_DRM_EXYNOS5433_DECON) += exynos5433_drm_decon.o exynosdrm-$(CONFIG_DRM_EXYNOS7_DECON) += exynos7_drm_decon.o diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c index aef487dd8731..5b4e0e8b23bc 100644 --- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c +++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c @@ -25,7 +25,6 @@ #include "exynos_drm_crtc.h" #include "exynos_drm_fb.h" #include "exynos_drm_plane.h" -#include "exynos_drm_iommu.h" #include "regs-decon5433.h" #define DSD_CFG_MUX 0x1004 @@ -84,6 +83,14 @@ static const enum drm_plane_type decon_win_types[WINDOWS_NR] = { [CURSON_WIN] = DRM_PLANE_TYPE_CURSOR, }; +static const unsigned int capabilities[WINDOWS_NR] = { + 0, + EXYNOS_DRM_PLANE_CAP_WIN_BLEND | EXYNOS_DRM_PLANE_CAP_PIX_BLEND, + EXYNOS_DRM_PLANE_CAP_WIN_BLEND | EXYNOS_DRM_PLANE_CAP_PIX_BLEND, + EXYNOS_DRM_PLANE_CAP_WIN_BLEND | EXYNOS_DRM_PLANE_CAP_PIX_BLEND, + EXYNOS_DRM_PLANE_CAP_WIN_BLEND | EXYNOS_DRM_PLANE_CAP_PIX_BLEND, +}; + static inline void decon_set_bits(struct decon_context *ctx, u32 reg, u32 mask, u32 val) { @@ -252,11 +259,76 @@ static void decon_commit(struct exynos_drm_crtc *crtc) decon_set_bits(ctx, DECON_UPDATE, STANDALONE_UPDATE_F, ~0); } +static void decon_win_set_bldeq(struct decon_context *ctx, unsigned int win, + unsigned int alpha, unsigned int pixel_alpha) +{ + u32 mask = BLENDERQ_A_FUNC_F(0xf) | BLENDERQ_B_FUNC_F(0xf); + u32 val = 0; + + switch (pixel_alpha) { + case DRM_MODE_BLEND_PIXEL_NONE: + case DRM_MODE_BLEND_COVERAGE: + val |= BLENDERQ_A_FUNC_F(BLENDERQ_ALPHA_A); + val |= BLENDERQ_B_FUNC_F(BLENDERQ_ONE_MINUS_ALPHA_A); + break; + case DRM_MODE_BLEND_PREMULTI: + default: + if (alpha != DRM_BLEND_ALPHA_OPAQUE) { + val |= BLENDERQ_A_FUNC_F(BLENDERQ_ALPHA0); + val |= BLENDERQ_B_FUNC_F(BLENDERQ_ONE_MINUS_ALPHA_A); + } else { + val |= BLENDERQ_A_FUNC_F(BLENDERQ_ONE); + val |= BLENDERQ_B_FUNC_F(BLENDERQ_ONE_MINUS_ALPHA_A); + } + break; + } + decon_set_bits(ctx, DECON_BLENDERQx(win), mask, val); +} + +static void decon_win_set_bldmod(struct decon_context *ctx, unsigned int win, + unsigned int alpha, unsigned int pixel_alpha) +{ + u32 win_alpha = alpha >> 8; + u32 val = 0; + + switch (pixel_alpha) { + case DRM_MODE_BLEND_PIXEL_NONE: + break; + case DRM_MODE_BLEND_COVERAGE: + case DRM_MODE_BLEND_PREMULTI: + default: + val |= WINCONx_ALPHA_SEL_F; + val |= WINCONx_BLD_PIX_F; + val |= WINCONx_ALPHA_MUL_F; + break; + } + decon_set_bits(ctx, DECON_WINCONx(win), WINCONx_BLEND_MODE_MASK, val); + + if (alpha != DRM_BLEND_ALPHA_OPAQUE) { + val = VIDOSD_Wx_ALPHA_R_F(win_alpha) | + VIDOSD_Wx_ALPHA_G_F(win_alpha) | + VIDOSD_Wx_ALPHA_B_F(win_alpha); + decon_set_bits(ctx, DECON_VIDOSDxC(win), + VIDOSDxC_ALPHA0_RGB_MASK, val); + decon_set_bits(ctx, DECON_BLENDCON, BLEND_NEW, BLEND_NEW); + } +} + static void decon_win_set_pixfmt(struct decon_context *ctx, unsigned int win, struct drm_framebuffer *fb) { + struct exynos_drm_plane plane = ctx->planes[win]; + struct exynos_drm_plane_state *state = + to_exynos_plane_state(plane.base.state); + unsigned int alpha = state->base.alpha; + unsigned int pixel_alpha; unsigned long val; + if (fb->format->has_alpha) + pixel_alpha = state->base.pixel_blend_mode; + else + pixel_alpha = DRM_MODE_BLEND_PIXEL_NONE; + val = readl(ctx->addr + DECON_WINCONx(win)); val &= WINCONx_ENWIN_F; @@ -279,7 +351,7 @@ static void decon_win_set_pixfmt(struct decon_context *ctx, unsigned int win, case DRM_FORMAT_ARGB8888: default: val |= WINCONx_BPPMODE_32BPP_A8888; - val |= WINCONx_WSWP_F | WINCONx_BLD_PIX_F | WINCONx_ALPHA_SEL_F; + val |= WINCONx_WSWP_F; val |= WINCONx_BURSTLEN_16WORD; break; } @@ -298,8 +370,12 @@ static void decon_win_set_pixfmt(struct decon_context *ctx, unsigned int win, val &= ~WINCONx_BURSTLEN_MASK; val |= WINCONx_BURSTLEN_8WORD; } + decon_set_bits(ctx, DECON_WINCONx(win), ~WINCONx_BLEND_MODE_MASK, val); - writel(val, ctx->addr + DECON_WINCONx(win)); + if (win > 0) { + decon_win_set_bldmod(ctx, win, alpha, pixel_alpha); + decon_win_set_bldeq(ctx, win, alpha, pixel_alpha); + } } static void decon_shadow_protect(struct decon_context *ctx, bool protect) @@ -552,6 +628,7 @@ static int decon_bind(struct device *dev, struct device *master, void *data) ctx->configs[win].num_pixel_formats = ARRAY_SIZE(decon_formats); ctx->configs[win].zpos = win - ctx->first_win; ctx->configs[win].type = decon_win_types[win]; + ctx->configs[win].capabilities = capabilities[win]; ret = exynos_plane_init(drm_dev, &ctx->planes[win], win, &ctx->configs[win]); @@ -569,7 +646,7 @@ static int decon_bind(struct device *dev, struct device *master, void *data) decon_clear_channels(ctx->crtc); - return drm_iommu_attach_device(drm_dev, dev); + return exynos_drm_register_dma(drm_dev, dev); } static void decon_unbind(struct device *dev, struct device *master, void *data) @@ -579,7 +656,7 @@ static void decon_unbind(struct device *dev, struct device *master, void *data) decon_disable(ctx->crtc); /* detach this sub driver from iommu mapping if supported. */ - drm_iommu_detach_device(ctx->drm_dev, ctx->dev); + exynos_drm_unregister_dma(ctx->drm_dev, ctx->dev); } static const struct component_ops decon_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos7_drm_decon.c b/drivers/gpu/drm/exynos/exynos7_drm_decon.c index 88cbd000eb09..381aa3d60e37 100644 --- a/drivers/gpu/drm/exynos/exynos7_drm_decon.c +++ b/drivers/gpu/drm/exynos/exynos7_drm_decon.c @@ -30,7 +30,6 @@ #include "exynos_drm_plane.h" #include "exynos_drm_drv.h" #include "exynos_drm_fb.h" -#include "exynos_drm_iommu.h" #include "regs-decon7.h" /* @@ -133,13 +132,13 @@ static int decon_ctx_initialize(struct decon_context *ctx, decon_clear_channels(ctx->crtc); - return drm_iommu_attach_device(drm_dev, ctx->dev); + return exynos_drm_register_dma(drm_dev, ctx->dev); } static void decon_ctx_remove(struct decon_context *ctx) { /* detach this sub driver from iommu mapping if supported. */ - drm_iommu_detach_device(ctx->drm_dev, ctx->dev); + exynos_drm_unregister_dma(ctx->drm_dev, ctx->dev); } static u32 decon_calc_clkdiv(struct decon_context *ctx, diff --git a/drivers/gpu/drm/exynos/exynos_drm_dma.c b/drivers/gpu/drm/exynos/exynos_drm_dma.c new file mode 100644 index 000000000000..3432c5ee9f0c --- /dev/null +++ b/drivers/gpu/drm/exynos/exynos_drm_dma.c @@ -0,0 +1,157 @@ +// SPDX-License-Identifier: GPL-2.0 +// +// Copyright (c) 2012 Samsung Electronics Co., Ltd. +// Author: Inki Dae <inki.dae@samsung.com> +// Author: Andrzej Hajda <a.hajda@samsung.com> + +#include <drm/drmP.h> +#include <drm/exynos_drm.h> +#include <linux/dma-iommu.h> +#include <linux/dma-mapping.h> +#include <linux/iommu.h> + +#include "exynos_drm_drv.h" + +#if defined(CONFIG_ARM_DMA_USE_IOMMU) +#include <asm/dma-iommu.h> +#else +#define arm_iommu_create_mapping(...) ({ NULL; }) +#define arm_iommu_attach_device(...) ({ -ENODEV; }) +#define arm_iommu_release_mapping(...) ({ }) +#define arm_iommu_detach_device(...) ({ }) +#define to_dma_iommu_mapping(dev) NULL +#endif + +#if !defined(CONFIG_IOMMU_DMA) +#define iommu_dma_init_domain(...) ({ -EINVAL; }) +#endif + +#define EXYNOS_DEV_ADDR_START 0x20000000 +#define EXYNOS_DEV_ADDR_SIZE 0x40000000 + +static inline int configure_dma_max_seg_size(struct device *dev) +{ + if (!dev->dma_parms) + dev->dma_parms = kzalloc(sizeof(*dev->dma_parms), GFP_KERNEL); + if (!dev->dma_parms) + return -ENOMEM; + + dma_set_max_seg_size(dev, DMA_BIT_MASK(32)); + return 0; +} + +static inline void clear_dma_max_seg_size(struct device *dev) +{ + kfree(dev->dma_parms); + dev->dma_parms = NULL; +} + +/* + * drm_iommu_attach_device- attach device to iommu mapping + * + * @drm_dev: DRM device + * @subdrv_dev: device to be attach + * + * This function should be called by sub drivers to attach it to iommu + * mapping. + */ +static int drm_iommu_attach_device(struct drm_device *drm_dev, + struct device *subdrv_dev) +{ + struct exynos_drm_private *priv = drm_dev->dev_private; + int ret; + + if (get_dma_ops(priv->dma_dev) != get_dma_ops(subdrv_dev)) { + DRM_ERROR("Device %s lacks support for IOMMU\n", + dev_name(subdrv_dev)); + return -EINVAL; + } + + ret = configure_dma_max_seg_size(subdrv_dev); + if (ret) + return ret; + + if (IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU)) { + if (to_dma_iommu_mapping(subdrv_dev)) + arm_iommu_detach_device(subdrv_dev); + + ret = arm_iommu_attach_device(subdrv_dev, priv->mapping); + } else if (IS_ENABLED(CONFIG_IOMMU_DMA)) { + ret = iommu_attach_device(priv->mapping, subdrv_dev); + } + + if (ret) + clear_dma_max_seg_size(subdrv_dev); + + return 0; +} + +/* + * drm_iommu_detach_device -detach device address space mapping from device + * + * @drm_dev: DRM device + * @subdrv_dev: device to be detached + * + * This function should be called by sub drivers to detach it from iommu + * mapping + */ +static void drm_iommu_detach_device(struct drm_device *drm_dev, + struct device *subdrv_dev) +{ + struct exynos_drm_private *priv = drm_dev->dev_private; + + if (IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU)) + arm_iommu_detach_device(subdrv_dev); + else if (IS_ENABLED(CONFIG_IOMMU_DMA)) + iommu_detach_device(priv->mapping, subdrv_dev); + + clear_dma_max_seg_size(subdrv_dev); +} + +int exynos_drm_register_dma(struct drm_device *drm, struct device *dev) +{ + struct exynos_drm_private *priv = drm->dev_private; + + if (!priv->dma_dev) { + priv->dma_dev = dev; + DRM_INFO("Exynos DRM: using %s device for DMA mapping operations\n", + dev_name(dev)); + } + + if (!IS_ENABLED(CONFIG_EXYNOS_IOMMU)) + return 0; + + if (!priv->mapping) { + void *mapping; + + if (IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU)) + mapping = arm_iommu_create_mapping(&platform_bus_type, + EXYNOS_DEV_ADDR_START, EXYNOS_DEV_ADDR_SIZE); + else if (IS_ENABLED(CONFIG_IOMMU_DMA)) + mapping = iommu_get_domain_for_dev(priv->dma_dev); + + if (IS_ERR(mapping)) + return PTR_ERR(mapping); + priv->mapping = mapping; + } + + return drm_iommu_attach_device(drm, dev); +} + +void exynos_drm_unregister_dma(struct drm_device *drm, struct device *dev) +{ + if (IS_ENABLED(CONFIG_EXYNOS_IOMMU)) + drm_iommu_detach_device(drm, dev); +} + +void exynos_drm_cleanup_dma(struct drm_device *drm) +{ + struct exynos_drm_private *priv = drm->dev_private; + + if (!IS_ENABLED(CONFIG_EXYNOS_IOMMU)) + return; + + arm_iommu_release_mapping(priv->mapping); + priv->mapping = NULL; + priv->dma_dev = NULL; +} diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c index 6f76baf4550a..2c75e789b2a7 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.c +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c @@ -30,7 +30,6 @@ #include "exynos_drm_ipp.h" #include "exynos_drm_vidi.h" #include "exynos_drm_g2d.h" -#include "exynos_drm_iommu.h" #define DRIVER_NAME "exynos" #define DRIVER_DESC "Samsung SoC DRM" @@ -175,8 +174,7 @@ struct exynos_drm_driver_info { #define DRM_COMPONENT_DRIVER BIT(0) /* supports component framework */ #define DRM_VIRTUAL_DEVICE BIT(1) /* create virtual platform device */ -#define DRM_DMA_DEVICE BIT(2) /* can be used for dma allocations */ -#define DRM_FIMC_DEVICE BIT(3) /* devices shared with V4L2 subsystem */ +#define DRM_FIMC_DEVICE BIT(2) /* devices shared with V4L2 subsystem */ #define DRV_PTR(drv, cond) (IS_ENABLED(cond) ? &drv : NULL) @@ -187,16 +185,16 @@ struct exynos_drm_driver_info { static struct exynos_drm_driver_info exynos_drm_drivers[] = { { DRV_PTR(fimd_driver, CONFIG_DRM_EXYNOS_FIMD), - DRM_COMPONENT_DRIVER | DRM_DMA_DEVICE + DRM_COMPONENT_DRIVER }, { DRV_PTR(exynos5433_decon_driver, CONFIG_DRM_EXYNOS5433_DECON), - DRM_COMPONENT_DRIVER | DRM_DMA_DEVICE + DRM_COMPONENT_DRIVER }, { DRV_PTR(decon_driver, CONFIG_DRM_EXYNOS7_DECON), - DRM_COMPONENT_DRIVER | DRM_DMA_DEVICE + DRM_COMPONENT_DRIVER }, { DRV_PTR(mixer_driver, CONFIG_DRM_EXYNOS_MIXER), - DRM_COMPONENT_DRIVER | DRM_DMA_DEVICE + DRM_COMPONENT_DRIVER }, { DRV_PTR(mic_driver, CONFIG_DRM_EXYNOS_MIC), DRM_COMPONENT_DRIVER @@ -267,27 +265,6 @@ static struct component_match *exynos_drm_match_add(struct device *dev) return match ?: ERR_PTR(-ENODEV); } -static struct device *exynos_drm_get_dma_device(void) -{ - int i; - - for (i = 0; i < ARRAY_SIZE(exynos_drm_drivers); ++i) { - struct exynos_drm_driver_info *info = &exynos_drm_drivers[i]; - struct device *dev; - - if (!info->driver || !(info->flags & DRM_DMA_DEVICE)) - continue; - - while ((dev = bus_find_device(&platform_bus_type, NULL, - &info->driver->driver, - (void *)platform_bus_type.match))) { - put_device(dev); - return dev; - } - } - return NULL; -} - static int exynos_drm_bind(struct device *dev) { struct exynos_drm_private *private; @@ -312,23 +289,6 @@ static int exynos_drm_bind(struct device *dev) dev_set_drvdata(dev, drm); drm->dev_private = (void *)private; - /* the first real CRTC device is used for all dma mapping operations */ - private->dma_dev = exynos_drm_get_dma_device(); - if (!private->dma_dev) { - DRM_ERROR("no device found for DMA mapping operations.\n"); - ret = -ENODEV; - goto err_free_private; - } - DRM_INFO("Exynos DRM: using %s device for DMA mapping operations\n", - dev_name(private->dma_dev)); - - /* create common IOMMU mapping for all devices attached to Exynos DRM */ - ret = drm_create_iommu_mapping(drm); - if (ret < 0) { - DRM_ERROR("failed to create iommu mapping.\n"); - goto err_free_private; - } - drm_mode_config_init(drm); exynos_drm_mode_config_init(drm); @@ -385,8 +345,7 @@ err_unbind_all: component_unbind_all(drm->dev, drm); err_mode_config_cleanup: drm_mode_config_cleanup(drm); - drm_release_iommu_mapping(drm); -err_free_private: + exynos_drm_cleanup_dma(drm); kfree(private); err_free_drm: drm_dev_put(drm); @@ -405,7 +364,7 @@ static void exynos_drm_unbind(struct device *dev) component_unbind_all(drm->dev, drm); drm_mode_config_cleanup(drm); - drm_release_iommu_mapping(drm); + exynos_drm_cleanup_dma(drm); kfree(drm->dev_private); drm->dev_private = NULL; diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h index 5e61e707f955..71eb240bc1f4 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.h +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h @@ -214,6 +214,17 @@ static inline struct device *to_dma_dev(struct drm_device *dev) return priv->dma_dev; } +static inline bool is_drm_iommu_supported(struct drm_device *drm_dev) +{ + struct exynos_drm_private *priv = drm_dev->dev_private; + + return priv->mapping ? true : false; +} + +int exynos_drm_register_dma(struct drm_device *drm, struct device *dev); +void exynos_drm_unregister_dma(struct drm_device *drm, struct device *dev); +void exynos_drm_cleanup_dma(struct drm_device *drm); + #ifdef CONFIG_DRM_EXYNOS_DPI struct drm_encoder *exynos_dpi_probe(struct device *dev); int exynos_dpi_remove(struct drm_encoder *encoder); diff --git a/drivers/gpu/drm/exynos/exynos_drm_fb.c b/drivers/gpu/drm/exynos/exynos_drm_fb.c index 9f52382e19ee..31eb538a44ae 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fb.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fb.c @@ -24,7 +24,6 @@ #include "exynos_drm_drv.h" #include "exynos_drm_fb.h" #include "exynos_drm_fbdev.h" -#include "exynos_drm_iommu.h" #include "exynos_drm_crtc.h" static int check_fb_gem_memory_type(struct drm_device *drm_dev, diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c index 01d182289efa..ce9604ca8041 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c @@ -23,7 +23,6 @@ #include "exynos_drm_drv.h" #include "exynos_drm_fb.h" #include "exynos_drm_fbdev.h" -#include "exynos_drm_iommu.h" #define MAX_CONNECTOR 4 #define PREFERRED_BPP 32 diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c index e8d0670bb5f8..90dfea0aec4d 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c @@ -25,7 +25,6 @@ #include <drm/exynos_drm.h> #include "regs-fimc.h" #include "exynos_drm_drv.h" -#include "exynos_drm_iommu.h" #include "exynos_drm_ipp.h" /* @@ -1129,7 +1128,7 @@ static int fimc_bind(struct device *dev, struct device *master, void *data) struct exynos_drm_ipp *ipp = &ctx->ipp; ctx->drm_dev = drm_dev; - drm_iommu_attach_device(drm_dev, dev); + exynos_drm_register_dma(drm_dev, dev); exynos_drm_ipp_register(drm_dev, ipp, &ipp_funcs, DRM_EXYNOS_IPP_CAP_CROP | DRM_EXYNOS_IPP_CAP_ROTATE | @@ -1149,7 +1148,7 @@ static void fimc_unbind(struct device *dev, struct device *master, struct exynos_drm_ipp *ipp = &ctx->ipp; exynos_drm_ipp_unregister(drm_dev, ipp); - drm_iommu_detach_device(drm_dev, dev); + exynos_drm_unregister_dma(drm_dev, dev); } static const struct component_ops fimc_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c index b7f56935a46b..786a8ee6f10f 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c @@ -32,7 +32,6 @@ #include "exynos_drm_fb.h" #include "exynos_drm_crtc.h" #include "exynos_drm_plane.h" -#include "exynos_drm_iommu.h" /* * FIMD stands for Fully Interactive Mobile Display and @@ -229,6 +228,21 @@ static const uint32_t fimd_formats[] = { DRM_FORMAT_ARGB8888, }; +static const unsigned int capabilities[WINDOWS_NR] = { + 0, + EXYNOS_DRM_PLANE_CAP_WIN_BLEND | EXYNOS_DRM_PLANE_CAP_PIX_BLEND, + EXYNOS_DRM_PLANE_CAP_WIN_BLEND | EXYNOS_DRM_PLANE_CAP_PIX_BLEND, + EXYNOS_DRM_PLANE_CAP_WIN_BLEND | EXYNOS_DRM_PLANE_CAP_PIX_BLEND, + EXYNOS_DRM_PLANE_CAP_WIN_BLEND | EXYNOS_DRM_PLANE_CAP_PIX_BLEND, +}; + +static inline void fimd_set_bits(struct fimd_context *ctx, u32 reg, u32 mask, + u32 val) +{ + val = (val & mask) | (readl(ctx->regs + reg) & ~mask); + writel(val, ctx->regs + reg); +} + static int fimd_enable_vblank(struct exynos_drm_crtc *crtc) { struct fimd_context *ctx = crtc->ctx; @@ -552,13 +566,88 @@ static void fimd_commit(struct exynos_drm_crtc *crtc) writel(val, ctx->regs + VIDCON0); } +static void fimd_win_set_bldeq(struct fimd_context *ctx, unsigned int win, + unsigned int alpha, unsigned int pixel_alpha) +{ + u32 mask = BLENDEQ_A_FUNC_F(0xf) | BLENDEQ_B_FUNC_F(0xf); + u32 val = 0; + + switch (pixel_alpha) { + case DRM_MODE_BLEND_PIXEL_NONE: + case DRM_MODE_BLEND_COVERAGE: + val |= BLENDEQ_A_FUNC_F(BLENDEQ_ALPHA_A); + val |= BLENDEQ_B_FUNC_F(BLENDEQ_ONE_MINUS_ALPHA_A); + break; + case DRM_MODE_BLEND_PREMULTI: + default: + if (alpha != DRM_BLEND_ALPHA_OPAQUE) { + val |= BLENDEQ_A_FUNC_F(BLENDEQ_ALPHA0); + val |= BLENDEQ_B_FUNC_F(BLENDEQ_ONE_MINUS_ALPHA_A); + } else { + val |= BLENDEQ_A_FUNC_F(BLENDEQ_ONE); + val |= BLENDEQ_B_FUNC_F(BLENDEQ_ONE_MINUS_ALPHA_A); + } + break; + } + fimd_set_bits(ctx, BLENDEQx(win), mask, val); +} -static void fimd_win_set_pixfmt(struct fimd_context *ctx, unsigned int win, - uint32_t pixel_format, int width) +static void fimd_win_set_bldmod(struct fimd_context *ctx, unsigned int win, + unsigned int alpha, unsigned int pixel_alpha) { - unsigned long val; + u32 win_alpha_l = (alpha >> 8) & 0xf; + u32 win_alpha_h = alpha >> 12; + u32 val = 0; - val = WINCONx_ENWIN; + switch (pixel_alpha) { + case DRM_MODE_BLEND_PIXEL_NONE: + break; + case DRM_MODE_BLEND_COVERAGE: + case DRM_MODE_BLEND_PREMULTI: + default: + val |= WINCON1_ALPHA_SEL; + val |= WINCON1_BLD_PIX; + val |= WINCON1_ALPHA_MUL; + break; + } + fimd_set_bits(ctx, WINCON(win), WINCONx_BLEND_MODE_MASK, val); + + /* OSD alpha */ + val = VIDISD14C_ALPHA0_R(win_alpha_h) | + VIDISD14C_ALPHA0_G(win_alpha_h) | + VIDISD14C_ALPHA0_B(win_alpha_h) | + VIDISD14C_ALPHA1_R(0x0) | + VIDISD14C_ALPHA1_G(0x0) | + VIDISD14C_ALPHA1_B(0x0); + writel(val, ctx->regs + VIDOSD_C(win)); + + val = VIDW_ALPHA_R(win_alpha_l) | VIDW_ALPHA_G(win_alpha_l) | + VIDW_ALPHA_B(win_alpha_l); + writel(val, ctx->regs + VIDWnALPHA0(win)); + + val = VIDW_ALPHA_R(0x0) | VIDW_ALPHA_G(0x0) | + VIDW_ALPHA_B(0x0); + writel(val, ctx->regs + VIDWnALPHA1(win)); + + fimd_set_bits(ctx, BLENDCON, BLENDCON_NEW_MASK, + BLENDCON_NEW_8BIT_ALPHA_VALUE); +} + +static void fimd_win_set_pixfmt(struct fimd_context *ctx, unsigned int win, + struct drm_framebuffer *fb, int width) +{ + struct exynos_drm_plane plane = ctx->planes[win]; + struct exynos_drm_plane_state *state = + to_exynos_plane_state(plane.base.state); + uint32_t pixel_format = fb->format->format; + unsigned int alpha = state->base.alpha; + u32 val = WINCONx_ENWIN; + unsigned int pixel_alpha; + + if (fb->format->has_alpha) + pixel_alpha = state->base.pixel_blend_mode; + else + pixel_alpha = DRM_MODE_BLEND_PIXEL_NONE; /* * In case of s3c64xx, window 0 doesn't support alpha channel. @@ -592,8 +681,7 @@ static void fimd_win_set_pixfmt(struct fimd_context *ctx, unsigned int win, break; case DRM_FORMAT_ARGB8888: default: - val |= WINCON1_BPPMODE_25BPP_A1888 - | WINCON1_BLD_PIX | WINCON1_ALPHA_SEL; + val |= WINCON1_BPPMODE_25BPP_A1888; val |= WINCONx_WSWP; val |= WINCONx_BURSTLEN_16WORD; break; @@ -611,25 +699,12 @@ static void fimd_win_set_pixfmt(struct fimd_context *ctx, unsigned int win, val &= ~WINCONx_BURSTLEN_MASK; val |= WINCONx_BURSTLEN_4WORD; } - - writel(val, ctx->regs + WINCON(win)); + fimd_set_bits(ctx, WINCON(win), ~WINCONx_BLEND_MODE_MASK, val); /* hardware window 0 doesn't support alpha channel. */ if (win != 0) { - /* OSD alpha */ - val = VIDISD14C_ALPHA0_R(0xf) | - VIDISD14C_ALPHA0_G(0xf) | - VIDISD14C_ALPHA0_B(0xf) | - VIDISD14C_ALPHA1_R(0xf) | - VIDISD14C_ALPHA1_G(0xf) | - VIDISD14C_ALPHA1_B(0xf); - - writel(val, ctx->regs + VIDOSD_C(win)); - - val = VIDW_ALPHA_R(0xf) | VIDW_ALPHA_G(0xf) | - VIDW_ALPHA_G(0xf); - writel(val, ctx->regs + VIDWnALPHA0(win)); - writel(val, ctx->regs + VIDWnALPHA1(win)); + fimd_win_set_bldmod(ctx, win, alpha, pixel_alpha); + fimd_win_set_bldeq(ctx, win, alpha, pixel_alpha); } } @@ -786,7 +861,7 @@ static void fimd_update_plane(struct exynos_drm_crtc *crtc, DRM_DEBUG_KMS("osd size = 0x%x\n", (unsigned int)val); } - fimd_win_set_pixfmt(ctx, win, fb->format->format, state->src.w); + fimd_win_set_pixfmt(ctx, win, fb, state->src.w); /* hardware window 0 doesn't support color key. */ if (win != 0) @@ -988,6 +1063,7 @@ static int fimd_bind(struct device *dev, struct device *master, void *data) ctx->configs[i].num_pixel_formats = ARRAY_SIZE(fimd_formats); ctx->configs[i].zpos = i; ctx->configs[i].type = fimd_win_types[i]; + ctx->configs[i].capabilities = capabilities[i]; ret = exynos_plane_init(drm_dev, &ctx->planes[i], i, &ctx->configs[i]); if (ret) @@ -1011,7 +1087,7 @@ static int fimd_bind(struct device *dev, struct device *master, void *data) if (is_drm_iommu_supported(drm_dev)) fimd_clear_channels(ctx->crtc); - return drm_iommu_attach_device(drm_dev, dev); + return exynos_drm_register_dma(drm_dev, dev); } static void fimd_unbind(struct device *dev, struct device *master, @@ -1021,7 +1097,7 @@ static void fimd_unbind(struct device *dev, struct device *master, fimd_disable(ctx->crtc); - drm_iommu_detach_device(ctx->drm_dev, ctx->dev); + exynos_drm_unregister_dma(ctx->drm_dev, ctx->dev); if (ctx->encoder) exynos_dpi_remove(ctx->encoder); diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.c b/drivers/gpu/drm/exynos/exynos_drm_g2d.c index f2481a2014bb..24c536d6d9cf 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_g2d.c +++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.c @@ -25,7 +25,6 @@ #include "exynos_drm_drv.h" #include "exynos_drm_g2d.h" #include "exynos_drm_gem.h" -#include "exynos_drm_iommu.h" #define G2D_HW_MAJOR_VER 4 #define G2D_HW_MINOR_VER 1 @@ -1405,7 +1404,7 @@ static int g2d_bind(struct device *dev, struct device *master, void *data) return ret; } - ret = drm_iommu_attach_device(drm_dev, dev); + ret = exynos_drm_register_dma(drm_dev, dev); if (ret < 0) { dev_err(dev, "failed to enable iommu.\n"); g2d_fini_cmdlist(g2d); @@ -1430,7 +1429,7 @@ static void g2d_unbind(struct device *dev, struct device *master, void *data) priv->g2d_dev = NULL; cancel_work_sync(&g2d->runqueue_work); - drm_iommu_detach_device(g2d->drm_dev, dev); + exynos_drm_unregister_dma(g2d->drm_dev, dev); } static const struct component_ops g2d_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index 34ace85feb68..df66c383a877 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -19,7 +19,6 @@ #include "exynos_drm_drv.h" #include "exynos_drm_gem.h" -#include "exynos_drm_iommu.h" static int exynos_drm_alloc_buf(struct exynos_drm_gem *exynos_gem) { diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c index ce15d46bfce8..f048d97fe9e2 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c @@ -24,7 +24,6 @@ #include <drm/exynos_drm.h> #include "regs-gsc.h" #include "exynos_drm_drv.h" -#include "exynos_drm_iommu.h" #include "exynos_drm_ipp.h" /* @@ -1170,7 +1169,7 @@ static int gsc_bind(struct device *dev, struct device *master, void *data) struct exynos_drm_ipp *ipp = &ctx->ipp; ctx->drm_dev = drm_dev; - drm_iommu_attach_device(drm_dev, dev); + exynos_drm_register_dma(drm_dev, dev); exynos_drm_ipp_register(drm_dev, ipp, &ipp_funcs, DRM_EXYNOS_IPP_CAP_CROP | DRM_EXYNOS_IPP_CAP_ROTATE | @@ -1190,7 +1189,7 @@ static void gsc_unbind(struct device *dev, struct device *master, struct exynos_drm_ipp *ipp = &ctx->ipp; exynos_drm_ipp_unregister(drm_dev, ipp); - drm_iommu_detach_device(drm_dev, dev); + exynos_drm_unregister_dma(drm_dev, dev); } static const struct component_ops gsc_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos_drm_iommu.c b/drivers/gpu/drm/exynos/exynos_drm_iommu.c deleted file mode 100644 index 0f373702414e..000000000000 --- a/drivers/gpu/drm/exynos/exynos_drm_iommu.c +++ /dev/null @@ -1,111 +0,0 @@ -/* exynos_drm_iommu.c - * - * Copyright (c) 2012 Samsung Electronics Co., Ltd. - * Author: Inki Dae <inki.dae@samsung.com> - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License as published by the - * Free Software Foundation; either version 2 of the License, or (at your - * option) any later version. - */ - -#include <drm/drmP.h> -#include <drm/exynos_drm.h> - -#include <linux/dma-mapping.h> -#include <linux/iommu.h> - -#include "exynos_drm_drv.h" -#include "exynos_drm_iommu.h" - -static inline int configure_dma_max_seg_size(struct device *dev) -{ - if (!dev->dma_parms) - dev->dma_parms = kzalloc(sizeof(*dev->dma_parms), GFP_KERNEL); - if (!dev->dma_parms) - return -ENOMEM; - - dma_set_max_seg_size(dev, DMA_BIT_MASK(32)); - return 0; -} - -static inline void clear_dma_max_seg_size(struct device *dev) -{ - kfree(dev->dma_parms); - dev->dma_parms = NULL; -} - -/* - * drm_create_iommu_mapping - create a mapping structure - * - * @drm_dev: DRM device - */ -int drm_create_iommu_mapping(struct drm_device *drm_dev) -{ - struct exynos_drm_private *priv = drm_dev->dev_private; - - return __exynos_iommu_create_mapping(priv, EXYNOS_DEV_ADDR_START, - EXYNOS_DEV_ADDR_SIZE); -} - -/* - * drm_release_iommu_mapping - release iommu mapping structure - * - * @drm_dev: DRM device - */ -void drm_release_iommu_mapping(struct drm_device *drm_dev) -{ - struct exynos_drm_private *priv = drm_dev->dev_private; - - __exynos_iommu_release_mapping(priv); -} - -/* - * drm_iommu_attach_device- attach device to iommu mapping - * - * @drm_dev: DRM device - * @subdrv_dev: device to be attach - * - * This function should be called by sub drivers to attach it to iommu - * mapping. - */ -int drm_iommu_attach_device(struct drm_device *drm_dev, - struct device *subdrv_dev) -{ - struct exynos_drm_private *priv = drm_dev->dev_private; - int ret; - - if (get_dma_ops(priv->dma_dev) != get_dma_ops(subdrv_dev)) { - DRM_ERROR("Device %s lacks support for IOMMU\n", - dev_name(subdrv_dev)); - return -EINVAL; - } - - ret = configure_dma_max_seg_size(subdrv_dev); - if (ret) - return ret; - - ret = __exynos_iommu_attach(priv, subdrv_dev); - if (ret) - clear_dma_max_seg_size(subdrv_dev); - - return 0; -} - -/* - * drm_iommu_detach_device -detach device address space mapping from device - * - * @drm_dev: DRM device - * @subdrv_dev: device to be detached - * - * This function should be called by sub drivers to detach it from iommu - * mapping - */ -void drm_iommu_detach_device(struct drm_device *drm_dev, - struct device *subdrv_dev) -{ - struct exynos_drm_private *priv = drm_dev->dev_private; - - __exynos_iommu_detach(priv, subdrv_dev); - clear_dma_max_seg_size(subdrv_dev); -} diff --git a/drivers/gpu/drm/exynos/exynos_drm_iommu.h b/drivers/gpu/drm/exynos/exynos_drm_iommu.h deleted file mode 100644 index 797d9ee5f15a..000000000000 --- a/drivers/gpu/drm/exynos/exynos_drm_iommu.h +++ /dev/null @@ -1,134 +0,0 @@ -/* exynos_drm_iommu.h - * - * Copyright (c) 2012 Samsung Electronics Co., Ltd. - * Authoer: Inki Dae <inki.dae@samsung.com> - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of the GNU General Public License as published by the - * Free Software Foundation; either version 2 of the License, or (at your - * option) any later version. - */ - -#ifndef _EXYNOS_DRM_IOMMU_H_ -#define _EXYNOS_DRM_IOMMU_H_ - -#define EXYNOS_DEV_ADDR_START 0x20000000 -#define EXYNOS_DEV_ADDR_SIZE 0x40000000 - -#ifdef CONFIG_DRM_EXYNOS_IOMMU - -#if defined(CONFIG_ARM_DMA_USE_IOMMU) -#include <asm/dma-iommu.h> - -static inline int __exynos_iommu_create_mapping(struct exynos_drm_private *priv, - unsigned long start, unsigned long size) -{ - priv->mapping = arm_iommu_create_mapping(&platform_bus_type, start, - size); - return IS_ERR(priv->mapping); -} - -static inline void -__exynos_iommu_release_mapping(struct exynos_drm_private *priv) -{ - arm_iommu_release_mapping(priv->mapping); -} - -static inline int __exynos_iommu_attach(struct exynos_drm_private *priv, - struct device *dev) -{ - if (dev->archdata.mapping) - arm_iommu_detach_device(dev); - - return arm_iommu_attach_device(dev, priv->mapping); -} - -static inline void __exynos_iommu_detach(struct exynos_drm_private *priv, - struct device *dev) -{ - arm_iommu_detach_device(dev); -} - -#elif defined(CONFIG_IOMMU_DMA) -#include <linux/dma-iommu.h> - -static inline int __exynos_iommu_create_mapping(struct exynos_drm_private *priv, - unsigned long start, unsigned long size) -{ - priv->mapping = iommu_get_domain_for_dev(priv->dma_dev); - return 0; -} - -static inline void __exynos_iommu_release_mapping(struct exynos_drm_private *priv) -{ - priv->mapping = NULL; -} - -static inline int __exynos_iommu_attach(struct exynos_drm_private *priv, - struct device *dev) -{ - struct iommu_domain *domain = priv->mapping; - - if (dev != priv->dma_dev) - return iommu_attach_device(domain, dev); - return 0; -} - -static inline void __exynos_iommu_detach(struct exynos_drm_private *priv, - struct device *dev) -{ - struct iommu_domain *domain = priv->mapping; - - if (dev != priv->dma_dev) - iommu_detach_device(domain, dev); -} -#else -#error Unsupported architecture and IOMMU/DMA-mapping glue code -#endif - -int drm_create_iommu_mapping(struct drm_device *drm_dev); - -void drm_release_iommu_mapping(struct drm_device *drm_dev); - -int drm_iommu_attach_device(struct drm_device *drm_dev, - struct device *subdrv_dev); - -void drm_iommu_detach_device(struct drm_device *dev_dev, - struct device *subdrv_dev); - -static inline bool is_drm_iommu_supported(struct drm_device *drm_dev) -{ - struct exynos_drm_private *priv = drm_dev->dev_private; - - return priv->mapping ? true : false; -} - -#else - -static inline int drm_create_iommu_mapping(struct drm_device *drm_dev) -{ - return 0; -} - -static inline void drm_release_iommu_mapping(struct drm_device *drm_dev) -{ -} - -static inline int drm_iommu_attach_device(struct drm_device *drm_dev, - struct device *subdrv_dev) -{ - return 0; -} - -static inline void drm_iommu_detach_device(struct drm_device *drm_dev, - struct device *subdrv_dev) -{ -} - -static inline bool is_drm_iommu_supported(struct drm_device *drm_dev) -{ - return false; -} - -#endif -#endif diff --git a/drivers/gpu/drm/exynos/exynos_drm_rotator.c b/drivers/gpu/drm/exynos/exynos_drm_rotator.c index a820a68429b9..8d67b2a54be3 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_rotator.c +++ b/drivers/gpu/drm/exynos/exynos_drm_rotator.c @@ -23,7 +23,6 @@ #include <drm/exynos_drm.h> #include "regs-rotator.h" #include "exynos_drm_drv.h" -#include "exynos_drm_iommu.h" #include "exynos_drm_ipp.h" /* @@ -244,7 +243,7 @@ static int rotator_bind(struct device *dev, struct device *master, void *data) struct exynos_drm_ipp *ipp = &rot->ipp; rot->drm_dev = drm_dev; - drm_iommu_attach_device(drm_dev, dev); + exynos_drm_register_dma(drm_dev, dev); exynos_drm_ipp_register(drm_dev, ipp, &ipp_funcs, DRM_EXYNOS_IPP_CAP_CROP | DRM_EXYNOS_IPP_CAP_ROTATE, @@ -263,7 +262,7 @@ static void rotator_unbind(struct device *dev, struct device *master, struct exynos_drm_ipp *ipp = &rot->ipp; exynos_drm_ipp_unregister(drm_dev, ipp); - drm_iommu_detach_device(rot->drm_dev, rot->dev); + exynos_drm_unregister_dma(rot->drm_dev, rot->dev); } static const struct component_ops rotator_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos_drm_scaler.c b/drivers/gpu/drm/exynos/exynos_drm_scaler.c index cd66774e817d..71270efa64f3 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_scaler.c +++ b/drivers/gpu/drm/exynos/exynos_drm_scaler.c @@ -23,7 +23,6 @@ #include "regs-scaler.h" #include "exynos_drm_fb.h" #include "exynos_drm_drv.h" -#include "exynos_drm_iommu.h" #include "exynos_drm_ipp.h" #define scaler_read(offset) readl(scaler->regs + (offset)) @@ -452,7 +451,7 @@ static int scaler_bind(struct device *dev, struct device *master, void *data) struct exynos_drm_ipp *ipp = &scaler->ipp; scaler->drm_dev = drm_dev; - drm_iommu_attach_device(drm_dev, dev); + exynos_drm_register_dma(drm_dev, dev); exynos_drm_ipp_register(drm_dev, ipp, &ipp_funcs, DRM_EXYNOS_IPP_CAP_CROP | DRM_EXYNOS_IPP_CAP_ROTATE | @@ -473,7 +472,7 @@ static void scaler_unbind(struct device *dev, struct device *master, struct exynos_drm_ipp *ipp = &scaler->ipp; exynos_drm_ipp_unregister(drm_dev, ipp); - drm_iommu_detach_device(scaler->drm_dev, scaler->dev); + exynos_drm_unregister_dma(scaler->drm_dev, scaler->dev); } static const struct component_ops scaler_component_ops = { diff --git a/drivers/gpu/drm/exynos/exynos_mixer.c b/drivers/gpu/drm/exynos/exynos_mixer.c index e3a4ecbc503b..0573eab0e190 100644 --- a/drivers/gpu/drm/exynos/exynos_mixer.c +++ b/drivers/gpu/drm/exynos/exynos_mixer.c @@ -40,7 +40,6 @@ #include "exynos_drm_crtc.h" #include "exynos_drm_fb.h" #include "exynos_drm_plane.h" -#include "exynos_drm_iommu.h" #define MIXER_WIN_NR 3 #define VP_DEFAULT_WIN 2 @@ -381,19 +380,16 @@ static void mixer_cfg_scan(struct mixer_context *ctx, int width, int height) mixer_reg_writemask(ctx, MXR_CFG, val, MXR_CFG_SCAN_MASK); } -static void mixer_cfg_rgb_fmt(struct mixer_context *ctx, unsigned int height) +static void mixer_cfg_rgb_fmt(struct mixer_context *ctx, struct drm_display_mode *mode) { + enum hdmi_quantization_range range = drm_default_rgb_quant_range(mode); u32 val; - switch (height) { - case 480: - case 576: - val = MXR_CFG_RGB601_0_255; - break; - case 720: - case 1080: - default: - val = MXR_CFG_RGB709_16_235; + if (mode->vdisplay < 720) { + val = MXR_CFG_RGB601; + } else { + val = MXR_CFG_RGB709; + /* Configure the BT.709 CSC matrix for full range RGB. */ mixer_reg_write(ctx, MXR_CM_COEFF_Y, MXR_CSC_CT( 0.184, 0.614, 0.063) | @@ -402,9 +398,13 @@ static void mixer_cfg_rgb_fmt(struct mixer_context *ctx, unsigned int height) MXR_CSC_CT(-0.102, -0.338, 0.440)); mixer_reg_write(ctx, MXR_CM_COEFF_CR, MXR_CSC_CT( 0.440, -0.399, -0.040)); - break; } + if (range == HDMI_QUANTIZATION_RANGE_FULL) + val |= MXR_CFG_QUANT_RANGE_FULL; + else + val |= MXR_CFG_QUANT_RANGE_LIMITED; + mixer_reg_writemask(ctx, MXR_CFG, val, MXR_CFG_RGB_FMT_MASK); } @@ -461,7 +461,7 @@ static void mixer_commit(struct mixer_context *ctx) struct drm_display_mode *mode = &ctx->crtc->base.state->adjusted_mode; mixer_cfg_scan(ctx, mode->hdisplay, mode->vdisplay); - mixer_cfg_rgb_fmt(ctx, mode->vdisplay); + mixer_cfg_rgb_fmt(ctx, mode); mixer_run(ctx); } @@ -878,12 +878,12 @@ static int mixer_initialize(struct mixer_context *mixer_ctx, } } - return drm_iommu_attach_device(drm_dev, mixer_ctx->dev); + return exynos_drm_register_dma(drm_dev, mixer_ctx->dev); } static void mixer_ctx_remove(struct mixer_context *mixer_ctx) { - drm_iommu_detach_device(mixer_ctx->drm_dev, mixer_ctx->dev); + exynos_drm_unregister_dma(mixer_ctx->drm_dev, mixer_ctx->dev); } static int mixer_enable_vblank(struct exynos_drm_crtc *crtc) diff --git a/drivers/gpu/drm/exynos/regs-decon5433.h b/drivers/gpu/drm/exynos/regs-decon5433.h index 19ad9e47945e..63db6974bf14 100644 --- a/drivers/gpu/drm/exynos/regs-decon5433.h +++ b/drivers/gpu/drm/exynos/regs-decon5433.h @@ -104,6 +104,7 @@ #define WINCONx_BURSTLEN_16WORD (0x0 << 10) #define WINCONx_BURSTLEN_8WORD (0x1 << 10) #define WINCONx_BURSTLEN_4WORD (0x2 << 10) +#define WINCONx_ALPHA_MUL_F (1 << 7) #define WINCONx_BLD_PIX_F (1 << 6) #define WINCONx_BPPMODE_MASK (0xf << 2) #define WINCONx_BPPMODE_16BPP_565 (0x5 << 2) @@ -116,11 +117,15 @@ #define WINCONx_BPPMODE_16BPP_A4444 (0xe << 2) #define WINCONx_ALPHA_SEL_F (1 << 1) #define WINCONx_ENWIN_F (1 << 0) +#define WINCONx_BLEND_MODE_MASK (0xc2) /* SHADOWCON */ #define SHADOWCON_PROTECT_MASK GENMASK(14, 10) #define SHADOWCON_Wx_PROTECT(n) (1 << (10 + (n))) +/* VIDOSDxC */ +#define VIDOSDxC_ALPHA0_RGB_MASK (0xffffff) + /* VIDOSDxD */ #define VIDOSD_Wx_ALPHA_R_F(n) (((n) & 0xff) << 16) #define VIDOSD_Wx_ALPHA_G_F(n) (((n) & 0xff) << 8) @@ -206,4 +211,21 @@ #define CRCCTRL_CRCEN (0x1 << 0) #define CRCCTRL_MASK (0x7) +/* BLENDCON */ +#define BLEND_NEW (1 << 0) + +/* BLENDERQx */ +#define BLENDERQ_ZERO 0x0 +#define BLENDERQ_ONE 0x1 +#define BLENDERQ_ALPHA_A 0x2 +#define BLENDERQ_ONE_MINUS_ALPHA_A 0x3 +#define BLENDERQ_ALPHA0 0x6 +#define BLENDERQ_Q_FUNC_F(n) (n << 18) +#define BLENDERQ_P_FUNC_F(n) (n << 12) +#define BLENDERQ_B_FUNC_F(n) (n << 6) +#define BLENDERQ_A_FUNC_F(n) (n << 0) + +/* BLENDCON */ +#define BLEND_NEW (1 << 0) + #endif /* EXYNOS_REGS_DECON5433_H */ diff --git a/drivers/gpu/drm/exynos/regs-mixer.h b/drivers/gpu/drm/exynos/regs-mixer.h index d2b8194a07bf..5ff095b0c1b3 100644 --- a/drivers/gpu/drm/exynos/regs-mixer.h +++ b/drivers/gpu/drm/exynos/regs-mixer.h @@ -85,10 +85,11 @@ /* bits for MXR_CFG */ #define MXR_CFG_LAYER_UPDATE (1 << 31) #define MXR_CFG_LAYER_UPDATE_COUNT_MASK (3 << 29) -#define MXR_CFG_RGB601_0_255 (0 << 9) -#define MXR_CFG_RGB601_16_235 (1 << 9) -#define MXR_CFG_RGB709_0_255 (2 << 9) -#define MXR_CFG_RGB709_16_235 (3 << 9) +#define MXR_CFG_QUANT_RANGE_FULL (0 << 9) +#define MXR_CFG_QUANT_RANGE_LIMITED (1 << 9) +#define MXR_CFG_RGB601 (0 << 10) +#define MXR_CFG_RGB709 (1 << 10) + #define MXR_CFG_RGB_FMT_MASK 0x600 #define MXR_CFG_OUT_YUV444 (0 << 8) #define MXR_CFG_OUT_RGB888 (1 << 8) diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig index 33a458b7f1fc..148be8e1a090 100644 --- a/drivers/gpu/drm/i915/Kconfig +++ b/drivers/gpu/drm/i915/Kconfig @@ -131,5 +131,5 @@ config DRM_I915_GVT_KVMGT menu "drm/i915 Debugging" depends on DRM_I915 depends on EXPERT -source drivers/gpu/drm/i915/Kconfig.debug +source "drivers/gpu/drm/i915/Kconfig.debug" endmenu diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index 0ff878c994e2..19b5fe5016bf 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -157,14 +157,17 @@ i915-y += dvo_ch7017.o \ intel_sdvo.o \ intel_tv.o \ vlv_dsi.o \ - vlv_dsi_pll.o + vlv_dsi_pll.o \ + intel_vdsc.o # Post-mortem debug and GPU hang state capture i915-$(CONFIG_DRM_I915_CAPTURE_ERROR) += i915_gpu_error.o i915-$(CONFIG_DRM_I915_SELFTEST) += \ selftests/i915_random.o \ selftests/i915_selftest.o \ - selftests/igt_flush_test.o + selftests/igt_flush_test.o \ + selftests/igt_reset.o \ + selftests/igt_spinner.o # virtual gpu code i915-y += i915_vgpu.o diff --git a/drivers/gpu/drm/i915/gvt/aperture_gm.c b/drivers/gpu/drm/i915/gvt/aperture_gm.c index fe754022e356..359d37d5c958 100644 --- a/drivers/gpu/drm/i915/gvt/aperture_gm.c +++ b/drivers/gpu/drm/i915/gvt/aperture_gm.c @@ -61,10 +61,12 @@ static int alloc_gm(struct intel_vgpu *vgpu, bool high_gm) } mutex_lock(&dev_priv->drm.struct_mutex); + mmio_hw_access_pre(dev_priv); ret = i915_gem_gtt_insert(&dev_priv->ggtt.vm, node, size, I915_GTT_PAGE_SIZE, I915_COLOR_UNEVICTABLE, start, end, flags); + mmio_hw_access_post(dev_priv); mutex_unlock(&dev_priv->drm.struct_mutex); if (ret) gvt_err("fail to alloc %s gm space from host\n", diff --git a/drivers/gpu/drm/i915/gvt/cmd_parser.c b/drivers/gpu/drm/i915/gvt/cmd_parser.c index 77edbfcb0f75..77ae634eb11c 100644 --- a/drivers/gpu/drm/i915/gvt/cmd_parser.c +++ b/drivers/gpu/drm/i915/gvt/cmd_parser.c @@ -1900,11 +1900,11 @@ static struct cmd_info cmd_info[] = { {"MI_URB_CLEAR", OP_MI_URB_CLEAR, F_LEN_VAR, R_RCS, D_ALL, 0, 8, NULL}, - {"ME_SEMAPHORE_SIGNAL", OP_MI_SEMAPHORE_SIGNAL, F_LEN_VAR, R_ALL, + {"MI_SEMAPHORE_SIGNAL", OP_MI_SEMAPHORE_SIGNAL, F_LEN_VAR, R_ALL, D_BDW_PLUS, 0, 8, NULL}, - {"ME_SEMAPHORE_WAIT", OP_MI_SEMAPHORE_WAIT, F_LEN_VAR, R_ALL, D_BDW_PLUS, - ADDR_FIX_1(2), 8, cmd_handler_mi_semaphore_wait}, + {"MI_SEMAPHORE_WAIT", OP_MI_SEMAPHORE_WAIT, F_LEN_VAR, R_ALL, + D_BDW_PLUS, ADDR_FIX_1(2), 8, cmd_handler_mi_semaphore_wait}, {"MI_STORE_DATA_IMM", OP_MI_STORE_DATA_IMM, F_LEN_VAR, R_ALL, D_BDW_PLUS, ADDR_FIX_1(1), 10, cmd_handler_mi_store_data_imm}, diff --git a/drivers/gpu/drm/i915/gvt/fb_decoder.c b/drivers/gpu/drm/i915/gvt/fb_decoder.c index 481896fb712a..85e6736f0a32 100644 --- a/drivers/gpu/drm/i915/gvt/fb_decoder.c +++ b/drivers/gpu/drm/i915/gvt/fb_decoder.c @@ -235,7 +235,7 @@ int intel_vgpu_decode_primary_plane(struct intel_vgpu *vgpu, plane->bpp = skl_pixel_formats[fmt].bpp; plane->drm_format = skl_pixel_formats[fmt].drm_format; } else { - plane->tiled = !!(val & DISPPLANE_TILED); + plane->tiled = val & DISPPLANE_TILED; fmt = bdw_format_to_drm(val & DISPPLANE_PIXFORMAT_MASK); plane->bpp = bdw_pixel_formats[fmt].bpp; plane->drm_format = bdw_pixel_formats[fmt].drm_format; diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c index 58e166effa45..c7103dd2d8d5 100644 --- a/drivers/gpu/drm/i915/gvt/gtt.c +++ b/drivers/gpu/drm/i915/gvt/gtt.c @@ -2447,10 +2447,11 @@ static void intel_vgpu_destroy_all_ppgtt_mm(struct intel_vgpu *vgpu) static void intel_vgpu_destroy_ggtt_mm(struct intel_vgpu *vgpu) { - struct intel_gvt_partial_pte *pos; + struct intel_gvt_partial_pte *pos, *next; - list_for_each_entry(pos, - &vgpu->gtt.ggtt_mm->ggtt_mm.partial_pte_list, list) { + list_for_each_entry_safe(pos, next, + &vgpu->gtt.ggtt_mm->ggtt_mm.partial_pte_list, + list) { gvt_dbg_mm("partial PTE update on hold 0x%lx : 0x%llx\n", pos->offset, pos->data); kfree(pos); diff --git a/drivers/gpu/drm/i915/gvt/gvt.c b/drivers/gpu/drm/i915/gvt/gvt.c index 6ef5a7fc70df..733a2a0d0c30 100644 --- a/drivers/gpu/drm/i915/gvt/gvt.c +++ b/drivers/gpu/drm/i915/gvt/gvt.c @@ -437,7 +437,7 @@ int intel_gvt_init_device(struct drm_i915_private *dev_priv) ret = intel_gvt_debugfs_init(gvt); if (ret) - gvt_err("debugfs registeration failed, go on.\n"); + gvt_err("debugfs registration failed, go on.\n"); gvt_dbg_core("gvt device initialization is done\n"); dev_priv->gvt = gvt; diff --git a/drivers/gpu/drm/i915/gvt/gvt.h b/drivers/gpu/drm/i915/gvt/gvt.h index 31f6cdbe5c42..b4ab1dad0143 100644 --- a/drivers/gpu/drm/i915/gvt/gvt.h +++ b/drivers/gpu/drm/i915/gvt/gvt.h @@ -159,6 +159,10 @@ struct intel_vgpu_submission { struct kmem_cache *workloads; atomic_t running_workload_num; struct i915_gem_context *shadow_ctx; + union { + u64 i915_context_pml4; + u64 i915_context_pdps[GEN8_3LVL_PDPES]; + }; DECLARE_BITMAP(shadow_ctx_desc_updated, I915_NUM_ENGINES); DECLARE_BITMAP(tlb_handle_pending, I915_NUM_ENGINES); void *ring_scan_buffer[I915_NUM_ENGINES]; diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c index aa280bb07125..b5475c91e2ef 100644 --- a/drivers/gpu/drm/i915/gvt/handlers.c +++ b/drivers/gpu/drm/i915/gvt/handlers.c @@ -475,6 +475,7 @@ static i915_reg_t force_nonpriv_white_list[] = { _MMIO(0x7704), _MMIO(0x7708), _MMIO(0x770c), + _MMIO(0x83a8), _MMIO(0xb110), GEN8_L3SQCREG4,//_MMIO(0xb118) _MMIO(0xe100), diff --git a/drivers/gpu/drm/i915/gvt/interrupt.c b/drivers/gpu/drm/i915/gvt/interrupt.c index 5daa23ae566b..6b9d1354ff29 100644 --- a/drivers/gpu/drm/i915/gvt/interrupt.c +++ b/drivers/gpu/drm/i915/gvt/interrupt.c @@ -126,7 +126,7 @@ static const char * const irq_name[INTEL_GVT_EVENT_MAX] = { [FDI_RX_INTERRUPTS_TRANSCODER_C] = "FDI RX Interrupts Combined C", [AUDIO_CP_CHANGE_TRANSCODER_C] = "Audio CP Change Transcoder C", [AUDIO_CP_REQUEST_TRANSCODER_C] = "Audio CP Request Transcoder C", - [ERR_AND_DBG] = "South Error and Debug Interupts Combined", + [ERR_AND_DBG] = "South Error and Debug Interrupts Combined", [GMBUS] = "Gmbus", [SDVO_B_HOTPLUG] = "SDVO B hotplug", [CRT_HOTPLUG] = "CRT Hotplug", diff --git a/drivers/gpu/drm/i915/gvt/mmio_context.c b/drivers/gpu/drm/i915/gvt/mmio_context.c index 36a5147cd01e..d6e02c15ef97 100644 --- a/drivers/gpu/drm/i915/gvt/mmio_context.c +++ b/drivers/gpu/drm/i915/gvt/mmio_context.c @@ -158,6 +158,8 @@ static void load_render_mocs(struct drm_i915_private *dev_priv) int ring_id, i; for (ring_id = 0; ring_id < ARRAY_SIZE(regs); ring_id++) { + if (!HAS_ENGINE(dev_priv, ring_id)) + continue; offset.reg = regs[ring_id]; for (i = 0; i < GEN9_MOCS_SIZE; i++) { gen9_render_mocs.control_table[ring_id][i] = diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c index b8fbe3fabea3..1ad8c5e1455d 100644 --- a/drivers/gpu/drm/i915/gvt/scheduler.c +++ b/drivers/gpu/drm/i915/gvt/scheduler.c @@ -1079,6 +1079,21 @@ err: return ret; } +static void +i915_context_ppgtt_root_restore(struct intel_vgpu_submission *s) +{ + struct i915_hw_ppgtt *i915_ppgtt = s->shadow_ctx->ppgtt; + int i; + + if (i915_vm_is_48bit(&i915_ppgtt->vm)) + px_dma(&i915_ppgtt->pml4) = s->i915_context_pml4; + else { + for (i = 0; i < GEN8_3LVL_PDPES; i++) + px_dma(i915_ppgtt->pdp.page_directory[i]) = + s->i915_context_pdps[i]; + } +} + /** * intel_vgpu_clean_submission - free submission-related resource for vGPU * @vgpu: a vGPU @@ -1091,6 +1106,7 @@ void intel_vgpu_clean_submission(struct intel_vgpu *vgpu) struct intel_vgpu_submission *s = &vgpu->submission; intel_vgpu_select_submission_ops(vgpu, ALL_ENGINES, 0); + i915_context_ppgtt_root_restore(s); i915_gem_context_put(s->shadow_ctx); kmem_cache_destroy(s->workloads); } @@ -1116,6 +1132,21 @@ void intel_vgpu_reset_submission(struct intel_vgpu *vgpu, s->ops->reset(vgpu, engine_mask); } +static void +i915_context_ppgtt_root_save(struct intel_vgpu_submission *s) +{ + struct i915_hw_ppgtt *i915_ppgtt = s->shadow_ctx->ppgtt; + int i; + + if (i915_vm_is_48bit(&i915_ppgtt->vm)) + s->i915_context_pml4 = px_dma(&i915_ppgtt->pml4); + else { + for (i = 0; i < GEN8_3LVL_PDPES; i++) + s->i915_context_pdps[i] = + px_dma(i915_ppgtt->pdp.page_directory[i]); + } +} + /** * intel_vgpu_setup_submission - setup submission-related resource for vGPU * @vgpu: a vGPU @@ -1138,6 +1169,8 @@ int intel_vgpu_setup_submission(struct intel_vgpu *vgpu) if (IS_ERR(s->shadow_ctx)) return PTR_ERR(s->shadow_ctx); + i915_context_ppgtt_root_save(s); + bitmap_zero(s->shadow_ctx_desc_updated, I915_NUM_ENGINES); s->workloads = kmem_cache_create_usercopy("gvt-g_vgpu_workload", diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index a63d084c8e96..9bad6a32adae 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -943,30 +943,30 @@ static int i915_gem_fence_regs_info(struct seq_file *m, void *data) static ssize_t gpu_state_read(struct file *file, char __user *ubuf, size_t count, loff_t *pos) { - struct i915_gpu_state *error = file->private_data; - struct drm_i915_error_state_buf str; + struct i915_gpu_state *error; ssize_t ret; - loff_t tmp; + void *buf; + error = file->private_data; if (!error) return 0; - ret = i915_error_state_buf_init(&str, error->i915, count, *pos); - if (ret) - return ret; + /* Bounce buffer required because of kernfs __user API convenience. */ + buf = kmalloc(count, GFP_KERNEL); + if (!buf) + return -ENOMEM; - ret = i915_error_state_to_str(&str, error); - if (ret) + ret = i915_gpu_state_copy_to_buffer(error, buf, *pos, count); + if (ret <= 0) goto out; - tmp = 0; - ret = simple_read_from_buffer(ubuf, count, &tmp, str.buf, str.bytes); - if (ret < 0) - goto out; + if (!copy_to_user(ubuf, buf, ret)) + *pos += ret; + else + ret = -EFAULT; - *pos = str.start + ret; out: - i915_error_state_buf_release(&str); + kfree(buf); return ret; } @@ -3368,13 +3368,15 @@ static int i915_shared_dplls_info(struct seq_file *m, void *unused) static int i915_wa_registers(struct seq_file *m, void *unused) { - struct i915_workarounds *wa = &node_to_i915(m->private)->workarounds; - int i; + struct drm_i915_private *i915 = node_to_i915(m->private); + const struct i915_wa_list *wal = &i915->engine[RCS]->ctx_wa_list; + struct i915_wa *wa; + unsigned int i; - seq_printf(m, "Workarounds applied: %d\n", wa->count); - for (i = 0; i < wa->count; ++i) + seq_printf(m, "Workarounds applied: %u\n", wal->count); + for (i = 0, wa = wal->list; i < wal->count; i++, wa++) seq_printf(m, "0x%X: 0x%08X, mask: 0x%08X\n", - wa->reg[i].addr, wa->reg[i].value, wa->reg[i].mask); + i915_mmio_reg_offset(wa->reg), wa->val, wa->mask); return 0; } @@ -3434,31 +3436,32 @@ static int i915_ddb_info(struct seq_file *m, void *unused) { struct drm_i915_private *dev_priv = node_to_i915(m->private); struct drm_device *dev = &dev_priv->drm; - struct skl_ddb_allocation *ddb; struct skl_ddb_entry *entry; - enum pipe pipe; - int plane; + struct intel_crtc *crtc; if (INTEL_GEN(dev_priv) < 9) return -ENODEV; drm_modeset_lock_all(dev); - ddb = &dev_priv->wm.skl_hw.ddb; - seq_printf(m, "%-15s%8s%8s%8s\n", "", "Start", "End", "Size"); - for_each_pipe(dev_priv, pipe) { + for_each_intel_crtc(&dev_priv->drm, crtc) { + struct intel_crtc_state *crtc_state = + to_intel_crtc_state(crtc->base.state); + enum pipe pipe = crtc->pipe; + enum plane_id plane_id; + seq_printf(m, "Pipe %c\n", pipe_name(pipe)); - for_each_universal_plane(dev_priv, pipe, plane) { - entry = &ddb->plane[pipe][plane]; - seq_printf(m, " Plane%-8d%8u%8u%8u\n", plane + 1, + for_each_plane_id_on_crtc(crtc, plane_id) { + entry = &crtc_state->wm.skl.plane_ddb_y[plane_id]; + seq_printf(m, " Plane%-8d%8u%8u%8u\n", plane_id + 1, entry->start, entry->end, skl_ddb_entry_size(entry)); } - entry = &ddb->plane[pipe][PLANE_CURSOR]; + entry = &crtc_state->wm.skl.plane_ddb_y[PLANE_CURSOR]; seq_printf(m, " %-13s%8u%8u%8u\n", "Cursor", entry->start, entry->end, skl_ddb_entry_size(entry)); } @@ -4585,6 +4588,13 @@ static int i915_hpd_storm_ctl_show(struct seq_file *m, void *data) struct drm_i915_private *dev_priv = m->private; struct i915_hotplug *hotplug = &dev_priv->hotplug; + /* Synchronize with everything first in case there's been an HPD + * storm, but we haven't finished handling it in the kernel yet + */ + synchronize_irq(dev_priv->drm.irq); + flush_work(&dev_priv->hotplug.dig_port_work); + flush_work(&dev_priv->hotplug.hotplug_work); + seq_printf(m, "Threshold: %d\n", hotplug->hpd_storm_threshold); seq_printf(m, "Detected: %s\n", yesno(delayed_work_pending(&hotplug->reenable_work))); diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index b1d23c73c147..b310a897a4ad 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -53,6 +53,7 @@ #include "i915_vgpu.h" #include "intel_drv.h" #include "intel_uc.h" +#include "intel_workarounds.h" static struct drm_driver driver; @@ -287,7 +288,7 @@ static void intel_detect_pch(struct drm_i915_private *dev_priv) * Use PCH_NOP (PCH but no South Display) for PCH platforms without * display. */ - if (pch && INTEL_INFO(dev_priv)->num_pipes == 0) { + if (pch && !HAS_DISPLAY(dev_priv)) { DRM_DEBUG_KMS("Display disabled, reverting to NOP PCH\n"); dev_priv->pch_type = PCH_NOP; dev_priv->pch_id = 0; @@ -645,7 +646,7 @@ static int i915_load_modeset_init(struct drm_device *dev) if (i915_inject_load_failure()) return -ENODEV; - if (INTEL_INFO(dev_priv)->num_pipes) { + if (HAS_DISPLAY(dev_priv)) { ret = drm_vblank_init(&dev_priv->drm, INTEL_INFO(dev_priv)->num_pipes); if (ret) @@ -696,7 +697,7 @@ static int i915_load_modeset_init(struct drm_device *dev) intel_overlay_setup(dev_priv); - if (INTEL_INFO(dev_priv)->num_pipes == 0) + if (!HAS_DISPLAY(dev_priv)) return 0; ret = intel_fbdev_init(dev); @@ -868,6 +869,7 @@ static void intel_detect_preproduction_hw(struct drm_i915_private *dev_priv) pre |= IS_HSW_EARLY_SDV(dev_priv); pre |= IS_SKL_REVID(dev_priv, 0, SKL_REVID_F0); pre |= IS_BXT_REVID(dev_priv, 0, BXT_REVID_B_LAST); + pre |= IS_KBL_REVID(dev_priv, 0, KBL_REVID_A0); if (pre) { DRM_ERROR("This is a pre-production stepping. " @@ -1383,6 +1385,20 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv) } } + if (HAS_EXECLISTS(dev_priv)) { + /* + * Older GVT emulation depends upon intercepting CSB mmio, + * which we no longer use, preferring to use the HWSP cache + * instead. + */ + if (intel_vgpu_active(dev_priv) && + !intel_vgpu_has_hwsp_emulation(dev_priv)) { + i915_report_error(dev_priv, + "old vGPU host found, support for HWSP emulation required\n"); + return -ENXIO; + } + } + intel_sanitize_options(dev_priv); i915_perf_init(dev_priv); @@ -1452,6 +1468,7 @@ static int i915_driver_init_hw(struct drm_i915_private *dev_priv) intel_uncore_sanitize(dev_priv); + intel_gt_init_workarounds(dev_priv); i915_gem_load_init_fences(dev_priv); /* On the 945G/GM, the chipset reports the MSI capability on the @@ -1551,7 +1568,7 @@ static void i915_driver_register(struct drm_i915_private *dev_priv) } else DRM_ERROR("Failed to register driver for userspace access!\n"); - if (INTEL_INFO(dev_priv)->num_pipes) { + if (HAS_DISPLAY(dev_priv)) { /* Must be done after probing outputs */ intel_opregion_register(dev_priv); acpi_video_register(); @@ -1575,7 +1592,7 @@ static void i915_driver_register(struct drm_i915_private *dev_priv) * We need to coordinate the hotplugs with the asynchronous fbdev * configuration, for which we use the fbdev->async_cookie. */ - if (INTEL_INFO(dev_priv)->num_pipes) + if (HAS_DISPLAY(dev_priv)) drm_kms_helper_poll_init(dev); intel_power_domains_enable(dev_priv); diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 4064e49dbf70..b1c31967194b 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -53,6 +53,7 @@ #include <drm/drm_auth.h> #include <drm/drm_cache.h> #include <drm/drm_util.h> +#include <drm/drm_dsc.h> #include "i915_fixed.h" #include "i915_params.h" @@ -68,6 +69,7 @@ #include "intel_ringbuffer.h" #include "intel_uncore.h" #include "intel_wopcm.h" +#include "intel_workarounds.h" #include "intel_uc.h" #include "i915_gem.h" @@ -88,8 +90,8 @@ #define DRIVER_NAME "i915" #define DRIVER_DESC "Intel Graphics" -#define DRIVER_DATE "20181122" -#define DRIVER_TIMESTAMP 1542898187 +#define DRIVER_DATE "20181204" +#define DRIVER_TIMESTAMP 1543944377 /* Use I915_STATE_WARN(x) and I915_STATE_WARN_ON() (rather than WARN() and * WARN_ON()) for hw state sanity checks to check for unexpected conditions @@ -494,6 +496,7 @@ struct i915_psr { bool sink_support; bool prepared, enabled; struct intel_dp *dp; + enum pipe pipe; bool active; struct work_struct work; unsigned busy_frontbuffer_bits; @@ -504,6 +507,8 @@ struct i915_psr { u8 sink_sync_latency; ktime_t last_entry_attempt; ktime_t last_exit; + bool sink_not_reliable; + bool irq_aux_error; }; enum intel_pch { @@ -1093,9 +1098,6 @@ static inline bool skl_ddb_entry_equal(const struct skl_ddb_entry *e1, } struct skl_ddb_allocation { - /* packed/y */ - struct skl_ddb_entry plane[I915_MAX_PIPES][I915_MAX_PLANES]; - struct skl_ddb_entry uv_plane[I915_MAX_PIPES][I915_MAX_PLANES]; u8 enabled_slices; /* GEN11 has configurable 2 slices */ }; @@ -1188,20 +1190,6 @@ struct i915_frontbuffer_tracking { unsigned flip_bits; }; -struct i915_wa_reg { - u32 addr; - u32 value; - /* bitmask representing WA bits */ - u32 mask; -}; - -#define I915_MAX_WA_REGS 16 - -struct i915_workarounds { - struct i915_wa_reg reg[I915_MAX_WA_REGS]; - u32 count; -}; - struct i915_virtual_gpu { bool active; u32 caps; @@ -1651,7 +1639,7 @@ struct drm_i915_private { int dpio_phy_iosf_port[I915_NUM_PHYS_VLV]; - struct i915_workarounds workarounds; + struct i915_wa_list gt_wa_list; struct i915_frontbuffer_tracking fb_tracking; @@ -1995,6 +1983,8 @@ struct drm_i915_private { struct delayed_work idle_work; ktime_t last_init_time; + + struct i915_vma *scratch; } gt; /* perform PHY state sanity checks? */ @@ -2448,9 +2438,9 @@ intel_info(const struct drm_i915_private *dev_priv) ((sizes) & ~(dev_priv)->info.page_sizes) == 0; \ }) -#define HAS_OVERLAY(dev_priv) ((dev_priv)->info.has_overlay) +#define HAS_OVERLAY(dev_priv) ((dev_priv)->info.display.has_overlay) #define OVERLAY_NEEDS_PHYSICAL(dev_priv) \ - ((dev_priv)->info.overlay_needs_physical) + ((dev_priv)->info.display.overlay_needs_physical) /* Early gen2 have a totally busted CS tlb and require pinned batches. */ #define HAS_BROKEN_CS_TLB(dev_priv) (IS_I830(dev_priv) || IS_I845G(dev_priv)) @@ -2471,31 +2461,31 @@ intel_info(const struct drm_i915_private *dev_priv) #define HAS_128_BYTE_Y_TILING(dev_priv) (!IS_GEN2(dev_priv) && \ !(IS_I915G(dev_priv) || \ IS_I915GM(dev_priv))) -#define SUPPORTS_TV(dev_priv) ((dev_priv)->info.supports_tv) -#define I915_HAS_HOTPLUG(dev_priv) ((dev_priv)->info.has_hotplug) +#define SUPPORTS_TV(dev_priv) ((dev_priv)->info.display.supports_tv) +#define I915_HAS_HOTPLUG(dev_priv) ((dev_priv)->info.display.has_hotplug) #define HAS_FW_BLC(dev_priv) (INTEL_GEN(dev_priv) > 2) -#define HAS_FBC(dev_priv) ((dev_priv)->info.has_fbc) +#define HAS_FBC(dev_priv) ((dev_priv)->info.display.has_fbc) #define HAS_CUR_FBC(dev_priv) (!HAS_GMCH_DISPLAY(dev_priv) && INTEL_GEN(dev_priv) >= 7) #define HAS_IPS(dev_priv) (IS_HSW_ULT(dev_priv) || IS_BROADWELL(dev_priv)) -#define HAS_DP_MST(dev_priv) ((dev_priv)->info.has_dp_mst) +#define HAS_DP_MST(dev_priv) ((dev_priv)->info.display.has_dp_mst) -#define HAS_DDI(dev_priv) ((dev_priv)->info.has_ddi) +#define HAS_DDI(dev_priv) ((dev_priv)->info.display.has_ddi) #define HAS_FPGA_DBG_UNCLAIMED(dev_priv) ((dev_priv)->info.has_fpga_dbg) -#define HAS_PSR(dev_priv) ((dev_priv)->info.has_psr) +#define HAS_PSR(dev_priv) ((dev_priv)->info.display.has_psr) #define HAS_RC6(dev_priv) ((dev_priv)->info.has_rc6) #define HAS_RC6p(dev_priv) ((dev_priv)->info.has_rc6p) #define HAS_RC6pp(dev_priv) (false) /* HW was never validated */ -#define HAS_CSR(dev_priv) ((dev_priv)->info.has_csr) +#define HAS_CSR(dev_priv) ((dev_priv)->info.display.has_csr) #define HAS_RUNTIME_PM(dev_priv) ((dev_priv)->info.has_runtime_pm) #define HAS_64BIT_RELOC(dev_priv) ((dev_priv)->info.has_64bit_reloc) -#define HAS_IPC(dev_priv) ((dev_priv)->info.has_ipc) +#define HAS_IPC(dev_priv) ((dev_priv)->info.display.has_ipc) /* * For now, anything with a GuC requires uCode loading, and then supports @@ -2556,7 +2546,7 @@ intel_info(const struct drm_i915_private *dev_priv) #define HAS_PCH_NOP(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_NOP) #define HAS_PCH_SPLIT(dev_priv) (INTEL_PCH_TYPE(dev_priv) != PCH_NONE) -#define HAS_GMCH_DISPLAY(dev_priv) ((dev_priv)->info.has_gmch_display) +#define HAS_GMCH_DISPLAY(dev_priv) ((dev_priv)->info.display.has_gmch_display) #define HAS_LSPCON(dev_priv) (INTEL_GEN(dev_priv) >= 9) @@ -2568,6 +2558,8 @@ intel_info(const struct drm_i915_private *dev_priv) #define GT_FREQUENCY_MULTIPLIER 50 #define GEN9_FREQ_SCALER 3 +#define HAS_DISPLAY(dev_priv) (INTEL_INFO(dev_priv)->num_pipes > 0) + #include "i915_trace.h" static inline bool intel_vtd_active(void) @@ -3340,6 +3332,9 @@ extern void intel_rps_mark_interactive(struct drm_i915_private *i915, bool interactive); extern bool intel_set_memory_cxsr(struct drm_i915_private *dev_priv, bool enable); +void intel_dsc_enable(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state); +void intel_dsc_disable(const struct intel_crtc_state *crtc_state); int i915_reg_read_ioctl(struct drm_device *dev, void *data, struct drm_file *file); @@ -3720,4 +3715,9 @@ static inline int intel_hws_csb_write_index(struct drm_i915_private *i915) return I915_HWS_CSB_WRITE_INDEX; } +static inline u32 i915_scratch_offset(const struct drm_i915_private *i915) +{ + return i915_ggtt_offset(i915->gt.scratch); +} + #endif diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index b70de1a9e899..7399ac7a5629 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -1282,8 +1282,7 @@ i915_gem_pread_ioctl(struct drm_device *dev, void *data, if (args->size == 0) return 0; - if (!access_ok(VERIFY_WRITE, - u64_to_user_ptr(args->data_ptr), + if (!access_ok(u64_to_user_ptr(args->data_ptr), args->size)) return -EFAULT; @@ -1609,9 +1608,7 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void *data, if (args->size == 0) return 0; - if (!access_ok(VERIFY_READ, - u64_to_user_ptr(args->data_ptr), - args->size)) + if (!access_ok(u64_to_user_ptr(args->data_ptr), args->size)) return -EFAULT; obj = i915_gem_object_lookup(file, args->handle); @@ -2559,7 +2556,7 @@ static int i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj) * If there's no chance of allocating enough pages for the whole * object, bail early. */ - if (page_count > totalram_pages) + if (page_count > totalram_pages()) return -ENOMEM; st = kmalloc(sizeof(*st), GFP_KERNEL); @@ -3309,16 +3306,6 @@ void i915_gem_reset_finish(struct drm_i915_private *dev_priv) static void nop_submit_request(struct i915_request *request) { - GEM_TRACE("%s fence %llx:%d -> -EIO\n", - request->engine->name, - request->fence.context, request->fence.seqno); - dma_fence_set_error(&request->fence, -EIO); - - i915_request_submit(request); -} - -static void nop_complete_submit_request(struct i915_request *request) -{ unsigned long flags; GEM_TRACE("%s fence %llx:%lld -> -EIO\n", @@ -3354,57 +3341,33 @@ void i915_gem_set_wedged(struct drm_i915_private *i915) * rolling the global seqno forward (since this would complete requests * for which we haven't set the fence error to EIO yet). */ - for_each_engine(engine, i915, id) { + for_each_engine(engine, i915, id) i915_gem_reset_prepare_engine(engine); - engine->submit_request = nop_submit_request; - engine->schedule = NULL; - } - i915->caps.scheduler = 0; - /* Even if the GPU reset fails, it should still stop the engines */ if (INTEL_GEN(i915) >= 5) intel_gpu_reset(i915, ALL_ENGINES); - /* - * Make sure no one is running the old callback before we proceed with - * cancelling requests and resetting the completion tracking. Otherwise - * we might submit a request to the hardware which never completes. - */ - synchronize_rcu(); - for_each_engine(engine, i915, id) { - /* Mark all executing requests as skipped */ - engine->cancel_requests(engine); - - /* - * Only once we've force-cancelled all in-flight requests can we - * start to complete all requests. - */ - engine->submit_request = nop_complete_submit_request; + engine->submit_request = nop_submit_request; + engine->schedule = NULL; } + i915->caps.scheduler = 0; /* * Make sure no request can slip through without getting completed by * either this call here to intel_engine_init_global_seqno, or the one - * in nop_complete_submit_request. + * in nop_submit_request. */ synchronize_rcu(); - for_each_engine(engine, i915, id) { - unsigned long flags; - - /* - * Mark all pending requests as complete so that any concurrent - * (lockless) lookup doesn't try and wait upon the request as we - * reset it. - */ - spin_lock_irqsave(&engine->timeline.lock, flags); - intel_engine_init_global_seqno(engine, - intel_engine_last_submit(engine)); - spin_unlock_irqrestore(&engine->timeline.lock, flags); + /* Mark all executing requests as skipped */ + for_each_engine(engine, i915, id) + engine->cancel_requests(engine); + for_each_engine(engine, i915, id) { i915_gem_reset_finish_engine(engine); + intel_engine_wakeup(engine); } out: @@ -5334,7 +5297,10 @@ int i915_gem_init_hw(struct drm_i915_private *dev_priv) I915_WRITE(MI_PREDICATE_RESULT_2, IS_HSW_GT3(dev_priv) ? LOWER_SLICE_ENABLED : LOWER_SLICE_DISABLED); - intel_gt_workarounds_apply(dev_priv); + /* Apply the GT workarounds... */ + intel_gt_apply_workarounds(dev_priv); + /* ...and determine whether they are sticking. */ + intel_gt_verify_workarounds(dev_priv, "init"); i915_gem_init_swizzling(dev_priv); @@ -5529,6 +5495,44 @@ err_active: goto out_ctx; } +static int +i915_gem_init_scratch(struct drm_i915_private *i915, unsigned int size) +{ + struct drm_i915_gem_object *obj; + struct i915_vma *vma; + int ret; + + obj = i915_gem_object_create_stolen(i915, size); + if (!obj) + obj = i915_gem_object_create_internal(i915, size); + if (IS_ERR(obj)) { + DRM_ERROR("Failed to allocate scratch page\n"); + return PTR_ERR(obj); + } + + vma = i915_vma_instance(obj, &i915->ggtt.vm, NULL); + if (IS_ERR(vma)) { + ret = PTR_ERR(vma); + goto err_unref; + } + + ret = i915_vma_pin(vma, 0, 0, PIN_GLOBAL | PIN_HIGH); + if (ret) + goto err_unref; + + i915->gt.scratch = vma; + return 0; + +err_unref: + i915_gem_object_put(obj); + return ret; +} + +static void i915_gem_fini_scratch(struct drm_i915_private *i915) +{ + i915_vma_unpin_and_release(&i915->gt.scratch, 0); +} + int i915_gem_init(struct drm_i915_private *dev_priv) { int ret; @@ -5575,12 +5579,19 @@ int i915_gem_init(struct drm_i915_private *dev_priv) goto err_unlock; } - ret = i915_gem_contexts_init(dev_priv); + ret = i915_gem_init_scratch(dev_priv, + IS_GEN2(dev_priv) ? SZ_256K : PAGE_SIZE); if (ret) { GEM_BUG_ON(ret == -EIO); goto err_ggtt; } + ret = i915_gem_contexts_init(dev_priv); + if (ret) { + GEM_BUG_ON(ret == -EIO); + goto err_scratch; + } + ret = intel_engines_init(dev_priv); if (ret) { GEM_BUG_ON(ret == -EIO); @@ -5653,6 +5664,8 @@ err_pm: err_context: if (ret != -EIO) i915_gem_contexts_fini(dev_priv); +err_scratch: + i915_gem_fini_scratch(dev_priv); err_ggtt: err_unlock: intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL); @@ -5704,8 +5717,11 @@ void i915_gem_fini(struct drm_i915_private *dev_priv) intel_uc_fini(dev_priv); i915_gem_cleanup_engines(dev_priv); i915_gem_contexts_fini(dev_priv); + i915_gem_fini_scratch(dev_priv); mutex_unlock(&dev_priv->drm.struct_mutex); + intel_wa_list_free(&dev_priv->gt_wa_list); + intel_cleanup_gt_powersave(dev_priv); intel_uc_fini_misc(dev_priv); diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c index 66117a8281ef..4ec386950f75 100644 --- a/drivers/gpu/drm/i915/i915_gem_context.c +++ b/drivers/gpu/drm/i915/i915_gem_context.c @@ -535,16 +535,12 @@ static bool needs_preempt_context(struct drm_i915_private *i915) int i915_gem_contexts_init(struct drm_i915_private *dev_priv) { struct i915_gem_context *ctx; - int ret; /* Reassure ourselves we are only called once */ GEM_BUG_ON(dev_priv->kernel_context); GEM_BUG_ON(dev_priv->preempt_context); - ret = intel_ctx_workarounds_init(dev_priv); - if (ret) - return ret; - + intel_engine_init_ctx_wa(dev_priv->engine[RCS]); init_contexts(dev_priv); /* lowest priority; idle task */ diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c index 10a4afb4f235..485b259127c3 100644 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c @@ -26,7 +26,7 @@ * */ -#include <linux/dma_remapping.h> +#include <linux/intel-iommu.h> #include <linux/reservation.h> #include <linux/sync_file.h> #include <linux/uaccess.h> @@ -1268,7 +1268,7 @@ relocate_entry(struct i915_vma *vma, else if (gen >= 4) len = 4; else - len = 6; + len = 3; batch = reloc_gpu(eb, vma, len); if (IS_ERR(batch)) @@ -1309,11 +1309,6 @@ relocate_entry(struct i915_vma *vma, *batch++ = MI_STORE_DWORD_IMM | MI_MEM_VIRTUAL; *batch++ = addr; *batch++ = target_offset; - - /* And again for good measure (blb/pnv) */ - *batch++ = MI_STORE_DWORD_IMM | MI_MEM_VIRTUAL; - *batch++ = addr; - *batch++ = target_offset; } goto out; @@ -1452,7 +1447,7 @@ static int eb_relocate_vma(struct i915_execbuffer *eb, struct i915_vma *vma) * to read. However, if the array is not writable the user loses * the updated relocation values. */ - if (unlikely(!access_ok(VERIFY_READ, urelocs, remain*sizeof(*urelocs)))) + if (unlikely(!access_ok(urelocs, remain*sizeof(*urelocs)))) return -EFAULT; do { @@ -1559,7 +1554,7 @@ static int check_relocations(const struct drm_i915_gem_exec_object2 *entry) addr = u64_to_user_ptr(entry->relocs_ptr); size *= sizeof(struct drm_i915_gem_relocation_entry); - if (!access_ok(VERIFY_READ, addr, size)) + if (!access_ok(addr, size)) return -EFAULT; end = addr + size; @@ -1610,6 +1605,7 @@ static int eb_copy_relocations(const struct i915_execbuffer *eb) (char __user *)urelocs + copied, len)) { end_user: + user_access_end(); kvfree(relocs); err = -EFAULT; goto err; @@ -1628,7 +1624,9 @@ end_user: * happened we would make the mistake of assuming that the * relocations were valid. */ - user_access_begin(); + if (!user_access_begin(urelocs, size)) + goto end_user; + for (copied = 0; copied < nreloc; copied++) unsafe_put_user(-1, &urelocs[copied].presumed_offset, @@ -2095,7 +2093,7 @@ get_fence_array(struct drm_i915_gem_execbuffer2 *args, return ERR_PTR(-EINVAL); user = u64_to_user_ptr(args->cliprects_ptr); - if (!access_ok(VERIFY_READ, user, nfences * sizeof(*user))) + if (!access_ok(user, nfences * sizeof(*user))) return ERR_PTR(-EFAULT); fences = kvmalloc_array(nfences, sizeof(*fences), @@ -2610,7 +2608,16 @@ i915_gem_execbuffer2_ioctl(struct drm_device *dev, void *data, unsigned int i; /* Copy the new buffer offsets back to the user's exec list. */ - user_access_begin(); + /* + * Note: count * sizeof(*user_exec_list) does not overflow, + * because we checked 'count' in check_buffer_count(). + * + * And this range already got effectively checked earlier + * when we did the "copy_from_user()" above. + */ + if (!user_access_begin(user_exec_list, count * sizeof(*user_exec_list))) + goto end_user; + for (i = 0; i < args->buffer_count; i++) { if (!(exec2_list[i].offset & UPDATE)) continue; diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c index 2c9b284036d1..9558582c105e 100644 --- a/drivers/gpu/drm/i915/i915_gem_userptr.c +++ b/drivers/gpu/drm/i915/i915_gem_userptr.c @@ -113,27 +113,25 @@ static void del_object(struct i915_mmu_object *mo) } static int i915_gem_userptr_mn_invalidate_range_start(struct mmu_notifier *_mn, - struct mm_struct *mm, - unsigned long start, - unsigned long end, - bool blockable) + const struct mmu_notifier_range *range) { struct i915_mmu_notifier *mn = container_of(_mn, struct i915_mmu_notifier, mn); struct i915_mmu_object *mo; struct interval_tree_node *it; LIST_HEAD(cancelled); + unsigned long end; if (RB_EMPTY_ROOT(&mn->objects.rb_root)) return 0; /* interval ranges are inclusive, but invalidate range is exclusive */ - end--; + end = range->end - 1; spin_lock(&mn->lock); - it = interval_tree_iter_first(&mn->objects, start, end); + it = interval_tree_iter_first(&mn->objects, range->start, end); while (it) { - if (!blockable) { + if (!range->blockable) { spin_unlock(&mn->lock); return -EAGAIN; } @@ -151,7 +149,7 @@ static int i915_gem_userptr_mn_invalidate_range_start(struct mmu_notifier *_mn, queue_work(mn->wq, &mo->work); list_add(&mo->link, &cancelled); - it = interval_tree_iter_next(it, start, end); + it = interval_tree_iter_next(it, range->start, end); } list_for_each_entry(mo, &cancelled, link) del_object(mo); @@ -791,8 +789,7 @@ i915_gem_userptr_ioctl(struct drm_device *dev, if (offset_in_page(args->user_ptr | args->user_size)) return -EINVAL; - if (!access_ok(args->flags & I915_USERPTR_READ_ONLY ? VERIFY_READ : VERIFY_WRITE, - (char __user *)(unsigned long)args->user_ptr, args->user_size)) + if (!access_ok((char __user *)(unsigned long)args->user_ptr, args->user_size)) return -EFAULT; if (args->flags & I915_USERPTR_READ_ONLY) { diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index 8123bf0e4807..07465123c166 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -27,11 +27,14 @@ * */ -#include <linux/utsname.h> +#include <linux/ascii85.h> +#include <linux/nmi.h> +#include <linux/scatterlist.h> #include <linux/stop_machine.h> +#include <linux/utsname.h> #include <linux/zlib.h> + #include <drm/drm_print.h> -#include <linux/ascii85.h> #include "i915_gpu_error.h" #include "i915_drv.h" @@ -77,112 +80,110 @@ static const char *purgeable_flag(int purgeable) return purgeable ? " purgeable" : ""; } -static bool __i915_error_ok(struct drm_i915_error_state_buf *e) +static void __sg_set_buf(struct scatterlist *sg, + void *addr, unsigned int len, loff_t it) { - - if (!e->err && WARN(e->bytes > (e->size - 1), "overflow")) { - e->err = -ENOSPC; - return false; - } - - if (e->bytes == e->size - 1 || e->err) - return false; - - return true; + sg->page_link = (unsigned long)virt_to_page(addr); + sg->offset = offset_in_page(addr); + sg->length = len; + sg->dma_address = it; } -static bool __i915_error_seek(struct drm_i915_error_state_buf *e, - unsigned len) +static bool __i915_error_grow(struct drm_i915_error_state_buf *e, size_t len) { - if (e->pos + len <= e->start) { - e->pos += len; + if (!len) return false; - } - /* First vsnprintf needs to fit in its entirety for memmove */ - if (len >= e->size) { - e->err = -EIO; - return false; - } + if (e->bytes + len + 1 <= e->size) + return true; - return true; -} + if (e->bytes) { + __sg_set_buf(e->cur++, e->buf, e->bytes, e->iter); + e->iter += e->bytes; + e->buf = NULL; + e->bytes = 0; + } -static void __i915_error_advance(struct drm_i915_error_state_buf *e, - unsigned len) -{ - /* If this is first printf in this window, adjust it so that - * start position matches start of the buffer - */ + if (e->cur == e->end) { + struct scatterlist *sgl; - if (e->pos < e->start) { - const size_t off = e->start - e->pos; + sgl = (typeof(sgl))__get_free_page(GFP_KERNEL); + if (!sgl) { + e->err = -ENOMEM; + return false; + } - /* Should not happen but be paranoid */ - if (off > len || e->bytes) { - e->err = -EIO; - return; + if (e->cur) { + e->cur->offset = 0; + e->cur->length = 0; + e->cur->page_link = + (unsigned long)sgl | SG_CHAIN; + } else { + e->sgl = sgl; } - memmove(e->buf, e->buf + off, len - off); - e->bytes = len - off; - e->pos = e->start; - return; + e->cur = sgl; + e->end = sgl + SG_MAX_SINGLE_ALLOC - 1; } - e->bytes += len; - e->pos += len; + e->size = ALIGN(len + 1, SZ_64K); + e->buf = kmalloc(e->size, GFP_KERNEL | __GFP_NOWARN | __GFP_NORETRY); + if (!e->buf) { + e->size = PAGE_ALIGN(len + 1); + e->buf = kmalloc(e->size, GFP_KERNEL); + } + if (!e->buf) { + e->err = -ENOMEM; + return false; + } + + return true; } __printf(2, 0) static void i915_error_vprintf(struct drm_i915_error_state_buf *e, - const char *f, va_list args) + const char *fmt, va_list args) { - unsigned len; + va_list ap; + int len; - if (!__i915_error_ok(e)) + if (e->err) return; - /* Seek the first printf which is hits start position */ - if (e->pos < e->start) { - va_list tmp; - - va_copy(tmp, args); - len = vsnprintf(NULL, 0, f, tmp); - va_end(tmp); - - if (!__i915_error_seek(e, len)) - return; + va_copy(ap, args); + len = vsnprintf(NULL, 0, fmt, ap); + va_end(ap); + if (len <= 0) { + e->err = len; + return; } - len = vsnprintf(e->buf + e->bytes, e->size - e->bytes, f, args); - if (len >= e->size - e->bytes) - len = e->size - e->bytes - 1; + if (!__i915_error_grow(e, len)) + return; - __i915_error_advance(e, len); + GEM_BUG_ON(e->bytes >= e->size); + len = vscnprintf(e->buf + e->bytes, e->size - e->bytes, fmt, args); + if (len < 0) { + e->err = len; + return; + } + e->bytes += len; } -static void i915_error_puts(struct drm_i915_error_state_buf *e, - const char *str) +static void i915_error_puts(struct drm_i915_error_state_buf *e, const char *str) { unsigned len; - if (!__i915_error_ok(e)) + if (e->err || !str) return; len = strlen(str); + if (!__i915_error_grow(e, len)) + return; - /* Seek the first printf which is hits start position */ - if (e->pos < e->start) { - if (!__i915_error_seek(e, len)) - return; - } - - if (len >= e->size - e->bytes) - len = e->size - e->bytes - 1; + GEM_BUG_ON(e->bytes + len > e->size); memcpy(e->buf + e->bytes, str, len); - - __i915_error_advance(e, len); + e->bytes += len; } #define err_printf(e, ...) i915_error_printf(e, __VA_ARGS__) @@ -268,6 +269,8 @@ static int compress_page(struct compress *c, if (zlib_deflate(zstream, Z_NO_FLUSH) != Z_OK) return -EIO; + + touch_nmi_watchdog(); } while (zstream->avail_in); /* Fallback to uncompressed if we increase size? */ @@ -635,21 +638,29 @@ static void err_print_uc(struct drm_i915_error_state_buf *m, print_error_obj(m, NULL, "GuC log buffer", error_uc->guc_log); } -int i915_error_state_to_str(struct drm_i915_error_state_buf *m, - const struct i915_gpu_state *error) +static void err_free_sgl(struct scatterlist *sgl) { - struct drm_i915_private *dev_priv = m->i915; - struct drm_i915_error_object *obj; - struct timespec64 ts; - int i, j; + while (sgl) { + struct scatterlist *sg; - if (!error) { - err_printf(m, "No error state collected\n"); - return 0; + for (sg = sgl; !sg_is_chain(sg); sg++) { + kfree(sg_virt(sg)); + if (sg_is_last(sg)) + break; + } + + sg = sg_is_last(sg) ? NULL : sg_chain_ptr(sg); + free_page((unsigned long)sgl); + sgl = sg; } +} - if (IS_ERR(error)) - return PTR_ERR(error); +static void __err_print_to_sgl(struct drm_i915_error_state_buf *m, + struct i915_gpu_state *error) +{ + struct drm_i915_error_object *obj; + struct timespec64 ts; + int i, j; if (*error->error_msg) err_printf(m, "%s\n", error->error_msg); @@ -683,12 +694,12 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, err_printf(m, "Reset count: %u\n", error->reset_count); err_printf(m, "Suspend count: %u\n", error->suspend_count); err_printf(m, "Platform: %s\n", intel_platform_name(error->device_info.platform)); - err_print_pciid(m, error->i915); + err_print_pciid(m, m->i915); err_printf(m, "IOMMU enabled?: %d\n", error->iommu); - if (HAS_CSR(dev_priv)) { - struct intel_csr *csr = &dev_priv->csr; + if (HAS_CSR(m->i915)) { + struct intel_csr *csr = &m->i915->csr; err_printf(m, "DMC loaded: %s\n", yesno(csr->dmc_payload != NULL)); @@ -708,22 +719,23 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, err_printf(m, "FORCEWAKE: 0x%08x\n", error->forcewake); err_printf(m, "DERRMR: 0x%08x\n", error->derrmr); err_printf(m, "CCID: 0x%08x\n", error->ccid); - err_printf(m, "Missed interrupts: 0x%08lx\n", dev_priv->gpu_error.missed_irq_rings); + err_printf(m, "Missed interrupts: 0x%08lx\n", + m->i915->gpu_error.missed_irq_rings); for (i = 0; i < error->nfence; i++) err_printf(m, " fence[%d] = %08llx\n", i, error->fence[i]); - if (INTEL_GEN(dev_priv) >= 6) { + if (INTEL_GEN(m->i915) >= 6) { err_printf(m, "ERROR: 0x%08x\n", error->error); - if (INTEL_GEN(dev_priv) >= 8) + if (INTEL_GEN(m->i915) >= 8) err_printf(m, "FAULT_TLB_DATA: 0x%08x 0x%08x\n", error->fault_data1, error->fault_data0); err_printf(m, "DONE_REG: 0x%08x\n", error->done_reg); } - if (IS_GEN7(dev_priv)) + if (IS_GEN7(m->i915)) err_printf(m, "ERR_INT: 0x%08x\n", error->err_int); for (i = 0; i < ARRAY_SIZE(error->engine); i++) { @@ -745,7 +757,7 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, len += scnprintf(buf + len, sizeof(buf), "%s%s", first ? "" : ", ", - dev_priv->engine[j]->name); + m->i915->engine[j]->name); first = 0; } scnprintf(buf + len, sizeof(buf), ")"); @@ -763,7 +775,7 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, obj = ee->batchbuffer; if (obj) { - err_puts(m, dev_priv->engine[i]->name); + err_puts(m, m->i915->engine[i]->name); if (ee->context.pid) err_printf(m, " (submitted by %s [%d], ctx %d [%d], score %d%s)", ee->context.comm, @@ -775,16 +787,16 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, err_printf(m, " --- gtt_offset = 0x%08x %08x\n", upper_32_bits(obj->gtt_offset), lower_32_bits(obj->gtt_offset)); - print_error_obj(m, dev_priv->engine[i], NULL, obj); + print_error_obj(m, m->i915->engine[i], NULL, obj); } for (j = 0; j < ee->user_bo_count; j++) - print_error_obj(m, dev_priv->engine[i], + print_error_obj(m, m->i915->engine[i], "user", ee->user_bo[j]); if (ee->num_requests) { err_printf(m, "%s --- %d requests\n", - dev_priv->engine[i]->name, + m->i915->engine[i]->name, ee->num_requests); for (j = 0; j < ee->num_requests; j++) error_print_request(m, " ", @@ -794,10 +806,10 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, if (IS_ERR(ee->waiters)) { err_printf(m, "%s --- ? waiters [unable to acquire spinlock]\n", - dev_priv->engine[i]->name); + m->i915->engine[i]->name); } else if (ee->num_waiters) { err_printf(m, "%s --- %d waiters\n", - dev_priv->engine[i]->name, + m->i915->engine[i]->name, ee->num_waiters); for (j = 0; j < ee->num_waiters; j++) { err_printf(m, " seqno 0x%08x for %s [%d]\n", @@ -807,22 +819,22 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, } } - print_error_obj(m, dev_priv->engine[i], + print_error_obj(m, m->i915->engine[i], "ringbuffer", ee->ringbuffer); - print_error_obj(m, dev_priv->engine[i], + print_error_obj(m, m->i915->engine[i], "HW Status", ee->hws_page); - print_error_obj(m, dev_priv->engine[i], + print_error_obj(m, m->i915->engine[i], "HW context", ee->ctx); - print_error_obj(m, dev_priv->engine[i], + print_error_obj(m, m->i915->engine[i], "WA context", ee->wa_ctx); - print_error_obj(m, dev_priv->engine[i], + print_error_obj(m, m->i915->engine[i], "WA batchbuffer", ee->wa_batchbuffer); - print_error_obj(m, dev_priv->engine[i], + print_error_obj(m, m->i915->engine[i], "NULL context", ee->default_state); } @@ -835,43 +847,107 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, err_print_capabilities(m, &error->device_info, &error->driver_caps); err_print_params(m, &error->params); err_print_uc(m, &error->uc); +} + +static int err_print_to_sgl(struct i915_gpu_state *error) +{ + struct drm_i915_error_state_buf m; + + if (IS_ERR(error)) + return PTR_ERR(error); + + if (READ_ONCE(error->sgl)) + return 0; + + memset(&m, 0, sizeof(m)); + m.i915 = error->i915; + + __err_print_to_sgl(&m, error); + + if (m.buf) { + __sg_set_buf(m.cur++, m.buf, m.bytes, m.iter); + m.bytes = 0; + m.buf = NULL; + } + if (m.cur) { + GEM_BUG_ON(m.end < m.cur); + sg_mark_end(m.cur - 1); + } + GEM_BUG_ON(m.sgl && !m.cur); + + if (m.err) { + err_free_sgl(m.sgl); + return m.err; + } - if (m->bytes == 0 && m->err) - return m->err; + if (cmpxchg(&error->sgl, NULL, m.sgl)) + err_free_sgl(m.sgl); return 0; } -int i915_error_state_buf_init(struct drm_i915_error_state_buf *ebuf, - struct drm_i915_private *i915, - size_t count, loff_t pos) +ssize_t i915_gpu_state_copy_to_buffer(struct i915_gpu_state *error, + char *buf, loff_t off, size_t rem) { - memset(ebuf, 0, sizeof(*ebuf)); - ebuf->i915 = i915; + struct scatterlist *sg; + size_t count; + loff_t pos; + int err; - /* We need to have enough room to store any i915_error_state printf - * so that we can move it to start position. - */ - ebuf->size = count + 1 > PAGE_SIZE ? count + 1 : PAGE_SIZE; - ebuf->buf = kmalloc(ebuf->size, - GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN); + if (!error || !rem) + return 0; - if (ebuf->buf == NULL) { - ebuf->size = PAGE_SIZE; - ebuf->buf = kmalloc(ebuf->size, GFP_KERNEL); - } + err = err_print_to_sgl(error); + if (err) + return err; - if (ebuf->buf == NULL) { - ebuf->size = 128; - ebuf->buf = kmalloc(ebuf->size, GFP_KERNEL); - } + sg = READ_ONCE(error->fit); + if (!sg || off < sg->dma_address) + sg = error->sgl; + if (!sg) + return 0; - if (ebuf->buf == NULL) - return -ENOMEM; + pos = sg->dma_address; + count = 0; + do { + size_t len, start; + + if (sg_is_chain(sg)) { + sg = sg_chain_ptr(sg); + GEM_BUG_ON(sg_is_chain(sg)); + } + + len = sg->length; + if (pos + len <= off) { + pos += len; + continue; + } - ebuf->start = pos; + start = sg->offset; + if (pos < off) { + GEM_BUG_ON(off - pos > len); + len -= off - pos; + start += off - pos; + pos = off; + } - return 0; + len = min(len, rem); + GEM_BUG_ON(!len || len > sg->length); + + memcpy(buf, page_address(sg_page(sg)) + start, len); + + count += len; + pos += len; + + buf += len; + rem -= len; + if (!rem) { + WRITE_ONCE(error->fit, sg); + break; + } + } while (!sg_is_last(sg++)); + + return count; } static void i915_error_object_free(struct drm_i915_error_object *obj) @@ -944,6 +1020,7 @@ void __i915_gpu_state_free(struct kref *error_ref) cleanup_params(error); cleanup_uc_state(error); + err_free_sgl(error->sgl); kfree(error); } @@ -1494,7 +1571,7 @@ static void gem_record_rings(struct i915_gpu_state *error) if (HAS_BROKEN_CS_TLB(i915)) ee->wa_batchbuffer = i915_error_object_create(i915, - engine->scratch); + i915->gt.scratch); request_record_user_bo(request, ee); ee->ctx = diff --git a/drivers/gpu/drm/i915/i915_gpu_error.h b/drivers/gpu/drm/i915/i915_gpu_error.h index 3ec89a504de5..ff2652bbb0b0 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.h +++ b/drivers/gpu/drm/i915/i915_gpu_error.h @@ -192,6 +192,8 @@ struct i915_gpu_state { } *active_bo[I915_NUM_ENGINES], *pinned_bo; u32 active_bo_count[I915_NUM_ENGINES], pinned_bo_count; struct i915_address_space *active_vm[I915_NUM_ENGINES]; + + struct scatterlist *sgl, *fit; }; struct i915_gpu_error { @@ -298,29 +300,20 @@ struct i915_gpu_error { struct drm_i915_error_state_buf { struct drm_i915_private *i915; - unsigned int bytes; - unsigned int size; + struct scatterlist *sgl, *cur, *end; + + char *buf; + size_t bytes; + size_t size; + loff_t iter; + int err; - u8 *buf; - loff_t start; - loff_t pos; }; #if IS_ENABLED(CONFIG_DRM_I915_CAPTURE_ERROR) __printf(2, 3) void i915_error_printf(struct drm_i915_error_state_buf *e, const char *f, ...); -int i915_error_state_to_str(struct drm_i915_error_state_buf *estr, - const struct i915_gpu_state *gpu); -int i915_error_state_buf_init(struct drm_i915_error_state_buf *eb, - struct drm_i915_private *i915, - size_t count, loff_t pos); - -static inline void -i915_error_state_buf_release(struct drm_i915_error_state_buf *eb) -{ - kfree(eb->buf); -} struct i915_gpu_state *i915_capture_gpu_state(struct drm_i915_private *i915); void i915_capture_error_state(struct drm_i915_private *dev_priv, @@ -334,6 +327,9 @@ i915_gpu_state_get(struct i915_gpu_state *gpu) return gpu; } +ssize_t i915_gpu_state_copy_to_buffer(struct i915_gpu_state *error, + char *buf, loff_t offset, size_t count); + void __i915_gpu_state_free(struct kref *kref); static inline void i915_gpu_state_put(struct i915_gpu_state *gpu) { diff --git a/drivers/gpu/drm/i915/i915_ioc32.c b/drivers/gpu/drm/i915/i915_ioc32.c index 0e5c580d117c..e869daf9c8a9 100644 --- a/drivers/gpu/drm/i915/i915_ioc32.c +++ b/drivers/gpu/drm/i915/i915_ioc32.c @@ -52,7 +52,7 @@ static int compat_i915_getparam(struct file *file, unsigned int cmd, return -EFAULT; request = compat_alloc_user_space(sizeof(*request)); - if (!access_ok(VERIFY_WRITE, request, sizeof(*request)) || + if (!access_ok(request, sizeof(*request)) || __put_user(req32.param, &request->param) || __put_user((void __user *)(unsigned long)req32.value, &request->value)) diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c index 1b81d7cb209e..6350db5503cd 100644 --- a/drivers/gpu/drm/i915/i915_pci.c +++ b/drivers/gpu/drm/i915/i915_pci.c @@ -79,8 +79,9 @@ #define GEN2_FEATURES \ GEN(2), \ .num_pipes = 1, \ - .has_overlay = 1, .overlay_needs_physical = 1, \ - .has_gmch_display = 1, \ + .display.has_overlay = 1, \ + .display.overlay_needs_physical = 1, \ + .display.has_gmch_display = 1, \ .hws_needs_physical = 1, \ .unfenced_needs_alignment = 1, \ .ring_mask = RENDER_RING, \ @@ -93,7 +94,8 @@ static const struct intel_device_info intel_i830_info = { GEN2_FEATURES, PLATFORM(INTEL_I830), - .is_mobile = 1, .cursor_needs_physical = 1, + .is_mobile = 1, + .display.cursor_needs_physical = 1, .num_pipes = 2, /* legal, last one wins */ }; @@ -107,8 +109,8 @@ static const struct intel_device_info intel_i85x_info = { PLATFORM(INTEL_I85X), .is_mobile = 1, .num_pipes = 2, /* legal, last one wins */ - .cursor_needs_physical = 1, - .has_fbc = 1, + .display.cursor_needs_physical = 1, + .display.has_fbc = 1, }; static const struct intel_device_info intel_i865g_info = { @@ -119,7 +121,7 @@ static const struct intel_device_info intel_i865g_info = { #define GEN3_FEATURES \ GEN(3), \ .num_pipes = 2, \ - .has_gmch_display = 1, \ + .display.has_gmch_display = 1, \ .ring_mask = RENDER_RING, \ .has_snoop = true, \ .has_coherent_ggtt = true, \ @@ -131,8 +133,9 @@ static const struct intel_device_info intel_i915g_info = { GEN3_FEATURES, PLATFORM(INTEL_I915G), .has_coherent_ggtt = false, - .cursor_needs_physical = 1, - .has_overlay = 1, .overlay_needs_physical = 1, + .display.cursor_needs_physical = 1, + .display.has_overlay = 1, + .display.overlay_needs_physical = 1, .hws_needs_physical = 1, .unfenced_needs_alignment = 1, }; @@ -141,10 +144,11 @@ static const struct intel_device_info intel_i915gm_info = { GEN3_FEATURES, PLATFORM(INTEL_I915GM), .is_mobile = 1, - .cursor_needs_physical = 1, - .has_overlay = 1, .overlay_needs_physical = 1, - .supports_tv = 1, - .has_fbc = 1, + .display.cursor_needs_physical = 1, + .display.has_overlay = 1, + .display.overlay_needs_physical = 1, + .display.supports_tv = 1, + .display.has_fbc = 1, .hws_needs_physical = 1, .unfenced_needs_alignment = 1, }; @@ -152,8 +156,10 @@ static const struct intel_device_info intel_i915gm_info = { static const struct intel_device_info intel_i945g_info = { GEN3_FEATURES, PLATFORM(INTEL_I945G), - .has_hotplug = 1, .cursor_needs_physical = 1, - .has_overlay = 1, .overlay_needs_physical = 1, + .display.has_hotplug = 1, + .display.cursor_needs_physical = 1, + .display.has_overlay = 1, + .display.overlay_needs_physical = 1, .hws_needs_physical = 1, .unfenced_needs_alignment = 1, }; @@ -162,10 +168,12 @@ static const struct intel_device_info intel_i945gm_info = { GEN3_FEATURES, PLATFORM(INTEL_I945GM), .is_mobile = 1, - .has_hotplug = 1, .cursor_needs_physical = 1, - .has_overlay = 1, .overlay_needs_physical = 1, - .supports_tv = 1, - .has_fbc = 1, + .display.has_hotplug = 1, + .display.cursor_needs_physical = 1, + .display.has_overlay = 1, + .display.overlay_needs_physical = 1, + .display.supports_tv = 1, + .display.has_fbc = 1, .hws_needs_physical = 1, .unfenced_needs_alignment = 1, }; @@ -173,23 +181,23 @@ static const struct intel_device_info intel_i945gm_info = { static const struct intel_device_info intel_g33_info = { GEN3_FEATURES, PLATFORM(INTEL_G33), - .has_hotplug = 1, - .has_overlay = 1, + .display.has_hotplug = 1, + .display.has_overlay = 1, }; static const struct intel_device_info intel_pineview_info = { GEN3_FEATURES, PLATFORM(INTEL_PINEVIEW), .is_mobile = 1, - .has_hotplug = 1, - .has_overlay = 1, + .display.has_hotplug = 1, + .display.has_overlay = 1, }; #define GEN4_FEATURES \ GEN(4), \ .num_pipes = 2, \ - .has_hotplug = 1, \ - .has_gmch_display = 1, \ + .display.has_hotplug = 1, \ + .display.has_gmch_display = 1, \ .ring_mask = RENDER_RING, \ .has_snoop = true, \ .has_coherent_ggtt = true, \ @@ -200,7 +208,7 @@ static const struct intel_device_info intel_pineview_info = { static const struct intel_device_info intel_i965g_info = { GEN4_FEATURES, PLATFORM(INTEL_I965G), - .has_overlay = 1, + .display.has_overlay = 1, .hws_needs_physical = 1, .has_snoop = false, }; @@ -208,9 +216,10 @@ static const struct intel_device_info intel_i965g_info = { static const struct intel_device_info intel_i965gm_info = { GEN4_FEATURES, PLATFORM(INTEL_I965GM), - .is_mobile = 1, .has_fbc = 1, - .has_overlay = 1, - .supports_tv = 1, + .is_mobile = 1, + .display.has_fbc = 1, + .display.has_overlay = 1, + .display.supports_tv = 1, .hws_needs_physical = 1, .has_snoop = false, }; @@ -224,15 +233,16 @@ static const struct intel_device_info intel_g45_info = { static const struct intel_device_info intel_gm45_info = { GEN4_FEATURES, PLATFORM(INTEL_GM45), - .is_mobile = 1, .has_fbc = 1, - .supports_tv = 1, + .is_mobile = 1, + .display.has_fbc = 1, + .display.supports_tv = 1, .ring_mask = RENDER_RING | BSD_RING, }; #define GEN5_FEATURES \ GEN(5), \ .num_pipes = 2, \ - .has_hotplug = 1, \ + .display.has_hotplug = 1, \ .ring_mask = RENDER_RING | BSD_RING, \ .has_snoop = true, \ .has_coherent_ggtt = true, \ @@ -250,14 +260,15 @@ static const struct intel_device_info intel_ironlake_d_info = { static const struct intel_device_info intel_ironlake_m_info = { GEN5_FEATURES, PLATFORM(INTEL_IRONLAKE), - .is_mobile = 1, .has_fbc = 1, + .is_mobile = 1, + .display.has_fbc = 1, }; #define GEN6_FEATURES \ GEN(6), \ .num_pipes = 2, \ - .has_hotplug = 1, \ - .has_fbc = 1, \ + .display.has_hotplug = 1, \ + .display.has_fbc = 1, \ .ring_mask = RENDER_RING | BSD_RING | BLT_RING, \ .has_coherent_ggtt = true, \ .has_llc = 1, \ @@ -301,8 +312,8 @@ static const struct intel_device_info intel_sandybridge_m_gt2_info = { #define GEN7_FEATURES \ GEN(7), \ .num_pipes = 3, \ - .has_hotplug = 1, \ - .has_fbc = 1, \ + .display.has_hotplug = 1, \ + .display.has_fbc = 1, \ .ring_mask = RENDER_RING | BSD_RING | BLT_RING, \ .has_coherent_ggtt = true, \ .has_llc = 1, \ @@ -359,8 +370,8 @@ static const struct intel_device_info intel_valleyview_info = { .num_pipes = 2, .has_runtime_pm = 1, .has_rc6 = 1, - .has_gmch_display = 1, - .has_hotplug = 1, + .display.has_gmch_display = 1, + .display.has_hotplug = 1, .ppgtt = INTEL_PPGTT_FULL, .has_snoop = true, .has_coherent_ggtt = false, @@ -374,10 +385,10 @@ static const struct intel_device_info intel_valleyview_info = { #define G75_FEATURES \ GEN7_FEATURES, \ .ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING, \ - .has_ddi = 1, \ + .display.has_ddi = 1, \ .has_fpga_dbg = 1, \ - .has_psr = 1, \ - .has_dp_mst = 1, \ + .display.has_psr = 1, \ + .display.has_dp_mst = 1, \ .has_rc6p = 0 /* RC6p removed-by HSW */, \ .has_runtime_pm = 1 @@ -444,14 +455,14 @@ static const struct intel_device_info intel_cherryview_info = { PLATFORM(INTEL_CHERRYVIEW), GEN(8), .num_pipes = 3, - .has_hotplug = 1, + .display.has_hotplug = 1, .is_lp = 1, .ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING, .has_64bit_reloc = 1, .has_runtime_pm = 1, .has_rc6 = 1, .has_logical_ring_contexts = 1, - .has_gmch_display = 1, + .display.has_gmch_display = 1, .ppgtt = INTEL_PPGTT_FULL, .has_reset_engine = 1, .has_snoop = true, @@ -473,15 +484,15 @@ static const struct intel_device_info intel_cherryview_info = { GEN(9), \ GEN9_DEFAULT_PAGE_SIZES, \ .has_logical_ring_preemption = 1, \ - .has_csr = 1, \ + .display.has_csr = 1, \ .has_guc = 1, \ - .has_ipc = 1, \ + .display.has_ipc = 1, \ .ddb_size = 896 #define SKL_PLATFORM \ GEN9_FEATURES, \ /* Display WA #0477 WaDisableIPC: skl */ \ - .has_ipc = 0, \ + .display.has_ipc = 0, \ PLATFORM(INTEL_SKYLAKE) static const struct intel_device_info intel_skylake_gt1_info = { @@ -512,19 +523,19 @@ static const struct intel_device_info intel_skylake_gt4_info = { #define GEN9_LP_FEATURES \ GEN(9), \ .is_lp = 1, \ - .has_hotplug = 1, \ + .display.has_hotplug = 1, \ .ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING, \ .num_pipes = 3, \ .has_64bit_reloc = 1, \ - .has_ddi = 1, \ + .display.has_ddi = 1, \ .has_fpga_dbg = 1, \ - .has_fbc = 1, \ - .has_psr = 1, \ + .display.has_fbc = 1, \ + .display.has_psr = 1, \ .has_runtime_pm = 1, \ .has_pooled_eu = 0, \ - .has_csr = 1, \ + .display.has_csr = 1, \ .has_rc6 = 1, \ - .has_dp_mst = 1, \ + .display.has_dp_mst = 1, \ .has_logical_ring_contexts = 1, \ .has_logical_ring_preemption = 1, \ .has_guc = 1, \ @@ -532,7 +543,7 @@ static const struct intel_device_info intel_skylake_gt4_info = { .has_reset_engine = 1, \ .has_snoop = true, \ .has_coherent_ggtt = false, \ - .has_ipc = 1, \ + .display.has_ipc = 1, \ GEN9_DEFAULT_PAGE_SIZES, \ GEN_DEFAULT_PIPEOFFSETS, \ IVB_CURSOR_OFFSETS, \ diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c index 4529edfdcfc8..2b2eb57ca71f 100644 --- a/drivers/gpu/drm/i915/i915_perf.c +++ b/drivers/gpu/drm/i915/i915_perf.c @@ -3052,7 +3052,7 @@ static struct i915_oa_reg *alloc_oa_regs(struct drm_i915_private *dev_priv, if (!n_regs) return NULL; - if (!access_ok(VERIFY_READ, regs, n_regs * sizeof(u32) * 2)) + if (!access_ok(regs, n_regs * sizeof(u32) * 2)) return ERR_PTR(-EFAULT); /* No is_valid function means we're not allowing any register to be programmed. */ diff --git a/drivers/gpu/drm/i915/i915_query.c b/drivers/gpu/drm/i915/i915_query.c index 6fc4b8eeab42..fe56465cdfd6 100644 --- a/drivers/gpu/drm/i915/i915_query.c +++ b/drivers/gpu/drm/i915/i915_query.c @@ -46,7 +46,7 @@ static int query_topology_info(struct drm_i915_private *dev_priv, if (topo.flags != 0) return -EINVAL; - if (!access_ok(VERIFY_WRITE, u64_to_user_ptr(query_item->data_ptr), + if (!access_ok(u64_to_user_ptr(query_item->data_ptr), total_length)) return -EFAULT; diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 47baf2fe8f71..0a7d60509ca7 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -4570,6 +4570,7 @@ enum { * of the infoframe structure specified by CEA-861. */ #define VIDEO_DIP_DATA_SIZE 32 #define VIDEO_DIP_VSC_DATA_SIZE 36 +#define VIDEO_DIP_PPS_DATA_SIZE 132 #define VIDEO_DIP_CTL _MMIO(0x61170) /* Pre HSW: */ #define VIDEO_DIP_ENABLE (1 << 31) @@ -4617,6 +4618,17 @@ enum { #define _PP_STATUS 0x61200 #define PP_STATUS(pps_idx) _MMIO_PPS(pps_idx, _PP_STATUS) #define PP_ON (1 << 31) + +#define _PP_CONTROL_1 0xc7204 +#define _PP_CONTROL_2 0xc7304 +#define ICP_PP_CONTROL(x) _MMIO(((x) == 1) ? _PP_CONTROL_1 : \ + _PP_CONTROL_2) +#define POWER_CYCLE_DELAY_MASK (0x1f << 4) +#define POWER_CYCLE_DELAY_SHIFT 4 +#define VDD_OVERRIDE_FORCE (1 << 3) +#define BACKLIGHT_ENABLE (1 << 2) +#define PWR_DOWN_ON_RESET (1 << 1) +#define PWR_STATE_TARGET (1 << 0) /* * Indicates that all dependencies of the panel are on: * @@ -7750,6 +7762,7 @@ enum { #define ICP_DDIB_HPD_LONG_DETECT (2 << 4) #define ICP_DDIB_HPD_SHORT_LONG_DETECT (3 << 4) #define ICP_DDIA_HPD_ENABLE (1 << 3) +#define ICP_DDIA_HPD_OP_DRIVE_1 (1 << 2) #define ICP_DDIA_HPD_STATUS_MASK (3 << 0) #define ICP_DDIA_HPD_NO_DETECT (0 << 0) #define ICP_DDIA_HPD_SHORT_DETECT (1 << 0) @@ -9197,6 +9210,7 @@ enum skl_power_gate { #define _DP_TP_CTL_B 0x64140 #define DP_TP_CTL(port) _MMIO_PORT(port, _DP_TP_CTL_A, _DP_TP_CTL_B) #define DP_TP_CTL_ENABLE (1 << 31) +#define DP_TP_CTL_FEC_ENABLE (1 << 30) #define DP_TP_CTL_MODE_SST (0 << 27) #define DP_TP_CTL_MODE_MST (1 << 27) #define DP_TP_CTL_FORCE_ACT (1 << 25) @@ -9215,6 +9229,7 @@ enum skl_power_gate { #define _DP_TP_STATUS_A 0x64044 #define _DP_TP_STATUS_B 0x64144 #define DP_TP_STATUS(port) _MMIO_PORT(port, _DP_TP_STATUS_A, _DP_TP_STATUS_B) +#define DP_TP_STATUS_FEC_ENABLE_LIVE (1 << 28) #define DP_TP_STATUS_IDLE_DONE (1 << 25) #define DP_TP_STATUS_ACT_SENT (1 << 24) #define DP_TP_STATUS_MODE_STATUS_MST (1 << 23) diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c index 191703986c7b..cefefc11d922 100644 --- a/drivers/gpu/drm/i915/i915_request.c +++ b/drivers/gpu/drm/i915/i915_request.c @@ -136,6 +136,9 @@ static int reset_all_global_seqno(struct drm_i915_private *i915, u32 seqno) intel_engine_get_seqno(engine), seqno); + if (seqno == engine->timeline.seqno) + continue; + kthread_park(engine->breadcrumbs.signaler); if (!i915_seqno_passed(seqno, engine->timeline.seqno)) { diff --git a/drivers/gpu/drm/i915/i915_sw_fence.c b/drivers/gpu/drm/i915/i915_sw_fence.c index 11bcdabd5177..7c58b049ecb5 100644 --- a/drivers/gpu/drm/i915/i915_sw_fence.c +++ b/drivers/gpu/drm/i915/i915_sw_fence.c @@ -1,10 +1,7 @@ /* - * (C) Copyright 2016 Intel Corporation + * SPDX-License-Identifier: MIT * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; version 2 - * of the License. + * (C) Copyright 2016 Intel Corporation */ #include <linux/slab.h> diff --git a/drivers/gpu/drm/i915/i915_sw_fence.h b/drivers/gpu/drm/i915/i915_sw_fence.h index fe2ef4dadfc6..0e055ea0179f 100644 --- a/drivers/gpu/drm/i915/i915_sw_fence.h +++ b/drivers/gpu/drm/i915/i915_sw_fence.h @@ -1,10 +1,9 @@ /* + * SPDX-License-Identifier: MIT + * * i915_sw_fence.h - library routines for N:M synchronisation points * * Copyright (C) 2016 Intel Corporation - * - * This file is released under the GPLv2. - * */ #ifndef _I915_SW_FENCE_H_ diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c index e5e6f6bb2b05..535caebd9813 100644 --- a/drivers/gpu/drm/i915/i915_sysfs.c +++ b/drivers/gpu/drm/i915/i915_sysfs.c @@ -483,7 +483,7 @@ static ssize_t gt_rp_mhz_show(struct device *kdev, struct device_attribute *attr return snprintf(buf, PAGE_SIZE, "%d\n", val); } -static const struct attribute *gen6_attrs[] = { +static const struct attribute * const gen6_attrs[] = { &dev_attr_gt_act_freq_mhz.attr, &dev_attr_gt_cur_freq_mhz.attr, &dev_attr_gt_boost_freq_mhz.attr, @@ -495,7 +495,7 @@ static const struct attribute *gen6_attrs[] = { NULL, }; -static const struct attribute *vlv_attrs[] = { +static const struct attribute * const vlv_attrs[] = { &dev_attr_gt_act_freq_mhz.attr, &dev_attr_gt_cur_freq_mhz.attr, &dev_attr_gt_boost_freq_mhz.attr, @@ -516,26 +516,21 @@ static ssize_t error_state_read(struct file *filp, struct kobject *kobj, { struct device *kdev = kobj_to_dev(kobj); - struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev); - struct drm_i915_error_state_buf error_str; + struct drm_i915_private *i915 = kdev_minor_to_i915(kdev); struct i915_gpu_state *gpu; ssize_t ret; - ret = i915_error_state_buf_init(&error_str, dev_priv, count, off); - if (ret) - return ret; - - gpu = i915_first_error_state(dev_priv); - ret = i915_error_state_to_str(&error_str, gpu); - if (ret) - goto out; - - ret = count < error_str.bytes ? count : error_str.bytes; - memcpy(buf, error_str.buf, ret); + gpu = i915_first_error_state(i915); + if (gpu) { + ret = i915_gpu_state_copy_to_buffer(gpu, buf, off, count); + i915_gpu_state_put(gpu); + } else { + const char *str = "No error state collected\n"; + size_t len = strlen(str); -out: - i915_gpu_state_put(gpu); - i915_error_state_buf_release(&error_str); + ret = min_t(size_t, count, len - off); + memcpy(buf, str + off, ret); + } return ret; } diff --git a/drivers/gpu/drm/i915/i915_utils.h b/drivers/gpu/drm/i915/i915_utils.h index 5858a43e19da..9726df37c4c4 100644 --- a/drivers/gpu/drm/i915/i915_utils.h +++ b/drivers/gpu/drm/i915/i915_utils.h @@ -44,16 +44,19 @@ __stringify(x), (long)(x)) #if defined(GCC_VERSION) && GCC_VERSION >= 70000 -#define add_overflows(A, B) \ - __builtin_add_overflow_p((A), (B), (typeof((A) + (B)))0) +#define add_overflows_t(T, A, B) \ + __builtin_add_overflow_p((A), (B), (T)0) #else -#define add_overflows(A, B) ({ \ +#define add_overflows_t(T, A, B) ({ \ typeof(A) a = (A); \ typeof(B) b = (B); \ - a + b < a; \ + (T)(a + b) < a; \ }) #endif +#define add_overflows(A, B) \ + add_overflows_t(typeof((A) + (B)), (A), (B)) + #define range_overflows(start, size, max) ({ \ typeof(start) start__ = (start); \ typeof(size) size__ = (size); \ diff --git a/drivers/gpu/drm/i915/icl_dsi.c b/drivers/gpu/drm/i915/icl_dsi.c index 01f422df8c23..4dd793b78996 100644 --- a/drivers/gpu/drm/i915/icl_dsi.c +++ b/drivers/gpu/drm/i915/icl_dsi.c @@ -26,6 +26,7 @@ */ #include <drm/drm_mipi_dsi.h> +#include <drm/drm_atomic_helper.h> #include "intel_dsi.h" static inline int header_credits_available(struct drm_i915_private *dev_priv, @@ -107,6 +108,90 @@ static void wait_for_cmds_dispatched_to_panel(struct intel_encoder *encoder) } } +static bool add_payld_to_queue(struct intel_dsi_host *host, const u8 *data, + u32 len) +{ + struct intel_dsi *intel_dsi = host->intel_dsi; + struct drm_i915_private *dev_priv = to_i915(intel_dsi->base.base.dev); + enum transcoder dsi_trans = dsi_port_to_transcoder(host->port); + int free_credits; + int i, j; + + for (i = 0; i < len; i += 4) { + u32 tmp = 0; + + free_credits = payload_credits_available(dev_priv, dsi_trans); + if (free_credits < 1) { + DRM_ERROR("Payload credit not available\n"); + return false; + } + + for (j = 0; j < min_t(u32, len - i, 4); j++) + tmp |= *data++ << 8 * j; + + I915_WRITE(DSI_CMD_TXPYLD(dsi_trans), tmp); + } + + return true; +} + +static int dsi_send_pkt_hdr(struct intel_dsi_host *host, + struct mipi_dsi_packet pkt, bool enable_lpdt) +{ + struct intel_dsi *intel_dsi = host->intel_dsi; + struct drm_i915_private *dev_priv = to_i915(intel_dsi->base.base.dev); + enum transcoder dsi_trans = dsi_port_to_transcoder(host->port); + u32 tmp; + int free_credits; + + /* check if header credit available */ + free_credits = header_credits_available(dev_priv, dsi_trans); + if (free_credits < 1) { + DRM_ERROR("send pkt header failed, not enough hdr credits\n"); + return -1; + } + + tmp = I915_READ(DSI_CMD_TXHDR(dsi_trans)); + + if (pkt.payload) + tmp |= PAYLOAD_PRESENT; + else + tmp &= ~PAYLOAD_PRESENT; + + tmp &= ~VBLANK_FENCE; + + if (enable_lpdt) + tmp |= LP_DATA_TRANSFER; + + tmp &= ~(PARAM_WC_MASK | VC_MASK | DT_MASK); + tmp |= ((pkt.header[0] & VC_MASK) << VC_SHIFT); + tmp |= ((pkt.header[0] & DT_MASK) << DT_SHIFT); + tmp |= (pkt.header[1] << PARAM_WC_LOWER_SHIFT); + tmp |= (pkt.header[2] << PARAM_WC_UPPER_SHIFT); + I915_WRITE(DSI_CMD_TXHDR(dsi_trans), tmp); + + return 0; +} + +static int dsi_send_pkt_payld(struct intel_dsi_host *host, + struct mipi_dsi_packet pkt) +{ + /* payload queue can accept *256 bytes*, check limit */ + if (pkt.payload_length > MAX_PLOAD_CREDIT * 4) { + DRM_ERROR("payload size exceeds max queue limit\n"); + return -1; + } + + /* load data into command payload queue */ + if (!add_payld_to_queue(host, pkt.payload, + pkt.payload_length)) { + DRM_ERROR("adding payload to queue failed\n"); + return -1; + } + + return 0; +} + static void dsi_program_swing_and_deemphasis(struct intel_encoder *encoder) { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); @@ -172,6 +257,45 @@ static void dsi_program_swing_and_deemphasis(struct intel_encoder *encoder) } } +static void configure_dual_link_mode(struct intel_encoder *encoder, + const struct intel_crtc_state *pipe_config) +{ + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base); + u32 dss_ctl1; + + dss_ctl1 = I915_READ(DSS_CTL1); + dss_ctl1 |= SPLITTER_ENABLE; + dss_ctl1 &= ~OVERLAP_PIXELS_MASK; + dss_ctl1 |= OVERLAP_PIXELS(intel_dsi->pixel_overlap); + + if (intel_dsi->dual_link == DSI_DUAL_LINK_FRONT_BACK) { + const struct drm_display_mode *adjusted_mode = + &pipe_config->base.adjusted_mode; + u32 dss_ctl2; + u16 hactive = adjusted_mode->crtc_hdisplay; + u16 dl_buffer_depth; + + dss_ctl1 &= ~DUAL_LINK_MODE_INTERLEAVE; + dl_buffer_depth = hactive / 2 + intel_dsi->pixel_overlap; + + if (dl_buffer_depth > MAX_DL_BUFFER_TARGET_DEPTH) + DRM_ERROR("DL buffer depth exceed max value\n"); + + dss_ctl1 &= ~LEFT_DL_BUF_TARGET_DEPTH_MASK; + dss_ctl1 |= LEFT_DL_BUF_TARGET_DEPTH(dl_buffer_depth); + dss_ctl2 = I915_READ(DSS_CTL2); + dss_ctl2 &= ~RIGHT_DL_BUF_TARGET_DEPTH_MASK; + dss_ctl2 |= RIGHT_DL_BUF_TARGET_DEPTH(dl_buffer_depth); + I915_WRITE(DSS_CTL2, dss_ctl2); + } else { + /* Interleave */ + dss_ctl1 |= DUAL_LINK_MODE_INTERLEAVE; + } + + I915_WRITE(DSS_CTL1, dss_ctl1); +} + static void gen11_dsi_program_esc_clk_div(struct intel_encoder *encoder) { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); @@ -412,6 +536,62 @@ static void gen11_dsi_setup_dphy_timings(struct intel_encoder *encoder) } } +static void gen11_dsi_gate_clocks(struct intel_encoder *encoder) +{ + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base); + u32 tmp; + enum port port; + + mutex_lock(&dev_priv->dpll_lock); + tmp = I915_READ(DPCLKA_CFGCR0_ICL); + for_each_dsi_port(port, intel_dsi->ports) { + tmp |= DPCLKA_CFGCR0_DDI_CLK_OFF(port); + } + + I915_WRITE(DPCLKA_CFGCR0_ICL, tmp); + mutex_unlock(&dev_priv->dpll_lock); +} + +static void gen11_dsi_ungate_clocks(struct intel_encoder *encoder) +{ + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base); + u32 tmp; + enum port port; + + mutex_lock(&dev_priv->dpll_lock); + tmp = I915_READ(DPCLKA_CFGCR0_ICL); + for_each_dsi_port(port, intel_dsi->ports) { + tmp &= ~DPCLKA_CFGCR0_DDI_CLK_OFF(port); + } + + I915_WRITE(DPCLKA_CFGCR0_ICL, tmp); + mutex_unlock(&dev_priv->dpll_lock); +} + +static void gen11_dsi_map_pll(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base); + struct intel_shared_dpll *pll = crtc_state->shared_dpll; + enum port port; + u32 val; + + mutex_lock(&dev_priv->dpll_lock); + + val = I915_READ(DPCLKA_CFGCR0_ICL); + for_each_dsi_port(port, intel_dsi->ports) { + val &= ~DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(port); + val |= DPCLKA_CFGCR0_DDI_CLK_SEL(pll->info->id, port); + } + I915_WRITE(DPCLKA_CFGCR0_ICL, val); + POSTING_READ(DPCLKA_CFGCR0_ICL); + + mutex_unlock(&dev_priv->dpll_lock); +} + static void gen11_dsi_configure_transcoder(struct intel_encoder *encoder, const struct intel_crtc_state *pipe_config) @@ -506,7 +686,8 @@ gen11_dsi_configure_transcoder(struct intel_encoder *encoder, I915_WRITE(TRANS_DDI_FUNC_CTL2(dsi_trans), tmp); } - //TODO: configure DSS_CTL1 + /* configure stream splitting */ + configure_dual_link_mode(encoder, pipe_config); } for_each_dsi_port(port, intel_dsi->ports) { @@ -758,6 +939,9 @@ gen11_dsi_enable_port_and_phy(struct intel_encoder *encoder, /* Step (4h, 4i, 4j, 4k): Configure transcoder */ gen11_dsi_configure_transcoder(encoder, pipe_config); + + /* Step 4l: Gate DDI clocks */ + gen11_dsi_gate_clocks(encoder); } static void gen11_dsi_powerup_panel(struct intel_encoder *encoder) @@ -799,18 +983,25 @@ static void gen11_dsi_powerup_panel(struct intel_encoder *encoder) wait_for_cmds_dispatched_to_panel(encoder); } -static void __attribute__((unused)) -gen11_dsi_pre_enable(struct intel_encoder *encoder, - const struct intel_crtc_state *pipe_config, - const struct drm_connector_state *conn_state) +static void gen11_dsi_pre_pll_enable(struct intel_encoder *encoder, + const struct intel_crtc_state *pipe_config, + const struct drm_connector_state *conn_state) { - struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base); - /* step2: enable IO power */ gen11_dsi_enable_io_power(encoder); /* step3: enable DSI PLL */ gen11_dsi_program_esc_clk_div(encoder); +} + +static void gen11_dsi_pre_enable(struct intel_encoder *encoder, + const struct intel_crtc_state *pipe_config, + const struct drm_connector_state *conn_state) +{ + struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base); + + /* step3b */ + gen11_dsi_map_pll(encoder, pipe_config); /* step4: enable DSI port and DPHY */ gen11_dsi_enable_port_and_phy(encoder, pipe_config); @@ -912,6 +1103,7 @@ static void gen11_dsi_disable_port(struct intel_encoder *encoder) u32 tmp; enum port port; + gen11_dsi_ungate_clocks(encoder); for_each_dsi_port(port, intel_dsi->ports) { tmp = I915_READ(DDI_BUF_CTL(port)); tmp &= ~DDI_BUF_CTL_ENABLE; @@ -923,6 +1115,7 @@ static void gen11_dsi_disable_port(struct intel_encoder *encoder) DRM_ERROR("DDI port:%c buffer not idle\n", port_name(port)); } + gen11_dsi_ungate_clocks(encoder); } static void gen11_dsi_disable_io_power(struct intel_encoder *encoder) @@ -945,10 +1138,9 @@ static void gen11_dsi_disable_io_power(struct intel_encoder *encoder) } } -static void __attribute__((unused)) gen11_dsi_disable( - struct intel_encoder *encoder, - const struct intel_crtc_state *old_crtc_state, - const struct drm_connector_state *old_conn_state) +static void gen11_dsi_disable(struct intel_encoder *encoder, + const struct intel_crtc_state *old_crtc_state, + const struct drm_connector_state *old_conn_state) { struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base); @@ -972,10 +1164,289 @@ static void __attribute__((unused)) gen11_dsi_disable( gen11_dsi_disable_io_power(encoder); } +static void gen11_dsi_get_config(struct intel_encoder *encoder, + struct intel_crtc_state *pipe_config) +{ + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base); + u32 pll_id; + + /* FIXME: adapt icl_ddi_clock_get() for DSI and use that? */ + pll_id = intel_get_shared_dpll_id(dev_priv, pipe_config->shared_dpll); + pipe_config->port_clock = cnl_calc_wrpll_link(dev_priv, pll_id); + pipe_config->base.adjusted_mode.crtc_clock = intel_dsi->pclk; + pipe_config->output_types |= BIT(INTEL_OUTPUT_DSI); +} + +static bool gen11_dsi_compute_config(struct intel_encoder *encoder, + struct intel_crtc_state *pipe_config, + struct drm_connector_state *conn_state) +{ + struct intel_dsi *intel_dsi = container_of(encoder, struct intel_dsi, + base); + struct intel_connector *intel_connector = intel_dsi->attached_connector; + struct intel_crtc *crtc = to_intel_crtc(pipe_config->base.crtc); + const struct drm_display_mode *fixed_mode = + intel_connector->panel.fixed_mode; + struct drm_display_mode *adjusted_mode = + &pipe_config->base.adjusted_mode; + + intel_fixed_panel_mode(fixed_mode, adjusted_mode); + intel_pch_panel_fitting(crtc, pipe_config, conn_state->scaling_mode); + + adjusted_mode->flags = 0; + + /* Dual link goes to trancoder DSI'0' */ + if (intel_dsi->ports == BIT(PORT_B)) + pipe_config->cpu_transcoder = TRANSCODER_DSI_1; + else + pipe_config->cpu_transcoder = TRANSCODER_DSI_0; + + pipe_config->clock_set = true; + pipe_config->port_clock = intel_dsi_bitrate(intel_dsi) / 5; + + return true; +} + +static u64 gen11_dsi_get_power_domains(struct intel_encoder *encoder, + struct intel_crtc_state *crtc_state) +{ + struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base); + u64 domains = 0; + enum port port; + + for_each_dsi_port(port, intel_dsi->ports) + if (port == PORT_A) + domains |= BIT_ULL(POWER_DOMAIN_PORT_DDI_A_IO); + else + domains |= BIT_ULL(POWER_DOMAIN_PORT_DDI_B_IO); + + return domains; +} + +static bool gen11_dsi_get_hw_state(struct intel_encoder *encoder, + enum pipe *pipe) +{ + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base); + u32 tmp; + enum port port; + enum transcoder dsi_trans; + bool ret = false; + + if (!intel_display_power_get_if_enabled(dev_priv, + encoder->power_domain)) + return false; + + for_each_dsi_port(port, intel_dsi->ports) { + dsi_trans = dsi_port_to_transcoder(port); + tmp = I915_READ(TRANS_DDI_FUNC_CTL(dsi_trans)); + switch (tmp & TRANS_DDI_EDP_INPUT_MASK) { + case TRANS_DDI_EDP_INPUT_A_ON: + *pipe = PIPE_A; + break; + case TRANS_DDI_EDP_INPUT_B_ONOFF: + *pipe = PIPE_B; + break; + case TRANS_DDI_EDP_INPUT_C_ONOFF: + *pipe = PIPE_C; + break; + default: + DRM_ERROR("Invalid PIPE input\n"); + goto out; + } + + tmp = I915_READ(PIPECONF(dsi_trans)); + ret = tmp & PIPECONF_ENABLE; + } +out: + intel_display_power_put(dev_priv, encoder->power_domain); + return ret; +} + +static void gen11_dsi_encoder_destroy(struct drm_encoder *encoder) +{ + intel_encoder_destroy(encoder); +} + +static const struct drm_encoder_funcs gen11_dsi_encoder_funcs = { + .destroy = gen11_dsi_encoder_destroy, +}; + +static const struct drm_connector_funcs gen11_dsi_connector_funcs = { + .late_register = intel_connector_register, + .early_unregister = intel_connector_unregister, + .destroy = intel_connector_destroy, + .fill_modes = drm_helper_probe_single_connector_modes, + .atomic_get_property = intel_digital_connector_atomic_get_property, + .atomic_set_property = intel_digital_connector_atomic_set_property, + .atomic_destroy_state = drm_atomic_helper_connector_destroy_state, + .atomic_duplicate_state = intel_digital_connector_duplicate_state, +}; + +static const struct drm_connector_helper_funcs gen11_dsi_connector_helper_funcs = { + .get_modes = intel_dsi_get_modes, + .mode_valid = intel_dsi_mode_valid, + .atomic_check = intel_digital_connector_atomic_check, +}; + +static int gen11_dsi_host_attach(struct mipi_dsi_host *host, + struct mipi_dsi_device *dsi) +{ + return 0; +} + +static int gen11_dsi_host_detach(struct mipi_dsi_host *host, + struct mipi_dsi_device *dsi) +{ + return 0; +} + +static ssize_t gen11_dsi_host_transfer(struct mipi_dsi_host *host, + const struct mipi_dsi_msg *msg) +{ + struct intel_dsi_host *intel_dsi_host = to_intel_dsi_host(host); + struct mipi_dsi_packet dsi_pkt; + ssize_t ret; + bool enable_lpdt = false; + + ret = mipi_dsi_create_packet(&dsi_pkt, msg); + if (ret < 0) + return ret; + + if (msg->flags & MIPI_DSI_MSG_USE_LPM) + enable_lpdt = true; + + /* send packet header */ + ret = dsi_send_pkt_hdr(intel_dsi_host, dsi_pkt, enable_lpdt); + if (ret < 0) + return ret; + + /* only long packet contains payload */ + if (mipi_dsi_packet_format_is_long(msg->type)) { + ret = dsi_send_pkt_payld(intel_dsi_host, dsi_pkt); + if (ret < 0) + return ret; + } + + //TODO: add payload receive code if needed + + ret = sizeof(dsi_pkt.header) + dsi_pkt.payload_length; + + return ret; +} + +static const struct mipi_dsi_host_ops gen11_dsi_host_ops = { + .attach = gen11_dsi_host_attach, + .detach = gen11_dsi_host_detach, + .transfer = gen11_dsi_host_transfer, +}; + void icl_dsi_init(struct drm_i915_private *dev_priv) { + struct drm_device *dev = &dev_priv->drm; + struct intel_dsi *intel_dsi; + struct intel_encoder *encoder; + struct intel_connector *intel_connector; + struct drm_connector *connector; + struct drm_display_mode *scan, *fixed_mode = NULL; enum port port; if (!intel_bios_is_dsi_present(dev_priv, &port)) return; + + intel_dsi = kzalloc(sizeof(*intel_dsi), GFP_KERNEL); + if (!intel_dsi) + return; + + intel_connector = intel_connector_alloc(); + if (!intel_connector) { + kfree(intel_dsi); + return; + } + + encoder = &intel_dsi->base; + intel_dsi->attached_connector = intel_connector; + connector = &intel_connector->base; + + /* register DSI encoder with DRM subsystem */ + drm_encoder_init(dev, &encoder->base, &gen11_dsi_encoder_funcs, + DRM_MODE_ENCODER_DSI, "DSI %c", port_name(port)); + + encoder->pre_pll_enable = gen11_dsi_pre_pll_enable; + encoder->pre_enable = gen11_dsi_pre_enable; + encoder->disable = gen11_dsi_disable; + encoder->port = port; + encoder->get_config = gen11_dsi_get_config; + encoder->compute_config = gen11_dsi_compute_config; + encoder->get_hw_state = gen11_dsi_get_hw_state; + encoder->type = INTEL_OUTPUT_DSI; + encoder->cloneable = 0; + encoder->crtc_mask = BIT(PIPE_A) | BIT(PIPE_B) | BIT(PIPE_C); + encoder->power_domain = POWER_DOMAIN_PORT_DSI; + encoder->get_power_domains = gen11_dsi_get_power_domains; + + /* register DSI connector with DRM subsystem */ + drm_connector_init(dev, connector, &gen11_dsi_connector_funcs, + DRM_MODE_CONNECTOR_DSI); + drm_connector_helper_add(connector, &gen11_dsi_connector_helper_funcs); + connector->display_info.subpixel_order = SubPixelHorizontalRGB; + connector->interlace_allowed = false; + connector->doublescan_allowed = false; + intel_connector->get_hw_state = intel_connector_get_hw_state; + + /* attach connector to encoder */ + intel_connector_attach_encoder(intel_connector, encoder); + + /* fill mode info from VBT */ + mutex_lock(&dev->mode_config.mutex); + intel_dsi_vbt_get_modes(intel_dsi); + list_for_each_entry(scan, &connector->probed_modes, head) { + if (scan->type & DRM_MODE_TYPE_PREFERRED) { + fixed_mode = drm_mode_duplicate(dev, scan); + break; + } + } + mutex_unlock(&dev->mode_config.mutex); + + if (!fixed_mode) { + DRM_ERROR("DSI fixed mode info missing\n"); + goto err; + } + + connector->display_info.width_mm = fixed_mode->width_mm; + connector->display_info.height_mm = fixed_mode->height_mm; + intel_panel_init(&intel_connector->panel, fixed_mode, NULL); + intel_panel_setup_backlight(connector, INVALID_PIPE); + + + if (dev_priv->vbt.dsi.config->dual_link) + intel_dsi->ports = BIT(PORT_A) | BIT(PORT_B); + else + intel_dsi->ports = BIT(port); + + intel_dsi->dcs_backlight_ports = dev_priv->vbt.dsi.bl_ports; + intel_dsi->dcs_cabc_ports = dev_priv->vbt.dsi.cabc_ports; + + for_each_dsi_port(port, intel_dsi->ports) { + struct intel_dsi_host *host; + + host = intel_dsi_host_init(intel_dsi, &gen11_dsi_host_ops, port); + if (!host) + goto err; + + intel_dsi->dsi_hosts[port] = host; + } + + if (!intel_dsi_vbt_init(intel_dsi, MIPI_DSI_GENERIC_PANEL_ID)) { + DRM_DEBUG_KMS("no device found\n"); + goto err; + } + + return; + +err: + drm_encoder_cleanup(&encoder->base); + kfree(intel_dsi); + kfree(intel_connector); } diff --git a/drivers/gpu/drm/i915/intel_atomic.c b/drivers/gpu/drm/i915/intel_atomic.c index a5a2c8fe58a7..8cb02f28d30c 100644 --- a/drivers/gpu/drm/i915/intel_atomic.c +++ b/drivers/gpu/drm/i915/intel_atomic.c @@ -184,6 +184,7 @@ intel_crtc_duplicate_state(struct drm_crtc *crtc) crtc_state->fifo_changed = false; crtc_state->wm.need_postvbl_update = false; crtc_state->fb_bits = 0; + crtc_state->update_planes = 0; return &crtc_state->base; } diff --git a/drivers/gpu/drm/i915/intel_atomic_plane.c b/drivers/gpu/drm/i915/intel_atomic_plane.c index 905f8ef3ba4f..0a73e6e65c20 100644 --- a/drivers/gpu/drm/i915/intel_atomic_plane.c +++ b/drivers/gpu/drm/i915/intel_atomic_plane.c @@ -139,6 +139,9 @@ int intel_plane_atomic_check_with_state(const struct intel_crtc_state *old_crtc_ if (state->visible && state->fb->format->format == DRM_FORMAT_NV12) crtc_state->nv12_planes |= BIT(intel_plane->id); + if (state->visible || old_plane_state->base.visible) + crtc_state->update_planes |= BIT(intel_plane->id); + return intel_plane_atomic_calc_changes(old_crtc_state, &crtc_state->base, old_plane_state, @@ -168,27 +171,75 @@ static int intel_plane_atomic_check(struct drm_plane *plane, to_intel_plane_state(new_plane_state)); } -void intel_update_planes_on_crtc(struct intel_atomic_state *old_state, - struct intel_crtc *crtc, - struct intel_crtc_state *old_crtc_state, - struct intel_crtc_state *new_crtc_state) +static struct intel_plane * +skl_next_plane_to_commit(struct intel_atomic_state *state, + struct intel_crtc *crtc, + struct skl_ddb_entry entries_y[I915_MAX_PLANES], + struct skl_ddb_entry entries_uv[I915_MAX_PLANES], + unsigned int *update_mask) { - struct intel_plane_state *new_plane_state; + struct intel_crtc_state *crtc_state = + intel_atomic_get_new_crtc_state(state, crtc); + struct intel_plane_state *plane_state; struct intel_plane *plane; - u32 update_mask; int i; - update_mask = old_crtc_state->active_planes; - update_mask |= new_crtc_state->active_planes; + if (*update_mask == 0) + return NULL; + + for_each_new_intel_plane_in_state(state, plane, plane_state, i) { + enum plane_id plane_id = plane->id; - for_each_new_intel_plane_in_state(old_state, plane, new_plane_state, i) { if (crtc->pipe != plane->pipe || - !(update_mask & BIT(plane->id))) + !(*update_mask & BIT(plane_id))) + continue; + + if (skl_ddb_allocation_overlaps(&crtc_state->wm.skl.plane_ddb_y[plane_id], + entries_y, + I915_MAX_PLANES, plane_id) || + skl_ddb_allocation_overlaps(&crtc_state->wm.skl.plane_ddb_uv[plane_id], + entries_uv, + I915_MAX_PLANES, plane_id)) continue; + *update_mask &= ~BIT(plane_id); + entries_y[plane_id] = crtc_state->wm.skl.plane_ddb_y[plane_id]; + entries_uv[plane_id] = crtc_state->wm.skl.plane_ddb_uv[plane_id]; + + return plane; + } + + /* should never happen */ + WARN_ON(1); + + return NULL; +} + +void skl_update_planes_on_crtc(struct intel_atomic_state *state, + struct intel_crtc *crtc) +{ + struct intel_crtc_state *old_crtc_state = + intel_atomic_get_old_crtc_state(state, crtc); + struct intel_crtc_state *new_crtc_state = + intel_atomic_get_new_crtc_state(state, crtc); + struct skl_ddb_entry entries_y[I915_MAX_PLANES]; + struct skl_ddb_entry entries_uv[I915_MAX_PLANES]; + u32 update_mask = new_crtc_state->update_planes; + struct intel_plane *plane; + + memcpy(entries_y, old_crtc_state->wm.skl.plane_ddb_y, + sizeof(old_crtc_state->wm.skl.plane_ddb_y)); + memcpy(entries_uv, old_crtc_state->wm.skl.plane_ddb_uv, + sizeof(old_crtc_state->wm.skl.plane_ddb_uv)); + + while ((plane = skl_next_plane_to_commit(state, crtc, + entries_y, entries_uv, + &update_mask))) { + struct intel_plane_state *new_plane_state = + intel_atomic_get_new_plane_state(state, plane); + if (new_plane_state->base.visible) { trace_intel_update_plane(&plane->base, crtc); - plane->update_plane(plane, new_crtc_state, new_plane_state); } else if (new_plane_state->slave) { struct intel_plane *master = @@ -204,15 +255,38 @@ void intel_update_planes_on_crtc(struct intel_atomic_state *old_state, * plane_state. */ new_plane_state = - intel_atomic_get_new_plane_state(old_state, master); + intel_atomic_get_new_plane_state(state, master); trace_intel_update_plane(&plane->base, crtc); - plane->update_slave(plane, new_crtc_state, new_plane_state); } else { trace_intel_disable_plane(&plane->base, crtc); + plane->disable_plane(plane, new_crtc_state); + } + } +} + +void i9xx_update_planes_on_crtc(struct intel_atomic_state *state, + struct intel_crtc *crtc) +{ + struct intel_crtc_state *new_crtc_state = + intel_atomic_get_new_crtc_state(state, crtc); + u32 update_mask = new_crtc_state->update_planes; + struct intel_plane_state *new_plane_state; + struct intel_plane *plane; + int i; - plane->disable_plane(plane, crtc); + for_each_new_intel_plane_in_state(state, plane, new_plane_state, i) { + if (crtc->pipe != plane->pipe || + !(update_mask & BIT(plane->id))) + continue; + + if (new_plane_state->base.visible) { + trace_intel_update_plane(&plane->base, crtc); + plane->update_plane(plane, new_crtc_state, new_plane_state); + } else { + trace_intel_disable_plane(&plane->base, crtc); + plane->disable_plane(plane, new_crtc_state); } } } diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c index 0694aa8bb9bc..6d3e0260d49c 100644 --- a/drivers/gpu/drm/i915/intel_bios.c +++ b/drivers/gpu/drm/i915/intel_bios.c @@ -1752,7 +1752,7 @@ void intel_bios_init(struct drm_i915_private *dev_priv) const struct bdb_header *bdb; u8 __iomem *bios = NULL; - if (INTEL_INFO(dev_priv)->num_pipes == 0) { + if (!HAS_DISPLAY(dev_priv)) { DRM_DEBUG_KMS("Skipping VBT init due to disabled display.\n"); return; } diff --git a/drivers/gpu/drm/i915/intel_breadcrumbs.c b/drivers/gpu/drm/i915/intel_breadcrumbs.c index 84bf8d827136..447c5256f63a 100644 --- a/drivers/gpu/drm/i915/intel_breadcrumbs.c +++ b/drivers/gpu/drm/i915/intel_breadcrumbs.c @@ -27,11 +27,7 @@ #include "i915_drv.h" -#ifdef CONFIG_SMP -#define task_asleep(tsk) ((tsk)->state & TASK_NORMAL && !(tsk)->on_cpu) -#else -#define task_asleep(tsk) ((tsk)->state & TASK_NORMAL) -#endif +#define task_asleep(tsk) ((tsk)->state & TASK_NORMAL && !(tsk)->on_rq) static unsigned int __intel_breadcrumbs_wakeup(struct intel_breadcrumbs *b) { diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c index ad11540ac436..f3e1d6a0b7dd 100644 --- a/drivers/gpu/drm/i915/intel_ddi.c +++ b/drivers/gpu/drm/i915/intel_ddi.c @@ -28,6 +28,7 @@ #include <drm/drm_scdc_helper.h> #include "i915_drv.h" #include "intel_drv.h" +#include "intel_dsi.h" struct ddi_buf_trans { u32 trans1; /* balance leg enable, de-emph level */ @@ -1363,8 +1364,8 @@ static int skl_calc_wrpll_link(struct drm_i915_private *dev_priv, return dco_freq / (p0 * p1 * p2 * 5); } -static int cnl_calc_wrpll_link(struct drm_i915_private *dev_priv, - enum intel_dpll_id pll_id) +int cnl_calc_wrpll_link(struct drm_i915_private *dev_priv, + enum intel_dpll_id pll_id) { uint32_t cfgcr0, cfgcr1; uint32_t p0, p1, p2, dco_freq, ref_clock; @@ -2154,6 +2155,12 @@ static u64 intel_ddi_get_power_domains(struct intel_encoder *encoder, intel_port_is_tc(dev_priv, encoder->port)) domains |= BIT_ULL(intel_ddi_main_link_aux_domain(dig_port)); + /* + * VDSC power is needed when DSC is enabled + */ + if (crtc_state->dsc_params.compression_enable) + domains |= BIT_ULL(intel_dsc_power_domain(crtc_state)); + return domains; } @@ -2785,77 +2792,54 @@ uint32_t icl_dpclka_cfgcr0_clk_off(struct drm_i915_private *dev_priv, return 0; } -void icl_map_plls_to_ports(struct drm_crtc *crtc, - struct intel_crtc_state *crtc_state, - struct drm_atomic_state *old_state) +static void icl_map_plls_to_ports(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) { + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); struct intel_shared_dpll *pll = crtc_state->shared_dpll; - struct drm_i915_private *dev_priv = to_i915(crtc->dev); - struct drm_connector_state *conn_state; - struct drm_connector *conn; - int i; - - for_each_new_connector_in_state(old_state, conn, conn_state, i) { - struct intel_encoder *encoder = - to_intel_encoder(conn_state->best_encoder); - enum port port; - uint32_t val; - - if (conn_state->crtc != crtc) - continue; - - port = encoder->port; - mutex_lock(&dev_priv->dpll_lock); + enum port port = encoder->port; + u32 val; - val = I915_READ(DPCLKA_CFGCR0_ICL); - WARN_ON((val & icl_dpclka_cfgcr0_clk_off(dev_priv, port)) == 0); + mutex_lock(&dev_priv->dpll_lock); - if (intel_port_is_combophy(dev_priv, port)) { - val &= ~DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(port); - val |= DPCLKA_CFGCR0_DDI_CLK_SEL(pll->info->id, port); - I915_WRITE(DPCLKA_CFGCR0_ICL, val); - POSTING_READ(DPCLKA_CFGCR0_ICL); - } + val = I915_READ(DPCLKA_CFGCR0_ICL); + WARN_ON((val & icl_dpclka_cfgcr0_clk_off(dev_priv, port)) == 0); - val &= ~icl_dpclka_cfgcr0_clk_off(dev_priv, port); + if (intel_port_is_combophy(dev_priv, port)) { + val &= ~DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(port); + val |= DPCLKA_CFGCR0_DDI_CLK_SEL(pll->info->id, port); I915_WRITE(DPCLKA_CFGCR0_ICL, val); - - mutex_unlock(&dev_priv->dpll_lock); + POSTING_READ(DPCLKA_CFGCR0_ICL); } + + val &= ~icl_dpclka_cfgcr0_clk_off(dev_priv, port); + I915_WRITE(DPCLKA_CFGCR0_ICL, val); + + mutex_unlock(&dev_priv->dpll_lock); } -void icl_unmap_plls_to_ports(struct drm_crtc *crtc, - struct intel_crtc_state *crtc_state, - struct drm_atomic_state *old_state) +static void icl_unmap_plls_to_ports(struct intel_encoder *encoder) { - struct drm_i915_private *dev_priv = to_i915(crtc->dev); - struct drm_connector_state *old_conn_state; - struct drm_connector *conn; - int i; + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + enum port port = encoder->port; + u32 val; - for_each_old_connector_in_state(old_state, conn, old_conn_state, i) { - struct intel_encoder *encoder = - to_intel_encoder(old_conn_state->best_encoder); - enum port port; + mutex_lock(&dev_priv->dpll_lock); - if (old_conn_state->crtc != crtc) - continue; + val = I915_READ(DPCLKA_CFGCR0_ICL); + val |= icl_dpclka_cfgcr0_clk_off(dev_priv, port); + I915_WRITE(DPCLKA_CFGCR0_ICL, val); - port = encoder->port; - mutex_lock(&dev_priv->dpll_lock); - I915_WRITE(DPCLKA_CFGCR0_ICL, - I915_READ(DPCLKA_CFGCR0_ICL) | - icl_dpclka_cfgcr0_clk_off(dev_priv, port)); - mutex_unlock(&dev_priv->dpll_lock); - } + mutex_unlock(&dev_priv->dpll_lock); } void icl_sanitize_encoder_pll_mapping(struct intel_encoder *encoder) { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); u32 val; - enum port port = encoder->port; - bool clk_enabled; + enum port port; + u32 port_mask; + bool ddi_clk_needed; /* * In case of DP MST, we sanitize the primary encoder only, not the @@ -2864,9 +2848,6 @@ void icl_sanitize_encoder_pll_mapping(struct intel_encoder *encoder) if (encoder->type == INTEL_OUTPUT_DP_MST) return; - val = I915_READ(DPCLKA_CFGCR0_ICL); - clk_enabled = !(val & icl_dpclka_cfgcr0_clk_off(dev_priv, port)); - if (!encoder->base.crtc && intel_encoder_is_dp(encoder)) { u8 pipe_mask; bool is_mst; @@ -2880,20 +2861,52 @@ void icl_sanitize_encoder_pll_mapping(struct intel_encoder *encoder) return; } - if (clk_enabled == !!encoder->base.crtc) - return; + port_mask = BIT(encoder->port); + ddi_clk_needed = encoder->base.crtc; - /* - * Punt on the case now where clock is disabled, but the encoder is - * enabled, something else is really broken then. - */ - if (WARN_ON(!clk_enabled)) - return; + if (encoder->type == INTEL_OUTPUT_DSI) { + struct intel_encoder *other_encoder; - DRM_NOTE("Port %c is disabled but it has a mapped PLL, unmap it\n", - port_name(port)); - val |= icl_dpclka_cfgcr0_clk_off(dev_priv, port); - I915_WRITE(DPCLKA_CFGCR0_ICL, val); + port_mask = intel_dsi_encoder_ports(encoder); + /* + * Sanity check that we haven't incorrectly registered another + * encoder using any of the ports of this DSI encoder. + */ + for_each_intel_encoder(&dev_priv->drm, other_encoder) { + if (other_encoder == encoder) + continue; + + if (WARN_ON(port_mask & BIT(other_encoder->port))) + return; + } + /* + * DSI ports should have their DDI clock ungated when disabled + * and gated when enabled. + */ + ddi_clk_needed = !encoder->base.crtc; + } + + val = I915_READ(DPCLKA_CFGCR0_ICL); + for_each_port_masked(port, port_mask) { + bool ddi_clk_ungated = !(val & + icl_dpclka_cfgcr0_clk_off(dev_priv, + port)); + + if (ddi_clk_needed == ddi_clk_ungated) + continue; + + /* + * Punt on the case now where clock is gated, but it would + * be needed by the port. Something else is really broken then. + */ + if (WARN_ON(ddi_clk_needed)) + continue; + + DRM_NOTE("Port %c is disabled/in DSI mode with an ungated DDI clock, gate it\n", + port_name(port)); + val |= icl_dpclka_cfgcr0_clk_off(dev_priv, port); + I915_WRITE(DPCLKA_CFGCR0_ICL, val); + } } static void intel_ddi_clk_select(struct intel_encoder *encoder, @@ -3096,6 +3109,53 @@ static void icl_program_mg_dp_mode(struct intel_digital_port *intel_dig_port) I915_WRITE(MG_DP_MODE(port, 1), ln1); } +static void intel_dp_sink_set_fec_ready(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state) +{ + if (!crtc_state->fec_enable) + return; + + if (drm_dp_dpcd_writeb(&intel_dp->aux, DP_FEC_CONFIGURATION, DP_FEC_READY) <= 0) + DRM_DEBUG_KMS("Failed to set FEC_READY in the sink\n"); +} + +static void intel_ddi_enable_fec(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + enum port port = encoder->port; + u32 val; + + if (!crtc_state->fec_enable) + return; + + val = I915_READ(DP_TP_CTL(port)); + val |= DP_TP_CTL_FEC_ENABLE; + I915_WRITE(DP_TP_CTL(port), val); + + if (intel_wait_for_register(dev_priv, DP_TP_STATUS(port), + DP_TP_STATUS_FEC_ENABLE_LIVE, + DP_TP_STATUS_FEC_ENABLE_LIVE, + 1)) + DRM_ERROR("Timed out waiting for FEC Enable Status\n"); +} + +static void intel_ddi_disable_fec_state(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) +{ + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + enum port port = encoder->port; + u32 val; + + if (!crtc_state->fec_enable) + return; + + val = I915_READ(DP_TP_CTL(port)); + val &= ~DP_TP_CTL_FEC_ENABLE; + I915_WRITE(DP_TP_CTL(port), val); + POSTING_READ(DP_TP_CTL(port)); +} + static void intel_ddi_pre_enable_dp(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state, const struct drm_connector_state *conn_state) @@ -3134,14 +3194,21 @@ static void intel_ddi_pre_enable_dp(struct intel_encoder *encoder, intel_ddi_init_dp_buf_reg(encoder); if (!is_mst) intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON); + intel_dp_sink_set_decompression_state(intel_dp, crtc_state, + true); + intel_dp_sink_set_fec_ready(intel_dp, crtc_state); intel_dp_start_link_train(intel_dp); if (port != PORT_A || INTEL_GEN(dev_priv) >= 9) intel_dp_stop_link_train(intel_dp); + intel_ddi_enable_fec(encoder, crtc_state); + icl_enable_phy_clock_gating(dig_port); if (!is_mst) intel_ddi_enable_pipe_clock(crtc_state); + + intel_dsc_enable(encoder, crtc_state); } static void intel_ddi_pre_enable_hdmi(struct intel_encoder *encoder, @@ -3208,6 +3275,9 @@ static void intel_ddi_pre_enable(struct intel_encoder *encoder, WARN_ON(crtc_state->has_pch_encoder); + if (INTEL_GEN(dev_priv) >= 11) + icl_map_plls_to_ports(encoder, crtc_state); + intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, true); if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) { @@ -3228,7 +3298,8 @@ static void intel_ddi_pre_enable(struct intel_encoder *encoder, } } -static void intel_disable_ddi_buf(struct intel_encoder *encoder) +static void intel_disable_ddi_buf(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) { struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); enum port port = encoder->port; @@ -3247,6 +3318,9 @@ static void intel_disable_ddi_buf(struct intel_encoder *encoder) val |= DP_TP_CTL_LINK_TRAIN_PAT1; I915_WRITE(DP_TP_CTL(port), val); + /* Disable FEC in DP Sink */ + intel_ddi_disable_fec_state(encoder, crtc_state); + if (wait) intel_wait_ddi_buf_idle(dev_priv, port); } @@ -3270,7 +3344,7 @@ static void intel_ddi_post_disable_dp(struct intel_encoder *encoder, intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_OFF); } - intel_disable_ddi_buf(encoder); + intel_disable_ddi_buf(encoder, old_crtc_state); intel_edp_panel_vdd_on(intel_dp); intel_edp_panel_off(intel_dp); @@ -3293,7 +3367,7 @@ static void intel_ddi_post_disable_hdmi(struct intel_encoder *encoder, intel_ddi_disable_pipe_clock(old_crtc_state); - intel_disable_ddi_buf(encoder); + intel_disable_ddi_buf(encoder, old_crtc_state); intel_display_power_put(dev_priv, dig_port->ddi_io_power_domain); @@ -3306,6 +3380,8 @@ static void intel_ddi_post_disable(struct intel_encoder *encoder, const struct intel_crtc_state *old_crtc_state, const struct drm_connector_state *old_conn_state) { + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + /* * When called from DP MST code: * - old_conn_state will be NULL @@ -3325,6 +3401,9 @@ static void intel_ddi_post_disable(struct intel_encoder *encoder, else intel_ddi_post_disable_dp(encoder, old_crtc_state, old_conn_state); + + if (INTEL_GEN(dev_priv) >= 11) + icl_unmap_plls_to_ports(encoder); } void intel_ddi_fdi_post_disable(struct intel_encoder *encoder, @@ -3344,7 +3423,7 @@ void intel_ddi_fdi_post_disable(struct intel_encoder *encoder, val &= ~FDI_RX_ENABLE; I915_WRITE(FDI_RX_CTL(PIPE_A), val); - intel_disable_ddi_buf(encoder); + intel_disable_ddi_buf(encoder, old_crtc_state); intel_ddi_clk_disable(encoder); val = I915_READ(FDI_RX_MISC(PIPE_A)); @@ -3491,6 +3570,9 @@ static void intel_disable_ddi_dp(struct intel_encoder *encoder, intel_edp_drrs_disable(intel_dp, old_crtc_state); intel_psr_disable(intel_dp, old_crtc_state); intel_edp_backlight_off(old_conn_state); + /* Disable the decompression in DP Sink */ + intel_dp_sink_set_decompression_state(intel_dp, old_crtc_state, + false); } static void intel_disable_ddi_hdmi(struct intel_encoder *encoder, diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c index ceecb5bd5226..1e56319334f3 100644 --- a/drivers/gpu/drm/i915/intel_device_info.c +++ b/drivers/gpu/drm/i915/intel_device_info.c @@ -77,6 +77,10 @@ void intel_device_info_dump_flags(const struct intel_device_info *info, #define PRINT_FLAG(name) drm_printf(p, "%s: %s\n", #name, yesno(info->name)); DEV_INFO_FOR_EACH_FLAG(PRINT_FLAG); #undef PRINT_FLAG + +#define PRINT_FLAG(name) drm_printf(p, "%s: %s\n", #name, yesno(info->display.name)); + DEV_INFO_DISPLAY_FOR_EACH_FLAG(PRINT_FLAG); +#undef PRINT_FLAG } static void sseu_dump(const struct sseu_dev_info *sseu, struct drm_printer *p) @@ -782,7 +786,7 @@ void intel_device_info_runtime_init(struct intel_device_info *info) if (i915_modparams.disable_display) { DRM_INFO("Display disabled (module parameter)\n"); info->num_pipes = 0; - } else if (info->num_pipes > 0 && + } else if (HAS_DISPLAY(dev_priv) && (IS_GEN7(dev_priv) || IS_GEN8(dev_priv)) && HAS_PCH_SPLIT(dev_priv)) { u32 fuse_strap = I915_READ(FUSE_STRAP); @@ -807,7 +811,7 @@ void intel_device_info_runtime_init(struct intel_device_info *info) DRM_INFO("PipeC fused off\n"); info->num_pipes -= 1; } - } else if (info->num_pipes > 0 && IS_GEN9(dev_priv)) { + } else if (HAS_DISPLAY(dev_priv) && IS_GEN9(dev_priv)) { u32 dfsm = I915_READ(SKL_DFSM); u8 disabled_mask = 0; bool invalid; diff --git a/drivers/gpu/drm/i915/intel_device_info.h b/drivers/gpu/drm/i915/intel_device_info.h index 88f97210dc49..1caf24e2cf0b 100644 --- a/drivers/gpu/drm/i915/intel_device_info.h +++ b/drivers/gpu/drm/i915/intel_device_info.h @@ -89,35 +89,38 @@ enum intel_ppgtt { func(is_alpha_support); \ /* Keep has_* in alphabetical order */ \ func(has_64bit_reloc); \ - func(has_csr); \ - func(has_ddi); \ - func(has_dp_mst); \ func(has_reset_engine); \ - func(has_fbc); \ func(has_fpga_dbg); \ - func(has_gmch_display); \ func(has_guc); \ func(has_guc_ct); \ - func(has_hotplug); \ func(has_l3_dpf); \ func(has_llc); \ func(has_logical_ring_contexts); \ func(has_logical_ring_elsq); \ func(has_logical_ring_preemption); \ - func(has_overlay); \ func(has_pooled_eu); \ - func(has_psr); \ func(has_rc6); \ func(has_rc6p); \ func(has_runtime_pm); \ func(has_snoop); \ func(has_coherent_ggtt); \ func(unfenced_needs_alignment); \ + func(hws_needs_physical); + +#define DEV_INFO_DISPLAY_FOR_EACH_FLAG(func) \ + /* Keep in alphabetical order */ \ func(cursor_needs_physical); \ - func(hws_needs_physical); \ + func(has_csr); \ + func(has_ddi); \ + func(has_dp_mst); \ + func(has_fbc); \ + func(has_gmch_display); \ + func(has_hotplug); \ + func(has_ipc); \ + func(has_overlay); \ + func(has_psr); \ func(overlay_needs_physical); \ - func(supports_tv); \ - func(has_ipc); + func(supports_tv); #define GEN_MAX_SLICES (6) /* CNL upper bound */ #define GEN_MAX_SUBSLICES (8) /* ICL upper bound */ @@ -172,6 +175,13 @@ struct intel_device_info { #define DEFINE_FLAG(name) u8 name:1 DEV_INFO_FOR_EACH_FLAG(DEFINE_FLAG); #undef DEFINE_FLAG + + struct { +#define DEFINE_FLAG(name) u8 name:1 + DEV_INFO_DISPLAY_FOR_EACH_FLAG(DEFINE_FLAG); +#undef DEFINE_FLAG + } display; + u16 ddb_size; /* in blocks */ /* Register offsets for the various display pipes and transcoders */ diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index 5a679af03a04..20beb1977a27 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -46,7 +46,7 @@ #include <drm/drm_plane_helper.h> #include <drm/drm_rect.h> #include <drm/drm_atomic_uapi.h> -#include <linux/dma_remapping.h> +#include <linux/intel-iommu.h> #include <linux/reservation.h> /* Primary plane formats for gen <= 3 */ @@ -2341,10 +2341,26 @@ static int intel_fb_offset_to_xy(int *x, int *y, int color_plane) { struct drm_i915_private *dev_priv = to_i915(fb->dev); + unsigned int height; if (fb->modifier != DRM_FORMAT_MOD_LINEAR && - fb->offsets[color_plane] % intel_tile_size(dev_priv)) + fb->offsets[color_plane] % intel_tile_size(dev_priv)) { + DRM_DEBUG_KMS("Misaligned offset 0x%08x for color plane %d\n", + fb->offsets[color_plane], color_plane); return -EINVAL; + } + + height = drm_framebuffer_plane_height(fb->height, fb, color_plane); + height = ALIGN(height, intel_tile_height(fb, color_plane)); + + /* Catch potential overflows early */ + if (add_overflows_t(u32, mul_u32_u32(height, fb->pitches[color_plane]), + fb->offsets[color_plane])) { + DRM_DEBUG_KMS("Bad offset 0x%08x or pitch %d for color plane %d\n", + fb->offsets[color_plane], fb->pitches[color_plane], + color_plane); + return -ERANGE; + } *x = 0; *y = 0; @@ -2767,7 +2783,7 @@ static void intel_plane_disable_noatomic(struct intel_crtc *crtc, intel_pre_disable_primary_noatomic(&crtc->base); trace_intel_disable_plane(&plane->base, crtc); - plane->disable_plane(plane, crtc); + plane->disable_plane(plane, crtc_state); } static void @@ -3315,7 +3331,6 @@ static void i9xx_update_plane(struct intel_plane *plane, enum i9xx_plane_id i9xx_plane = plane->i9xx_plane; u32 linear_offset; u32 dspcntr = plane_state->ctl; - i915_reg_t reg = DSPCNTR(i9xx_plane); int x = plane_state->color_plane[0].x; int y = plane_state->color_plane[0].y; unsigned long irqflags; @@ -3330,47 +3345,51 @@ static void i9xx_update_plane(struct intel_plane *plane, spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); + I915_WRITE_FW(DSPSTRIDE(i9xx_plane), plane_state->color_plane[0].stride); + if (INTEL_GEN(dev_priv) < 4) { /* pipesrc and dspsize control the size that is scaled from, * which should always be the user's requested size. */ + I915_WRITE_FW(DSPPOS(i9xx_plane), 0); I915_WRITE_FW(DSPSIZE(i9xx_plane), ((crtc_state->pipe_src_h - 1) << 16) | (crtc_state->pipe_src_w - 1)); - I915_WRITE_FW(DSPPOS(i9xx_plane), 0); } else if (IS_CHERRYVIEW(dev_priv) && i9xx_plane == PLANE_B) { + I915_WRITE_FW(PRIMPOS(i9xx_plane), 0); I915_WRITE_FW(PRIMSIZE(i9xx_plane), ((crtc_state->pipe_src_h - 1) << 16) | (crtc_state->pipe_src_w - 1)); - I915_WRITE_FW(PRIMPOS(i9xx_plane), 0); I915_WRITE_FW(PRIMCNSTALPHA(i9xx_plane), 0); } - I915_WRITE_FW(reg, dspcntr); - - I915_WRITE_FW(DSPSTRIDE(i9xx_plane), plane_state->color_plane[0].stride); if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) { - I915_WRITE_FW(DSPSURF(i9xx_plane), - intel_plane_ggtt_offset(plane_state) + - dspaddr_offset); I915_WRITE_FW(DSPOFFSET(i9xx_plane), (y << 16) | x); } else if (INTEL_GEN(dev_priv) >= 4) { + I915_WRITE_FW(DSPLINOFF(i9xx_plane), linear_offset); + I915_WRITE_FW(DSPTILEOFF(i9xx_plane), (y << 16) | x); + } + + /* + * The control register self-arms if the plane was previously + * disabled. Try to make the plane enable atomic by writing + * the control register just before the surface register. + */ + I915_WRITE_FW(DSPCNTR(i9xx_plane), dspcntr); + if (INTEL_GEN(dev_priv) >= 4) I915_WRITE_FW(DSPSURF(i9xx_plane), intel_plane_ggtt_offset(plane_state) + dspaddr_offset); - I915_WRITE_FW(DSPTILEOFF(i9xx_plane), (y << 16) | x); - I915_WRITE_FW(DSPLINOFF(i9xx_plane), linear_offset); - } else { + else I915_WRITE_FW(DSPADDR(i9xx_plane), intel_plane_ggtt_offset(plane_state) + dspaddr_offset); - } spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags); } static void i9xx_disable_plane(struct intel_plane *plane, - struct intel_crtc *crtc) + const struct intel_crtc_state *crtc_state) { struct drm_i915_private *dev_priv = to_i915(plane->base.dev); enum i9xx_plane_id i9xx_plane = plane->i9xx_plane; @@ -3456,6 +3475,21 @@ static void skl_detach_scalers(const struct intel_crtc_state *crtc_state) } } +static unsigned int skl_plane_stride_mult(const struct drm_framebuffer *fb, + int color_plane, unsigned int rotation) +{ + /* + * The stride is either expressed as a multiple of 64 bytes chunks for + * linear buffers or in number of tiles for tiled buffers. + */ + if (fb->modifier == DRM_FORMAT_MOD_LINEAR) + return 64; + else if (drm_rotation_90_or_270(rotation)) + return intel_tile_height(fb, color_plane); + else + return intel_tile_width_bytes(fb, color_plane); +} + u32 skl_plane_stride(const struct intel_plane_state *plane_state, int color_plane) { @@ -3466,16 +3500,7 @@ u32 skl_plane_stride(const struct intel_plane_state *plane_state, if (color_plane >= fb->format->num_planes) return 0; - /* - * The stride is either expressed as a multiple of 64 bytes chunks for - * linear buffers or in number of tiles for tiled buffers. - */ - if (drm_rotation_90_or_270(rotation)) - stride /= intel_tile_height(fb, color_plane); - else - stride /= intel_fb_stride_alignment(fb, color_plane); - - return stride; + return stride / skl_plane_stride_mult(fb, color_plane, rotation); } static u32 skl_plane_ctl_format(uint32_t pixel_format) @@ -5403,23 +5428,32 @@ static void intel_pre_plane_update(struct intel_crtc_state *old_crtc_state, intel_update_watermarks(crtc); } -static void intel_crtc_disable_planes(struct intel_crtc *crtc, unsigned plane_mask) +static void intel_crtc_disable_planes(struct intel_atomic_state *state, + struct intel_crtc *crtc) { - struct drm_device *dev = crtc->base.dev; + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + const struct intel_crtc_state *new_crtc_state = + intel_atomic_get_new_crtc_state(state, crtc); + unsigned int update_mask = new_crtc_state->update_planes; + const struct intel_plane_state *old_plane_state; struct intel_plane *plane; unsigned fb_bits = 0; + int i; intel_crtc_dpms_overlay_disable(crtc); - for_each_intel_plane_on_crtc(dev, crtc, plane) { - if (plane_mask & BIT(plane->id)) { - plane->disable_plane(plane, crtc); + for_each_old_intel_plane_in_state(state, plane, old_plane_state, i) { + if (crtc->pipe != plane->pipe || + !(update_mask & BIT(plane->id))) + continue; + plane->disable_plane(plane, new_crtc_state); + + if (old_plane_state->base.visible) fb_bits |= plane->frontbuffer_bit; - } } - intel_frontbuffer_flip(to_i915(dev), fb_bits); + intel_frontbuffer_flip(dev_priv, fb_bits); } static void intel_encoders_pre_pll_enable(struct drm_crtc *crtc, @@ -5692,9 +5726,6 @@ static void haswell_crtc_enable(struct intel_crtc_state *pipe_config, if (pipe_config->shared_dpll) intel_enable_shared_dpll(pipe_config); - if (INTEL_GEN(dev_priv) >= 11) - icl_map_plls_to_ports(crtc, pipe_config, old_state); - intel_encoders_pre_enable(crtc, pipe_config, old_state); if (intel_crtc_has_dp_encoder(pipe_config)) @@ -5889,6 +5920,8 @@ static void haswell_crtc_disable(struct intel_crtc_state *old_crtc_state, if (!transcoder_is_dsi(cpu_transcoder)) intel_ddi_disable_transcoder_func(old_crtc_state); + intel_dsc_disable(old_crtc_state); + if (INTEL_GEN(dev_priv) >= 9) skylake_scaler_disable(intel_crtc); else @@ -5896,9 +5929,6 @@ static void haswell_crtc_disable(struct intel_crtc_state *old_crtc_state, intel_encoders_post_disable(crtc, old_crtc_state, old_state); - if (INTEL_GEN(dev_priv) >= 11) - icl_unmap_plls_to_ports(crtc, old_crtc_state, old_state); - intel_encoders_post_pll_disable(crtc, old_crtc_state, old_state); } @@ -6724,7 +6754,7 @@ static void compute_m_n(unsigned int m, unsigned int n, } void -intel_link_compute_m_n(int bits_per_pixel, int nlanes, +intel_link_compute_m_n(u16 bits_per_pixel, int nlanes, int pixel_clock, int link_clock, struct intel_link_m_n *m_n, bool constant_n) @@ -8939,7 +8969,7 @@ skylake_get_initial_plane_config(struct intel_crtc *crtc, fb->width = ((val >> 0) & 0x1fff) + 1; val = I915_READ(PLANE_STRIDE(pipe, plane_id)); - stride_mult = intel_fb_stride_alignment(fb, 0); + stride_mult = skl_plane_stride_mult(fb, 0, DRM_MODE_ROTATE_0); fb->pitches[0] = (val & 0x3ff) * stride_mult; aligned_height = intel_fb_align_height(fb, 0, fb->height); @@ -9303,10 +9333,12 @@ void hsw_disable_pc8(struct drm_i915_private *dev_priv) static int haswell_crtc_compute_clock(struct intel_crtc *crtc, struct intel_crtc_state *crtc_state) { + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); struct intel_atomic_state *state = to_intel_atomic_state(crtc_state->base.state); - if (!intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DSI)) { + if (!intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DSI) || + IS_ICELAKE(dev_priv)) { struct intel_encoder *encoder = intel_get_crtc_new_encoder(state, crtc_state); @@ -9444,11 +9476,18 @@ static bool hsw_get_transcoder_state(struct intel_crtc *crtc, struct drm_device *dev = crtc->base.dev; struct drm_i915_private *dev_priv = to_i915(dev); enum intel_display_power_domain power_domain; + unsigned long panel_transcoder_mask = BIT(TRANSCODER_EDP); + unsigned long enabled_panel_transcoders = 0; + enum transcoder panel_transcoder; u32 tmp; + if (IS_ICELAKE(dev_priv)) + panel_transcoder_mask |= + BIT(TRANSCODER_DSI_0) | BIT(TRANSCODER_DSI_1); + /* * The pipe->transcoder mapping is fixed with the exception of the eDP - * transcoder handled below. + * and DSI transcoders handled below. */ pipe_config->cpu_transcoder = (enum transcoder) crtc->pipe; @@ -9456,29 +9495,49 @@ static bool hsw_get_transcoder_state(struct intel_crtc *crtc, * XXX: Do intel_display_power_get_if_enabled before reading this (for * consistency and less surprising code; it's in always on power). */ - tmp = I915_READ(TRANS_DDI_FUNC_CTL(TRANSCODER_EDP)); - if (tmp & TRANS_DDI_FUNC_ENABLE) { - enum pipe trans_edp_pipe; + for_each_set_bit(panel_transcoder, &panel_transcoder_mask, 32) { + enum pipe trans_pipe; + + tmp = I915_READ(TRANS_DDI_FUNC_CTL(panel_transcoder)); + if (!(tmp & TRANS_DDI_FUNC_ENABLE)) + continue; + + /* + * Log all enabled ones, only use the first one. + * + * FIXME: This won't work for two separate DSI displays. + */ + enabled_panel_transcoders |= BIT(panel_transcoder); + if (enabled_panel_transcoders != BIT(panel_transcoder)) + continue; + switch (tmp & TRANS_DDI_EDP_INPUT_MASK) { default: - WARN(1, "unknown pipe linked to edp transcoder\n"); + WARN(1, "unknown pipe linked to transcoder %s\n", + transcoder_name(panel_transcoder)); /* fall through */ case TRANS_DDI_EDP_INPUT_A_ONOFF: case TRANS_DDI_EDP_INPUT_A_ON: - trans_edp_pipe = PIPE_A; + trans_pipe = PIPE_A; break; case TRANS_DDI_EDP_INPUT_B_ONOFF: - trans_edp_pipe = PIPE_B; + trans_pipe = PIPE_B; break; case TRANS_DDI_EDP_INPUT_C_ONOFF: - trans_edp_pipe = PIPE_C; + trans_pipe = PIPE_C; break; } - if (trans_edp_pipe == crtc->pipe) - pipe_config->cpu_transcoder = TRANSCODER_EDP; + if (trans_pipe == crtc->pipe) + pipe_config->cpu_transcoder = panel_transcoder; } + /* + * Valid combos: none, eDP, DSI0, DSI1, DSI0+DSI1 + */ + WARN_ON((enabled_panel_transcoders & BIT(TRANSCODER_EDP)) && + enabled_panel_transcoders != BIT(TRANSCODER_EDP)); + power_domain = POWER_DOMAIN_TRANSCODER(pipe_config->cpu_transcoder); if (!intel_display_power_get_if_enabled(dev_priv, power_domain)) return false; @@ -9611,7 +9670,8 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc, if (!active) goto out; - if (!transcoder_is_dsi(pipe_config->cpu_transcoder)) { + if (!transcoder_is_dsi(pipe_config->cpu_transcoder) || + IS_ICELAKE(dev_priv)) { haswell_get_ddi_port_state(crtc, pipe_config); intel_get_pipe_timings(crtc, pipe_config); } @@ -9667,7 +9727,7 @@ static u32 intel_cursor_base(const struct intel_plane_state *plane_state) const struct drm_i915_gem_object *obj = intel_fb_obj(fb); u32 base; - if (INTEL_INFO(dev_priv)->cursor_needs_physical) + if (INTEL_INFO(dev_priv)->display.cursor_needs_physical) base = obj->phys_handle->busaddr; else base = intel_plane_ggtt_offset(plane_state); @@ -9894,9 +9954,9 @@ static void i845_update_cursor(struct intel_plane *plane, } static void i845_disable_cursor(struct intel_plane *plane, - struct intel_crtc *crtc) + const struct intel_crtc_state *crtc_state) { - i845_update_cursor(plane, NULL, NULL); + i845_update_cursor(plane, crtc_state, NULL); } static bool i845_cursor_get_hw_state(struct intel_plane *plane, @@ -10087,8 +10147,8 @@ static void i9xx_update_cursor(struct intel_plane *plane, * On some platforms writing CURCNTR first will also * cause CURPOS to be armed by the CURBASE write. * Without the CURCNTR write the CURPOS write would - * arm itself. Thus we always start the full update - * with a CURCNTR write. + * arm itself. Thus we always update CURCNTR before + * CURPOS. * * On other platforms CURPOS always requires the * CURBASE write to arm the update. Additonally @@ -10098,15 +10158,20 @@ static void i9xx_update_cursor(struct intel_plane *plane, * cursor that doesn't appear to move, or even change * shape. Thus we always write CURBASE. * - * CURCNTR and CUR_FBC_CTL are always - * armed by the CURBASE write only. + * The other registers are armed by by the CURBASE write + * except when the plane is getting enabled at which time + * the CURCNTR write arms the update. */ + + if (INTEL_GEN(dev_priv) >= 9) + skl_write_cursor_wm(plane, crtc_state); + if (plane->cursor.base != base || plane->cursor.size != fbc_ctl || plane->cursor.cntl != cntl) { - I915_WRITE_FW(CURCNTR(pipe), cntl); if (HAS_CUR_FBC(dev_priv)) I915_WRITE_FW(CUR_FBC_CTL(pipe), fbc_ctl); + I915_WRITE_FW(CURCNTR(pipe), cntl); I915_WRITE_FW(CURPOS(pipe), pos); I915_WRITE_FW(CURBASE(pipe), base); @@ -10122,9 +10187,9 @@ static void i9xx_update_cursor(struct intel_plane *plane, } static void i9xx_disable_cursor(struct intel_plane *plane, - struct intel_crtc *crtc) + const struct intel_crtc_state *crtc_state) { - i9xx_update_cursor(plane, NULL, NULL); + i9xx_update_cursor(plane, crtc_state, NULL); } static bool i9xx_cursor_get_hw_state(struct intel_plane *plane, @@ -10835,8 +10900,10 @@ static int icl_check_nv12_planes(struct intel_crtc_state *crtc_state) continue; plane_state->linked_plane = NULL; - if (plane_state->slave && !plane_state->base.visible) + if (plane_state->slave && !plane_state->base.visible) { crtc_state->active_planes &= ~BIT(plane->id); + crtc_state->update_planes |= BIT(plane->id); + } plane_state->slave = false; } @@ -10877,6 +10944,7 @@ static int icl_check_nv12_planes(struct intel_crtc_state *crtc_state) linked_state->slave = true; linked_state->linked_plane = plane; crtc_state->active_planes |= BIT(linked->id); + crtc_state->update_planes |= BIT(linked->id); DRM_DEBUG_KMS("Using %s as Y plane for %s\n", linked->base.name, plane->base.name); } @@ -11887,6 +11955,8 @@ static void verify_wm_state(struct drm_crtc *crtc, struct skl_pipe_wm hw_wm, *sw_wm; struct skl_plane_wm *hw_plane_wm, *sw_plane_wm; struct skl_ddb_entry *hw_ddb_entry, *sw_ddb_entry; + struct skl_ddb_entry hw_ddb_y[I915_MAX_PLANES]; + struct skl_ddb_entry hw_ddb_uv[I915_MAX_PLANES]; struct intel_crtc *intel_crtc = to_intel_crtc(crtc); const enum pipe pipe = intel_crtc->pipe; int plane, level, max_level = ilk_wm_max_level(dev_priv); @@ -11897,6 +11967,8 @@ static void verify_wm_state(struct drm_crtc *crtc, skl_pipe_wm_get_hw_state(crtc, &hw_wm); sw_wm = &to_intel_crtc_state(new_state)->wm.skl.optimal; + skl_pipe_ddb_get_hw_state(intel_crtc, hw_ddb_y, hw_ddb_uv); + skl_ddb_get_hw_state(dev_priv, &hw_ddb); sw_ddb = &dev_priv->wm.skl_hw.ddb; @@ -11939,8 +12011,8 @@ static void verify_wm_state(struct drm_crtc *crtc, } /* DDB */ - hw_ddb_entry = &hw_ddb.plane[pipe][plane]; - sw_ddb_entry = &sw_ddb->plane[pipe][plane]; + hw_ddb_entry = &hw_ddb_y[plane]; + sw_ddb_entry = &to_intel_crtc_state(new_state)->wm.skl.plane_ddb_y[plane]; if (!skl_ddb_entry_equal(hw_ddb_entry, sw_ddb_entry)) { DRM_ERROR("mismatch in DDB state pipe %c plane %d (expected (%u,%u), found (%u,%u))\n", @@ -11989,8 +12061,8 @@ static void verify_wm_state(struct drm_crtc *crtc, } /* DDB */ - hw_ddb_entry = &hw_ddb.plane[pipe][PLANE_CURSOR]; - sw_ddb_entry = &sw_ddb->plane[pipe][PLANE_CURSOR]; + hw_ddb_entry = &hw_ddb_y[PLANE_CURSOR]; + sw_ddb_entry = &to_intel_crtc_state(new_state)->wm.skl.plane_ddb_y[PLANE_CURSOR]; if (!skl_ddb_entry_equal(hw_ddb_entry, sw_ddb_entry)) { DRM_ERROR("mismatch in DDB state pipe %c cursor (expected (%u,%u), found (%u,%u))\n", @@ -12672,7 +12744,6 @@ static void intel_update_crtc(struct drm_crtc *crtc, struct drm_device *dev = crtc->dev; struct drm_i915_private *dev_priv = to_i915(dev); struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - struct intel_crtc_state *old_intel_cstate = to_intel_crtc_state(old_crtc_state); struct intel_crtc_state *pipe_config = to_intel_crtc_state(new_crtc_state); bool modeset = needs_modeset(new_crtc_state); struct intel_plane_state *new_plane_state = @@ -12695,8 +12766,10 @@ static void intel_update_crtc(struct drm_crtc *crtc, intel_begin_crtc_commit(crtc, old_crtc_state); - intel_update_planes_on_crtc(to_intel_atomic_state(state), intel_crtc, - old_intel_cstate, pipe_config); + if (INTEL_GEN(dev_priv) >= 9) + skl_update_planes_on_crtc(to_intel_atomic_state(state), intel_crtc); + else + i9xx_update_planes_on_crtc(to_intel_atomic_state(state), intel_crtc); intel_finish_crtc_commit(crtc, old_crtc_state); } @@ -12889,7 +12962,7 @@ static void intel_atomic_commit_tail(struct drm_atomic_state *state) intel_pre_plane_update(old_intel_crtc_state, new_intel_crtc_state); if (old_crtc_state->active) { - intel_crtc_disable_planes(intel_crtc, old_intel_crtc_state->active_planes); + intel_crtc_disable_planes(intel_state, intel_crtc); /* * We need to disable pipe CRC before disabling the pipe, @@ -13244,7 +13317,7 @@ static int intel_plane_pin_fb(struct intel_plane_state *plane_state) struct i915_vma *vma; if (plane->id == PLANE_CURSOR && - INTEL_INFO(dev_priv)->cursor_needs_physical) { + INTEL_INFO(dev_priv)->display.cursor_needs_physical) { struct drm_i915_gem_object *obj = intel_fb_obj(fb); const int align = intel_cursor_alignment(dev_priv); int err; @@ -13739,7 +13812,7 @@ intel_legacy_cursor_update(struct drm_plane *plane, to_intel_plane_state(plane->state)); } else { trace_intel_disable_plane(plane, to_intel_crtc(crtc)); - intel_plane->disable_plane(intel_plane, to_intel_crtc(crtc)); + intel_plane->disable_plane(intel_plane, crtc_state); } intel_plane_unpin_fb(to_intel_plane_state(old_plane_state)); @@ -14190,7 +14263,7 @@ static void intel_setup_outputs(struct drm_i915_private *dev_priv) intel_pps_init(dev_priv); - if (INTEL_INFO(dev_priv)->num_pipes == 0) + if (!HAS_DISPLAY(dev_priv)) return; /* @@ -14455,7 +14528,6 @@ static int intel_framebuffer_init(struct intel_framebuffer *intel_fb, { struct drm_i915_private *dev_priv = to_i915(obj->base.dev); struct drm_framebuffer *fb = &intel_fb->base; - struct drm_format_name_buf format_name; u32 pitch_limit; unsigned int tiling, stride; int ret = -EINVAL; @@ -14486,39 +14558,14 @@ static int intel_framebuffer_init(struct intel_framebuffer *intel_fb, } } - /* Passed in modifier sanity checking. */ - switch (mode_cmd->modifier[0]) { - case I915_FORMAT_MOD_Y_TILED_CCS: - case I915_FORMAT_MOD_Yf_TILED_CCS: - switch (mode_cmd->pixel_format) { - case DRM_FORMAT_XBGR8888: - case DRM_FORMAT_ABGR8888: - case DRM_FORMAT_XRGB8888: - case DRM_FORMAT_ARGB8888: - break; - default: - DRM_DEBUG_KMS("RC supported only with RGB8888 formats\n"); - goto err; - } - /* fall through */ - case I915_FORMAT_MOD_Yf_TILED: - if (mode_cmd->pixel_format == DRM_FORMAT_C8) { - DRM_DEBUG_KMS("Indexed format does not support Yf tiling\n"); - goto err; - } - /* fall through */ - case I915_FORMAT_MOD_Y_TILED: - if (INTEL_GEN(dev_priv) < 9) { - DRM_DEBUG_KMS("Unsupported tiling 0x%llx!\n", - mode_cmd->modifier[0]); - goto err; - } - break; - case DRM_FORMAT_MOD_LINEAR: - case I915_FORMAT_MOD_X_TILED: - break; - default: - DRM_DEBUG_KMS("Unsupported fb modifier 0x%llx!\n", + if (!drm_any_plane_has_format(&dev_priv->drm, + mode_cmd->pixel_format, + mode_cmd->modifier[0])) { + struct drm_format_name_buf format_name; + + DRM_DEBUG_KMS("unsupported pixel format %s / modifier 0x%llx\n", + drm_get_format_name(mode_cmd->pixel_format, + &format_name), mode_cmd->modifier[0]); goto err; } @@ -14553,69 +14600,6 @@ static int intel_framebuffer_init(struct intel_framebuffer *intel_fb, goto err; } - /* Reject formats not supported by any plane early. */ - switch (mode_cmd->pixel_format) { - case DRM_FORMAT_C8: - case DRM_FORMAT_RGB565: - case DRM_FORMAT_XRGB8888: - case DRM_FORMAT_ARGB8888: - break; - case DRM_FORMAT_XRGB1555: - if (INTEL_GEN(dev_priv) > 3) { - DRM_DEBUG_KMS("unsupported pixel format: %s\n", - drm_get_format_name(mode_cmd->pixel_format, &format_name)); - goto err; - } - break; - case DRM_FORMAT_ABGR8888: - if (!IS_VALLEYVIEW(dev_priv) && !IS_CHERRYVIEW(dev_priv) && - INTEL_GEN(dev_priv) < 9) { - DRM_DEBUG_KMS("unsupported pixel format: %s\n", - drm_get_format_name(mode_cmd->pixel_format, &format_name)); - goto err; - } - break; - case DRM_FORMAT_XBGR8888: - case DRM_FORMAT_XRGB2101010: - case DRM_FORMAT_XBGR2101010: - if (INTEL_GEN(dev_priv) < 4) { - DRM_DEBUG_KMS("unsupported pixel format: %s\n", - drm_get_format_name(mode_cmd->pixel_format, &format_name)); - goto err; - } - break; - case DRM_FORMAT_ABGR2101010: - if (!IS_VALLEYVIEW(dev_priv) && !IS_CHERRYVIEW(dev_priv)) { - DRM_DEBUG_KMS("unsupported pixel format: %s\n", - drm_get_format_name(mode_cmd->pixel_format, &format_name)); - goto err; - } - break; - case DRM_FORMAT_YUYV: - case DRM_FORMAT_UYVY: - case DRM_FORMAT_YVYU: - case DRM_FORMAT_VYUY: - if (INTEL_GEN(dev_priv) < 5 && !IS_G4X(dev_priv)) { - DRM_DEBUG_KMS("unsupported pixel format: %s\n", - drm_get_format_name(mode_cmd->pixel_format, &format_name)); - goto err; - } - break; - case DRM_FORMAT_NV12: - if (INTEL_GEN(dev_priv) < 9 || IS_SKYLAKE(dev_priv) || - IS_BROXTON(dev_priv)) { - DRM_DEBUG_KMS("unsupported pixel format: %s\n", - drm_get_format_name(mode_cmd->pixel_format, - &format_name)); - goto err; - } - break; - default: - DRM_DEBUG_KMS("unsupported pixel format: %s\n", - drm_get_format_name(mode_cmd->pixel_format, &format_name)); - goto err; - } - /* FIXME need to adjust LINOFF/TILEOFF accordingly. */ if (mode_cmd->offsets[0] != 0) goto err; @@ -16070,7 +16054,7 @@ intel_display_capture_error_state(struct drm_i915_private *dev_priv) }; int i; - if (INTEL_INFO(dev_priv)->num_pipes == 0) + if (!HAS_DISPLAY(dev_priv)) return NULL; error = kzalloc(sizeof(*error), GFP_ATOMIC); diff --git a/drivers/gpu/drm/i915/intel_display.h b/drivers/gpu/drm/i915/intel_display.h index 5f2955b944da..4262452963b3 100644 --- a/drivers/gpu/drm/i915/intel_display.h +++ b/drivers/gpu/drm/i915/intel_display.h @@ -242,6 +242,7 @@ enum intel_display_power_domain { POWER_DOMAIN_TRANSCODER_B, POWER_DOMAIN_TRANSCODER_C, POWER_DOMAIN_TRANSCODER_EDP, + POWER_DOMAIN_TRANSCODER_EDP_VDSC, POWER_DOMAIN_TRANSCODER_DSI_A, POWER_DOMAIN_TRANSCODER_DSI_C, POWER_DOMAIN_PORT_DDI_A_LANES, @@ -398,6 +399,14 @@ struct intel_link_m_n { for_each_power_well_reverse(__dev_priv, __power_well) \ for_each_if((__power_well)->desc->domains & (__domain_mask)) +#define for_each_old_intel_plane_in_state(__state, plane, old_plane_state, __i) \ + for ((__i) = 0; \ + (__i) < (__state)->base.dev->mode_config.num_total_plane && \ + ((plane) = to_intel_plane((__state)->base.planes[__i].ptr), \ + (old_plane_state) = to_intel_plane_state((__state)->base.planes[__i].old_state), 1); \ + (__i)++) \ + for_each_if(plane) + #define for_each_new_intel_plane_in_state(__state, plane, new_plane_state, __i) \ for ((__i) = 0; \ (__i) < (__state)->base.dev->mode_config.num_total_plane && \ @@ -423,10 +432,18 @@ struct intel_link_m_n { (__i)++) \ for_each_if(plane) -void intel_link_compute_m_n(int bpp, int nlanes, +#define for_each_oldnew_intel_crtc_in_state(__state, crtc, old_crtc_state, new_crtc_state, __i) \ + for ((__i) = 0; \ + (__i) < (__state)->base.dev->mode_config.num_crtc && \ + ((crtc) = to_intel_crtc((__state)->base.crtcs[__i].ptr), \ + (old_crtc_state) = to_intel_crtc_state((__state)->base.crtcs[__i].old_state), \ + (new_crtc_state) = to_intel_crtc_state((__state)->base.crtcs[__i].new_state), 1); \ + (__i)++) \ + for_each_if(crtc) + +void intel_link_compute_m_n(u16 bpp, int nlanes, int pixel_clock, int link_clock, struct intel_link_m_n *m_n, bool constant_n); - bool is_ccs_modifier(u64 modifier); #endif diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 7699f9b7b2d2..fdd2cbc56fa3 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -47,6 +47,8 @@ /* DP DSC small joiner has 2 FIFOs each of 640 x 6 bytes */ #define DP_DSC_MAX_SMALL_JOINER_RAM_BUFFER 61440 +#define DP_DSC_MIN_SUPPORTED_BPC 8 +#define DP_DSC_MAX_SUPPORTED_BPC 10 /* DP DSC throughput values used for slice count calculations KPixels/s */ #define DP_DSC_PEAK_PIXEL_RATE 2720000 @@ -543,7 +545,7 @@ intel_dp_mode_valid(struct drm_connector *connector, dsc_slice_count = drm_dp_dsc_sink_max_slice_count(intel_dp->dsc_dpcd, true); - } else { + } else if (drm_dp_sink_supports_fec(intel_dp->fec_capable)) { dsc_max_output_bpp = intel_dp_dsc_get_output_bpp(max_link_clock, max_lanes, @@ -1708,6 +1710,41 @@ struct link_config_limits { int min_bpp, max_bpp; }; +static bool intel_dp_source_supports_fec(struct intel_dp *intel_dp, + const struct intel_crtc_state *pipe_config) +{ + struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + + return INTEL_GEN(dev_priv) >= 11 && + pipe_config->cpu_transcoder != TRANSCODER_A; +} + +static bool intel_dp_supports_fec(struct intel_dp *intel_dp, + const struct intel_crtc_state *pipe_config) +{ + return intel_dp_source_supports_fec(intel_dp, pipe_config) && + drm_dp_sink_supports_fec(intel_dp->fec_capable); +} + +static bool intel_dp_source_supports_dsc(struct intel_dp *intel_dp, + const struct intel_crtc_state *pipe_config) +{ + struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + + return INTEL_GEN(dev_priv) >= 10 && + pipe_config->cpu_transcoder != TRANSCODER_A; +} + +static bool intel_dp_supports_dsc(struct intel_dp *intel_dp, + const struct intel_crtc_state *pipe_config) +{ + if (!intel_dp_is_edp(intel_dp) && !pipe_config->fec_enable) + return false; + + return intel_dp_source_supports_dsc(intel_dp, pipe_config) && + drm_dp_sink_supports_dsc(intel_dp->dsc_dpcd); +} + static int intel_dp_compute_bpp(struct intel_dp *intel_dp, struct intel_crtc_state *pipe_config) { @@ -1842,14 +1879,122 @@ intel_dp_compute_link_config_fast(struct intel_dp *intel_dp, return false; } +static int intel_dp_dsc_compute_bpp(struct intel_dp *intel_dp, u8 dsc_max_bpc) +{ + int i, num_bpc; + u8 dsc_bpc[3] = {0}; + + num_bpc = drm_dp_dsc_sink_supported_input_bpcs(intel_dp->dsc_dpcd, + dsc_bpc); + for (i = 0; i < num_bpc; i++) { + if (dsc_max_bpc >= dsc_bpc[i]) + return dsc_bpc[i] * 3; + } + + return 0; +} + +static bool intel_dp_dsc_compute_config(struct intel_dp *intel_dp, + struct intel_crtc_state *pipe_config, + struct drm_connector_state *conn_state, + struct link_config_limits *limits) +{ + struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); + struct drm_i915_private *dev_priv = to_i915(dig_port->base.base.dev); + struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode; + u8 dsc_max_bpc; + int pipe_bpp; + + if (!intel_dp_supports_dsc(intel_dp, pipe_config)) + return false; + + dsc_max_bpc = min_t(u8, DP_DSC_MAX_SUPPORTED_BPC, + conn_state->max_requested_bpc); + + pipe_bpp = intel_dp_dsc_compute_bpp(intel_dp, dsc_max_bpc); + if (pipe_bpp < DP_DSC_MIN_SUPPORTED_BPC * 3) { + DRM_DEBUG_KMS("No DSC support for less than 8bpc\n"); + return false; + } + + /* + * For now enable DSC for max bpp, max link rate, max lane count. + * Optimize this later for the minimum possible link rate/lane count + * with DSC enabled for the requested mode. + */ + pipe_config->pipe_bpp = pipe_bpp; + pipe_config->port_clock = intel_dp->common_rates[limits->max_clock]; + pipe_config->lane_count = limits->max_lane_count; + + if (intel_dp_is_edp(intel_dp)) { + pipe_config->dsc_params.compressed_bpp = + min_t(u16, drm_edp_dsc_sink_output_bpp(intel_dp->dsc_dpcd) >> 4, + pipe_config->pipe_bpp); + pipe_config->dsc_params.slice_count = + drm_dp_dsc_sink_max_slice_count(intel_dp->dsc_dpcd, + true); + } else { + u16 dsc_max_output_bpp; + u8 dsc_dp_slice_count; + + dsc_max_output_bpp = + intel_dp_dsc_get_output_bpp(pipe_config->port_clock, + pipe_config->lane_count, + adjusted_mode->crtc_clock, + adjusted_mode->crtc_hdisplay); + dsc_dp_slice_count = + intel_dp_dsc_get_slice_count(intel_dp, + adjusted_mode->crtc_clock, + adjusted_mode->crtc_hdisplay); + if (!dsc_max_output_bpp || !dsc_dp_slice_count) { + DRM_DEBUG_KMS("Compressed BPP/Slice Count not supported\n"); + return false; + } + pipe_config->dsc_params.compressed_bpp = min_t(u16, + dsc_max_output_bpp >> 4, + pipe_config->pipe_bpp); + pipe_config->dsc_params.slice_count = dsc_dp_slice_count; + } + /* + * VDSC engine operates at 1 Pixel per clock, so if peak pixel rate + * is greater than the maximum Cdclock and if slice count is even + * then we need to use 2 VDSC instances. + */ + if (adjusted_mode->crtc_clock > dev_priv->max_cdclk_freq) { + if (pipe_config->dsc_params.slice_count > 1) { + pipe_config->dsc_params.dsc_split = true; + } else { + DRM_DEBUG_KMS("Cannot split stream to use 2 VDSC instances\n"); + return false; + } + } + if (intel_dp_compute_dsc_params(intel_dp, pipe_config) < 0) { + DRM_DEBUG_KMS("Cannot compute valid DSC parameters for Input Bpp = %d " + "Compressed BPP = %d\n", + pipe_config->pipe_bpp, + pipe_config->dsc_params.compressed_bpp); + return false; + } + pipe_config->dsc_params.compression_enable = true; + DRM_DEBUG_KMS("DP DSC computed with Input Bpp = %d " + "Compressed Bpp = %d Slice Count = %d\n", + pipe_config->pipe_bpp, + pipe_config->dsc_params.compressed_bpp, + pipe_config->dsc_params.slice_count); + + return true; +} + static bool intel_dp_compute_link_config(struct intel_encoder *encoder, - struct intel_crtc_state *pipe_config) + struct intel_crtc_state *pipe_config, + struct drm_connector_state *conn_state) { struct drm_display_mode *adjusted_mode = &pipe_config->base.adjusted_mode; struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base); struct link_config_limits limits; int common_len; + bool ret; common_len = intel_dp_common_len_rate_limit(intel_dp, intel_dp->max_link_rate); @@ -1888,7 +2033,7 @@ intel_dp_compute_link_config(struct intel_encoder *encoder, intel_dp->common_rates[limits.max_clock], limits.max_bpp, adjusted_mode->crtc_clock); - if (intel_dp_is_edp(intel_dp)) { + if (intel_dp_is_edp(intel_dp)) /* * Optimize for fast and narrow. eDP 1.3 section 3.3 and eDP 1.4 * section A.1: "It is recommended that the minimum number of @@ -1898,26 +2043,42 @@ intel_dp_compute_link_config(struct intel_encoder *encoder, * Note that we use the max clock and lane count for eDP 1.3 and * earlier, and fast vs. wide is irrelevant. */ - if (!intel_dp_compute_link_config_fast(intel_dp, pipe_config, - &limits)) - return false; - } else { + ret = intel_dp_compute_link_config_fast(intel_dp, pipe_config, + &limits); + else /* Optimize for slow and wide. */ - if (!intel_dp_compute_link_config_wide(intel_dp, pipe_config, - &limits)) + ret = intel_dp_compute_link_config_wide(intel_dp, pipe_config, + &limits); + + /* enable compression if the mode doesn't fit available BW */ + if (!ret) { + if (!intel_dp_dsc_compute_config(intel_dp, pipe_config, + conn_state, &limits)) return false; } - DRM_DEBUG_KMS("DP lane count %d clock %d bpp %d\n", - pipe_config->lane_count, pipe_config->port_clock, - pipe_config->pipe_bpp); + if (pipe_config->dsc_params.compression_enable) { + DRM_DEBUG_KMS("DP lane count %d clock %d Input bpp %d Compressed bpp %d\n", + pipe_config->lane_count, pipe_config->port_clock, + pipe_config->pipe_bpp, + pipe_config->dsc_params.compressed_bpp); - DRM_DEBUG_KMS("DP link rate required %i available %i\n", - intel_dp_link_required(adjusted_mode->crtc_clock, - pipe_config->pipe_bpp), - intel_dp_max_data_rate(pipe_config->port_clock, - pipe_config->lane_count)); + DRM_DEBUG_KMS("DP link rate required %i available %i\n", + intel_dp_link_required(adjusted_mode->crtc_clock, + pipe_config->dsc_params.compressed_bpp), + intel_dp_max_data_rate(pipe_config->port_clock, + pipe_config->lane_count)); + } else { + DRM_DEBUG_KMS("DP lane count %d clock %d bpp %d\n", + pipe_config->lane_count, pipe_config->port_clock, + pipe_config->pipe_bpp); + DRM_DEBUG_KMS("DP link rate required %i available %i\n", + intel_dp_link_required(adjusted_mode->crtc_clock, + pipe_config->pipe_bpp), + intel_dp_max_data_rate(pipe_config->port_clock, + pipe_config->lane_count)); + } return true; } @@ -1983,7 +2144,10 @@ intel_dp_compute_config(struct intel_encoder *encoder, if (adjusted_mode->flags & DRM_MODE_FLAG_DBLCLK) return false; - if (!intel_dp_compute_link_config(encoder, pipe_config)) + pipe_config->fec_enable = !intel_dp_is_edp(intel_dp) && + intel_dp_supports_fec(intel_dp, pipe_config); + + if (!intel_dp_compute_link_config(encoder, pipe_config, conn_state)) return false; if (intel_conn_state->broadcast_rgb == INTEL_BROADCAST_RGB_AUTO) { @@ -2001,11 +2165,20 @@ intel_dp_compute_config(struct intel_encoder *encoder, intel_conn_state->broadcast_rgb == INTEL_BROADCAST_RGB_LIMITED; } - intel_link_compute_m_n(pipe_config->pipe_bpp, pipe_config->lane_count, - adjusted_mode->crtc_clock, - pipe_config->port_clock, - &pipe_config->dp_m_n, - constant_n); + if (!pipe_config->dsc_params.compression_enable) + intel_link_compute_m_n(pipe_config->pipe_bpp, + pipe_config->lane_count, + adjusted_mode->crtc_clock, + pipe_config->port_clock, + &pipe_config->dp_m_n, + constant_n); + else + intel_link_compute_m_n(pipe_config->dsc_params.compressed_bpp, + pipe_config->lane_count, + adjusted_mode->crtc_clock, + pipe_config->port_clock, + &pipe_config->dp_m_n, + constant_n); if (intel_connector->panel.downclock_mode != NULL && dev_priv->drrs.type == SEAMLESS_DRRS_SUPPORT) { @@ -2702,6 +2875,22 @@ static bool downstream_hpd_needs_d0(struct intel_dp *intel_dp) intel_dp->downstream_ports[0] & DP_DS_PORT_HPD; } +void intel_dp_sink_set_decompression_state(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state, + bool enable) +{ + int ret; + + if (!crtc_state->dsc_params.compression_enable) + return; + + ret = drm_dp_dpcd_writeb(&intel_dp->aux, DP_DSC_ENABLE, + enable ? DP_DECOMPRESSION_EN : 0); + if (ret < 0) + DRM_DEBUG_KMS("Failed to %s sink decompression state\n", + enable ? "enable" : "disable"); +} + /* If the sink supports it, try to set the power state appropriately */ void intel_dp_sink_dpms(struct intel_dp *intel_dp, int mode) { @@ -3837,15 +4026,14 @@ static void intel_dp_get_dsc_sink_cap(struct intel_dp *intel_dp) DRM_DEBUG_KMS("DSC DPCD: %*ph\n", (int)sizeof(intel_dp->dsc_dpcd), intel_dp->dsc_dpcd); + /* FEC is supported only on DP 1.4 */ - if (!intel_dp_is_edp(intel_dp)) { - if (drm_dp_dpcd_readb(&intel_dp->aux, DP_FEC_CAPABILITY, - &intel_dp->fec_capable) < 0) - DRM_ERROR("Failed to read FEC DPCD register\n"); + if (!intel_dp_is_edp(intel_dp) && + drm_dp_dpcd_readb(&intel_dp->aux, DP_FEC_CAPABILITY, + &intel_dp->fec_capable) < 0) + DRM_ERROR("Failed to read FEC DPCD register\n"); - DRM_DEBUG_KMS("FEC CAPABILITY: %x\n", - intel_dp->fec_capable); - } + DRM_DEBUG_KMS("FEC CAPABILITY: %x\n", intel_dp->fec_capable); } } @@ -3936,8 +4124,6 @@ intel_edp_init_dpcd(struct intel_dp *intel_dp) static bool intel_dp_get_dpcd(struct intel_dp *intel_dp) { - u8 sink_count; - if (!intel_dp_read_dpcd(intel_dp)) return false; @@ -3947,25 +4133,35 @@ intel_dp_get_dpcd(struct intel_dp *intel_dp) intel_dp_set_common_rates(intel_dp); } - if (drm_dp_dpcd_readb(&intel_dp->aux, DP_SINK_COUNT, &sink_count) <= 0) - return false; - /* - * Sink count can change between short pulse hpd hence - * a member variable in intel_dp will track any changes - * between short pulse interrupts. + * Some eDP panels do not set a valid value for sink count, that is why + * it don't care about read it here and in intel_edp_init_dpcd(). */ - intel_dp->sink_count = DP_GET_SINK_COUNT(sink_count); + if (!intel_dp_is_edp(intel_dp)) { + u8 count; + ssize_t r; - /* - * SINK_COUNT == 0 and DOWNSTREAM_PORT_PRESENT == 1 implies that - * a dongle is present but no display. Unless we require to know - * if a dongle is present or not, we don't need to update - * downstream port information. So, an early return here saves - * time from performing other operations which are not required. - */ - if (!intel_dp_is_edp(intel_dp) && !intel_dp->sink_count) - return false; + r = drm_dp_dpcd_readb(&intel_dp->aux, DP_SINK_COUNT, &count); + if (r < 1) + return false; + + /* + * Sink count can change between short pulse hpd hence + * a member variable in intel_dp will track any changes + * between short pulse interrupts. + */ + intel_dp->sink_count = DP_GET_SINK_COUNT(count); + + /* + * SINK_COUNT == 0 and DOWNSTREAM_PORT_PRESENT == 1 implies that + * a dongle is present but no display. Unless we require to know + * if a dongle is present or not, we don't need to update + * downstream port information. So, an early return here saves + * time from performing other operations which are not required. + */ + if (!intel_dp->sink_count) + return false; + } if (!drm_dp_is_branch(intel_dp->dpcd)) return true; /* native DP sink */ @@ -4375,6 +4571,17 @@ intel_dp_needs_link_retrain(struct intel_dp *intel_dp) if (!intel_dp->link_trained) return false; + /* + * While PSR source HW is enabled, it will control main-link sending + * frames, enabling and disabling it so trying to do a retrain will fail + * as the link would or not be on or it could mix training patterns + * and frame data at the same time causing retrain to fail. + * Also when exiting PSR, HW will retrain the link anyways fixing + * any link status error. + */ + if (intel_psr_enabled(intel_dp)) + return false; + if (!intel_dp_get_link_status(intel_dp, link_status)) return false; diff --git a/drivers/gpu/drm/i915/intel_dpll_mgr.c b/drivers/gpu/drm/i915/intel_dpll_mgr.c index 901e15063b24..d513ca875c67 100644 --- a/drivers/gpu/drm/i915/intel_dpll_mgr.c +++ b/drivers/gpu/drm/i915/intel_dpll_mgr.c @@ -2523,7 +2523,8 @@ static bool icl_calc_dpll_state(struct intel_crtc_state *crtc_state, if (intel_port_is_tc(dev_priv, encoder->port)) ret = icl_calc_tbt_pll(dev_priv, clock, &pll_params); - else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI)) + else if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_HDMI) || + intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DSI)) ret = cnl_ddi_calculate_wrpll(clock, dev_priv, &pll_params); else ret = icl_calc_dp_combo_pll(dev_priv, clock, &pll_params); diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 153ee47d9c03..a3252064b8d2 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -706,6 +706,8 @@ struct intel_crtc_wm_state { /* gen9+ only needs 1-step wm programming */ struct skl_pipe_wm optimal; struct skl_ddb_entry ddb; + struct skl_ddb_entry plane_ddb_y[I915_MAX_PLANES]; + struct skl_ddb_entry plane_ddb_uv[I915_MAX_PLANES]; } skl; struct { @@ -926,6 +928,9 @@ struct intel_crtc_state { u8 active_planes; u8 nv12_planes; + /* bitmask of planes that will be updated during the commit */ + u8 update_planes; + /* HDMI scrambling status */ bool hdmi_scrambling; @@ -937,6 +942,18 @@ struct intel_crtc_state { /* Output down scaling is done in LSPCON device */ bool lspcon_downsampling; + + /* Display Stream compression state */ + struct { + bool compression_enable; + bool dsc_split; + u16 compressed_bpp; + u8 slice_count; + } dsc_params; + struct drm_dsc_config dp_dsc_cfg; + + /* Forward Error correction State */ + bool fec_enable; }; struct intel_crtc { @@ -1013,7 +1030,7 @@ struct intel_plane { const struct intel_crtc_state *crtc_state, const struct intel_plane_state *plane_state); void (*disable_plane)(struct intel_plane *plane, - struct intel_crtc *crtc); + const struct intel_crtc_state *crtc_state); bool (*get_hw_state)(struct intel_plane *plane, enum pipe *pipe); int (*check_plane)(struct intel_crtc_state *crtc_state, struct intel_plane_state *plane_state); @@ -1516,13 +1533,9 @@ u8 intel_ddi_dp_pre_emphasis_max(struct intel_encoder *encoder, u8 voltage_swing); int intel_ddi_toggle_hdcp_signalling(struct intel_encoder *intel_encoder, bool enable); -void icl_map_plls_to_ports(struct drm_crtc *crtc, - struct intel_crtc_state *crtc_state, - struct drm_atomic_state *old_state); -void icl_unmap_plls_to_ports(struct drm_crtc *crtc, - struct intel_crtc_state *crtc_state, - struct drm_atomic_state *old_state); void icl_sanitize_encoder_pll_mapping(struct intel_encoder *encoder); +int cnl_calc_wrpll_link(struct drm_i915_private *dev_priv, + enum intel_dpll_id pll_id); unsigned int intel_fb_align_height(const struct drm_framebuffer *fb, int color_plane, unsigned int height); @@ -1787,6 +1800,9 @@ void intel_dp_stop_link_train(struct intel_dp *intel_dp); int intel_dp_retrain_link(struct intel_encoder *encoder, struct drm_modeset_acquire_ctx *ctx); void intel_dp_sink_dpms(struct intel_dp *intel_dp, int mode); +void intel_dp_sink_set_decompression_state(struct intel_dp *intel_dp, + const struct intel_crtc_state *crtc_state, + bool enable); void intel_dp_encoder_reset(struct drm_encoder *encoder); void intel_dp_encoder_suspend(struct intel_encoder *intel_encoder); void intel_dp_encoder_destroy(struct drm_encoder *encoder); @@ -1842,6 +1858,12 @@ uint16_t intel_dp_dsc_get_output_bpp(int link_clock, uint8_t lane_count, uint8_t intel_dp_dsc_get_slice_count(struct intel_dp *intel_dp, int mode_clock, int mode_hdisplay); +/* intel_vdsc.c */ +int intel_dp_compute_dsc_params(struct intel_dp *intel_dp, + struct intel_crtc_state *pipe_config); +enum intel_display_power_domain +intel_dsc_power_domain(const struct intel_crtc_state *crtc_state); + static inline unsigned int intel_dp_unused_lane_mask(int lane_count) { return ~((1 << lane_count) - 1) & 0xf; @@ -2046,6 +2068,7 @@ void intel_psr_irq_handler(struct drm_i915_private *dev_priv, u32 psr_iir); void intel_psr_short_pulse(struct intel_dp *intel_dp); int intel_psr_wait_for_idle(const struct intel_crtc_state *new_crtc_state, u32 *out_value); +bool intel_psr_enabled(struct intel_dp *intel_dp); /* intel_quirks.c */ void intel_init_quirks(struct drm_i915_private *dev_priv); @@ -2180,6 +2203,9 @@ void g4x_wm_get_hw_state(struct drm_device *dev); void vlv_wm_get_hw_state(struct drm_device *dev); void ilk_wm_get_hw_state(struct drm_device *dev); void skl_wm_get_hw_state(struct drm_device *dev); +void skl_pipe_ddb_get_hw_state(struct intel_crtc *crtc, + struct skl_ddb_entry *ddb_y, + struct skl_ddb_entry *ddb_uv); void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv, struct skl_ddb_allocation *ddb /* out */); void skl_pipe_wm_get_hw_state(struct drm_crtc *crtc, @@ -2194,6 +2220,10 @@ bool skl_wm_level_equals(const struct skl_wm_level *l1, bool skl_ddb_allocation_overlaps(const struct skl_ddb_entry *ddb, const struct skl_ddb_entry entries[], int num_entries, int ignore_idx); +void skl_write_plane_wm(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state); +void skl_write_cursor_wm(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state); bool ilk_disable_lp_wm(struct drm_device *dev); int skl_check_pipe_max_pixel_rate(struct intel_crtc *intel_crtc, struct intel_crtc_state *cstate); @@ -2286,10 +2316,10 @@ struct drm_plane_state *intel_plane_duplicate_state(struct drm_plane *plane); void intel_plane_destroy_state(struct drm_plane *plane, struct drm_plane_state *state); extern const struct drm_plane_helper_funcs intel_plane_helper_funcs; -void intel_update_planes_on_crtc(struct intel_atomic_state *old_state, - struct intel_crtc *crtc, - struct intel_crtc_state *old_crtc_state, - struct intel_crtc_state *new_crtc_state); +void skl_update_planes_on_crtc(struct intel_atomic_state *state, + struct intel_crtc *crtc); +void i9xx_update_planes_on_crtc(struct intel_atomic_state *state, + struct intel_crtc *crtc); int intel_plane_atomic_check_with_state(const struct intel_crtc_state *old_crtc_state, struct intel_crtc_state *crtc_state, const struct intel_plane_state *old_plane_state, diff --git a/drivers/gpu/drm/i915/intel_dsi.h b/drivers/gpu/drm/i915/intel_dsi.h index ee93137f4433..d968f1f13e09 100644 --- a/drivers/gpu/drm/i915/intel_dsi.h +++ b/drivers/gpu/drm/i915/intel_dsi.h @@ -146,6 +146,11 @@ static inline bool is_cmd_mode(struct intel_dsi *intel_dsi) return intel_dsi->operation_mode == INTEL_DSI_COMMAND_MODE; } +static inline u16 intel_dsi_encoder_ports(struct intel_encoder *encoder) +{ + return enc_to_intel_dsi(&encoder->base)->ports; +} + /* intel_dsi.c */ int intel_dsi_bitrate(const struct intel_dsi *intel_dsi); int intel_dsi_tlpx_ns(const struct intel_dsi *intel_dsi); diff --git a/drivers/gpu/drm/i915/intel_dsi_vbt.c b/drivers/gpu/drm/i915/intel_dsi_vbt.c index a72de81f4832..a1a8b3790e61 100644 --- a/drivers/gpu/drm/i915/intel_dsi_vbt.c +++ b/drivers/gpu/drm/i915/intel_dsi_vbt.c @@ -103,6 +103,18 @@ static struct gpio_map vlv_gpio_table[] = { #define CHV_GPIO_PAD_CFG1(f, i) (0x4400 + (f) * 0x400 + (i) * 8 + 4) #define CHV_GPIO_CFGLOCK (1 << 31) +/* ICL DSI Display GPIO Pins */ +#define ICL_GPIO_DDSP_HPD_A 0 +#define ICL_GPIO_L_VDDEN_1 1 +#define ICL_GPIO_L_BKLTEN_1 2 +#define ICL_GPIO_DDPA_CTRLCLK_1 3 +#define ICL_GPIO_DDPA_CTRLDATA_1 4 +#define ICL_GPIO_DDSP_HPD_B 5 +#define ICL_GPIO_L_VDDEN_2 6 +#define ICL_GPIO_L_BKLTEN_2 7 +#define ICL_GPIO_DDPA_CTRLCLK_2 8 +#define ICL_GPIO_DDPA_CTRLDATA_2 9 + static inline enum port intel_dsi_seq_port_to_port(u8 port) { return port ? PORT_C : PORT_A; @@ -324,6 +336,12 @@ static void bxt_exec_gpio(struct drm_i915_private *dev_priv, gpiod_set_value(gpio_desc, value); } +static void icl_exec_gpio(struct drm_i915_private *dev_priv, + u8 gpio_source, u8 gpio_index, bool value) +{ + DRM_DEBUG_KMS("Skipping ICL GPIO element execution\n"); +} + static const u8 *mipi_exec_gpio(struct intel_dsi *intel_dsi, const u8 *data) { struct drm_device *dev = intel_dsi->base.base.dev; @@ -347,7 +365,9 @@ static const u8 *mipi_exec_gpio(struct intel_dsi *intel_dsi, const u8 *data) /* pull up/down */ value = *data++ & 1; - if (IS_VALLEYVIEW(dev_priv)) + if (IS_ICELAKE(dev_priv)) + icl_exec_gpio(dev_priv, gpio_source, gpio_index, value); + else if (IS_VALLEYVIEW(dev_priv)) vlv_exec_gpio(dev_priv, gpio_source, gpio_number, value); else if (IS_CHERRYVIEW(dev_priv)) chv_exec_gpio(dev_priv, gpio_source, gpio_number, value); diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c index dfafa79171df..ff5b7bc692ce 100644 --- a/drivers/gpu/drm/i915/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/intel_engine_cs.c @@ -493,46 +493,6 @@ void intel_engine_setup_common(struct intel_engine_cs *engine) intel_engine_init_cmd_parser(engine); } -int intel_engine_create_scratch(struct intel_engine_cs *engine, - unsigned int size) -{ - struct drm_i915_gem_object *obj; - struct i915_vma *vma; - int ret; - - WARN_ON(engine->scratch); - - obj = i915_gem_object_create_stolen(engine->i915, size); - if (!obj) - obj = i915_gem_object_create_internal(engine->i915, size); - if (IS_ERR(obj)) { - DRM_ERROR("Failed to allocate scratch page\n"); - return PTR_ERR(obj); - } - - vma = i915_vma_instance(obj, &engine->i915->ggtt.vm, NULL); - if (IS_ERR(vma)) { - ret = PTR_ERR(vma); - goto err_unref; - } - - ret = i915_vma_pin(vma, 0, 0, PIN_GLOBAL | PIN_HIGH); - if (ret) - goto err_unref; - - engine->scratch = vma; - return 0; - -err_unref: - i915_gem_object_put(obj); - return ret; -} - -void intel_engine_cleanup_scratch(struct intel_engine_cs *engine) -{ - i915_vma_unpin_and_release(&engine->scratch, 0); -} - static void cleanup_status_page(struct intel_engine_cs *engine) { if (HWS_NEEDS_PHYSICAL(engine->i915)) { @@ -707,8 +667,6 @@ void intel_engine_cleanup_common(struct intel_engine_cs *engine) { struct drm_i915_private *i915 = engine->i915; - intel_engine_cleanup_scratch(engine); - cleanup_status_page(engine); intel_engine_fini_breadcrumbs(engine); @@ -723,6 +681,10 @@ void intel_engine_cleanup_common(struct intel_engine_cs *engine) __intel_context_unpin(i915->kernel_context, engine); i915_timeline_fini(&engine->timeline); + + intel_wa_list_free(&engine->ctx_wa_list); + intel_wa_list_free(&engine->wa_list); + intel_wa_list_free(&engine->whitelist); } u64 intel_engine_get_active_head(const struct intel_engine_cs *engine) diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c index 14cbaf4a0e93..f23570c44323 100644 --- a/drivers/gpu/drm/i915/intel_fbc.c +++ b/drivers/gpu/drm/i915/intel_fbc.c @@ -1309,7 +1309,7 @@ void intel_fbc_init(struct drm_i915_private *dev_priv) fbc->active = false; if (need_fbc_vtd_wa(dev_priv)) - mkwrite_device_info(dev_priv)->has_fbc = false; + mkwrite_device_info(dev_priv)->display.has_fbc = false; i915_modparams.enable_fbc = intel_sanitize_fbc_option(dev_priv); DRM_DEBUG_KMS("Sanitized enable_fbc value: %d\n", diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c index 2480c7d6edee..fb5bb5b32a60 100644 --- a/drivers/gpu/drm/i915/intel_fbdev.c +++ b/drivers/gpu/drm/i915/intel_fbdev.c @@ -672,7 +672,7 @@ int intel_fbdev_init(struct drm_device *dev) struct intel_fbdev *ifbdev; int ret; - if (WARN_ON(INTEL_INFO(dev_priv)->num_pipes == 0)) + if (WARN_ON(!HAS_DISPLAY(dev_priv))) return -ENODEV; ifbdev = kzalloc(sizeof(struct intel_fbdev), GFP_KERNEL); diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c index 4d264e577e3b..55aeb97dd66d 100644 --- a/drivers/gpu/drm/i915/intel_hdmi.c +++ b/drivers/gpu/drm/i915/intel_hdmi.c @@ -115,6 +115,8 @@ static u32 hsw_infoframe_enable(unsigned int type) switch (type) { case DP_SDP_VSC: return VIDEO_DIP_ENABLE_VSC_HSW; + case DP_SDP_PPS: + return VDIP_ENABLE_PPS; case HDMI_INFOFRAME_TYPE_AVI: return VIDEO_DIP_ENABLE_AVI_HSW; case HDMI_INFOFRAME_TYPE_SPD: @@ -136,6 +138,8 @@ hsw_dip_data_reg(struct drm_i915_private *dev_priv, switch (type) { case DP_SDP_VSC: return HSW_TVIDEO_DIP_VSC_DATA(cpu_transcoder, i); + case DP_SDP_PPS: + return ICL_VIDEO_DIP_PPS_DATA(cpu_transcoder, i); case HDMI_INFOFRAME_TYPE_AVI: return HSW_TVIDEO_DIP_AVI_DATA(cpu_transcoder, i); case HDMI_INFOFRAME_TYPE_SPD: @@ -148,6 +152,18 @@ hsw_dip_data_reg(struct drm_i915_private *dev_priv, } } +static int hsw_dip_data_size(unsigned int type) +{ + switch (type) { + case DP_SDP_VSC: + return VIDEO_DIP_VSC_DATA_SIZE; + case DP_SDP_PPS: + return VIDEO_DIP_PPS_DATA_SIZE; + default: + return VIDEO_DIP_DATA_SIZE; + } +} + static void g4x_write_infoframe(struct intel_encoder *encoder, const struct intel_crtc_state *crtc_state, unsigned int type, @@ -382,11 +398,12 @@ static void hsw_write_infoframe(struct intel_encoder *encoder, struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); enum transcoder cpu_transcoder = crtc_state->cpu_transcoder; i915_reg_t ctl_reg = HSW_TVIDEO_DIP_CTL(cpu_transcoder); - int data_size = type == DP_SDP_VSC ? - VIDEO_DIP_VSC_DATA_SIZE : VIDEO_DIP_DATA_SIZE; + int data_size; int i; u32 val = I915_READ(ctl_reg); + data_size = hsw_dip_data_size(type); + val &= ~hsw_infoframe_enable(type); I915_WRITE(ctl_reg, val); diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c index 33d87ab93fdd..802d0394ccc4 100644 --- a/drivers/gpu/drm/i915/intel_i2c.c +++ b/drivers/gpu/drm/i915/intel_i2c.c @@ -817,7 +817,7 @@ int intel_setup_gmbus(struct drm_i915_private *dev_priv) unsigned int pin; int ret; - if (INTEL_INFO(dev_priv)->num_pipes == 0) + if (!HAS_DISPLAY(dev_priv)) return 0; if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index 9399db3260ad..d84c7815ee0c 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -398,8 +398,13 @@ static u64 execlists_update_context(struct i915_request *rq) * may not be visible to the HW prior to the completion of the UC * register write and that we may begin execution from the context * before its image is complete leading to invalid PD chasing. + * + * Furthermore, Braswell, at least, wants a full mb to be sure that + * the writes are coherent in memory (visible to the GPU) prior to + * execution, and not just visible to other CPUs (as is the result of + * wmb). */ - wmb(); + mb(); return ce->lrc_desc; } @@ -767,6 +772,8 @@ execlists_cancel_port_requests(struct intel_engine_execlists * const execlists) static void reset_csb_pointers(struct intel_engine_execlists *execlists) { + const unsigned int reset_value = GEN8_CSB_ENTRIES - 1; + /* * After a reset, the HW starts writing into CSB entry [0]. We * therefore have to set our HEAD pointer back one entry so that @@ -776,8 +783,8 @@ static void reset_csb_pointers(struct intel_engine_execlists *execlists) * inline comparison of our cached head position against the last HW * write works even before the first interrupt. */ - execlists->csb_head = execlists->csb_write_reset; - WRITE_ONCE(*execlists->csb_write, execlists->csb_write_reset); + execlists->csb_head = reset_value; + WRITE_ONCE(*execlists->csb_write, reset_value); } static void nop_submission_tasklet(unsigned long data) @@ -818,8 +825,11 @@ static void execlists_cancel_requests(struct intel_engine_cs *engine) /* Mark all executing requests as skipped. */ list_for_each_entry(rq, &engine->timeline.requests, link) { GEM_BUG_ON(!rq->global_seqno); - if (!i915_request_completed(rq)) - dma_fence_set_error(&rq->fence, -EIO); + + if (test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &rq->fence.flags)) + continue; + + dma_fence_set_error(&rq->fence, -EIO); } /* Flush the queued requests to the timeline list (for retiring). */ @@ -839,6 +849,10 @@ static void execlists_cancel_requests(struct intel_engine_cs *engine) kmem_cache_free(engine->i915->priorities, p); } + intel_write_status_page(engine, + I915_GEM_HWS_INDEX, + intel_engine_last_submit(engine)); + /* Remaining _unready_ requests will be nop'ed when submitted */ execlists->queue_priority = INT_MIN; @@ -1279,9 +1293,10 @@ static int execlists_request_alloc(struct i915_request *request) static u32 * gen8_emit_flush_coherentl3_wa(struct intel_engine_cs *engine, u32 *batch) { + /* NB no one else is allowed to scribble over scratch + 256! */ *batch++ = MI_STORE_REGISTER_MEM_GEN8 | MI_SRM_LRM_GLOBAL_GTT; *batch++ = i915_mmio_reg_offset(GEN8_L3SQCREG4); - *batch++ = i915_ggtt_offset(engine->scratch) + 256; + *batch++ = i915_scratch_offset(engine->i915) + 256; *batch++ = 0; *batch++ = MI_LOAD_REGISTER_IMM(1); @@ -1295,7 +1310,7 @@ gen8_emit_flush_coherentl3_wa(struct intel_engine_cs *engine, u32 *batch) *batch++ = MI_LOAD_REGISTER_MEM_GEN8 | MI_SRM_LRM_GLOBAL_GTT; *batch++ = i915_mmio_reg_offset(GEN8_L3SQCREG4); - *batch++ = i915_ggtt_offset(engine->scratch) + 256; + *batch++ = i915_scratch_offset(engine->i915) + 256; *batch++ = 0; return batch; @@ -1332,7 +1347,7 @@ static u32 *gen8_init_indirectctx_bb(struct intel_engine_cs *engine, u32 *batch) PIPE_CONTROL_GLOBAL_GTT_IVB | PIPE_CONTROL_CS_STALL | PIPE_CONTROL_QW_WRITE, - i915_ggtt_offset(engine->scratch) + + i915_scratch_offset(engine->i915) + 2 * CACHELINE_BYTES); *batch++ = MI_ARB_ON_OFF | MI_ARB_ENABLE; @@ -1401,18 +1416,6 @@ static u32 *gen9_init_indirectctx_bb(struct intel_engine_cs *engine, u32 *batch) batch = emit_lri(batch, lri, ARRAY_SIZE(lri)); - /* WaClearSlmSpaceAtContextSwitch:kbl */ - /* Actual scratch location is at 128 bytes offset */ - if (IS_KBL_REVID(engine->i915, 0, KBL_REVID_A0)) { - batch = gen8_emit_pipe_control(batch, - PIPE_CONTROL_FLUSH_L3 | - PIPE_CONTROL_GLOBAL_GTT_IVB | - PIPE_CONTROL_CS_STALL | - PIPE_CONTROL_QW_WRITE, - i915_ggtt_offset(engine->scratch) - + 2 * CACHELINE_BYTES); - } - /* WaMediaPoolStateCmdInWABB:bxt,glk */ if (HAS_POOLED_EU(engine->i915)) { /* @@ -1629,6 +1632,8 @@ static bool unexpected_starting_state(struct intel_engine_cs *engine) static int gen8_init_common_ring(struct intel_engine_cs *engine) { + intel_engine_apply_workarounds(engine); + intel_mocs_init_engine(engine); intel_engine_reset_breadcrumbs(engine); @@ -1653,7 +1658,7 @@ static int gen8_init_render_ring(struct intel_engine_cs *engine) if (ret) return ret; - intel_whitelist_workarounds_apply(engine); + intel_engine_apply_whitelist(engine); /* We need to disable the AsyncFlip performance optimisations in order * to use MI_WAIT_FOR_EVENT within the CS. It should already be @@ -1676,7 +1681,7 @@ static int gen9_init_render_ring(struct intel_engine_cs *engine) if (ret) return ret; - intel_whitelist_workarounds_apply(engine); + intel_engine_apply_whitelist(engine); return 0; } @@ -1974,7 +1979,7 @@ static int gen8_emit_flush_render(struct i915_request *request, { struct intel_engine_cs *engine = request->engine; u32 scratch_addr = - i915_ggtt_offset(engine->scratch) + 2 * CACHELINE_BYTES; + i915_scratch_offset(engine->i915) + 2 * CACHELINE_BYTES; bool vf_flush_wa = false, dc_flush_wa = false; u32 *cs, flags = 0; int len; @@ -2088,7 +2093,7 @@ static int gen8_init_rcs_context(struct i915_request *rq) { int ret; - ret = intel_ctx_workarounds_emit(rq); + ret = intel_engine_emit_ctx_wa(rq); if (ret) return ret; @@ -2229,12 +2234,6 @@ logical_ring_setup(struct intel_engine_cs *engine) logical_ring_default_irqs(engine); } -static bool csb_force_mmio(struct drm_i915_private *i915) -{ - /* Older GVT emulation depends upon intercepting CSB mmio */ - return intel_vgpu_active(i915) && !intel_vgpu_has_hwsp_emulation(i915); -} - static int logical_ring_init(struct intel_engine_cs *engine) { struct drm_i915_private *i915 = engine->i915; @@ -2264,24 +2263,12 @@ static int logical_ring_init(struct intel_engine_cs *engine) upper_32_bits(ce->lrc_desc); } - execlists->csb_read = - i915->regs + i915_mmio_reg_offset(RING_CONTEXT_STATUS_PTR(engine)); - if (csb_force_mmio(i915)) { - execlists->csb_status = (u32 __force *) - (i915->regs + i915_mmio_reg_offset(RING_CONTEXT_STATUS_BUF_LO(engine, 0))); + execlists->csb_status = + &engine->status_page.page_addr[I915_HWS_CSB_BUF0_INDEX]; - execlists->csb_write = (u32 __force *)execlists->csb_read; - execlists->csb_write_reset = - _MASKED_FIELD(GEN8_CSB_WRITE_PTR_MASK, - GEN8_CSB_ENTRIES - 1); - } else { - execlists->csb_status = - &engine->status_page.page_addr[I915_HWS_CSB_BUF0_INDEX]; + execlists->csb_write = + &engine->status_page.page_addr[intel_hws_csb_write_index(i915)]; - execlists->csb_write = - &engine->status_page.page_addr[intel_hws_csb_write_index(i915)]; - execlists->csb_write_reset = GEN8_CSB_ENTRIES - 1; - } reset_csb_pointers(execlists); return 0; @@ -2311,10 +2298,6 @@ int logical_render_ring_init(struct intel_engine_cs *engine) if (ret) return ret; - ret = intel_engine_create_scratch(engine, PAGE_SIZE); - if (ret) - goto err_cleanup_common; - ret = intel_init_workaround_bb(engine); if (ret) { /* @@ -2326,11 +2309,10 @@ int logical_render_ring_init(struct intel_engine_cs *engine) ret); } - return 0; + intel_engine_init_whitelist(engine); + intel_engine_init_workarounds(engine); -err_cleanup_common: - intel_engine_cleanup_common(engine); - return ret; + return 0; } int logical_xcs_ring_init(struct intel_engine_cs *engine) diff --git a/drivers/gpu/drm/i915/intel_opregion.h b/drivers/gpu/drm/i915/intel_opregion.h index d84b6d2d2fae..4aa68ffbd30e 100644 --- a/drivers/gpu/drm/i915/intel_opregion.h +++ b/drivers/gpu/drm/i915/intel_opregion.h @@ -87,12 +87,12 @@ static inline void intel_opregion_unregister(struct drm_i915_private *dev_priv) { } -void intel_opregion_resume(struct drm_i915_private *dev_priv) +static inline void intel_opregion_resume(struct drm_i915_private *dev_priv) { } -void intel_opregion_suspend(struct drm_i915_private *dev_priv, - pci_power_t state) +static inline void intel_opregion_suspend(struct drm_i915_private *dev_priv, + pci_power_t state) { } diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 897a791662c5..a26b4eddda25 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -3951,68 +3951,68 @@ static void skl_ddb_get_hw_plane_state(struct drm_i915_private *dev_priv, const enum pipe pipe, const enum plane_id plane_id, - struct skl_ddb_allocation *ddb /* out */) + struct skl_ddb_entry *ddb_y, + struct skl_ddb_entry *ddb_uv) { - u32 val, val2 = 0; - int fourcc, pixel_format; + u32 val, val2; + u32 fourcc = 0; /* Cursor doesn't support NV12/planar, so no extra calculation needed */ if (plane_id == PLANE_CURSOR) { val = I915_READ(CUR_BUF_CFG(pipe)); - skl_ddb_entry_init_from_hw(dev_priv, - &ddb->plane[pipe][plane_id], val); + skl_ddb_entry_init_from_hw(dev_priv, ddb_y, val); return; } val = I915_READ(PLANE_CTL(pipe, plane_id)); /* No DDB allocated for disabled planes */ - if (!(val & PLANE_CTL_ENABLE)) - return; - - pixel_format = val & PLANE_CTL_FORMAT_MASK; - fourcc = skl_format_to_fourcc(pixel_format, - val & PLANE_CTL_ORDER_RGBX, - val & PLANE_CTL_ALPHA_MASK); + if (val & PLANE_CTL_ENABLE) + fourcc = skl_format_to_fourcc(val & PLANE_CTL_FORMAT_MASK, + val & PLANE_CTL_ORDER_RGBX, + val & PLANE_CTL_ALPHA_MASK); - val = I915_READ(PLANE_BUF_CFG(pipe, plane_id)); - if (fourcc == DRM_FORMAT_NV12 && INTEL_GEN(dev_priv) < 11) { + if (INTEL_GEN(dev_priv) >= 11) { + val = I915_READ(PLANE_BUF_CFG(pipe, plane_id)); + skl_ddb_entry_init_from_hw(dev_priv, ddb_y, val); + } else { + val = I915_READ(PLANE_BUF_CFG(pipe, plane_id)); val2 = I915_READ(PLANE_NV12_BUF_CFG(pipe, plane_id)); - skl_ddb_entry_init_from_hw(dev_priv, - &ddb->plane[pipe][plane_id], val2); - skl_ddb_entry_init_from_hw(dev_priv, - &ddb->uv_plane[pipe][plane_id], val); - } else { - skl_ddb_entry_init_from_hw(dev_priv, - &ddb->plane[pipe][plane_id], val); + if (fourcc == DRM_FORMAT_NV12) + swap(val, val2); + + skl_ddb_entry_init_from_hw(dev_priv, ddb_y, val); + skl_ddb_entry_init_from_hw(dev_priv, ddb_uv, val2); } } -void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv, - struct skl_ddb_allocation *ddb /* out */) +void skl_pipe_ddb_get_hw_state(struct intel_crtc *crtc, + struct skl_ddb_entry *ddb_y, + struct skl_ddb_entry *ddb_uv) { - struct intel_crtc *crtc; - - memset(ddb, 0, sizeof(*ddb)); - - ddb->enabled_slices = intel_enabled_dbuf_slices_num(dev_priv); + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + enum intel_display_power_domain power_domain; + enum pipe pipe = crtc->pipe; + enum plane_id plane_id; - for_each_intel_crtc(&dev_priv->drm, crtc) { - enum intel_display_power_domain power_domain; - enum plane_id plane_id; - enum pipe pipe = crtc->pipe; + power_domain = POWER_DOMAIN_PIPE(pipe); + if (!intel_display_power_get_if_enabled(dev_priv, power_domain)) + return; - power_domain = POWER_DOMAIN_PIPE(pipe); - if (!intel_display_power_get_if_enabled(dev_priv, power_domain)) - continue; + for_each_plane_id_on_crtc(crtc, plane_id) + skl_ddb_get_hw_plane_state(dev_priv, pipe, + plane_id, + &ddb_y[plane_id], + &ddb_uv[plane_id]); - for_each_plane_id_on_crtc(crtc, plane_id) - skl_ddb_get_hw_plane_state(dev_priv, pipe, - plane_id, ddb); + intel_display_power_put(dev_priv, power_domain); +} - intel_display_power_put(dev_priv, power_domain); - } +void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv, + struct skl_ddb_allocation *ddb /* out */) +{ + ddb->enabled_slices = intel_enabled_dbuf_slices_num(dev_priv); } /* @@ -4410,7 +4410,6 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate, struct drm_crtc *crtc = cstate->base.crtc; struct drm_i915_private *dev_priv = to_i915(crtc->dev); struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - enum pipe pipe = intel_crtc->pipe; struct skl_ddb_entry *alloc = &cstate->wm.skl.ddb; uint16_t alloc_size, start; uint16_t minimum[I915_MAX_PLANES] = {}; @@ -4423,8 +4422,8 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate, uint16_t total_min_blocks = 0; /* Clear the partitioning for disabled planes. */ - memset(ddb->plane[pipe], 0, sizeof(ddb->plane[pipe])); - memset(ddb->uv_plane[pipe], 0, sizeof(ddb->uv_plane[pipe])); + memset(cstate->wm.skl.plane_ddb_y, 0, sizeof(cstate->wm.skl.plane_ddb_y)); + memset(cstate->wm.skl.plane_ddb_uv, 0, sizeof(cstate->wm.skl.plane_ddb_uv)); if (WARN_ON(!state)) return 0; @@ -4471,8 +4470,8 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate, } alloc_size -= total_min_blocks; - ddb->plane[pipe][PLANE_CURSOR].start = alloc->end - minimum[PLANE_CURSOR]; - ddb->plane[pipe][PLANE_CURSOR].end = alloc->end; + cstate->wm.skl.plane_ddb_y[PLANE_CURSOR].start = alloc->end - minimum[PLANE_CURSOR]; + cstate->wm.skl.plane_ddb_y[PLANE_CURSOR].end = alloc->end; /* * 2. Distribute the remaining space in proportion to the amount of @@ -4503,8 +4502,8 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate, /* Leave disabled planes at (0,0) */ if (data_rate) { - ddb->plane[pipe][plane_id].start = start; - ddb->plane[pipe][plane_id].end = start + plane_blocks; + cstate->wm.skl.plane_ddb_y[plane_id].start = start; + cstate->wm.skl.plane_ddb_y[plane_id].end = start + plane_blocks; } start += plane_blocks; @@ -4519,8 +4518,8 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate, WARN_ON(INTEL_GEN(dev_priv) >= 11 && uv_plane_blocks); if (uv_data_rate) { - ddb->uv_plane[pipe][plane_id].start = start; - ddb->uv_plane[pipe][plane_id].end = + cstate->wm.skl.plane_ddb_uv[plane_id].start = start; + cstate->wm.skl.plane_ddb_uv[plane_id].end = start + uv_plane_blocks; } @@ -4617,12 +4616,12 @@ skl_adjusted_plane_pixel_rate(const struct intel_crtc_state *cstate, } static int -skl_compute_plane_wm_params(const struct drm_i915_private *dev_priv, - const struct intel_crtc_state *cstate, +skl_compute_plane_wm_params(const struct intel_crtc_state *cstate, const struct intel_plane_state *intel_pstate, - struct skl_wm_params *wp, int plane_id) + struct skl_wm_params *wp, int color_plane) { struct intel_plane *plane = to_intel_plane(intel_pstate->base.plane); + struct drm_i915_private *dev_priv = to_i915(plane->base.dev); const struct drm_plane_state *pstate = &intel_pstate->base; const struct drm_framebuffer *fb = pstate->fb; uint32_t interm_pbpl; @@ -4630,11 +4629,8 @@ skl_compute_plane_wm_params(const struct drm_i915_private *dev_priv, to_intel_atomic_state(cstate->base.state); bool apply_memory_bw_wa = skl_needs_memory_bw_wa(state); - if (!intel_wm_plane_visible(cstate, intel_pstate)) - return 0; - /* only NV12 format has two planes */ - if (plane_id == 1 && fb->format->format != DRM_FORMAT_NV12) { + if (color_plane == 1 && fb->format->format != DRM_FORMAT_NV12) { DRM_DEBUG_KMS("Non NV12 format have single plane\n"); return -EINVAL; } @@ -4659,10 +4655,10 @@ skl_compute_plane_wm_params(const struct drm_i915_private *dev_priv, wp->width = drm_rect_width(&intel_pstate->base.src) >> 16; } - if (plane_id == 1 && wp->is_planar) + if (color_plane == 1 && wp->is_planar) wp->width /= 2; - wp->cpp = fb->format->cpp[plane_id]; + wp->cpp = fb->format->cpp[color_plane]; wp->plane_pixel_rate = skl_adjusted_plane_pixel_rate(cstate, intel_pstate); @@ -4724,8 +4720,7 @@ skl_compute_plane_wm_params(const struct drm_i915_private *dev_priv, return 0; } -static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, - const struct intel_crtc_state *cstate, +static int skl_compute_plane_wm(const struct intel_crtc_state *cstate, const struct intel_plane_state *intel_pstate, uint16_t ddb_allocation, int level, @@ -4733,6 +4728,8 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, const struct skl_wm_level *result_prev, struct skl_wm_level *result /* out */) { + struct drm_i915_private *dev_priv = + to_i915(intel_pstate->base.plane->dev); const struct drm_plane_state *pstate = &intel_pstate->base; uint32_t latency = dev_priv->wm.skl_latency[level]; uint_fixed_16_16_t method1, method2; @@ -4743,11 +4740,8 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, bool apply_memory_bw_wa = skl_needs_memory_bw_wa(state); uint32_t min_disp_buf_needed; - if (latency == 0 || - !intel_wm_plane_visible(cstate, intel_pstate)) { - result->plane_en = false; - return 0; - } + if (latency == 0) + return level == 0 ? -EINVAL : 0; /* Display WA #1141: kbl,cfl */ if ((IS_KABYLAKE(dev_priv) || IS_COFFEELAKE(dev_priv) || @@ -4844,8 +4838,6 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, if ((level > 0 && res_lines > 31) || res_blocks >= ddb_allocation || min_disp_buf_needed >= ddb_allocation) { - result->plane_en = false; - /* * If there are no valid level 0 watermarks, then we can't * support this display configuration. @@ -4872,27 +4864,22 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, } static int -skl_compute_wm_levels(const struct drm_i915_private *dev_priv, - struct skl_ddb_allocation *ddb, - const struct intel_crtc_state *cstate, +skl_compute_wm_levels(const struct intel_crtc_state *cstate, const struct intel_plane_state *intel_pstate, uint16_t ddb_blocks, const struct skl_wm_params *wm_params, - struct skl_plane_wm *wm, struct skl_wm_level *levels) { + struct drm_i915_private *dev_priv = + to_i915(intel_pstate->base.plane->dev); int level, max_level = ilk_wm_max_level(dev_priv); struct skl_wm_level *result_prev = &levels[0]; int ret; - if (WARN_ON(!intel_pstate->base.fb)) - return -EINVAL; - for (level = 0; level <= max_level; level++) { struct skl_wm_level *result = &levels[level]; - ret = skl_compute_plane_wm(dev_priv, - cstate, + ret = skl_compute_plane_wm(cstate, intel_pstate, ddb_blocks, level, @@ -4905,9 +4892,6 @@ skl_compute_wm_levels(const struct drm_i915_private *dev_priv, result_prev = result; } - if (intel_pstate->base.fb->format->format == DRM_FORMAT_NV12) - wm->is_planar = true; - return 0; } @@ -4935,10 +4919,9 @@ skl_compute_linetime_wm(const struct intel_crtc_state *cstate) } static void skl_compute_transition_wm(const struct intel_crtc_state *cstate, - struct skl_wm_params *wp, - struct skl_wm_level *wm_l0, - uint16_t ddb_allocation, - struct skl_wm_level *trans_wm /* out */) + const struct skl_wm_params *wp, + struct skl_plane_wm *wm, + uint16_t ddb_allocation) { struct drm_device *dev = cstate->base.crtc->dev; const struct drm_i915_private *dev_priv = to_i915(dev); @@ -4946,16 +4929,13 @@ static void skl_compute_transition_wm(const struct intel_crtc_state *cstate, const uint16_t trans_amount = 10; /* This is configurable amount */ uint16_t wm0_sel_res_b, trans_offset_b, res_blocks; - if (!cstate->base.active) - goto exit; - /* Transition WM are not recommended by HW team for GEN9 */ if (INTEL_GEN(dev_priv) <= 9) - goto exit; + return; /* Transition WM don't make any sense if ipc is disabled */ if (!dev_priv->ipc_enabled) - goto exit; + return; trans_min = 14; if (INTEL_GEN(dev_priv) >= 11) @@ -4973,7 +4953,7 @@ static void skl_compute_transition_wm(const struct intel_crtc_state *cstate, * Result Blocks is Result Blocks minus 1 and it should work for the * current platforms. */ - wm0_sel_res_b = wm_l0->plane_res_b - 1; + wm0_sel_res_b = wm->wm[0].plane_res_b - 1; if (wp->y_tiled) { trans_y_tile_min = (uint16_t) mul_round_up_u32_fixed16(2, @@ -4992,107 +4972,129 @@ static void skl_compute_transition_wm(const struct intel_crtc_state *cstate, res_blocks += 1; if (res_blocks < ddb_allocation) { - trans_wm->plane_res_b = res_blocks; - trans_wm->plane_en = true; - return; + wm->trans_wm.plane_res_b = res_blocks; + wm->trans_wm.plane_en = true; } - -exit: - trans_wm->plane_en = false; } -static int __skl_build_plane_wm_single(struct skl_ddb_allocation *ddb, - struct skl_pipe_wm *pipe_wm, - enum plane_id plane_id, - const struct intel_crtc_state *cstate, - const struct intel_plane_state *pstate, - int color_plane) +static int skl_build_plane_wm_single(struct intel_crtc_state *crtc_state, + const struct intel_plane_state *plane_state, + enum plane_id plane_id, int color_plane) { - struct drm_i915_private *dev_priv = to_i915(pstate->base.plane->dev); - struct skl_plane_wm *wm = &pipe_wm->planes[plane_id]; - enum pipe pipe = to_intel_plane(pstate->base.plane)->pipe; + struct skl_plane_wm *wm = &crtc_state->wm.skl.optimal.planes[plane_id]; + u16 ddb_blocks = skl_ddb_entry_size(&crtc_state->wm.skl.plane_ddb_y[plane_id]); struct skl_wm_params wm_params; - uint16_t ddb_blocks = skl_ddb_entry_size(&ddb->plane[pipe][plane_id]); int ret; - ret = skl_compute_plane_wm_params(dev_priv, cstate, pstate, + ret = skl_compute_plane_wm_params(crtc_state, plane_state, &wm_params, color_plane); if (ret) return ret; - ret = skl_compute_wm_levels(dev_priv, ddb, cstate, pstate, - ddb_blocks, &wm_params, wm, wm->wm); - + ret = skl_compute_wm_levels(crtc_state, plane_state, + ddb_blocks, &wm_params, wm->wm); if (ret) return ret; - skl_compute_transition_wm(cstate, &wm_params, &wm->wm[0], - ddb_blocks, &wm->trans_wm); + skl_compute_transition_wm(crtc_state, &wm_params, wm, ddb_blocks); return 0; } -static int skl_build_plane_wm_single(struct skl_ddb_allocation *ddb, - struct skl_pipe_wm *pipe_wm, - const struct intel_crtc_state *cstate, - const struct intel_plane_state *pstate) -{ - enum plane_id plane_id = to_intel_plane(pstate->base.plane)->id; - - return __skl_build_plane_wm_single(ddb, pipe_wm, plane_id, cstate, pstate, 0); -} - -static int skl_build_plane_wm_planar(struct skl_ddb_allocation *ddb, - struct skl_pipe_wm *pipe_wm, - const struct intel_crtc_state *cstate, - const struct intel_plane_state *pstate) +static int skl_build_plane_wm_uv(struct intel_crtc_state *crtc_state, + const struct intel_plane_state *plane_state, + enum plane_id plane_id) { - struct intel_plane *plane = to_intel_plane(pstate->base.plane); - struct drm_i915_private *dev_priv = to_i915(plane->base.dev); - enum plane_id plane_id = plane->id; - struct skl_plane_wm *wm = &pipe_wm->planes[plane_id]; + struct skl_plane_wm *wm = &crtc_state->wm.skl.optimal.planes[plane_id]; + u16 ddb_blocks = skl_ddb_entry_size(&crtc_state->wm.skl.plane_ddb_uv[plane_id]); struct skl_wm_params wm_params; - enum pipe pipe = plane->pipe; - uint16_t ddb_blocks = skl_ddb_entry_size(&ddb->plane[pipe][plane_id]); int ret; - ret = __skl_build_plane_wm_single(ddb, pipe_wm, plane_id, cstate, pstate, 0); - if (ret) - return ret; + wm->is_planar = true; /* uv plane watermarks must also be validated for NV12/Planar */ - ddb_blocks = skl_ddb_entry_size(&ddb->uv_plane[pipe][plane_id]); + ret = skl_compute_plane_wm_params(crtc_state, plane_state, + &wm_params, 1); + if (ret) + return ret; - ret = skl_compute_plane_wm_params(dev_priv, cstate, pstate, &wm_params, 1); + ret = skl_compute_wm_levels(crtc_state, plane_state, + ddb_blocks, &wm_params, wm->uv_wm); if (ret) return ret; - return skl_compute_wm_levels(dev_priv, ddb, cstate, pstate, - ddb_blocks, &wm_params, wm, wm->uv_wm); + return 0; } -static int icl_build_plane_wm_planar(struct skl_ddb_allocation *ddb, - struct skl_pipe_wm *pipe_wm, - const struct intel_crtc_state *cstate, - const struct intel_plane_state *pstate) +static int skl_build_plane_wm(struct skl_pipe_wm *pipe_wm, + struct intel_crtc_state *crtc_state, + const struct intel_plane_state *plane_state) { + struct intel_plane *plane = to_intel_plane(plane_state->base.plane); + const struct drm_framebuffer *fb = plane_state->base.fb; + enum plane_id plane_id = plane->id; int ret; - enum plane_id y_plane_id = pstate->linked_plane->id; - enum plane_id uv_plane_id = to_intel_plane(pstate->base.plane)->id; - ret = __skl_build_plane_wm_single(ddb, pipe_wm, y_plane_id, - cstate, pstate, 0); + if (!intel_wm_plane_visible(crtc_state, plane_state)) + return 0; + + ret = skl_build_plane_wm_single(crtc_state, plane_state, + plane_id, 0); if (ret) return ret; - return __skl_build_plane_wm_single(ddb, pipe_wm, uv_plane_id, - cstate, pstate, 1); + if (fb->format->is_yuv && fb->format->num_planes > 1) { + ret = skl_build_plane_wm_uv(crtc_state, plane_state, + plane_id); + if (ret) + return ret; + } + + return 0; +} + +static int icl_build_plane_wm(struct skl_pipe_wm *pipe_wm, + struct intel_crtc_state *crtc_state, + const struct intel_plane_state *plane_state) +{ + enum plane_id plane_id = to_intel_plane(plane_state->base.plane)->id; + int ret; + + /* Watermarks calculated in master */ + if (plane_state->slave) + return 0; + + if (plane_state->linked_plane) { + const struct drm_framebuffer *fb = plane_state->base.fb; + enum plane_id y_plane_id = plane_state->linked_plane->id; + + WARN_ON(!intel_wm_plane_visible(crtc_state, plane_state)); + WARN_ON(!fb->format->is_yuv || + fb->format->num_planes == 1); + + ret = skl_build_plane_wm_single(crtc_state, plane_state, + y_plane_id, 0); + if (ret) + return ret; + + ret = skl_build_plane_wm_single(crtc_state, plane_state, + plane_id, 1); + if (ret) + return ret; + } else if (intel_wm_plane_visible(crtc_state, plane_state)) { + ret = skl_build_plane_wm_single(crtc_state, plane_state, + plane_id, 0); + if (ret) + return ret; + } + + return 0; } static int skl_build_pipe_wm(struct intel_crtc_state *cstate, - struct skl_ddb_allocation *ddb, struct skl_pipe_wm *pipe_wm) { + struct drm_i915_private *dev_priv = to_i915(cstate->base.crtc->dev); struct drm_crtc_state *crtc_state = &cstate->base; struct drm_plane *plane; const struct drm_plane_state *pstate; @@ -5108,18 +5110,12 @@ static int skl_build_pipe_wm(struct intel_crtc_state *cstate, const struct intel_plane_state *intel_pstate = to_intel_plane_state(pstate); - /* Watermarks calculated in master */ - if (intel_pstate->slave) - continue; - - if (intel_pstate->linked_plane) - ret = icl_build_plane_wm_planar(ddb, pipe_wm, cstate, intel_pstate); - else if (intel_pstate->base.fb && - intel_pstate->base.fb->format->format == DRM_FORMAT_NV12) - ret = skl_build_plane_wm_planar(ddb, pipe_wm, cstate, intel_pstate); + if (INTEL_GEN(dev_priv) >= 11) + ret = icl_build_plane_wm(pipe_wm, + cstate, intel_pstate); else - ret = skl_build_plane_wm_single(ddb, pipe_wm, cstate, intel_pstate); - + ret = skl_build_plane_wm(pipe_wm, + cstate, intel_pstate); if (ret) return ret; } @@ -5134,9 +5130,9 @@ static void skl_ddb_entry_write(struct drm_i915_private *dev_priv, const struct skl_ddb_entry *entry) { if (entry->end) - I915_WRITE(reg, (entry->end - 1) << 16 | entry->start); + I915_WRITE_FW(reg, (entry->end - 1) << 16 | entry->start); else - I915_WRITE(reg, 0); + I915_WRITE_FW(reg, 0); } static void skl_write_wm_level(struct drm_i915_private *dev_priv, @@ -5151,19 +5147,22 @@ static void skl_write_wm_level(struct drm_i915_private *dev_priv, val |= level->plane_res_l << PLANE_WM_LINES_SHIFT; } - I915_WRITE(reg, val); + I915_WRITE_FW(reg, val); } -static void skl_write_plane_wm(struct intel_crtc *intel_crtc, - const struct skl_plane_wm *wm, - const struct skl_ddb_allocation *ddb, - enum plane_id plane_id) +void skl_write_plane_wm(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state) { - struct drm_crtc *crtc = &intel_crtc->base; - struct drm_device *dev = crtc->dev; - struct drm_i915_private *dev_priv = to_i915(dev); + struct drm_i915_private *dev_priv = to_i915(plane->base.dev); int level, max_level = ilk_wm_max_level(dev_priv); - enum pipe pipe = intel_crtc->pipe; + enum plane_id plane_id = plane->id; + enum pipe pipe = plane->pipe; + const struct skl_plane_wm *wm = + &crtc_state->wm.skl.optimal.planes[plane_id]; + const struct skl_ddb_entry *ddb_y = + &crtc_state->wm.skl.plane_ddb_y[plane_id]; + const struct skl_ddb_entry *ddb_uv = + &crtc_state->wm.skl.plane_ddb_uv[plane_id]; for (level = 0; level <= max_level; level++) { skl_write_wm_level(dev_priv, PLANE_WM(pipe, plane_id, level), @@ -5172,29 +5171,32 @@ static void skl_write_plane_wm(struct intel_crtc *intel_crtc, skl_write_wm_level(dev_priv, PLANE_WM_TRANS(pipe, plane_id), &wm->trans_wm); - if (wm->is_planar && INTEL_GEN(dev_priv) < 11) { - skl_ddb_entry_write(dev_priv, PLANE_BUF_CFG(pipe, plane_id), - &ddb->uv_plane[pipe][plane_id]); + if (INTEL_GEN(dev_priv) >= 11) { skl_ddb_entry_write(dev_priv, - PLANE_NV12_BUF_CFG(pipe, plane_id), - &ddb->plane[pipe][plane_id]); - } else { - skl_ddb_entry_write(dev_priv, PLANE_BUF_CFG(pipe, plane_id), - &ddb->plane[pipe][plane_id]); - if (INTEL_GEN(dev_priv) < 11) - I915_WRITE(PLANE_NV12_BUF_CFG(pipe, plane_id), 0x0); + PLANE_BUF_CFG(pipe, plane_id), ddb_y); + return; } + + if (wm->is_planar) + swap(ddb_y, ddb_uv); + + skl_ddb_entry_write(dev_priv, + PLANE_BUF_CFG(pipe, plane_id), ddb_y); + skl_ddb_entry_write(dev_priv, + PLANE_NV12_BUF_CFG(pipe, plane_id), ddb_uv); } -static void skl_write_cursor_wm(struct intel_crtc *intel_crtc, - const struct skl_plane_wm *wm, - const struct skl_ddb_allocation *ddb) +void skl_write_cursor_wm(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state) { - struct drm_crtc *crtc = &intel_crtc->base; - struct drm_device *dev = crtc->dev; - struct drm_i915_private *dev_priv = to_i915(dev); + struct drm_i915_private *dev_priv = to_i915(plane->base.dev); int level, max_level = ilk_wm_max_level(dev_priv); - enum pipe pipe = intel_crtc->pipe; + enum plane_id plane_id = plane->id; + enum pipe pipe = plane->pipe; + const struct skl_plane_wm *wm = + &crtc_state->wm.skl.optimal.planes[plane_id]; + const struct skl_ddb_entry *ddb = + &crtc_state->wm.skl.plane_ddb_y[plane_id]; for (level = 0; level <= max_level; level++) { skl_write_wm_level(dev_priv, CUR_WM(pipe, level), @@ -5202,22 +5204,30 @@ static void skl_write_cursor_wm(struct intel_crtc *intel_crtc, } skl_write_wm_level(dev_priv, CUR_WM_TRANS(pipe), &wm->trans_wm); - skl_ddb_entry_write(dev_priv, CUR_BUF_CFG(pipe), - &ddb->plane[pipe][PLANE_CURSOR]); + skl_ddb_entry_write(dev_priv, CUR_BUF_CFG(pipe), ddb); } bool skl_wm_level_equals(const struct skl_wm_level *l1, const struct skl_wm_level *l2) { - if (l1->plane_en != l2->plane_en) - return false; + return l1->plane_en == l2->plane_en && + l1->plane_res_l == l2->plane_res_l && + l1->plane_res_b == l2->plane_res_b; +} - /* If both planes aren't enabled, the rest shouldn't matter */ - if (!l1->plane_en) - return true; +static bool skl_plane_wm_equals(struct drm_i915_private *dev_priv, + const struct skl_plane_wm *wm1, + const struct skl_plane_wm *wm2) +{ + int level, max_level = ilk_wm_max_level(dev_priv); - return (l1->plane_res_l == l2->plane_res_l && - l1->plane_res_b == l2->plane_res_b); + for (level = 0; level <= max_level; level++) { + if (!skl_wm_level_equals(&wm1->wm[level], &wm2->wm[level]) || + !skl_wm_level_equals(&wm1->uv_wm[level], &wm2->uv_wm[level])) + return false; + } + + return skl_wm_level_equals(&wm1->trans_wm, &wm2->trans_wm); } static inline bool skl_ddb_entries_overlap(const struct skl_ddb_entry *a, @@ -5244,13 +5254,12 @@ bool skl_ddb_allocation_overlaps(const struct skl_ddb_entry *ddb, static int skl_update_pipe_wm(struct drm_crtc_state *cstate, const struct skl_pipe_wm *old_pipe_wm, struct skl_pipe_wm *pipe_wm, /* out */ - struct skl_ddb_allocation *ddb, /* out */ bool *changed /* out */) { struct intel_crtc_state *intel_cstate = to_intel_crtc_state(cstate); int ret; - ret = skl_build_pipe_wm(intel_cstate, ddb, pipe_wm); + ret = skl_build_pipe_wm(intel_cstate, pipe_wm); if (ret) return ret; @@ -5276,42 +5285,29 @@ pipes_modified(struct drm_atomic_state *state) } static int -skl_ddb_add_affected_planes(struct intel_crtc_state *cstate) +skl_ddb_add_affected_planes(const struct intel_crtc_state *old_crtc_state, + struct intel_crtc_state *new_crtc_state) { - struct drm_atomic_state *state = cstate->base.state; - struct drm_device *dev = state->dev; - struct drm_crtc *crtc = cstate->base.crtc; - struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - struct drm_i915_private *dev_priv = to_i915(dev); - struct intel_atomic_state *intel_state = to_intel_atomic_state(state); - struct skl_ddb_allocation *new_ddb = &intel_state->wm_results.ddb; - struct skl_ddb_allocation *cur_ddb = &dev_priv->wm.skl_hw.ddb; - struct drm_plane *plane; - enum pipe pipe = intel_crtc->pipe; + struct intel_atomic_state *state = to_intel_atomic_state(new_crtc_state->base.state); + struct intel_crtc *crtc = to_intel_crtc(new_crtc_state->base.crtc); + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + struct intel_plane *plane; - drm_for_each_plane_mask(plane, dev, cstate->base.plane_mask) { - struct drm_plane_state *plane_state; - struct intel_plane *linked; - enum plane_id plane_id = to_intel_plane(plane)->id; + for_each_intel_plane_on_crtc(&dev_priv->drm, crtc, plane) { + struct intel_plane_state *plane_state; + enum plane_id plane_id = plane->id; - if (skl_ddb_entry_equal(&cur_ddb->plane[pipe][plane_id], - &new_ddb->plane[pipe][plane_id]) && - skl_ddb_entry_equal(&cur_ddb->uv_plane[pipe][plane_id], - &new_ddb->uv_plane[pipe][plane_id])) + if (skl_ddb_entry_equal(&old_crtc_state->wm.skl.plane_ddb_y[plane_id], + &new_crtc_state->wm.skl.plane_ddb_y[plane_id]) && + skl_ddb_entry_equal(&old_crtc_state->wm.skl.plane_ddb_uv[plane_id], + &new_crtc_state->wm.skl.plane_ddb_uv[plane_id])) continue; - plane_state = drm_atomic_get_plane_state(state, plane); + plane_state = intel_atomic_get_plane_state(state, plane); if (IS_ERR(plane_state)) return PTR_ERR(plane_state); - /* Make sure linked plane is updated too */ - linked = to_intel_plane_state(plane_state)->linked_plane; - if (!linked) - continue; - - plane_state = drm_atomic_get_plane_state(state, &linked->base); - if (IS_ERR(plane_state)) - return PTR_ERR(plane_state); + new_crtc_state->update_planes |= BIT(plane_id); } return 0; @@ -5323,18 +5319,21 @@ skl_compute_ddb(struct drm_atomic_state *state) const struct drm_i915_private *dev_priv = to_i915(state->dev); struct intel_atomic_state *intel_state = to_intel_atomic_state(state); struct skl_ddb_allocation *ddb = &intel_state->wm_results.ddb; + struct intel_crtc_state *old_crtc_state; + struct intel_crtc_state *new_crtc_state; struct intel_crtc *crtc; - struct intel_crtc_state *cstate; int ret, i; memcpy(ddb, &dev_priv->wm.skl_hw.ddb, sizeof(*ddb)); - for_each_new_intel_crtc_in_state(intel_state, crtc, cstate, i) { - ret = skl_allocate_pipe_ddb(cstate, ddb); + for_each_oldnew_intel_crtc_in_state(intel_state, crtc, old_crtc_state, + new_crtc_state, i) { + ret = skl_allocate_pipe_ddb(new_crtc_state, ddb); if (ret) return ret; - ret = skl_ddb_add_affected_planes(cstate); + ret = skl_ddb_add_affected_planes(old_crtc_state, + new_crtc_state); if (ret) return ret; } @@ -5343,36 +5342,29 @@ skl_compute_ddb(struct drm_atomic_state *state) } static void -skl_print_wm_changes(const struct drm_atomic_state *state) +skl_print_wm_changes(struct intel_atomic_state *state) { - const struct drm_device *dev = state->dev; - const struct drm_i915_private *dev_priv = to_i915(dev); - const struct intel_atomic_state *intel_state = - to_intel_atomic_state(state); - const struct drm_crtc *crtc; - const struct drm_crtc_state *cstate; - const struct intel_plane *intel_plane; - const struct skl_ddb_allocation *old_ddb = &dev_priv->wm.skl_hw.ddb; - const struct skl_ddb_allocation *new_ddb = &intel_state->wm_results.ddb; + struct drm_i915_private *dev_priv = to_i915(state->base.dev); + const struct intel_crtc_state *old_crtc_state; + const struct intel_crtc_state *new_crtc_state; + struct intel_plane *plane; + struct intel_crtc *crtc; int i; - for_each_new_crtc_in_state(state, crtc, cstate, i) { - const struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - enum pipe pipe = intel_crtc->pipe; - - for_each_intel_plane_on_crtc(dev, intel_crtc, intel_plane) { - enum plane_id plane_id = intel_plane->id; + for_each_oldnew_intel_crtc_in_state(state, crtc, old_crtc_state, + new_crtc_state, i) { + for_each_intel_plane_on_crtc(&dev_priv->drm, crtc, plane) { + enum plane_id plane_id = plane->id; const struct skl_ddb_entry *old, *new; - old = &old_ddb->plane[pipe][plane_id]; - new = &new_ddb->plane[pipe][plane_id]; + old = &old_crtc_state->wm.skl.plane_ddb_y[plane_id]; + new = &new_crtc_state->wm.skl.plane_ddb_y[plane_id]; if (skl_ddb_entry_equal(old, new)) continue; DRM_DEBUG_KMS("[PLANE:%d:%s] ddb (%d - %d) -> (%d - %d)\n", - intel_plane->base.base.id, - intel_plane->base.name, + plane->base.base.id, plane->base.name, old->start, old->end, new->start, new->end); } @@ -5469,6 +5461,66 @@ skl_ddb_add_affected_pipes(struct drm_atomic_state *state, bool *changed) return 0; } +/* + * To make sure the cursor watermark registers are always consistent + * with our computed state the following scenario needs special + * treatment: + * + * 1. enable cursor + * 2. move cursor entirely offscreen + * 3. disable cursor + * + * Step 2. does call .disable_plane() but does not zero the watermarks + * (since we consider an offscreen cursor still active for the purposes + * of watermarks). Step 3. would not normally call .disable_plane() + * because the actual plane visibility isn't changing, and we don't + * deallocate the cursor ddb until the pipe gets disabled. So we must + * force step 3. to call .disable_plane() to update the watermark + * registers properly. + * + * Other planes do not suffer from this issues as their watermarks are + * calculated based on the actual plane visibility. The only time this + * can trigger for the other planes is during the initial readout as the + * default value of the watermarks registers is not zero. + */ +static int skl_wm_add_affected_planes(struct intel_atomic_state *state, + struct intel_crtc *crtc) +{ + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + const struct intel_crtc_state *old_crtc_state = + intel_atomic_get_old_crtc_state(state, crtc); + struct intel_crtc_state *new_crtc_state = + intel_atomic_get_new_crtc_state(state, crtc); + struct intel_plane *plane; + + for_each_intel_plane_on_crtc(&dev_priv->drm, crtc, plane) { + struct intel_plane_state *plane_state; + enum plane_id plane_id = plane->id; + + /* + * Force a full wm update for every plane on modeset. + * Required because the reset value of the wm registers + * is non-zero, whereas we want all disabled planes to + * have zero watermarks. So if we turn off the relevant + * power well the hardware state will go out of sync + * with the software state. + */ + if (!drm_atomic_crtc_needs_modeset(&new_crtc_state->base) && + skl_plane_wm_equals(dev_priv, + &old_crtc_state->wm.skl.optimal.planes[plane_id], + &new_crtc_state->wm.skl.optimal.planes[plane_id])) + continue; + + plane_state = intel_atomic_get_plane_state(state, plane); + if (IS_ERR(plane_state)) + return PTR_ERR(plane_state); + + new_crtc_state->update_planes |= BIT(plane_id); + } + + return 0; +} + static int skl_compute_wm(struct drm_atomic_state *state) { @@ -5508,8 +5560,12 @@ skl_compute_wm(struct drm_atomic_state *state) &to_intel_crtc_state(crtc->state)->wm.skl.optimal; pipe_wm = &intel_cstate->wm.skl.optimal; - ret = skl_update_pipe_wm(cstate, old_pipe_wm, pipe_wm, - &results->ddb, &changed); + ret = skl_update_pipe_wm(cstate, old_pipe_wm, pipe_wm, &changed); + if (ret) + return ret; + + ret = skl_wm_add_affected_planes(intel_state, + to_intel_crtc(crtc)); if (ret) return ret; @@ -5523,7 +5579,7 @@ skl_compute_wm(struct drm_atomic_state *state) intel_cstate->update_wm_pre = true; } - skl_print_wm_changes(state); + skl_print_wm_changes(intel_state); return 0; } @@ -5534,23 +5590,12 @@ static void skl_atomic_update_crtc_wm(struct intel_atomic_state *state, struct intel_crtc *crtc = to_intel_crtc(cstate->base.crtc); struct drm_i915_private *dev_priv = to_i915(state->base.dev); struct skl_pipe_wm *pipe_wm = &cstate->wm.skl.optimal; - const struct skl_ddb_allocation *ddb = &state->wm_results.ddb; enum pipe pipe = crtc->pipe; - enum plane_id plane_id; if (!(state->wm_results.dirty_pipes & drm_crtc_mask(&crtc->base))) return; I915_WRITE(PIPE_WM_LINETIME(pipe), pipe_wm->linetime); - - for_each_plane_id_on_crtc(crtc, plane_id) { - if (plane_id != PLANE_CURSOR) - skl_write_plane_wm(crtc, &pipe_wm->planes[plane_id], - ddb, plane_id); - else - skl_write_cursor_wm(crtc, &pipe_wm->planes[plane_id], - ddb); - } } static void skl_initial_wm(struct intel_atomic_state *state, @@ -5560,8 +5605,6 @@ static void skl_initial_wm(struct intel_atomic_state *state, struct drm_device *dev = intel_crtc->base.dev; struct drm_i915_private *dev_priv = to_i915(dev); struct skl_ddb_values *results = &state->wm_results; - struct skl_ddb_values *hw_vals = &dev_priv->wm.skl_hw; - enum pipe pipe = intel_crtc->pipe; if ((results->dirty_pipes & drm_crtc_mask(&intel_crtc->base)) == 0) return; @@ -5571,11 +5614,6 @@ static void skl_initial_wm(struct intel_atomic_state *state, if (cstate->base.active_changed) skl_atomic_update_crtc_wm(state, cstate); - memcpy(hw_vals->ddb.uv_plane[pipe], results->ddb.uv_plane[pipe], - sizeof(hw_vals->ddb.uv_plane[pipe])); - memcpy(hw_vals->ddb.plane[pipe], results->ddb.plane[pipe], - sizeof(hw_vals->ddb.plane[pipe])); - mutex_unlock(&dev_priv->wm.wm_mutex); } @@ -5726,13 +5764,6 @@ void skl_wm_get_hw_state(struct drm_device *dev) if (dev_priv->active_crtcs) { /* Fully recompute DDB on first atomic commit */ dev_priv->wm.distrust_bios_wm = true; - } else { - /* - * Easy/common case; just sanitize DDB now if everything off - * Keep dbuf slice info intact - */ - memset(ddb->plane, 0, sizeof(ddb->plane)); - memset(ddb->uv_plane, 0, sizeof(ddb->uv_plane)); } } diff --git a/drivers/gpu/drm/i915/intel_psr.c b/drivers/gpu/drm/i915/intel_psr.c index 54fa17a5596a..419e56342523 100644 --- a/drivers/gpu/drm/i915/intel_psr.c +++ b/drivers/gpu/drm/i915/intel_psr.c @@ -75,6 +75,10 @@ static bool intel_psr2_enabled(struct drm_i915_private *dev_priv, if (i915_modparams.enable_psr == -1) return false; + /* Cannot enable DSC and PSR2 simultaneously */ + WARN_ON(crtc_state->dsc_params.compression_enable && + crtc_state->has_psr2); + switch (dev_priv->psr.debug & I915_PSR_DEBUG_MODE_MASK) { case I915_PSR_DEBUG_FORCE_PSR1: return false; @@ -169,6 +173,7 @@ void intel_psr_irq_handler(struct drm_i915_private *dev_priv, u32 psr_iir) u32 transcoders = BIT(TRANSCODER_EDP); enum transcoder cpu_transcoder; ktime_t time_ns = ktime_get(); + u32 mask = 0; if (INTEL_GEN(dev_priv) >= 8) transcoders |= BIT(TRANSCODER_A) | @@ -178,10 +183,22 @@ void intel_psr_irq_handler(struct drm_i915_private *dev_priv, u32 psr_iir) for_each_cpu_transcoder_masked(dev_priv, cpu_transcoder, transcoders) { int shift = edp_psr_shift(cpu_transcoder); - /* FIXME: Exit PSR and link train manually when this happens. */ - if (psr_iir & EDP_PSR_ERROR(shift)) - DRM_DEBUG_KMS("[transcoder %s] PSR aux error\n", - transcoder_name(cpu_transcoder)); + if (psr_iir & EDP_PSR_ERROR(shift)) { + DRM_WARN("[transcoder %s] PSR aux error\n", + transcoder_name(cpu_transcoder)); + + dev_priv->psr.irq_aux_error = true; + + /* + * If this interruption is not masked it will keep + * interrupting so fast that it prevents the scheduled + * work to run. + * Also after a PSR error, we don't want to arm PSR + * again so we don't care about unmask the interruption + * or unset irq_aux_error. + */ + mask |= EDP_PSR_ERROR(shift); + } if (psr_iir & EDP_PSR_PRE_ENTRY(shift)) { dev_priv->psr.last_entry_attempt = time_ns; @@ -203,6 +220,13 @@ void intel_psr_irq_handler(struct drm_i915_private *dev_priv, u32 psr_iir) } } } + + if (mask) { + mask |= I915_READ(EDP_PSR_IMR); + I915_WRITE(EDP_PSR_IMR, mask); + + schedule_work(&dev_priv->psr.work); + } } static bool intel_dp_get_colorimetry_status(struct intel_dp *intel_dp) @@ -482,6 +506,16 @@ static bool intel_psr2_config_valid(struct intel_dp *intel_dp, if (!dev_priv->psr.sink_psr2_support) return false; + /* + * DSC and PSR2 cannot be enabled simultaneously. If a requested + * resolution requires DSC to be enabled, priority is given to DSC + * over PSR2. + */ + if (crtc_state->dsc_params.compression_enable) { + DRM_DEBUG_KMS("PSR2 cannot be enabled since DSC is enabled\n"); + return false; + } + if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) { psr_max_h = 4096; psr_max_v = 2304; @@ -527,10 +561,8 @@ void intel_psr_compute_config(struct intel_dp *intel_dp, return; } - if (IS_HASWELL(dev_priv) && - I915_READ(HSW_STEREO_3D_CTL(crtc_state->cpu_transcoder)) & - S3D_ENABLE) { - DRM_DEBUG_KMS("PSR condition failed: Stereo 3D is Enabled\n"); + if (dev_priv->psr.sink_not_reliable) { + DRM_DEBUG_KMS("PSR sink implementation is not reliable\n"); return; } @@ -687,6 +719,7 @@ void intel_psr_enable(struct intel_dp *intel_dp, dev_priv->psr.psr2_enabled = intel_psr2_enabled(dev_priv, crtc_state); dev_priv->psr.busy_frontbuffer_bits = 0; dev_priv->psr.prepared = true; + dev_priv->psr.pipe = to_intel_crtc(crtc_state->base.crtc)->pipe; if (psr_global_enabled(dev_priv->psr.debug)) intel_psr_enable_locked(dev_priv, crtc_state); @@ -933,6 +966,16 @@ int intel_psr_set_debugfs_mode(struct drm_i915_private *dev_priv, return ret; } +static void intel_psr_handle_irq(struct drm_i915_private *dev_priv) +{ + struct i915_psr *psr = &dev_priv->psr; + + intel_psr_disable_locked(psr->dp); + psr->sink_not_reliable = true; + /* let's make sure that sink is awaken */ + drm_dp_dpcd_writeb(&psr->dp->aux, DP_SET_POWER, DP_SET_POWER_D0); +} + static void intel_psr_work(struct work_struct *work) { struct drm_i915_private *dev_priv = @@ -943,6 +986,9 @@ static void intel_psr_work(struct work_struct *work) if (!dev_priv->psr.enabled) goto unlock; + if (READ_ONCE(dev_priv->psr.irq_aux_error)) + intel_psr_handle_irq(dev_priv); + /* * We have to make sure PSR is ready for re-enable * otherwise it keeps disabled until next full enable/disable cycle. @@ -981,9 +1027,6 @@ unlock: void intel_psr_invalidate(struct drm_i915_private *dev_priv, unsigned frontbuffer_bits, enum fb_op_origin origin) { - struct drm_crtc *crtc; - enum pipe pipe; - if (!CAN_PSR(dev_priv)) return; @@ -996,10 +1039,7 @@ void intel_psr_invalidate(struct drm_i915_private *dev_priv, return; } - crtc = dp_to_dig_port(dev_priv->psr.dp)->base.base.crtc; - pipe = to_intel_crtc(crtc)->pipe; - - frontbuffer_bits &= INTEL_FRONTBUFFER_ALL_MASK(pipe); + frontbuffer_bits &= INTEL_FRONTBUFFER_ALL_MASK(dev_priv->psr.pipe); dev_priv->psr.busy_frontbuffer_bits |= frontbuffer_bits; if (frontbuffer_bits) @@ -1024,9 +1064,6 @@ void intel_psr_invalidate(struct drm_i915_private *dev_priv, void intel_psr_flush(struct drm_i915_private *dev_priv, unsigned frontbuffer_bits, enum fb_op_origin origin) { - struct drm_crtc *crtc; - enum pipe pipe; - if (!CAN_PSR(dev_priv)) return; @@ -1039,10 +1076,7 @@ void intel_psr_flush(struct drm_i915_private *dev_priv, return; } - crtc = dp_to_dig_port(dev_priv->psr.dp)->base.base.crtc; - pipe = to_intel_crtc(crtc)->pipe; - - frontbuffer_bits &= INTEL_FRONTBUFFER_ALL_MASK(pipe); + frontbuffer_bits &= INTEL_FRONTBUFFER_ALL_MASK(dev_priv->psr.pipe); dev_priv->psr.busy_frontbuffer_bits &= ~frontbuffer_bits; /* By definition flush = invalidate + flush */ @@ -1056,7 +1090,7 @@ void intel_psr_flush(struct drm_i915_private *dev_priv, * but it makes more sense write to the current active * pipe. */ - I915_WRITE(CURSURFLIVE(pipe), 0); + I915_WRITE(CURSURFLIVE(dev_priv->psr.pipe), 0); } if (!dev_priv->psr.active && !dev_priv->psr.busy_frontbuffer_bits) @@ -1073,6 +1107,8 @@ void intel_psr_flush(struct drm_i915_private *dev_priv, */ void intel_psr_init(struct drm_i915_private *dev_priv) { + u32 val; + if (!HAS_PSR(dev_priv)) return; @@ -1086,6 +1122,22 @@ void intel_psr_init(struct drm_i915_private *dev_priv) if (INTEL_GEN(dev_priv) < 9 || !dev_priv->vbt.psr.enable) i915_modparams.enable_psr = 0; + /* + * If a PSR error happened and the driver is reloaded, the EDP_PSR_IIR + * will still keep the error set even after the reset done in the + * irq_preinstall and irq_uninstall hooks. + * And enabling in this situation cause the screen to freeze in the + * first time that PSR HW tries to activate so lets keep PSR disabled + * to avoid any rendering problems. + */ + val = I915_READ(EDP_PSR_IIR); + val &= EDP_PSR_ERROR(edp_psr_shift(TRANSCODER_EDP)); + if (val) { + DRM_DEBUG_KMS("PSR interruption error set\n"); + dev_priv->psr.sink_not_reliable = true; + return; + } + /* Set link_standby x link_off defaults */ if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) /* HSW and BDW require workarounds that we don't implement. */ @@ -1123,6 +1175,7 @@ void intel_psr_short_pulse(struct intel_dp *intel_dp) if ((val & DP_PSR_SINK_STATE_MASK) == DP_PSR_SINK_INTERNAL_ERROR) { DRM_DEBUG_KMS("PSR sink internal error, disabling PSR\n"); intel_psr_disable_locked(intel_dp); + psr->sink_not_reliable = true; } if (drm_dp_dpcd_readb(&intel_dp->aux, DP_PSR_ERROR_STATUS, &val) != 1) { @@ -1140,10 +1193,27 @@ void intel_psr_short_pulse(struct intel_dp *intel_dp) if (val & ~errors) DRM_ERROR("PSR_ERROR_STATUS unhandled errors %x\n", val & ~errors); - if (val & errors) + if (val & errors) { intel_psr_disable_locked(intel_dp); + psr->sink_not_reliable = true; + } /* clear status register */ drm_dp_dpcd_writeb(&intel_dp->aux, DP_PSR_ERROR_STATUS, val); exit: mutex_unlock(&psr->lock); } + +bool intel_psr_enabled(struct intel_dp *intel_dp) +{ + struct drm_i915_private *dev_priv = dp_to_i915(intel_dp); + bool ret; + + if (!CAN_PSR(dev_priv) || !intel_dp_is_edp(intel_dp)) + return false; + + mutex_lock(&dev_priv->psr.lock); + ret = (dev_priv->psr.dp == intel_dp && dev_priv->psr.enabled); + mutex_unlock(&dev_priv->psr.lock); + + return ret; +} diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index 87eebc13c0d8..fbeaec3994e7 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -69,19 +69,28 @@ unsigned int intel_ring_update_space(struct intel_ring *ring) static int gen2_render_ring_flush(struct i915_request *rq, u32 mode) { + unsigned int num_store_dw; u32 cmd, *cs; cmd = MI_FLUSH; - + num_store_dw = 0; if (mode & EMIT_INVALIDATE) cmd |= MI_READ_FLUSH; + if (mode & EMIT_FLUSH) + num_store_dw = 4; - cs = intel_ring_begin(rq, 2); + cs = intel_ring_begin(rq, 2 + 3 * num_store_dw); if (IS_ERR(cs)) return PTR_ERR(cs); *cs++ = cmd; - *cs++ = MI_NOOP; + while (num_store_dw--) { + *cs++ = MI_STORE_DWORD_IMM | MI_MEM_VIRTUAL; + *cs++ = i915_scratch_offset(rq->i915); + *cs++ = 0; + } + *cs++ = MI_FLUSH | MI_NO_WRITE_FLUSH; + intel_ring_advance(rq, cs); return 0; @@ -150,8 +159,7 @@ gen4_render_ring_flush(struct i915_request *rq, u32 mode) */ if (mode & EMIT_INVALIDATE) { *cs++ = GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE; - *cs++ = i915_ggtt_offset(rq->engine->scratch) | - PIPE_CONTROL_GLOBAL_GTT; + *cs++ = i915_scratch_offset(rq->i915) | PIPE_CONTROL_GLOBAL_GTT; *cs++ = 0; *cs++ = 0; @@ -159,8 +167,7 @@ gen4_render_ring_flush(struct i915_request *rq, u32 mode) *cs++ = MI_FLUSH; *cs++ = GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE; - *cs++ = i915_ggtt_offset(rq->engine->scratch) | - PIPE_CONTROL_GLOBAL_GTT; + *cs++ = i915_scratch_offset(rq->i915) | PIPE_CONTROL_GLOBAL_GTT; *cs++ = 0; *cs++ = 0; } @@ -212,8 +219,7 @@ gen4_render_ring_flush(struct i915_request *rq, u32 mode) static int intel_emit_post_sync_nonzero_flush(struct i915_request *rq) { - u32 scratch_addr = - i915_ggtt_offset(rq->engine->scratch) + 2 * CACHELINE_BYTES; + u32 scratch_addr = i915_scratch_offset(rq->i915) + 2 * CACHELINE_BYTES; u32 *cs; cs = intel_ring_begin(rq, 6); @@ -246,8 +252,7 @@ intel_emit_post_sync_nonzero_flush(struct i915_request *rq) static int gen6_render_ring_flush(struct i915_request *rq, u32 mode) { - u32 scratch_addr = - i915_ggtt_offset(rq->engine->scratch) + 2 * CACHELINE_BYTES; + u32 scratch_addr = i915_scratch_offset(rq->i915) + 2 * CACHELINE_BYTES; u32 *cs, flags = 0; int ret; @@ -316,8 +321,7 @@ gen7_render_ring_cs_stall_wa(struct i915_request *rq) static int gen7_render_ring_flush(struct i915_request *rq, u32 mode) { - u32 scratch_addr = - i915_ggtt_offset(rq->engine->scratch) + 2 * CACHELINE_BYTES; + u32 scratch_addr = i915_scratch_offset(rq->i915) + 2 * CACHELINE_BYTES; u32 *cs, flags = 0; /* @@ -529,6 +533,13 @@ static int init_ring_common(struct intel_engine_cs *engine) intel_engine_reset_breadcrumbs(engine); + if (HAS_LEGACY_SEMAPHORES(engine->i915)) { + I915_WRITE(RING_SYNC_0(engine->mmio_base), 0); + I915_WRITE(RING_SYNC_1(engine->mmio_base), 0); + if (HAS_VEBOX(dev_priv)) + I915_WRITE(RING_SYNC_2(engine->mmio_base), 0); + } + /* Enforce ordering by reading HEAD register back */ I915_READ_HEAD(engine); @@ -546,10 +557,11 @@ static int init_ring_common(struct intel_engine_cs *engine) /* Check that the ring offsets point within the ring! */ GEM_BUG_ON(!intel_ring_offset_valid(ring, ring->head)); GEM_BUG_ON(!intel_ring_offset_valid(ring, ring->tail)); - intel_ring_update_space(ring); + + /* First wake the ring up to an empty/idle ring */ I915_WRITE_HEAD(engine, ring->head); - I915_WRITE_TAIL(engine, ring->tail); + I915_WRITE_TAIL(engine, ring->head); (void)I915_READ_TAIL(engine); I915_WRITE_CTL(engine, RING_CTL_SIZE(ring->size) | RING_VALID); @@ -574,6 +586,12 @@ static int init_ring_common(struct intel_engine_cs *engine) if (INTEL_GEN(dev_priv) > 2) I915_WRITE_MODE(engine, _MASKED_BIT_DISABLE(STOP_RING)); + /* Now awake, let it get started */ + if (ring->tail != ring->head) { + I915_WRITE_TAIL(engine, ring->tail); + (void)I915_READ_TAIL(engine); + } + /* Papering over lost _interrupts_ immediately following the restart */ intel_engine_wakeup(engine); out: @@ -642,7 +660,7 @@ static int intel_rcs_ctx_init(struct i915_request *rq) { int ret; - ret = intel_ctx_workarounds_emit(rq); + ret = intel_engine_emit_ctx_wa(rq); if (ret != 0) return ret; @@ -660,8 +678,6 @@ static int init_render_ring(struct intel_engine_cs *engine) if (ret) return ret; - intel_whitelist_workarounds_apply(engine); - /* WaTimedSingleVertexDispatch:cl,bw,ctg,elk,ilk,snb */ if (IS_GEN(dev_priv, 4, 6)) I915_WRITE(MI_MODE, _MASKED_BIT_ENABLE(VS_TIMER_DISPATCH)); @@ -743,9 +759,18 @@ static void cancel_requests(struct intel_engine_cs *engine) /* Mark all submitted requests as skipped. */ list_for_each_entry(request, &engine->timeline.requests, link) { GEM_BUG_ON(!request->global_seqno); - if (!i915_request_completed(request)) - dma_fence_set_error(&request->fence, -EIO); + + if (test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, + &request->fence.flags)) + continue; + + dma_fence_set_error(&request->fence, -EIO); } + + intel_write_status_page(engine, + I915_GEM_HWS_INDEX, + intel_engine_last_submit(engine)); + /* Remaining _unready_ requests will be nop'ed when submitted */ spin_unlock_irqrestore(&engine->timeline.lock, flags); @@ -973,7 +998,7 @@ i965_emit_bb_start(struct i915_request *rq, } /* Just userspace ABI convention to limit the wa batch bo to a resonable size */ -#define I830_BATCH_LIMIT (256*1024) +#define I830_BATCH_LIMIT SZ_256K #define I830_TLB_ENTRIES (2) #define I830_WA_SIZE max(I830_TLB_ENTRIES*4096, I830_BATCH_LIMIT) static int @@ -981,7 +1006,9 @@ i830_emit_bb_start(struct i915_request *rq, u64 offset, u32 len, unsigned int dispatch_flags) { - u32 *cs, cs_offset = i915_ggtt_offset(rq->engine->scratch); + u32 *cs, cs_offset = i915_scratch_offset(rq->i915); + + GEM_BUG_ON(rq->i915->gt.scratch->size < I830_WA_SIZE); cs = intel_ring_begin(rq, 6); if (IS_ERR(cs)) @@ -1438,7 +1465,6 @@ static int intel_init_ring_buffer(struct intel_engine_cs *engine) { struct i915_timeline *timeline; struct intel_ring *ring; - unsigned int size; int err; intel_engine_setup_common(engine); @@ -1463,21 +1489,12 @@ static int intel_init_ring_buffer(struct intel_engine_cs *engine) GEM_BUG_ON(engine->buffer); engine->buffer = ring; - size = PAGE_SIZE; - if (HAS_BROKEN_CS_TLB(engine->i915)) - size = I830_WA_SIZE; - err = intel_engine_create_scratch(engine, size); - if (err) - goto err_unpin; - err = intel_engine_init_common(engine); if (err) - goto err_scratch; + goto err_unpin; return 0; -err_scratch: - intel_engine_cleanup_scratch(engine); err_unpin: intel_ring_unpin(ring); err_ring: @@ -1551,7 +1568,7 @@ static int flush_pd_dir(struct i915_request *rq) /* Stall until the page table load is complete */ *cs++ = MI_STORE_REGISTER_MEM | MI_SRM_LRM_GLOBAL_GTT; *cs++ = i915_mmio_reg_offset(RING_PP_DIR_BASE(engine)); - *cs++ = i915_ggtt_offset(engine->scratch); + *cs++ = i915_scratch_offset(rq->i915); *cs++ = MI_NOOP; intel_ring_advance(rq, cs); @@ -1660,7 +1677,7 @@ static inline int mi_set_context(struct i915_request *rq, u32 flags) /* Insert a delay before the next switch! */ *cs++ = MI_STORE_REGISTER_MEM | MI_SRM_LRM_GLOBAL_GTT; *cs++ = i915_mmio_reg_offset(last_reg); - *cs++ = i915_ggtt_offset(engine->scratch); + *cs++ = i915_scratch_offset(rq->i915); *cs++ = MI_NOOP; } *cs++ = MI_ARB_ON_OFF | MI_ARB_ENABLE; diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index 8a2270b209b0..72edaa7ff411 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -15,6 +15,7 @@ #include "i915_selftest.h" #include "i915_timeline.h" #include "intel_gpu_commands.h" +#include "intel_workarounds.h" struct drm_printer; struct i915_sched_attr; @@ -313,13 +314,6 @@ struct intel_engine_execlists { struct rb_root_cached queue; /** - * @csb_read: control register for Context Switch buffer - * - * Note this register is always in mmio. - */ - u32 __iomem *csb_read; - - /** * @csb_write: control register for Context Switch buffer * * Note this register may be either mmio or HWSP shadow. @@ -339,15 +333,6 @@ struct intel_engine_execlists { u32 preempt_complete_status; /** - * @csb_write_reset: reset value for CSB write pointer - * - * As the CSB write pointer maybe either in HWSP or as a field - * inside an mmio register, we want to reprogram it slightly - * differently to avoid later confusion. - */ - u32 csb_write_reset; - - /** * @csb_head: context status buffer head */ u8 csb_head; @@ -451,7 +436,9 @@ struct intel_engine_cs { struct intel_hw_status_page status_page; struct i915_ctx_workarounds wa_ctx; - struct i915_vma *scratch; + struct i915_wa_list ctx_wa_list; + struct i915_wa_list wa_list; + struct i915_wa_list whitelist; u32 irq_keep_mask; /* always keep these interrupts */ u32 irq_enable_mask; /* bitmask to enable ring interrupt */ @@ -908,10 +895,6 @@ void intel_engine_setup_common(struct intel_engine_cs *engine); int intel_engine_init_common(struct intel_engine_cs *engine); void intel_engine_cleanup_common(struct intel_engine_cs *engine); -int intel_engine_create_scratch(struct intel_engine_cs *engine, - unsigned int size); -void intel_engine_cleanup_scratch(struct intel_engine_cs *engine); - int intel_init_render_ring_buffer(struct intel_engine_cs *engine); int intel_init_bsd_ring_buffer(struct intel_engine_cs *engine); int intel_init_blt_ring_buffer(struct intel_engine_cs *engine); diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c index 1c2de9b69a19..4350a5270423 100644 --- a/drivers/gpu/drm/i915/intel_runtime_pm.c +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c @@ -76,6 +76,8 @@ intel_display_power_domain_str(enum intel_display_power_domain domain) return "TRANSCODER_C"; case POWER_DOMAIN_TRANSCODER_EDP: return "TRANSCODER_EDP"; + case POWER_DOMAIN_TRANSCODER_EDP_VDSC: + return "TRANSCODER_EDP_VDSC"; case POWER_DOMAIN_TRANSCODER_DSI_A: return "TRANSCODER_DSI_A"; case POWER_DOMAIN_TRANSCODER_DSI_C: @@ -2028,9 +2030,9 @@ void intel_display_power_put(struct drm_i915_private *dev_priv, */ #define ICL_PW_2_POWER_DOMAINS ( \ ICL_PW_3_POWER_DOMAINS | \ + BIT_ULL(POWER_DOMAIN_TRANSCODER_EDP_VDSC) | \ BIT_ULL(POWER_DOMAIN_INIT)) /* - * - eDP/DSI VDSC * - KVMR (HW control) */ #define ICL_DISPLAY_DC_OFF_POWER_DOMAINS ( \ diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c index abe193815ccc..d2e003d8f3db 100644 --- a/drivers/gpu/drm/i915/intel_sprite.c +++ b/drivers/gpu/drm/i915/intel_sprite.c @@ -373,14 +373,12 @@ skl_program_scaler(struct intel_plane *plane, #define BOFF(x) (((x) & 0xffff) << 16) static void -icl_program_input_csc_coeff(const struct intel_crtc_state *crtc_state, - const struct intel_plane_state *plane_state) +icl_program_input_csc(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state, + const struct intel_plane_state *plane_state) { - struct drm_i915_private *dev_priv = - to_i915(plane_state->base.plane->dev); - struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc); - enum pipe pipe = crtc->pipe; - struct intel_plane *plane = to_intel_plane(plane_state->base.plane); + struct drm_i915_private *dev_priv = to_i915(plane->base.dev); + enum pipe pipe = plane->pipe; enum plane_id plane_id = plane->id; static const u16 input_csc_matrix[][9] = { @@ -508,28 +506,12 @@ skl_program_plane(struct intel_plane *plane, spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); - if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) - I915_WRITE_FW(PLANE_COLOR_CTL(pipe, plane_id), - plane_state->color_ctl); - - if (fb->format->is_yuv && icl_is_hdr_plane(plane)) - icl_program_input_csc_coeff(crtc_state, plane_state); - - I915_WRITE_FW(PLANE_KEYVAL(pipe, plane_id), key->min_value); - I915_WRITE_FW(PLANE_KEYMAX(pipe, plane_id), keymax); - I915_WRITE_FW(PLANE_KEYMSK(pipe, plane_id), keymsk); - - I915_WRITE_FW(PLANE_OFFSET(pipe, plane_id), (y << 16) | x); I915_WRITE_FW(PLANE_STRIDE(pipe, plane_id), stride); + I915_WRITE_FW(PLANE_POS(pipe, plane_id), (crtc_y << 16) | crtc_x); I915_WRITE_FW(PLANE_SIZE(pipe, plane_id), (src_h << 16) | src_w); I915_WRITE_FW(PLANE_AUX_DIST(pipe, plane_id), (plane_state->color_plane[1].offset - surf_addr) | aux_stride); - if (INTEL_GEN(dev_priv) < 11) - I915_WRITE_FW(PLANE_AUX_OFFSET(pipe, plane_id), - (plane_state->color_plane[1].y << 16) | - plane_state->color_plane[1].x); - if (icl_is_hdr_plane(plane)) { u32 cus_ctl = 0; @@ -551,15 +533,38 @@ skl_program_plane(struct intel_plane *plane, I915_WRITE_FW(PLANE_CUS_CTL(pipe, plane_id), cus_ctl); } - if (!slave && plane_state->scaler_id >= 0) - skl_program_scaler(plane, crtc_state, plane_state); + if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) + I915_WRITE_FW(PLANE_COLOR_CTL(pipe, plane_id), + plane_state->color_ctl); - I915_WRITE_FW(PLANE_POS(pipe, plane_id), (crtc_y << 16) | crtc_x); + if (fb->format->is_yuv && icl_is_hdr_plane(plane)) + icl_program_input_csc(plane, crtc_state, plane_state); + + skl_write_plane_wm(plane, crtc_state); + I915_WRITE_FW(PLANE_KEYVAL(pipe, plane_id), key->min_value); + I915_WRITE_FW(PLANE_KEYMSK(pipe, plane_id), keymsk); + I915_WRITE_FW(PLANE_KEYMAX(pipe, plane_id), keymax); + + I915_WRITE_FW(PLANE_OFFSET(pipe, plane_id), (y << 16) | x); + + if (INTEL_GEN(dev_priv) < 11) + I915_WRITE_FW(PLANE_AUX_OFFSET(pipe, plane_id), + (plane_state->color_plane[1].y << 16) | + plane_state->color_plane[1].x); + + /* + * The control register self-arms if the plane was previously + * disabled. Try to make the plane enable atomic by writing + * the control register just before the surface register. + */ I915_WRITE_FW(PLANE_CTL(pipe, plane_id), plane_ctl); I915_WRITE_FW(PLANE_SURF(pipe, plane_id), intel_plane_ggtt_offset(plane_state) + surf_addr); + if (!slave && plane_state->scaler_id >= 0) + skl_program_scaler(plane, crtc_state, plane_state); + spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags); } @@ -589,7 +594,8 @@ icl_update_slave(struct intel_plane *plane, } static void -skl_disable_plane(struct intel_plane *plane, struct intel_crtc *crtc) +skl_disable_plane(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state) { struct drm_i915_private *dev_priv = to_i915(plane->base.dev); enum plane_id plane_id = plane->id; @@ -598,6 +604,8 @@ skl_disable_plane(struct intel_plane *plane, struct intel_crtc *crtc) spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); + skl_write_plane_wm(plane, crtc_state); + I915_WRITE_FW(PLANE_CTL(pipe, plane_id), 0); I915_WRITE_FW(PLANE_SURF(pipe, plane_id), 0); @@ -819,35 +827,41 @@ vlv_update_plane(struct intel_plane *plane, spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); - vlv_update_clrc(plane_state); + I915_WRITE_FW(SPSTRIDE(pipe, plane_id), + plane_state->color_plane[0].stride); + I915_WRITE_FW(SPPOS(pipe, plane_id), (crtc_y << 16) | crtc_x); + I915_WRITE_FW(SPSIZE(pipe, plane_id), (crtc_h << 16) | crtc_w); + I915_WRITE_FW(SPCONSTALPHA(pipe, plane_id), 0); if (IS_CHERRYVIEW(dev_priv) && pipe == PIPE_B) chv_update_csc(plane_state); if (key->flags) { I915_WRITE_FW(SPKEYMINVAL(pipe, plane_id), key->min_value); - I915_WRITE_FW(SPKEYMAXVAL(pipe, plane_id), key->max_value); I915_WRITE_FW(SPKEYMSK(pipe, plane_id), key->channel_mask); + I915_WRITE_FW(SPKEYMAXVAL(pipe, plane_id), key->max_value); } - I915_WRITE_FW(SPSTRIDE(pipe, plane_id), - plane_state->color_plane[0].stride); - I915_WRITE_FW(SPPOS(pipe, plane_id), (crtc_y << 16) | crtc_x); - I915_WRITE_FW(SPTILEOFF(pipe, plane_id), (y << 16) | x); I915_WRITE_FW(SPLINOFF(pipe, plane_id), linear_offset); + I915_WRITE_FW(SPTILEOFF(pipe, plane_id), (y << 16) | x); - I915_WRITE_FW(SPCONSTALPHA(pipe, plane_id), 0); - - I915_WRITE_FW(SPSIZE(pipe, plane_id), (crtc_h << 16) | crtc_w); + /* + * The control register self-arms if the plane was previously + * disabled. Try to make the plane enable atomic by writing + * the control register just before the surface register. + */ I915_WRITE_FW(SPCNTR(pipe, plane_id), sprctl); I915_WRITE_FW(SPSURF(pipe, plane_id), intel_plane_ggtt_offset(plane_state) + sprsurf_offset); + vlv_update_clrc(plane_state); + spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags); } static void -vlv_disable_plane(struct intel_plane *plane, struct intel_crtc *crtc) +vlv_disable_plane(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state) { struct drm_i915_private *dev_priv = to_i915(plane->base.dev); enum pipe pipe = plane->pipe; @@ -980,27 +994,32 @@ ivb_update_plane(struct intel_plane *plane, spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); + I915_WRITE_FW(SPRSTRIDE(pipe), plane_state->color_plane[0].stride); + I915_WRITE_FW(SPRPOS(pipe), (crtc_y << 16) | crtc_x); + I915_WRITE_FW(SPRSIZE(pipe), (crtc_h << 16) | crtc_w); + if (IS_IVYBRIDGE(dev_priv)) + I915_WRITE_FW(SPRSCALE(pipe), sprscale); + if (key->flags) { I915_WRITE_FW(SPRKEYVAL(pipe), key->min_value); - I915_WRITE_FW(SPRKEYMAX(pipe), key->max_value); I915_WRITE_FW(SPRKEYMSK(pipe), key->channel_mask); + I915_WRITE_FW(SPRKEYMAX(pipe), key->max_value); } - I915_WRITE_FW(SPRSTRIDE(pipe), plane_state->color_plane[0].stride); - I915_WRITE_FW(SPRPOS(pipe), (crtc_y << 16) | crtc_x); - /* HSW consolidates SPRTILEOFF and SPRLINOFF into a single SPROFFSET * register */ if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) { I915_WRITE_FW(SPROFFSET(pipe), (y << 16) | x); } else { - I915_WRITE_FW(SPRTILEOFF(pipe), (y << 16) | x); I915_WRITE_FW(SPRLINOFF(pipe), linear_offset); + I915_WRITE_FW(SPRTILEOFF(pipe), (y << 16) | x); } - I915_WRITE_FW(SPRSIZE(pipe), (crtc_h << 16) | crtc_w); - if (IS_IVYBRIDGE(dev_priv)) - I915_WRITE_FW(SPRSCALE(pipe), sprscale); + /* + * The control register self-arms if the plane was previously + * disabled. Try to make the plane enable atomic by writing + * the control register just before the surface register. + */ I915_WRITE_FW(SPRCTL(pipe), sprctl); I915_WRITE_FW(SPRSURF(pipe), intel_plane_ggtt_offset(plane_state) + sprsurf_offset); @@ -1009,7 +1028,8 @@ ivb_update_plane(struct intel_plane *plane, } static void -ivb_disable_plane(struct intel_plane *plane, struct intel_crtc *crtc) +ivb_disable_plane(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state) { struct drm_i915_private *dev_priv = to_i915(plane->base.dev); enum pipe pipe = plane->pipe; @@ -1018,7 +1038,7 @@ ivb_disable_plane(struct intel_plane *plane, struct intel_crtc *crtc) spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); I915_WRITE_FW(SPRCTL(pipe), 0); - /* Can't leave the scaler enabled... */ + /* Disable the scaler */ if (IS_IVYBRIDGE(dev_priv)) I915_WRITE_FW(SPRSCALE(pipe), 0); I915_WRITE_FW(SPRSURF(pipe), 0); @@ -1148,20 +1168,25 @@ g4x_update_plane(struct intel_plane *plane, spin_lock_irqsave(&dev_priv->uncore.lock, irqflags); + I915_WRITE_FW(DVSSTRIDE(pipe), plane_state->color_plane[0].stride); + I915_WRITE_FW(DVSPOS(pipe), (crtc_y << 16) | crtc_x); + I915_WRITE_FW(DVSSIZE(pipe), (crtc_h << 16) | crtc_w); + I915_WRITE_FW(DVSSCALE(pipe), dvsscale); + if (key->flags) { I915_WRITE_FW(DVSKEYVAL(pipe), key->min_value); - I915_WRITE_FW(DVSKEYMAX(pipe), key->max_value); I915_WRITE_FW(DVSKEYMSK(pipe), key->channel_mask); + I915_WRITE_FW(DVSKEYMAX(pipe), key->max_value); } - I915_WRITE_FW(DVSSTRIDE(pipe), plane_state->color_plane[0].stride); - I915_WRITE_FW(DVSPOS(pipe), (crtc_y << 16) | crtc_x); - - I915_WRITE_FW(DVSTILEOFF(pipe), (y << 16) | x); I915_WRITE_FW(DVSLINOFF(pipe), linear_offset); + I915_WRITE_FW(DVSTILEOFF(pipe), (y << 16) | x); - I915_WRITE_FW(DVSSIZE(pipe), (crtc_h << 16) | crtc_w); - I915_WRITE_FW(DVSSCALE(pipe), dvsscale); + /* + * The control register self-arms if the plane was previously + * disabled. Try to make the plane enable atomic by writing + * the control register just before the surface register. + */ I915_WRITE_FW(DVSCNTR(pipe), dvscntr); I915_WRITE_FW(DVSSURF(pipe), intel_plane_ggtt_offset(plane_state) + dvssurf_offset); @@ -1170,7 +1195,8 @@ g4x_update_plane(struct intel_plane *plane, } static void -g4x_disable_plane(struct intel_plane *plane, struct intel_crtc *crtc) +g4x_disable_plane(struct intel_plane *plane, + const struct intel_crtc_state *crtc_state) { struct drm_i915_private *dev_priv = to_i915(plane->base.dev); enum pipe pipe = plane->pipe; diff --git a/drivers/gpu/drm/i915/intel_vdsc.c b/drivers/gpu/drm/i915/intel_vdsc.c new file mode 100644 index 000000000000..c56ba0e04044 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_vdsc.c @@ -0,0 +1,1088 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2018 Intel Corporation + * + * Author: Gaurav K Singh <gaurav.k.singh@intel.com> + * Manasi Navare <manasi.d.navare@intel.com> + */ + +#include <drm/drmP.h> +#include <drm/i915_drm.h> +#include "i915_drv.h" +#include "intel_drv.h" + +enum ROW_INDEX_BPP { + ROW_INDEX_6BPP = 0, + ROW_INDEX_8BPP, + ROW_INDEX_10BPP, + ROW_INDEX_12BPP, + ROW_INDEX_15BPP, + MAX_ROW_INDEX +}; + +enum COLUMN_INDEX_BPC { + COLUMN_INDEX_8BPC = 0, + COLUMN_INDEX_10BPC, + COLUMN_INDEX_12BPC, + COLUMN_INDEX_14BPC, + COLUMN_INDEX_16BPC, + MAX_COLUMN_INDEX +}; + +#define DSC_SUPPORTED_VERSION_MIN 1 + +/* From DSC_v1.11 spec, rc_parameter_Set syntax element typically constant */ +static u16 rc_buf_thresh[] = { + 896, 1792, 2688, 3584, 4480, 5376, 6272, 6720, 7168, 7616, + 7744, 7872, 8000, 8064 +}; + +struct rc_parameters { + u16 initial_xmit_delay; + u8 first_line_bpg_offset; + u16 initial_offset; + u8 flatness_min_qp; + u8 flatness_max_qp; + u8 rc_quant_incr_limit0; + u8 rc_quant_incr_limit1; + struct drm_dsc_rc_range_parameters rc_range_params[DSC_NUM_BUF_RANGES]; +}; + +/* + * Selected Rate Control Related Parameter Recommended Values + * from DSC_v1.11 spec & C Model release: DSC_model_20161212 + */ +static struct rc_parameters rc_params[][MAX_COLUMN_INDEX] = { +{ + /* 6BPP/8BPC */ + { 768, 15, 6144, 3, 13, 11, 11, { + { 0, 4, 0 }, { 1, 6, -2 }, { 3, 8, -2 }, { 4, 8, -4 }, + { 5, 9, -6 }, { 5, 9, -6 }, { 6, 9, -6 }, { 6, 10, -8 }, + { 7, 11, -8 }, { 8, 12, -10 }, { 9, 12, -10 }, { 10, 12, -12 }, + { 10, 12, -12 }, { 11, 12, -12 }, { 13, 14, -12 } + } + }, + /* 6BPP/10BPC */ + { 768, 15, 6144, 7, 17, 15, 15, { + { 0, 8, 0 }, { 3, 10, -2 }, { 7, 12, -2 }, { 8, 12, -4 }, + { 9, 13, -6 }, { 9, 13, -6 }, { 10, 13, -6 }, { 10, 14, -8 }, + { 11, 15, -8 }, { 12, 16, -10 }, { 13, 16, -10 }, + { 14, 16, -12 }, { 14, 16, -12 }, { 15, 16, -12 }, + { 17, 18, -12 } + } + }, + /* 6BPP/12BPC */ + { 768, 15, 6144, 11, 21, 19, 19, { + { 0, 12, 0 }, { 5, 14, -2 }, { 11, 16, -2 }, { 12, 16, -4 }, + { 13, 17, -6 }, { 13, 17, -6 }, { 14, 17, -6 }, { 14, 18, -8 }, + { 15, 19, -8 }, { 16, 20, -10 }, { 17, 20, -10 }, + { 18, 20, -12 }, { 18, 20, -12 }, { 19, 20, -12 }, + { 21, 22, -12 } + } + }, + /* 6BPP/14BPC */ + { 768, 15, 6144, 15, 25, 23, 27, { + { 0, 16, 0 }, { 7, 18, -2 }, { 15, 20, -2 }, { 16, 20, -4 }, + { 17, 21, -6 }, { 17, 21, -6 }, { 18, 21, -6 }, { 18, 22, -8 }, + { 19, 23, -8 }, { 20, 24, -10 }, { 21, 24, -10 }, + { 22, 24, -12 }, { 22, 24, -12 }, { 23, 24, -12 }, + { 25, 26, -12 } + } + }, + /* 6BPP/16BPC */ + { 768, 15, 6144, 19, 29, 27, 27, { + { 0, 20, 0 }, { 9, 22, -2 }, { 19, 24, -2 }, { 20, 24, -4 }, + { 21, 25, -6 }, { 21, 25, -6 }, { 22, 25, -6 }, { 22, 26, -8 }, + { 23, 27, -8 }, { 24, 28, -10 }, { 25, 28, -10 }, + { 26, 28, -12 }, { 26, 28, -12 }, { 27, 28, -12 }, + { 29, 30, -12 } + } + }, +}, +{ + /* 8BPP/8BPC */ + { 512, 12, 6144, 3, 12, 11, 11, { + { 0, 4, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 }, + { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, + { 3, 9, -8 }, { 3, 10, -10 }, { 5, 11, -10 }, { 5, 12, -12 }, + { 5, 13, -12 }, { 7, 13, -12 }, { 13, 15, -12 } + } + }, + /* 8BPP/10BPC */ + { 512, 12, 6144, 7, 16, 15, 15, { + { 0, 4, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 5, 10, -2 }, + { 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, + { 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 }, + { 9, 17, -12 }, { 11, 17, -12 }, { 17, 19, -12 } + } + }, + /* 8BPP/12BPC */ + { 512, 12, 6144, 11, 20, 19, 19, { + { 0, 12, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 9, 14, -2 }, + { 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, + { 11, 17, -8 }, { 11, 18, -10 }, { 13, 19, -10 }, + { 13, 20, -12 }, { 13, 21, -12 }, { 15, 21, -12 }, + { 21, 23, -12 } + } + }, + /* 8BPP/14BPC */ + { 512, 12, 6144, 15, 24, 23, 23, { + { 0, 12, 0 }, { 5, 13, 0 }, { 11, 15, 0 }, { 12, 17, -2 }, + { 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 }, + { 15, 21, -8 }, { 15, 22, -10 }, { 17, 22, -10 }, + { 17, 23, -12 }, { 17, 23, -12 }, { 21, 24, -12 }, + { 24, 25, -12 } + } + }, + /* 8BPP/16BPC */ + { 512, 12, 6144, 19, 28, 27, 27, { + { 0, 12, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 15, 20, -2 }, + { 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 }, + { 19, 25, -8 }, { 19, 26, -10 }, { 21, 26, -10 }, + { 21, 27, -12 }, { 21, 27, -12 }, { 25, 28, -12 }, + { 28, 29, -12 } + } + }, +}, +{ + /* 10BPP/8BPC */ + { 410, 15, 5632, 3, 12, 11, 11, { + { 0, 3, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 2, 6, -2 }, + { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, + { 3, 9, -8 }, { 3, 9, -10 }, { 5, 10, -10 }, { 5, 10, -10 }, + { 5, 11, -12 }, { 7, 11, -12 }, { 11, 12, -12 } + } + }, + /* 10BPP/10BPC */ + { 410, 15, 5632, 7, 16, 15, 15, { + { 0, 7, 2 }, { 4, 8, 0 }, { 5, 9, 0 }, { 6, 10, -2 }, + { 7, 11, -4 }, { 7, 11, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, + { 7, 13, -8 }, { 7, 13, -10 }, { 9, 14, -10 }, { 9, 14, -10 }, + { 9, 15, -12 }, { 11, 15, -12 }, { 15, 16, -12 } + } + }, + /* 10BPP/12BPC */ + { 410, 15, 5632, 11, 20, 19, 19, { + { 0, 11, 2 }, { 4, 12, 0 }, { 9, 13, 0 }, { 10, 14, -2 }, + { 11, 15, -4 }, { 11, 15, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, + { 11, 17, -8 }, { 11, 17, -10 }, { 13, 18, -10 }, + { 13, 18, -10 }, { 13, 19, -12 }, { 15, 19, -12 }, + { 19, 20, -12 } + } + }, + /* 10BPP/14BPC */ + { 410, 15, 5632, 15, 24, 23, 23, { + { 0, 11, 2 }, { 5, 13, 0 }, { 11, 15, 0 }, { 13, 18, -2 }, + { 15, 19, -4 }, { 15, 19, -6 }, { 15, 19, -8 }, { 15, 20, -8 }, + { 15, 21, -8 }, { 15, 21, -10 }, { 17, 22, -10 }, + { 17, 22, -10 }, { 17, 23, -12 }, { 19, 23, -12 }, + { 23, 24, -12 } + } + }, + /* 10BPP/16BPC */ + { 410, 15, 5632, 19, 28, 27, 27, { + { 0, 11, 2 }, { 6, 14, 0 }, { 13, 17, 0 }, { 16, 20, -2 }, + { 19, 23, -4 }, { 19, 23, -6 }, { 19, 23, -8 }, { 19, 24, -8 }, + { 19, 25, -8 }, { 19, 25, -10 }, { 21, 26, -10 }, + { 21, 26, -10 }, { 21, 27, -12 }, { 23, 27, -12 }, + { 27, 28, -12 } + } + }, +}, +{ + /* 12BPP/8BPC */ + { 341, 15, 2048, 3, 12, 11, 11, { + { 0, 2, 2 }, { 0, 4, 0 }, { 1, 5, 0 }, { 1, 6, -2 }, + { 3, 7, -4 }, { 3, 7, -6 }, { 3, 7, -8 }, { 3, 8, -8 }, + { 3, 9, -8 }, { 3, 10, -10 }, { 5, 11, -10 }, + { 5, 12, -12 }, { 5, 13, -12 }, { 7, 13, -12 }, { 13, 15, -12 } + } + }, + /* 12BPP/10BPC */ + { 341, 15, 2048, 7, 16, 15, 15, { + { 0, 2, 2 }, { 2, 5, 0 }, { 3, 7, 0 }, { 4, 8, -2 }, + { 6, 9, -4 }, { 7, 10, -6 }, { 7, 11, -8 }, { 7, 12, -8 }, + { 7, 13, -8 }, { 7, 14, -10 }, { 9, 15, -10 }, { 9, 16, -12 }, + { 9, 17, -12 }, { 11, 17, -12 }, { 17, 19, -12 } + } + }, + /* 12BPP/12BPC */ + { 341, 15, 2048, 11, 20, 19, 19, { + { 0, 6, 2 }, { 4, 9, 0 }, { 7, 11, 0 }, { 8, 12, -2 }, + { 10, 13, -4 }, { 11, 14, -6 }, { 11, 15, -8 }, { 11, 16, -8 }, + { 11, 17, -8 }, { 11, 18, -10 }, { 13, 19, -10 }, + { 13, 20, -12 }, { 13, 21, -12 }, { 15, 21, -12 }, + { 21, 23, -12 } + } + }, + /* 12BPP/14BPC */ + { 341, 15, 2048, 15, 24, 23, 23, { + { 0, 6, 2 }, { 7, 10, 0 }, { 9, 13, 0 }, { 11, 16, -2 }, + { 14, 17, -4 }, { 15, 18, -6 }, { 15, 19, -8 }, { 15, 20, -8 }, + { 15, 20, -8 }, { 15, 21, -10 }, { 17, 21, -10 }, + { 17, 21, -12 }, { 17, 21, -12 }, { 19, 22, -12 }, + { 22, 23, -12 } + } + }, + /* 12BPP/16BPC */ + { 341, 15, 2048, 19, 28, 27, 27, { + { 0, 6, 2 }, { 6, 11, 0 }, { 11, 15, 0 }, { 14, 18, -2 }, + { 18, 21, -4 }, { 19, 22, -6 }, { 19, 23, -8 }, { 19, 24, -8 }, + { 19, 24, -8 }, { 19, 25, -10 }, { 21, 25, -10 }, + { 21, 25, -12 }, { 21, 25, -12 }, { 23, 26, -12 }, + { 26, 27, -12 } + } + }, +}, +{ + /* 15BPP/8BPC */ + { 273, 15, 2048, 3, 12, 11, 11, { + { 0, 0, 10 }, { 0, 1, 8 }, { 0, 1, 6 }, { 0, 2, 4 }, + { 1, 2, 2 }, { 1, 3, 0 }, { 1, 3, -2 }, { 2, 4, -4 }, + { 2, 5, -6 }, { 3, 5, -8 }, { 4, 6, -10 }, { 4, 7, -10 }, + { 5, 7, -12 }, { 7, 8, -12 }, { 8, 9, -12 } + } + }, + /* 15BPP/10BPC */ + { 273, 15, 2048, 7, 16, 15, 15, { + { 0, 2, 10 }, { 2, 5, 8 }, { 3, 5, 6 }, { 4, 6, 4 }, + { 5, 6, 2 }, { 5, 7, 0 }, { 5, 7, -2 }, { 6, 8, -4 }, + { 6, 9, -6 }, { 7, 9, -8 }, { 8, 10, -10 }, { 8, 11, -10 }, + { 9, 11, -12 }, { 11, 12, -12 }, { 12, 13, -12 } + } + }, + /* 15BPP/12BPC */ + { 273, 15, 2048, 11, 20, 19, 19, { + { 0, 4, 10 }, { 2, 7, 8 }, { 4, 9, 6 }, { 6, 11, 4 }, + { 9, 11, 2 }, { 9, 11, 0 }, { 9, 12, -2 }, { 10, 12, -4 }, + { 11, 13, -6 }, { 11, 13, -8 }, { 12, 14, -10 }, + { 13, 15, -10 }, { 13, 15, -12 }, { 15, 16, -12 }, + { 16, 17, -12 } + } + }, + /* 15BPP/14BPC */ + { 273, 15, 2048, 15, 24, 23, 23, { + { 0, 4, 10 }, { 3, 8, 8 }, { 6, 11, 6 }, { 9, 14, 4 }, + { 13, 15, 2 }, { 13, 15, 0 }, { 13, 16, -2 }, { 14, 16, -4 }, + { 15, 17, -6 }, { 15, 17, -8 }, { 16, 18, -10 }, + { 17, 19, -10 }, { 17, 19, -12 }, { 19, 20, -12 }, + { 20, 21, -12 } + } + }, + /* 15BPP/16BPC */ + { 273, 15, 2048, 19, 28, 27, 27, { + { 0, 4, 10 }, { 4, 9, 8 }, { 8, 13, 6 }, { 12, 17, 4 }, + { 17, 19, 2 }, { 17, 20, 0 }, { 17, 20, -2 }, { 18, 20, -4 }, + { 19, 21, -6 }, { 19, 21, -8 }, { 20, 22, -10 }, + { 21, 23, -10 }, { 21, 23, -12 }, { 23, 24, -12 }, + { 24, 25, -12 } + } + } +} + +}; + +static int get_row_index_for_rc_params(u16 compressed_bpp) +{ + switch (compressed_bpp) { + case 6: + return ROW_INDEX_6BPP; + case 8: + return ROW_INDEX_8BPP; + case 10: + return ROW_INDEX_10BPP; + case 12: + return ROW_INDEX_12BPP; + case 15: + return ROW_INDEX_15BPP; + default: + return -EINVAL; + } +} + +static int get_column_index_for_rc_params(u8 bits_per_component) +{ + switch (bits_per_component) { + case 8: + return COLUMN_INDEX_8BPC; + case 10: + return COLUMN_INDEX_10BPC; + case 12: + return COLUMN_INDEX_12BPC; + case 14: + return COLUMN_INDEX_14BPC; + case 16: + return COLUMN_INDEX_16BPC; + default: + return -EINVAL; + } +} + +static int intel_compute_rc_parameters(struct drm_dsc_config *vdsc_cfg) +{ + unsigned long groups_per_line = 0; + unsigned long groups_total = 0; + unsigned long num_extra_mux_bits = 0; + unsigned long slice_bits = 0; + unsigned long hrd_delay = 0; + unsigned long final_scale = 0; + unsigned long rbs_min = 0; + + /* Number of groups used to code each line of a slice */ + groups_per_line = DIV_ROUND_UP(vdsc_cfg->slice_width, + DSC_RC_PIXELS_PER_GROUP); + + /* chunksize in Bytes */ + vdsc_cfg->slice_chunk_size = DIV_ROUND_UP(vdsc_cfg->slice_width * + vdsc_cfg->bits_per_pixel, + (8 * 16)); + + if (vdsc_cfg->convert_rgb) + num_extra_mux_bits = 3 * (vdsc_cfg->mux_word_size + + (4 * vdsc_cfg->bits_per_component + 4) + - 2); + else + num_extra_mux_bits = 3 * vdsc_cfg->mux_word_size + + (4 * vdsc_cfg->bits_per_component + 4) + + 2 * (4 * vdsc_cfg->bits_per_component) - 2; + /* Number of bits in one Slice */ + slice_bits = 8 * vdsc_cfg->slice_chunk_size * vdsc_cfg->slice_height; + + while ((num_extra_mux_bits > 0) && + ((slice_bits - num_extra_mux_bits) % vdsc_cfg->mux_word_size)) + num_extra_mux_bits--; + + if (groups_per_line < vdsc_cfg->initial_scale_value - 8) + vdsc_cfg->initial_scale_value = groups_per_line + 8; + + /* scale_decrement_interval calculation according to DSC spec 1.11 */ + if (vdsc_cfg->initial_scale_value > 8) + vdsc_cfg->scale_decrement_interval = groups_per_line / + (vdsc_cfg->initial_scale_value - 8); + else + vdsc_cfg->scale_decrement_interval = DSC_SCALE_DECREMENT_INTERVAL_MAX; + + vdsc_cfg->final_offset = vdsc_cfg->rc_model_size - + (vdsc_cfg->initial_xmit_delay * + vdsc_cfg->bits_per_pixel + 8) / 16 + num_extra_mux_bits; + + if (vdsc_cfg->final_offset >= vdsc_cfg->rc_model_size) { + DRM_DEBUG_KMS("FinalOfs < RcModelSze for this InitialXmitDelay\n"); + return -ERANGE; + } + + final_scale = (vdsc_cfg->rc_model_size * 8) / + (vdsc_cfg->rc_model_size - vdsc_cfg->final_offset); + if (vdsc_cfg->slice_height > 1) + /* + * NflBpgOffset is 16 bit value with 11 fractional bits + * hence we multiply by 2^11 for preserving the + * fractional part + */ + vdsc_cfg->nfl_bpg_offset = DIV_ROUND_UP((vdsc_cfg->first_line_bpg_offset << 11), + (vdsc_cfg->slice_height - 1)); + else + vdsc_cfg->nfl_bpg_offset = 0; + + /* 2^16 - 1 */ + if (vdsc_cfg->nfl_bpg_offset > 65535) { + DRM_DEBUG_KMS("NflBpgOffset is too large for this slice height\n"); + return -ERANGE; + } + + /* Number of groups used to code the entire slice */ + groups_total = groups_per_line * vdsc_cfg->slice_height; + + /* slice_bpg_offset is 16 bit value with 11 fractional bits */ + vdsc_cfg->slice_bpg_offset = DIV_ROUND_UP(((vdsc_cfg->rc_model_size - + vdsc_cfg->initial_offset + + num_extra_mux_bits) << 11), + groups_total); + + if (final_scale > 9) { + /* + * ScaleIncrementInterval = + * finaloffset/((NflBpgOffset + SliceBpgOffset)*8(finalscale - 1.125)) + * as (NflBpgOffset + SliceBpgOffset) has 11 bit fractional value, + * we need divide by 2^11 from pstDscCfg values + */ + vdsc_cfg->scale_increment_interval = + (vdsc_cfg->final_offset * (1 << 11)) / + ((vdsc_cfg->nfl_bpg_offset + + vdsc_cfg->slice_bpg_offset) * + (final_scale - 9)); + } else { + /* + * If finalScaleValue is less than or equal to 9, a value of 0 should + * be used to disable the scale increment at the end of the slice + */ + vdsc_cfg->scale_increment_interval = 0; + } + + if (vdsc_cfg->scale_increment_interval > 65535) { + DRM_DEBUG_KMS("ScaleIncrementInterval is large for slice height\n"); + return -ERANGE; + } + + /* + * DSC spec mentions that bits_per_pixel specifies the target + * bits/pixel (bpp) rate that is used by the encoder, + * in steps of 1/16 of a bit per pixel + */ + rbs_min = vdsc_cfg->rc_model_size - vdsc_cfg->initial_offset + + DIV_ROUND_UP(vdsc_cfg->initial_xmit_delay * + vdsc_cfg->bits_per_pixel, 16) + + groups_per_line * vdsc_cfg->first_line_bpg_offset; + + hrd_delay = DIV_ROUND_UP((rbs_min * 16), vdsc_cfg->bits_per_pixel); + vdsc_cfg->rc_bits = (hrd_delay * vdsc_cfg->bits_per_pixel) / 16; + vdsc_cfg->initial_dec_delay = hrd_delay - vdsc_cfg->initial_xmit_delay; + + return 0; +} + +int intel_dp_compute_dsc_params(struct intel_dp *intel_dp, + struct intel_crtc_state *pipe_config) +{ + struct drm_dsc_config *vdsc_cfg = &pipe_config->dp_dsc_cfg; + u16 compressed_bpp = pipe_config->dsc_params.compressed_bpp; + u8 i = 0; + int row_index = 0; + int column_index = 0; + u8 line_buf_depth = 0; + + vdsc_cfg->pic_width = pipe_config->base.adjusted_mode.crtc_hdisplay; + vdsc_cfg->pic_height = pipe_config->base.adjusted_mode.crtc_vdisplay; + vdsc_cfg->slice_width = DIV_ROUND_UP(vdsc_cfg->pic_width, + pipe_config->dsc_params.slice_count); + /* + * Slice Height of 8 works for all currently available panels. So start + * with that if pic_height is an integral multiple of 8. + * Eventually add logic to try multiple slice heights. + */ + if (vdsc_cfg->pic_height % 8 == 0) + vdsc_cfg->slice_height = 8; + else if (vdsc_cfg->pic_height % 4 == 0) + vdsc_cfg->slice_height = 4; + else + vdsc_cfg->slice_height = 2; + + /* Values filled from DSC Sink DPCD */ + vdsc_cfg->dsc_version_major = + (intel_dp->dsc_dpcd[DP_DSC_REV - DP_DSC_SUPPORT] & + DP_DSC_MAJOR_MASK) >> DP_DSC_MAJOR_SHIFT; + vdsc_cfg->dsc_version_minor = + min(DSC_SUPPORTED_VERSION_MIN, + (intel_dp->dsc_dpcd[DP_DSC_REV - DP_DSC_SUPPORT] & + DP_DSC_MINOR_MASK) >> DP_DSC_MINOR_SHIFT); + + vdsc_cfg->convert_rgb = intel_dp->dsc_dpcd[DP_DSC_DEC_COLOR_FORMAT_CAP - DP_DSC_SUPPORT] & + DP_DSC_RGB; + + line_buf_depth = drm_dp_dsc_sink_line_buf_depth(intel_dp->dsc_dpcd); + if (!line_buf_depth) { + DRM_DEBUG_KMS("DSC Sink Line Buffer Depth invalid\n"); + return -EINVAL; + } + if (vdsc_cfg->dsc_version_minor == 2) + vdsc_cfg->line_buf_depth = (line_buf_depth == DSC_1_2_MAX_LINEBUF_DEPTH_BITS) ? + DSC_1_2_MAX_LINEBUF_DEPTH_VAL : line_buf_depth; + else + vdsc_cfg->line_buf_depth = (line_buf_depth > DSC_1_1_MAX_LINEBUF_DEPTH_BITS) ? + DSC_1_1_MAX_LINEBUF_DEPTH_BITS : line_buf_depth; + + /* Gen 11 does not support YCbCr */ + vdsc_cfg->enable422 = false; + /* Gen 11 does not support VBR */ + vdsc_cfg->vbr_enable = false; + vdsc_cfg->block_pred_enable = + intel_dp->dsc_dpcd[DP_DSC_BLK_PREDICTION_SUPPORT - DP_DSC_SUPPORT] & + DP_DSC_BLK_PREDICTION_IS_SUPPORTED; + + /* Gen 11 only supports integral values of bpp */ + vdsc_cfg->bits_per_pixel = compressed_bpp << 4; + vdsc_cfg->bits_per_component = pipe_config->pipe_bpp / 3; + + for (i = 0; i < DSC_NUM_BUF_RANGES - 1; i++) { + /* + * six 0s are appended to the lsb of each threshold value + * internally in h/w. + * Only 8 bits are allowed for programming RcBufThreshold + */ + vdsc_cfg->rc_buf_thresh[i] = rc_buf_thresh[i] >> 6; + } + + /* + * For 6bpp, RC Buffer threshold 12 and 13 need a different value + * as per C Model + */ + if (compressed_bpp == 6) { + vdsc_cfg->rc_buf_thresh[12] = 0x7C; + vdsc_cfg->rc_buf_thresh[13] = 0x7D; + } + + row_index = get_row_index_for_rc_params(compressed_bpp); + column_index = + get_column_index_for_rc_params(vdsc_cfg->bits_per_component); + + if (row_index < 0 || column_index < 0) + return -EINVAL; + + vdsc_cfg->first_line_bpg_offset = + rc_params[row_index][column_index].first_line_bpg_offset; + vdsc_cfg->initial_xmit_delay = + rc_params[row_index][column_index].initial_xmit_delay; + vdsc_cfg->initial_offset = + rc_params[row_index][column_index].initial_offset; + vdsc_cfg->flatness_min_qp = + rc_params[row_index][column_index].flatness_min_qp; + vdsc_cfg->flatness_max_qp = + rc_params[row_index][column_index].flatness_max_qp; + vdsc_cfg->rc_quant_incr_limit0 = + rc_params[row_index][column_index].rc_quant_incr_limit0; + vdsc_cfg->rc_quant_incr_limit1 = + rc_params[row_index][column_index].rc_quant_incr_limit1; + + for (i = 0; i < DSC_NUM_BUF_RANGES; i++) { + vdsc_cfg->rc_range_params[i].range_min_qp = + rc_params[row_index][column_index].rc_range_params[i].range_min_qp; + vdsc_cfg->rc_range_params[i].range_max_qp = + rc_params[row_index][column_index].rc_range_params[i].range_max_qp; + /* + * Range BPG Offset uses 2's complement and is only a 6 bits. So + * mask it to get only 6 bits. + */ + vdsc_cfg->rc_range_params[i].range_bpg_offset = + rc_params[row_index][column_index].rc_range_params[i].range_bpg_offset & + DSC_RANGE_BPG_OFFSET_MASK; + } + + /* + * BitsPerComponent value determines mux_word_size: + * When BitsPerComponent is 12bpc, muxWordSize will be equal to 64 bits + * When BitsPerComponent is 8 or 10bpc, muxWordSize will be equal to + * 48 bits + */ + if (vdsc_cfg->bits_per_component == 8 || + vdsc_cfg->bits_per_component == 10) + vdsc_cfg->mux_word_size = DSC_MUX_WORD_SIZE_8_10_BPC; + else if (vdsc_cfg->bits_per_component == 12) + vdsc_cfg->mux_word_size = DSC_MUX_WORD_SIZE_12_BPC; + + /* RC_MODEL_SIZE is a constant across all configurations */ + vdsc_cfg->rc_model_size = DSC_RC_MODEL_SIZE_CONST; + /* InitialScaleValue is a 6 bit value with 3 fractional bits (U3.3) */ + vdsc_cfg->initial_scale_value = (vdsc_cfg->rc_model_size << 3) / + (vdsc_cfg->rc_model_size - vdsc_cfg->initial_offset); + + return intel_compute_rc_parameters(vdsc_cfg); +} + +enum intel_display_power_domain +intel_dsc_power_domain(const struct intel_crtc_state *crtc_state) +{ + enum transcoder cpu_transcoder = crtc_state->cpu_transcoder; + + /* + * On ICL VDSC/joining for eDP transcoder uses a separate power well PW2 + * This requires POWER_DOMAIN_TRANSCODER_EDP_VDSC power domain. + * For any other transcoder, VDSC/joining uses the power well associated + * with the pipe/transcoder in use. Hence another reference on the + * transcoder power domain will suffice. + */ + if (cpu_transcoder == TRANSCODER_EDP) + return POWER_DOMAIN_TRANSCODER_EDP_VDSC; + else + return POWER_DOMAIN_TRANSCODER(cpu_transcoder); +} + +static void intel_configure_pps_for_dsc_encoder(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc); + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + const struct drm_dsc_config *vdsc_cfg = &crtc_state->dp_dsc_cfg; + enum pipe pipe = crtc->pipe; + enum transcoder cpu_transcoder = crtc_state->cpu_transcoder; + u32 pps_val = 0; + u32 rc_buf_thresh_dword[4]; + u32 rc_range_params_dword[8]; + u8 num_vdsc_instances = (crtc_state->dsc_params.dsc_split) ? 2 : 1; + int i = 0; + + /* Populate PICTURE_PARAMETER_SET_0 registers */ + pps_val = DSC_VER_MAJ | vdsc_cfg->dsc_version_minor << + DSC_VER_MIN_SHIFT | + vdsc_cfg->bits_per_component << DSC_BPC_SHIFT | + vdsc_cfg->line_buf_depth << DSC_LINE_BUF_DEPTH_SHIFT; + if (vdsc_cfg->block_pred_enable) + pps_val |= DSC_BLOCK_PREDICTION; + if (vdsc_cfg->convert_rgb) + pps_val |= DSC_COLOR_SPACE_CONVERSION; + if (vdsc_cfg->enable422) + pps_val |= DSC_422_ENABLE; + if (vdsc_cfg->vbr_enable) + pps_val |= DSC_VBR_ENABLE; + DRM_INFO("PPS0 = 0x%08x\n", pps_val); + if (cpu_transcoder == TRANSCODER_EDP) { + I915_WRITE(DSCA_PICTURE_PARAMETER_SET_0, pps_val); + /* + * If 2 VDSC instances are needed, configure PPS for second + * VDSC + */ + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(DSCC_PICTURE_PARAMETER_SET_0, pps_val); + } else { + I915_WRITE(ICL_DSC0_PICTURE_PARAMETER_SET_0(pipe), pps_val); + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(ICL_DSC1_PICTURE_PARAMETER_SET_0(pipe), + pps_val); + } + + /* Populate PICTURE_PARAMETER_SET_1 registers */ + pps_val = 0; + pps_val |= DSC_BPP(vdsc_cfg->bits_per_pixel); + DRM_INFO("PPS1 = 0x%08x\n", pps_val); + if (cpu_transcoder == TRANSCODER_EDP) { + I915_WRITE(DSCA_PICTURE_PARAMETER_SET_1, pps_val); + /* + * If 2 VDSC instances are needed, configure PPS for second + * VDSC + */ + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(DSCC_PICTURE_PARAMETER_SET_1, pps_val); + } else { + I915_WRITE(ICL_DSC0_PICTURE_PARAMETER_SET_1(pipe), pps_val); + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(ICL_DSC1_PICTURE_PARAMETER_SET_1(pipe), + pps_val); + } + + /* Populate PICTURE_PARAMETER_SET_2 registers */ + pps_val = 0; + pps_val |= DSC_PIC_HEIGHT(vdsc_cfg->pic_height) | + DSC_PIC_WIDTH(vdsc_cfg->pic_width / num_vdsc_instances); + DRM_INFO("PPS2 = 0x%08x\n", pps_val); + if (encoder->type == INTEL_OUTPUT_EDP) { + I915_WRITE(DSCA_PICTURE_PARAMETER_SET_2, pps_val); + /* + * If 2 VDSC instances are needed, configure PPS for second + * VDSC + */ + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(DSCC_PICTURE_PARAMETER_SET_2, pps_val); + } else { + I915_WRITE(ICL_DSC0_PICTURE_PARAMETER_SET_2(pipe), pps_val); + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(ICL_DSC1_PICTURE_PARAMETER_SET_2(pipe), + pps_val); + } + + /* Populate PICTURE_PARAMETER_SET_3 registers */ + pps_val = 0; + pps_val |= DSC_SLICE_HEIGHT(vdsc_cfg->slice_height) | + DSC_SLICE_WIDTH(vdsc_cfg->slice_width); + DRM_INFO("PPS3 = 0x%08x\n", pps_val); + if (cpu_transcoder == TRANSCODER_EDP) { + I915_WRITE(DSCA_PICTURE_PARAMETER_SET_3, pps_val); + /* + * If 2 VDSC instances are needed, configure PPS for second + * VDSC + */ + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(DSCC_PICTURE_PARAMETER_SET_3, pps_val); + } else { + I915_WRITE(ICL_DSC0_PICTURE_PARAMETER_SET_3(pipe), pps_val); + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(ICL_DSC1_PICTURE_PARAMETER_SET_3(pipe), + pps_val); + } + + /* Populate PICTURE_PARAMETER_SET_4 registers */ + pps_val = 0; + pps_val |= DSC_INITIAL_XMIT_DELAY(vdsc_cfg->initial_xmit_delay) | + DSC_INITIAL_DEC_DELAY(vdsc_cfg->initial_dec_delay); + DRM_INFO("PPS4 = 0x%08x\n", pps_val); + if (cpu_transcoder == TRANSCODER_EDP) { + I915_WRITE(DSCA_PICTURE_PARAMETER_SET_4, pps_val); + /* + * If 2 VDSC instances are needed, configure PPS for second + * VDSC + */ + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(DSCC_PICTURE_PARAMETER_SET_4, pps_val); + } else { + I915_WRITE(ICL_DSC0_PICTURE_PARAMETER_SET_4(pipe), pps_val); + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(ICL_DSC1_PICTURE_PARAMETER_SET_4(pipe), + pps_val); + } + + /* Populate PICTURE_PARAMETER_SET_5 registers */ + pps_val = 0; + pps_val |= DSC_SCALE_INC_INT(vdsc_cfg->scale_increment_interval) | + DSC_SCALE_DEC_INT(vdsc_cfg->scale_decrement_interval); + DRM_INFO("PPS5 = 0x%08x\n", pps_val); + if (cpu_transcoder == TRANSCODER_EDP) { + I915_WRITE(DSCA_PICTURE_PARAMETER_SET_5, pps_val); + /* + * If 2 VDSC instances are needed, configure PPS for second + * VDSC + */ + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(DSCC_PICTURE_PARAMETER_SET_5, pps_val); + } else { + I915_WRITE(ICL_DSC0_PICTURE_PARAMETER_SET_5(pipe), pps_val); + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(ICL_DSC1_PICTURE_PARAMETER_SET_5(pipe), + pps_val); + } + + /* Populate PICTURE_PARAMETER_SET_6 registers */ + pps_val = 0; + pps_val |= DSC_INITIAL_SCALE_VALUE(vdsc_cfg->initial_scale_value) | + DSC_FIRST_LINE_BPG_OFFSET(vdsc_cfg->first_line_bpg_offset) | + DSC_FLATNESS_MIN_QP(vdsc_cfg->flatness_min_qp) | + DSC_FLATNESS_MAX_QP(vdsc_cfg->flatness_max_qp); + DRM_INFO("PPS6 = 0x%08x\n", pps_val); + if (cpu_transcoder == TRANSCODER_EDP) { + I915_WRITE(DSCA_PICTURE_PARAMETER_SET_6, pps_val); + /* + * If 2 VDSC instances are needed, configure PPS for second + * VDSC + */ + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(DSCC_PICTURE_PARAMETER_SET_6, pps_val); + } else { + I915_WRITE(ICL_DSC0_PICTURE_PARAMETER_SET_6(pipe), pps_val); + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(ICL_DSC1_PICTURE_PARAMETER_SET_6(pipe), + pps_val); + } + + /* Populate PICTURE_PARAMETER_SET_7 registers */ + pps_val = 0; + pps_val |= DSC_SLICE_BPG_OFFSET(vdsc_cfg->slice_bpg_offset) | + DSC_NFL_BPG_OFFSET(vdsc_cfg->nfl_bpg_offset); + DRM_INFO("PPS7 = 0x%08x\n", pps_val); + if (cpu_transcoder == TRANSCODER_EDP) { + I915_WRITE(DSCA_PICTURE_PARAMETER_SET_7, pps_val); + /* + * If 2 VDSC instances are needed, configure PPS for second + * VDSC + */ + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(DSCC_PICTURE_PARAMETER_SET_7, pps_val); + } else { + I915_WRITE(ICL_DSC0_PICTURE_PARAMETER_SET_7(pipe), pps_val); + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(ICL_DSC1_PICTURE_PARAMETER_SET_7(pipe), + pps_val); + } + + /* Populate PICTURE_PARAMETER_SET_8 registers */ + pps_val = 0; + pps_val |= DSC_FINAL_OFFSET(vdsc_cfg->final_offset) | + DSC_INITIAL_OFFSET(vdsc_cfg->initial_offset); + DRM_INFO("PPS8 = 0x%08x\n", pps_val); + if (cpu_transcoder == TRANSCODER_EDP) { + I915_WRITE(DSCA_PICTURE_PARAMETER_SET_8, pps_val); + /* + * If 2 VDSC instances are needed, configure PPS for second + * VDSC + */ + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(DSCC_PICTURE_PARAMETER_SET_8, pps_val); + } else { + I915_WRITE(ICL_DSC0_PICTURE_PARAMETER_SET_8(pipe), pps_val); + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(ICL_DSC1_PICTURE_PARAMETER_SET_8(pipe), + pps_val); + } + + /* Populate PICTURE_PARAMETER_SET_9 registers */ + pps_val = 0; + pps_val |= DSC_RC_MODEL_SIZE(DSC_RC_MODEL_SIZE_CONST) | + DSC_RC_EDGE_FACTOR(DSC_RC_EDGE_FACTOR_CONST); + DRM_INFO("PPS9 = 0x%08x\n", pps_val); + if (cpu_transcoder == TRANSCODER_EDP) { + I915_WRITE(DSCA_PICTURE_PARAMETER_SET_9, pps_val); + /* + * If 2 VDSC instances are needed, configure PPS for second + * VDSC + */ + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(DSCC_PICTURE_PARAMETER_SET_9, pps_val); + } else { + I915_WRITE(ICL_DSC0_PICTURE_PARAMETER_SET_9(pipe), pps_val); + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(ICL_DSC1_PICTURE_PARAMETER_SET_9(pipe), + pps_val); + } + + /* Populate PICTURE_PARAMETER_SET_10 registers */ + pps_val = 0; + pps_val |= DSC_RC_QUANT_INC_LIMIT0(vdsc_cfg->rc_quant_incr_limit0) | + DSC_RC_QUANT_INC_LIMIT1(vdsc_cfg->rc_quant_incr_limit1) | + DSC_RC_TARGET_OFF_HIGH(DSC_RC_TGT_OFFSET_HI_CONST) | + DSC_RC_TARGET_OFF_LOW(DSC_RC_TGT_OFFSET_LO_CONST); + DRM_INFO("PPS10 = 0x%08x\n", pps_val); + if (cpu_transcoder == TRANSCODER_EDP) { + I915_WRITE(DSCA_PICTURE_PARAMETER_SET_10, pps_val); + /* + * If 2 VDSC instances are needed, configure PPS for second + * VDSC + */ + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(DSCC_PICTURE_PARAMETER_SET_10, pps_val); + } else { + I915_WRITE(ICL_DSC0_PICTURE_PARAMETER_SET_10(pipe), pps_val); + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(ICL_DSC1_PICTURE_PARAMETER_SET_10(pipe), + pps_val); + } + + /* Populate Picture parameter set 16 */ + pps_val = 0; + pps_val |= DSC_SLICE_CHUNK_SIZE(vdsc_cfg->slice_chunk_size) | + DSC_SLICE_PER_LINE((vdsc_cfg->pic_width / num_vdsc_instances) / + vdsc_cfg->slice_width) | + DSC_SLICE_ROW_PER_FRAME(vdsc_cfg->pic_height / + vdsc_cfg->slice_height); + DRM_INFO("PPS16 = 0x%08x\n", pps_val); + if (cpu_transcoder == TRANSCODER_EDP) { + I915_WRITE(DSCA_PICTURE_PARAMETER_SET_16, pps_val); + /* + * If 2 VDSC instances are needed, configure PPS for second + * VDSC + */ + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(DSCC_PICTURE_PARAMETER_SET_16, pps_val); + } else { + I915_WRITE(ICL_DSC0_PICTURE_PARAMETER_SET_16(pipe), pps_val); + if (crtc_state->dsc_params.dsc_split) + I915_WRITE(ICL_DSC1_PICTURE_PARAMETER_SET_16(pipe), + pps_val); + } + + /* Populate the RC_BUF_THRESH registers */ + memset(rc_buf_thresh_dword, 0, sizeof(rc_buf_thresh_dword)); + for (i = 0; i < DSC_NUM_BUF_RANGES - 1; i++) { + rc_buf_thresh_dword[i / 4] |= + (u32)(vdsc_cfg->rc_buf_thresh[i] << + BITS_PER_BYTE * (i % 4)); + DRM_INFO(" RC_BUF_THRESH%d = 0x%08x\n", i, + rc_buf_thresh_dword[i / 4]); + } + if (cpu_transcoder == TRANSCODER_EDP) { + I915_WRITE(DSCA_RC_BUF_THRESH_0, rc_buf_thresh_dword[0]); + I915_WRITE(DSCA_RC_BUF_THRESH_0_UDW, rc_buf_thresh_dword[1]); + I915_WRITE(DSCA_RC_BUF_THRESH_1, rc_buf_thresh_dword[2]); + I915_WRITE(DSCA_RC_BUF_THRESH_1_UDW, rc_buf_thresh_dword[3]); + if (crtc_state->dsc_params.dsc_split) { + I915_WRITE(DSCC_RC_BUF_THRESH_0, + rc_buf_thresh_dword[0]); + I915_WRITE(DSCC_RC_BUF_THRESH_0_UDW, + rc_buf_thresh_dword[1]); + I915_WRITE(DSCC_RC_BUF_THRESH_1, + rc_buf_thresh_dword[2]); + I915_WRITE(DSCC_RC_BUF_THRESH_1_UDW, + rc_buf_thresh_dword[3]); + } + } else { + I915_WRITE(ICL_DSC0_RC_BUF_THRESH_0(pipe), + rc_buf_thresh_dword[0]); + I915_WRITE(ICL_DSC0_RC_BUF_THRESH_0_UDW(pipe), + rc_buf_thresh_dword[1]); + I915_WRITE(ICL_DSC0_RC_BUF_THRESH_1(pipe), + rc_buf_thresh_dword[2]); + I915_WRITE(ICL_DSC0_RC_BUF_THRESH_1_UDW(pipe), + rc_buf_thresh_dword[3]); + if (crtc_state->dsc_params.dsc_split) { + I915_WRITE(ICL_DSC1_RC_BUF_THRESH_0(pipe), + rc_buf_thresh_dword[0]); + I915_WRITE(ICL_DSC1_RC_BUF_THRESH_0_UDW(pipe), + rc_buf_thresh_dword[1]); + I915_WRITE(ICL_DSC1_RC_BUF_THRESH_1(pipe), + rc_buf_thresh_dword[2]); + I915_WRITE(ICL_DSC1_RC_BUF_THRESH_1_UDW(pipe), + rc_buf_thresh_dword[3]); + } + } + + /* Populate the RC_RANGE_PARAMETERS registers */ + memset(rc_range_params_dword, 0, sizeof(rc_range_params_dword)); + for (i = 0; i < DSC_NUM_BUF_RANGES; i++) { + rc_range_params_dword[i / 2] |= + (u32)(((vdsc_cfg->rc_range_params[i].range_bpg_offset << + RC_BPG_OFFSET_SHIFT) | + (vdsc_cfg->rc_range_params[i].range_max_qp << + RC_MAX_QP_SHIFT) | + (vdsc_cfg->rc_range_params[i].range_min_qp << + RC_MIN_QP_SHIFT)) << 16 * (i % 2)); + DRM_INFO(" RC_RANGE_PARAM_%d = 0x%08x\n", i, + rc_range_params_dword[i / 2]); + } + if (cpu_transcoder == TRANSCODER_EDP) { + I915_WRITE(DSCA_RC_RANGE_PARAMETERS_0, + rc_range_params_dword[0]); + I915_WRITE(DSCA_RC_RANGE_PARAMETERS_0_UDW, + rc_range_params_dword[1]); + I915_WRITE(DSCA_RC_RANGE_PARAMETERS_1, + rc_range_params_dword[2]); + I915_WRITE(DSCA_RC_RANGE_PARAMETERS_1_UDW, + rc_range_params_dword[3]); + I915_WRITE(DSCA_RC_RANGE_PARAMETERS_2, + rc_range_params_dword[4]); + I915_WRITE(DSCA_RC_RANGE_PARAMETERS_2_UDW, + rc_range_params_dword[5]); + I915_WRITE(DSCA_RC_RANGE_PARAMETERS_3, + rc_range_params_dword[6]); + I915_WRITE(DSCA_RC_RANGE_PARAMETERS_3_UDW, + rc_range_params_dword[7]); + if (crtc_state->dsc_params.dsc_split) { + I915_WRITE(DSCC_RC_RANGE_PARAMETERS_0, + rc_range_params_dword[0]); + I915_WRITE(DSCC_RC_RANGE_PARAMETERS_0_UDW, + rc_range_params_dword[1]); + I915_WRITE(DSCC_RC_RANGE_PARAMETERS_1, + rc_range_params_dword[2]); + I915_WRITE(DSCC_RC_RANGE_PARAMETERS_1_UDW, + rc_range_params_dword[3]); + I915_WRITE(DSCC_RC_RANGE_PARAMETERS_2, + rc_range_params_dword[4]); + I915_WRITE(DSCC_RC_RANGE_PARAMETERS_2_UDW, + rc_range_params_dword[5]); + I915_WRITE(DSCC_RC_RANGE_PARAMETERS_3, + rc_range_params_dword[6]); + I915_WRITE(DSCC_RC_RANGE_PARAMETERS_3_UDW, + rc_range_params_dword[7]); + } + } else { + I915_WRITE(ICL_DSC0_RC_RANGE_PARAMETERS_0(pipe), + rc_range_params_dword[0]); + I915_WRITE(ICL_DSC0_RC_RANGE_PARAMETERS_0_UDW(pipe), + rc_range_params_dword[1]); + I915_WRITE(ICL_DSC0_RC_RANGE_PARAMETERS_1(pipe), + rc_range_params_dword[2]); + I915_WRITE(ICL_DSC0_RC_RANGE_PARAMETERS_1_UDW(pipe), + rc_range_params_dword[3]); + I915_WRITE(ICL_DSC0_RC_RANGE_PARAMETERS_2(pipe), + rc_range_params_dword[4]); + I915_WRITE(ICL_DSC0_RC_RANGE_PARAMETERS_2_UDW(pipe), + rc_range_params_dword[5]); + I915_WRITE(ICL_DSC0_RC_RANGE_PARAMETERS_3(pipe), + rc_range_params_dword[6]); + I915_WRITE(ICL_DSC0_RC_RANGE_PARAMETERS_3_UDW(pipe), + rc_range_params_dword[7]); + if (crtc_state->dsc_params.dsc_split) { + I915_WRITE(ICL_DSC1_RC_RANGE_PARAMETERS_0(pipe), + rc_range_params_dword[0]); + I915_WRITE(ICL_DSC1_RC_RANGE_PARAMETERS_0_UDW(pipe), + rc_range_params_dword[1]); + I915_WRITE(ICL_DSC1_RC_RANGE_PARAMETERS_1(pipe), + rc_range_params_dword[2]); + I915_WRITE(ICL_DSC1_RC_RANGE_PARAMETERS_1_UDW(pipe), + rc_range_params_dword[3]); + I915_WRITE(ICL_DSC1_RC_RANGE_PARAMETERS_2(pipe), + rc_range_params_dword[4]); + I915_WRITE(ICL_DSC1_RC_RANGE_PARAMETERS_2_UDW(pipe), + rc_range_params_dword[5]); + I915_WRITE(ICL_DSC1_RC_RANGE_PARAMETERS_3(pipe), + rc_range_params_dword[6]); + I915_WRITE(ICL_DSC1_RC_RANGE_PARAMETERS_3_UDW(pipe), + rc_range_params_dword[7]); + } + } +} + +static void intel_dp_write_dsc_pps_sdp(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) +{ + struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base); + struct intel_digital_port *intel_dig_port = dp_to_dig_port(intel_dp); + const struct drm_dsc_config *vdsc_cfg = &crtc_state->dp_dsc_cfg; + struct drm_dsc_pps_infoframe dp_dsc_pps_sdp; + + /* Prepare DP SDP PPS header as per DP 1.4 spec, Table 2-123 */ + drm_dsc_dp_pps_header_init(&dp_dsc_pps_sdp); + + /* Fill the PPS payload bytes as per DSC spec 1.2 Table 4-1 */ + drm_dsc_pps_infoframe_pack(&dp_dsc_pps_sdp, vdsc_cfg); + + intel_dig_port->write_infoframe(encoder, crtc_state, + DP_SDP_PPS, &dp_dsc_pps_sdp, + sizeof(dp_dsc_pps_sdp)); +} + +void intel_dsc_enable(struct intel_encoder *encoder, + const struct intel_crtc_state *crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(crtc_state->base.crtc); + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + enum pipe pipe = crtc->pipe; + i915_reg_t dss_ctl1_reg, dss_ctl2_reg; + u32 dss_ctl1_val = 0; + u32 dss_ctl2_val = 0; + + if (!crtc_state->dsc_params.compression_enable) + return; + + /* Enable Power wells for VDSC/joining */ + intel_display_power_get(dev_priv, + intel_dsc_power_domain(crtc_state)); + + intel_configure_pps_for_dsc_encoder(encoder, crtc_state); + + intel_dp_write_dsc_pps_sdp(encoder, crtc_state); + + if (crtc_state->cpu_transcoder == TRANSCODER_EDP) { + dss_ctl1_reg = DSS_CTL1; + dss_ctl2_reg = DSS_CTL2; + } else { + dss_ctl1_reg = ICL_PIPE_DSS_CTL1(pipe); + dss_ctl2_reg = ICL_PIPE_DSS_CTL2(pipe); + } + dss_ctl2_val |= LEFT_BRANCH_VDSC_ENABLE; + if (crtc_state->dsc_params.dsc_split) { + dss_ctl2_val |= RIGHT_BRANCH_VDSC_ENABLE; + dss_ctl1_val |= JOINER_ENABLE; + } + I915_WRITE(dss_ctl1_reg, dss_ctl1_val); + I915_WRITE(dss_ctl2_reg, dss_ctl2_val); +} + +void intel_dsc_disable(const struct intel_crtc_state *old_crtc_state) +{ + struct intel_crtc *crtc = to_intel_crtc(old_crtc_state->base.crtc); + struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); + enum pipe pipe = crtc->pipe; + i915_reg_t dss_ctl1_reg, dss_ctl2_reg; + u32 dss_ctl1_val = 0, dss_ctl2_val = 0; + + if (!old_crtc_state->dsc_params.compression_enable) + return; + + if (old_crtc_state->cpu_transcoder == TRANSCODER_EDP) { + dss_ctl1_reg = DSS_CTL1; + dss_ctl2_reg = DSS_CTL2; + } else { + dss_ctl1_reg = ICL_PIPE_DSS_CTL1(pipe); + dss_ctl2_reg = ICL_PIPE_DSS_CTL2(pipe); + } + dss_ctl1_val = I915_READ(dss_ctl1_reg); + if (dss_ctl1_val & JOINER_ENABLE) + dss_ctl1_val &= ~JOINER_ENABLE; + I915_WRITE(dss_ctl1_reg, dss_ctl1_val); + + dss_ctl2_val = I915_READ(dss_ctl2_reg); + if (dss_ctl2_val & LEFT_BRANCH_VDSC_ENABLE || + dss_ctl2_val & RIGHT_BRANCH_VDSC_ENABLE) + dss_ctl2_val &= ~(LEFT_BRANCH_VDSC_ENABLE | + RIGHT_BRANCH_VDSC_ENABLE); + I915_WRITE(dss_ctl2_reg, dss_ctl2_val); + + /* Disable Power wells for VDSC/joining */ + intel_display_power_put(dev_priv, + intel_dsc_power_domain(old_crtc_state)); +} diff --git a/drivers/gpu/drm/i915/intel_workarounds.c b/drivers/gpu/drm/i915/intel_workarounds.c index ca1f78a42b17..4f41e326f3f3 100644 --- a/drivers/gpu/drm/i915/intel_workarounds.c +++ b/drivers/gpu/drm/i915/intel_workarounds.c @@ -48,58 +48,112 @@ * - Public functions to init or apply the given workaround type. */ -static void wa_add(struct drm_i915_private *i915, - i915_reg_t reg, const u32 mask, const u32 val) +static void wa_init_start(struct i915_wa_list *wal, const char *name) { - struct i915_workarounds *wa = &i915->workarounds; - unsigned int start = 0, end = wa->count; - unsigned int addr = i915_mmio_reg_offset(reg); - struct i915_wa_reg *r; + wal->name = name; +} + +#define WA_LIST_CHUNK (1 << 4) + +static void wa_init_finish(struct i915_wa_list *wal) +{ + /* Trim unused entries. */ + if (!IS_ALIGNED(wal->count, WA_LIST_CHUNK)) { + struct i915_wa *list = kmemdup(wal->list, + wal->count * sizeof(*list), + GFP_KERNEL); + + if (list) { + kfree(wal->list); + wal->list = list; + } + } + + if (!wal->count) + return; + + DRM_DEBUG_DRIVER("Initialized %u %s workarounds\n", + wal->wa_count, wal->name); +} + +static void _wa_add(struct i915_wa_list *wal, const struct i915_wa *wa) +{ + unsigned int addr = i915_mmio_reg_offset(wa->reg); + unsigned int start = 0, end = wal->count; + const unsigned int grow = WA_LIST_CHUNK; + struct i915_wa *wa_; + + GEM_BUG_ON(!is_power_of_2(grow)); + + if (IS_ALIGNED(wal->count, grow)) { /* Either uninitialized or full. */ + struct i915_wa *list; + + list = kmalloc_array(ALIGN(wal->count + 1, grow), sizeof(*wa), + GFP_KERNEL); + if (!list) { + DRM_ERROR("No space for workaround init!\n"); + return; + } + + if (wal->list) + memcpy(list, wal->list, sizeof(*wa) * wal->count); + + wal->list = list; + } while (start < end) { unsigned int mid = start + (end - start) / 2; - if (wa->reg[mid].addr < addr) { + if (i915_mmio_reg_offset(wal->list[mid].reg) < addr) { start = mid + 1; - } else if (wa->reg[mid].addr > addr) { + } else if (i915_mmio_reg_offset(wal->list[mid].reg) > addr) { end = mid; } else { - r = &wa->reg[mid]; + wa_ = &wal->list[mid]; - if ((mask & ~r->mask) == 0) { + if ((wa->mask & ~wa_->mask) == 0) { DRM_ERROR("Discarding overwritten w/a for reg %04x (mask: %08x, value: %08x)\n", - addr, r->mask, r->value); + i915_mmio_reg_offset(wa_->reg), + wa_->mask, wa_->val); - r->value &= ~mask; + wa_->val &= ~wa->mask; } - r->value |= val; - r->mask |= mask; + wal->wa_count++; + wa_->val |= wa->val; + wa_->mask |= wa->mask; return; } } - if (WARN_ON_ONCE(wa->count >= I915_MAX_WA_REGS)) { - DRM_ERROR("Dropping w/a for reg %04x (mask: %08x, value: %08x)\n", - addr, mask, val); - return; - } - - r = &wa->reg[wa->count++]; - r->addr = addr; - r->value = val; - r->mask = mask; + wal->wa_count++; + wa_ = &wal->list[wal->count++]; + *wa_ = *wa; - while (r-- > wa->reg) { - GEM_BUG_ON(r[0].addr == r[1].addr); - if (r[1].addr > r[0].addr) + while (wa_-- > wal->list) { + GEM_BUG_ON(i915_mmio_reg_offset(wa_[0].reg) == + i915_mmio_reg_offset(wa_[1].reg)); + if (i915_mmio_reg_offset(wa_[1].reg) > + i915_mmio_reg_offset(wa_[0].reg)) break; - swap(r[1], r[0]); + swap(wa_[1], wa_[0]); } } -#define WA_REG(addr, mask, val) wa_add(dev_priv, (addr), (mask), (val)) +static void +__wa_add(struct i915_wa_list *wal, i915_reg_t reg, u32 mask, u32 val) +{ + struct i915_wa wa = { + .reg = reg, + .mask = mask, + .val = val + }; + + _wa_add(wal, &wa); +} + +#define WA_REG(addr, mask, val) __wa_add(wal, (addr), (mask), (val)) #define WA_SET_BIT_MASKED(addr, mask) \ WA_REG(addr, (mask), _MASKED_BIT_ENABLE(mask)) @@ -110,8 +164,10 @@ static void wa_add(struct drm_i915_private *i915, #define WA_SET_FIELD_MASKED(addr, mask, value) \ WA_REG(addr, (mask), _MASKED_FIELD(mask, value)) -static int gen8_ctx_workarounds_init(struct drm_i915_private *dev_priv) +static void gen8_ctx_workarounds_init(struct intel_engine_cs *engine) { + struct i915_wa_list *wal = &engine->ctx_wa_list; + WA_SET_BIT_MASKED(INSTPM, INSTPM_FORCE_ORDERING); /* WaDisableAsyncFlipPerfMode:bdw,chv */ @@ -155,17 +211,14 @@ static int gen8_ctx_workarounds_init(struct drm_i915_private *dev_priv) WA_SET_FIELD_MASKED(GEN7_GT_MODE, GEN6_WIZ_HASHING_MASK, GEN6_WIZ_HASHING_16x4); - - return 0; } -static int bdw_ctx_workarounds_init(struct drm_i915_private *dev_priv) +static void bdw_ctx_workarounds_init(struct intel_engine_cs *engine) { - int ret; + struct drm_i915_private *i915 = engine->i915; + struct i915_wa_list *wal = &engine->ctx_wa_list; - ret = gen8_ctx_workarounds_init(dev_priv); - if (ret) - return ret; + gen8_ctx_workarounds_init(engine); /* WaDisableThreadStallDopClockGating:bdw (pre-production) */ WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN, STALL_DOP_GATING_DISABLE); @@ -185,31 +238,28 @@ static int bdw_ctx_workarounds_init(struct drm_i915_private *dev_priv) /* WaForceContextSaveRestoreNonCoherent:bdw */ HDC_FORCE_CONTEXT_SAVE_RESTORE_NON_COHERENT | /* WaDisableFenceDestinationToSLM:bdw (pre-prod) */ - (IS_BDW_GT3(dev_priv) ? HDC_FENCE_DEST_SLM_DISABLE : 0)); - - return 0; + (IS_BDW_GT3(i915) ? HDC_FENCE_DEST_SLM_DISABLE : 0)); } -static int chv_ctx_workarounds_init(struct drm_i915_private *dev_priv) +static void chv_ctx_workarounds_init(struct intel_engine_cs *engine) { - int ret; + struct i915_wa_list *wal = &engine->ctx_wa_list; - ret = gen8_ctx_workarounds_init(dev_priv); - if (ret) - return ret; + gen8_ctx_workarounds_init(engine); /* WaDisableThreadStallDopClockGating:chv */ WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN, STALL_DOP_GATING_DISABLE); /* Improve HiZ throughput on CHV. */ WA_SET_BIT_MASKED(HIZ_CHICKEN, CHV_HZ_8X8_MODE_IN_1X); - - return 0; } -static int gen9_ctx_workarounds_init(struct drm_i915_private *dev_priv) +static void gen9_ctx_workarounds_init(struct intel_engine_cs *engine) { - if (HAS_LLC(dev_priv)) { + struct drm_i915_private *i915 = engine->i915; + struct i915_wa_list *wal = &engine->ctx_wa_list; + + if (HAS_LLC(i915)) { /* WaCompressedResourceSamplerPbeMediaNewHashMode:skl,kbl * * Must match Display Engine. See @@ -228,7 +278,7 @@ static int gen9_ctx_workarounds_init(struct drm_i915_private *dev_priv) PARTIAL_INSTRUCTION_SHOOTDOWN_DISABLE); /* Syncing dependencies between camera and graphics:skl,bxt,kbl */ - if (!IS_COFFEELAKE(dev_priv)) + if (!IS_COFFEELAKE(i915)) WA_SET_BIT_MASKED(HALF_SLICE_CHICKEN3, GEN9_DISABLE_OCL_OOB_SUPPRESS_LOGIC); @@ -271,9 +321,7 @@ static int gen9_ctx_workarounds_init(struct drm_i915_private *dev_priv) HDC_FORCE_NON_COHERENT); /* WaDisableSamplerPowerBypassForSOPingPong:skl,bxt,kbl,cfl */ - if (IS_SKYLAKE(dev_priv) || - IS_KABYLAKE(dev_priv) || - IS_COFFEELAKE(dev_priv)) + if (IS_SKYLAKE(i915) || IS_KABYLAKE(i915) || IS_COFFEELAKE(i915)) WA_SET_BIT_MASKED(HALF_SLICE_CHICKEN3, GEN8_SAMPLER_POWER_BYPASS_DIS); @@ -300,14 +348,14 @@ static int gen9_ctx_workarounds_init(struct drm_i915_private *dev_priv) GEN9_PREEMPT_GPGPU_COMMAND_LEVEL); /* WaClearHIZ_WM_CHICKEN3:bxt,glk */ - if (IS_GEN9_LP(dev_priv)) + if (IS_GEN9_LP(i915)) WA_SET_BIT_MASKED(GEN9_WM_CHICKEN3, GEN9_FACTOR_IN_CLR_VAL_HIZ); - - return 0; } -static int skl_tune_iz_hashing(struct drm_i915_private *dev_priv) +static void skl_tune_iz_hashing(struct intel_engine_cs *engine) { + struct drm_i915_private *i915 = engine->i915; + struct i915_wa_list *wal = &engine->ctx_wa_list; u8 vals[3] = { 0, 0, 0 }; unsigned int i; @@ -318,7 +366,7 @@ static int skl_tune_iz_hashing(struct drm_i915_private *dev_priv) * Only consider slices where one, and only one, subslice has 7 * EUs */ - if (!is_power_of_2(INTEL_INFO(dev_priv)->sseu.subslice_7eu[i])) + if (!is_power_of_2(INTEL_INFO(i915)->sseu.subslice_7eu[i])) continue; /* @@ -327,12 +375,12 @@ static int skl_tune_iz_hashing(struct drm_i915_private *dev_priv) * * -> 0 <= ss <= 3; */ - ss = ffs(INTEL_INFO(dev_priv)->sseu.subslice_7eu[i]) - 1; + ss = ffs(INTEL_INFO(i915)->sseu.subslice_7eu[i]) - 1; vals[i] = 3 - ss; } if (vals[0] == 0 && vals[1] == 0 && vals[2] == 0) - return 0; + return; /* Tune IZ hashing. See intel_device_info_runtime_init() */ WA_SET_FIELD_MASKED(GEN7_GT_MODE, @@ -342,28 +390,19 @@ static int skl_tune_iz_hashing(struct drm_i915_private *dev_priv) GEN9_IZ_HASHING(2, vals[2]) | GEN9_IZ_HASHING(1, vals[1]) | GEN9_IZ_HASHING(0, vals[0])); - - return 0; } -static int skl_ctx_workarounds_init(struct drm_i915_private *dev_priv) +static void skl_ctx_workarounds_init(struct intel_engine_cs *engine) { - int ret; - - ret = gen9_ctx_workarounds_init(dev_priv); - if (ret) - return ret; - - return skl_tune_iz_hashing(dev_priv); + gen9_ctx_workarounds_init(engine); + skl_tune_iz_hashing(engine); } -static int bxt_ctx_workarounds_init(struct drm_i915_private *dev_priv) +static void bxt_ctx_workarounds_init(struct intel_engine_cs *engine) { - int ret; + struct i915_wa_list *wal = &engine->ctx_wa_list; - ret = gen9_ctx_workarounds_init(dev_priv); - if (ret) - return ret; + gen9_ctx_workarounds_init(engine); /* WaDisableThreadStallDopClockGating:bxt */ WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN, @@ -372,57 +411,41 @@ static int bxt_ctx_workarounds_init(struct drm_i915_private *dev_priv) /* WaToEnableHwFixForPushConstHWBug:bxt */ WA_SET_BIT_MASKED(COMMON_SLICE_CHICKEN2, GEN8_SBE_DISABLE_REPLAY_BUF_OPTIMIZATION); - - return 0; } -static int kbl_ctx_workarounds_init(struct drm_i915_private *dev_priv) +static void kbl_ctx_workarounds_init(struct intel_engine_cs *engine) { - int ret; - - ret = gen9_ctx_workarounds_init(dev_priv); - if (ret) - return ret; + struct drm_i915_private *i915 = engine->i915; + struct i915_wa_list *wal = &engine->ctx_wa_list; - /* WaDisableFenceDestinationToSLM:kbl (pre-prod) */ - if (IS_KBL_REVID(dev_priv, KBL_REVID_A0, KBL_REVID_A0)) - WA_SET_BIT_MASKED(HDC_CHICKEN0, - HDC_FENCE_DEST_SLM_DISABLE); + gen9_ctx_workarounds_init(engine); /* WaToEnableHwFixForPushConstHWBug:kbl */ - if (IS_KBL_REVID(dev_priv, KBL_REVID_C0, REVID_FOREVER)) + if (IS_KBL_REVID(i915, KBL_REVID_C0, REVID_FOREVER)) WA_SET_BIT_MASKED(COMMON_SLICE_CHICKEN2, GEN8_SBE_DISABLE_REPLAY_BUF_OPTIMIZATION); /* WaDisableSbeCacheDispatchPortSharing:kbl */ WA_SET_BIT_MASKED(GEN7_HALF_SLICE_CHICKEN1, GEN7_SBE_SS_CACHE_DISPATCH_PORT_SHARING_DISABLE); - - return 0; } -static int glk_ctx_workarounds_init(struct drm_i915_private *dev_priv) +static void glk_ctx_workarounds_init(struct intel_engine_cs *engine) { - int ret; + struct i915_wa_list *wal = &engine->ctx_wa_list; - ret = gen9_ctx_workarounds_init(dev_priv); - if (ret) - return ret; + gen9_ctx_workarounds_init(engine); /* WaToEnableHwFixForPushConstHWBug:glk */ WA_SET_BIT_MASKED(COMMON_SLICE_CHICKEN2, GEN8_SBE_DISABLE_REPLAY_BUF_OPTIMIZATION); - - return 0; } -static int cfl_ctx_workarounds_init(struct drm_i915_private *dev_priv) +static void cfl_ctx_workarounds_init(struct intel_engine_cs *engine) { - int ret; + struct i915_wa_list *wal = &engine->ctx_wa_list; - ret = gen9_ctx_workarounds_init(dev_priv); - if (ret) - return ret; + gen9_ctx_workarounds_init(engine); /* WaToEnableHwFixForPushConstHWBug:cfl */ WA_SET_BIT_MASKED(COMMON_SLICE_CHICKEN2, @@ -431,18 +454,19 @@ static int cfl_ctx_workarounds_init(struct drm_i915_private *dev_priv) /* WaDisableSbeCacheDispatchPortSharing:cfl */ WA_SET_BIT_MASKED(GEN7_HALF_SLICE_CHICKEN1, GEN7_SBE_SS_CACHE_DISPATCH_PORT_SHARING_DISABLE); - - return 0; } -static int cnl_ctx_workarounds_init(struct drm_i915_private *dev_priv) +static void cnl_ctx_workarounds_init(struct intel_engine_cs *engine) { + struct drm_i915_private *i915 = engine->i915; + struct i915_wa_list *wal = &engine->ctx_wa_list; + /* WaForceContextSaveRestoreNonCoherent:cnl */ WA_SET_BIT_MASKED(CNL_HDC_CHICKEN0, HDC_FORCE_CONTEXT_SAVE_RESTORE_NON_COHERENT); /* WaThrottleEUPerfToAvoidTDBackPressure:cnl(pre-prod) */ - if (IS_CNL_REVID(dev_priv, CNL_REVID_B0, CNL_REVID_B0)) + if (IS_CNL_REVID(i915, CNL_REVID_B0, CNL_REVID_B0)) WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN, THROTTLE_12_5); /* WaDisableReplayBufferBankArbitrationOptimization:cnl */ @@ -450,7 +474,7 @@ static int cnl_ctx_workarounds_init(struct drm_i915_private *dev_priv) GEN8_SBE_DISABLE_REPLAY_BUF_OPTIMIZATION); /* WaDisableEnhancedSBEVertexCaching:cnl (pre-prod) */ - if (IS_CNL_REVID(dev_priv, 0, CNL_REVID_B0)) + if (IS_CNL_REVID(i915, 0, CNL_REVID_B0)) WA_SET_BIT_MASKED(COMMON_SLICE_CHICKEN2, GEN8_CSC2_SBE_VUE_CACHE_CONSERVATIVE); @@ -470,16 +494,17 @@ static int cnl_ctx_workarounds_init(struct drm_i915_private *dev_priv) /* WaDisableEarlyEOT:cnl */ WA_SET_BIT_MASKED(GEN8_ROW_CHICKEN, DISABLE_EARLY_EOT); - - return 0; } -static int icl_ctx_workarounds_init(struct drm_i915_private *dev_priv) +static void icl_ctx_workarounds_init(struct intel_engine_cs *engine) { + struct drm_i915_private *i915 = engine->i915; + struct i915_wa_list *wal = &engine->ctx_wa_list; + /* Wa_1604370585:icl (pre-prod) * Formerly known as WaPushConstantDereferenceHoldDisable */ - if (IS_ICL_REVID(dev_priv, ICL_REVID_A0, ICL_REVID_B0)) + if (IS_ICL_REVID(i915, ICL_REVID_A0, ICL_REVID_B0)) WA_SET_BIT_MASKED(GEN7_ROW_CHICKEN2, PUSH_CONSTANT_DEREF_DISABLE); @@ -495,7 +520,7 @@ static int icl_ctx_workarounds_init(struct drm_i915_private *dev_priv) /* Wa_2006611047:icl (pre-prod) * Formerly known as WaDisableImprovedTdlClkGating */ - if (IS_ICL_REVID(dev_priv, ICL_REVID_A0, ICL_REVID_A0)) + if (IS_ICL_REVID(i915, ICL_REVID_A0, ICL_REVID_A0)) WA_SET_BIT_MASKED(GEN7_ROW_CHICKEN2, GEN11_TDL_CLOCK_GATING_FIX_DISABLE); @@ -504,70 +529,67 @@ static int icl_ctx_workarounds_init(struct drm_i915_private *dev_priv) GEN11_STATE_CACHE_REDIRECT_TO_CS); /* Wa_2006665173:icl (pre-prod) */ - if (IS_ICL_REVID(dev_priv, ICL_REVID_A0, ICL_REVID_A0)) + if (IS_ICL_REVID(i915, ICL_REVID_A0, ICL_REVID_A0)) WA_SET_BIT_MASKED(GEN11_COMMON_SLICE_CHICKEN3, GEN11_BLEND_EMB_FIX_DISABLE_IN_RCC); - - return 0; } -int intel_ctx_workarounds_init(struct drm_i915_private *dev_priv) +void intel_engine_init_ctx_wa(struct intel_engine_cs *engine) { - int err = 0; - - dev_priv->workarounds.count = 0; - - if (INTEL_GEN(dev_priv) < 8) - err = 0; - else if (IS_BROADWELL(dev_priv)) - err = bdw_ctx_workarounds_init(dev_priv); - else if (IS_CHERRYVIEW(dev_priv)) - err = chv_ctx_workarounds_init(dev_priv); - else if (IS_SKYLAKE(dev_priv)) - err = skl_ctx_workarounds_init(dev_priv); - else if (IS_BROXTON(dev_priv)) - err = bxt_ctx_workarounds_init(dev_priv); - else if (IS_KABYLAKE(dev_priv)) - err = kbl_ctx_workarounds_init(dev_priv); - else if (IS_GEMINILAKE(dev_priv)) - err = glk_ctx_workarounds_init(dev_priv); - else if (IS_COFFEELAKE(dev_priv)) - err = cfl_ctx_workarounds_init(dev_priv); - else if (IS_CANNONLAKE(dev_priv)) - err = cnl_ctx_workarounds_init(dev_priv); - else if (IS_ICELAKE(dev_priv)) - err = icl_ctx_workarounds_init(dev_priv); + struct drm_i915_private *i915 = engine->i915; + struct i915_wa_list *wal = &engine->ctx_wa_list; + + wa_init_start(wal, "context"); + + if (INTEL_GEN(i915) < 8) + return; + else if (IS_BROADWELL(i915)) + bdw_ctx_workarounds_init(engine); + else if (IS_CHERRYVIEW(i915)) + chv_ctx_workarounds_init(engine); + else if (IS_SKYLAKE(i915)) + skl_ctx_workarounds_init(engine); + else if (IS_BROXTON(i915)) + bxt_ctx_workarounds_init(engine); + else if (IS_KABYLAKE(i915)) + kbl_ctx_workarounds_init(engine); + else if (IS_GEMINILAKE(i915)) + glk_ctx_workarounds_init(engine); + else if (IS_COFFEELAKE(i915)) + cfl_ctx_workarounds_init(engine); + else if (IS_CANNONLAKE(i915)) + cnl_ctx_workarounds_init(engine); + else if (IS_ICELAKE(i915)) + icl_ctx_workarounds_init(engine); else - MISSING_CASE(INTEL_GEN(dev_priv)); - if (err) - return err; + MISSING_CASE(INTEL_GEN(i915)); - DRM_DEBUG_DRIVER("Number of context specific w/a: %d\n", - dev_priv->workarounds.count); - return 0; + wa_init_finish(wal); } -int intel_ctx_workarounds_emit(struct i915_request *rq) +int intel_engine_emit_ctx_wa(struct i915_request *rq) { - struct i915_workarounds *w = &rq->i915->workarounds; + struct i915_wa_list *wal = &rq->engine->ctx_wa_list; + struct i915_wa *wa; + unsigned int i; u32 *cs; - int ret, i; + int ret; - if (w->count == 0) + if (wal->count == 0) return 0; ret = rq->engine->emit_flush(rq, EMIT_BARRIER); if (ret) return ret; - cs = intel_ring_begin(rq, (w->count * 2 + 2)); + cs = intel_ring_begin(rq, (wal->count * 2 + 2)); if (IS_ERR(cs)) return PTR_ERR(cs); - *cs++ = MI_LOAD_REGISTER_IMM(w->count); - for (i = 0; i < w->count; i++) { - *cs++ = w->reg[i].addr; - *cs++ = w->reg[i].value; + *cs++ = MI_LOAD_REGISTER_IMM(wal->count); + for (i = 0, wa = wal->list; i < wal->count; i++, wa++) { + *cs++ = i915_mmio_reg_offset(wa->reg); + *cs++ = wa->val; } *cs++ = MI_NOOP; @@ -580,160 +602,149 @@ int intel_ctx_workarounds_emit(struct i915_request *rq) return 0; } -static void bdw_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void +wa_masked_en(struct i915_wa_list *wal, i915_reg_t reg, u32 val) +{ + struct i915_wa wa = { + .reg = reg, + .mask = val, + .val = _MASKED_BIT_ENABLE(val) + }; + + _wa_add(wal, &wa); +} + +static void +wa_write_masked_or(struct i915_wa_list *wal, i915_reg_t reg, u32 mask, + u32 val) { + struct i915_wa wa = { + .reg = reg, + .mask = mask, + .val = val + }; + + _wa_add(wal, &wa); } -static void chv_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void +wa_write(struct i915_wa_list *wal, i915_reg_t reg, u32 val) { + wa_write_masked_or(wal, reg, ~0, val); } -static void gen9_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void +wa_write_or(struct i915_wa_list *wal, i915_reg_t reg, u32 val) { - /* WaContextSwitchWithConcurrentTLBInvalidate:skl,bxt,kbl,glk,cfl */ - I915_WRITE(GEN9_CSFE_CHICKEN1_RCS, - _MASKED_BIT_ENABLE(GEN9_PREEMPT_GPGPU_SYNC_SWITCH_DISABLE)); + wa_write_masked_or(wal, reg, val, val); +} - /* WaEnableLbsSlaRetryTimerDecrement:skl,bxt,kbl,glk,cfl */ - I915_WRITE(BDW_SCRATCH1, I915_READ(BDW_SCRATCH1) | - GEN9_LBS_SLA_RETRY_TIMER_DECREMENT_ENABLE); +static void gen9_gt_workarounds_init(struct drm_i915_private *i915) +{ + struct i915_wa_list *wal = &i915->gt_wa_list; /* WaDisableKillLogic:bxt,skl,kbl */ - if (!IS_COFFEELAKE(dev_priv)) - I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) | - ECOCHK_DIS_TLB); + if (!IS_COFFEELAKE(i915)) + wa_write_or(wal, + GAM_ECOCHK, + ECOCHK_DIS_TLB); - if (HAS_LLC(dev_priv)) { + if (HAS_LLC(i915)) { /* WaCompressedResourceSamplerPbeMediaNewHashMode:skl,kbl * * Must match Display Engine. See * WaCompressedResourceDisplayNewHashMode. */ - I915_WRITE(MMCD_MISC_CTRL, - I915_READ(MMCD_MISC_CTRL) | - MMCD_PCLA | - MMCD_HOTSPOT_EN); + wa_write_or(wal, + MMCD_MISC_CTRL, + MMCD_PCLA | MMCD_HOTSPOT_EN); } /* WaDisableHDCInvalidation:skl,bxt,kbl,cfl */ - I915_WRITE(GAM_ECOCHK, I915_READ(GAM_ECOCHK) | - BDW_DISABLE_HDC_INVALIDATION); - - /* WaProgramL3SqcReg1DefaultForPerf:bxt,glk */ - if (IS_GEN9_LP(dev_priv)) { - u32 val = I915_READ(GEN8_L3SQCREG1); - - val &= ~L3_PRIO_CREDITS_MASK; - val |= L3_GENERAL_PRIO_CREDITS(62) | L3_HIGH_PRIO_CREDITS(2); - I915_WRITE(GEN8_L3SQCREG1, val); - } - - /* WaOCLCoherentLineFlush:skl,bxt,kbl,cfl */ - I915_WRITE(GEN8_L3SQCREG4, - I915_READ(GEN8_L3SQCREG4) | GEN8_LQSC_FLUSH_COHERENT_LINES); - - /* WaEnablePreemptionGranularityControlByUMD:skl,bxt,kbl,cfl,[cnl] */ - I915_WRITE(GEN7_FF_SLICE_CS_CHICKEN1, - _MASKED_BIT_ENABLE(GEN9_FFSC_PERCTX_PREEMPT_CTRL)); + wa_write_or(wal, + GAM_ECOCHK, + BDW_DISABLE_HDC_INVALIDATION); } -static void skl_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void skl_gt_workarounds_init(struct drm_i915_private *i915) { - gen9_gt_workarounds_apply(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; - /* WaEnableGapsTsvCreditFix:skl */ - I915_WRITE(GEN8_GARBCNTL, - I915_READ(GEN8_GARBCNTL) | GEN9_GAPS_TSV_CREDIT_DISABLE); + gen9_gt_workarounds_init(i915); /* WaDisableGafsUnitClkGating:skl */ - I915_WRITE(GEN7_UCGCTL4, - I915_READ(GEN7_UCGCTL4) | GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); + wa_write_or(wal, + GEN7_UCGCTL4, + GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); /* WaInPlaceDecompressionHang:skl */ - if (IS_SKL_REVID(dev_priv, SKL_REVID_H0, REVID_FOREVER)) - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, - I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); + if (IS_SKL_REVID(i915, SKL_REVID_H0, REVID_FOREVER)) + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); } -static void bxt_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void bxt_gt_workarounds_init(struct drm_i915_private *i915) { - gen9_gt_workarounds_apply(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; - /* WaDisablePooledEuLoadBalancingFix:bxt */ - I915_WRITE(FF_SLICE_CS_CHICKEN2, - _MASKED_BIT_ENABLE(GEN9_POOLED_EU_LOAD_BALANCING_FIX_DISABLE)); + gen9_gt_workarounds_init(i915); /* WaInPlaceDecompressionHang:bxt */ - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, - I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); } -static void kbl_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void kbl_gt_workarounds_init(struct drm_i915_private *i915) { - gen9_gt_workarounds_apply(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; - /* WaEnableGapsTsvCreditFix:kbl */ - I915_WRITE(GEN8_GARBCNTL, - I915_READ(GEN8_GARBCNTL) | GEN9_GAPS_TSV_CREDIT_DISABLE); + gen9_gt_workarounds_init(i915); /* WaDisableDynamicCreditSharing:kbl */ - if (IS_KBL_REVID(dev_priv, 0, KBL_REVID_B0)) - I915_WRITE(GAMT_CHKN_BIT_REG, - I915_READ(GAMT_CHKN_BIT_REG) | - GAMT_CHKN_DISABLE_DYNAMIC_CREDIT_SHARING); + if (IS_KBL_REVID(i915, 0, KBL_REVID_B0)) + wa_write_or(wal, + GAMT_CHKN_BIT_REG, + GAMT_CHKN_DISABLE_DYNAMIC_CREDIT_SHARING); /* WaDisableGafsUnitClkGating:kbl */ - I915_WRITE(GEN7_UCGCTL4, - I915_READ(GEN7_UCGCTL4) | GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); + wa_write_or(wal, + GEN7_UCGCTL4, + GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); /* WaInPlaceDecompressionHang:kbl */ - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, - I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); - - /* WaKBLVECSSemaphoreWaitPoll:kbl */ - if (IS_KBL_REVID(dev_priv, KBL_REVID_A0, KBL_REVID_E0)) { - struct intel_engine_cs *engine; - unsigned int tmp; - - for_each_engine(engine, dev_priv, tmp) { - if (engine->id == RCS) - continue; - - I915_WRITE(RING_SEMA_WAIT_POLL(engine->mmio_base), 1); - } - } + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); } -static void glk_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void glk_gt_workarounds_init(struct drm_i915_private *i915) { - gen9_gt_workarounds_apply(dev_priv); + gen9_gt_workarounds_init(i915); } -static void cfl_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void cfl_gt_workarounds_init(struct drm_i915_private *i915) { - gen9_gt_workarounds_apply(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; - /* WaEnableGapsTsvCreditFix:cfl */ - I915_WRITE(GEN8_GARBCNTL, - I915_READ(GEN8_GARBCNTL) | GEN9_GAPS_TSV_CREDIT_DISABLE); + gen9_gt_workarounds_init(i915); /* WaDisableGafsUnitClkGating:cfl */ - I915_WRITE(GEN7_UCGCTL4, - I915_READ(GEN7_UCGCTL4) | GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); + wa_write_or(wal, + GEN7_UCGCTL4, + GEN8_EU_GAUNIT_CLOCK_GATE_DISABLE); /* WaInPlaceDecompressionHang:cfl */ - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, - I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); } static void wa_init_mcr(struct drm_i915_private *dev_priv) { const struct sseu_dev_info *sseu = &(INTEL_INFO(dev_priv)->sseu); - u32 mcr; + struct i915_wa_list *wal = &dev_priv->gt_wa_list; u32 mcr_slice_subslice_mask; /* @@ -770,8 +781,6 @@ static void wa_init_mcr(struct drm_i915_private *dev_priv) WARN_ON((enabled_mask & disabled_mask) != enabled_mask); } - mcr = I915_READ(GEN8_MCR_SELECTOR); - if (INTEL_GEN(dev_priv) >= 11) mcr_slice_subslice_mask = GEN11_MCR_SLICE_MASK | GEN11_MCR_SUBSLICE_MASK; @@ -789,186 +798,220 @@ static void wa_init_mcr(struct drm_i915_private *dev_priv) * occasions, such as INSTDONE, where this value is dependent * on s/ss combo, the read should be done with read_subslice_reg. */ - mcr &= ~mcr_slice_subslice_mask; - mcr |= intel_calculate_mcr_s_ss_select(dev_priv); - I915_WRITE(GEN8_MCR_SELECTOR, mcr); + wa_write_masked_or(wal, + GEN8_MCR_SELECTOR, + mcr_slice_subslice_mask, + intel_calculate_mcr_s_ss_select(dev_priv)); } -static void cnl_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void cnl_gt_workarounds_init(struct drm_i915_private *i915) { - wa_init_mcr(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; + + wa_init_mcr(i915); /* WaDisableI2mCycleOnWRPort:cnl (pre-prod) */ - if (IS_CNL_REVID(dev_priv, CNL_REVID_B0, CNL_REVID_B0)) - I915_WRITE(GAMT_CHKN_BIT_REG, - I915_READ(GAMT_CHKN_BIT_REG) | - GAMT_CHKN_DISABLE_I2M_CYCLE_ON_WR_PORT); + if (IS_CNL_REVID(i915, CNL_REVID_B0, CNL_REVID_B0)) + wa_write_or(wal, + GAMT_CHKN_BIT_REG, + GAMT_CHKN_DISABLE_I2M_CYCLE_ON_WR_PORT); /* WaInPlaceDecompressionHang:cnl */ - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, - I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); - - /* WaEnablePreemptionGranularityControlByUMD:cnl */ - I915_WRITE(GEN7_FF_SLICE_CS_CHICKEN1, - _MASKED_BIT_ENABLE(GEN9_FFSC_PERCTX_PREEMPT_CTRL)); + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); } -static void icl_gt_workarounds_apply(struct drm_i915_private *dev_priv) +static void icl_gt_workarounds_init(struct drm_i915_private *i915) { - wa_init_mcr(dev_priv); + struct i915_wa_list *wal = &i915->gt_wa_list; - /* This is not an Wa. Enable for better image quality */ - I915_WRITE(_3D_CHICKEN3, - _MASKED_BIT_ENABLE(_3D_CHICKEN3_AA_LINE_QUALITY_FIX_ENABLE)); + wa_init_mcr(i915); /* WaInPlaceDecompressionHang:icl */ - I915_WRITE(GEN9_GAMT_ECO_REG_RW_IA, - I915_READ(GEN9_GAMT_ECO_REG_RW_IA) | - GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); - - /* WaPipelineFlushCoherentLines:icl */ - I915_WRITE(GEN8_L3SQCREG4, - I915_READ(GEN8_L3SQCREG4) | - GEN8_LQSC_FLUSH_COHERENT_LINES); - - /* Wa_1405543622:icl - * Formerly known as WaGAPZPriorityScheme - */ - I915_WRITE(GEN8_GARBCNTL, - I915_READ(GEN8_GARBCNTL) | - GEN11_ARBITRATION_PRIO_ORDER_MASK); - - /* Wa_1604223664:icl - * Formerly known as WaL3BankAddressHashing - */ - I915_WRITE(GEN8_GARBCNTL, - (I915_READ(GEN8_GARBCNTL) & ~GEN11_HASH_CTRL_EXCL_MASK) | - GEN11_HASH_CTRL_EXCL_BIT0); - I915_WRITE(GEN11_GLBLINVL, - (I915_READ(GEN11_GLBLINVL) & ~GEN11_BANK_HASH_ADDR_EXCL_MASK) | - GEN11_BANK_HASH_ADDR_EXCL_BIT0); + wa_write_or(wal, + GEN9_GAMT_ECO_REG_RW_IA, + GAMT_ECO_ENABLE_IN_PLACE_DECOMPRESS); /* WaModifyGamTlbPartitioning:icl */ - I915_WRITE(GEN11_GACB_PERF_CTRL, - (I915_READ(GEN11_GACB_PERF_CTRL) & ~GEN11_HASH_CTRL_MASK) | - GEN11_HASH_CTRL_BIT0 | GEN11_HASH_CTRL_BIT4); - - /* Wa_1405733216:icl - * Formerly known as WaDisableCleanEvicts - */ - I915_WRITE(GEN8_L3SQCREG4, - I915_READ(GEN8_L3SQCREG4) | - GEN11_LQSC_CLEAN_EVICT_DISABLE); + wa_write_masked_or(wal, + GEN11_GACB_PERF_CTRL, + GEN11_HASH_CTRL_MASK, + GEN11_HASH_CTRL_BIT0 | GEN11_HASH_CTRL_BIT4); /* Wa_1405766107:icl * Formerly known as WaCL2SFHalfMaxAlloc */ - I915_WRITE(GEN11_LSN_UNSLCVC, - I915_READ(GEN11_LSN_UNSLCVC) | - GEN11_LSN_UNSLCVC_GAFS_HALF_SF_MAXALLOC | - GEN11_LSN_UNSLCVC_GAFS_HALF_CL2_MAXALLOC); + wa_write_or(wal, + GEN11_LSN_UNSLCVC, + GEN11_LSN_UNSLCVC_GAFS_HALF_SF_MAXALLOC | + GEN11_LSN_UNSLCVC_GAFS_HALF_CL2_MAXALLOC); /* Wa_220166154:icl * Formerly known as WaDisCtxReload */ - I915_WRITE(GEN8_GAMW_ECO_DEV_RW_IA, - I915_READ(GEN8_GAMW_ECO_DEV_RW_IA) | - GAMW_ECO_DEV_CTX_RELOAD_DISABLE); + wa_write_or(wal, + GEN8_GAMW_ECO_DEV_RW_IA, + GAMW_ECO_DEV_CTX_RELOAD_DISABLE); /* Wa_1405779004:icl (pre-prod) */ - if (IS_ICL_REVID(dev_priv, ICL_REVID_A0, ICL_REVID_A0)) - I915_WRITE(SLICE_UNIT_LEVEL_CLKGATE, - I915_READ(SLICE_UNIT_LEVEL_CLKGATE) | - MSCUNIT_CLKGATE_DIS); + if (IS_ICL_REVID(i915, ICL_REVID_A0, ICL_REVID_A0)) + wa_write_or(wal, + SLICE_UNIT_LEVEL_CLKGATE, + MSCUNIT_CLKGATE_DIS); /* Wa_1406680159:icl */ - I915_WRITE(SUBSLICE_UNIT_LEVEL_CLKGATE, - I915_READ(SUBSLICE_UNIT_LEVEL_CLKGATE) | - GWUNIT_CLKGATE_DIS); - - /* Wa_1604302699:icl */ - I915_WRITE(GEN10_L3_CHICKEN_MODE_REGISTER, - I915_READ(GEN10_L3_CHICKEN_MODE_REGISTER) | - GEN11_I2M_WRITE_DISABLE); + wa_write_or(wal, + SUBSLICE_UNIT_LEVEL_CLKGATE, + GWUNIT_CLKGATE_DIS); /* Wa_1406838659:icl (pre-prod) */ - if (IS_ICL_REVID(dev_priv, ICL_REVID_A0, ICL_REVID_B0)) - I915_WRITE(INF_UNIT_LEVEL_CLKGATE, - I915_READ(INF_UNIT_LEVEL_CLKGATE) | - CGPSF_CLKGATE_DIS); - - /* WaForwardProgressSoftReset:icl */ - I915_WRITE(GEN10_SCRATCH_LNCF2, - I915_READ(GEN10_SCRATCH_LNCF2) | - PMFLUSHDONE_LNICRSDROP | - PMFLUSH_GAPL3UNBLOCK | - PMFLUSHDONE_LNEBLK); + if (IS_ICL_REVID(i915, ICL_REVID_A0, ICL_REVID_B0)) + wa_write_or(wal, + INF_UNIT_LEVEL_CLKGATE, + CGPSF_CLKGATE_DIS); /* Wa_1406463099:icl * Formerly known as WaGamTlbPendError */ - I915_WRITE(GAMT_CHKN_BIT_REG, - I915_READ(GAMT_CHKN_BIT_REG) | - GAMT_CHKN_DISABLE_L3_COH_PIPE); - - /* Wa_1406609255:icl (pre-prod) */ - if (IS_ICL_REVID(dev_priv, ICL_REVID_A0, ICL_REVID_B0)) - I915_WRITE(GEN7_SARCHKMD, - I915_READ(GEN7_SARCHKMD) | - GEN7_DISABLE_DEMAND_PREFETCH | - GEN7_DISABLE_SAMPLER_PREFETCH); + wa_write_or(wal, + GAMT_CHKN_BIT_REG, + GAMT_CHKN_DISABLE_L3_COH_PIPE); } -void intel_gt_workarounds_apply(struct drm_i915_private *dev_priv) +void intel_gt_init_workarounds(struct drm_i915_private *i915) { - if (INTEL_GEN(dev_priv) < 8) + struct i915_wa_list *wal = &i915->gt_wa_list; + + wa_init_start(wal, "GT"); + + if (INTEL_GEN(i915) < 8) + return; + else if (IS_BROADWELL(i915)) return; - else if (IS_BROADWELL(dev_priv)) - bdw_gt_workarounds_apply(dev_priv); - else if (IS_CHERRYVIEW(dev_priv)) - chv_gt_workarounds_apply(dev_priv); - else if (IS_SKYLAKE(dev_priv)) - skl_gt_workarounds_apply(dev_priv); - else if (IS_BROXTON(dev_priv)) - bxt_gt_workarounds_apply(dev_priv); - else if (IS_KABYLAKE(dev_priv)) - kbl_gt_workarounds_apply(dev_priv); - else if (IS_GEMINILAKE(dev_priv)) - glk_gt_workarounds_apply(dev_priv); - else if (IS_COFFEELAKE(dev_priv)) - cfl_gt_workarounds_apply(dev_priv); - else if (IS_CANNONLAKE(dev_priv)) - cnl_gt_workarounds_apply(dev_priv); - else if (IS_ICELAKE(dev_priv)) - icl_gt_workarounds_apply(dev_priv); + else if (IS_CHERRYVIEW(i915)) + return; + else if (IS_SKYLAKE(i915)) + skl_gt_workarounds_init(i915); + else if (IS_BROXTON(i915)) + bxt_gt_workarounds_init(i915); + else if (IS_KABYLAKE(i915)) + kbl_gt_workarounds_init(i915); + else if (IS_GEMINILAKE(i915)) + glk_gt_workarounds_init(i915); + else if (IS_COFFEELAKE(i915)) + cfl_gt_workarounds_init(i915); + else if (IS_CANNONLAKE(i915)) + cnl_gt_workarounds_init(i915); + else if (IS_ICELAKE(i915)) + icl_gt_workarounds_init(i915); else - MISSING_CASE(INTEL_GEN(dev_priv)); + MISSING_CASE(INTEL_GEN(i915)); + + wa_init_finish(wal); } -struct whitelist { - i915_reg_t reg[RING_MAX_NONPRIV_SLOTS]; - unsigned int count; - u32 nopid; -}; +static enum forcewake_domains +wal_get_fw_for_rmw(struct drm_i915_private *dev_priv, + const struct i915_wa_list *wal) +{ + enum forcewake_domains fw = 0; + struct i915_wa *wa; + unsigned int i; -static void whitelist_reg(struct whitelist *w, i915_reg_t reg) + for (i = 0, wa = wal->list; i < wal->count; i++, wa++) + fw |= intel_uncore_forcewake_for_reg(dev_priv, + wa->reg, + FW_REG_READ | + FW_REG_WRITE); + + return fw; +} + +static void +wa_list_apply(struct drm_i915_private *dev_priv, const struct i915_wa_list *wal) { - if (GEM_DEBUG_WARN_ON(w->count >= RING_MAX_NONPRIV_SLOTS)) + enum forcewake_domains fw; + unsigned long flags; + struct i915_wa *wa; + unsigned int i; + + if (!wal->count) return; - w->reg[w->count++] = reg; + fw = wal_get_fw_for_rmw(dev_priv, wal); + + spin_lock_irqsave(&dev_priv->uncore.lock, flags); + intel_uncore_forcewake_get__locked(dev_priv, fw); + + for (i = 0, wa = wal->list; i < wal->count; i++, wa++) { + u32 val = I915_READ_FW(wa->reg); + + val &= ~wa->mask; + val |= wa->val; + + I915_WRITE_FW(wa->reg, val); + } + + intel_uncore_forcewake_put__locked(dev_priv, fw); + spin_unlock_irqrestore(&dev_priv->uncore.lock, flags); + + DRM_DEBUG_DRIVER("Applied %u %s workarounds\n", wal->count, wal->name); +} + +void intel_gt_apply_workarounds(struct drm_i915_private *dev_priv) +{ + wa_list_apply(dev_priv, &dev_priv->gt_wa_list); +} + +static bool +wa_verify(const struct i915_wa *wa, u32 cur, const char *name, const char *from) +{ + if ((cur ^ wa->val) & wa->mask) { + DRM_ERROR("%s workaround lost on %s! (%x=%x/%x, expected %x, mask=%x)\n", + name, from, i915_mmio_reg_offset(wa->reg), cur, + cur & wa->mask, wa->val, wa->mask); + + return false; + } + + return true; } -static void bdw_whitelist_build(struct whitelist *w) +static bool wa_list_verify(struct drm_i915_private *dev_priv, + const struct i915_wa_list *wal, + const char *from) { + struct i915_wa *wa; + unsigned int i; + bool ok = true; + + for (i = 0, wa = wal->list; i < wal->count; i++, wa++) + ok &= wa_verify(wa, I915_READ(wa->reg), wal->name, from); + + return ok; } -static void chv_whitelist_build(struct whitelist *w) +bool intel_gt_verify_workarounds(struct drm_i915_private *dev_priv, + const char *from) { + return wa_list_verify(dev_priv, &dev_priv->gt_wa_list, from); } -static void gen9_whitelist_build(struct whitelist *w) +static void +whitelist_reg(struct i915_wa_list *wal, i915_reg_t reg) +{ + struct i915_wa wa = { + .reg = reg + }; + + if (GEM_DEBUG_WARN_ON(wal->count >= RING_MAX_NONPRIV_SLOTS)) + return; + + _wa_add(wal, &wa); +} + +static void gen9_whitelist_build(struct i915_wa_list *w) { /* WaVFEStateAfterPipeControlwithMediaStateClear:skl,bxt,glk,cfl */ whitelist_reg(w, GEN9_CTX_PREEMPT_REG); @@ -980,7 +1023,7 @@ static void gen9_whitelist_build(struct whitelist *w) whitelist_reg(w, GEN8_HDC_CHICKEN1); } -static void skl_whitelist_build(struct whitelist *w) +static void skl_whitelist_build(struct i915_wa_list *w) { gen9_whitelist_build(w); @@ -988,12 +1031,12 @@ static void skl_whitelist_build(struct whitelist *w) whitelist_reg(w, GEN8_L3SQCREG4); } -static void bxt_whitelist_build(struct whitelist *w) +static void bxt_whitelist_build(struct i915_wa_list *w) { gen9_whitelist_build(w); } -static void kbl_whitelist_build(struct whitelist *w) +static void kbl_whitelist_build(struct i915_wa_list *w) { gen9_whitelist_build(w); @@ -1001,7 +1044,7 @@ static void kbl_whitelist_build(struct whitelist *w) whitelist_reg(w, GEN8_L3SQCREG4); } -static void glk_whitelist_build(struct whitelist *w) +static void glk_whitelist_build(struct i915_wa_list *w) { gen9_whitelist_build(w); @@ -1009,18 +1052,18 @@ static void glk_whitelist_build(struct whitelist *w) whitelist_reg(w, GEN9_SLICE_COMMON_ECO_CHICKEN1); } -static void cfl_whitelist_build(struct whitelist *w) +static void cfl_whitelist_build(struct i915_wa_list *w) { gen9_whitelist_build(w); } -static void cnl_whitelist_build(struct whitelist *w) +static void cnl_whitelist_build(struct i915_wa_list *w) { /* WaEnablePreemptionGranularityControlByUMD:cnl */ whitelist_reg(w, GEN8_CS_CHICKEN1); } -static void icl_whitelist_build(struct whitelist *w) +static void icl_whitelist_build(struct i915_wa_list *w) { /* WaAllowUMDToModifyHalfSliceChicken7:icl */ whitelist_reg(w, GEN9_HALF_SLICE_CHICKEN7); @@ -1029,22 +1072,21 @@ static void icl_whitelist_build(struct whitelist *w) whitelist_reg(w, GEN10_SAMPLER_MODE); } -static struct whitelist *whitelist_build(struct intel_engine_cs *engine, - struct whitelist *w) +void intel_engine_init_whitelist(struct intel_engine_cs *engine) { struct drm_i915_private *i915 = engine->i915; + struct i915_wa_list *w = &engine->whitelist; GEM_BUG_ON(engine->id != RCS); - w->count = 0; - w->nopid = i915_mmio_reg_offset(RING_NOPID(engine->mmio_base)); + wa_init_start(w, "whitelist"); if (INTEL_GEN(i915) < 8) - return NULL; + return; else if (IS_BROADWELL(i915)) - bdw_whitelist_build(w); + return; else if (IS_CHERRYVIEW(i915)) - chv_whitelist_build(w); + return; else if (IS_SKYLAKE(i915)) skl_whitelist_build(w); else if (IS_BROXTON(i915)) @@ -1062,39 +1104,180 @@ static struct whitelist *whitelist_build(struct intel_engine_cs *engine, else MISSING_CASE(INTEL_GEN(i915)); - return w; + wa_init_finish(w); } -static void whitelist_apply(struct intel_engine_cs *engine, - const struct whitelist *w) +void intel_engine_apply_whitelist(struct intel_engine_cs *engine) { struct drm_i915_private *dev_priv = engine->i915; + const struct i915_wa_list *wal = &engine->whitelist; const u32 base = engine->mmio_base; + struct i915_wa *wa; unsigned int i; - if (!w) + if (!wal->count) return; - intel_uncore_forcewake_get(engine->i915, FORCEWAKE_ALL); - - for (i = 0; i < w->count; i++) - I915_WRITE_FW(RING_FORCE_TO_NONPRIV(base, i), - i915_mmio_reg_offset(w->reg[i])); + for (i = 0, wa = wal->list; i < wal->count; i++, wa++) + I915_WRITE(RING_FORCE_TO_NONPRIV(base, i), + i915_mmio_reg_offset(wa->reg)); /* And clear the rest just in case of garbage */ for (; i < RING_MAX_NONPRIV_SLOTS; i++) - I915_WRITE_FW(RING_FORCE_TO_NONPRIV(base, i), w->nopid); + I915_WRITE(RING_FORCE_TO_NONPRIV(base, i), + i915_mmio_reg_offset(RING_NOPID(base))); - intel_uncore_forcewake_put(engine->i915, FORCEWAKE_ALL); + DRM_DEBUG_DRIVER("Applied %u %s workarounds\n", wal->count, wal->name); } -void intel_whitelist_workarounds_apply(struct intel_engine_cs *engine) +static void rcs_engine_wa_init(struct intel_engine_cs *engine) { - struct whitelist w; + struct drm_i915_private *i915 = engine->i915; + struct i915_wa_list *wal = &engine->wa_list; + + if (IS_ICELAKE(i915)) { + /* This is not an Wa. Enable for better image quality */ + wa_masked_en(wal, + _3D_CHICKEN3, + _3D_CHICKEN3_AA_LINE_QUALITY_FIX_ENABLE); + + /* WaPipelineFlushCoherentLines:icl */ + wa_write_or(wal, + GEN8_L3SQCREG4, + GEN8_LQSC_FLUSH_COHERENT_LINES); + + /* + * Wa_1405543622:icl + * Formerly known as WaGAPZPriorityScheme + */ + wa_write_or(wal, + GEN8_GARBCNTL, + GEN11_ARBITRATION_PRIO_ORDER_MASK); + + /* + * Wa_1604223664:icl + * Formerly known as WaL3BankAddressHashing + */ + wa_write_masked_or(wal, + GEN8_GARBCNTL, + GEN11_HASH_CTRL_EXCL_MASK, + GEN11_HASH_CTRL_EXCL_BIT0); + wa_write_masked_or(wal, + GEN11_GLBLINVL, + GEN11_BANK_HASH_ADDR_EXCL_MASK, + GEN11_BANK_HASH_ADDR_EXCL_BIT0); - whitelist_apply(engine, whitelist_build(engine, &w)); + /* + * Wa_1405733216:icl + * Formerly known as WaDisableCleanEvicts + */ + wa_write_or(wal, + GEN8_L3SQCREG4, + GEN11_LQSC_CLEAN_EVICT_DISABLE); + + /* WaForwardProgressSoftReset:icl */ + wa_write_or(wal, + GEN10_SCRATCH_LNCF2, + PMFLUSHDONE_LNICRSDROP | + PMFLUSH_GAPL3UNBLOCK | + PMFLUSHDONE_LNEBLK); + + /* Wa_1406609255:icl (pre-prod) */ + if (IS_ICL_REVID(i915, ICL_REVID_A0, ICL_REVID_B0)) + wa_write_or(wal, + GEN7_SARCHKMD, + GEN7_DISABLE_DEMAND_PREFETCH | + GEN7_DISABLE_SAMPLER_PREFETCH); + } + + if (IS_GEN9(i915) || IS_CANNONLAKE(i915)) { + /* WaEnablePreemptionGranularityControlByUMD:skl,bxt,kbl,cfl,cnl */ + wa_masked_en(wal, + GEN7_FF_SLICE_CS_CHICKEN1, + GEN9_FFSC_PERCTX_PREEMPT_CTRL); + } + + if (IS_SKYLAKE(i915) || IS_KABYLAKE(i915) || IS_COFFEELAKE(i915)) { + /* WaEnableGapsTsvCreditFix:skl,kbl,cfl */ + wa_write_or(wal, + GEN8_GARBCNTL, + GEN9_GAPS_TSV_CREDIT_DISABLE); + } + + if (IS_BROXTON(i915)) { + /* WaDisablePooledEuLoadBalancingFix:bxt */ + wa_masked_en(wal, + FF_SLICE_CS_CHICKEN2, + GEN9_POOLED_EU_LOAD_BALANCING_FIX_DISABLE); + } + + if (IS_GEN9(i915)) { + /* WaContextSwitchWithConcurrentTLBInvalidate:skl,bxt,kbl,glk,cfl */ + wa_masked_en(wal, + GEN9_CSFE_CHICKEN1_RCS, + GEN9_PREEMPT_GPGPU_SYNC_SWITCH_DISABLE); + + /* WaEnableLbsSlaRetryTimerDecrement:skl,bxt,kbl,glk,cfl */ + wa_write_or(wal, + BDW_SCRATCH1, + GEN9_LBS_SLA_RETRY_TIMER_DECREMENT_ENABLE); + + /* WaProgramL3SqcReg1DefaultForPerf:bxt,glk */ + if (IS_GEN9_LP(i915)) + wa_write_masked_or(wal, + GEN8_L3SQCREG1, + L3_PRIO_CREDITS_MASK, + L3_GENERAL_PRIO_CREDITS(62) | + L3_HIGH_PRIO_CREDITS(2)); + + /* WaOCLCoherentLineFlush:skl,bxt,kbl,cfl */ + wa_write_or(wal, + GEN8_L3SQCREG4, + GEN8_LQSC_FLUSH_COHERENT_LINES); + } +} + +static void xcs_engine_wa_init(struct intel_engine_cs *engine) +{ + struct drm_i915_private *i915 = engine->i915; + struct i915_wa_list *wal = &engine->wa_list; + + /* WaKBLVECSSemaphoreWaitPoll:kbl */ + if (IS_KBL_REVID(i915, KBL_REVID_A0, KBL_REVID_E0)) { + wa_write(wal, + RING_SEMA_WAIT_POLL(engine->mmio_base), + 1); + } +} + +void intel_engine_init_workarounds(struct intel_engine_cs *engine) +{ + struct i915_wa_list *wal = &engine->wa_list; + + if (GEM_WARN_ON(INTEL_GEN(engine->i915) < 8)) + return; + + wa_init_start(wal, engine->name); + + if (engine->id == RCS) + rcs_engine_wa_init(engine); + else + xcs_engine_wa_init(engine); + + wa_init_finish(wal); +} + +void intel_engine_apply_workarounds(struct intel_engine_cs *engine) +{ + wa_list_apply(engine->i915, &engine->wa_list); } #if IS_ENABLED(CONFIG_DRM_I915_SELFTEST) +static bool intel_engine_verify_workarounds(struct intel_engine_cs *engine, + const char *from) +{ + return wa_list_verify(engine->i915, &engine->wa_list, from); +} + #include "selftests/intel_workarounds.c" #endif diff --git a/drivers/gpu/drm/i915/intel_workarounds.h b/drivers/gpu/drm/i915/intel_workarounds.h index b11d0623e626..7c734714b05e 100644 --- a/drivers/gpu/drm/i915/intel_workarounds.h +++ b/drivers/gpu/drm/i915/intel_workarounds.h @@ -7,11 +7,39 @@ #ifndef _I915_WORKAROUNDS_H_ #define _I915_WORKAROUNDS_H_ -int intel_ctx_workarounds_init(struct drm_i915_private *dev_priv); -int intel_ctx_workarounds_emit(struct i915_request *rq); +#include <linux/slab.h> -void intel_gt_workarounds_apply(struct drm_i915_private *dev_priv); +struct i915_wa { + i915_reg_t reg; + u32 mask; + u32 val; +}; -void intel_whitelist_workarounds_apply(struct intel_engine_cs *engine); +struct i915_wa_list { + const char *name; + struct i915_wa *list; + unsigned int count; + unsigned int wa_count; +}; + +static inline void intel_wa_list_free(struct i915_wa_list *wal) +{ + kfree(wal->list); + memset(wal, 0, sizeof(*wal)); +} + +void intel_engine_init_ctx_wa(struct intel_engine_cs *engine); +int intel_engine_emit_ctx_wa(struct i915_request *rq); + +void intel_gt_init_workarounds(struct drm_i915_private *dev_priv); +void intel_gt_apply_workarounds(struct drm_i915_private *dev_priv); +bool intel_gt_verify_workarounds(struct drm_i915_private *dev_priv, + const char *from); + +void intel_engine_init_whitelist(struct intel_engine_cs *engine); +void intel_engine_apply_whitelist(struct intel_engine_cs *engine); + +void intel_engine_init_workarounds(struct intel_engine_cs *engine); +void intel_engine_apply_workarounds(struct intel_engine_cs *engine); #endif diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c index 69fe86b30fbb..a9ed0ecc94e2 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c @@ -170,7 +170,7 @@ static int igt_ppgtt_alloc(void *arg) * This should ensure that we do not run into the oomkiller during * the test and take down the machine wilfully. */ - limit = totalram_pages << PAGE_SHIFT; + limit = totalram_pages() << PAGE_SHIFT; limit = min(ppgtt->vm.total, limit); /* Check we can allocate the entire range */ @@ -1244,7 +1244,7 @@ static int exercise_mock(struct drm_i915_private *i915, u64 hole_start, u64 hole_end, unsigned long end_time)) { - const u64 limit = totalram_pages << PAGE_SHIFT; + const u64 limit = totalram_pages() << PAGE_SHIFT; struct i915_gem_context *ctx; struct i915_hw_ppgtt *ppgtt; IGT_TIMEOUT(end_time); diff --git a/drivers/gpu/drm/i915/selftests/igt_reset.c b/drivers/gpu/drm/i915/selftests/igt_reset.c new file mode 100644 index 000000000000..208a966da8ca --- /dev/null +++ b/drivers/gpu/drm/i915/selftests/igt_reset.c @@ -0,0 +1,44 @@ +/* + * SPDX-License-Identifier: MIT + * + * Copyright © 2018 Intel Corporation + */ + +#include "igt_reset.h" + +#include "../i915_drv.h" +#include "../intel_ringbuffer.h" + +void igt_global_reset_lock(struct drm_i915_private *i915) +{ + struct intel_engine_cs *engine; + enum intel_engine_id id; + + pr_debug("%s: current gpu_error=%08lx\n", + __func__, i915->gpu_error.flags); + + while (test_and_set_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags)) + wait_event(i915->gpu_error.reset_queue, + !test_bit(I915_RESET_BACKOFF, + &i915->gpu_error.flags)); + + for_each_engine(engine, i915, id) { + while (test_and_set_bit(I915_RESET_ENGINE + id, + &i915->gpu_error.flags)) + wait_on_bit(&i915->gpu_error.flags, + I915_RESET_ENGINE + id, + TASK_UNINTERRUPTIBLE); + } +} + +void igt_global_reset_unlock(struct drm_i915_private *i915) +{ + struct intel_engine_cs *engine; + enum intel_engine_id id; + + for_each_engine(engine, i915, id) + clear_bit(I915_RESET_ENGINE + id, &i915->gpu_error.flags); + + clear_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); + wake_up_all(&i915->gpu_error.reset_queue); +} diff --git a/drivers/gpu/drm/i915/selftests/igt_reset.h b/drivers/gpu/drm/i915/selftests/igt_reset.h new file mode 100644 index 000000000000..5f0234d045d5 --- /dev/null +++ b/drivers/gpu/drm/i915/selftests/igt_reset.h @@ -0,0 +1,15 @@ +/* + * SPDX-License-Identifier: MIT + * + * Copyright © 2018 Intel Corporation + */ + +#ifndef __I915_SELFTESTS_IGT_RESET_H__ +#define __I915_SELFTESTS_IGT_RESET_H__ + +#include "../i915_drv.h" + +void igt_global_reset_lock(struct drm_i915_private *i915); +void igt_global_reset_unlock(struct drm_i915_private *i915); + +#endif diff --git a/drivers/gpu/drm/i915/selftests/igt_spinner.c b/drivers/gpu/drm/i915/selftests/igt_spinner.c new file mode 100644 index 000000000000..8cd34f6e6859 --- /dev/null +++ b/drivers/gpu/drm/i915/selftests/igt_spinner.c @@ -0,0 +1,199 @@ +/* + * SPDX-License-Identifier: MIT + * + * Copyright © 2018 Intel Corporation + */ + +#include "igt_spinner.h" + +int igt_spinner_init(struct igt_spinner *spin, struct drm_i915_private *i915) +{ + unsigned int mode; + void *vaddr; + int err; + + GEM_BUG_ON(INTEL_GEN(i915) < 8); + + memset(spin, 0, sizeof(*spin)); + spin->i915 = i915; + + spin->hws = i915_gem_object_create_internal(i915, PAGE_SIZE); + if (IS_ERR(spin->hws)) { + err = PTR_ERR(spin->hws); + goto err; + } + + spin->obj = i915_gem_object_create_internal(i915, PAGE_SIZE); + if (IS_ERR(spin->obj)) { + err = PTR_ERR(spin->obj); + goto err_hws; + } + + i915_gem_object_set_cache_level(spin->hws, I915_CACHE_LLC); + vaddr = i915_gem_object_pin_map(spin->hws, I915_MAP_WB); + if (IS_ERR(vaddr)) { + err = PTR_ERR(vaddr); + goto err_obj; + } + spin->seqno = memset(vaddr, 0xff, PAGE_SIZE); + + mode = i915_coherent_map_type(i915); + vaddr = i915_gem_object_pin_map(spin->obj, mode); + if (IS_ERR(vaddr)) { + err = PTR_ERR(vaddr); + goto err_unpin_hws; + } + spin->batch = vaddr; + + return 0; + +err_unpin_hws: + i915_gem_object_unpin_map(spin->hws); +err_obj: + i915_gem_object_put(spin->obj); +err_hws: + i915_gem_object_put(spin->hws); +err: + return err; +} + +static unsigned int seqno_offset(u64 fence) +{ + return offset_in_page(sizeof(u32) * fence); +} + +static u64 hws_address(const struct i915_vma *hws, + const struct i915_request *rq) +{ + return hws->node.start + seqno_offset(rq->fence.context); +} + +static int emit_recurse_batch(struct igt_spinner *spin, + struct i915_request *rq, + u32 arbitration_command) +{ + struct i915_address_space *vm = &rq->gem_context->ppgtt->vm; + struct i915_vma *hws, *vma; + u32 *batch; + int err; + + vma = i915_vma_instance(spin->obj, vm, NULL); + if (IS_ERR(vma)) + return PTR_ERR(vma); + + hws = i915_vma_instance(spin->hws, vm, NULL); + if (IS_ERR(hws)) + return PTR_ERR(hws); + + err = i915_vma_pin(vma, 0, 0, PIN_USER); + if (err) + return err; + + err = i915_vma_pin(hws, 0, 0, PIN_USER); + if (err) + goto unpin_vma; + + err = i915_vma_move_to_active(vma, rq, 0); + if (err) + goto unpin_hws; + + if (!i915_gem_object_has_active_reference(vma->obj)) { + i915_gem_object_get(vma->obj); + i915_gem_object_set_active_reference(vma->obj); + } + + err = i915_vma_move_to_active(hws, rq, 0); + if (err) + goto unpin_hws; + + if (!i915_gem_object_has_active_reference(hws->obj)) { + i915_gem_object_get(hws->obj); + i915_gem_object_set_active_reference(hws->obj); + } + + batch = spin->batch; + + *batch++ = MI_STORE_DWORD_IMM_GEN4; + *batch++ = lower_32_bits(hws_address(hws, rq)); + *batch++ = upper_32_bits(hws_address(hws, rq)); + *batch++ = rq->fence.seqno; + + *batch++ = arbitration_command; + + *batch++ = MI_BATCH_BUFFER_START | 1 << 8 | 1; + *batch++ = lower_32_bits(vma->node.start); + *batch++ = upper_32_bits(vma->node.start); + *batch++ = MI_BATCH_BUFFER_END; /* not reached */ + + i915_gem_chipset_flush(spin->i915); + + err = rq->engine->emit_bb_start(rq, vma->node.start, PAGE_SIZE, 0); + +unpin_hws: + i915_vma_unpin(hws); +unpin_vma: + i915_vma_unpin(vma); + return err; +} + +struct i915_request * +igt_spinner_create_request(struct igt_spinner *spin, + struct i915_gem_context *ctx, + struct intel_engine_cs *engine, + u32 arbitration_command) +{ + struct i915_request *rq; + int err; + + rq = i915_request_alloc(engine, ctx); + if (IS_ERR(rq)) + return rq; + + err = emit_recurse_batch(spin, rq, arbitration_command); + if (err) { + i915_request_add(rq); + return ERR_PTR(err); + } + + return rq; +} + +static u32 +hws_seqno(const struct igt_spinner *spin, const struct i915_request *rq) +{ + u32 *seqno = spin->seqno + seqno_offset(rq->fence.context); + + return READ_ONCE(*seqno); +} + +void igt_spinner_end(struct igt_spinner *spin) +{ + *spin->batch = MI_BATCH_BUFFER_END; + i915_gem_chipset_flush(spin->i915); +} + +void igt_spinner_fini(struct igt_spinner *spin) +{ + igt_spinner_end(spin); + + i915_gem_object_unpin_map(spin->obj); + i915_gem_object_put(spin->obj); + + i915_gem_object_unpin_map(spin->hws); + i915_gem_object_put(spin->hws); +} + +bool igt_wait_for_spinner(struct igt_spinner *spin, struct i915_request *rq) +{ + if (!wait_event_timeout(rq->execute, + READ_ONCE(rq->global_seqno), + msecs_to_jiffies(10))) + return false; + + return !(wait_for_us(i915_seqno_passed(hws_seqno(spin, rq), + rq->fence.seqno), + 10) && + wait_for(i915_seqno_passed(hws_seqno(spin, rq), + rq->fence.seqno), + 1000)); +} diff --git a/drivers/gpu/drm/i915/selftests/igt_spinner.h b/drivers/gpu/drm/i915/selftests/igt_spinner.h new file mode 100644 index 000000000000..391777c76dc7 --- /dev/null +++ b/drivers/gpu/drm/i915/selftests/igt_spinner.h @@ -0,0 +1,37 @@ +/* + * SPDX-License-Identifier: MIT + * + * Copyright © 2018 Intel Corporation + */ + +#ifndef __I915_SELFTESTS_IGT_SPINNER_H__ +#define __I915_SELFTESTS_IGT_SPINNER_H__ + +#include "../i915_selftest.h" + +#include "../i915_drv.h" +#include "../i915_request.h" +#include "../intel_ringbuffer.h" +#include "../i915_gem_context.h" + +struct igt_spinner { + struct drm_i915_private *i915; + struct drm_i915_gem_object *hws; + struct drm_i915_gem_object *obj; + u32 *batch; + void *seqno; +}; + +int igt_spinner_init(struct igt_spinner *spin, struct drm_i915_private *i915); +void igt_spinner_fini(struct igt_spinner *spin); + +struct i915_request * +igt_spinner_create_request(struct igt_spinner *spin, + struct i915_gem_context *ctx, + struct intel_engine_cs *engine, + u32 arbitration_command); +void igt_spinner_end(struct igt_spinner *spin); + +bool igt_wait_for_spinner(struct igt_spinner *spin, struct i915_request *rq); + +#endif diff --git a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c index 33494d922fab..5910da3e7d79 100644 --- a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c +++ b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c @@ -27,6 +27,7 @@ #include "../i915_selftest.h" #include "i915_random.h" #include "igt_flush_test.h" +#include "igt_reset.h" #include "igt_wedge_me.h" #include "mock_context.h" @@ -308,6 +309,7 @@ static int igt_hang_sanitycheck(void *arg) goto unlock; for_each_engine(engine, i915, id) { + struct igt_wedge_me w; long timeout; if (!intel_engine_can_store_dword(engine)) @@ -328,9 +330,14 @@ static int igt_hang_sanitycheck(void *arg) i915_request_add(rq); - timeout = i915_request_wait(rq, - I915_WAIT_LOCKED, - MAX_SCHEDULE_TIMEOUT); + timeout = 0; + igt_wedge_on_timeout(&w, i915, HZ / 10 /* 100ms timeout*/) + timeout = i915_request_wait(rq, + I915_WAIT_LOCKED, + MAX_SCHEDULE_TIMEOUT); + if (i915_terminally_wedged(&i915->gpu_error)) + timeout = -EIO; + i915_request_put(rq); if (timeout < 0) { @@ -348,40 +355,6 @@ unlock: return err; } -static void global_reset_lock(struct drm_i915_private *i915) -{ - struct intel_engine_cs *engine; - enum intel_engine_id id; - - pr_debug("%s: current gpu_error=%08lx\n", - __func__, i915->gpu_error.flags); - - while (test_and_set_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags)) - wait_event(i915->gpu_error.reset_queue, - !test_bit(I915_RESET_BACKOFF, - &i915->gpu_error.flags)); - - for_each_engine(engine, i915, id) { - while (test_and_set_bit(I915_RESET_ENGINE + id, - &i915->gpu_error.flags)) - wait_on_bit(&i915->gpu_error.flags, - I915_RESET_ENGINE + id, - TASK_UNINTERRUPTIBLE); - } -} - -static void global_reset_unlock(struct drm_i915_private *i915) -{ - struct intel_engine_cs *engine; - enum intel_engine_id id; - - for_each_engine(engine, i915, id) - clear_bit(I915_RESET_ENGINE + id, &i915->gpu_error.flags); - - clear_bit(I915_RESET_BACKOFF, &i915->gpu_error.flags); - wake_up_all(&i915->gpu_error.reset_queue); -} - static int igt_global_reset(void *arg) { struct drm_i915_private *i915 = arg; @@ -390,7 +363,7 @@ static int igt_global_reset(void *arg) /* Check that we can issue a global GPU reset */ - global_reset_lock(i915); + igt_global_reset_lock(i915); set_bit(I915_RESET_HANDOFF, &i915->gpu_error.flags); mutex_lock(&i915->drm.struct_mutex); @@ -405,7 +378,7 @@ static int igt_global_reset(void *arg) mutex_unlock(&i915->drm.struct_mutex); GEM_BUG_ON(test_bit(I915_RESET_HANDOFF, &i915->gpu_error.flags)); - global_reset_unlock(i915); + igt_global_reset_unlock(i915); if (i915_terminally_wedged(&i915->gpu_error)) err = -EIO; @@ -936,7 +909,7 @@ static int igt_reset_wait(void *arg) /* Check that we detect a stuck waiter and issue a reset */ - global_reset_lock(i915); + igt_global_reset_lock(i915); mutex_lock(&i915->drm.struct_mutex); err = hang_init(&h, i915); @@ -988,7 +961,7 @@ fini: hang_fini(&h); unlock: mutex_unlock(&i915->drm.struct_mutex); - global_reset_unlock(i915); + igt_global_reset_unlock(i915); if (i915_terminally_wedged(&i915->gpu_error)) return -EIO; @@ -1066,7 +1039,7 @@ static int __igt_reset_evict_vma(struct drm_i915_private *i915, /* Check that we can recover an unbind stuck on a hanging request */ - global_reset_lock(i915); + igt_global_reset_lock(i915); mutex_lock(&i915->drm.struct_mutex); err = hang_init(&h, i915); @@ -1186,7 +1159,7 @@ fini: hang_fini(&h); unlock: mutex_unlock(&i915->drm.struct_mutex); - global_reset_unlock(i915); + igt_global_reset_unlock(i915); if (i915_terminally_wedged(&i915->gpu_error)) return -EIO; @@ -1266,7 +1239,7 @@ static int igt_reset_queue(void *arg) /* Check that we replay pending requests following a hang */ - global_reset_lock(i915); + igt_global_reset_lock(i915); mutex_lock(&i915->drm.struct_mutex); err = hang_init(&h, i915); @@ -1397,7 +1370,7 @@ fini: hang_fini(&h); unlock: mutex_unlock(&i915->drm.struct_mutex); - global_reset_unlock(i915); + igt_global_reset_unlock(i915); if (i915_terminally_wedged(&i915->gpu_error)) return -EIO; diff --git a/drivers/gpu/drm/i915/selftests/intel_lrc.c b/drivers/gpu/drm/i915/selftests/intel_lrc.c index 94fc0e5c8766..ca461e3a5f27 100644 --- a/drivers/gpu/drm/i915/selftests/intel_lrc.c +++ b/drivers/gpu/drm/i915/selftests/intel_lrc.c @@ -6,216 +6,18 @@ #include "../i915_selftest.h" #include "igt_flush_test.h" +#include "igt_spinner.h" #include "i915_random.h" #include "mock_context.h" -struct spinner { - struct drm_i915_private *i915; - struct drm_i915_gem_object *hws; - struct drm_i915_gem_object *obj; - u32 *batch; - void *seqno; -}; - -static int spinner_init(struct spinner *spin, struct drm_i915_private *i915) -{ - unsigned int mode; - void *vaddr; - int err; - - GEM_BUG_ON(INTEL_GEN(i915) < 8); - - memset(spin, 0, sizeof(*spin)); - spin->i915 = i915; - - spin->hws = i915_gem_object_create_internal(i915, PAGE_SIZE); - if (IS_ERR(spin->hws)) { - err = PTR_ERR(spin->hws); - goto err; - } - - spin->obj = i915_gem_object_create_internal(i915, PAGE_SIZE); - if (IS_ERR(spin->obj)) { - err = PTR_ERR(spin->obj); - goto err_hws; - } - - i915_gem_object_set_cache_level(spin->hws, I915_CACHE_LLC); - vaddr = i915_gem_object_pin_map(spin->hws, I915_MAP_WB); - if (IS_ERR(vaddr)) { - err = PTR_ERR(vaddr); - goto err_obj; - } - spin->seqno = memset(vaddr, 0xff, PAGE_SIZE); - - mode = i915_coherent_map_type(i915); - vaddr = i915_gem_object_pin_map(spin->obj, mode); - if (IS_ERR(vaddr)) { - err = PTR_ERR(vaddr); - goto err_unpin_hws; - } - spin->batch = vaddr; - - return 0; - -err_unpin_hws: - i915_gem_object_unpin_map(spin->hws); -err_obj: - i915_gem_object_put(spin->obj); -err_hws: - i915_gem_object_put(spin->hws); -err: - return err; -} - -static unsigned int seqno_offset(u64 fence) -{ - return offset_in_page(sizeof(u32) * fence); -} - -static u64 hws_address(const struct i915_vma *hws, - const struct i915_request *rq) -{ - return hws->node.start + seqno_offset(rq->fence.context); -} - -static int emit_recurse_batch(struct spinner *spin, - struct i915_request *rq, - u32 arbitration_command) -{ - struct i915_address_space *vm = &rq->gem_context->ppgtt->vm; - struct i915_vma *hws, *vma; - u32 *batch; - int err; - - vma = i915_vma_instance(spin->obj, vm, NULL); - if (IS_ERR(vma)) - return PTR_ERR(vma); - - hws = i915_vma_instance(spin->hws, vm, NULL); - if (IS_ERR(hws)) - return PTR_ERR(hws); - - err = i915_vma_pin(vma, 0, 0, PIN_USER); - if (err) - return err; - - err = i915_vma_pin(hws, 0, 0, PIN_USER); - if (err) - goto unpin_vma; - - err = i915_vma_move_to_active(vma, rq, 0); - if (err) - goto unpin_hws; - - if (!i915_gem_object_has_active_reference(vma->obj)) { - i915_gem_object_get(vma->obj); - i915_gem_object_set_active_reference(vma->obj); - } - - err = i915_vma_move_to_active(hws, rq, 0); - if (err) - goto unpin_hws; - - if (!i915_gem_object_has_active_reference(hws->obj)) { - i915_gem_object_get(hws->obj); - i915_gem_object_set_active_reference(hws->obj); - } - - batch = spin->batch; - - *batch++ = MI_STORE_DWORD_IMM_GEN4; - *batch++ = lower_32_bits(hws_address(hws, rq)); - *batch++ = upper_32_bits(hws_address(hws, rq)); - *batch++ = rq->fence.seqno; - - *batch++ = arbitration_command; - - *batch++ = MI_BATCH_BUFFER_START | 1 << 8 | 1; - *batch++ = lower_32_bits(vma->node.start); - *batch++ = upper_32_bits(vma->node.start); - *batch++ = MI_BATCH_BUFFER_END; /* not reached */ - - i915_gem_chipset_flush(spin->i915); - - err = rq->engine->emit_bb_start(rq, vma->node.start, PAGE_SIZE, 0); - -unpin_hws: - i915_vma_unpin(hws); -unpin_vma: - i915_vma_unpin(vma); - return err; -} - -static struct i915_request * -spinner_create_request(struct spinner *spin, - struct i915_gem_context *ctx, - struct intel_engine_cs *engine, - u32 arbitration_command) -{ - struct i915_request *rq; - int err; - - rq = i915_request_alloc(engine, ctx); - if (IS_ERR(rq)) - return rq; - - err = emit_recurse_batch(spin, rq, arbitration_command); - if (err) { - i915_request_add(rq); - return ERR_PTR(err); - } - - return rq; -} - -static u32 hws_seqno(const struct spinner *spin, const struct i915_request *rq) -{ - u32 *seqno = spin->seqno + seqno_offset(rq->fence.context); - - return READ_ONCE(*seqno); -} - -static void spinner_end(struct spinner *spin) -{ - *spin->batch = MI_BATCH_BUFFER_END; - i915_gem_chipset_flush(spin->i915); -} - -static void spinner_fini(struct spinner *spin) -{ - spinner_end(spin); - - i915_gem_object_unpin_map(spin->obj); - i915_gem_object_put(spin->obj); - - i915_gem_object_unpin_map(spin->hws); - i915_gem_object_put(spin->hws); -} - -static bool wait_for_spinner(struct spinner *spin, struct i915_request *rq) -{ - if (!wait_event_timeout(rq->execute, - READ_ONCE(rq->global_seqno), - msecs_to_jiffies(10))) - return false; - - return !(wait_for_us(i915_seqno_passed(hws_seqno(spin, rq), - rq->fence.seqno), - 10) && - wait_for(i915_seqno_passed(hws_seqno(spin, rq), - rq->fence.seqno), - 1000)); -} - static int live_sanitycheck(void *arg) { struct drm_i915_private *i915 = arg; struct intel_engine_cs *engine; struct i915_gem_context *ctx; enum intel_engine_id id; - struct spinner spin; + struct igt_spinner spin; int err = -ENOMEM; if (!HAS_LOGICAL_RING_CONTEXTS(i915)) @@ -224,7 +26,7 @@ static int live_sanitycheck(void *arg) mutex_lock(&i915->drm.struct_mutex); intel_runtime_pm_get(i915); - if (spinner_init(&spin, i915)) + if (igt_spinner_init(&spin, i915)) goto err_unlock; ctx = kernel_context(i915); @@ -234,14 +36,14 @@ static int live_sanitycheck(void *arg) for_each_engine(engine, i915, id) { struct i915_request *rq; - rq = spinner_create_request(&spin, ctx, engine, MI_NOOP); + rq = igt_spinner_create_request(&spin, ctx, engine, MI_NOOP); if (IS_ERR(rq)) { err = PTR_ERR(rq); goto err_ctx; } i915_request_add(rq); - if (!wait_for_spinner(&spin, rq)) { + if (!igt_wait_for_spinner(&spin, rq)) { GEM_TRACE("spinner failed to start\n"); GEM_TRACE_DUMP(); i915_gem_set_wedged(i915); @@ -249,7 +51,7 @@ static int live_sanitycheck(void *arg) goto err_ctx; } - spinner_end(&spin); + igt_spinner_end(&spin); if (igt_flush_test(i915, I915_WAIT_LOCKED)) { err = -EIO; goto err_ctx; @@ -260,7 +62,7 @@ static int live_sanitycheck(void *arg) err_ctx: kernel_context_close(ctx); err_spin: - spinner_fini(&spin); + igt_spinner_fini(&spin); err_unlock: igt_flush_test(i915, I915_WAIT_LOCKED); intel_runtime_pm_put(i915); @@ -272,7 +74,7 @@ static int live_preempt(void *arg) { struct drm_i915_private *i915 = arg; struct i915_gem_context *ctx_hi, *ctx_lo; - struct spinner spin_hi, spin_lo; + struct igt_spinner spin_hi, spin_lo; struct intel_engine_cs *engine; enum intel_engine_id id; int err = -ENOMEM; @@ -283,10 +85,10 @@ static int live_preempt(void *arg) mutex_lock(&i915->drm.struct_mutex); intel_runtime_pm_get(i915); - if (spinner_init(&spin_hi, i915)) + if (igt_spinner_init(&spin_hi, i915)) goto err_unlock; - if (spinner_init(&spin_lo, i915)) + if (igt_spinner_init(&spin_lo, i915)) goto err_spin_hi; ctx_hi = kernel_context(i915); @@ -304,15 +106,15 @@ static int live_preempt(void *arg) for_each_engine(engine, i915, id) { struct i915_request *rq; - rq = spinner_create_request(&spin_lo, ctx_lo, engine, - MI_ARB_CHECK); + rq = igt_spinner_create_request(&spin_lo, ctx_lo, engine, + MI_ARB_CHECK); if (IS_ERR(rq)) { err = PTR_ERR(rq); goto err_ctx_lo; } i915_request_add(rq); - if (!wait_for_spinner(&spin_lo, rq)) { + if (!igt_wait_for_spinner(&spin_lo, rq)) { GEM_TRACE("lo spinner failed to start\n"); GEM_TRACE_DUMP(); i915_gem_set_wedged(i915); @@ -320,16 +122,16 @@ static int live_preempt(void *arg) goto err_ctx_lo; } - rq = spinner_create_request(&spin_hi, ctx_hi, engine, - MI_ARB_CHECK); + rq = igt_spinner_create_request(&spin_hi, ctx_hi, engine, + MI_ARB_CHECK); if (IS_ERR(rq)) { - spinner_end(&spin_lo); + igt_spinner_end(&spin_lo); err = PTR_ERR(rq); goto err_ctx_lo; } i915_request_add(rq); - if (!wait_for_spinner(&spin_hi, rq)) { + if (!igt_wait_for_spinner(&spin_hi, rq)) { GEM_TRACE("hi spinner failed to start\n"); GEM_TRACE_DUMP(); i915_gem_set_wedged(i915); @@ -337,8 +139,8 @@ static int live_preempt(void *arg) goto err_ctx_lo; } - spinner_end(&spin_hi); - spinner_end(&spin_lo); + igt_spinner_end(&spin_hi); + igt_spinner_end(&spin_lo); if (igt_flush_test(i915, I915_WAIT_LOCKED)) { err = -EIO; goto err_ctx_lo; @@ -351,9 +153,9 @@ err_ctx_lo: err_ctx_hi: kernel_context_close(ctx_hi); err_spin_lo: - spinner_fini(&spin_lo); + igt_spinner_fini(&spin_lo); err_spin_hi: - spinner_fini(&spin_hi); + igt_spinner_fini(&spin_hi); err_unlock: igt_flush_test(i915, I915_WAIT_LOCKED); intel_runtime_pm_put(i915); @@ -365,7 +167,7 @@ static int live_late_preempt(void *arg) { struct drm_i915_private *i915 = arg; struct i915_gem_context *ctx_hi, *ctx_lo; - struct spinner spin_hi, spin_lo; + struct igt_spinner spin_hi, spin_lo; struct intel_engine_cs *engine; struct i915_sched_attr attr = {}; enum intel_engine_id id; @@ -377,10 +179,10 @@ static int live_late_preempt(void *arg) mutex_lock(&i915->drm.struct_mutex); intel_runtime_pm_get(i915); - if (spinner_init(&spin_hi, i915)) + if (igt_spinner_init(&spin_hi, i915)) goto err_unlock; - if (spinner_init(&spin_lo, i915)) + if (igt_spinner_init(&spin_lo, i915)) goto err_spin_hi; ctx_hi = kernel_context(i915); @@ -394,28 +196,29 @@ static int live_late_preempt(void *arg) for_each_engine(engine, i915, id) { struct i915_request *rq; - rq = spinner_create_request(&spin_lo, ctx_lo, engine, - MI_ARB_CHECK); + rq = igt_spinner_create_request(&spin_lo, ctx_lo, engine, + MI_ARB_CHECK); if (IS_ERR(rq)) { err = PTR_ERR(rq); goto err_ctx_lo; } i915_request_add(rq); - if (!wait_for_spinner(&spin_lo, rq)) { + if (!igt_wait_for_spinner(&spin_lo, rq)) { pr_err("First context failed to start\n"); goto err_wedged; } - rq = spinner_create_request(&spin_hi, ctx_hi, engine, MI_NOOP); + rq = igt_spinner_create_request(&spin_hi, ctx_hi, engine, + MI_NOOP); if (IS_ERR(rq)) { - spinner_end(&spin_lo); + igt_spinner_end(&spin_lo); err = PTR_ERR(rq); goto err_ctx_lo; } i915_request_add(rq); - if (wait_for_spinner(&spin_hi, rq)) { + if (igt_wait_for_spinner(&spin_hi, rq)) { pr_err("Second context overtook first?\n"); goto err_wedged; } @@ -423,14 +226,14 @@ static int live_late_preempt(void *arg) attr.priority = I915_USER_PRIORITY(I915_PRIORITY_MAX); engine->schedule(rq, &attr); - if (!wait_for_spinner(&spin_hi, rq)) { + if (!igt_wait_for_spinner(&spin_hi, rq)) { pr_err("High priority context failed to preempt the low priority context\n"); GEM_TRACE_DUMP(); goto err_wedged; } - spinner_end(&spin_hi); - spinner_end(&spin_lo); + igt_spinner_end(&spin_hi); + igt_spinner_end(&spin_lo); if (igt_flush_test(i915, I915_WAIT_LOCKED)) { err = -EIO; goto err_ctx_lo; @@ -443,9 +246,9 @@ err_ctx_lo: err_ctx_hi: kernel_context_close(ctx_hi); err_spin_lo: - spinner_fini(&spin_lo); + igt_spinner_fini(&spin_lo); err_spin_hi: - spinner_fini(&spin_hi); + igt_spinner_fini(&spin_hi); err_unlock: igt_flush_test(i915, I915_WAIT_LOCKED); intel_runtime_pm_put(i915); @@ -453,8 +256,8 @@ err_unlock: return err; err_wedged: - spinner_end(&spin_hi); - spinner_end(&spin_lo); + igt_spinner_end(&spin_hi); + igt_spinner_end(&spin_lo); i915_gem_set_wedged(i915); err = -EIO; goto err_ctx_lo; @@ -464,7 +267,7 @@ static int live_preempt_hang(void *arg) { struct drm_i915_private *i915 = arg; struct i915_gem_context *ctx_hi, *ctx_lo; - struct spinner spin_hi, spin_lo; + struct igt_spinner spin_hi, spin_lo; struct intel_engine_cs *engine; enum intel_engine_id id; int err = -ENOMEM; @@ -478,10 +281,10 @@ static int live_preempt_hang(void *arg) mutex_lock(&i915->drm.struct_mutex); intel_runtime_pm_get(i915); - if (spinner_init(&spin_hi, i915)) + if (igt_spinner_init(&spin_hi, i915)) goto err_unlock; - if (spinner_init(&spin_lo, i915)) + if (igt_spinner_init(&spin_lo, i915)) goto err_spin_hi; ctx_hi = kernel_context(i915); @@ -500,15 +303,15 @@ static int live_preempt_hang(void *arg) if (!intel_engine_has_preemption(engine)) continue; - rq = spinner_create_request(&spin_lo, ctx_lo, engine, - MI_ARB_CHECK); + rq = igt_spinner_create_request(&spin_lo, ctx_lo, engine, + MI_ARB_CHECK); if (IS_ERR(rq)) { err = PTR_ERR(rq); goto err_ctx_lo; } i915_request_add(rq); - if (!wait_for_spinner(&spin_lo, rq)) { + if (!igt_wait_for_spinner(&spin_lo, rq)) { GEM_TRACE("lo spinner failed to start\n"); GEM_TRACE_DUMP(); i915_gem_set_wedged(i915); @@ -516,10 +319,10 @@ static int live_preempt_hang(void *arg) goto err_ctx_lo; } - rq = spinner_create_request(&spin_hi, ctx_hi, engine, - MI_ARB_CHECK); + rq = igt_spinner_create_request(&spin_hi, ctx_hi, engine, + MI_ARB_CHECK); if (IS_ERR(rq)) { - spinner_end(&spin_lo); + igt_spinner_end(&spin_lo); err = PTR_ERR(rq); goto err_ctx_lo; } @@ -544,7 +347,7 @@ static int live_preempt_hang(void *arg) engine->execlists.preempt_hang.inject_hang = false; - if (!wait_for_spinner(&spin_hi, rq)) { + if (!igt_wait_for_spinner(&spin_hi, rq)) { GEM_TRACE("hi spinner failed to start\n"); GEM_TRACE_DUMP(); i915_gem_set_wedged(i915); @@ -552,8 +355,8 @@ static int live_preempt_hang(void *arg) goto err_ctx_lo; } - spinner_end(&spin_hi); - spinner_end(&spin_lo); + igt_spinner_end(&spin_hi); + igt_spinner_end(&spin_lo); if (igt_flush_test(i915, I915_WAIT_LOCKED)) { err = -EIO; goto err_ctx_lo; @@ -566,9 +369,9 @@ err_ctx_lo: err_ctx_hi: kernel_context_close(ctx_hi); err_spin_lo: - spinner_fini(&spin_lo); + igt_spinner_fini(&spin_lo); err_spin_hi: - spinner_fini(&spin_hi); + igt_spinner_fini(&spin_hi); err_unlock: igt_flush_test(i915, I915_WAIT_LOCKED); intel_runtime_pm_put(i915); diff --git a/drivers/gpu/drm/i915/selftests/intel_workarounds.c b/drivers/gpu/drm/i915/selftests/intel_workarounds.c index d1a0923d2f38..67017d5175b8 100644 --- a/drivers/gpu/drm/i915/selftests/intel_workarounds.c +++ b/drivers/gpu/drm/i915/selftests/intel_workarounds.c @@ -6,6 +6,9 @@ #include "../i915_selftest.h" +#include "igt_flush_test.h" +#include "igt_reset.h" +#include "igt_spinner.h" #include "igt_wedge_me.h" #include "mock_context.h" @@ -91,17 +94,23 @@ err_obj: return ERR_PTR(err); } -static u32 get_whitelist_reg(const struct whitelist *w, unsigned int i) +static u32 +get_whitelist_reg(const struct intel_engine_cs *engine, unsigned int i) { - return i < w->count ? i915_mmio_reg_offset(w->reg[i]) : w->nopid; + i915_reg_t reg = i < engine->whitelist.count ? + engine->whitelist.list[i].reg : + RING_NOPID(engine->mmio_base); + + return i915_mmio_reg_offset(reg); } -static void print_results(const struct whitelist *w, const u32 *results) +static void +print_results(const struct intel_engine_cs *engine, const u32 *results) { unsigned int i; for (i = 0; i < RING_MAX_NONPRIV_SLOTS; i++) { - u32 expected = get_whitelist_reg(w, i); + u32 expected = get_whitelist_reg(engine, i); u32 actual = results[i]; pr_info("RING_NONPRIV[%d]: expected 0x%08x, found 0x%08x\n", @@ -109,8 +118,7 @@ static void print_results(const struct whitelist *w, const u32 *results) } } -static int check_whitelist(const struct whitelist *w, - struct i915_gem_context *ctx, +static int check_whitelist(struct i915_gem_context *ctx, struct intel_engine_cs *engine) { struct drm_i915_gem_object *results; @@ -138,11 +146,11 @@ static int check_whitelist(const struct whitelist *w, } for (i = 0; i < RING_MAX_NONPRIV_SLOTS; i++) { - u32 expected = get_whitelist_reg(w, i); + u32 expected = get_whitelist_reg(engine, i); u32 actual = vaddr[i]; if (expected != actual) { - print_results(w, vaddr); + print_results(engine, vaddr); pr_err("Invalid RING_NONPRIV[%d], expected 0x%08x, found 0x%08x\n", i, expected, actual); @@ -159,66 +167,107 @@ out_put: static int do_device_reset(struct intel_engine_cs *engine) { - i915_reset(engine->i915, ENGINE_MASK(engine->id), NULL); + set_bit(I915_RESET_HANDOFF, &engine->i915->gpu_error.flags); + i915_reset(engine->i915, ENGINE_MASK(engine->id), "live_workarounds"); return 0; } static int do_engine_reset(struct intel_engine_cs *engine) { - return i915_reset_engine(engine, NULL); + return i915_reset_engine(engine, "live_workarounds"); } -static int switch_to_scratch_context(struct intel_engine_cs *engine) +static int +switch_to_scratch_context(struct intel_engine_cs *engine, + struct igt_spinner *spin) { struct i915_gem_context *ctx; struct i915_request *rq; + int err = 0; ctx = kernel_context(engine->i915); if (IS_ERR(ctx)) return PTR_ERR(ctx); intel_runtime_pm_get(engine->i915); - rq = i915_request_alloc(engine, ctx); + + if (spin) + rq = igt_spinner_create_request(spin, ctx, engine, MI_NOOP); + else + rq = i915_request_alloc(engine, ctx); + intel_runtime_pm_put(engine->i915); kernel_context_close(ctx); - if (IS_ERR(rq)) - return PTR_ERR(rq); + + if (IS_ERR(rq)) { + spin = NULL; + err = PTR_ERR(rq); + goto err; + } i915_request_add(rq); - return 0; + if (spin && !igt_wait_for_spinner(spin, rq)) { + pr_err("Spinner failed to start\n"); + err = -ETIMEDOUT; + } + +err: + if (err && spin) + igt_spinner_end(spin); + + return err; } static int check_whitelist_across_reset(struct intel_engine_cs *engine, int (*reset)(struct intel_engine_cs *), - const struct whitelist *w, const char *name) { + struct drm_i915_private *i915 = engine->i915; + bool want_spin = reset == do_engine_reset; struct i915_gem_context *ctx; + struct igt_spinner spin; int err; - ctx = kernel_context(engine->i915); + pr_info("Checking %d whitelisted registers (RING_NONPRIV) [%s]\n", + engine->whitelist.count, name); + + if (want_spin) { + err = igt_spinner_init(&spin, i915); + if (err) + return err; + } + + ctx = kernel_context(i915); if (IS_ERR(ctx)) return PTR_ERR(ctx); - err = check_whitelist(w, ctx, engine); + err = check_whitelist(ctx, engine); if (err) { pr_err("Invalid whitelist *before* %s reset!\n", name); goto out; } - err = switch_to_scratch_context(engine); + err = switch_to_scratch_context(engine, want_spin ? &spin : NULL); if (err) goto out; + intel_runtime_pm_get(i915); err = reset(engine); + intel_runtime_pm_put(i915); + + if (want_spin) { + igt_spinner_end(&spin); + igt_spinner_fini(&spin); + } + if (err) { pr_err("%s reset failed\n", name); goto out; } - err = check_whitelist(w, ctx, engine); + err = check_whitelist(ctx, engine); if (err) { pr_err("Whitelist not preserved in context across %s reset!\n", name); @@ -227,11 +276,11 @@ static int check_whitelist_across_reset(struct intel_engine_cs *engine, kernel_context_close(ctx); - ctx = kernel_context(engine->i915); + ctx = kernel_context(i915); if (IS_ERR(ctx)) return PTR_ERR(ctx); - err = check_whitelist(w, ctx, engine); + err = check_whitelist(ctx, engine); if (err) { pr_err("Invalid whitelist *after* %s reset in fresh context!\n", name); @@ -247,26 +296,18 @@ static int live_reset_whitelist(void *arg) { struct drm_i915_private *i915 = arg; struct intel_engine_cs *engine = i915->engine[RCS]; - struct i915_gpu_error *error = &i915->gpu_error; - struct whitelist w; int err = 0; /* If we reset the gpu, we should not lose the RING_NONPRIV */ - if (!engine) - return 0; - - if (!whitelist_build(engine, &w)) + if (!engine || engine->whitelist.count == 0) return 0; - pr_info("Checking %d whitelisted registers (RING_NONPRIV)\n", w.count); - - set_bit(I915_RESET_BACKOFF, &error->flags); - set_bit(I915_RESET_ENGINE + engine->id, &error->flags); + igt_global_reset_lock(i915); if (intel_has_reset_engine(i915)) { err = check_whitelist_across_reset(engine, - do_engine_reset, &w, + do_engine_reset, "engine"); if (err) goto out; @@ -274,22 +315,156 @@ static int live_reset_whitelist(void *arg) if (intel_has_gpu_reset(i915)) { err = check_whitelist_across_reset(engine, - do_device_reset, &w, + do_device_reset, "device"); if (err) goto out; } out: - clear_bit(I915_RESET_ENGINE + engine->id, &error->flags); - clear_bit(I915_RESET_BACKOFF, &error->flags); + igt_global_reset_unlock(i915); return err; } +static bool verify_gt_engine_wa(struct drm_i915_private *i915, const char *str) +{ + struct intel_engine_cs *engine; + enum intel_engine_id id; + bool ok = true; + + ok &= intel_gt_verify_workarounds(i915, str); + + for_each_engine(engine, i915, id) + ok &= intel_engine_verify_workarounds(engine, str); + + return ok; +} + +static int +live_gpu_reset_gt_engine_workarounds(void *arg) +{ + struct drm_i915_private *i915 = arg; + struct i915_gpu_error *error = &i915->gpu_error; + bool ok; + + if (!intel_has_gpu_reset(i915)) + return 0; + + pr_info("Verifying after GPU reset...\n"); + + igt_global_reset_lock(i915); + + ok = verify_gt_engine_wa(i915, "before reset"); + if (!ok) + goto out; + + intel_runtime_pm_get(i915); + set_bit(I915_RESET_HANDOFF, &error->flags); + i915_reset(i915, ALL_ENGINES, "live_workarounds"); + intel_runtime_pm_put(i915); + + ok = verify_gt_engine_wa(i915, "after reset"); + +out: + igt_global_reset_unlock(i915); + + return ok ? 0 : -ESRCH; +} + +static int +live_engine_reset_gt_engine_workarounds(void *arg) +{ + struct drm_i915_private *i915 = arg; + struct intel_engine_cs *engine; + struct i915_gem_context *ctx; + struct igt_spinner spin; + enum intel_engine_id id; + struct i915_request *rq; + int ret = 0; + + if (!intel_has_reset_engine(i915)) + return 0; + + ctx = kernel_context(i915); + if (IS_ERR(ctx)) + return PTR_ERR(ctx); + + igt_global_reset_lock(i915); + + for_each_engine(engine, i915, id) { + bool ok; + + pr_info("Verifying after %s reset...\n", engine->name); + + ok = verify_gt_engine_wa(i915, "before reset"); + if (!ok) { + ret = -ESRCH; + goto err; + } + + intel_runtime_pm_get(i915); + i915_reset_engine(engine, "live_workarounds"); + intel_runtime_pm_put(i915); + + ok = verify_gt_engine_wa(i915, "after idle reset"); + if (!ok) { + ret = -ESRCH; + goto err; + } + + ret = igt_spinner_init(&spin, i915); + if (ret) + goto err; + + intel_runtime_pm_get(i915); + + rq = igt_spinner_create_request(&spin, ctx, engine, MI_NOOP); + if (IS_ERR(rq)) { + ret = PTR_ERR(rq); + igt_spinner_fini(&spin); + intel_runtime_pm_put(i915); + goto err; + } + + i915_request_add(rq); + + if (!igt_wait_for_spinner(&spin, rq)) { + pr_err("Spinner failed to start\n"); + igt_spinner_fini(&spin); + intel_runtime_pm_put(i915); + ret = -ETIMEDOUT; + goto err; + } + + i915_reset_engine(engine, "live_workarounds"); + + intel_runtime_pm_put(i915); + + igt_spinner_end(&spin); + igt_spinner_fini(&spin); + + ok = verify_gt_engine_wa(i915, "after busy reset"); + if (!ok) { + ret = -ESRCH; + goto err; + } + } + +err: + igt_global_reset_unlock(i915); + kernel_context_close(ctx); + + igt_flush_test(i915, I915_WAIT_LOCKED); + + return ret; +} + int intel_workarounds_live_selftests(struct drm_i915_private *i915) { static const struct i915_subtest tests[] = { SUBTEST(live_reset_whitelist), + SUBTEST(live_gpu_reset_gt_engine_workarounds), + SUBTEST(live_engine_reset_gt_engine_workarounds), }; int err; diff --git a/drivers/gpu/drm/imx/dw_hdmi-imx.c b/drivers/gpu/drm/imx/dw_hdmi-imx.c index fe6becdcc29e..77a26fd3a44a 100644 --- a/drivers/gpu/drm/imx/dw_hdmi-imx.c +++ b/drivers/gpu/drm/imx/dw_hdmi-imx.c @@ -1,10 +1,7 @@ +// SPDX-License-Identifier: GPL-2.0 /* Copyright (C) 2011-2013 Freescale Semiconductor, Inc. * * derived from imx-hdmi.c(renamed to bridge/dw_hdmi.c now) - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. */ #include <linux/module.h> #include <linux/platform_device.h> diff --git a/drivers/gpu/drm/imx/imx-drm-core.c b/drivers/gpu/drm/imx/imx-drm-core.c index 0e6942f21a4e..820c7e3878f0 100644 --- a/drivers/gpu/drm/imx/imx-drm-core.c +++ b/drivers/gpu/drm/imx/imx-drm-core.c @@ -1,17 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0+ /* * Freescale i.MX drm driver * * Copyright (C) 2011 Sascha Hauer, Pengutronix - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version 2 - * of the License, or (at your option) any later version. - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * */ #include <linux/component.h> #include <linux/device.h> diff --git a/drivers/gpu/drm/imx/imx-ldb.c b/drivers/gpu/drm/imx/imx-ldb.c index 3bd0f8a18e74..2c5bbe317353 100644 --- a/drivers/gpu/drm/imx/imx-ldb.c +++ b/drivers/gpu/drm/imx/imx-ldb.c @@ -1,16 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0+ /* * i.MX drm driver - LVDS display bridge * * Copyright (C) 2012 Sascha Hauer, Pengutronix - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version 2 - * of the License, or (at your option) any later version. - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. */ #include <linux/module.h> diff --git a/drivers/gpu/drm/imx/imx-tve.c b/drivers/gpu/drm/imx/imx-tve.c index cffd3310240e..293dd5752583 100644 --- a/drivers/gpu/drm/imx/imx-tve.c +++ b/drivers/gpu/drm/imx/imx-tve.c @@ -1,16 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0+ /* * i.MX drm driver - Television Encoder (TVEv2) * * Copyright (C) 2013 Philipp Zabel, Pengutronix - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version 2 - * of the License, or (at your option) any later version. - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. */ #include <linux/clk.h> @@ -442,7 +434,7 @@ static int clk_tve_di_set_rate(struct clk_hw *hw, unsigned long rate, return 0; } -static struct clk_ops clk_tve_di_ops = { +static const struct clk_ops clk_tve_di_ops = { .round_rate = clk_tve_di_round_rate, .set_rate = clk_tve_di_set_rate, .recalc_rate = clk_tve_di_recalc_rate, diff --git a/drivers/gpu/drm/imx/ipuv3-crtc.c b/drivers/gpu/drm/imx/ipuv3-crtc.c index 7d4b710b837a..058b53c0aa7e 100644 --- a/drivers/gpu/drm/imx/ipuv3-crtc.c +++ b/drivers/gpu/drm/imx/ipuv3-crtc.c @@ -1,16 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0+ /* * i.MX IPUv3 Graphics driver * * Copyright (C) 2011 Sascha Hauer, Pengutronix - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version 2 - * of the License, or (at your option) any later version. - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. */ #include <linux/component.h> #include <linux/module.h> diff --git a/drivers/gpu/drm/imx/ipuv3-plane.c b/drivers/gpu/drm/imx/ipuv3-plane.c index 40605fdf0e33..c390924de93d 100644 --- a/drivers/gpu/drm/imx/ipuv3-plane.c +++ b/drivers/gpu/drm/imx/ipuv3-plane.c @@ -1,16 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0+ /* * i.MX IPUv3 DP Overlay Planes * * Copyright (C) 2013 Philipp Zabel, Pengutronix - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version 2 - * of the License, or (at your option) any later version. - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. */ #include <drm/drmP.h> @@ -236,9 +228,15 @@ static void ipu_plane_enable(struct ipu_plane *ipu_plane) void ipu_plane_disable(struct ipu_plane *ipu_plane, bool disable_dp_channel) { + int ret; + DRM_DEBUG_KMS("[%d] %s\n", __LINE__, __func__); - ipu_idmac_wait_busy(ipu_plane->ipu_ch, 50); + ret = ipu_idmac_wait_busy(ipu_plane->ipu_ch, 50); + if (ret == -ETIMEDOUT) { + DRM_ERROR("[PLANE:%d] IDMAC timeout\n", + ipu_plane->base.base.id); + } if (ipu_plane->dp && disable_dp_channel) ipu_dp_disable_channel(ipu_plane->dp, false); diff --git a/drivers/gpu/drm/imx/parallel-display.c b/drivers/gpu/drm/imx/parallel-display.c index aefd04e18f93..f3ce51121dd6 100644 --- a/drivers/gpu/drm/imx/parallel-display.c +++ b/drivers/gpu/drm/imx/parallel-display.c @@ -1,16 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0+ /* * i.MX drm driver - parallel display implementation * * Copyright (C) 2012 Sascha Hauer, Pengutronix - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version 2 - * of the License, or (at your option) any later version. - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. */ #include <linux/component.h> diff --git a/drivers/gpu/drm/mediatek/mtk_dsi.c b/drivers/gpu/drm/mediatek/mtk_dsi.c index 66df1b177959..27b507eb4a99 100644 --- a/drivers/gpu/drm/mediatek/mtk_dsi.c +++ b/drivers/gpu/drm/mediatek/mtk_dsi.c @@ -818,10 +818,13 @@ static int mtk_dsi_create_conn_enc(struct drm_device *drm, struct mtk_dsi *dsi) dsi->encoder.possible_crtcs = 1; /* If there's a bridge, attach to it and let it create the connector */ - ret = drm_bridge_attach(&dsi->encoder, dsi->bridge, NULL); - if (ret) { - DRM_ERROR("Failed to attach bridge to drm\n"); - + if (dsi->bridge) { + ret = drm_bridge_attach(&dsi->encoder, dsi->bridge, NULL); + if (ret) { + DRM_ERROR("Failed to attach bridge to drm\n"); + goto err_encoder_cleanup; + } + } else { /* Otherwise create our own connector and attach to a panel */ ret = mtk_dsi_create_connector(drm, dsi); if (ret) diff --git a/drivers/gpu/drm/meson/meson_crtc.c b/drivers/gpu/drm/meson/meson_crtc.c index d78168f979db..75d97f1b2e8f 100644 --- a/drivers/gpu/drm/meson/meson_crtc.c +++ b/drivers/gpu/drm/meson/meson_crtc.c @@ -46,6 +46,7 @@ struct meson_crtc { struct drm_crtc base; struct drm_pending_vblank_event *event; struct meson_drm *priv; + bool enabled; }; #define to_meson_crtc(x) container_of(x, struct meson_crtc, base) @@ -81,8 +82,7 @@ static const struct drm_crtc_funcs meson_crtc_funcs = { }; -static void meson_crtc_atomic_enable(struct drm_crtc *crtc, - struct drm_crtc_state *old_state) +static void meson_crtc_enable(struct drm_crtc *crtc) { struct meson_crtc *meson_crtc = to_meson_crtc(crtc); struct drm_crtc_state *crtc_state = crtc->state; @@ -106,6 +106,22 @@ static void meson_crtc_atomic_enable(struct drm_crtc *crtc, writel_bits_relaxed(VPP_POSTBLEND_ENABLE, VPP_POSTBLEND_ENABLE, priv->io_base + _REG(VPP_MISC)); + drm_crtc_vblank_on(crtc); + + meson_crtc->enabled = true; +} + +static void meson_crtc_atomic_enable(struct drm_crtc *crtc, + struct drm_crtc_state *old_state) +{ + struct meson_crtc *meson_crtc = to_meson_crtc(crtc); + struct meson_drm *priv = meson_crtc->priv; + + DRM_DEBUG_DRIVER("\n"); + + if (!meson_crtc->enabled) + meson_crtc_enable(crtc); + priv->viu.osd1_enabled = true; } @@ -117,6 +133,8 @@ static void meson_crtc_atomic_disable(struct drm_crtc *crtc, DRM_DEBUG_DRIVER("\n"); + drm_crtc_vblank_off(crtc); + priv->viu.osd1_enabled = false; priv->viu.osd1_commit = false; @@ -135,6 +153,8 @@ static void meson_crtc_atomic_disable(struct drm_crtc *crtc, crtc->state->event = NULL; } + + meson_crtc->enabled = false; } static void meson_crtc_atomic_begin(struct drm_crtc *crtc, @@ -143,6 +163,9 @@ static void meson_crtc_atomic_begin(struct drm_crtc *crtc, struct meson_crtc *meson_crtc = to_meson_crtc(crtc); unsigned long flags; + if (crtc->state->enable && !meson_crtc->enabled) + meson_crtc_enable(crtc); + if (crtc->state->event) { WARN_ON(drm_crtc_vblank_get(crtc) != 0); diff --git a/drivers/gpu/drm/meson/meson_dw_hdmi.c b/drivers/gpu/drm/meson/meson_dw_hdmi.c index 0b6c29cdd934..bc25001b8207 100644 --- a/drivers/gpu/drm/meson/meson_dw_hdmi.c +++ b/drivers/gpu/drm/meson/meson_dw_hdmi.c @@ -690,6 +690,7 @@ static const struct regmap_config meson_dw_hdmi_regmap_config = { .reg_read = meson_dw_hdmi_reg_read, .reg_write = meson_dw_hdmi_reg_write, .max_register = 0x10000, + .fast_io = true, }; static bool meson_hdmi_connector_is_available(struct device *dev) diff --git a/drivers/gpu/drm/meson/meson_venc.c b/drivers/gpu/drm/meson/meson_venc.c index e95e0e7a7fa1..0ba04f6813e6 100644 --- a/drivers/gpu/drm/meson/meson_venc.c +++ b/drivers/gpu/drm/meson/meson_venc.c @@ -71,6 +71,7 @@ */ /* HHI Registers */ +#define HHI_GCLK_MPEG2 0x148 /* 0x52 offset in data sheet */ #define HHI_VDAC_CNTL0 0x2F4 /* 0xbd offset in data sheet */ #define HHI_VDAC_CNTL1 0x2F8 /* 0xbe offset in data sheet */ #define HHI_HDMI_PHY_CNTL0 0x3a0 /* 0xe8 offset in data sheet */ @@ -840,6 +841,7 @@ struct meson_hdmi_venc_vic_mode { { 5, &meson_hdmi_encp_mode_1080i60 }, { 20, &meson_hdmi_encp_mode_1080i50 }, { 32, &meson_hdmi_encp_mode_1080p24 }, + { 33, &meson_hdmi_encp_mode_1080p50 }, { 34, &meson_hdmi_encp_mode_1080p30 }, { 31, &meson_hdmi_encp_mode_1080p50 }, { 16, &meson_hdmi_encp_mode_1080p60 }, @@ -1659,10 +1661,12 @@ unsigned int meson_venci_get_field(struct meson_drm *priv) void meson_venc_enable_vsync(struct meson_drm *priv) { writel_relaxed(2, priv->io_base + _REG(VENC_INTCTRL)); + regmap_update_bits(priv->hhi, HHI_GCLK_MPEG2, BIT(25), BIT(25)); } void meson_venc_disable_vsync(struct meson_drm *priv) { + regmap_update_bits(priv->hhi, HHI_GCLK_MPEG2, BIT(25), 0); writel_relaxed(0, priv->io_base + _REG(VENC_INTCTRL)); } diff --git a/drivers/gpu/drm/meson/meson_viu.c b/drivers/gpu/drm/meson/meson_viu.c index 0ba87ff95530..e46e05f50bad 100644 --- a/drivers/gpu/drm/meson/meson_viu.c +++ b/drivers/gpu/drm/meson/meson_viu.c @@ -184,18 +184,18 @@ void meson_viu_set_osd_lut(struct meson_drm *priv, enum viu_lut_sel_e lut_sel, if (lut_sel == VIU_LUT_OSD_OETF) { writel(0, priv->io_base + _REG(addr_port)); - for (i = 0; i < 20; i++) + for (i = 0; i < (OSD_OETF_LUT_SIZE / 2); i++) writel(r_map[i * 2] | (r_map[i * 2 + 1] << 16), priv->io_base + _REG(data_port)); writel(r_map[OSD_OETF_LUT_SIZE - 1] | (g_map[0] << 16), priv->io_base + _REG(data_port)); - for (i = 0; i < 20; i++) + for (i = 0; i < (OSD_OETF_LUT_SIZE / 2); i++) writel(g_map[i * 2 + 1] | (g_map[i * 2 + 2] << 16), priv->io_base + _REG(data_port)); - for (i = 0; i < 20; i++) + for (i = 0; i < (OSD_OETF_LUT_SIZE / 2); i++) writel(b_map[i * 2] | (b_map[i * 2 + 1] << 16), priv->io_base + _REG(data_port)); @@ -211,18 +211,18 @@ void meson_viu_set_osd_lut(struct meson_drm *priv, enum viu_lut_sel_e lut_sel, } else if (lut_sel == VIU_LUT_OSD_EOTF) { writel(0, priv->io_base + _REG(addr_port)); - for (i = 0; i < 20; i++) + for (i = 0; i < (OSD_EOTF_LUT_SIZE / 2); i++) writel(r_map[i * 2] | (r_map[i * 2 + 1] << 16), priv->io_base + _REG(data_port)); writel(r_map[OSD_EOTF_LUT_SIZE - 1] | (g_map[0] << 16), priv->io_base + _REG(data_port)); - for (i = 0; i < 20; i++) + for (i = 0; i < (OSD_EOTF_LUT_SIZE / 2); i++) writel(g_map[i * 2 + 1] | (g_map[i * 2 + 2] << 16), priv->io_base + _REG(data_port)); - for (i = 0; i < 20; i++) + for (i = 0; i < (OSD_EOTF_LUT_SIZE / 2); i++) writel(b_map[i * 2] | (b_map[i * 2 + 1] << 16), priv->io_base + _REG(data_port)); diff --git a/drivers/gpu/drm/msm/Kconfig b/drivers/gpu/drm/msm/Kconfig index 843a9d40c05e..cf549f1ed403 100644 --- a/drivers/gpu/drm/msm/Kconfig +++ b/drivers/gpu/drm/msm/Kconfig @@ -2,7 +2,7 @@ config DRM_MSM tristate "MSM DRM" depends on DRM - depends on ARCH_QCOM || (ARM && COMPILE_TEST) + depends on ARCH_QCOM || SOC_IMX5 || (ARM && COMPILE_TEST) depends on OF && COMMON_CLK depends on MMU select QCOM_MDT_LOADER if ARCH_QCOM @@ -11,7 +11,7 @@ config DRM_MSM select DRM_PANEL select SHMEM select TMPFS - select QCOM_SCM + select QCOM_SCM if ARCH_QCOM select WANT_DEV_COREDUMP select SND_SOC_HDMI_CODEC if SND_SOC select SYNC_FILE diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile index 19ab521d4c3a..56a70c74af4e 100644 --- a/drivers/gpu/drm/msm/Makefile +++ b/drivers/gpu/drm/msm/Makefile @@ -6,6 +6,7 @@ ccflags-$(CONFIG_DRM_MSM_DSI) += -Idrivers/gpu/drm/msm/dsi msm-y := \ adreno/adreno_device.o \ adreno/adreno_gpu.o \ + adreno/a2xx_gpu.o \ adreno/a3xx_gpu.o \ adreno/a4xx_gpu.o \ adreno/a5xx_gpu.o \ @@ -14,6 +15,7 @@ msm-y := \ adreno/a6xx_gpu.o \ adreno/a6xx_gmu.o \ adreno/a6xx_hfi.o \ + adreno/a6xx_gpu_state.o \ hdmi/hdmi.o \ hdmi/hdmi_audio.o \ hdmi/hdmi_bridge.o \ @@ -68,11 +70,9 @@ msm-y := \ disp/dpu1/dpu_hw_util.o \ disp/dpu1/dpu_hw_vbif.o \ disp/dpu1/dpu_io_util.o \ - disp/dpu1/dpu_irq.o \ disp/dpu1/dpu_kms.o \ disp/dpu1/dpu_mdss.o \ disp/dpu1/dpu_plane.o \ - disp/dpu1/dpu_power_handle.o \ disp/dpu1/dpu_rm.o \ disp/dpu1/dpu_vbif.o \ msm_atomic.o \ @@ -90,10 +90,11 @@ msm-y := \ msm_perf.o \ msm_rd.o \ msm_ringbuffer.o \ - msm_submitqueue.o + msm_submitqueue.o \ + msm_gpu_tracepoints.o \ + msm_gpummu.o -msm-$(CONFIG_DEBUG_FS) += adreno/a5xx_debugfs.o \ - disp/dpu1/dpu_dbg.o +msm-$(CONFIG_DEBUG_FS) += adreno/a5xx_debugfs.o msm-$(CONFIG_DRM_FBDEV_EMULATION) += msm_fbdev.o msm-$(CONFIG_COMMON_CLK) += disp/mdp4/mdp4_lvds_pll.o diff --git a/drivers/gpu/drm/msm/adreno/a2xx.xml.h b/drivers/gpu/drm/msm/adreno/a2xx.xml.h index 12b0ba270b5e..14eb52f3e605 100644 --- a/drivers/gpu/drm/msm/adreno/a2xx.xml.h +++ b/drivers/gpu/drm/msm/adreno/a2xx.xml.h @@ -10,13 +10,13 @@ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: - /home/robclark/src/envytools/rnndb/adreno.xml ( 501 bytes, from 2018-07-03 19:37:13) - /home/robclark/src/envytools/rnndb/freedreno_copyright.xml ( 1572 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/a2xx.xml ( 36805 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/adreno_common.xml ( 13634 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/adreno_pm4.xml ( 42585 bytes, from 2018-10-04 19:06:37) +- /home/robclark/src/envytools/rnndb/adreno/a2xx.xml ( 42463 bytes, from 2018-11-19 13:44:03) +- /home/robclark/src/envytools/rnndb/adreno/adreno_common.xml ( 14201 bytes, from 2018-12-02 17:29:54) +- /home/robclark/src/envytools/rnndb/adreno/adreno_pm4.xml ( 43052 bytes, from 2018-12-02 17:29:54) - /home/robclark/src/envytools/rnndb/adreno/a3xx.xml ( 83840 bytes, from 2018-07-03 19:37:13) - /home/robclark/src/envytools/rnndb/adreno/a4xx.xml ( 112086 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/a5xx.xml ( 147240 bytes, from 2018-10-04 19:06:37) -- /home/robclark/src/envytools/rnndb/adreno/a6xx.xml ( 139581 bytes, from 2018-10-04 19:06:42) +- /home/robclark/src/envytools/rnndb/adreno/a5xx.xml ( 147240 bytes, from 2018-12-02 17:29:54) +- /home/robclark/src/envytools/rnndb/adreno/a6xx.xml ( 140790 bytes, from 2018-12-02 17:29:54) - /home/robclark/src/envytools/rnndb/adreno/a6xx_gmu.xml ( 10431 bytes, from 2018-09-14 13:03:07) - /home/robclark/src/envytools/rnndb/adreno/ocmem.xml ( 1773 bytes, from 2018-07-03 19:37:13) @@ -239,7 +239,63 @@ enum sq_tex_swiz { enum sq_tex_filter { SQ_TEX_FILTER_POINT = 0, SQ_TEX_FILTER_BILINEAR = 1, - SQ_TEX_FILTER_BICUBIC = 2, + SQ_TEX_FILTER_BASEMAP = 2, + SQ_TEX_FILTER_USE_FETCH_CONST = 3, +}; + +enum sq_tex_aniso_filter { + SQ_TEX_ANISO_FILTER_DISABLED = 0, + SQ_TEX_ANISO_FILTER_MAX_1_1 = 1, + SQ_TEX_ANISO_FILTER_MAX_2_1 = 2, + SQ_TEX_ANISO_FILTER_MAX_4_1 = 3, + SQ_TEX_ANISO_FILTER_MAX_8_1 = 4, + SQ_TEX_ANISO_FILTER_MAX_16_1 = 5, + SQ_TEX_ANISO_FILTER_USE_FETCH_CONST = 7, +}; + +enum sq_tex_dimension { + SQ_TEX_DIMENSION_1D = 0, + SQ_TEX_DIMENSION_2D = 1, + SQ_TEX_DIMENSION_3D = 2, + SQ_TEX_DIMENSION_CUBE = 3, +}; + +enum sq_tex_border_color { + SQ_TEX_BORDER_COLOR_BLACK = 0, + SQ_TEX_BORDER_COLOR_WHITE = 1, + SQ_TEX_BORDER_COLOR_ACBYCR_BLACK = 2, + SQ_TEX_BORDER_COLOR_ACBCRY_BLACK = 3, +}; + +enum sq_tex_sign { + SQ_TEX_SIGN_UNISIGNED = 0, + SQ_TEX_SIGN_SIGNED = 1, + SQ_TEX_SIGN_UNISIGNED_BIASED = 2, + SQ_TEX_SIGN_GAMMA = 3, +}; + +enum sq_tex_endian { + SQ_TEX_ENDIAN_NONE = 0, + SQ_TEX_ENDIAN_8IN16 = 1, + SQ_TEX_ENDIAN_8IN32 = 2, + SQ_TEX_ENDIAN_16IN32 = 3, +}; + +enum sq_tex_clamp_policy { + SQ_TEX_CLAMP_POLICY_D3D = 0, + SQ_TEX_CLAMP_POLICY_OGL = 1, +}; + +enum sq_tex_num_format { + SQ_TEX_NUM_FORMAT_FRAC = 0, + SQ_TEX_NUM_FORMAT_INT = 1, +}; + +enum sq_tex_type { + SQ_TEX_TYPE_0 = 0, + SQ_TEX_TYPE_1 = 1, + SQ_TEX_TYPE_2 = 2, + SQ_TEX_TYPE_3 = 3, }; #define REG_A2XX_RBBM_PATCH_RELEASE 0x00000001 @@ -323,6 +379,18 @@ static inline uint32_t A2XX_MH_MMU_CONFIG_PA_W_CLNT_BEHAVIOR(enum adreno_mmu_cln } #define REG_A2XX_MH_MMU_VA_RANGE 0x00000041 +#define A2XX_MH_MMU_VA_RANGE_NUM_64KB_REGIONS__MASK 0x00000fff +#define A2XX_MH_MMU_VA_RANGE_NUM_64KB_REGIONS__SHIFT 0 +static inline uint32_t A2XX_MH_MMU_VA_RANGE_NUM_64KB_REGIONS(uint32_t val) +{ + return ((val) << A2XX_MH_MMU_VA_RANGE_NUM_64KB_REGIONS__SHIFT) & A2XX_MH_MMU_VA_RANGE_NUM_64KB_REGIONS__MASK; +} +#define A2XX_MH_MMU_VA_RANGE_VA_BASE__MASK 0xfffff000 +#define A2XX_MH_MMU_VA_RANGE_VA_BASE__SHIFT 12 +static inline uint32_t A2XX_MH_MMU_VA_RANGE_VA_BASE(uint32_t val) +{ + return ((val) << A2XX_MH_MMU_VA_RANGE_VA_BASE__SHIFT) & A2XX_MH_MMU_VA_RANGE_VA_BASE__MASK; +} #define REG_A2XX_MH_MMU_PT_BASE 0x00000042 @@ -331,6 +399,8 @@ static inline uint32_t A2XX_MH_MMU_CONFIG_PA_W_CLNT_BEHAVIOR(enum adreno_mmu_cln #define REG_A2XX_MH_MMU_TRAN_ERROR 0x00000044 #define REG_A2XX_MH_MMU_INVALIDATE 0x00000045 +#define A2XX_MH_MMU_INVALIDATE_INVALIDATE_ALL 0x00000001 +#define A2XX_MH_MMU_INVALIDATE_INVALIDATE_TC 0x00000002 #define REG_A2XX_MH_MMU_MPU_BASE 0x00000046 @@ -389,12 +459,19 @@ static inline uint32_t A2XX_MH_MMU_CONFIG_PA_W_CLNT_BEHAVIOR(enum adreno_mmu_cln #define REG_A2XX_RBBM_READ_ERROR 0x000003b3 #define REG_A2XX_RBBM_INT_CNTL 0x000003b4 +#define A2XX_RBBM_INT_CNTL_RDERR_INT_MASK 0x00000001 +#define A2XX_RBBM_INT_CNTL_DISPLAY_UPDATE_INT_MASK 0x00000002 +#define A2XX_RBBM_INT_CNTL_GUI_IDLE_INT_MASK 0x00080000 #define REG_A2XX_RBBM_INT_STATUS 0x000003b5 #define REG_A2XX_RBBM_INT_ACK 0x000003b6 #define REG_A2XX_MASTER_INT_SIGNAL 0x000003b7 +#define A2XX_MASTER_INT_SIGNAL_MH_INT_STAT 0x00000020 +#define A2XX_MASTER_INT_SIGNAL_SQ_INT_STAT 0x04000000 +#define A2XX_MASTER_INT_SIGNAL_CP_INT_STAT 0x40000000 +#define A2XX_MASTER_INT_SIGNAL_RBBM_INT_STAT 0x80000000 #define REG_A2XX_RBBM_PERIPHID1 0x000003f9 @@ -467,6 +544,19 @@ static inline uint32_t A2XX_MH_ARBITER_CONFIG_IN_FLIGHT_LIMIT(uint32_t val) #define A2XX_MH_ARBITER_CONFIG_RB_CLNT_ENABLE 0x02000000 #define A2XX_MH_ARBITER_CONFIG_PA_CLNT_ENABLE 0x04000000 +#define REG_A2XX_MH_INTERRUPT_MASK 0x00000a42 +#define A2XX_MH_INTERRUPT_MASK_AXI_READ_ERROR 0x00000001 +#define A2XX_MH_INTERRUPT_MASK_AXI_WRITE_ERROR 0x00000002 +#define A2XX_MH_INTERRUPT_MASK_MMU_PAGE_FAULT 0x00000004 + +#define REG_A2XX_MH_INTERRUPT_STATUS 0x00000a43 + +#define REG_A2XX_MH_INTERRUPT_CLEAR 0x00000a44 + +#define REG_A2XX_MH_CLNT_INTF_CTRL_CONFIG1 0x00000a54 + +#define REG_A2XX_MH_CLNT_INTF_CTRL_CONFIG2 0x00000a55 + #define REG_A2XX_A220_VSC_BIN_SIZE 0x00000c01 #define A2XX_A220_VSC_BIN_SIZE_WIDTH__MASK 0x0000001f #define A2XX_A220_VSC_BIN_SIZE_WIDTH__SHIFT 0 @@ -648,6 +738,18 @@ static inline uint32_t A2XX_RB_BC_CONTROL_MEM_EXPORT_TIMEOUT_SELECT(uint32_t val #define REG_A2XX_RB_DEBUG_DATA 0x00000f27 #define REG_A2XX_RB_SURFACE_INFO 0x00002000 +#define A2XX_RB_SURFACE_INFO_SURFACE_PITCH__MASK 0x00003fff +#define A2XX_RB_SURFACE_INFO_SURFACE_PITCH__SHIFT 0 +static inline uint32_t A2XX_RB_SURFACE_INFO_SURFACE_PITCH(uint32_t val) +{ + return ((val) << A2XX_RB_SURFACE_INFO_SURFACE_PITCH__SHIFT) & A2XX_RB_SURFACE_INFO_SURFACE_PITCH__MASK; +} +#define A2XX_RB_SURFACE_INFO_MSAA_SAMPLES__MASK 0x0000c000 +#define A2XX_RB_SURFACE_INFO_MSAA_SAMPLES__SHIFT 14 +static inline uint32_t A2XX_RB_SURFACE_INFO_MSAA_SAMPLES(uint32_t val) +{ + return ((val) << A2XX_RB_SURFACE_INFO_MSAA_SAMPLES__SHIFT) & A2XX_RB_SURFACE_INFO_MSAA_SAMPLES__MASK; +} #define REG_A2XX_RB_COLOR_INFO 0x00002001 #define A2XX_RB_COLOR_INFO_FORMAT__MASK 0x0000000f @@ -679,7 +781,7 @@ static inline uint32_t A2XX_RB_COLOR_INFO_SWAP(uint32_t val) #define A2XX_RB_COLOR_INFO_BASE__SHIFT 12 static inline uint32_t A2XX_RB_COLOR_INFO_BASE(uint32_t val) { - return ((val >> 10) << A2XX_RB_COLOR_INFO_BASE__SHIFT) & A2XX_RB_COLOR_INFO_BASE__MASK; + return ((val >> 12) << A2XX_RB_COLOR_INFO_BASE__SHIFT) & A2XX_RB_COLOR_INFO_BASE__MASK; } #define REG_A2XX_RB_DEPTH_INFO 0x00002002 @@ -693,7 +795,7 @@ static inline uint32_t A2XX_RB_DEPTH_INFO_DEPTH_FORMAT(enum adreno_rb_depth_form #define A2XX_RB_DEPTH_INFO_DEPTH_BASE__SHIFT 12 static inline uint32_t A2XX_RB_DEPTH_INFO_DEPTH_BASE(uint32_t val) { - return ((val >> 10) << A2XX_RB_DEPTH_INFO_DEPTH_BASE__SHIFT) & A2XX_RB_DEPTH_INFO_DEPTH_BASE__MASK; + return ((val >> 12) << A2XX_RB_DEPTH_INFO_DEPTH_BASE__SHIFT) & A2XX_RB_DEPTH_INFO_DEPTH_BASE__MASK; } #define REG_A2XX_A225_RB_COLOR_INFO3 0x00002005 @@ -1757,6 +1859,36 @@ static inline uint32_t A2XX_RB_COPY_DEST_OFFSET_Y(uint32_t val) #define REG_A2XX_COHER_STATUS_PM4 0x00000a2b #define REG_A2XX_SQ_TEX_0 0x00000000 +#define A2XX_SQ_TEX_0_TYPE__MASK 0x00000003 +#define A2XX_SQ_TEX_0_TYPE__SHIFT 0 +static inline uint32_t A2XX_SQ_TEX_0_TYPE(enum sq_tex_type val) +{ + return ((val) << A2XX_SQ_TEX_0_TYPE__SHIFT) & A2XX_SQ_TEX_0_TYPE__MASK; +} +#define A2XX_SQ_TEX_0_SIGN_X__MASK 0x0000000c +#define A2XX_SQ_TEX_0_SIGN_X__SHIFT 2 +static inline uint32_t A2XX_SQ_TEX_0_SIGN_X(enum sq_tex_sign val) +{ + return ((val) << A2XX_SQ_TEX_0_SIGN_X__SHIFT) & A2XX_SQ_TEX_0_SIGN_X__MASK; +} +#define A2XX_SQ_TEX_0_SIGN_Y__MASK 0x00000030 +#define A2XX_SQ_TEX_0_SIGN_Y__SHIFT 4 +static inline uint32_t A2XX_SQ_TEX_0_SIGN_Y(enum sq_tex_sign val) +{ + return ((val) << A2XX_SQ_TEX_0_SIGN_Y__SHIFT) & A2XX_SQ_TEX_0_SIGN_Y__MASK; +} +#define A2XX_SQ_TEX_0_SIGN_Z__MASK 0x000000c0 +#define A2XX_SQ_TEX_0_SIGN_Z__SHIFT 6 +static inline uint32_t A2XX_SQ_TEX_0_SIGN_Z(enum sq_tex_sign val) +{ + return ((val) << A2XX_SQ_TEX_0_SIGN_Z__SHIFT) & A2XX_SQ_TEX_0_SIGN_Z__MASK; +} +#define A2XX_SQ_TEX_0_SIGN_W__MASK 0x00000300 +#define A2XX_SQ_TEX_0_SIGN_W__SHIFT 8 +static inline uint32_t A2XX_SQ_TEX_0_SIGN_W(enum sq_tex_sign val) +{ + return ((val) << A2XX_SQ_TEX_0_SIGN_W__SHIFT) & A2XX_SQ_TEX_0_SIGN_W__MASK; +} #define A2XX_SQ_TEX_0_CLAMP_X__MASK 0x00001c00 #define A2XX_SQ_TEX_0_CLAMP_X__SHIFT 10 static inline uint32_t A2XX_SQ_TEX_0_CLAMP_X(enum sq_tex_clamp val) @@ -1775,14 +1907,46 @@ static inline uint32_t A2XX_SQ_TEX_0_CLAMP_Z(enum sq_tex_clamp val) { return ((val) << A2XX_SQ_TEX_0_CLAMP_Z__SHIFT) & A2XX_SQ_TEX_0_CLAMP_Z__MASK; } -#define A2XX_SQ_TEX_0_PITCH__MASK 0xffc00000 +#define A2XX_SQ_TEX_0_PITCH__MASK 0x7fc00000 #define A2XX_SQ_TEX_0_PITCH__SHIFT 22 static inline uint32_t A2XX_SQ_TEX_0_PITCH(uint32_t val) { return ((val >> 5) << A2XX_SQ_TEX_0_PITCH__SHIFT) & A2XX_SQ_TEX_0_PITCH__MASK; } +#define A2XX_SQ_TEX_0_TILED 0x00000002 #define REG_A2XX_SQ_TEX_1 0x00000001 +#define A2XX_SQ_TEX_1_FORMAT__MASK 0x0000003f +#define A2XX_SQ_TEX_1_FORMAT__SHIFT 0 +static inline uint32_t A2XX_SQ_TEX_1_FORMAT(enum a2xx_sq_surfaceformat val) +{ + return ((val) << A2XX_SQ_TEX_1_FORMAT__SHIFT) & A2XX_SQ_TEX_1_FORMAT__MASK; +} +#define A2XX_SQ_TEX_1_ENDIANNESS__MASK 0x000000c0 +#define A2XX_SQ_TEX_1_ENDIANNESS__SHIFT 6 +static inline uint32_t A2XX_SQ_TEX_1_ENDIANNESS(enum sq_tex_endian val) +{ + return ((val) << A2XX_SQ_TEX_1_ENDIANNESS__SHIFT) & A2XX_SQ_TEX_1_ENDIANNESS__MASK; +} +#define A2XX_SQ_TEX_1_REQUEST_SIZE__MASK 0x00000300 +#define A2XX_SQ_TEX_1_REQUEST_SIZE__SHIFT 8 +static inline uint32_t A2XX_SQ_TEX_1_REQUEST_SIZE(uint32_t val) +{ + return ((val) << A2XX_SQ_TEX_1_REQUEST_SIZE__SHIFT) & A2XX_SQ_TEX_1_REQUEST_SIZE__MASK; +} +#define A2XX_SQ_TEX_1_STACKED 0x00000400 +#define A2XX_SQ_TEX_1_CLAMP_POLICY__MASK 0x00000800 +#define A2XX_SQ_TEX_1_CLAMP_POLICY__SHIFT 11 +static inline uint32_t A2XX_SQ_TEX_1_CLAMP_POLICY(enum sq_tex_clamp_policy val) +{ + return ((val) << A2XX_SQ_TEX_1_CLAMP_POLICY__SHIFT) & A2XX_SQ_TEX_1_CLAMP_POLICY__MASK; +} +#define A2XX_SQ_TEX_1_BASE_ADDRESS__MASK 0xfffff000 +#define A2XX_SQ_TEX_1_BASE_ADDRESS__SHIFT 12 +static inline uint32_t A2XX_SQ_TEX_1_BASE_ADDRESS(uint32_t val) +{ + return ((val >> 12) << A2XX_SQ_TEX_1_BASE_ADDRESS__SHIFT) & A2XX_SQ_TEX_1_BASE_ADDRESS__MASK; +} #define REG_A2XX_SQ_TEX_2 0x00000002 #define A2XX_SQ_TEX_2_WIDTH__MASK 0x00001fff @@ -1797,8 +1961,20 @@ static inline uint32_t A2XX_SQ_TEX_2_HEIGHT(uint32_t val) { return ((val) << A2XX_SQ_TEX_2_HEIGHT__SHIFT) & A2XX_SQ_TEX_2_HEIGHT__MASK; } +#define A2XX_SQ_TEX_2_DEPTH__MASK 0xfc000000 +#define A2XX_SQ_TEX_2_DEPTH__SHIFT 26 +static inline uint32_t A2XX_SQ_TEX_2_DEPTH(uint32_t val) +{ + return ((val) << A2XX_SQ_TEX_2_DEPTH__SHIFT) & A2XX_SQ_TEX_2_DEPTH__MASK; +} #define REG_A2XX_SQ_TEX_3 0x00000003 +#define A2XX_SQ_TEX_3_NUM_FORMAT__MASK 0x00000001 +#define A2XX_SQ_TEX_3_NUM_FORMAT__SHIFT 0 +static inline uint32_t A2XX_SQ_TEX_3_NUM_FORMAT(enum sq_tex_num_format val) +{ + return ((val) << A2XX_SQ_TEX_3_NUM_FORMAT__SHIFT) & A2XX_SQ_TEX_3_NUM_FORMAT__MASK; +} #define A2XX_SQ_TEX_3_SWIZ_X__MASK 0x0000000e #define A2XX_SQ_TEX_3_SWIZ_X__SHIFT 1 static inline uint32_t A2XX_SQ_TEX_3_SWIZ_X(enum sq_tex_swiz val) @@ -1823,6 +1999,12 @@ static inline uint32_t A2XX_SQ_TEX_3_SWIZ_W(enum sq_tex_swiz val) { return ((val) << A2XX_SQ_TEX_3_SWIZ_W__SHIFT) & A2XX_SQ_TEX_3_SWIZ_W__MASK; } +#define A2XX_SQ_TEX_3_EXP_ADJUST__MASK 0x0007e000 +#define A2XX_SQ_TEX_3_EXP_ADJUST__SHIFT 13 +static inline uint32_t A2XX_SQ_TEX_3_EXP_ADJUST(uint32_t val) +{ + return ((val) << A2XX_SQ_TEX_3_EXP_ADJUST__SHIFT) & A2XX_SQ_TEX_3_EXP_ADJUST__MASK; +} #define A2XX_SQ_TEX_3_XY_MAG_FILTER__MASK 0x00180000 #define A2XX_SQ_TEX_3_XY_MAG_FILTER__SHIFT 19 static inline uint32_t A2XX_SQ_TEX_3_XY_MAG_FILTER(enum sq_tex_filter val) @@ -1835,6 +2017,104 @@ static inline uint32_t A2XX_SQ_TEX_3_XY_MIN_FILTER(enum sq_tex_filter val) { return ((val) << A2XX_SQ_TEX_3_XY_MIN_FILTER__SHIFT) & A2XX_SQ_TEX_3_XY_MIN_FILTER__MASK; } +#define A2XX_SQ_TEX_3_MIP_FILTER__MASK 0x01800000 +#define A2XX_SQ_TEX_3_MIP_FILTER__SHIFT 23 +static inline uint32_t A2XX_SQ_TEX_3_MIP_FILTER(enum sq_tex_filter val) +{ + return ((val) << A2XX_SQ_TEX_3_MIP_FILTER__SHIFT) & A2XX_SQ_TEX_3_MIP_FILTER__MASK; +} +#define A2XX_SQ_TEX_3_ANISO_FILTER__MASK 0x0e000000 +#define A2XX_SQ_TEX_3_ANISO_FILTER__SHIFT 25 +static inline uint32_t A2XX_SQ_TEX_3_ANISO_FILTER(enum sq_tex_aniso_filter val) +{ + return ((val) << A2XX_SQ_TEX_3_ANISO_FILTER__SHIFT) & A2XX_SQ_TEX_3_ANISO_FILTER__MASK; +} +#define A2XX_SQ_TEX_3_BORDER_SIZE__MASK 0x80000000 +#define A2XX_SQ_TEX_3_BORDER_SIZE__SHIFT 31 +static inline uint32_t A2XX_SQ_TEX_3_BORDER_SIZE(uint32_t val) +{ + return ((val) << A2XX_SQ_TEX_3_BORDER_SIZE__SHIFT) & A2XX_SQ_TEX_3_BORDER_SIZE__MASK; +} + +#define REG_A2XX_SQ_TEX_4 0x00000004 +#define A2XX_SQ_TEX_4_VOL_MAG_FILTER__MASK 0x00000001 +#define A2XX_SQ_TEX_4_VOL_MAG_FILTER__SHIFT 0 +static inline uint32_t A2XX_SQ_TEX_4_VOL_MAG_FILTER(enum sq_tex_filter val) +{ + return ((val) << A2XX_SQ_TEX_4_VOL_MAG_FILTER__SHIFT) & A2XX_SQ_TEX_4_VOL_MAG_FILTER__MASK; +} +#define A2XX_SQ_TEX_4_VOL_MIN_FILTER__MASK 0x00000002 +#define A2XX_SQ_TEX_4_VOL_MIN_FILTER__SHIFT 1 +static inline uint32_t A2XX_SQ_TEX_4_VOL_MIN_FILTER(enum sq_tex_filter val) +{ + return ((val) << A2XX_SQ_TEX_4_VOL_MIN_FILTER__SHIFT) & A2XX_SQ_TEX_4_VOL_MIN_FILTER__MASK; +} +#define A2XX_SQ_TEX_4_MIP_MIN_LEVEL__MASK 0x0000003c +#define A2XX_SQ_TEX_4_MIP_MIN_LEVEL__SHIFT 2 +static inline uint32_t A2XX_SQ_TEX_4_MIP_MIN_LEVEL(uint32_t val) +{ + return ((val) << A2XX_SQ_TEX_4_MIP_MIN_LEVEL__SHIFT) & A2XX_SQ_TEX_4_MIP_MIN_LEVEL__MASK; +} +#define A2XX_SQ_TEX_4_MIP_MAX_LEVEL__MASK 0x000003c0 +#define A2XX_SQ_TEX_4_MIP_MAX_LEVEL__SHIFT 6 +static inline uint32_t A2XX_SQ_TEX_4_MIP_MAX_LEVEL(uint32_t val) +{ + return ((val) << A2XX_SQ_TEX_4_MIP_MAX_LEVEL__SHIFT) & A2XX_SQ_TEX_4_MIP_MAX_LEVEL__MASK; +} +#define A2XX_SQ_TEX_4_MAX_ANISO_WALK 0x00000400 +#define A2XX_SQ_TEX_4_MIN_ANISO_WALK 0x00000800 +#define A2XX_SQ_TEX_4_LOD_BIAS__MASK 0x003ff000 +#define A2XX_SQ_TEX_4_LOD_BIAS__SHIFT 12 +static inline uint32_t A2XX_SQ_TEX_4_LOD_BIAS(float val) +{ + return ((((int32_t)(val * 32.0))) << A2XX_SQ_TEX_4_LOD_BIAS__SHIFT) & A2XX_SQ_TEX_4_LOD_BIAS__MASK; +} +#define A2XX_SQ_TEX_4_GRAD_EXP_ADJUST_H__MASK 0x07c00000 +#define A2XX_SQ_TEX_4_GRAD_EXP_ADJUST_H__SHIFT 22 +static inline uint32_t A2XX_SQ_TEX_4_GRAD_EXP_ADJUST_H(uint32_t val) +{ + return ((val) << A2XX_SQ_TEX_4_GRAD_EXP_ADJUST_H__SHIFT) & A2XX_SQ_TEX_4_GRAD_EXP_ADJUST_H__MASK; +} +#define A2XX_SQ_TEX_4_GRAD_EXP_ADJUST_V__MASK 0xf8000000 +#define A2XX_SQ_TEX_4_GRAD_EXP_ADJUST_V__SHIFT 27 +static inline uint32_t A2XX_SQ_TEX_4_GRAD_EXP_ADJUST_V(uint32_t val) +{ + return ((val) << A2XX_SQ_TEX_4_GRAD_EXP_ADJUST_V__SHIFT) & A2XX_SQ_TEX_4_GRAD_EXP_ADJUST_V__MASK; +} + +#define REG_A2XX_SQ_TEX_5 0x00000005 +#define A2XX_SQ_TEX_5_BORDER_COLOR__MASK 0x00000003 +#define A2XX_SQ_TEX_5_BORDER_COLOR__SHIFT 0 +static inline uint32_t A2XX_SQ_TEX_5_BORDER_COLOR(enum sq_tex_border_color val) +{ + return ((val) << A2XX_SQ_TEX_5_BORDER_COLOR__SHIFT) & A2XX_SQ_TEX_5_BORDER_COLOR__MASK; +} +#define A2XX_SQ_TEX_5_FORCE_BCW_MAX 0x00000004 +#define A2XX_SQ_TEX_5_TRI_CLAMP__MASK 0x00000018 +#define A2XX_SQ_TEX_5_TRI_CLAMP__SHIFT 3 +static inline uint32_t A2XX_SQ_TEX_5_TRI_CLAMP(uint32_t val) +{ + return ((val) << A2XX_SQ_TEX_5_TRI_CLAMP__SHIFT) & A2XX_SQ_TEX_5_TRI_CLAMP__MASK; +} +#define A2XX_SQ_TEX_5_ANISO_BIAS__MASK 0x000001e0 +#define A2XX_SQ_TEX_5_ANISO_BIAS__SHIFT 5 +static inline uint32_t A2XX_SQ_TEX_5_ANISO_BIAS(float val) +{ + return ((((int32_t)(val * 1.0))) << A2XX_SQ_TEX_5_ANISO_BIAS__SHIFT) & A2XX_SQ_TEX_5_ANISO_BIAS__MASK; +} +#define A2XX_SQ_TEX_5_DIMENSION__MASK 0x00000600 +#define A2XX_SQ_TEX_5_DIMENSION__SHIFT 9 +static inline uint32_t A2XX_SQ_TEX_5_DIMENSION(enum sq_tex_dimension val) +{ + return ((val) << A2XX_SQ_TEX_5_DIMENSION__SHIFT) & A2XX_SQ_TEX_5_DIMENSION__MASK; +} +#define A2XX_SQ_TEX_5_PACKED_MIPS 0x00000800 +#define A2XX_SQ_TEX_5_MIP_ADDRESS__MASK 0xfffff000 +#define A2XX_SQ_TEX_5_MIP_ADDRESS__SHIFT 12 +static inline uint32_t A2XX_SQ_TEX_5_MIP_ADDRESS(uint32_t val) +{ + return ((val >> 12) << A2XX_SQ_TEX_5_MIP_ADDRESS__SHIFT) & A2XX_SQ_TEX_5_MIP_ADDRESS__MASK; +} #endif /* A2XX_XML */ diff --git a/drivers/gpu/drm/msm/adreno/a2xx_gpu.c b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c new file mode 100644 index 000000000000..1f83bc18d500 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a2xx_gpu.c @@ -0,0 +1,492 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2018 The Linux Foundation. All rights reserved. */ + +#include "a2xx_gpu.h" +#include "msm_gem.h" +#include "msm_mmu.h" + +extern bool hang_debug; + +static void a2xx_dump(struct msm_gpu *gpu); +static bool a2xx_idle(struct msm_gpu *gpu); + +static bool a2xx_me_init(struct msm_gpu *gpu) +{ + struct msm_ringbuffer *ring = gpu->rb[0]; + + OUT_PKT3(ring, CP_ME_INIT, 18); + + /* All fields present (bits 9:0) */ + OUT_RING(ring, 0x000003ff); + /* Disable/Enable Real-Time Stream processing (present but ignored) */ + OUT_RING(ring, 0x00000000); + /* Enable (2D <-> 3D) implicit synchronization (present but ignored) */ + OUT_RING(ring, 0x00000000); + + OUT_RING(ring, REG_A2XX_RB_SURFACE_INFO - 0x2000); + OUT_RING(ring, REG_A2XX_PA_SC_WINDOW_OFFSET - 0x2000); + OUT_RING(ring, REG_A2XX_VGT_MAX_VTX_INDX - 0x2000); + OUT_RING(ring, REG_A2XX_SQ_PROGRAM_CNTL - 0x2000); + OUT_RING(ring, REG_A2XX_RB_DEPTHCONTROL - 0x2000); + OUT_RING(ring, REG_A2XX_PA_SU_POINT_SIZE - 0x2000); + OUT_RING(ring, REG_A2XX_PA_SC_LINE_CNTL - 0x2000); + OUT_RING(ring, REG_A2XX_PA_SU_POLY_OFFSET_FRONT_SCALE - 0x2000); + + /* Vertex and Pixel Shader Start Addresses in instructions + * (3 DWORDS per instruction) */ + OUT_RING(ring, 0x80000180); + /* Maximum Contexts */ + OUT_RING(ring, 0x00000001); + /* Write Confirm Interval and The CP will wait the + * wait_interval * 16 clocks between polling */ + OUT_RING(ring, 0x00000000); + /* NQ and External Memory Swap */ + OUT_RING(ring, 0x00000000); + /* protected mode error checking (0x1f2 is REG_AXXX_CP_INT_CNTL) */ + OUT_RING(ring, 0x200001f2); + /* Disable header dumping and Header dump address */ + OUT_RING(ring, 0x00000000); + /* Header dump size */ + OUT_RING(ring, 0x00000000); + + /* enable protected mode */ + OUT_PKT3(ring, CP_SET_PROTECTED_MODE, 1); + OUT_RING(ring, 1); + + gpu->funcs->flush(gpu, ring); + return a2xx_idle(gpu); +} + +static int a2xx_hw_init(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + dma_addr_t pt_base, tran_error; + uint32_t *ptr, len; + int i, ret; + + msm_gpummu_params(gpu->aspace->mmu, &pt_base, &tran_error); + + DBG("%s", gpu->name); + + /* halt ME to avoid ucode upload issues on a20x */ + gpu_write(gpu, REG_AXXX_CP_ME_CNTL, AXXX_CP_ME_CNTL_HALT); + + gpu_write(gpu, REG_A2XX_RBBM_PM_OVERRIDE1, 0xfffffffe); + gpu_write(gpu, REG_A2XX_RBBM_PM_OVERRIDE2, 0xffffffff); + + /* note: kgsl uses 0x00000001 after first reset on a22x */ + gpu_write(gpu, REG_A2XX_RBBM_SOFT_RESET, 0xffffffff); + msleep(30); + gpu_write(gpu, REG_A2XX_RBBM_SOFT_RESET, 0x00000000); + + if (adreno_is_a225(adreno_gpu)) + gpu_write(gpu, REG_A2XX_SQ_FLOW_CONTROL, 0x18000000); + + /* note: kgsl uses 0x0000ffff for a20x */ + gpu_write(gpu, REG_A2XX_RBBM_CNTL, 0x00004442); + + /* MPU: physical range */ + gpu_write(gpu, REG_A2XX_MH_MMU_MPU_BASE, 0x00000000); + gpu_write(gpu, REG_A2XX_MH_MMU_MPU_END, 0xfffff000); + + gpu_write(gpu, REG_A2XX_MH_MMU_CONFIG, A2XX_MH_MMU_CONFIG_MMU_ENABLE | + A2XX_MH_MMU_CONFIG_RB_W_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_CP_W_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_CP_R0_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_CP_R1_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_CP_R2_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_CP_R3_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_CP_R4_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_VGT_R0_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_VGT_R1_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_TC_R_CLNT_BEHAVIOR(BEH_TRAN_RNG) | + A2XX_MH_MMU_CONFIG_PA_W_CLNT_BEHAVIOR(BEH_TRAN_RNG)); + + /* same as parameters in adreno_gpu */ + gpu_write(gpu, REG_A2XX_MH_MMU_VA_RANGE, SZ_16M | + A2XX_MH_MMU_VA_RANGE_NUM_64KB_REGIONS(0xfff)); + + gpu_write(gpu, REG_A2XX_MH_MMU_PT_BASE, pt_base); + gpu_write(gpu, REG_A2XX_MH_MMU_TRAN_ERROR, tran_error); + + gpu_write(gpu, REG_A2XX_MH_MMU_INVALIDATE, + A2XX_MH_MMU_INVALIDATE_INVALIDATE_ALL | + A2XX_MH_MMU_INVALIDATE_INVALIDATE_TC); + + gpu_write(gpu, REG_A2XX_MH_ARBITER_CONFIG, + A2XX_MH_ARBITER_CONFIG_SAME_PAGE_LIMIT(16) | + A2XX_MH_ARBITER_CONFIG_L1_ARB_ENABLE | + A2XX_MH_ARBITER_CONFIG_L1_ARB_HOLD_ENABLE | + A2XX_MH_ARBITER_CONFIG_PAGE_SIZE(1) | + A2XX_MH_ARBITER_CONFIG_TC_REORDER_ENABLE | + A2XX_MH_ARBITER_CONFIG_TC_ARB_HOLD_ENABLE | + A2XX_MH_ARBITER_CONFIG_IN_FLIGHT_LIMIT_ENABLE | + A2XX_MH_ARBITER_CONFIG_IN_FLIGHT_LIMIT(8) | + A2XX_MH_ARBITER_CONFIG_CP_CLNT_ENABLE | + A2XX_MH_ARBITER_CONFIG_VGT_CLNT_ENABLE | + A2XX_MH_ARBITER_CONFIG_TC_CLNT_ENABLE | + A2XX_MH_ARBITER_CONFIG_RB_CLNT_ENABLE | + A2XX_MH_ARBITER_CONFIG_PA_CLNT_ENABLE); + if (!adreno_is_a20x(adreno_gpu)) + gpu_write(gpu, REG_A2XX_MH_CLNT_INTF_CTRL_CONFIG1, 0x00032f07); + + gpu_write(gpu, REG_A2XX_SQ_VS_PROGRAM, 0x00000000); + gpu_write(gpu, REG_A2XX_SQ_PS_PROGRAM, 0x00000000); + + gpu_write(gpu, REG_A2XX_RBBM_PM_OVERRIDE1, 0); /* 0x200 for msm8960? */ + gpu_write(gpu, REG_A2XX_RBBM_PM_OVERRIDE2, 0); /* 0x80/0x1a0 for a22x? */ + + /* note: gsl doesn't set this */ + gpu_write(gpu, REG_A2XX_RBBM_DEBUG, 0x00080000); + + gpu_write(gpu, REG_A2XX_RBBM_INT_CNTL, + A2XX_RBBM_INT_CNTL_RDERR_INT_MASK); + gpu_write(gpu, REG_AXXX_CP_INT_CNTL, + AXXX_CP_INT_CNTL_T0_PACKET_IN_IB_MASK | + AXXX_CP_INT_CNTL_OPCODE_ERROR_MASK | + AXXX_CP_INT_CNTL_PROTECTED_MODE_ERROR_MASK | + AXXX_CP_INT_CNTL_RESERVED_BIT_ERROR_MASK | + AXXX_CP_INT_CNTL_IB_ERROR_MASK | + AXXX_CP_INT_CNTL_IB1_INT_MASK | + AXXX_CP_INT_CNTL_RB_INT_MASK); + gpu_write(gpu, REG_A2XX_SQ_INT_CNTL, 0); + gpu_write(gpu, REG_A2XX_MH_INTERRUPT_MASK, + A2XX_MH_INTERRUPT_MASK_AXI_READ_ERROR | + A2XX_MH_INTERRUPT_MASK_AXI_WRITE_ERROR | + A2XX_MH_INTERRUPT_MASK_MMU_PAGE_FAULT); + + for (i = 3; i <= 5; i++) + if ((SZ_16K << i) == adreno_gpu->gmem) + break; + gpu_write(gpu, REG_A2XX_RB_EDRAM_INFO, i); + + ret = adreno_hw_init(gpu); + if (ret) + return ret; + + /* NOTE: PM4/micro-engine firmware registers look to be the same + * for a2xx and a3xx.. we could possibly push that part down to + * adreno_gpu base class. Or push both PM4 and PFP but + * parameterize the pfp ucode addr/data registers.. + */ + + /* Load PM4: */ + ptr = (uint32_t *)(adreno_gpu->fw[ADRENO_FW_PM4]->data); + len = adreno_gpu->fw[ADRENO_FW_PM4]->size / 4; + DBG("loading PM4 ucode version: %x", ptr[1]); + + gpu_write(gpu, REG_AXXX_CP_DEBUG, + AXXX_CP_DEBUG_MIU_128BIT_WRITE_ENABLE); + gpu_write(gpu, REG_AXXX_CP_ME_RAM_WADDR, 0); + for (i = 1; i < len; i++) + gpu_write(gpu, REG_AXXX_CP_ME_RAM_DATA, ptr[i]); + + /* Load PFP: */ + ptr = (uint32_t *)(adreno_gpu->fw[ADRENO_FW_PFP]->data); + len = adreno_gpu->fw[ADRENO_FW_PFP]->size / 4; + DBG("loading PFP ucode version: %x", ptr[5]); + + gpu_write(gpu, REG_A2XX_CP_PFP_UCODE_ADDR, 0); + for (i = 1; i < len; i++) + gpu_write(gpu, REG_A2XX_CP_PFP_UCODE_DATA, ptr[i]); + + gpu_write(gpu, REG_AXXX_CP_QUEUE_THRESHOLDS, 0x000C0804); + + /* clear ME_HALT to start micro engine */ + gpu_write(gpu, REG_AXXX_CP_ME_CNTL, 0); + + return a2xx_me_init(gpu) ? 0 : -EINVAL; +} + +static void a2xx_recover(struct msm_gpu *gpu) +{ + int i; + + adreno_dump_info(gpu); + + for (i = 0; i < 8; i++) { + printk("CP_SCRATCH_REG%d: %u\n", i, + gpu_read(gpu, REG_AXXX_CP_SCRATCH_REG0 + i)); + } + + /* dump registers before resetting gpu, if enabled: */ + if (hang_debug) + a2xx_dump(gpu); + + gpu_write(gpu, REG_A2XX_RBBM_SOFT_RESET, 1); + gpu_read(gpu, REG_A2XX_RBBM_SOFT_RESET); + gpu_write(gpu, REG_A2XX_RBBM_SOFT_RESET, 0); + adreno_recover(gpu); +} + +static void a2xx_destroy(struct msm_gpu *gpu) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a2xx_gpu *a2xx_gpu = to_a2xx_gpu(adreno_gpu); + + DBG("%s", gpu->name); + + adreno_gpu_cleanup(adreno_gpu); + + kfree(a2xx_gpu); +} + +static bool a2xx_idle(struct msm_gpu *gpu) +{ + /* wait for ringbuffer to drain: */ + if (!adreno_idle(gpu, gpu->rb[0])) + return false; + + /* then wait for GPU to finish: */ + if (spin_until(!(gpu_read(gpu, REG_A2XX_RBBM_STATUS) & + A2XX_RBBM_STATUS_GUI_ACTIVE))) { + DRM_ERROR("%s: timeout waiting for GPU to idle!\n", gpu->name); + + /* TODO maybe we need to reset GPU here to recover from hang? */ + return false; + } + + return true; +} + +static irqreturn_t a2xx_irq(struct msm_gpu *gpu) +{ + uint32_t mstatus, status; + + mstatus = gpu_read(gpu, REG_A2XX_MASTER_INT_SIGNAL); + + if (mstatus & A2XX_MASTER_INT_SIGNAL_MH_INT_STAT) { + status = gpu_read(gpu, REG_A2XX_MH_INTERRUPT_STATUS); + + dev_warn(gpu->dev->dev, "MH_INT: %08X\n", status); + dev_warn(gpu->dev->dev, "MMU_PAGE_FAULT: %08X\n", + gpu_read(gpu, REG_A2XX_MH_MMU_PAGE_FAULT)); + + gpu_write(gpu, REG_A2XX_MH_INTERRUPT_CLEAR, status); + } + + if (mstatus & A2XX_MASTER_INT_SIGNAL_CP_INT_STAT) { + status = gpu_read(gpu, REG_AXXX_CP_INT_STATUS); + + /* only RB_INT is expected */ + if (status & ~AXXX_CP_INT_CNTL_RB_INT_MASK) + dev_warn(gpu->dev->dev, "CP_INT: %08X\n", status); + + gpu_write(gpu, REG_AXXX_CP_INT_ACK, status); + } + + if (mstatus & A2XX_MASTER_INT_SIGNAL_RBBM_INT_STAT) { + status = gpu_read(gpu, REG_A2XX_RBBM_INT_STATUS); + + dev_warn(gpu->dev->dev, "RBBM_INT: %08X\n", status); + + gpu_write(gpu, REG_A2XX_RBBM_INT_ACK, status); + } + + msm_gpu_retire(gpu); + + return IRQ_HANDLED; +} + +static const unsigned int a200_registers[] = { + 0x0000, 0x0002, 0x0004, 0x000B, 0x003B, 0x003D, 0x0040, 0x0044, + 0x0046, 0x0047, 0x01C0, 0x01C1, 0x01C3, 0x01C8, 0x01D5, 0x01D9, + 0x01DC, 0x01DD, 0x01EA, 0x01EA, 0x01EE, 0x01F3, 0x01F6, 0x01F7, + 0x01FC, 0x01FF, 0x0391, 0x0392, 0x039B, 0x039E, 0x03B2, 0x03B5, + 0x03B7, 0x03B7, 0x03F8, 0x03FB, 0x0440, 0x0440, 0x0443, 0x0444, + 0x044B, 0x044B, 0x044D, 0x044F, 0x0452, 0x0452, 0x0454, 0x045B, + 0x047F, 0x047F, 0x0578, 0x0587, 0x05C9, 0x05C9, 0x05D0, 0x05D0, + 0x0601, 0x0604, 0x0606, 0x0609, 0x060B, 0x060E, 0x0613, 0x0614, + 0x0A29, 0x0A2B, 0x0A2F, 0x0A31, 0x0A40, 0x0A43, 0x0A45, 0x0A45, + 0x0A4E, 0x0A4F, 0x0C2C, 0x0C2C, 0x0C30, 0x0C30, 0x0C38, 0x0C3C, + 0x0C40, 0x0C40, 0x0C44, 0x0C44, 0x0C80, 0x0C86, 0x0C88, 0x0C94, + 0x0C99, 0x0C9A, 0x0CA4, 0x0CA5, 0x0D00, 0x0D03, 0x0D06, 0x0D06, + 0x0D08, 0x0D0B, 0x0D34, 0x0D35, 0x0DAE, 0x0DC1, 0x0DC8, 0x0DD4, + 0x0DD8, 0x0DD9, 0x0E00, 0x0E00, 0x0E02, 0x0E04, 0x0E17, 0x0E1E, + 0x0EC0, 0x0EC9, 0x0ECB, 0x0ECC, 0x0ED0, 0x0ED0, 0x0ED4, 0x0ED7, + 0x0EE0, 0x0EE2, 0x0F01, 0x0F02, 0x0F0C, 0x0F0C, 0x0F0E, 0x0F12, + 0x0F26, 0x0F2A, 0x0F2C, 0x0F2C, 0x2000, 0x2002, 0x2006, 0x200F, + 0x2080, 0x2082, 0x2100, 0x2109, 0x210C, 0x2114, 0x2180, 0x2184, + 0x21F5, 0x21F7, 0x2200, 0x2208, 0x2280, 0x2283, 0x2293, 0x2294, + 0x2300, 0x2308, 0x2312, 0x2312, 0x2316, 0x231D, 0x2324, 0x2326, + 0x2380, 0x2383, 0x2400, 0x2402, 0x2406, 0x240F, 0x2480, 0x2482, + 0x2500, 0x2509, 0x250C, 0x2514, 0x2580, 0x2584, 0x25F5, 0x25F7, + 0x2600, 0x2608, 0x2680, 0x2683, 0x2693, 0x2694, 0x2700, 0x2708, + 0x2712, 0x2712, 0x2716, 0x271D, 0x2724, 0x2726, 0x2780, 0x2783, + 0x4000, 0x4003, 0x4800, 0x4805, 0x4900, 0x4900, 0x4908, 0x4908, + ~0 /* sentinel */ +}; + +static const unsigned int a220_registers[] = { + 0x0000, 0x0002, 0x0004, 0x000B, 0x003B, 0x003D, 0x0040, 0x0044, + 0x0046, 0x0047, 0x01C0, 0x01C1, 0x01C3, 0x01C8, 0x01D5, 0x01D9, + 0x01DC, 0x01DD, 0x01EA, 0x01EA, 0x01EE, 0x01F3, 0x01F6, 0x01F7, + 0x01FC, 0x01FF, 0x0391, 0x0392, 0x039B, 0x039E, 0x03B2, 0x03B5, + 0x03B7, 0x03B7, 0x03F8, 0x03FB, 0x0440, 0x0440, 0x0443, 0x0444, + 0x044B, 0x044B, 0x044D, 0x044F, 0x0452, 0x0452, 0x0454, 0x045B, + 0x047F, 0x047F, 0x0578, 0x0587, 0x05C9, 0x05C9, 0x05D0, 0x05D0, + 0x0601, 0x0604, 0x0606, 0x0609, 0x060B, 0x060E, 0x0613, 0x0614, + 0x0A29, 0x0A2B, 0x0A2F, 0x0A31, 0x0A40, 0x0A40, 0x0A42, 0x0A43, + 0x0A45, 0x0A45, 0x0A4E, 0x0A4F, 0x0C30, 0x0C30, 0x0C38, 0x0C39, + 0x0C3C, 0x0C3C, 0x0C80, 0x0C81, 0x0C88, 0x0C93, 0x0D00, 0x0D03, + 0x0D05, 0x0D06, 0x0D08, 0x0D0B, 0x0D34, 0x0D35, 0x0DAE, 0x0DC1, + 0x0DC8, 0x0DD4, 0x0DD8, 0x0DD9, 0x0E00, 0x0E00, 0x0E02, 0x0E04, + 0x0E17, 0x0E1E, 0x0EC0, 0x0EC9, 0x0ECB, 0x0ECC, 0x0ED0, 0x0ED0, + 0x0ED4, 0x0ED7, 0x0EE0, 0x0EE2, 0x0F01, 0x0F02, 0x2000, 0x2002, + 0x2006, 0x200F, 0x2080, 0x2082, 0x2100, 0x2102, 0x2104, 0x2109, + 0x210C, 0x2114, 0x2180, 0x2184, 0x21F5, 0x21F7, 0x2200, 0x2202, + 0x2204, 0x2204, 0x2208, 0x2208, 0x2280, 0x2282, 0x2294, 0x2294, + 0x2300, 0x2308, 0x2309, 0x230A, 0x2312, 0x2312, 0x2316, 0x2316, + 0x2318, 0x231D, 0x2324, 0x2326, 0x2380, 0x2383, 0x2400, 0x2402, + 0x2406, 0x240F, 0x2480, 0x2482, 0x2500, 0x2502, 0x2504, 0x2509, + 0x250C, 0x2514, 0x2580, 0x2584, 0x25F5, 0x25F7, 0x2600, 0x2602, + 0x2604, 0x2606, 0x2608, 0x2608, 0x2680, 0x2682, 0x2694, 0x2694, + 0x2700, 0x2708, 0x2712, 0x2712, 0x2716, 0x2716, 0x2718, 0x271D, + 0x2724, 0x2726, 0x2780, 0x2783, 0x4000, 0x4003, 0x4800, 0x4805, + 0x4900, 0x4900, 0x4908, 0x4908, + ~0 /* sentinel */ +}; + +static const unsigned int a225_registers[] = { + 0x0000, 0x0002, 0x0004, 0x000B, 0x003B, 0x003D, 0x0040, 0x0044, + 0x0046, 0x0047, 0x013C, 0x013C, 0x0140, 0x014F, 0x01C0, 0x01C1, + 0x01C3, 0x01C8, 0x01D5, 0x01D9, 0x01DC, 0x01DD, 0x01EA, 0x01EA, + 0x01EE, 0x01F3, 0x01F6, 0x01F7, 0x01FC, 0x01FF, 0x0391, 0x0392, + 0x039B, 0x039E, 0x03B2, 0x03B5, 0x03B7, 0x03B7, 0x03F8, 0x03FB, + 0x0440, 0x0440, 0x0443, 0x0444, 0x044B, 0x044B, 0x044D, 0x044F, + 0x0452, 0x0452, 0x0454, 0x045B, 0x047F, 0x047F, 0x0578, 0x0587, + 0x05C9, 0x05C9, 0x05D0, 0x05D0, 0x0601, 0x0604, 0x0606, 0x0609, + 0x060B, 0x060E, 0x0613, 0x0614, 0x0A29, 0x0A2B, 0x0A2F, 0x0A31, + 0x0A40, 0x0A40, 0x0A42, 0x0A43, 0x0A45, 0x0A45, 0x0A4E, 0x0A4F, + 0x0C01, 0x0C1D, 0x0C30, 0x0C30, 0x0C38, 0x0C39, 0x0C3C, 0x0C3C, + 0x0C80, 0x0C81, 0x0C88, 0x0C93, 0x0D00, 0x0D03, 0x0D05, 0x0D06, + 0x0D08, 0x0D0B, 0x0D34, 0x0D35, 0x0DAE, 0x0DC1, 0x0DC8, 0x0DD4, + 0x0DD8, 0x0DD9, 0x0E00, 0x0E00, 0x0E02, 0x0E04, 0x0E17, 0x0E1E, + 0x0EC0, 0x0EC9, 0x0ECB, 0x0ECC, 0x0ED0, 0x0ED0, 0x0ED4, 0x0ED7, + 0x0EE0, 0x0EE2, 0x0F01, 0x0F02, 0x2000, 0x200F, 0x2080, 0x2082, + 0x2100, 0x2109, 0x210C, 0x2114, 0x2180, 0x2184, 0x21F5, 0x21F7, + 0x2200, 0x2202, 0x2204, 0x2206, 0x2208, 0x2210, 0x2220, 0x2222, + 0x2280, 0x2282, 0x2294, 0x2294, 0x2297, 0x2297, 0x2300, 0x230A, + 0x2312, 0x2312, 0x2315, 0x2316, 0x2318, 0x231D, 0x2324, 0x2326, + 0x2340, 0x2357, 0x2360, 0x2360, 0x2380, 0x2383, 0x2400, 0x240F, + 0x2480, 0x2482, 0x2500, 0x2509, 0x250C, 0x2514, 0x2580, 0x2584, + 0x25F5, 0x25F7, 0x2600, 0x2602, 0x2604, 0x2606, 0x2608, 0x2610, + 0x2620, 0x2622, 0x2680, 0x2682, 0x2694, 0x2694, 0x2697, 0x2697, + 0x2700, 0x270A, 0x2712, 0x2712, 0x2715, 0x2716, 0x2718, 0x271D, + 0x2724, 0x2726, 0x2740, 0x2757, 0x2760, 0x2760, 0x2780, 0x2783, + 0x4000, 0x4003, 0x4800, 0x4806, 0x4808, 0x4808, 0x4900, 0x4900, + 0x4908, 0x4908, + ~0 /* sentinel */ +}; + +/* would be nice to not have to duplicate the _show() stuff with printk(): */ +static void a2xx_dump(struct msm_gpu *gpu) +{ + printk("status: %08x\n", + gpu_read(gpu, REG_A2XX_RBBM_STATUS)); + adreno_dump(gpu); +} + +static struct msm_gpu_state *a2xx_gpu_state_get(struct msm_gpu *gpu) +{ + struct msm_gpu_state *state = kzalloc(sizeof(*state), GFP_KERNEL); + + if (!state) + return ERR_PTR(-ENOMEM); + + adreno_gpu_state_get(gpu, state); + + state->rbbm_status = gpu_read(gpu, REG_A2XX_RBBM_STATUS); + + return state; +} + +/* Register offset defines for A2XX - copy of A3XX */ +static const unsigned int a2xx_register_offsets[REG_ADRENO_REGISTER_MAX] = { + REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_BASE, REG_AXXX_CP_RB_BASE), + REG_ADRENO_SKIP(REG_ADRENO_CP_RB_BASE_HI), + REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_RPTR_ADDR, REG_AXXX_CP_RB_RPTR_ADDR), + REG_ADRENO_SKIP(REG_ADRENO_CP_RB_RPTR_ADDR_HI), + REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_RPTR, REG_AXXX_CP_RB_RPTR), + REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_WPTR, REG_AXXX_CP_RB_WPTR), + REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_CNTL, REG_AXXX_CP_RB_CNTL), +}; + +static const struct adreno_gpu_funcs funcs = { + .base = { + .get_param = adreno_get_param, + .hw_init = a2xx_hw_init, + .pm_suspend = msm_gpu_pm_suspend, + .pm_resume = msm_gpu_pm_resume, + .recover = a2xx_recover, + .submit = adreno_submit, + .flush = adreno_flush, + .active_ring = adreno_active_ring, + .irq = a2xx_irq, + .destroy = a2xx_destroy, +#if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP) + .show = adreno_show, +#endif + .gpu_state_get = a2xx_gpu_state_get, + .gpu_state_put = adreno_gpu_state_put, + }, +}; + +static const struct msm_gpu_perfcntr perfcntrs[] = { +/* TODO */ +}; + +struct msm_gpu *a2xx_gpu_init(struct drm_device *dev) +{ + struct a2xx_gpu *a2xx_gpu = NULL; + struct adreno_gpu *adreno_gpu; + struct msm_gpu *gpu; + struct msm_drm_private *priv = dev->dev_private; + struct platform_device *pdev = priv->gpu_pdev; + int ret; + + if (!pdev) { + dev_err(dev->dev, "no a2xx device\n"); + ret = -ENXIO; + goto fail; + } + + a2xx_gpu = kzalloc(sizeof(*a2xx_gpu), GFP_KERNEL); + if (!a2xx_gpu) { + ret = -ENOMEM; + goto fail; + } + + adreno_gpu = &a2xx_gpu->base; + gpu = &adreno_gpu->base; + + gpu->perfcntrs = perfcntrs; + gpu->num_perfcntrs = ARRAY_SIZE(perfcntrs); + + if (adreno_is_a20x(adreno_gpu)) + adreno_gpu->registers = a200_registers; + else if (adreno_is_a225(adreno_gpu)) + adreno_gpu->registers = a225_registers; + else + adreno_gpu->registers = a220_registers; + + adreno_gpu->reg_offsets = a2xx_register_offsets; + + ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 1); + if (ret) + goto fail; + + if (!gpu->aspace) { + dev_err(dev->dev, "No memory protection without MMU\n"); + ret = -ENXIO; + goto fail; + } + + return gpu; + +fail: + if (a2xx_gpu) + a2xx_destroy(&a2xx_gpu->base.base); + + return ERR_PTR(ret); +} diff --git a/drivers/gpu/drm/msm/adreno/a2xx_gpu.h b/drivers/gpu/drm/msm/adreno/a2xx_gpu.h new file mode 100644 index 000000000000..02fba2cb8932 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a2xx_gpu.h @@ -0,0 +1,21 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2018 The Linux Foundation. All rights reserved. */ + +#ifndef __A2XX_GPU_H__ +#define __A2XX_GPU_H__ + +#include "adreno_gpu.h" + +/* arrg, somehow fb.h is getting pulled in: */ +#undef ROP_COPY +#undef ROP_XOR + +#include "a2xx.xml.h" + +struct a2xx_gpu { + struct adreno_gpu base; + bool pm_enabled; +}; +#define to_a2xx_gpu(x) container_of(x, struct a2xx_gpu, base) + +#endif /* __A2XX_GPU_H__ */ diff --git a/drivers/gpu/drm/msm/adreno/a3xx.xml.h b/drivers/gpu/drm/msm/adreno/a3xx.xml.h index a89f7bb8b5cc..17059f242a98 100644 --- a/drivers/gpu/drm/msm/adreno/a3xx.xml.h +++ b/drivers/gpu/drm/msm/adreno/a3xx.xml.h @@ -10,13 +10,13 @@ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: - /home/robclark/src/envytools/rnndb/adreno.xml ( 501 bytes, from 2018-07-03 19:37:13) - /home/robclark/src/envytools/rnndb/freedreno_copyright.xml ( 1572 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/a2xx.xml ( 36805 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/adreno_common.xml ( 13634 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/adreno_pm4.xml ( 42585 bytes, from 2018-10-04 19:06:37) +- /home/robclark/src/envytools/rnndb/adreno/a2xx.xml ( 42463 bytes, from 2018-11-19 13:44:03) +- /home/robclark/src/envytools/rnndb/adreno/adreno_common.xml ( 14201 bytes, from 2018-12-02 17:29:54) +- /home/robclark/src/envytools/rnndb/adreno/adreno_pm4.xml ( 43052 bytes, from 2018-12-02 17:29:54) - /home/robclark/src/envytools/rnndb/adreno/a3xx.xml ( 83840 bytes, from 2018-07-03 19:37:13) - /home/robclark/src/envytools/rnndb/adreno/a4xx.xml ( 112086 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/a5xx.xml ( 147240 bytes, from 2018-10-04 19:06:37) -- /home/robclark/src/envytools/rnndb/adreno/a6xx.xml ( 139581 bytes, from 2018-10-04 19:06:42) +- /home/robclark/src/envytools/rnndb/adreno/a5xx.xml ( 147240 bytes, from 2018-12-02 17:29:54) +- /home/robclark/src/envytools/rnndb/adreno/a6xx.xml ( 140790 bytes, from 2018-12-02 17:29:54) - /home/robclark/src/envytools/rnndb/adreno/a6xx_gmu.xml ( 10431 bytes, from 2018-09-14 13:03:07) - /home/robclark/src/envytools/rnndb/adreno/ocmem.xml ( 1773 bytes, from 2018-07-03 19:37:13) diff --git a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c index 669c2d4b070d..c3b4bc6e4155 100644 --- a/drivers/gpu/drm/msm/adreno/a3xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a3xx_gpu.c @@ -481,7 +481,7 @@ struct msm_gpu *a3xx_gpu_init(struct drm_device *dev) int ret; if (!pdev) { - dev_err(dev->dev, "no a3xx device\n"); + DRM_DEV_ERROR(dev->dev, "no a3xx device\n"); ret = -ENXIO; goto fail; } @@ -528,7 +528,7 @@ struct msm_gpu *a3xx_gpu_init(struct drm_device *dev) * to not be possible to restrict access, then we must * implement a cmdstream validator. */ - dev_err(dev->dev, "No memory protection without IOMMU\n"); + DRM_DEV_ERROR(dev->dev, "No memory protection without IOMMU\n"); ret = -ENXIO; goto fail; } diff --git a/drivers/gpu/drm/msm/adreno/a4xx.xml.h b/drivers/gpu/drm/msm/adreno/a4xx.xml.h index 858690f52854..9b51e25a9583 100644 --- a/drivers/gpu/drm/msm/adreno/a4xx.xml.h +++ b/drivers/gpu/drm/msm/adreno/a4xx.xml.h @@ -10,13 +10,13 @@ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: - /home/robclark/src/envytools/rnndb/adreno.xml ( 501 bytes, from 2018-07-03 19:37:13) - /home/robclark/src/envytools/rnndb/freedreno_copyright.xml ( 1572 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/a2xx.xml ( 36805 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/adreno_common.xml ( 13634 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/adreno_pm4.xml ( 42585 bytes, from 2018-10-04 19:06:37) +- /home/robclark/src/envytools/rnndb/adreno/a2xx.xml ( 42463 bytes, from 2018-11-19 13:44:03) +- /home/robclark/src/envytools/rnndb/adreno/adreno_common.xml ( 14201 bytes, from 2018-12-02 17:29:54) +- /home/robclark/src/envytools/rnndb/adreno/adreno_pm4.xml ( 43052 bytes, from 2018-12-02 17:29:54) - /home/robclark/src/envytools/rnndb/adreno/a3xx.xml ( 83840 bytes, from 2018-07-03 19:37:13) - /home/robclark/src/envytools/rnndb/adreno/a4xx.xml ( 112086 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/a5xx.xml ( 147240 bytes, from 2018-10-04 19:06:37) -- /home/robclark/src/envytools/rnndb/adreno/a6xx.xml ( 139581 bytes, from 2018-10-04 19:06:42) +- /home/robclark/src/envytools/rnndb/adreno/a5xx.xml ( 147240 bytes, from 2018-12-02 17:29:54) +- /home/robclark/src/envytools/rnndb/adreno/a6xx.xml ( 140790 bytes, from 2018-12-02 17:29:54) - /home/robclark/src/envytools/rnndb/adreno/a6xx_gmu.xml ( 10431 bytes, from 2018-09-14 13:03:07) - /home/robclark/src/envytools/rnndb/adreno/ocmem.xml ( 1773 bytes, from 2018-07-03 19:37:13) diff --git a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c index 7c4e6dc1ed59..18f9a8e0bf3b 100644 --- a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c @@ -561,7 +561,7 @@ struct msm_gpu *a4xx_gpu_init(struct drm_device *dev) int ret; if (!pdev) { - dev_err(dev->dev, "no a4xx device\n"); + DRM_DEV_ERROR(dev->dev, "no a4xx device\n"); ret = -ENXIO; goto fail; } @@ -608,7 +608,7 @@ struct msm_gpu *a4xx_gpu_init(struct drm_device *dev) * to not be possible to restrict access, then we must * implement a cmdstream validator. */ - dev_err(dev->dev, "No memory protection without IOMMU\n"); + DRM_DEV_ERROR(dev->dev, "No memory protection without IOMMU\n"); ret = -ENXIO; goto fail; } diff --git a/drivers/gpu/drm/msm/adreno/a5xx.xml.h b/drivers/gpu/drm/msm/adreno/a5xx.xml.h index b4944cc0e62f..cf4fe14ddd6e 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx.xml.h +++ b/drivers/gpu/drm/msm/adreno/a5xx.xml.h @@ -10,13 +10,13 @@ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: - /home/robclark/src/envytools/rnndb/adreno.xml ( 501 bytes, from 2018-07-03 19:37:13) - /home/robclark/src/envytools/rnndb/freedreno_copyright.xml ( 1572 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/a2xx.xml ( 36805 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/adreno_common.xml ( 13634 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/adreno_pm4.xml ( 42585 bytes, from 2018-10-04 19:06:37) +- /home/robclark/src/envytools/rnndb/adreno/a2xx.xml ( 42463 bytes, from 2018-11-19 13:44:03) +- /home/robclark/src/envytools/rnndb/adreno/adreno_common.xml ( 14201 bytes, from 2018-12-02 17:29:54) +- /home/robclark/src/envytools/rnndb/adreno/adreno_pm4.xml ( 43052 bytes, from 2018-12-02 17:29:54) - /home/robclark/src/envytools/rnndb/adreno/a3xx.xml ( 83840 bytes, from 2018-07-03 19:37:13) - /home/robclark/src/envytools/rnndb/adreno/a4xx.xml ( 112086 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/a5xx.xml ( 147240 bytes, from 2018-10-04 19:06:37) -- /home/robclark/src/envytools/rnndb/adreno/a6xx.xml ( 139581 bytes, from 2018-10-04 19:06:42) +- /home/robclark/src/envytools/rnndb/adreno/a5xx.xml ( 147240 bytes, from 2018-12-02 17:29:54) +- /home/robclark/src/envytools/rnndb/adreno/a6xx.xml ( 140790 bytes, from 2018-12-02 17:29:54) - /home/robclark/src/envytools/rnndb/adreno/a6xx_gmu.xml ( 10431 bytes, from 2018-09-14 13:03:07) - /home/robclark/src/envytools/rnndb/adreno/ocmem.xml ( 1773 bytes, from 2018-07-03 19:37:13) diff --git a/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c b/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c index d2127b1c4ece..d9af3aff690f 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_debugfs.c @@ -130,15 +130,13 @@ reset_set(void *data, u64 val) adreno_gpu->fw[ADRENO_FW_PFP] = NULL; if (a5xx_gpu->pm4_bo) { - if (a5xx_gpu->pm4_iova) - msm_gem_put_iova(a5xx_gpu->pm4_bo, gpu->aspace); + msm_gem_unpin_iova(a5xx_gpu->pm4_bo, gpu->aspace); drm_gem_object_put(a5xx_gpu->pm4_bo); a5xx_gpu->pm4_bo = NULL; } if (a5xx_gpu->pfp_bo) { - if (a5xx_gpu->pfp_iova) - msm_gem_put_iova(a5xx_gpu->pfp_bo, gpu->aspace); + msm_gem_unpin_iova(a5xx_gpu->pfp_bo, gpu->aspace); drm_gem_object_put(a5xx_gpu->pfp_bo); a5xx_gpu->pfp_bo = NULL; } @@ -173,7 +171,7 @@ int a5xx_debugfs_init(struct msm_gpu *gpu, struct drm_minor *minor) minor->debugfs_root, minor); if (ret) { - dev_err(dev->dev, "could not install a5xx_debugfs_list\n"); + DRM_DEV_ERROR(dev->dev, "could not install a5xx_debugfs_list\n"); return ret; } diff --git a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c index 8edd80bb0428..d5f5e56422f5 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_gpu.c @@ -20,7 +20,6 @@ #include <linux/soc/qcom/mdt_loader.h> #include <linux/pm_opp.h> #include <linux/nvmem-consumer.h> -#include <linux/iopoll.h> #include <linux/slab.h> #include "msm_gem.h" #include "msm_mmu.h" @@ -511,13 +510,16 @@ static int a5xx_ucode_init(struct msm_gpu *gpu) a5xx_gpu->pm4_bo = adreno_fw_create_bo(gpu, adreno_gpu->fw[ADRENO_FW_PM4], &a5xx_gpu->pm4_iova); + if (IS_ERR(a5xx_gpu->pm4_bo)) { ret = PTR_ERR(a5xx_gpu->pm4_bo); a5xx_gpu->pm4_bo = NULL; - dev_err(gpu->dev->dev, "could not allocate PM4: %d\n", + DRM_DEV_ERROR(gpu->dev->dev, "could not allocate PM4: %d\n", ret); return ret; } + + msm_gem_object_set_name(a5xx_gpu->pm4_bo, "pm4fw"); } if (!a5xx_gpu->pfp_bo) { @@ -527,10 +529,12 @@ static int a5xx_ucode_init(struct msm_gpu *gpu) if (IS_ERR(a5xx_gpu->pfp_bo)) { ret = PTR_ERR(a5xx_gpu->pfp_bo); a5xx_gpu->pfp_bo = NULL; - dev_err(gpu->dev->dev, "could not allocate PFP: %d\n", + DRM_DEV_ERROR(gpu->dev->dev, "could not allocate PFP: %d\n", ret); return ret; } + + msm_gem_object_set_name(a5xx_gpu->pfp_bo, "pfpfw"); } gpu_write64(gpu, REG_A5XX_CP_ME_INSTR_BASE_LO, @@ -841,20 +845,17 @@ static void a5xx_destroy(struct msm_gpu *gpu) a5xx_preempt_fini(gpu); if (a5xx_gpu->pm4_bo) { - if (a5xx_gpu->pm4_iova) - msm_gem_put_iova(a5xx_gpu->pm4_bo, gpu->aspace); + msm_gem_unpin_iova(a5xx_gpu->pm4_bo, gpu->aspace); drm_gem_object_put_unlocked(a5xx_gpu->pm4_bo); } if (a5xx_gpu->pfp_bo) { - if (a5xx_gpu->pfp_iova) - msm_gem_put_iova(a5xx_gpu->pfp_bo, gpu->aspace); + msm_gem_unpin_iova(a5xx_gpu->pfp_bo, gpu->aspace); drm_gem_object_put_unlocked(a5xx_gpu->pfp_bo); } if (a5xx_gpu->gpmu_bo) { - if (a5xx_gpu->gpmu_iova) - msm_gem_put_iova(a5xx_gpu->gpmu_bo, gpu->aspace); + msm_gem_unpin_iova(a5xx_gpu->gpmu_bo, gpu->aspace); drm_gem_object_put_unlocked(a5xx_gpu->gpmu_bo); } @@ -1028,7 +1029,7 @@ static void a5xx_fault_detect_irq(struct msm_gpu *gpu) struct msm_drm_private *priv = dev->dev_private; struct msm_ringbuffer *ring = gpu->funcs->active_ring(gpu); - dev_err(dev->dev, "gpu fault ring %d fence %x status %8.8X rb %4.4x/%4.4x ib1 %16.16llX/%4.4x ib2 %16.16llX/%4.4x\n", + DRM_DEV_ERROR(dev->dev, "gpu fault ring %d fence %x status %8.8X rb %4.4x/%4.4x ib1 %16.16llX/%4.4x ib2 %16.16llX/%4.4x\n", ring ? ring->id : -1, ring ? ring->seqno : 0, gpu_read(gpu, REG_A5XX_RBBM_STATUS), gpu_read(gpu, REG_A5XX_CP_RB_RPTR), @@ -1134,7 +1135,7 @@ static const u32 a5xx_registers[] = { static void a5xx_dump(struct msm_gpu *gpu) { - dev_info(gpu->dev->dev, "status: %08x\n", + DRM_DEV_INFO(gpu->dev->dev, "status: %08x\n", gpu_read(gpu, REG_A5XX_RBBM_STATUS)); adreno_dump(gpu); } @@ -1211,10 +1212,6 @@ struct a5xx_gpu_state { u32 *hlsqregs; }; -#define gpu_poll_timeout(gpu, addr, val, cond, interval, timeout) \ - readl_poll_timeout((gpu)->mmio + ((addr) << 2), val, cond, \ - interval, timeout) - static int a5xx_crashdumper_init(struct msm_gpu *gpu, struct a5xx_crashdumper *dumper) { @@ -1222,16 +1219,10 @@ static int a5xx_crashdumper_init(struct msm_gpu *gpu, SZ_1M, MSM_BO_UNCACHED, gpu->aspace, &dumper->bo, &dumper->iova); - return PTR_ERR_OR_ZERO(dumper->ptr); -} - -static void a5xx_crashdumper_free(struct msm_gpu *gpu, - struct a5xx_crashdumper *dumper) -{ - msm_gem_put_iova(dumper->bo, gpu->aspace); - msm_gem_put_vaddr(dumper->bo); + if (!IS_ERR(dumper->ptr)) + msm_gem_object_set_name(dumper->bo, "crashdump"); - drm_gem_object_put(dumper->bo); + return PTR_ERR_OR_ZERO(dumper->ptr); } static int a5xx_crashdumper_run(struct msm_gpu *gpu, @@ -1326,7 +1317,7 @@ static void a5xx_gpu_state_get_hlsq_regs(struct msm_gpu *gpu, if (a5xx_crashdumper_run(gpu, &dumper)) { kfree(a5xx_state->hlsqregs); - a5xx_crashdumper_free(gpu, &dumper); + msm_gem_kernel_put(dumper.bo, gpu->aspace, true); return; } @@ -1334,7 +1325,7 @@ static void a5xx_gpu_state_get_hlsq_regs(struct msm_gpu *gpu, memcpy(a5xx_state->hlsqregs, dumper.ptr + (256 * SZ_1K), count * sizeof(u32)); - a5xx_crashdumper_free(gpu, &dumper); + msm_gem_kernel_put(dumper.bo, gpu->aspace, true); } static struct msm_gpu_state *a5xx_gpu_state_get(struct msm_gpu *gpu) @@ -1505,7 +1496,7 @@ struct msm_gpu *a5xx_gpu_init(struct drm_device *dev) int ret; if (!pdev) { - dev_err(dev->dev, "No A5XX device is defined\n"); + DRM_DEV_ERROR(dev->dev, "No A5XX device is defined\n"); return ERR_PTR(-ENXIO); } diff --git a/drivers/gpu/drm/msm/adreno/a5xx_power.c b/drivers/gpu/drm/msm/adreno/a5xx_power.c index 7a41e1c147e4..70e65c94e525 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_power.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_power.c @@ -298,7 +298,9 @@ void a5xx_gpmu_ucode_init(struct msm_gpu *gpu) MSM_BO_UNCACHED | MSM_BO_GPU_READONLY, gpu->aspace, &a5xx_gpu->gpmu_bo, &a5xx_gpu->gpmu_iova); if (IS_ERR(ptr)) - goto err; + return; + + msm_gem_object_set_name(a5xx_gpu->gpmu_bo, "gpmufw"); while (cmds_size > 0) { int i; @@ -317,15 +319,4 @@ void a5xx_gpmu_ucode_init(struct msm_gpu *gpu) msm_gem_put_vaddr(a5xx_gpu->gpmu_bo); a5xx_gpu->gpmu_dwords = dwords; - - return; -err: - if (a5xx_gpu->gpmu_iova) - msm_gem_put_iova(a5xx_gpu->gpmu_bo, gpu->aspace); - if (a5xx_gpu->gpmu_bo) - drm_gem_object_put(a5xx_gpu->gpmu_bo); - - a5xx_gpu->gpmu_bo = NULL; - a5xx_gpu->gpmu_iova = 0; - a5xx_gpu->gpmu_dwords = 0; } diff --git a/drivers/gpu/drm/msm/adreno/a5xx_preempt.c b/drivers/gpu/drm/msm/adreno/a5xx_preempt.c index 4c357ead1be6..3d62310a535f 100644 --- a/drivers/gpu/drm/msm/adreno/a5xx_preempt.c +++ b/drivers/gpu/drm/msm/adreno/a5xx_preempt.c @@ -92,7 +92,7 @@ static void a5xx_preempt_timer(struct timer_list *t) if (!try_preempt_state(a5xx_gpu, PREEMPT_TRIGGERED, PREEMPT_FAULTED)) return; - dev_err(dev->dev, "%s: preemption timed out\n", gpu->name); + DRM_DEV_ERROR(dev->dev, "%s: preemption timed out\n", gpu->name); queue_work(priv->wq, &gpu->recover_work); } @@ -188,7 +188,7 @@ void a5xx_preempt_irq(struct msm_gpu *gpu) status = gpu_read(gpu, REG_A5XX_CP_CONTEXT_SWITCH_CNTL); if (unlikely(status)) { set_preempt_state(a5xx_gpu, PREEMPT_FAULTED); - dev_err(dev->dev, "%s: Preemption failed to complete\n", + DRM_DEV_ERROR(dev->dev, "%s: Preemption failed to complete\n", gpu->name); queue_work(priv->wq, &gpu->recover_work); return; @@ -245,6 +245,8 @@ static int preempt_init_ring(struct a5xx_gpu *a5xx_gpu, if (IS_ERR(ptr)) return PTR_ERR(ptr); + msm_gem_object_set_name(bo, "preempt"); + a5xx_gpu->preempt_bo[ring->id] = bo; a5xx_gpu->preempt_iova[ring->id] = iova; a5xx_gpu->preempt[ring->id] = ptr; @@ -267,18 +269,8 @@ void a5xx_preempt_fini(struct msm_gpu *gpu) struct a5xx_gpu *a5xx_gpu = to_a5xx_gpu(adreno_gpu); int i; - for (i = 0; i < gpu->nr_rings; i++) { - if (!a5xx_gpu->preempt_bo[i]) - continue; - - msm_gem_put_vaddr(a5xx_gpu->preempt_bo[i]); - - if (a5xx_gpu->preempt_iova[i]) - msm_gem_put_iova(a5xx_gpu->preempt_bo[i], gpu->aspace); - - drm_gem_object_put(a5xx_gpu->preempt_bo[i]); - a5xx_gpu->preempt_bo[i] = NULL; - } + for (i = 0; i < gpu->nr_rings; i++) + msm_gem_kernel_put(a5xx_gpu->preempt_bo[i], gpu->aspace, true); } void a5xx_preempt_init(struct msm_gpu *gpu) diff --git a/drivers/gpu/drm/msm/adreno/a6xx.xml.h b/drivers/gpu/drm/msm/adreno/a6xx.xml.h index a6f7c40454a6..f44553ec3193 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx.xml.h +++ b/drivers/gpu/drm/msm/adreno/a6xx.xml.h @@ -10,13 +10,13 @@ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: - /home/robclark/src/envytools/rnndb/adreno.xml ( 501 bytes, from 2018-07-03 19:37:13) - /home/robclark/src/envytools/rnndb/freedreno_copyright.xml ( 1572 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/a2xx.xml ( 36805 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/adreno_common.xml ( 13634 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/adreno_pm4.xml ( 42585 bytes, from 2018-10-04 19:06:37) +- /home/robclark/src/envytools/rnndb/adreno/a2xx.xml ( 42463 bytes, from 2018-11-19 13:44:03) +- /home/robclark/src/envytools/rnndb/adreno/adreno_common.xml ( 14201 bytes, from 2018-12-02 17:29:54) +- /home/robclark/src/envytools/rnndb/adreno/adreno_pm4.xml ( 43052 bytes, from 2018-12-02 17:29:54) - /home/robclark/src/envytools/rnndb/adreno/a3xx.xml ( 83840 bytes, from 2018-07-03 19:37:13) - /home/robclark/src/envytools/rnndb/adreno/a4xx.xml ( 112086 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/a5xx.xml ( 147240 bytes, from 2018-10-04 19:06:37) -- /home/robclark/src/envytools/rnndb/adreno/a6xx.xml ( 139581 bytes, from 2018-10-04 19:06:42) +- /home/robclark/src/envytools/rnndb/adreno/a5xx.xml ( 147240 bytes, from 2018-12-02 17:29:54) +- /home/robclark/src/envytools/rnndb/adreno/a6xx.xml ( 140790 bytes, from 2018-12-02 17:29:54) - /home/robclark/src/envytools/rnndb/adreno/a6xx_gmu.xml ( 10431 bytes, from 2018-09-14 13:03:07) - /home/robclark/src/envytools/rnndb/adreno/ocmem.xml ( 1773 bytes, from 2018-07-03 19:37:13) @@ -501,7 +501,7 @@ enum a6xx_vfd_perfcounter_select { PERF_VFDP_VS_STAGE_WAVES = 22, }; -enum a6xx_hslq_perfcounter_select { +enum a6xx_hlsq_perfcounter_select { PERF_HLSQ_BUSY_CYCLES = 0, PERF_HLSQ_STALL_CYCLES_UCHE = 1, PERF_HLSQ_STALL_CYCLES_SP_STATE = 2, @@ -2959,6 +2959,8 @@ static inline uint32_t A6XX_GRAS_SC_WINDOW_SCISSOR_BR_Y(uint32_t val) #define A6XX_GRAS_LRZ_CNTL_ENABLE 0x00000001 #define A6XX_GRAS_LRZ_CNTL_LRZ_WRITE 0x00000002 #define A6XX_GRAS_LRZ_CNTL_GREATER 0x00000004 +#define A6XX_GRAS_LRZ_CNTL_UNK3 0x00000008 +#define A6XX_GRAS_LRZ_CNTL_UNK4 0x00000010 #define REG_A6XX_GRAS_UNKNOWN_8101 0x00008101 @@ -2997,6 +2999,13 @@ static inline uint32_t A6XX_GRAS_LRZ_BUFFER_PITCH_ARRAY_PITCH(uint32_t val) #define REG_A6XX_GRAS_UNKNOWN_8110 0x00008110 #define REG_A6XX_GRAS_2D_BLIT_CNTL 0x00008400 +#define A6XX_GRAS_2D_BLIT_CNTL_COLOR_FORMAT__MASK 0x0000ff00 +#define A6XX_GRAS_2D_BLIT_CNTL_COLOR_FORMAT__SHIFT 8 +static inline uint32_t A6XX_GRAS_2D_BLIT_CNTL_COLOR_FORMAT(enum a6xx_color_fmt val) +{ + return ((val) << A6XX_GRAS_2D_BLIT_CNTL_COLOR_FORMAT__SHIFT) & A6XX_GRAS_2D_BLIT_CNTL_COLOR_FORMAT__MASK; +} +#define A6XX_GRAS_2D_BLIT_CNTL_SCISSOR 0x00010000 #define REG_A6XX_GRAS_2D_SRC_TL_X 0x00008401 #define A6XX_GRAS_2D_SRC_TL_X_X__MASK 0x00ffff00 @@ -3449,6 +3458,7 @@ static inline uint32_t A6XX_RB_BLEND_CNTL_ENABLE_BLEND(uint32_t val) return ((val) << A6XX_RB_BLEND_CNTL_ENABLE_BLEND__SHIFT) & A6XX_RB_BLEND_CNTL_ENABLE_BLEND__MASK; } #define A6XX_RB_BLEND_CNTL_INDEPENDENT_BLEND 0x00000100 +#define A6XX_RB_BLEND_CNTL_ALPHA_TO_COVERAGE 0x00000400 #define A6XX_RB_BLEND_CNTL_SAMPLE_MASK__MASK 0xffff0000 #define A6XX_RB_BLEND_CNTL_SAMPLE_MASK__SHIFT 16 static inline uint32_t A6XX_RB_BLEND_CNTL_SAMPLE_MASK(uint32_t val) @@ -3642,6 +3652,9 @@ static inline uint32_t A6XX_RB_WINDOW_OFFSET_Y(uint32_t val) #define REG_A6XX_RB_SAMPLE_COUNT_CONTROL 0x00008891 #define A6XX_RB_SAMPLE_COUNT_CONTROL_COPY 0x00000002 +#define REG_A6XX_RB_LRZ_CNTL 0x00008898 +#define A6XX_RB_LRZ_CNTL_ENABLE 0x00000001 + #define REG_A6XX_RB_UNKNOWN_88D0 0x000088d0 #define REG_A6XX_RB_BLIT_SCISSOR_TL 0x000088d1 @@ -3674,6 +3687,14 @@ static inline uint32_t A6XX_RB_BLIT_SCISSOR_BR_Y(uint32_t val) return ((val) << A6XX_RB_BLIT_SCISSOR_BR_Y__SHIFT) & A6XX_RB_BLIT_SCISSOR_BR_Y__MASK; } +#define REG_A6XX_RB_MSAA_CNTL 0x000088d5 +#define A6XX_RB_MSAA_CNTL_SAMPLES__MASK 0x00000018 +#define A6XX_RB_MSAA_CNTL_SAMPLES__SHIFT 3 +static inline uint32_t A6XX_RB_MSAA_CNTL_SAMPLES(enum a3xx_msaa_samples val) +{ + return ((val) << A6XX_RB_MSAA_CNTL_SAMPLES__SHIFT) & A6XX_RB_MSAA_CNTL_SAMPLES__MASK; +} + #define REG_A6XX_RB_BLIT_BASE_GMEM 0x000088d6 #define REG_A6XX_RB_BLIT_DST_INFO 0x000088d7 @@ -3684,6 +3705,12 @@ static inline uint32_t A6XX_RB_BLIT_DST_INFO_TILE_MODE(enum a6xx_tile_mode val) return ((val) << A6XX_RB_BLIT_DST_INFO_TILE_MODE__SHIFT) & A6XX_RB_BLIT_DST_INFO_TILE_MODE__MASK; } #define A6XX_RB_BLIT_DST_INFO_FLAGS 0x00000004 +#define A6XX_RB_BLIT_DST_INFO_SAMPLES__MASK 0x00000018 +#define A6XX_RB_BLIT_DST_INFO_SAMPLES__SHIFT 3 +static inline uint32_t A6XX_RB_BLIT_DST_INFO_SAMPLES(enum a3xx_msaa_samples val) +{ + return ((val) << A6XX_RB_BLIT_DST_INFO_SAMPLES__SHIFT) & A6XX_RB_BLIT_DST_INFO_SAMPLES__MASK; +} #define A6XX_RB_BLIT_DST_INFO_COLOR_FORMAT__MASK 0x00007f80 #define A6XX_RB_BLIT_DST_INFO_COLOR_FORMAT__SHIFT 7 static inline uint32_t A6XX_RB_BLIT_DST_INFO_COLOR_FORMAT(enum a6xx_color_fmt val) @@ -3780,6 +3807,9 @@ static inline uint32_t A6XX_RB_2D_BLIT_CNTL_COLOR_FORMAT(enum a6xx_color_fmt val { return ((val) << A6XX_RB_2D_BLIT_CNTL_COLOR_FORMAT__SHIFT) & A6XX_RB_2D_BLIT_CNTL_COLOR_FORMAT__MASK; } +#define A6XX_RB_2D_BLIT_CNTL_SCISSOR 0x00010000 + +#define REG_A6XX_RB_UNKNOWN_8C01 0x00008c01 #define REG_A6XX_RB_2D_DST_INFO 0x00008c17 #define A6XX_RB_2D_DST_INFO_COLOR_FORMAT__MASK 0x000000ff @@ -4465,6 +4495,7 @@ static inline uint32_t A6XX_SP_FS_CTRL_REG0_THREADSIZE(enum a3xx_threadsize val) #define REG_A6XX_SP_BLEND_CNTL 0x0000a989 #define A6XX_SP_BLEND_CNTL_ENABLED 0x00000001 #define A6XX_SP_BLEND_CNTL_UNK8 0x00000100 +#define A6XX_SP_BLEND_CNTL_ALPHA_TO_COVERAGE 0x00000400 #define REG_A6XX_SP_SRGB_CNTL 0x0000a98a #define A6XX_SP_SRGB_CNTL_SRGB_MRT0 0x00000001 @@ -4643,6 +4674,8 @@ static inline uint32_t A6XX_SP_FS_CONFIG_NSAMP(uint32_t val) #define REG_A6XX_SP_UNKNOWN_AB20 0x0000ab20 +#define REG_A6XX_SP_UNKNOWN_ACC0 0x0000acc0 + #define REG_A6XX_SP_UNKNOWN_AE00 0x0000ae00 #define REG_A6XX_SP_UNKNOWN_AE03 0x0000ae03 @@ -4700,11 +4733,34 @@ static inline uint32_t A6XX_SP_PS_2D_SRC_INFO_COLOR_SWAP(enum a3xx_color_swap va return ((val) << A6XX_SP_PS_2D_SRC_INFO_COLOR_SWAP__SHIFT) & A6XX_SP_PS_2D_SRC_INFO_COLOR_SWAP__MASK; } #define A6XX_SP_PS_2D_SRC_INFO_FLAGS 0x00001000 +#define A6XX_SP_PS_2D_SRC_INFO_FILTER 0x00010000 + +#define REG_A6XX_SP_PS_2D_SRC_SIZE 0x0000b4c1 +#define A6XX_SP_PS_2D_SRC_SIZE_WIDTH__MASK 0x00007fff +#define A6XX_SP_PS_2D_SRC_SIZE_WIDTH__SHIFT 0 +static inline uint32_t A6XX_SP_PS_2D_SRC_SIZE_WIDTH(uint32_t val) +{ + return ((val) << A6XX_SP_PS_2D_SRC_SIZE_WIDTH__SHIFT) & A6XX_SP_PS_2D_SRC_SIZE_WIDTH__MASK; +} +#define A6XX_SP_PS_2D_SRC_SIZE_HEIGHT__MASK 0x3fff8000 +#define A6XX_SP_PS_2D_SRC_SIZE_HEIGHT__SHIFT 15 +static inline uint32_t A6XX_SP_PS_2D_SRC_SIZE_HEIGHT(uint32_t val) +{ + return ((val) << A6XX_SP_PS_2D_SRC_SIZE_HEIGHT__SHIFT) & A6XX_SP_PS_2D_SRC_SIZE_HEIGHT__MASK; +} #define REG_A6XX_SP_PS_2D_SRC_LO 0x0000b4c2 #define REG_A6XX_SP_PS_2D_SRC_HI 0x0000b4c3 +#define REG_A6XX_SP_PS_2D_SRC_PITCH 0x0000b4c4 +#define A6XX_SP_PS_2D_SRC_PITCH_PITCH__MASK 0x01fffe00 +#define A6XX_SP_PS_2D_SRC_PITCH_PITCH__SHIFT 9 +static inline uint32_t A6XX_SP_PS_2D_SRC_PITCH_PITCH(uint32_t val) +{ + return ((val >> 6) << A6XX_SP_PS_2D_SRC_PITCH_PITCH__SHIFT) & A6XX_SP_PS_2D_SRC_PITCH_PITCH__MASK; +} + #define REG_A6XX_SP_PS_2D_SRC_FLAGS_LO 0x0000b4ca #define REG_A6XX_SP_PS_2D_SRC_FLAGS_HI 0x0000b4cb @@ -5033,6 +5089,12 @@ static inline uint32_t A6XX_TEX_CONST_0_MIPLVLS(uint32_t val) { return ((val) << A6XX_TEX_CONST_0_MIPLVLS__SHIFT) & A6XX_TEX_CONST_0_MIPLVLS__MASK; } +#define A6XX_TEX_CONST_0_SAMPLES__MASK 0x00300000 +#define A6XX_TEX_CONST_0_SAMPLES__SHIFT 20 +static inline uint32_t A6XX_TEX_CONST_0_SAMPLES(enum a3xx_msaa_samples val) +{ + return ((val) << A6XX_TEX_CONST_0_SAMPLES__SHIFT) & A6XX_TEX_CONST_0_SAMPLES__MASK; +} #define A6XX_TEX_CONST_0_FMT__MASK 0x3fc00000 #define A6XX_TEX_CONST_0_FMT__SHIFT 22 static inline uint32_t A6XX_TEX_CONST_0_FMT(enum a6xx_tex_fmt val) @@ -5365,5 +5427,9 @@ static inline uint32_t A6XX_CX_DBGC_CFG_DBGBUS_BYTEL_1_BYTEL15(uint32_t val) #define REG_A6XX_CX_DBGC_CFG_DBGBUS_TRACE_BUF2 0x00000030 +#define REG_A6XX_CX_MISC_SYSTEM_CACHE_CNTL_0 0x00000001 + +#define REG_A6XX_CX_MISC_SYSTEM_CACHE_CNTL_1 0x00000002 + #endif /* A6XX_XML */ diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c index d4e98e5876bc..5beb83d1cf87 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.c @@ -51,10 +51,31 @@ static irqreturn_t a6xx_hfi_irq(int irq, void *data) return IRQ_HANDLED; } +bool a6xx_gmu_sptprac_is_on(struct a6xx_gmu *gmu) +{ + u32 val; + + /* This can be called from gpu state code so make sure GMU is valid */ + if (IS_ERR_OR_NULL(gmu->mmio)) + return false; + + val = gmu_read(gmu, REG_A6XX_GMU_SPTPRAC_PWR_CLK_STATUS); + + return !(val & + (A6XX_GMU_SPTPRAC_PWR_CLK_STATUS_SPTPRAC_GDSC_POWER_OFF | + A6XX_GMU_SPTPRAC_PWR_CLK_STATUS_SP_CLOCK_OFF)); +} + /* Check to see if the GX rail is still powered */ -static bool a6xx_gmu_gx_is_on(struct a6xx_gmu *gmu) +bool a6xx_gmu_gx_is_on(struct a6xx_gmu *gmu) { - u32 val = gmu_read(gmu, REG_A6XX_GMU_SPTPRAC_PWR_CLK_STATUS); + u32 val; + + /* This can be called from gpu state code so make sure GMU is valid */ + if (IS_ERR_OR_NULL(gmu->mmio)) + return false; + + val = gmu_read(gmu, REG_A6XX_GMU_SPTPRAC_PWR_CLK_STATUS); return !(val & (A6XX_GMU_SPTPRAC_PWR_CLK_STATUS_GX_HM_GDSC_POWER_OFF | @@ -153,7 +174,7 @@ static int a6xx_gmu_start(struct a6xx_gmu *gmu) val == 0xbabeface, 100, 10000); if (ret) - dev_err(gmu->dev, "GMU firmware initialization timed out\n"); + DRM_DEV_ERROR(gmu->dev, "GMU firmware initialization timed out\n"); return ret; } @@ -168,7 +189,7 @@ static int a6xx_gmu_hfi_start(struct a6xx_gmu *gmu) ret = gmu_poll_timeout(gmu, REG_A6XX_GMU_HFI_CTRL_STATUS, val, val & 1, 100, 10000); if (ret) - dev_err(gmu->dev, "Unable to start the HFI queues\n"); + DRM_DEV_ERROR(gmu->dev, "Unable to start the HFI queues\n"); return ret; } @@ -209,7 +230,7 @@ int a6xx_gmu_set_oob(struct a6xx_gmu *gmu, enum a6xx_gmu_oob_state state) val & (1 << ack), 100, 10000); if (ret) - dev_err(gmu->dev, + DRM_DEV_ERROR(gmu->dev, "Timeout waiting for GMU OOB set %s: 0x%x\n", name, gmu_read(gmu, REG_A6XX_GMU_GMU2HOST_INTR_INFO)); @@ -251,7 +272,7 @@ static int a6xx_sptprac_enable(struct a6xx_gmu *gmu) (val & 0x38) == 0x28, 1, 100); if (ret) { - dev_err(gmu->dev, "Unable to power on SPTPRAC: 0x%x\n", + DRM_DEV_ERROR(gmu->dev, "Unable to power on SPTPRAC: 0x%x\n", gmu_read(gmu, REG_A6XX_GMU_SPTPRAC_PWR_CLK_STATUS)); } @@ -273,7 +294,7 @@ static void a6xx_sptprac_disable(struct a6xx_gmu *gmu) (val & 0x04), 100, 10000); if (ret) - dev_err(gmu->dev, "failed to power off SPTPRAC: 0x%x\n", + DRM_DEV_ERROR(gmu->dev, "failed to power off SPTPRAC: 0x%x\n", gmu_read(gmu, REG_A6XX_GMU_SPTPRAC_PWR_CLK_STATUS)); } @@ -317,7 +338,7 @@ static int a6xx_gmu_notify_slumber(struct a6xx_gmu *gmu) /* Check to see if the GMU really did slumber */ if (gmu_read(gmu, REG_A6XX_GPU_GMU_CX_GMU_RPMH_POWER_STATE) != 0x0f) { - dev_err(gmu->dev, "The GMU did not go into slumber\n"); + DRM_DEV_ERROR(gmu->dev, "The GMU did not go into slumber\n"); ret = -ETIMEDOUT; } } @@ -339,23 +360,27 @@ static int a6xx_rpmh_start(struct a6xx_gmu *gmu) ret = gmu_poll_timeout(gmu, REG_A6XX_GMU_RSCC_CONTROL_ACK, val, val & (1 << 1), 100, 10000); if (ret) { - dev_err(gmu->dev, "Unable to power on the GPU RSC\n"); + DRM_DEV_ERROR(gmu->dev, "Unable to power on the GPU RSC\n"); return ret; } ret = gmu_poll_timeout(gmu, REG_A6XX_RSCC_SEQ_BUSY_DRV0, val, !val, 100, 10000); - if (!ret) { - gmu_write(gmu, REG_A6XX_GMU_RSCC_CONTROL_REQ, 0); - - /* Re-enable the power counter */ - gmu_write(gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_ENABLE, 1); - return 0; + if (ret) { + DRM_DEV_ERROR(gmu->dev, "GPU RSC sequence stuck while waking up the GPU\n"); + return ret; } - dev_err(gmu->dev, "GPU RSC sequence stuck while waking up the GPU\n"); - return ret; + gmu_write(gmu, REG_A6XX_GMU_RSCC_CONTROL_REQ, 0); + + /* Set up CX GMU counter 0 to count busy ticks */ + gmu_write(gmu, REG_A6XX_GPU_GMU_AO_GPU_CX_BUSY_MASK, 0xff000000); + gmu_rmw(gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_SELECT_0, 0xff, 0x20); + + /* Enable the power counter */ + gmu_write(gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_ENABLE, 1); + return 0; } static void a6xx_rpmh_stop(struct a6xx_gmu *gmu) @@ -368,7 +393,7 @@ static void a6xx_rpmh_stop(struct a6xx_gmu *gmu) ret = gmu_poll_timeout(gmu, REG_A6XX_GPU_RSCC_RSC_STATUS0_DRV0, val, val & (1 << 16), 100, 10000); if (ret) - dev_err(gmu->dev, "Unable to power off the GPU RSC\n"); + DRM_DEV_ERROR(gmu->dev, "Unable to power off the GPU RSC\n"); gmu_write(gmu, REG_A6XX_GMU_RSCC_CONTROL_REQ, 0); } @@ -520,7 +545,7 @@ static int a6xx_gmu_fw_start(struct a6xx_gmu *gmu, unsigned int state) /* Sanity check the size of the firmware that was loaded */ if (adreno_gpu->fw[ADRENO_FW_GMU]->size > 0x8000) { - dev_err(gmu->dev, + DRM_DEV_ERROR(gmu->dev, "GMU firmware is bigger than the available region\n"); return -EINVAL; } @@ -764,7 +789,7 @@ int a6xx_gmu_stop(struct a6xx_gpu *a6xx_gpu) */ if (ret) - dev_err(gmu->dev, + DRM_DEV_ERROR(gmu->dev, "Unable to slumber GMU: status = 0%x/0%x\n", gmu_read(gmu, REG_A6XX_GPU_GMU_AO_GPU_CX_BUSY_STATUS), @@ -843,7 +868,7 @@ static struct a6xx_gmu_bo *a6xx_gmu_memory_alloc(struct a6xx_gmu *gmu, IOMMU_READ | IOMMU_WRITE); if (ret) { - dev_err(gmu->dev, "Unable to map GMU buffer object\n"); + DRM_DEV_ERROR(gmu->dev, "Unable to map GMU buffer object\n"); for (i = i - 1 ; i >= 0; i--) iommu_unmap(gmu->domain, @@ -902,26 +927,6 @@ static int a6xx_gmu_memory_probe(struct a6xx_gmu *gmu) return ret; } -/* Get the list of RPMh voltage levels from cmd-db */ -static int a6xx_gmu_rpmh_arc_cmds(const char *id, void *vals, int size) -{ - u32 len = cmd_db_read_aux_data_len(id); - - if (!len) - return 0; - - if (WARN_ON(len > size)) - return -EINVAL; - - cmd_db_read_aux_data(id, vals, len); - - /* - * The data comes back as an array of unsigned shorts so adjust the - * count accordingly - */ - return len >> 1; -} - /* Return the 'arc-level' for the given frequency */ static u32 a6xx_gmu_get_arc_level(struct device *dev, unsigned long freq) { @@ -949,11 +954,30 @@ static u32 a6xx_gmu_get_arc_level(struct device *dev, unsigned long freq) } static int a6xx_gmu_rpmh_arc_votes_init(struct device *dev, u32 *votes, - unsigned long *freqs, int freqs_count, - u16 *pri, int pri_count, - u16 *sec, int sec_count) + unsigned long *freqs, int freqs_count, const char *id) { int i, j; + const u16 *pri, *sec; + size_t pri_count, sec_count; + + pri = cmd_db_read_aux_data(id, &pri_count); + if (IS_ERR(pri)) + return PTR_ERR(pri); + /* + * The data comes back as an array of unsigned shorts so adjust the + * count accordingly + */ + pri_count >>= 1; + if (!pri_count) + return -EINVAL; + + sec = cmd_db_read_aux_data("mx.lvl", &sec_count); + if (IS_ERR(sec)) + return PTR_ERR(sec); + + sec_count >>= 1; + if (!sec_count) + return -EINVAL; /* Construct a vote for each frequency */ for (i = 0; i < freqs_count; i++) { @@ -969,12 +993,12 @@ static int a6xx_gmu_rpmh_arc_votes_init(struct device *dev, u32 *votes, } if (j == pri_count) { - dev_err(dev, + DRM_DEV_ERROR(dev, "Level %u not found in in the RPMh list\n", level); - dev_err(dev, "Available levels:\n"); + DRM_DEV_ERROR(dev, "Available levels:\n"); for (j = 0; j < pri_count; j++) - dev_err(dev, " %u\n", pri[j]); + DRM_DEV_ERROR(dev, " %u\n", pri[j]); return -EINVAL; } @@ -1012,25 +1036,15 @@ static int a6xx_gmu_rpmh_votes_init(struct a6xx_gmu *gmu) struct a6xx_gpu *a6xx_gpu = container_of(gmu, struct a6xx_gpu, gmu); struct adreno_gpu *adreno_gpu = &a6xx_gpu->base; struct msm_gpu *gpu = &adreno_gpu->base; - - u16 gx[16], cx[16], mx[16]; - u32 gxcount, cxcount, mxcount; int ret; - /* Get the list of available voltage levels for each component */ - gxcount = a6xx_gmu_rpmh_arc_cmds("gfx.lvl", gx, sizeof(gx)); - cxcount = a6xx_gmu_rpmh_arc_cmds("cx.lvl", cx, sizeof(cx)); - mxcount = a6xx_gmu_rpmh_arc_cmds("mx.lvl", mx, sizeof(mx)); - /* Build the GX votes */ ret = a6xx_gmu_rpmh_arc_votes_init(&gpu->pdev->dev, gmu->gx_arc_votes, - gmu->gpu_freqs, gmu->nr_gpu_freqs, - gx, gxcount, mx, mxcount); + gmu->gpu_freqs, gmu->nr_gpu_freqs, "gfx.lvl"); /* Build the CX votes */ ret |= a6xx_gmu_rpmh_arc_votes_init(gmu->dev, gmu->cx_arc_votes, - gmu->gmu_freqs, gmu->nr_gmu_freqs, - cx, cxcount, mx, mxcount); + gmu->gmu_freqs, gmu->nr_gmu_freqs, "cx.lvl"); return ret; } @@ -1081,7 +1095,7 @@ static int a6xx_gmu_pwrlevels_probe(struct a6xx_gmu *gmu) */ ret = dev_pm_opp_of_add_table(gmu->dev); if (ret) { - dev_err(gmu->dev, "Unable to set the OPP table for the GMU\n"); + DRM_DEV_ERROR(gmu->dev, "Unable to set the OPP table for the GMU\n"); return ret; } @@ -1122,13 +1136,13 @@ static void __iomem *a6xx_gmu_get_mmio(struct platform_device *pdev, IORESOURCE_MEM, name); if (!res) { - dev_err(&pdev->dev, "Unable to find the %s registers\n", name); + DRM_DEV_ERROR(&pdev->dev, "Unable to find the %s registers\n", name); return ERR_PTR(-EINVAL); } ret = devm_ioremap(&pdev->dev, res->start, resource_size(res)); if (!ret) { - dev_err(&pdev->dev, "Unable to map the %s registers\n", name); + DRM_DEV_ERROR(&pdev->dev, "Unable to map the %s registers\n", name); return ERR_PTR(-EINVAL); } @@ -1145,7 +1159,7 @@ static int a6xx_gmu_get_irq(struct a6xx_gmu *gmu, struct platform_device *pdev, ret = devm_request_irq(&pdev->dev, irq, handler, IRQF_TRIGGER_HIGH, name, gmu); if (ret) { - dev_err(&pdev->dev, "Unable to get interrupt %s\n", name); + DRM_DEV_ERROR(&pdev->dev, "Unable to get interrupt %s\n", name); return ret; } diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h index 35f765afae45..c721d9165d8e 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.h @@ -164,4 +164,7 @@ void a6xx_hfi_init(struct a6xx_gmu *gmu); int a6xx_hfi_start(struct a6xx_gmu *gmu, int boot_state); void a6xx_hfi_stop(struct a6xx_gmu *gmu); +bool a6xx_gmu_gx_is_on(struct a6xx_gmu *gmu); +bool a6xx_gmu_sptprac_is_on(struct a6xx_gmu *gmu); + #endif diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gmu.xml.h b/drivers/gpu/drm/msm/adreno/a6xx_gmu.xml.h index db56f263ed77..1cc1c135236b 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gmu.xml.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gmu.xml.h @@ -10,13 +10,13 @@ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: - /home/robclark/src/envytools/rnndb/adreno.xml ( 501 bytes, from 2018-07-03 19:37:13) - /home/robclark/src/envytools/rnndb/freedreno_copyright.xml ( 1572 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/a2xx.xml ( 36805 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/adreno_common.xml ( 13634 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/adreno_pm4.xml ( 42585 bytes, from 2018-10-04 19:06:37) +- /home/robclark/src/envytools/rnndb/adreno/a2xx.xml ( 42463 bytes, from 2018-11-19 13:44:03) +- /home/robclark/src/envytools/rnndb/adreno/adreno_common.xml ( 14201 bytes, from 2018-12-02 17:29:54) +- /home/robclark/src/envytools/rnndb/adreno/adreno_pm4.xml ( 43052 bytes, from 2018-12-02 17:29:54) - /home/robclark/src/envytools/rnndb/adreno/a3xx.xml ( 83840 bytes, from 2018-07-03 19:37:13) - /home/robclark/src/envytools/rnndb/adreno/a4xx.xml ( 112086 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/a5xx.xml ( 147240 bytes, from 2018-10-04 19:06:37) -- /home/robclark/src/envytools/rnndb/adreno/a6xx.xml ( 139581 bytes, from 2018-10-04 19:06:42) +- /home/robclark/src/envytools/rnndb/adreno/a5xx.xml ( 147240 bytes, from 2018-12-02 17:29:54) +- /home/robclark/src/envytools/rnndb/adreno/a6xx.xml ( 140790 bytes, from 2018-12-02 17:29:54) - /home/robclark/src/envytools/rnndb/adreno/a6xx_gmu.xml ( 10431 bytes, from 2018-09-14 13:03:07) - /home/robclark/src/envytools/rnndb/adreno/ocmem.xml ( 1773 bytes, from 2018-07-03 19:37:13) diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c index 631257c297fd..fefe773c989e 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c @@ -4,6 +4,7 @@ #include "msm_gem.h" #include "msm_mmu.h" +#include "msm_gpu_trace.h" #include "a6xx_gpu.h" #include "a6xx_gmu.xml.h" @@ -67,13 +68,36 @@ static void a6xx_flush(struct msm_gpu *gpu, struct msm_ringbuffer *ring) gpu_write(gpu, REG_A6XX_CP_RB_WPTR, wptr); } +static void get_stats_counter(struct msm_ringbuffer *ring, u32 counter, + u64 iova) +{ + OUT_PKT7(ring, CP_REG_TO_MEM, 3); + OUT_RING(ring, counter | (1 << 30) | (2 << 18)); + OUT_RING(ring, lower_32_bits(iova)); + OUT_RING(ring, upper_32_bits(iova)); +} + static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit, struct msm_file_private *ctx) { + unsigned int index = submit->seqno % MSM_GPU_SUBMIT_STATS_COUNT; struct msm_drm_private *priv = gpu->dev->dev_private; + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); struct msm_ringbuffer *ring = submit->ring; unsigned int i; + get_stats_counter(ring, REG_A6XX_RBBM_PERFCTR_CP_0_LO, + rbmemptr_stats(ring, index, cpcycles_start)); + + /* + * For PM4 the GMU register offsets are calculated from the base of the + * GPU registers so we need to add 0x1a800 to the register value on A630 + * to get the right value from PM4. + */ + get_stats_counter(ring, REG_A6XX_GMU_ALWAYS_ON_COUNTER_L + 0x1a800, + rbmemptr_stats(ring, index, alwayson_start)); + /* Invalidate CCU depth and color */ OUT_PKT7(ring, CP_EVENT_WRITE, 1); OUT_RING(ring, PC_CCU_INVALIDATE_DEPTH); @@ -98,6 +122,11 @@ static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit, } } + get_stats_counter(ring, REG_A6XX_RBBM_PERFCTR_CP_0_LO, + rbmemptr_stats(ring, index, cpcycles_end)); + get_stats_counter(ring, REG_A6XX_GMU_ALWAYS_ON_COUNTER_L + 0x1a800, + rbmemptr_stats(ring, index, alwayson_end)); + /* Write the fence to the scratch register */ OUT_PKT4(ring, REG_A6XX_CP_SCRATCH_REG(2), 1); OUT_RING(ring, submit->seqno); @@ -112,6 +141,10 @@ static void a6xx_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit, OUT_RING(ring, upper_32_bits(rbmemptr(ring, fence))); OUT_RING(ring, submit->seqno); + trace_msm_gpu_submit_flush(submit, + gmu_read64(&a6xx_gpu->gmu, REG_A6XX_GMU_ALWAYS_ON_COUNTER_L, + REG_A6XX_GMU_ALWAYS_ON_COUNTER_H)); + a6xx_flush(gpu, ring); } @@ -300,6 +333,8 @@ static int a6xx_ucode_init(struct msm_gpu *gpu) return ret; } + + msm_gem_object_set_name(a6xx_gpu->sqe_bo, "sqefw"); } gpu_write64(gpu, REG_A6XX_CP_SQE_INSTR_BASE_LO, @@ -387,14 +422,6 @@ static int a6xx_hw_init(struct msm_gpu *gpu) /* Select CP0 to always count cycles */ gpu_write(gpu, REG_A6XX_CP_PERFCTR_CP_SEL_0, PERF_CP_ALWAYS_COUNT); - /* FIXME: not sure if this should live here or in a6xx_gmu.c */ - gmu_write(&a6xx_gpu->gmu, REG_A6XX_GPU_GMU_AO_GPU_CX_BUSY_MASK, - 0xff000000); - gmu_rmw(&a6xx_gpu->gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_SELECT_0, - 0xff, 0x20); - gmu_write(&a6xx_gpu->gmu, REG_A6XX_GMU_CX_GMU_POWER_COUNTER_ENABLE, - 0x01); - gpu_write(gpu, REG_A6XX_RB_NC_MODE_CNTL, 2 << 1); gpu_write(gpu, REG_A6XX_TPL1_NC_MODE_CNTL, 2 << 1); gpu_write(gpu, REG_A6XX_SP_NC_MODE_CNTL, 2 << 1); @@ -481,7 +508,7 @@ out: static void a6xx_dump(struct msm_gpu *gpu) { - dev_info(&gpu->pdev->dev, "status: %08x\n", + DRM_DEV_INFO(&gpu->pdev->dev, "status: %08x\n", gpu_read(gpu, REG_A6XX_RBBM_STATUS)); adreno_dump(gpu); } @@ -498,7 +525,7 @@ static void a6xx_recover(struct msm_gpu *gpu) adreno_dump_info(gpu); for (i = 0; i < 8; i++) - dev_info(&gpu->pdev->dev, "CP_SCRATCH_REG%d: %u\n", i, + DRM_DEV_INFO(&gpu->pdev->dev, "CP_SCRATCH_REG%d: %u\n", i, gpu_read(gpu, REG_A6XX_CP_SCRATCH_REG(i))); if (hang_debug) @@ -645,33 +672,6 @@ static const u32 a6xx_register_offsets[REG_ADRENO_REGISTER_MAX] = { REG_ADRENO_DEFINE(REG_ADRENO_CP_RB_CNTL, REG_A6XX_CP_RB_CNTL), }; -static const u32 a6xx_registers[] = { - 0x0000, 0x0002, 0x0010, 0x0010, 0x0012, 0x0012, 0x0018, 0x001b, - 0x001e, 0x0032, 0x0038, 0x003c, 0x0042, 0x0042, 0x0044, 0x0044, - 0x0047, 0x0047, 0x0056, 0x0056, 0x00ad, 0x00ae, 0x00b0, 0x00fb, - 0x0100, 0x011d, 0x0200, 0x020d, 0x0210, 0x0213, 0x0218, 0x023d, - 0x0400, 0x04f9, 0x0500, 0x0500, 0x0505, 0x050b, 0x050e, 0x0511, - 0x0533, 0x0533, 0x0540, 0x0555, 0x0800, 0x0808, 0x0810, 0x0813, - 0x0820, 0x0821, 0x0823, 0x0827, 0x0830, 0x0833, 0x0840, 0x0843, - 0x084f, 0x086f, 0x0880, 0x088a, 0x08a0, 0x08ab, 0x08c0, 0x08c4, - 0x08d0, 0x08dd, 0x08f0, 0x08f3, 0x0900, 0x0903, 0x0908, 0x0911, - 0x0928, 0x093e, 0x0942, 0x094d, 0x0980, 0x0984, 0x098d, 0x0996, - 0x0998, 0x099e, 0x09a0, 0x09a6, 0x09a8, 0x09ae, 0x09b0, 0x09b1, - 0x09c2, 0x09c8, 0x0a00, 0x0a03, 0x0c00, 0x0c04, 0x0c06, 0x0c06, - 0x0c10, 0x0cd9, 0x0e00, 0x0e0e, 0x0e10, 0x0e13, 0x0e17, 0x0e19, - 0x0e1c, 0x0e2b, 0x0e30, 0x0e32, 0x0e38, 0x0e39, 0x8600, 0x8601, - 0x8610, 0x861b, 0x8620, 0x8620, 0x8628, 0x862b, 0x8630, 0x8637, - 0x8e01, 0x8e01, 0x8e04, 0x8e05, 0x8e07, 0x8e08, 0x8e0c, 0x8e0c, - 0x8e10, 0x8e1c, 0x8e20, 0x8e25, 0x8e28, 0x8e28, 0x8e2c, 0x8e2f, - 0x8e3b, 0x8e3e, 0x8e40, 0x8e43, 0x8e50, 0x8e5e, 0x8e70, 0x8e77, - 0x9600, 0x9604, 0x9624, 0x9637, 0x9e00, 0x9e01, 0x9e03, 0x9e0e, - 0x9e11, 0x9e16, 0x9e19, 0x9e19, 0x9e1c, 0x9e1c, 0x9e20, 0x9e23, - 0x9e30, 0x9e31, 0x9e34, 0x9e34, 0x9e70, 0x9e72, 0x9e78, 0x9e79, - 0x9e80, 0x9fff, 0xa600, 0xa601, 0xa603, 0xa603, 0xa60a, 0xa60a, - 0xa610, 0xa617, 0xa630, 0xa630, - ~0 -}; - static int a6xx_pm_resume(struct msm_gpu *gpu) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); @@ -724,14 +724,6 @@ static int a6xx_get_timestamp(struct msm_gpu *gpu, uint64_t *value) return 0; } -#if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP) -static void a6xx_show(struct msm_gpu *gpu, struct msm_gpu_state *state, - struct drm_printer *p) -{ - adreno_show(gpu, state, p); -} -#endif - static struct msm_ringbuffer *a6xx_active_ring(struct msm_gpu *gpu) { struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); @@ -746,8 +738,7 @@ static void a6xx_destroy(struct msm_gpu *gpu) struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); if (a6xx_gpu->sqe_bo) { - if (a6xx_gpu->sqe_iova) - msm_gem_put_iova(a6xx_gpu->sqe_bo, gpu->aspace); + msm_gem_unpin_iova(a6xx_gpu->sqe_bo, gpu->aspace); drm_gem_object_put_unlocked(a6xx_gpu->sqe_bo); } @@ -796,6 +787,8 @@ static const struct adreno_gpu_funcs funcs = { .gpu_busy = a6xx_gpu_busy, .gpu_get_freq = a6xx_gmu_get_freq, .gpu_set_freq = a6xx_gmu_set_freq, + .gpu_state_get = a6xx_gpu_state_get, + .gpu_state_put = a6xx_gpu_state_put, }, .get_timestamp = a6xx_get_timestamp, }; @@ -817,7 +810,7 @@ struct msm_gpu *a6xx_gpu_init(struct drm_device *dev) adreno_gpu = &a6xx_gpu->base; gpu = &adreno_gpu->base; - adreno_gpu->registers = a6xx_registers; + adreno_gpu->registers = NULL; adreno_gpu->reg_offsets = a6xx_register_offsets; ret = adreno_gpu_init(dev, pdev, adreno_gpu, &funcs, 1); diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h index 4127dcebc202..528a4cfe07cd 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.h +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.h @@ -56,6 +56,14 @@ void a6xx_gmu_clear_oob(struct a6xx_gmu *gmu, enum a6xx_gmu_oob_state state); int a6xx_gmu_probe(struct a6xx_gpu *a6xx_gpu, struct device_node *node); void a6xx_gmu_remove(struct a6xx_gpu *a6xx_gpu); + void a6xx_gmu_set_freq(struct msm_gpu *gpu, unsigned long freq); unsigned long a6xx_gmu_get_freq(struct msm_gpu *gpu); + +void a6xx_show(struct msm_gpu *gpu, struct msm_gpu_state *state, + struct drm_printer *p); + +struct msm_gpu_state *a6xx_gpu_state_get(struct msm_gpu *gpu); +int a6xx_gpu_state_put(struct msm_gpu_state *state); + #endif /* __A6XX_GPU_H__ */ diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c new file mode 100644 index 000000000000..e686331fa089 --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.c @@ -0,0 +1,1165 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2018 The Linux Foundation. All rights reserved. */ + +#include <linux/ascii85.h> +#include "msm_gem.h" +#include "a6xx_gpu.h" +#include "a6xx_gmu.h" +#include "a6xx_gpu_state.h" +#include "a6xx_gmu.xml.h" + +struct a6xx_gpu_state_obj { + const void *handle; + u32 *data; +}; + +struct a6xx_gpu_state { + struct msm_gpu_state base; + + struct a6xx_gpu_state_obj *gmu_registers; + int nr_gmu_registers; + + struct a6xx_gpu_state_obj *registers; + int nr_registers; + + struct a6xx_gpu_state_obj *shaders; + int nr_shaders; + + struct a6xx_gpu_state_obj *clusters; + int nr_clusters; + + struct a6xx_gpu_state_obj *dbgahb_clusters; + int nr_dbgahb_clusters; + + struct a6xx_gpu_state_obj *indexed_regs; + int nr_indexed_regs; + + struct a6xx_gpu_state_obj *debugbus; + int nr_debugbus; + + struct a6xx_gpu_state_obj *vbif_debugbus; + + struct a6xx_gpu_state_obj *cx_debugbus; + int nr_cx_debugbus; + + struct list_head objs; +}; + +static inline int CRASHDUMP_WRITE(u64 *in, u32 reg, u32 val) +{ + in[0] = val; + in[1] = (((u64) reg) << 44 | (1 << 21) | 1); + + return 2; +} + +static inline int CRASHDUMP_READ(u64 *in, u32 reg, u32 dwords, u64 target) +{ + in[0] = target; + in[1] = (((u64) reg) << 44 | dwords); + + return 2; +} + +static inline int CRASHDUMP_FINI(u64 *in) +{ + in[0] = 0; + in[1] = 0; + + return 2; +} + +struct a6xx_crashdumper { + void *ptr; + struct drm_gem_object *bo; + u64 iova; +}; + +struct a6xx_state_memobj { + struct list_head node; + unsigned long long data[]; +}; + +void *state_kcalloc(struct a6xx_gpu_state *a6xx_state, int nr, size_t objsize) +{ + struct a6xx_state_memobj *obj = + kzalloc((nr * objsize) + sizeof(*obj), GFP_KERNEL); + + if (!obj) + return NULL; + + list_add_tail(&obj->node, &a6xx_state->objs); + return &obj->data; +} + +void *state_kmemdup(struct a6xx_gpu_state *a6xx_state, void *src, + size_t size) +{ + void *dst = state_kcalloc(a6xx_state, 1, size); + + if (dst) + memcpy(dst, src, size); + return dst; +} + +/* + * Allocate 1MB for the crashdumper scratch region - 8k for the script and + * the rest for the data + */ +#define A6XX_CD_DATA_OFFSET 8192 +#define A6XX_CD_DATA_SIZE (SZ_1M - 8192) + +static int a6xx_crashdumper_init(struct msm_gpu *gpu, + struct a6xx_crashdumper *dumper) +{ + dumper->ptr = msm_gem_kernel_new_locked(gpu->dev, + SZ_1M, MSM_BO_UNCACHED, gpu->aspace, + &dumper->bo, &dumper->iova); + + if (!IS_ERR(dumper->ptr)) + msm_gem_object_set_name(dumper->bo, "crashdump"); + + return PTR_ERR_OR_ZERO(dumper->ptr); +} + +static int a6xx_crashdumper_run(struct msm_gpu *gpu, + struct a6xx_crashdumper *dumper) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + u32 val; + int ret; + + if (IS_ERR_OR_NULL(dumper->ptr)) + return -EINVAL; + + if (!a6xx_gmu_sptprac_is_on(&a6xx_gpu->gmu)) + return -EINVAL; + + /* Make sure all pending memory writes are posted */ + wmb(); + + gpu_write64(gpu, REG_A6XX_CP_CRASH_SCRIPT_BASE_LO, + REG_A6XX_CP_CRASH_SCRIPT_BASE_HI, dumper->iova); + + gpu_write(gpu, REG_A6XX_CP_CRASH_DUMP_CNTL, 1); + + ret = gpu_poll_timeout(gpu, REG_A6XX_CP_CRASH_DUMP_STATUS, val, + val & 0x02, 100, 10000); + + gpu_write(gpu, REG_A6XX_CP_CRASH_DUMP_CNTL, 0); + + return ret; +} + +/* read a value from the GX debug bus */ +static int debugbus_read(struct msm_gpu *gpu, u32 block, u32 offset, + u32 *data) +{ + u32 reg = A6XX_DBGC_CFG_DBGBUS_SEL_D_PING_INDEX(offset) | + A6XX_DBGC_CFG_DBGBUS_SEL_D_PING_BLK_SEL(block); + + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_SEL_A, reg); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_SEL_B, reg); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_SEL_C, reg); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_SEL_D, reg); + + /* Wait 1 us to make sure the data is flowing */ + udelay(1); + + data[0] = gpu_read(gpu, REG_A6XX_DBGC_CFG_DBGBUS_TRACE_BUF2); + data[1] = gpu_read(gpu, REG_A6XX_DBGC_CFG_DBGBUS_TRACE_BUF1); + + return 2; +} + +#define cxdbg_write(ptr, offset, val) \ + msm_writel((val), (ptr) + ((offset) << 2)) + +#define cxdbg_read(ptr, offset) \ + msm_readl((ptr) + ((offset) << 2)) + +/* read a value from the CX debug bus */ +static int cx_debugbus_read(void *__iomem cxdbg, u32 block, u32 offset, + u32 *data) +{ + u32 reg = A6XX_CX_DBGC_CFG_DBGBUS_SEL_A_PING_INDEX(offset) | + A6XX_CX_DBGC_CFG_DBGBUS_SEL_A_PING_BLK_SEL(block); + + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_SEL_A, reg); + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_SEL_B, reg); + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_SEL_C, reg); + cxdbg_write(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_SEL_D, reg); + + /* Wait 1 us to make sure the data is flowing */ + udelay(1); + + data[0] = cxdbg_read(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_TRACE_BUF2); + data[1] = cxdbg_read(cxdbg, REG_A6XX_CX_DBGC_CFG_DBGBUS_TRACE_BUF1); + + return 2; +} + +/* Read a chunk of data from the VBIF debug bus */ +static int vbif_debugbus_read(struct msm_gpu *gpu, u32 ctrl0, u32 ctrl1, + u32 reg, int count, u32 *data) +{ + int i; + + gpu_write(gpu, ctrl0, reg); + + for (i = 0; i < count; i++) { + gpu_write(gpu, ctrl1, i); + data[i] = gpu_read(gpu, REG_A6XX_VBIF_TEST_BUS_OUT); + } + + return count; +} + +#define AXI_ARB_BLOCKS 2 +#define XIN_AXI_BLOCKS 5 +#define XIN_CORE_BLOCKS 4 + +#define VBIF_DEBUGBUS_BLOCK_SIZE \ + ((16 * AXI_ARB_BLOCKS) + \ + (18 * XIN_AXI_BLOCKS) + \ + (12 * XIN_CORE_BLOCKS)) + +static void a6xx_get_vbif_debugbus_block(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + struct a6xx_gpu_state_obj *obj) +{ + u32 clk, *ptr; + int i; + + obj->data = state_kcalloc(a6xx_state, VBIF_DEBUGBUS_BLOCK_SIZE, + sizeof(u32)); + if (!obj->data) + return; + + obj->handle = NULL; + + /* Get the current clock setting */ + clk = gpu_read(gpu, REG_A6XX_VBIF_CLKON); + + /* Force on the bus so we can read it */ + gpu_write(gpu, REG_A6XX_VBIF_CLKON, + clk | A6XX_VBIF_CLKON_FORCE_ON_TESTBUS); + + /* We will read from BUS2 first, so disable BUS1 */ + gpu_write(gpu, REG_A6XX_VBIF_TEST_BUS1_CTRL0, 0); + + /* Enable the VBIF bus for reading */ + gpu_write(gpu, REG_A6XX_VBIF_TEST_BUS_OUT_CTRL, 1); + + ptr = obj->data; + + for (i = 0; i < AXI_ARB_BLOCKS; i++) + ptr += vbif_debugbus_read(gpu, + REG_A6XX_VBIF_TEST_BUS2_CTRL0, + REG_A6XX_VBIF_TEST_BUS2_CTRL1, + 1 << (i + 16), 16, ptr); + + for (i = 0; i < XIN_AXI_BLOCKS; i++) + ptr += vbif_debugbus_read(gpu, + REG_A6XX_VBIF_TEST_BUS2_CTRL0, + REG_A6XX_VBIF_TEST_BUS2_CTRL1, + 1 << i, 18, ptr); + + /* Stop BUS2 so we can turn on BUS1 */ + gpu_write(gpu, REG_A6XX_VBIF_TEST_BUS2_CTRL0, 0); + + for (i = 0; i < XIN_CORE_BLOCKS; i++) + ptr += vbif_debugbus_read(gpu, + REG_A6XX_VBIF_TEST_BUS1_CTRL0, + REG_A6XX_VBIF_TEST_BUS1_CTRL1, + 1 << i, 12, ptr); + + /* Restore the VBIF clock setting */ + gpu_write(gpu, REG_A6XX_VBIF_CLKON, clk); +} + +static void a6xx_get_debugbus_block(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_debugbus_block *block, + struct a6xx_gpu_state_obj *obj) +{ + int i; + u32 *ptr; + + obj->data = state_kcalloc(a6xx_state, block->count, sizeof(u64)); + if (!obj->data) + return; + + obj->handle = block; + + for (ptr = obj->data, i = 0; i < block->count; i++) + ptr += debugbus_read(gpu, block->id, i, ptr); +} + +static void a6xx_get_cx_debugbus_block(void __iomem *cxdbg, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_debugbus_block *block, + struct a6xx_gpu_state_obj *obj) +{ + int i; + u32 *ptr; + + obj->data = state_kcalloc(a6xx_state, block->count, sizeof(u64)); + if (!obj->data) + return; + + obj->handle = block; + + for (ptr = obj->data, i = 0; i < block->count; i++) + ptr += cx_debugbus_read(cxdbg, block->id, i, ptr); +} + +static void a6xx_get_debugbus(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state) +{ + struct resource *res; + void __iomem *cxdbg = NULL; + + /* Set up the GX debug bus */ + + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_CNTLT, + A6XX_DBGC_CFG_DBGBUS_CNTLT_SEGT(0xf)); + + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_CNTLM, + A6XX_DBGC_CFG_DBGBUS_CNTLM_ENABLE(0xf)); + + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_IVTL_0, 0); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_IVTL_1, 0); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_IVTL_2, 0); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_IVTL_3, 0); + + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_BYTEL_0, 0x76543210); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_BYTEL_1, 0xFEDCBA98); + + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_MASKL_0, 0); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_MASKL_1, 0); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_MASKL_2, 0); + gpu_write(gpu, REG_A6XX_DBGC_CFG_DBGBUS_MASKL_3, 0); + + /* Set up the CX debug bus - it lives elsewhere in the system so do a + * temporary ioremap for the registers + */ + res = platform_get_resource_byname(gpu->pdev, IORESOURCE_MEM, + "cx_dbgc"); + + if (res) + cxdbg = ioremap(res->start, resource_size(res)); + + if (cxdbg) { + cxdbg_write(cxdbg, REG_A6XX_DBGC_CFG_DBGBUS_CNTLT, + A6XX_DBGC_CFG_DBGBUS_CNTLT_SEGT(0xf)); + + cxdbg_write(cxdbg, REG_A6XX_DBGC_CFG_DBGBUS_CNTLM, + A6XX_DBGC_CFG_DBGBUS_CNTLM_ENABLE(0xf)); + + cxdbg_write(cxdbg, REG_A6XX_DBGC_CFG_DBGBUS_IVTL_0, 0); + cxdbg_write(cxdbg, REG_A6XX_DBGC_CFG_DBGBUS_IVTL_1, 0); + cxdbg_write(cxdbg, REG_A6XX_DBGC_CFG_DBGBUS_IVTL_2, 0); + cxdbg_write(cxdbg, REG_A6XX_DBGC_CFG_DBGBUS_IVTL_3, 0); + + cxdbg_write(cxdbg, REG_A6XX_DBGC_CFG_DBGBUS_BYTEL_0, + 0x76543210); + cxdbg_write(cxdbg, REG_A6XX_DBGC_CFG_DBGBUS_BYTEL_1, + 0xFEDCBA98); + + cxdbg_write(cxdbg, REG_A6XX_DBGC_CFG_DBGBUS_MASKL_0, 0); + cxdbg_write(cxdbg, REG_A6XX_DBGC_CFG_DBGBUS_MASKL_1, 0); + cxdbg_write(cxdbg, REG_A6XX_DBGC_CFG_DBGBUS_MASKL_2, 0); + cxdbg_write(cxdbg, REG_A6XX_DBGC_CFG_DBGBUS_MASKL_3, 0); + } + + a6xx_state->debugbus = state_kcalloc(a6xx_state, + ARRAY_SIZE(a6xx_debugbus_blocks), + sizeof(*a6xx_state->debugbus)); + + if (a6xx_state->debugbus) { + int i; + + for (i = 0; i < ARRAY_SIZE(a6xx_debugbus_blocks); i++) + a6xx_get_debugbus_block(gpu, + a6xx_state, + &a6xx_debugbus_blocks[i], + &a6xx_state->debugbus[i]); + + a6xx_state->nr_debugbus = ARRAY_SIZE(a6xx_debugbus_blocks); + } + + a6xx_state->vbif_debugbus = + state_kcalloc(a6xx_state, 1, + sizeof(*a6xx_state->vbif_debugbus)); + + if (a6xx_state->vbif_debugbus) + a6xx_get_vbif_debugbus_block(gpu, a6xx_state, + a6xx_state->vbif_debugbus); + + if (cxdbg) { + a6xx_state->cx_debugbus = + state_kcalloc(a6xx_state, + ARRAY_SIZE(a6xx_cx_debugbus_blocks), + sizeof(*a6xx_state->cx_debugbus)); + + if (a6xx_state->cx_debugbus) { + int i; + + for (i = 0; i < ARRAY_SIZE(a6xx_cx_debugbus_blocks); i++) + a6xx_get_cx_debugbus_block(cxdbg, + a6xx_state, + &a6xx_cx_debugbus_blocks[i], + &a6xx_state->cx_debugbus[i]); + + a6xx_state->nr_cx_debugbus = + ARRAY_SIZE(a6xx_cx_debugbus_blocks); + } + + iounmap(cxdbg); + } +} + +#define RANGE(reg, a) ((reg)[(a) + 1] - (reg)[(a)] + 1) + +/* Read a data cluster from behind the AHB aperture */ +static void a6xx_get_dbgahb_cluster(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_dbgahb_cluster *dbgahb, + struct a6xx_gpu_state_obj *obj, + struct a6xx_crashdumper *dumper) +{ + u64 *in = dumper->ptr; + u64 out = dumper->iova + A6XX_CD_DATA_OFFSET; + size_t datasize; + int i, regcount = 0; + + for (i = 0; i < A6XX_NUM_CONTEXTS; i++) { + int j; + + in += CRASHDUMP_WRITE(in, REG_A6XX_HLSQ_DBG_READ_SEL, + (dbgahb->statetype + i * 2) << 8); + + for (j = 0; j < dbgahb->count; j += 2) { + int count = RANGE(dbgahb->registers, j); + u32 offset = REG_A6XX_HLSQ_DBG_AHB_READ_APERTURE + + dbgahb->registers[j] - (dbgahb->base >> 2); + + in += CRASHDUMP_READ(in, offset, count, out); + + out += count * sizeof(u32); + + if (i == 0) + regcount += count; + } + } + + CRASHDUMP_FINI(in); + + datasize = regcount * A6XX_NUM_CONTEXTS * sizeof(u32); + + if (WARN_ON(datasize > A6XX_CD_DATA_SIZE)) + return; + + if (a6xx_crashdumper_run(gpu, dumper)) + return; + + obj->handle = dbgahb; + obj->data = state_kmemdup(a6xx_state, dumper->ptr + A6XX_CD_DATA_OFFSET, + datasize); +} + +static void a6xx_get_dbgahb_clusters(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + struct a6xx_crashdumper *dumper) +{ + int i; + + a6xx_state->dbgahb_clusters = state_kcalloc(a6xx_state, + ARRAY_SIZE(a6xx_dbgahb_clusters), + sizeof(*a6xx_state->dbgahb_clusters)); + + if (!a6xx_state->dbgahb_clusters) + return; + + a6xx_state->nr_dbgahb_clusters = ARRAY_SIZE(a6xx_dbgahb_clusters); + + for (i = 0; i < ARRAY_SIZE(a6xx_dbgahb_clusters); i++) + a6xx_get_dbgahb_cluster(gpu, a6xx_state, + &a6xx_dbgahb_clusters[i], + &a6xx_state->dbgahb_clusters[i], dumper); +} + +/* Read a data cluster from the CP aperture with the crashdumper */ +static void a6xx_get_cluster(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_cluster *cluster, + struct a6xx_gpu_state_obj *obj, + struct a6xx_crashdumper *dumper) +{ + u64 *in = dumper->ptr; + u64 out = dumper->iova + A6XX_CD_DATA_OFFSET; + size_t datasize; + int i, regcount = 0; + + /* Some clusters need a selector register to be programmed too */ + if (cluster->sel_reg) + in += CRASHDUMP_WRITE(in, cluster->sel_reg, cluster->sel_val); + + for (i = 0; i < A6XX_NUM_CONTEXTS; i++) { + int j; + + in += CRASHDUMP_WRITE(in, REG_A6XX_CP_APERTURE_CNTL_CD, + (cluster->id << 8) | (i << 4) | i); + + for (j = 0; j < cluster->count; j += 2) { + int count = RANGE(cluster->registers, j); + + in += CRASHDUMP_READ(in, cluster->registers[j], + count, out); + + out += count * sizeof(u32); + + if (i == 0) + regcount += count; + } + } + + CRASHDUMP_FINI(in); + + datasize = regcount * A6XX_NUM_CONTEXTS * sizeof(u32); + + if (WARN_ON(datasize > A6XX_CD_DATA_SIZE)) + return; + + if (a6xx_crashdumper_run(gpu, dumper)) + return; + + obj->handle = cluster; + obj->data = state_kmemdup(a6xx_state, dumper->ptr + A6XX_CD_DATA_OFFSET, + datasize); +} + +static void a6xx_get_clusters(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + struct a6xx_crashdumper *dumper) +{ + int i; + + a6xx_state->clusters = state_kcalloc(a6xx_state, + ARRAY_SIZE(a6xx_clusters), sizeof(*a6xx_state->clusters)); + + if (!a6xx_state->clusters) + return; + + a6xx_state->nr_clusters = ARRAY_SIZE(a6xx_clusters); + + for (i = 0; i < ARRAY_SIZE(a6xx_clusters); i++) + a6xx_get_cluster(gpu, a6xx_state, &a6xx_clusters[i], + &a6xx_state->clusters[i], dumper); +} + +/* Read a shader / debug block from the HLSQ aperture with the crashdumper */ +static void a6xx_get_shader_block(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_shader_block *block, + struct a6xx_gpu_state_obj *obj, + struct a6xx_crashdumper *dumper) +{ + u64 *in = dumper->ptr; + size_t datasize = block->size * A6XX_NUM_SHADER_BANKS * sizeof(u32); + int i; + + if (WARN_ON(datasize > A6XX_CD_DATA_SIZE)) + return; + + for (i = 0; i < A6XX_NUM_SHADER_BANKS; i++) { + in += CRASHDUMP_WRITE(in, REG_A6XX_HLSQ_DBG_READ_SEL, + (block->type << 8) | i); + + in += CRASHDUMP_READ(in, REG_A6XX_HLSQ_DBG_AHB_READ_APERTURE, + block->size, dumper->iova + A6XX_CD_DATA_OFFSET); + } + + CRASHDUMP_FINI(in); + + if (a6xx_crashdumper_run(gpu, dumper)) + return; + + obj->handle = block; + obj->data = state_kmemdup(a6xx_state, dumper->ptr + A6XX_CD_DATA_OFFSET, + datasize); +} + +static void a6xx_get_shaders(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + struct a6xx_crashdumper *dumper) +{ + int i; + + a6xx_state->shaders = state_kcalloc(a6xx_state, + ARRAY_SIZE(a6xx_shader_blocks), sizeof(*a6xx_state->shaders)); + + if (!a6xx_state->shaders) + return; + + a6xx_state->nr_shaders = ARRAY_SIZE(a6xx_shader_blocks); + + for (i = 0; i < ARRAY_SIZE(a6xx_shader_blocks); i++) + a6xx_get_shader_block(gpu, a6xx_state, &a6xx_shader_blocks[i], + &a6xx_state->shaders[i], dumper); +} + +/* Read registers from behind the HLSQ aperture with the crashdumper */ +static void a6xx_get_crashdumper_hlsq_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_registers *regs, + struct a6xx_gpu_state_obj *obj, + struct a6xx_crashdumper *dumper) + +{ + u64 *in = dumper->ptr; + u64 out = dumper->iova + A6XX_CD_DATA_OFFSET; + int i, regcount = 0; + + in += CRASHDUMP_WRITE(in, REG_A6XX_HLSQ_DBG_READ_SEL, regs->val1); + + for (i = 0; i < regs->count; i += 2) { + u32 count = RANGE(regs->registers, i); + u32 offset = REG_A6XX_HLSQ_DBG_AHB_READ_APERTURE + + regs->registers[i] - (regs->val0 >> 2); + + in += CRASHDUMP_READ(in, offset, count, out); + + out += count * sizeof(u32); + regcount += count; + } + + CRASHDUMP_FINI(in); + + if (WARN_ON((regcount * sizeof(u32)) > A6XX_CD_DATA_SIZE)) + return; + + if (a6xx_crashdumper_run(gpu, dumper)) + return; + + obj->handle = regs; + obj->data = state_kmemdup(a6xx_state, dumper->ptr + A6XX_CD_DATA_OFFSET, + regcount * sizeof(u32)); +} + +/* Read a block of registers using the crashdumper */ +static void a6xx_get_crashdumper_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_registers *regs, + struct a6xx_gpu_state_obj *obj, + struct a6xx_crashdumper *dumper) + +{ + u64 *in = dumper->ptr; + u64 out = dumper->iova + A6XX_CD_DATA_OFFSET; + int i, regcount = 0; + + /* Some blocks might need to program a selector register first */ + if (regs->val0) + in += CRASHDUMP_WRITE(in, regs->val0, regs->val1); + + for (i = 0; i < regs->count; i += 2) { + u32 count = RANGE(regs->registers, i); + + in += CRASHDUMP_READ(in, regs->registers[i], count, out); + + out += count * sizeof(u32); + regcount += count; + } + + CRASHDUMP_FINI(in); + + if (WARN_ON((regcount * sizeof(u32)) > A6XX_CD_DATA_SIZE)) + return; + + if (a6xx_crashdumper_run(gpu, dumper)) + return; + + obj->handle = regs; + obj->data = state_kmemdup(a6xx_state, dumper->ptr + A6XX_CD_DATA_OFFSET, + regcount * sizeof(u32)); +} + +/* Read a block of registers via AHB */ +static void a6xx_get_ahb_gpu_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_registers *regs, + struct a6xx_gpu_state_obj *obj) +{ + int i, regcount = 0, index = 0; + + for (i = 0; i < regs->count; i += 2) + regcount += RANGE(regs->registers, i); + + obj->handle = (const void *) regs; + obj->data = state_kcalloc(a6xx_state, regcount, sizeof(u32)); + if (!obj->data) + return; + + for (i = 0; i < regs->count; i += 2) { + u32 count = RANGE(regs->registers, i); + int j; + + for (j = 0; j < count; j++) + obj->data[index++] = gpu_read(gpu, + regs->registers[i] + j); + } +} + +/* Read a block of GMU registers */ +static void _a6xx_get_gmu_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_registers *regs, + struct a6xx_gpu_state_obj *obj) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gmu *gmu = &a6xx_gpu->gmu; + int i, regcount = 0, index = 0; + + for (i = 0; i < regs->count; i += 2) + regcount += RANGE(regs->registers, i); + + obj->handle = (const void *) regs; + obj->data = state_kcalloc(a6xx_state, regcount, sizeof(u32)); + if (!obj->data) + return; + + for (i = 0; i < regs->count; i += 2) { + u32 count = RANGE(regs->registers, i); + int j; + + for (j = 0; j < count; j++) + obj->data[index++] = gmu_read(gmu, + regs->registers[i] + j); + } +} + +static void a6xx_get_gmu_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state) +{ + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + + a6xx_state->gmu_registers = state_kcalloc(a6xx_state, + 2, sizeof(*a6xx_state->gmu_registers)); + + if (!a6xx_state->gmu_registers) + return; + + a6xx_state->nr_gmu_registers = 2; + + /* Get the CX GMU registers from AHB */ + _a6xx_get_gmu_registers(gpu, a6xx_state, &a6xx_gmu_reglist[0], + &a6xx_state->gmu_registers[0]); + + if (!a6xx_gmu_gx_is_on(&a6xx_gpu->gmu)) + return; + + /* Set the fence to ALLOW mode so we can access the registers */ + gpu_write(gpu, REG_A6XX_GMU_AO_AHB_FENCE_CTRL, 0); + + _a6xx_get_gmu_registers(gpu, a6xx_state, &a6xx_gmu_reglist[1], + &a6xx_state->gmu_registers[1]); +} + +static void a6xx_get_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + struct a6xx_crashdumper *dumper) +{ + int i, count = ARRAY_SIZE(a6xx_ahb_reglist) + + ARRAY_SIZE(a6xx_reglist) + + ARRAY_SIZE(a6xx_hlsq_reglist); + int index = 0; + + a6xx_state->registers = state_kcalloc(a6xx_state, + count, sizeof(*a6xx_state->registers)); + + if (!a6xx_state->registers) + return; + + a6xx_state->nr_registers = count; + + for (i = 0; i < ARRAY_SIZE(a6xx_ahb_reglist); i++) + a6xx_get_ahb_gpu_registers(gpu, + a6xx_state, &a6xx_ahb_reglist[i], + &a6xx_state->registers[index++]); + + for (i = 0; i < ARRAY_SIZE(a6xx_reglist); i++) + a6xx_get_crashdumper_registers(gpu, + a6xx_state, &a6xx_reglist[i], + &a6xx_state->registers[index++], + dumper); + + for (i = 0; i < ARRAY_SIZE(a6xx_hlsq_reglist); i++) + a6xx_get_crashdumper_hlsq_registers(gpu, + a6xx_state, &a6xx_hlsq_reglist[i], + &a6xx_state->registers[index++], + dumper); +} + +/* Read a block of data from an indexed register pair */ +static void a6xx_get_indexed_regs(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state, + const struct a6xx_indexed_registers *indexed, + struct a6xx_gpu_state_obj *obj) +{ + int i; + + obj->handle = (const void *) indexed; + obj->data = state_kcalloc(a6xx_state, indexed->count, sizeof(u32)); + if (!obj->data) + return; + + /* All the indexed banks start at address 0 */ + gpu_write(gpu, indexed->addr, 0); + + /* Read the data - each read increments the internal address by 1 */ + for (i = 0; i < indexed->count; i++) + obj->data[i] = gpu_read(gpu, indexed->data); +} + +static void a6xx_get_indexed_registers(struct msm_gpu *gpu, + struct a6xx_gpu_state *a6xx_state) +{ + u32 mempool_size; + int count = ARRAY_SIZE(a6xx_indexed_reglist) + 1; + int i; + + a6xx_state->indexed_regs = state_kcalloc(a6xx_state, count, + sizeof(a6xx_state->indexed_regs)); + if (!a6xx_state->indexed_regs) + return; + + for (i = 0; i < ARRAY_SIZE(a6xx_indexed_reglist); i++) + a6xx_get_indexed_regs(gpu, a6xx_state, &a6xx_indexed_reglist[i], + &a6xx_state->indexed_regs[i]); + + /* Set the CP mempool size to 0 to stabilize it while dumping */ + mempool_size = gpu_read(gpu, REG_A6XX_CP_MEM_POOL_SIZE); + gpu_write(gpu, REG_A6XX_CP_MEM_POOL_SIZE, 0); + + /* Get the contents of the CP mempool */ + a6xx_get_indexed_regs(gpu, a6xx_state, &a6xx_cp_mempool_indexed, + &a6xx_state->indexed_regs[i]); + + /* + * Offset 0x2000 in the mempool is the size - copy the saved size over + * so the data is consistent + */ + a6xx_state->indexed_regs[i].data[0x2000] = mempool_size; + + /* Restore the size in the hardware */ + gpu_write(gpu, REG_A6XX_CP_MEM_POOL_SIZE, mempool_size); + + a6xx_state->nr_indexed_regs = count; +} + +struct msm_gpu_state *a6xx_gpu_state_get(struct msm_gpu *gpu) +{ + struct a6xx_crashdumper dumper = { 0 }; + struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); + struct a6xx_gpu *a6xx_gpu = to_a6xx_gpu(adreno_gpu); + struct a6xx_gpu_state *a6xx_state = kzalloc(sizeof(*a6xx_state), + GFP_KERNEL); + + if (!a6xx_state) + return ERR_PTR(-ENOMEM); + + INIT_LIST_HEAD(&a6xx_state->objs); + + /* Get the generic state from the adreno core */ + adreno_gpu_state_get(gpu, &a6xx_state->base); + + a6xx_get_gmu_registers(gpu, a6xx_state); + + /* If GX isn't on the rest of the data isn't going to be accessible */ + if (!a6xx_gmu_gx_is_on(&a6xx_gpu->gmu)) + return &a6xx_state->base; + + /* Get the banks of indexed registers */ + a6xx_get_indexed_registers(gpu, a6xx_state); + + /* Try to initialize the crashdumper */ + if (!a6xx_crashdumper_init(gpu, &dumper)) { + a6xx_get_registers(gpu, a6xx_state, &dumper); + a6xx_get_shaders(gpu, a6xx_state, &dumper); + a6xx_get_clusters(gpu, a6xx_state, &dumper); + a6xx_get_dbgahb_clusters(gpu, a6xx_state, &dumper); + + msm_gem_kernel_put(dumper.bo, gpu->aspace, true); + } + + a6xx_get_debugbus(gpu, a6xx_state); + + return &a6xx_state->base; +} + +void a6xx_gpu_state_destroy(struct kref *kref) +{ + struct a6xx_state_memobj *obj, *tmp; + struct msm_gpu_state *state = container_of(kref, + struct msm_gpu_state, ref); + struct a6xx_gpu_state *a6xx_state = container_of(state, + struct a6xx_gpu_state, base); + + list_for_each_entry_safe(obj, tmp, &a6xx_state->objs, node) + kfree(obj); + + adreno_gpu_state_destroy(state); + kfree(a6xx_state); +} + +int a6xx_gpu_state_put(struct msm_gpu_state *state) +{ + if (IS_ERR_OR_NULL(state)) + return 1; + + return kref_put(&state->ref, a6xx_gpu_state_destroy); +} + +static void a6xx_show_registers(const u32 *registers, u32 *data, size_t count, + struct drm_printer *p) +{ + int i, index = 0; + + if (!data) + return; + + for (i = 0; i < count; i += 2) { + u32 count = RANGE(registers, i); + u32 offset = registers[i]; + int j; + + for (j = 0; j < count; index++, offset++, j++) { + if (data[index] == 0xdeafbead) + continue; + + drm_printf(p, " - { offset: 0x%06x, value: 0x%08x }\n", + offset << 2, data[index]); + } + } +} + +static void print_ascii85(struct drm_printer *p, size_t len, u32 *data) +{ + char out[ASCII85_BUFSZ]; + long i, l, datalen = 0; + + for (i = 0; i < len >> 2; i++) { + if (data[i]) + datalen = (i + 1) << 2; + } + + if (datalen == 0) + return; + + drm_puts(p, " data: !!ascii85 |\n"); + drm_puts(p, " "); + + + l = ascii85_encode_len(datalen); + + for (i = 0; i < l; i++) + drm_puts(p, ascii85_encode(data[i], out)); + + drm_puts(p, "\n"); +} + +static void print_name(struct drm_printer *p, const char *fmt, const char *name) +{ + drm_puts(p, fmt); + drm_puts(p, name); + drm_puts(p, "\n"); +} + +static void a6xx_show_shader(struct a6xx_gpu_state_obj *obj, + struct drm_printer *p) +{ + const struct a6xx_shader_block *block = obj->handle; + int i; + + if (!obj->handle) + return; + + print_name(p, " - type: ", block->name); + + for (i = 0; i < A6XX_NUM_SHADER_BANKS; i++) { + drm_printf(p, " - bank: %d\n", i); + drm_printf(p, " size: %d\n", block->size); + + if (!obj->data) + continue; + + print_ascii85(p, block->size << 2, + obj->data + (block->size * i)); + } +} + +static void a6xx_show_cluster_data(const u32 *registers, int size, u32 *data, + struct drm_printer *p) +{ + int ctx, index = 0; + + for (ctx = 0; ctx < A6XX_NUM_CONTEXTS; ctx++) { + int j; + + drm_printf(p, " - context: %d\n", ctx); + + for (j = 0; j < size; j += 2) { + u32 count = RANGE(registers, j); + u32 offset = registers[j]; + int k; + + for (k = 0; k < count; index++, offset++, k++) { + if (data[index] == 0xdeafbead) + continue; + + drm_printf(p, " - { offset: 0x%06x, value: 0x%08x }\n", + offset << 2, data[index]); + } + } + } +} + +static void a6xx_show_dbgahb_cluster(struct a6xx_gpu_state_obj *obj, + struct drm_printer *p) +{ + const struct a6xx_dbgahb_cluster *dbgahb = obj->handle; + + if (dbgahb) { + print_name(p, " - cluster-name: ", dbgahb->name); + a6xx_show_cluster_data(dbgahb->registers, dbgahb->count, + obj->data, p); + } +} + +static void a6xx_show_cluster(struct a6xx_gpu_state_obj *obj, + struct drm_printer *p) +{ + const struct a6xx_cluster *cluster = obj->handle; + + if (cluster) { + print_name(p, " - cluster-name: ", cluster->name); + a6xx_show_cluster_data(cluster->registers, cluster->count, + obj->data, p); + } +} + +static void a6xx_show_indexed_regs(struct a6xx_gpu_state_obj *obj, + struct drm_printer *p) +{ + const struct a6xx_indexed_registers *indexed = obj->handle; + + if (!indexed) + return; + + print_name(p, " - regs-name: ", indexed->name); + drm_printf(p, " dwords: %d\n", indexed->count); + + print_ascii85(p, indexed->count << 2, obj->data); +} + +static void a6xx_show_debugbus_block(const struct a6xx_debugbus_block *block, + u32 *data, struct drm_printer *p) +{ + if (block) { + print_name(p, " - debugbus-block: ", block->name); + + /* + * count for regular debugbus data is in quadwords, + * but print the size in dwords for consistency + */ + drm_printf(p, " count: %d\n", block->count << 1); + + print_ascii85(p, block->count << 3, data); + } +} + +static void a6xx_show_debugbus(struct a6xx_gpu_state *a6xx_state, + struct drm_printer *p) +{ + int i; + + for (i = 0; i < a6xx_state->nr_debugbus; i++) { + struct a6xx_gpu_state_obj *obj = &a6xx_state->debugbus[i]; + + a6xx_show_debugbus_block(obj->handle, obj->data, p); + } + + if (a6xx_state->vbif_debugbus) { + struct a6xx_gpu_state_obj *obj = a6xx_state->vbif_debugbus; + + drm_puts(p, " - debugbus-block: A6XX_DBGBUS_VBIF\n"); + drm_printf(p, " count: %d\n", VBIF_DEBUGBUS_BLOCK_SIZE); + + /* vbif debugbus data is in dwords. Confusing, huh? */ + print_ascii85(p, VBIF_DEBUGBUS_BLOCK_SIZE << 2, obj->data); + } + + for (i = 0; i < a6xx_state->nr_cx_debugbus; i++) { + struct a6xx_gpu_state_obj *obj = &a6xx_state->cx_debugbus[i]; + + a6xx_show_debugbus_block(obj->handle, obj->data, p); + } +} + +void a6xx_show(struct msm_gpu *gpu, struct msm_gpu_state *state, + struct drm_printer *p) +{ + struct a6xx_gpu_state *a6xx_state = container_of(state, + struct a6xx_gpu_state, base); + int i; + + if (IS_ERR_OR_NULL(state)) + return; + + adreno_show(gpu, state, p); + + drm_puts(p, "registers:\n"); + for (i = 0; i < a6xx_state->nr_registers; i++) { + struct a6xx_gpu_state_obj *obj = &a6xx_state->registers[i]; + const struct a6xx_registers *regs = obj->handle; + + if (!obj->handle) + continue; + + a6xx_show_registers(regs->registers, obj->data, regs->count, p); + } + + drm_puts(p, "registers-gmu:\n"); + for (i = 0; i < a6xx_state->nr_gmu_registers; i++) { + struct a6xx_gpu_state_obj *obj = &a6xx_state->gmu_registers[i]; + const struct a6xx_registers *regs = obj->handle; + + if (!obj->handle) + continue; + + a6xx_show_registers(regs->registers, obj->data, regs->count, p); + } + + drm_puts(p, "indexed-registers:\n"); + for (i = 0; i < a6xx_state->nr_indexed_regs; i++) + a6xx_show_indexed_regs(&a6xx_state->indexed_regs[i], p); + + drm_puts(p, "shader-blocks:\n"); + for (i = 0; i < a6xx_state->nr_shaders; i++) + a6xx_show_shader(&a6xx_state->shaders[i], p); + + drm_puts(p, "clusters:\n"); + for (i = 0; i < a6xx_state->nr_clusters; i++) + a6xx_show_cluster(&a6xx_state->clusters[i], p); + + for (i = 0; i < a6xx_state->nr_dbgahb_clusters; i++) + a6xx_show_dbgahb_cluster(&a6xx_state->dbgahb_clusters[i], p); + + drm_puts(p, "debugbus:\n"); + a6xx_show_debugbus(a6xx_state, p); +} diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h new file mode 100644 index 000000000000..68cccfa2870a --- /dev/null +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu_state.h @@ -0,0 +1,430 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2018 The Linux Foundation. All rights reserved. */ + +#ifndef _A6XX_CRASH_DUMP_H_ +#define _A6XX_CRASH_DUMP_H_ + +#include "a6xx.xml.h" + +#define A6XX_NUM_CONTEXTS 2 +#define A6XX_NUM_SHADER_BANKS 3 + +static const u32 a6xx_gras_cluster[] = { + 0x8000, 0x8006, 0x8010, 0x8092, 0x8094, 0x809d, 0x80a0, 0x80a6, + 0x80af, 0x80f1, 0x8100, 0x8107, 0x8109, 0x8109, 0x8110, 0x8110, + 0x8400, 0x840b, +}; + +static const u32 a6xx_ps_cluster_rac[] = { + 0x8800, 0x8806, 0x8809, 0x8811, 0x8818, 0x881e, 0x8820, 0x8865, + 0x8870, 0x8879, 0x8880, 0x8889, 0x8890, 0x8891, 0x8898, 0x8898, + 0x88c0, 0x88c1, 0x88d0, 0x88e3, 0x8900, 0x890c, 0x890f, 0x891a, + 0x8c00, 0x8c01, 0x8c08, 0x8c10, 0x8c17, 0x8c1f, 0x8c26, 0x8c33, +}; + +static const u32 a6xx_ps_cluster_rbp[] = { + 0x88f0, 0x88f3, 0x890d, 0x890e, 0x8927, 0x8928, 0x8bf0, 0x8bf1, + 0x8c02, 0x8c07, 0x8c11, 0x8c16, 0x8c20, 0x8c25, +}; + +static const u32 a6xx_ps_cluster[] = { + 0x9200, 0x9216, 0x9218, 0x9236, 0x9300, 0x9306, +}; + +static const u32 a6xx_fe_cluster[] = { + 0x9300, 0x9306, 0x9800, 0x9806, 0x9b00, 0x9b07, 0xa000, 0xa009, + 0xa00e, 0xa0ef, 0xa0f8, 0xa0f8, +}; + +static const u32 a6xx_pc_vs_cluster[] = { + 0x9100, 0x9108, 0x9300, 0x9306, 0x9980, 0x9981, 0x9b00, 0x9b07, +}; + +#define CLUSTER_FE 0 +#define CLUSTER_SP_VS 1 +#define CLUSTER_PC_VS 2 +#define CLUSTER_GRAS 3 +#define CLUSTER_SP_PS 4 +#define CLUSTER_PS 5 + +#define CLUSTER(_id, _reg, _sel_reg, _sel_val) \ + { .id = _id, .name = #_id,\ + .registers = _reg, \ + .count = ARRAY_SIZE(_reg), \ + .sel_reg = _sel_reg, .sel_val = _sel_val } + +static const struct a6xx_cluster { + u32 id; + const char *name; + const u32 *registers; + size_t count; + u32 sel_reg; + u32 sel_val; +} a6xx_clusters[] = { + CLUSTER(CLUSTER_GRAS, a6xx_gras_cluster, 0, 0), + CLUSTER(CLUSTER_PS, a6xx_ps_cluster_rac, REG_A6XX_RB_RB_SUB_BLOCK_SEL_CNTL_CD, 0x0), + CLUSTER(CLUSTER_PS, a6xx_ps_cluster_rbp, REG_A6XX_RB_RB_SUB_BLOCK_SEL_CNTL_CD, 0x9), + CLUSTER(CLUSTER_PS, a6xx_ps_cluster, 0, 0), + CLUSTER(CLUSTER_FE, a6xx_fe_cluster, 0, 0), + CLUSTER(CLUSTER_PC_VS, a6xx_pc_vs_cluster, 0, 0), +}; + +static const u32 a6xx_sp_vs_hlsq_cluster[] = { + 0xb800, 0xb803, 0xb820, 0xb822, +}; + +static const u32 a6xx_sp_vs_sp_cluster[] = { + 0xa800, 0xa824, 0xa830, 0xa83c, 0xa840, 0xa864, 0xa870, 0xa895, + 0xa8a0, 0xa8af, 0xa8c0, 0xa8c3, +}; + +static const u32 a6xx_hlsq_duplicate_cluster[] = { + 0xbb10, 0xbb11, 0xbb20, 0xbb29, +}; + +static const u32 a6xx_hlsq_2d_duplicate_cluster[] = { + 0xbd80, 0xbd80, +}; + +static const u32 a6xx_sp_duplicate_cluster[] = { + 0xab00, 0xab00, 0xab04, 0xab05, 0xab10, 0xab1b, 0xab20, 0xab20, +}; + +static const u32 a6xx_tp_duplicate_cluster[] = { + 0xb300, 0xb307, 0xb309, 0xb309, 0xb380, 0xb382, +}; + +static const u32 a6xx_sp_ps_hlsq_cluster[] = { + 0xb980, 0xb980, 0xb982, 0xb987, 0xb990, 0xb99b, 0xb9a0, 0xb9a2, + 0xb9c0, 0xb9c9, +}; + +static const u32 a6xx_sp_ps_hlsq_2d_cluster[] = { + 0xbd80, 0xbd80, +}; + +static const u32 a6xx_sp_ps_sp_cluster[] = { + 0xa980, 0xa9a8, 0xa9b0, 0xa9bc, 0xa9d0, 0xa9d3, 0xa9e0, 0xa9f3, + 0xaa00, 0xaa00, 0xaa30, 0xaa31, +}; + +static const u32 a6xx_sp_ps_sp_2d_cluster[] = { + 0xacc0, 0xacc0, +}; + +static const u32 a6xx_sp_ps_tp_cluster[] = { + 0xb180, 0xb183, 0xb190, 0xb191, +}; + +static const u32 a6xx_sp_ps_tp_2d_cluster[] = { + 0xb4c0, 0xb4d1, +}; + +#define CLUSTER_DBGAHB(_id, _base, _type, _reg) \ + { .name = #_id, .statetype = _type, .base = _base, \ + .registers = _reg, .count = ARRAY_SIZE(_reg) } + +static const struct a6xx_dbgahb_cluster { + const char *name; + u32 statetype; + u32 base; + const u32 *registers; + size_t count; +} a6xx_dbgahb_clusters[] = { + CLUSTER_DBGAHB(CLUSTER_SP_VS, 0x0002e000, 0x41, a6xx_sp_vs_hlsq_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_VS, 0x0002a000, 0x21, a6xx_sp_vs_sp_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_VS, 0x0002e000, 0x41, a6xx_hlsq_duplicate_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_VS, 0x0002f000, 0x45, a6xx_hlsq_2d_duplicate_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_VS, 0x0002a000, 0x21, a6xx_sp_duplicate_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_VS, 0x0002c000, 0x1, a6xx_tp_duplicate_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002e000, 0x42, a6xx_sp_ps_hlsq_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002f000, 0x46, a6xx_sp_ps_hlsq_2d_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002a000, 0x22, a6xx_sp_ps_sp_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002b000, 0x26, a6xx_sp_ps_sp_2d_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002c000, 0x2, a6xx_sp_ps_tp_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002d000, 0x6, a6xx_sp_ps_tp_2d_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002e000, 0x42, a6xx_hlsq_duplicate_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002a000, 0x22, a6xx_sp_duplicate_cluster), + CLUSTER_DBGAHB(CLUSTER_SP_PS, 0x0002c000, 0x2, a6xx_tp_duplicate_cluster), +}; + +static const u32 a6xx_hlsq_registers[] = { + 0xbe00, 0xbe01, 0xbe04, 0xbe05, 0xbe08, 0xbe09, 0xbe10, 0xbe15, + 0xbe20, 0xbe23, +}; + +static const u32 a6xx_sp_registers[] = { + 0xae00, 0xae04, 0xae0c, 0xae0c, 0xae0f, 0xae2b, 0xae30, 0xae32, + 0xae35, 0xae35, 0xae3a, 0xae3f, 0xae50, 0xae52, +}; + +static const u32 a6xx_tp_registers[] = { + 0xb600, 0xb601, 0xb604, 0xb605, 0xb610, 0xb61b, 0xb620, 0xb623, +}; + +struct a6xx_registers { + const u32 *registers; + size_t count; + u32 val0; + u32 val1; +}; + +#define HLSQ_DBG_REGS(_base, _type, _array) \ + { .val0 = _base, .val1 = _type, .registers = _array, \ + .count = ARRAY_SIZE(_array), } + +static const struct a6xx_registers a6xx_hlsq_reglist[] = { + HLSQ_DBG_REGS(0x0002F800, 0x40, a6xx_hlsq_registers), + HLSQ_DBG_REGS(0x0002B800, 0x20, a6xx_sp_registers), + HLSQ_DBG_REGS(0x0002D800, 0x0, a6xx_tp_registers), +}; + +#define SHADER(_type, _size) \ + { .type = _type, .name = #_type, .size = _size } + +static const struct a6xx_shader_block { + const char *name; + u32 type; + u32 size; +} a6xx_shader_blocks[] = { + SHADER(A6XX_TP0_TMO_DATA, 0x200), + SHADER(A6XX_TP0_SMO_DATA, 0x80), + SHADER(A6XX_TP0_MIPMAP_BASE_DATA, 0x3c0), + SHADER(A6XX_TP1_TMO_DATA, 0x200), + SHADER(A6XX_TP1_SMO_DATA, 0x80), + SHADER(A6XX_TP1_MIPMAP_BASE_DATA, 0x3c0), + SHADER(A6XX_SP_INST_DATA, 0x800), + SHADER(A6XX_SP_LB_0_DATA, 0x800), + SHADER(A6XX_SP_LB_1_DATA, 0x800), + SHADER(A6XX_SP_LB_2_DATA, 0x800), + SHADER(A6XX_SP_LB_3_DATA, 0x800), + SHADER(A6XX_SP_LB_4_DATA, 0x800), + SHADER(A6XX_SP_LB_5_DATA, 0x200), + SHADER(A6XX_SP_CB_BINDLESS_DATA, 0x2000), + SHADER(A6XX_SP_CB_LEGACY_DATA, 0x280), + SHADER(A6XX_SP_UAV_DATA, 0x80), + SHADER(A6XX_SP_INST_TAG, 0x80), + SHADER(A6XX_SP_CB_BINDLESS_TAG, 0x80), + SHADER(A6XX_SP_TMO_UMO_TAG, 0x80), + SHADER(A6XX_SP_SMO_TAG, 0x80), + SHADER(A6XX_SP_STATE_DATA, 0x3f), + SHADER(A6XX_HLSQ_CHUNK_CVS_RAM, 0x1c0), + SHADER(A6XX_HLSQ_CHUNK_CPS_RAM, 0x280), + SHADER(A6XX_HLSQ_CHUNK_CVS_RAM_TAG, 0x40), + SHADER(A6XX_HLSQ_CHUNK_CPS_RAM_TAG, 0x40), + SHADER(A6XX_HLSQ_ICB_CVS_CB_BASE_TAG, 0x4), + SHADER(A6XX_HLSQ_ICB_CPS_CB_BASE_TAG, 0x4), + SHADER(A6XX_HLSQ_CVS_MISC_RAM, 0x1c0), + SHADER(A6XX_HLSQ_CPS_MISC_RAM, 0x580), + SHADER(A6XX_HLSQ_INST_RAM, 0x800), + SHADER(A6XX_HLSQ_GFX_CVS_CONST_RAM, 0x800), + SHADER(A6XX_HLSQ_GFX_CPS_CONST_RAM, 0x800), + SHADER(A6XX_HLSQ_CVS_MISC_RAM_TAG, 0x8), + SHADER(A6XX_HLSQ_CPS_MISC_RAM_TAG, 0x4), + SHADER(A6XX_HLSQ_INST_RAM_TAG, 0x80), + SHADER(A6XX_HLSQ_GFX_CVS_CONST_RAM_TAG, 0xc), + SHADER(A6XX_HLSQ_GFX_CPS_CONST_RAM_TAG, 0x10), + SHADER(A6XX_HLSQ_PWR_REST_RAM, 0x28), + SHADER(A6XX_HLSQ_PWR_REST_TAG, 0x14), + SHADER(A6XX_HLSQ_DATAPATH_META, 0x40), + SHADER(A6XX_HLSQ_FRONTEND_META, 0x40), + SHADER(A6XX_HLSQ_INDIRECT_META, 0x40), +}; + +static const u32 a6xx_rb_rac_registers[] = { + 0x8e04, 0x8e05, 0x8e07, 0x8e08, 0x8e10, 0x8e1c, 0x8e20, 0x8e25, + 0x8e28, 0x8e28, 0x8e2c, 0x8e2f, 0x8e50, 0x8e52, +}; + +static const u32 a6xx_rb_rbp_registers[] = { + 0x8e01, 0x8e01, 0x8e0c, 0x8e0c, 0x8e3b, 0x8e3e, 0x8e40, 0x8e43, + 0x8e53, 0x8e5f, 0x8e70, 0x8e77, +}; + +static const u32 a6xx_registers[] = { + /* RBBM */ + 0x0000, 0x0002, 0x0010, 0x0010, 0x0012, 0x0012, 0x0018, 0x001b, + 0x001e, 0x0032, 0x0038, 0x003c, 0x0042, 0x0042, 0x0044, 0x0044, + 0x0047, 0x0047, 0x0056, 0x0056, 0x00ad, 0x00ae, 0x00b0, 0x00fb, + 0x0100, 0x011d, 0x0200, 0x020d, 0x0218, 0x023d, 0x0400, 0x04f9, + 0x0500, 0x0500, 0x0505, 0x050b, 0x050e, 0x0511, 0x0533, 0x0533, + 0x0540, 0x0555, + /* CP */ + 0x0800, 0x0808, 0x0810, 0x0813, 0x0820, 0x0821, 0x0823, 0x0824, + 0x0826, 0x0827, 0x0830, 0x0833, 0x0840, 0x0843, 0x084f, 0x086f, + 0x0880, 0x088a, 0x08a0, 0x08ab, 0x08c0, 0x08c4, 0x08d0, 0x08dd, + 0x08f0, 0x08f3, 0x0900, 0x0903, 0x0908, 0x0911, 0x0928, 0x093e, + 0x0942, 0x094d, 0x0980, 0x0984, 0x098d, 0x0996, 0x0998, 0x099e, + 0x09a0, 0x09a6, 0x09a8, 0x09ae, 0x09b0, 0x09b1, 0x09c2, 0x09c8, + 0x0a00, 0x0a03, + /* VSC */ + 0x0c00, 0x0c04, 0x0c06, 0x0c06, 0x0c10, 0x0cd9, 0x0e00, 0x0e0e, + /* UCHE */ + 0x0e10, 0x0e13, 0x0e17, 0x0e19, 0x0e1c, 0x0e2b, 0x0e30, 0x0e32, + 0x0e38, 0x0e39, + /* GRAS */ + 0x8600, 0x8601, 0x8610, 0x861b, 0x8620, 0x8620, 0x8628, 0x862b, + 0x8630, 0x8637, + /* VPC */ + 0x9600, 0x9604, 0x9624, 0x9637, + /* PC */ + 0x9e00, 0x9e01, 0x9e03, 0x9e0e, 0x9e11, 0x9e16, 0x9e19, 0x9e19, + 0x9e1c, 0x9e1c, 0x9e20, 0x9e23, 0x9e30, 0x9e31, 0x9e34, 0x9e34, + 0x9e70, 0x9e72, 0x9e78, 0x9e79, 0x9e80, 0x9fff, + /* VFD */ + 0xa600, 0xa601, 0xa603, 0xa603, 0xa60a, 0xa60a, 0xa610, 0xa617, + 0xa630, 0xa630, +}; + +#define REGS(_array, _sel_reg, _sel_val) \ + { .registers = _array, .count = ARRAY_SIZE(_array), \ + .val0 = _sel_reg, .val1 = _sel_val } + +static const struct a6xx_registers a6xx_reglist[] = { + REGS(a6xx_registers, 0, 0), + REGS(a6xx_rb_rac_registers, REG_A6XX_RB_RB_SUB_BLOCK_SEL_CNTL_CD, 0), + REGS(a6xx_rb_rbp_registers, REG_A6XX_RB_RB_SUB_BLOCK_SEL_CNTL_CD, 9), +}; + +static const u32 a6xx_ahb_registers[] = { + /* RBBM_STATUS - RBBM_STATUS3 */ + 0x210, 0x213, + /* CP_STATUS_1 */ + 0x825, 0x825, +}; + +static const u32 a6xx_vbif_registers[] = { + 0x3000, 0x3007, 0x300c, 0x3014, 0x3018, 0x302d, 0x3030, 0x3031, + 0x3034, 0x3036, 0x303c, 0x303d, 0x3040, 0x3040, 0x3042, 0x3042, + 0x3049, 0x3049, 0x3058, 0x3058, 0x305a, 0x3061, 0x3064, 0x3068, + 0x306c, 0x306d, 0x3080, 0x3088, 0x308b, 0x308c, 0x3090, 0x3094, + 0x3098, 0x3098, 0x309c, 0x309c, 0x30c0, 0x30c0, 0x30c8, 0x30c8, + 0x30d0, 0x30d0, 0x30d8, 0x30d8, 0x30e0, 0x30e0, 0x3100, 0x3100, + 0x3108, 0x3108, 0x3110, 0x3110, 0x3118, 0x3118, 0x3120, 0x3120, + 0x3124, 0x3125, 0x3129, 0x3129, 0x3131, 0x3131, 0x3154, 0x3154, + 0x3156, 0x3156, 0x3158, 0x3158, 0x315a, 0x315a, 0x315c, 0x315c, + 0x315e, 0x315e, 0x3160, 0x3160, 0x3162, 0x3162, 0x340c, 0x340c, + 0x3410, 0x3410, 0x3800, 0x3801, +}; + +static const struct a6xx_registers a6xx_ahb_reglist[] = { + REGS(a6xx_ahb_registers, 0, 0), + REGS(a6xx_vbif_registers, 0, 0), +}; + +static const u32 a6xx_gmu_gx_registers[] = { + /* GMU GX */ + 0x0000, 0x0000, 0x0010, 0x0013, 0x0016, 0x0016, 0x0018, 0x001b, + 0x001e, 0x001e, 0x0020, 0x0023, 0x0026, 0x0026, 0x0028, 0x002b, + 0x002e, 0x002e, 0x0030, 0x0033, 0x0036, 0x0036, 0x0038, 0x003b, + 0x003e, 0x003e, 0x0040, 0x0043, 0x0046, 0x0046, 0x0080, 0x0084, + 0x0100, 0x012b, 0x0140, 0x0140, +}; + +static const u32 a6xx_gmu_cx_registers[] = { + /* GMU CX */ + 0x4c00, 0x4c07, 0x4c10, 0x4c12, 0x4d00, 0x4d00, 0x4d07, 0x4d0a, + 0x5000, 0x5004, 0x5007, 0x5008, 0x500b, 0x500c, 0x500f, 0x501c, + 0x5024, 0x502a, 0x502d, 0x5030, 0x5040, 0x5053, 0x5087, 0x5089, + 0x50a0, 0x50a2, 0x50a4, 0x50af, 0x50c0, 0x50c3, 0x50d0, 0x50d0, + 0x50e4, 0x50e4, 0x50e8, 0x50ec, 0x5100, 0x5103, 0x5140, 0x5140, + 0x5142, 0x5144, 0x514c, 0x514d, 0x514f, 0x5151, 0x5154, 0x5154, + 0x5157, 0x5158, 0x515d, 0x515d, 0x5162, 0x5162, 0x5164, 0x5165, + 0x5180, 0x5186, 0x5190, 0x519e, 0x51c0, 0x51c0, 0x51c5, 0x51cc, + 0x51e0, 0x51e2, 0x51f0, 0x51f0, 0x5200, 0x5201, + /* GPU RSCC */ + 0x8c8c, 0x8c8c, 0x8d01, 0x8d02, 0x8f40, 0x8f42, 0x8f44, 0x8f47, + 0x8f4c, 0x8f87, 0x8fec, 0x8fef, 0x8ff4, 0x902f, 0x9094, 0x9097, + 0x909c, 0x90d7, 0x913c, 0x913f, 0x9144, 0x917f, + /* GMU AO */ + 0x9300, 0x9316, 0x9400, 0x9400, + /* GPU CC */ + 0x9800, 0x9812, 0x9840, 0x9852, 0x9c00, 0x9c04, 0x9c07, 0x9c0b, + 0x9c15, 0x9c1c, 0x9c1e, 0x9c2d, 0x9c3c, 0x9c3d, 0x9c3f, 0x9c40, + 0x9c42, 0x9c49, 0x9c58, 0x9c5a, 0x9d40, 0x9d5e, 0xa000, 0xa002, + 0xa400, 0xa402, 0xac00, 0xac02, 0xb000, 0xb002, 0xb400, 0xb402, + 0xb800, 0xb802, + /* GPU CC ACD */ + 0xbc00, 0xbc16, 0xbc20, 0xbc27, +}; + +static const struct a6xx_registers a6xx_gmu_reglist[] = { + REGS(a6xx_gmu_cx_registers, 0, 0), + REGS(a6xx_gmu_gx_registers, 0, 0), +}; + +static const struct a6xx_indexed_registers { + const char *name; + u32 addr; + u32 data; + u32 count; +} a6xx_indexed_reglist[] = { + { "CP_SEQ_STAT", REG_A6XX_CP_SQE_STAT_ADDR, + REG_A6XX_CP_SQE_STAT_DATA, 0x33 }, + { "CP_DRAW_STATE", REG_A6XX_CP_DRAW_STATE_ADDR, + REG_A6XX_CP_DRAW_STATE_DATA, 0x100 }, + { "CP_UCODE_DBG_DATA", REG_A6XX_CP_SQE_UCODE_DBG_ADDR, + REG_A6XX_CP_SQE_UCODE_DBG_DATA, 0x6000 }, + { "CP_ROQ", REG_A6XX_CP_ROQ_DBG_ADDR, + REG_A6XX_CP_ROQ_DBG_DATA, 0x400 }, +}; + +static const struct a6xx_indexed_registers a6xx_cp_mempool_indexed = { + "CP_MEMPOOOL", REG_A6XX_CP_MEM_POOL_DBG_ADDR, + REG_A6XX_CP_MEM_POOL_DBG_DATA, 0x2060, +}; + +#define DEBUGBUS(_id, _count) { .id = _id, .name = #_id, .count = _count } + +static const struct a6xx_debugbus_block { + const char *name; + u32 id; + u32 count; +} a6xx_debugbus_blocks[] = { + DEBUGBUS(A6XX_DBGBUS_CP, 0x100), + DEBUGBUS(A6XX_DBGBUS_RBBM, 0x100), + DEBUGBUS(A6XX_DBGBUS_HLSQ, 0x100), + DEBUGBUS(A6XX_DBGBUS_UCHE, 0x100), + DEBUGBUS(A6XX_DBGBUS_DPM, 0x100), + DEBUGBUS(A6XX_DBGBUS_TESS, 0x100), + DEBUGBUS(A6XX_DBGBUS_PC, 0x100), + DEBUGBUS(A6XX_DBGBUS_VFDP, 0x100), + DEBUGBUS(A6XX_DBGBUS_VPC, 0x100), + DEBUGBUS(A6XX_DBGBUS_TSE, 0x100), + DEBUGBUS(A6XX_DBGBUS_RAS, 0x100), + DEBUGBUS(A6XX_DBGBUS_VSC, 0x100), + DEBUGBUS(A6XX_DBGBUS_COM, 0x100), + DEBUGBUS(A6XX_DBGBUS_LRZ, 0x100), + DEBUGBUS(A6XX_DBGBUS_A2D, 0x100), + DEBUGBUS(A6XX_DBGBUS_CCUFCHE, 0x100), + DEBUGBUS(A6XX_DBGBUS_RBP, 0x100), + DEBUGBUS(A6XX_DBGBUS_DCS, 0x100), + DEBUGBUS(A6XX_DBGBUS_DBGC, 0x100), + DEBUGBUS(A6XX_DBGBUS_GMU_GX, 0x100), + DEBUGBUS(A6XX_DBGBUS_TPFCHE, 0x100), + DEBUGBUS(A6XX_DBGBUS_GPC, 0x100), + DEBUGBUS(A6XX_DBGBUS_LARC, 0x100), + DEBUGBUS(A6XX_DBGBUS_HLSQ_SPTP, 0x100), + DEBUGBUS(A6XX_DBGBUS_RB_0, 0x100), + DEBUGBUS(A6XX_DBGBUS_RB_1, 0x100), + DEBUGBUS(A6XX_DBGBUS_UCHE_WRAPPER, 0x100), + DEBUGBUS(A6XX_DBGBUS_CCU_0, 0x100), + DEBUGBUS(A6XX_DBGBUS_CCU_1, 0x100), + DEBUGBUS(A6XX_DBGBUS_VFD_0, 0x100), + DEBUGBUS(A6XX_DBGBUS_VFD_1, 0x100), + DEBUGBUS(A6XX_DBGBUS_VFD_2, 0x100), + DEBUGBUS(A6XX_DBGBUS_VFD_3, 0x100), + DEBUGBUS(A6XX_DBGBUS_SP_0, 0x100), + DEBUGBUS(A6XX_DBGBUS_SP_1, 0x100), + DEBUGBUS(A6XX_DBGBUS_TPL1_0, 0x100), + DEBUGBUS(A6XX_DBGBUS_TPL1_1, 0x100), + DEBUGBUS(A6XX_DBGBUS_TPL1_2, 0x100), + DEBUGBUS(A6XX_DBGBUS_TPL1_3, 0x100), +}; + +static const struct a6xx_debugbus_block a6xx_cx_debugbus_blocks[] = { + DEBUGBUS(A6XX_DBGBUS_GMU_CX, 0x100), + DEBUGBUS(A6XX_DBGBUS_CX, 0x100), +}; + +#endif diff --git a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c index 6ff9baec2658..eda11abc5f01 100644 --- a/drivers/gpu/drm/msm/adreno/a6xx_hfi.c +++ b/drivers/gpu/drm/msm/adreno/a6xx_hfi.c @@ -91,7 +91,7 @@ static int a6xx_hfi_wait_for_ack(struct a6xx_gmu *gmu, u32 id, u32 seqnum, val & A6XX_GMU_GMU2HOST_INTR_INFO_MSGQ, 100, 5000); if (ret) { - dev_err(gmu->dev, + DRM_DEV_ERROR(gmu->dev, "Message %s id %d timed out waiting for response\n", a6xx_hfi_msg_id[id], seqnum); return -ETIMEDOUT; @@ -110,7 +110,7 @@ static int a6xx_hfi_wait_for_ack(struct a6xx_gmu *gmu, u32 id, u32 seqnum, /* If the queue is empty our response never made it */ if (!ret) { - dev_err(gmu->dev, + DRM_DEV_ERROR(gmu->dev, "The HFI response queue is unexpectedly empty\n"); return -ENOENT; @@ -120,20 +120,20 @@ static int a6xx_hfi_wait_for_ack(struct a6xx_gmu *gmu, u32 id, u32 seqnum, struct a6xx_hfi_msg_error *error = (struct a6xx_hfi_msg_error *) &resp; - dev_err(gmu->dev, "GMU firmware error %d\n", + DRM_DEV_ERROR(gmu->dev, "GMU firmware error %d\n", error->code); continue; } if (seqnum != HFI_HEADER_SEQNUM(resp.ret_header)) { - dev_err(gmu->dev, + DRM_DEV_ERROR(gmu->dev, "Unexpected message id %d on the response queue\n", HFI_HEADER_SEQNUM(resp.ret_header)); continue; } if (resp.error) { - dev_err(gmu->dev, + DRM_DEV_ERROR(gmu->dev, "Message %s id %d returned error %d\n", a6xx_hfi_msg_id[id], seqnum, resp.error); return -EINVAL; @@ -163,7 +163,7 @@ static int a6xx_hfi_send_msg(struct a6xx_gmu *gmu, int id, ret = a6xx_hfi_queue_write(gmu, queue, data, dwords); if (ret) { - dev_err(gmu->dev, "Unable to send message %s id %d\n", + DRM_DEV_ERROR(gmu->dev, "Unable to send message %s id %d\n", a6xx_hfi_msg_id[id], seqnum); return ret; } @@ -317,7 +317,7 @@ void a6xx_hfi_stop(struct a6xx_gmu *gmu) continue; if (queue->header->read_index != queue->header->write_index) - dev_err(gmu->dev, "HFI queue %d is not empty\n", i); + DRM_DEV_ERROR(gmu->dev, "HFI queue %d is not empty\n", i); queue->header->read_index = 0; queue->header->write_index = 0; diff --git a/drivers/gpu/drm/msm/adreno/adreno_common.xml.h b/drivers/gpu/drm/msm/adreno/adreno_common.xml.h index 1318959d504d..641d3ba477b6 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_common.xml.h +++ b/drivers/gpu/drm/msm/adreno/adreno_common.xml.h @@ -10,13 +10,13 @@ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: - /home/robclark/src/envytools/rnndb/adreno.xml ( 501 bytes, from 2018-07-03 19:37:13) - /home/robclark/src/envytools/rnndb/freedreno_copyright.xml ( 1572 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/a2xx.xml ( 36805 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/adreno_common.xml ( 13634 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/adreno_pm4.xml ( 42585 bytes, from 2018-10-04 19:06:37) +- /home/robclark/src/envytools/rnndb/adreno/a2xx.xml ( 42463 bytes, from 2018-11-19 13:44:03) +- /home/robclark/src/envytools/rnndb/adreno/adreno_common.xml ( 14201 bytes, from 2018-12-02 17:29:54) +- /home/robclark/src/envytools/rnndb/adreno/adreno_pm4.xml ( 43052 bytes, from 2018-12-02 17:29:54) - /home/robclark/src/envytools/rnndb/adreno/a3xx.xml ( 83840 bytes, from 2018-07-03 19:37:13) - /home/robclark/src/envytools/rnndb/adreno/a4xx.xml ( 112086 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/a5xx.xml ( 147240 bytes, from 2018-10-04 19:06:37) -- /home/robclark/src/envytools/rnndb/adreno/a6xx.xml ( 139581 bytes, from 2018-10-04 19:06:42) +- /home/robclark/src/envytools/rnndb/adreno/a5xx.xml ( 147240 bytes, from 2018-12-02 17:29:54) +- /home/robclark/src/envytools/rnndb/adreno/a6xx.xml ( 140790 bytes, from 2018-12-02 17:29:54) - /home/robclark/src/envytools/rnndb/adreno/a6xx_gmu.xml ( 10431 bytes, from 2018-09-14 13:03:07) - /home/robclark/src/envytools/rnndb/adreno/ocmem.xml ( 1773 bytes, from 2018-07-03 19:37:13) @@ -339,6 +339,15 @@ static inline uint32_t AXXX_SCRATCH_UMSK_SWAP(uint32_t val) #define REG_AXXX_CP_STATE_DEBUG_DATA 0x000001ed #define REG_AXXX_CP_INT_CNTL 0x000001f2 +#define AXXX_CP_INT_CNTL_SW_INT_MASK 0x00080000 +#define AXXX_CP_INT_CNTL_T0_PACKET_IN_IB_MASK 0x00800000 +#define AXXX_CP_INT_CNTL_OPCODE_ERROR_MASK 0x01000000 +#define AXXX_CP_INT_CNTL_PROTECTED_MODE_ERROR_MASK 0x02000000 +#define AXXX_CP_INT_CNTL_RESERVED_BIT_ERROR_MASK 0x04000000 +#define AXXX_CP_INT_CNTL_IB_ERROR_MASK 0x08000000 +#define AXXX_CP_INT_CNTL_IB2_INT_MASK 0x20000000 +#define AXXX_CP_INT_CNTL_IB1_INT_MASK 0x40000000 +#define AXXX_CP_INT_CNTL_RB_INT_MASK 0x80000000 #define REG_AXXX_CP_INT_STATUS 0x000001f3 diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c index 86abdb2b3a9c..714ed6505e47 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_device.c +++ b/drivers/gpu/drm/msm/adreno/adreno_device.c @@ -27,6 +27,39 @@ module_param_named(hang_debug, hang_debug, bool, 0600); static const struct adreno_info gpulist[] = { { + .rev = ADRENO_REV(2, 0, 0, 0), + .revn = 200, + .name = "A200", + .fw = { + [ADRENO_FW_PM4] = "yamato_pm4.fw", + [ADRENO_FW_PFP] = "yamato_pfp.fw", + }, + .gmem = SZ_256K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .init = a2xx_gpu_init, + }, { /* a200 on i.mx51 has only 128kib gmem */ + .rev = ADRENO_REV(2, 0, 0, 1), + .revn = 201, + .name = "A200", + .fw = { + [ADRENO_FW_PM4] = "yamato_pm4.fw", + [ADRENO_FW_PFP] = "yamato_pfp.fw", + }, + .gmem = SZ_128K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .init = a2xx_gpu_init, + }, { + .rev = ADRENO_REV(2, 2, 0, ANY_ID), + .revn = 220, + .name = "A220", + .fw = { + [ADRENO_FW_PM4] = "leia_pm4_470.fw", + [ADRENO_FW_PFP] = "leia_pfp_470.fw", + }, + .gmem = SZ_512K, + .inactive_period = DRM_MSM_INACTIVE_PERIOD, + .init = a2xx_gpu_init, + }, { .rev = ADRENO_REV(3, 0, 5, ANY_ID), .revn = 305, .name = "A305", @@ -196,7 +229,7 @@ struct msm_gpu *adreno_load_gpu(struct drm_device *dev) ret = pm_runtime_get_sync(&pdev->dev); if (ret < 0) { - dev_err(dev->dev, "Couldn't power up the GPU: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "Couldn't power up the GPU: %d\n", ret); return NULL; } @@ -205,7 +238,7 @@ struct msm_gpu *adreno_load_gpu(struct drm_device *dev) mutex_unlock(&dev->struct_mutex); pm_runtime_put_autosuspend(&pdev->dev); if (ret) { - dev_err(dev->dev, "gpu hw init failed: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "gpu hw init failed: %d\n", ret); return NULL; } @@ -238,7 +271,8 @@ static int find_chipid(struct device *dev, struct adreno_rev *rev) if (ret == 0) { unsigned int r, patch; - if (sscanf(compat, "qcom,adreno-%u.%u", &r, &patch) == 2) { + if (sscanf(compat, "qcom,adreno-%u.%u", &r, &patch) == 2 || + sscanf(compat, "amd,imageon-%u.%u", &r, &patch) == 2) { rev->core = r / 100; r %= 100; rev->major = r / 10; @@ -253,7 +287,7 @@ static int find_chipid(struct device *dev, struct adreno_rev *rev) /* and if that fails, fall back to legacy "qcom,chipid" property: */ ret = of_property_read_u32(node, "qcom,chipid", &chipid); if (ret) { - dev_err(dev, "could not parse qcom,chipid: %d\n", ret); + DRM_DEV_ERROR(dev, "could not parse qcom,chipid: %d\n", ret); return ret; } @@ -274,6 +308,7 @@ static int adreno_bind(struct device *dev, struct device *master, void *data) static struct adreno_platform_config config = {}; const struct adreno_info *info; struct drm_device *drm = dev_get_drvdata(master); + struct msm_drm_private *priv = drm->dev_private; struct msm_gpu *gpu; int ret; @@ -296,6 +331,8 @@ static int adreno_bind(struct device *dev, struct device *master, void *data) DBG("Found GPU: %u.%u.%u.%u", config.rev.core, config.rev.major, config.rev.minor, config.rev.patchid); + priv->is_a2xx = config.rev.core == 2; + gpu = info->init(drm); if (IS_ERR(gpu)) { dev_warn(drm->dev, "failed to load adreno gpu\n"); @@ -323,9 +360,37 @@ static const struct component_ops a3xx_ops = { .unbind = adreno_unbind, }; +static void adreno_device_register_headless(void) +{ + /* on imx5, we don't have a top-level mdp/dpu node + * this creates a dummy node for the driver for that case + */ + struct platform_device_info dummy_info = { + .parent = NULL, + .name = "msm", + .id = -1, + .res = NULL, + .num_res = 0, + .data = NULL, + .size_data = 0, + .dma_mask = ~0, + }; + platform_device_register_full(&dummy_info); +} + static int adreno_probe(struct platform_device *pdev) { - return component_add(&pdev->dev, &a3xx_ops); + + int ret; + + ret = component_add(&pdev->dev, &a3xx_ops); + if (ret) + return ret; + + if (of_device_is_compatible(pdev->dev.of_node, "amd,imageon")) + adreno_device_register_headless(); + + return 0; } static int adreno_remove(struct platform_device *pdev) @@ -337,6 +402,8 @@ static int adreno_remove(struct platform_device *pdev) static const struct of_device_id dt_match[] = { { .compatible = "qcom,adreno" }, { .compatible = "qcom,adreno-3xx" }, + /* for compatibility with imx5 gpu: */ + { .compatible = "amd,imageon" }, /* for backwards compat w/ downstream kgsl DT files: */ { .compatible = "qcom,kgsl-3d0" }, {} diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c index 93d70f4a2154..2e4372ef17a3 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c @@ -89,12 +89,12 @@ adreno_request_fw(struct adreno_gpu *adreno_gpu, const char *fwname) ret = request_firmware_direct(&fw, newname, drm->dev); if (!ret) { - dev_info(drm->dev, "loaded %s from new location\n", + DRM_DEV_INFO(drm->dev, "loaded %s from new location\n", newname); adreno_gpu->fwloc = FW_LOCATION_NEW; goto out; } else if (adreno_gpu->fwloc != FW_LOCATION_UNKNOWN) { - dev_err(drm->dev, "failed to load %s: %d\n", + DRM_DEV_ERROR(drm->dev, "failed to load %s: %d\n", newname, ret); fw = ERR_PTR(ret); goto out; @@ -109,12 +109,12 @@ adreno_request_fw(struct adreno_gpu *adreno_gpu, const char *fwname) ret = request_firmware_direct(&fw, fwname, drm->dev); if (!ret) { - dev_info(drm->dev, "loaded %s from legacy location\n", + DRM_DEV_INFO(drm->dev, "loaded %s from legacy location\n", newname); adreno_gpu->fwloc = FW_LOCATION_LEGACY; goto out; } else if (adreno_gpu->fwloc != FW_LOCATION_UNKNOWN) { - dev_err(drm->dev, "failed to load %s: %d\n", + DRM_DEV_ERROR(drm->dev, "failed to load %s: %d\n", fwname, ret); fw = ERR_PTR(ret); goto out; @@ -130,19 +130,19 @@ adreno_request_fw(struct adreno_gpu *adreno_gpu, const char *fwname) ret = request_firmware(&fw, newname, drm->dev); if (!ret) { - dev_info(drm->dev, "loaded %s with helper\n", + DRM_DEV_INFO(drm->dev, "loaded %s with helper\n", newname); adreno_gpu->fwloc = FW_LOCATION_HELPER; goto out; } else if (adreno_gpu->fwloc != FW_LOCATION_UNKNOWN) { - dev_err(drm->dev, "failed to load %s: %d\n", + DRM_DEV_ERROR(drm->dev, "failed to load %s: %d\n", newname, ret); fw = ERR_PTR(ret); goto out; } } - dev_err(drm->dev, "failed to load %s\n", fwname); + DRM_DEV_ERROR(drm->dev, "failed to load %s\n", fwname); fw = ERR_PTR(-ENOENT); out: kfree(newname); @@ -209,14 +209,6 @@ int adreno_hw_init(struct msm_gpu *gpu) if (!ring) continue; - ret = msm_gem_get_iova(ring->bo, gpu->aspace, &ring->iova); - if (ret) { - ring->iova = 0; - dev_err(gpu->dev->dev, - "could not map ringbuffer %d: %d\n", i, ret); - return ret; - } - ring->cur = ring->start; ring->next = ring->start; @@ -277,7 +269,7 @@ void adreno_recover(struct msm_gpu *gpu) ret = msm_gpu_hw_init(gpu); if (ret) { - dev_err(dev->dev, "gpu hw init failed: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "gpu hw init failed: %d\n", ret); /* hmm, oh well? */ } } @@ -319,16 +311,27 @@ void adreno_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit, */ OUT_PKT3(ring, CP_EVENT_WRITE, 1); OUT_RING(ring, HLSQ_FLUSH); - - OUT_PKT3(ring, CP_WAIT_FOR_IDLE, 1); - OUT_RING(ring, 0x00000000); } - /* BIT(31) of CACHE_FLUSH_TS triggers CACHE_FLUSH_TS IRQ from GPU */ - OUT_PKT3(ring, CP_EVENT_WRITE, 3); - OUT_RING(ring, CACHE_FLUSH_TS | BIT(31)); - OUT_RING(ring, rbmemptr(ring, fence)); - OUT_RING(ring, submit->seqno); + /* wait for idle before cache flush/interrupt */ + OUT_PKT3(ring, CP_WAIT_FOR_IDLE, 1); + OUT_RING(ring, 0x00000000); + + if (!adreno_is_a2xx(adreno_gpu)) { + /* BIT(31) of CACHE_FLUSH_TS triggers CACHE_FLUSH_TS IRQ from GPU */ + OUT_PKT3(ring, CP_EVENT_WRITE, 3); + OUT_RING(ring, CACHE_FLUSH_TS | BIT(31)); + OUT_RING(ring, rbmemptr(ring, fence)); + OUT_RING(ring, submit->seqno); + } else { + /* BIT(31) means something else on a2xx */ + OUT_PKT3(ring, CP_EVENT_WRITE, 3); + OUT_RING(ring, CACHE_FLUSH_TS); + OUT_RING(ring, rbmemptr(ring, fence)); + OUT_RING(ring, submit->seqno); + OUT_PKT3(ring, CP_INTERRUPT, 1); + OUT_RING(ring, 0x80000000); + } #if 0 if (adreno_is_a3xx(adreno_gpu)) { @@ -406,7 +409,7 @@ int adreno_gpu_state_get(struct msm_gpu *gpu, struct msm_gpu_state *state) size = j + 1; if (size) { - state->ring[i].data = kmalloc(size << 2, GFP_KERNEL); + state->ring[i].data = kvmalloc(size << 2, GFP_KERNEL); if (state->ring[i].data) { memcpy(state->ring[i].data, gpu->rb[i]->start, size << 2); state->ring[i].data_size = size << 2; @@ -414,6 +417,10 @@ int adreno_gpu_state_get(struct msm_gpu *gpu, struct msm_gpu_state *state) } } + /* Some targets prefer to collect their own registers */ + if (!adreno_gpu->registers) + return 0; + /* Count the number of registers */ for (i = 0; adreno_gpu->registers[i] != ~0; i += 2) count += adreno_gpu->registers[i + 1] - @@ -445,7 +452,7 @@ void adreno_gpu_state_destroy(struct msm_gpu_state *state) int i; for (i = 0; i < ARRAY_SIZE(state->ring); i++) - kfree(state->ring[i].data); + kvfree(state->ring[i].data); for (i = 0; state->bos && i < state->nr_bos; i++) kvfree(state->bos[i].data); @@ -475,34 +482,74 @@ int adreno_gpu_state_put(struct msm_gpu_state *state) #if defined(CONFIG_DEBUG_FS) || defined(CONFIG_DEV_COREDUMP) -static void adreno_show_object(struct drm_printer *p, u32 *ptr, int len) +static char *adreno_gpu_ascii85_encode(u32 *src, size_t len) { + void *buf; + size_t buf_itr = 0, buffer_size; char out[ASCII85_BUFSZ]; - long l, datalen, i; + long l; + int i; - if (!ptr || !len) - return; + if (!src || !len) + return NULL; + + l = ascii85_encode_len(len); /* - * Only dump the non-zero part of the buffer - rarely will any data - * completely fill the entire allocated size of the buffer + * Ascii85 outputs either a 5 byte string or a 1 byte string. So we + * account for the worst case of 5 bytes per dword plus the 1 for '\0' */ - for (datalen = 0, i = 0; i < len >> 2; i++) { - if (ptr[i]) - datalen = (i << 2) + 1; - } + buffer_size = (l * 5) + 1; + + buf = kvmalloc(buffer_size, GFP_KERNEL); + if (!buf) + return NULL; + + for (i = 0; i < l; i++) + buf_itr += snprintf(buf + buf_itr, buffer_size - buf_itr, "%s", + ascii85_encode(src[i], out)); + + return buf; +} - /* Skip printing the object if it is empty */ - if (datalen == 0) +/* len is expected to be in bytes */ +static void adreno_show_object(struct drm_printer *p, void **ptr, int len, + bool *encoded) +{ + if (!*ptr || !len) return; - l = ascii85_encode_len(datalen); + if (!*encoded) { + long datalen, i; + u32 *buf = *ptr; + + /* + * Only dump the non-zero part of the buffer - rarely will + * any data completely fill the entire allocated size of + * the buffer. + */ + for (datalen = 0, i = 0; i < len >> 2; i++) + if (buf[i]) + datalen = ((i + 1) << 2); + + /* + * If we reach here, then the originally captured binary buffer + * will be replaced with the ascii85 encoded string + */ + *ptr = adreno_gpu_ascii85_encode(buf, datalen); + + kvfree(buf); + + *encoded = true; + } + + if (!*ptr) + return; drm_puts(p, " data: !!ascii85 |\n"); drm_puts(p, " "); - for (i = 0; i < l; i++) - drm_puts(p, ascii85_encode(ptr[i], out)); + drm_puts(p, *ptr); drm_puts(p, "\n"); } @@ -534,8 +581,8 @@ void adreno_show(struct msm_gpu *gpu, struct msm_gpu_state *state, drm_printf(p, " wptr: %d\n", state->ring[i].wptr); drm_printf(p, " size: %d\n", MSM_GPU_RINGBUFFER_SZ); - adreno_show_object(p, state->ring[i].data, - state->ring[i].data_size); + adreno_show_object(p, &state->ring[i].data, + state->ring[i].data_size, &state->ring[i].encoded); } if (state->bos) { @@ -546,17 +593,19 @@ void adreno_show(struct msm_gpu *gpu, struct msm_gpu_state *state, state->bos[i].iova); drm_printf(p, " size: %zd\n", state->bos[i].size); - adreno_show_object(p, state->bos[i].data, - state->bos[i].size); + adreno_show_object(p, &state->bos[i].data, + state->bos[i].size, &state->bos[i].encoded); } } - drm_puts(p, "registers:\n"); + if (state->nr_registers) { + drm_puts(p, "registers:\n"); - for (i = 0; i < state->nr_registers; i++) { - drm_printf(p, " - { offset: 0x%04x, value: 0x%08x }\n", - state->registers[i * 2] << 2, - state->registers[(i * 2) + 1]); + for (i = 0; i < state->nr_registers; i++) { + drm_printf(p, " - { offset: 0x%04x, value: 0x%08x }\n", + state->registers[i * 2] << 2, + state->registers[(i * 2) + 1]); + } } } #endif @@ -595,6 +644,9 @@ void adreno_dump(struct msm_gpu *gpu) struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu); int i; + if (!adreno_gpu->registers) + return; + /* dump these out in a form that can be parsed by demsm: */ printk("IO:region %s 00000000 00020000\n", gpu->name); for (i = 0; adreno_gpu->registers[i] != ~0; i += 2) { @@ -635,7 +687,7 @@ static int adreno_get_legacy_pwrlevels(struct device *dev) node = of_get_compatible_child(dev->of_node, "qcom,gpu-pwrlevels"); if (!node) { - dev_err(dev, "Could not find the GPU powerlevels\n"); + DRM_DEV_ERROR(dev, "Could not find the GPU powerlevels\n"); return -ENXIO; } @@ -674,7 +726,7 @@ static int adreno_get_pwrlevels(struct device *dev, else { ret = dev_pm_opp_of_add_table(dev); if (ret) - dev_err(dev, "Unable to set the OPP table\n"); + DRM_DEV_ERROR(dev, "Unable to set the OPP table\n"); } if (!ret) { @@ -717,6 +769,9 @@ int adreno_gpu_init(struct drm_device *drm, struct platform_device *pdev, adreno_gpu_config.va_start = SZ_16M; adreno_gpu_config.va_end = 0xffffffff; + /* maximum range of a2xx mmu */ + if (adreno_is_a2xx(adreno_gpu)) + adreno_gpu_config.va_end = SZ_16M + 0xfff * SZ_64K; adreno_gpu_config.nr_rings = nr_rings; diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h index de6e6ee42fba..5db459bc28a7 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h @@ -21,6 +21,7 @@ #define __ADRENO_GPU_H__ #include <linux/firmware.h> +#include <linux/iopoll.h> #include "msm_gpu.h" @@ -154,6 +155,20 @@ struct adreno_platform_config { __ret; \ }) +static inline bool adreno_is_a2xx(struct adreno_gpu *gpu) +{ + return (gpu->revn < 300); +} + +static inline bool adreno_is_a20x(struct adreno_gpu *gpu) +{ + return (gpu->revn < 210); +} + +static inline bool adreno_is_a225(struct adreno_gpu *gpu) +{ + return gpu->revn == 225; +} static inline bool adreno_is_a3xx(struct adreno_gpu *gpu) { @@ -334,6 +349,7 @@ static inline void adreno_gpu_write(struct adreno_gpu *gpu, gpu_write(&gpu->base, reg - 1, data); } +struct msm_gpu *a2xx_gpu_init(struct drm_device *dev); struct msm_gpu *a3xx_gpu_init(struct drm_device *dev); struct msm_gpu *a4xx_gpu_init(struct drm_device *dev); struct msm_gpu *a5xx_gpu_init(struct drm_device *dev); @@ -375,4 +391,9 @@ static inline uint32_t get_wptr(struct msm_ringbuffer *ring) ((1 << 29) \ ((ilog2((_len)) & 0x1F) << 24) | (((_reg) << 2) & 0xFFFFF)) + +#define gpu_poll_timeout(gpu, addr, val, cond, interval, timeout) \ + readl_poll_timeout((gpu)->mmio + ((addr) << 2), val, cond, \ + interval, timeout) + #endif /* __ADRENO_GPU_H__ */ diff --git a/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h b/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h index 15eb03bed984..79b907ac0b4b 100644 --- a/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h +++ b/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h @@ -10,13 +10,13 @@ git clone https://github.com/freedreno/envytools.git The rules-ng-ng source files this header was generated from are: - /home/robclark/src/envytools/rnndb/adreno.xml ( 501 bytes, from 2018-07-03 19:37:13) - /home/robclark/src/envytools/rnndb/freedreno_copyright.xml ( 1572 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/a2xx.xml ( 36805 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/adreno_common.xml ( 13634 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/adreno_pm4.xml ( 42585 bytes, from 2018-10-04 19:06:37) +- /home/robclark/src/envytools/rnndb/adreno/a2xx.xml ( 42463 bytes, from 2018-11-19 13:44:03) +- /home/robclark/src/envytools/rnndb/adreno/adreno_common.xml ( 14201 bytes, from 2018-12-02 17:29:54) +- /home/robclark/src/envytools/rnndb/adreno/adreno_pm4.xml ( 43052 bytes, from 2018-12-02 17:29:54) - /home/robclark/src/envytools/rnndb/adreno/a3xx.xml ( 83840 bytes, from 2018-07-03 19:37:13) - /home/robclark/src/envytools/rnndb/adreno/a4xx.xml ( 112086 bytes, from 2018-07-03 19:37:13) -- /home/robclark/src/envytools/rnndb/adreno/a5xx.xml ( 147240 bytes, from 2018-10-04 19:06:37) -- /home/robclark/src/envytools/rnndb/adreno/a6xx.xml ( 139581 bytes, from 2018-10-04 19:06:42) +- /home/robclark/src/envytools/rnndb/adreno/a5xx.xml ( 147240 bytes, from 2018-12-02 17:29:54) +- /home/robclark/src/envytools/rnndb/adreno/a6xx.xml ( 140790 bytes, from 2018-12-02 17:29:54) - /home/robclark/src/envytools/rnndb/adreno/a6xx_gmu.xml ( 10431 bytes, from 2018-09-14 13:03:07) - /home/robclark/src/envytools/rnndb/adreno/ocmem.xml ( 1773 bytes, from 2018-07-03 19:37:13) @@ -108,6 +108,13 @@ enum pc_di_src_sel { DI_SRC_SEL_RESERVED = 3, }; +enum pc_di_face_cull_sel { + DI_FACE_CULL_NONE = 0, + DI_FACE_CULL_FETCH = 1, + DI_FACE_BACKFACE_CULL = 2, + DI_FACE_FRONTFACE_CULL = 3, +}; + enum pc_di_index_size { INDEX_SIZE_IGN = 0, INDEX_SIZE_16_BIT = 0, @@ -356,6 +363,7 @@ enum a6xx_render_mode { RM6_GMEM = 4, RM6_BLIT2D = 5, RM6_RESOLVE = 6, + RM6_BLIT2DSCALE = 12, }; enum pseudo_reg { diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.c index 879c13fe74e0..e45c69044935 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.c @@ -319,10 +319,8 @@ static int dpu_debugfs_core_irq_show(struct seq_file *s, void *v) unsigned long irq_flags; int i, irq_count, enable_count, cb_count; - if (!irq_obj || !irq_obj->enable_counts || !irq_obj->irq_cb_tbl) { - DPU_ERROR("invalid parameters\n"); + if (WARN_ON(!irq_obj->enable_counts || !irq_obj->irq_cb_tbl)) return 0; - } for (i = 0; i < irq_obj->total_irqs; i++) { spin_lock_irqsave(&irq_obj->cb_lock, irq_flags); @@ -343,31 +341,11 @@ static int dpu_debugfs_core_irq_show(struct seq_file *s, void *v) DEFINE_DPU_DEBUGFS_SEQ_FOPS(dpu_debugfs_core_irq); -int dpu_debugfs_core_irq_init(struct dpu_kms *dpu_kms, - struct dentry *parent) -{ - dpu_kms->irq_obj.debugfs_file = debugfs_create_file("core_irq", 0600, - parent, &dpu_kms->irq_obj, - &dpu_debugfs_core_irq_fops); - - return 0; -} - -void dpu_debugfs_core_irq_destroy(struct dpu_kms *dpu_kms) -{ - debugfs_remove(dpu_kms->irq_obj.debugfs_file); - dpu_kms->irq_obj.debugfs_file = NULL; -} - -#else -int dpu_debugfs_core_irq_init(struct dpu_kms *dpu_kms, +void dpu_debugfs_core_irq_init(struct dpu_kms *dpu_kms, struct dentry *parent) { - return 0; -} - -void dpu_debugfs_core_irq_destroy(struct dpu_kms *dpu_kms) -{ + debugfs_create_file("core_irq", 0600, parent, &dpu_kms->irq_obj, + &dpu_debugfs_core_irq_fops); } #endif @@ -376,10 +354,7 @@ void dpu_core_irq_preinstall(struct dpu_kms *dpu_kms) struct msm_drm_private *priv; int i; - if (!dpu_kms) { - DPU_ERROR("invalid dpu_kms\n"); - return; - } else if (!dpu_kms->dev) { + if (!dpu_kms->dev) { DPU_ERROR("invalid drm device\n"); return; } else if (!dpu_kms->dev->dev_private) { @@ -410,20 +385,12 @@ void dpu_core_irq_preinstall(struct dpu_kms *dpu_kms) } } -int dpu_core_irq_postinstall(struct dpu_kms *dpu_kms) -{ - return 0; -} - void dpu_core_irq_uninstall(struct dpu_kms *dpu_kms) { struct msm_drm_private *priv; int i; - if (!dpu_kms) { - DPU_ERROR("invalid dpu_kms\n"); - return; - } else if (!dpu_kms->dev) { + if (!dpu_kms->dev) { DPU_ERROR("invalid drm device\n"); return; } else if (!dpu_kms->dev->dev_private) { diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.h index 5e98bba46af5..e9015a2b23fe 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_irq.h @@ -24,13 +24,6 @@ void dpu_core_irq_preinstall(struct dpu_kms *dpu_kms); /** - * dpu_core_irq_postinstall - perform post-installation of core IRQ handler - * @dpu_kms: DPU handle - * @return: 0 if success; error code otherwise - */ -int dpu_core_irq_postinstall(struct dpu_kms *dpu_kms); - -/** * dpu_core_irq_uninstall - uninstall core IRQ handler * @dpu_kms: DPU handle * @return: none @@ -139,15 +132,8 @@ int dpu_core_irq_unregister_callback( * dpu_debugfs_core_irq_init - register core irq debugfs * @dpu_kms: pointer to kms * @parent: debugfs directory root - * @Return: 0 on success */ -int dpu_debugfs_core_irq_init(struct dpu_kms *dpu_kms, +void dpu_debugfs_core_irq_init(struct dpu_kms *dpu_kms, struct dentry *parent); -/** - * dpu_debugfs_core_irq_destroy - deregister core irq debugfs - * @dpu_kms: pointer to kms - */ -void dpu_debugfs_core_irq_destroy(struct dpu_kms *dpu_kms); - #endif /* __DPU_CORE_IRQ_H__ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c index 41c5191f9056..9f20f397f77d 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.c @@ -24,8 +24,6 @@ #include "dpu_crtc.h" #include "dpu_core_perf.h" -#define DPU_PERF_MODE_STRING_SIZE 128 - /** * enum dpu_perf_mode - performance tuning mode * @DPU_PERF_MODE_NORMAL: performance controlled by user mode client @@ -57,31 +55,20 @@ static struct dpu_kms *_dpu_crtc_get_kms(struct drm_crtc *crtc) return to_dpu_kms(priv->kms); } -static bool _dpu_core_perf_crtc_is_power_on(struct drm_crtc *crtc) -{ - return dpu_crtc_is_enabled(crtc); -} - static bool _dpu_core_video_mode_intf_connected(struct drm_crtc *crtc) { struct drm_crtc *tmp_crtc; - bool intf_connected = false; - - if (!crtc) - goto end; drm_for_each_crtc(tmp_crtc, crtc->dev) { if ((dpu_crtc_get_intf_mode(tmp_crtc) == INTF_MODE_VIDEO) && - _dpu_core_perf_crtc_is_power_on(tmp_crtc)) { + tmp_crtc->enabled) { DPU_DEBUG("video interface connected crtc:%d\n", tmp_crtc->base.id); - intf_connected = true; - goto end; + return true; } } -end: - return intf_connected; + return false; } static void _dpu_core_perf_calc_crtc(struct dpu_kms *kms, @@ -101,20 +88,20 @@ static void _dpu_core_perf_calc_crtc(struct dpu_kms *kms, memset(perf, 0, sizeof(struct dpu_core_perf_params)); if (!dpu_cstate->bw_control) { - for (i = 0; i < DPU_POWER_HANDLE_DBUS_ID_MAX; i++) { + for (i = 0; i < DPU_CORE_PERF_DATA_BUS_ID_MAX; i++) { perf->bw_ctl[i] = kms->catalog->perf.max_bw_high * 1000ULL; perf->max_per_pipe_ib[i] = perf->bw_ctl[i]; } perf->core_clk_rate = kms->perf.max_core_clk_rate; } else if (kms->perf.perf_tune.mode == DPU_PERF_MODE_MINIMUM) { - for (i = 0; i < DPU_POWER_HANDLE_DBUS_ID_MAX; i++) { + for (i = 0; i < DPU_CORE_PERF_DATA_BUS_ID_MAX; i++) { perf->bw_ctl[i] = 0; perf->max_per_pipe_ib[i] = 0; } perf->core_clk_rate = 0; } else if (kms->perf.perf_tune.mode == DPU_PERF_MODE_FIXED) { - for (i = 0; i < DPU_POWER_HANDLE_DBUS_ID_MAX; i++) { + for (i = 0; i < DPU_CORE_PERF_DATA_BUS_ID_MAX; i++) { perf->bw_ctl[i] = kms->perf.fix_core_ab_vote; perf->max_per_pipe_ib[i] = kms->perf.fix_core_ib_vote; } @@ -124,12 +111,12 @@ static void _dpu_core_perf_calc_crtc(struct dpu_kms *kms, DPU_DEBUG( "crtc=%d clk_rate=%llu core_ib=%llu core_ab=%llu llcc_ib=%llu llcc_ab=%llu mem_ib=%llu mem_ab=%llu\n", crtc->base.id, perf->core_clk_rate, - perf->max_per_pipe_ib[DPU_POWER_HANDLE_DBUS_ID_MNOC], - perf->bw_ctl[DPU_POWER_HANDLE_DBUS_ID_MNOC], - perf->max_per_pipe_ib[DPU_POWER_HANDLE_DBUS_ID_LLCC], - perf->bw_ctl[DPU_POWER_HANDLE_DBUS_ID_LLCC], - perf->max_per_pipe_ib[DPU_POWER_HANDLE_DBUS_ID_EBI], - perf->bw_ctl[DPU_POWER_HANDLE_DBUS_ID_EBI]); + perf->max_per_pipe_ib[DPU_CORE_PERF_DATA_BUS_ID_MNOC], + perf->bw_ctl[DPU_CORE_PERF_DATA_BUS_ID_MNOC], + perf->max_per_pipe_ib[DPU_CORE_PERF_DATA_BUS_ID_LLCC], + perf->bw_ctl[DPU_CORE_PERF_DATA_BUS_ID_LLCC], + perf->max_per_pipe_ib[DPU_CORE_PERF_DATA_BUS_ID_EBI], + perf->bw_ctl[DPU_CORE_PERF_DATA_BUS_ID_EBI]); } int dpu_core_perf_crtc_check(struct drm_crtc *crtc, @@ -164,13 +151,13 @@ int dpu_core_perf_crtc_check(struct drm_crtc *crtc, /* obtain new values */ _dpu_core_perf_calc_crtc(kms, crtc, state, &dpu_cstate->new_perf); - for (i = DPU_POWER_HANDLE_DBUS_ID_MNOC; - i < DPU_POWER_HANDLE_DBUS_ID_MAX; i++) { + for (i = DPU_CORE_PERF_DATA_BUS_ID_MNOC; + i < DPU_CORE_PERF_DATA_BUS_ID_MAX; i++) { bw_sum_of_intfs = dpu_cstate->new_perf.bw_ctl[i]; curr_client_type = dpu_crtc_get_client_type(crtc); drm_for_each_crtc(tmp_crtc, crtc->dev) { - if (_dpu_core_perf_crtc_is_power_on(tmp_crtc) && + if (tmp_crtc->enabled && (dpu_crtc_get_client_type(tmp_crtc) == curr_client_type) && (tmp_crtc != crtc)) { @@ -229,7 +216,7 @@ static int _dpu_core_perf_crtc_update_bus(struct dpu_kms *kms, int ret = 0; drm_for_each_crtc(tmp_crtc, crtc->dev) { - if (_dpu_core_perf_crtc_is_power_on(tmp_crtc) && + if (tmp_crtc->enabled && curr_client_type == dpu_crtc_get_client_type(tmp_crtc)) { dpu_cstate = to_dpu_crtc_state(tmp_crtc->state); @@ -286,7 +273,7 @@ void dpu_core_perf_crtc_release_bw(struct drm_crtc *crtc) */ if (dpu_crtc_get_intf_mode(crtc) == INTF_MODE_CMD) drm_for_each_crtc(tmp_crtc, crtc->dev) { - if (_dpu_core_perf_crtc_is_power_on(tmp_crtc) && + if (tmp_crtc->enabled && dpu_crtc_get_intf_mode(tmp_crtc) == INTF_MODE_VIDEO) return; @@ -296,7 +283,7 @@ void dpu_core_perf_crtc_release_bw(struct drm_crtc *crtc) if (kms->perf.enable_bw_release) { trace_dpu_cmd_release_bw(crtc->base.id); DPU_DEBUG("Release BW crtc=%d\n", crtc->base.id); - for (i = 0; i < DPU_POWER_HANDLE_DBUS_ID_MAX; i++) { + for (i = 0; i < DPU_CORE_PERF_DATA_BUS_ID_MAX; i++) { dpu_crtc->cur_perf.bw_ctl[i] = 0; _dpu_core_perf_crtc_update_bus(kms, crtc, i); } @@ -321,7 +308,7 @@ static u64 _dpu_core_perf_get_core_clk_rate(struct dpu_kms *kms) struct dpu_crtc_state *dpu_cstate; drm_for_each_crtc(crtc, kms->dev) { - if (_dpu_core_perf_crtc_is_power_on(crtc)) { + if (crtc->enabled) { dpu_cstate = to_dpu_crtc_state(crtc->state); clk_rate = max(dpu_cstate->new_perf.core_clk_rate, clk_rate); @@ -372,8 +359,8 @@ int dpu_core_perf_crtc_update(struct drm_crtc *crtc, old = &dpu_crtc->cur_perf; new = &dpu_cstate->new_perf; - if (_dpu_core_perf_crtc_is_power_on(crtc) && !stop_req) { - for (i = 0; i < DPU_POWER_HANDLE_DBUS_ID_MAX; i++) { + if (crtc->enabled && !stop_req) { + for (i = 0; i < DPU_CORE_PERF_DATA_BUS_ID_MAX; i++) { /* * cases for bus bandwidth update. * 1. new bandwidth vote - "ab or ib vote" is higher @@ -415,13 +402,13 @@ int dpu_core_perf_crtc_update(struct drm_crtc *crtc, update_clk = 1; } trace_dpu_perf_crtc_update(crtc->base.id, - new->bw_ctl[DPU_POWER_HANDLE_DBUS_ID_MNOC], - new->bw_ctl[DPU_POWER_HANDLE_DBUS_ID_LLCC], - new->bw_ctl[DPU_POWER_HANDLE_DBUS_ID_EBI], + new->bw_ctl[DPU_CORE_PERF_DATA_BUS_ID_MNOC], + new->bw_ctl[DPU_CORE_PERF_DATA_BUS_ID_LLCC], + new->bw_ctl[DPU_CORE_PERF_DATA_BUS_ID_EBI], new->core_clk_rate, stop_req, update_bus, update_clk); - for (i = 0; i < DPU_POWER_HANDLE_DBUS_ID_MAX; i++) { + for (i = 0; i < DPU_CORE_PERF_DATA_BUS_ID_MAX; i++) { if (update_bus & BIT(i)) { ret = _dpu_core_perf_crtc_update_bus(kms, crtc, i); if (ret) { @@ -462,24 +449,14 @@ static ssize_t _dpu_core_perf_mode_write(struct file *file, struct dpu_core_perf *perf = file->private_data; struct dpu_perf_cfg *cfg = &perf->catalog->perf; u32 perf_mode = 0; - char buf[10]; - - if (!perf) - return -ENODEV; - - if (count >= sizeof(buf)) - return -EFAULT; - - if (copy_from_user(buf, user_buf, count)) - return -EFAULT; - - buf[count] = 0; /* end of string */ + int ret; - if (kstrtouint(buf, 0, &perf_mode)) - return -EFAULT; + ret = kstrtouint_from_user(user_buf, count, 0, &perf_mode); + if (ret) + return ret; if (perf_mode >= DPU_PERF_MODE_MAX) - return -EFAULT; + return -EINVAL; if (perf_mode == DPU_PERF_MODE_FIXED) { DRM_INFO("fix performance mode\n"); @@ -504,29 +481,16 @@ static ssize_t _dpu_core_perf_mode_read(struct file *file, char __user *buff, size_t count, loff_t *ppos) { struct dpu_core_perf *perf = file->private_data; - int len = 0; - char buf[DPU_PERF_MODE_STRING_SIZE] = {'\0'}; - - if (!perf) - return -ENODEV; + int len; + char buf[128]; - if (*ppos) - return 0; /* the end */ - - len = snprintf(buf, sizeof(buf), + len = scnprintf(buf, sizeof(buf), "mode %d min_mdp_clk %llu min_bus_vote %llu\n", perf->perf_tune.mode, perf->perf_tune.min_core_clk, perf->perf_tune.min_bus_vote); - if (len < 0 || len >= sizeof(buf)) - return 0; - - if ((count < sizeof(buf)) || copy_to_user(buff, buf, len)) - return -EFAULT; - - *ppos += len; /* increase offset */ - return len; + return simple_read_from_buffer(buff, count, ppos, buf, len); } static const struct file_operations dpu_core_perf_mode_fops = { @@ -535,70 +499,43 @@ static const struct file_operations dpu_core_perf_mode_fops = { .write = _dpu_core_perf_mode_write, }; -static void dpu_core_perf_debugfs_destroy(struct dpu_core_perf *perf) -{ - debugfs_remove_recursive(perf->debugfs_root); - perf->debugfs_root = NULL; -} - -int dpu_core_perf_debugfs_init(struct dpu_core_perf *perf, - struct dentry *parent) +int dpu_core_perf_debugfs_init(struct dpu_kms *dpu_kms, struct dentry *parent) { + struct dpu_core_perf *perf = &dpu_kms->perf; struct dpu_mdss_cfg *catalog = perf->catalog; - struct msm_drm_private *priv; - struct dpu_kms *dpu_kms; - - priv = perf->dev->dev_private; - if (!priv || !priv->kms) { - DPU_ERROR("invalid KMS reference\n"); - return -EINVAL; - } + struct dentry *entry; - dpu_kms = to_dpu_kms(priv->kms); - - perf->debugfs_root = debugfs_create_dir("core_perf", parent); - if (!perf->debugfs_root) { - DPU_ERROR("failed to create core perf debugfs\n"); + entry = debugfs_create_dir("core_perf", parent); + if (IS_ERR_OR_NULL(entry)) return -EINVAL; - } - debugfs_create_u64("max_core_clk_rate", 0600, perf->debugfs_root, + debugfs_create_u64("max_core_clk_rate", 0600, entry, &perf->max_core_clk_rate); - debugfs_create_u64("core_clk_rate", 0600, perf->debugfs_root, + debugfs_create_u64("core_clk_rate", 0600, entry, &perf->core_clk_rate); - debugfs_create_u32("enable_bw_release", 0600, perf->debugfs_root, + debugfs_create_u32("enable_bw_release", 0600, entry, (u32 *)&perf->enable_bw_release); - debugfs_create_u32("threshold_low", 0600, perf->debugfs_root, + debugfs_create_u32("threshold_low", 0600, entry, (u32 *)&catalog->perf.max_bw_low); - debugfs_create_u32("threshold_high", 0600, perf->debugfs_root, + debugfs_create_u32("threshold_high", 0600, entry, (u32 *)&catalog->perf.max_bw_high); - debugfs_create_u32("min_core_ib", 0600, perf->debugfs_root, + debugfs_create_u32("min_core_ib", 0600, entry, (u32 *)&catalog->perf.min_core_ib); - debugfs_create_u32("min_llcc_ib", 0600, perf->debugfs_root, + debugfs_create_u32("min_llcc_ib", 0600, entry, (u32 *)&catalog->perf.min_llcc_ib); - debugfs_create_u32("min_dram_ib", 0600, perf->debugfs_root, + debugfs_create_u32("min_dram_ib", 0600, entry, (u32 *)&catalog->perf.min_dram_ib); - debugfs_create_file("perf_mode", 0600, perf->debugfs_root, + debugfs_create_file("perf_mode", 0600, entry, (u32 *)perf, &dpu_core_perf_mode_fops); - debugfs_create_u64("fix_core_clk_rate", 0600, perf->debugfs_root, + debugfs_create_u64("fix_core_clk_rate", 0600, entry, &perf->fix_core_clk_rate); - debugfs_create_u64("fix_core_ib_vote", 0600, perf->debugfs_root, + debugfs_create_u64("fix_core_ib_vote", 0600, entry, &perf->fix_core_ib_vote); - debugfs_create_u64("fix_core_ab_vote", 0600, perf->debugfs_root, + debugfs_create_u64("fix_core_ab_vote", 0600, entry, &perf->fix_core_ab_vote); return 0; } -#else -static void dpu_core_perf_debugfs_destroy(struct dpu_core_perf *perf) -{ -} - -int dpu_core_perf_debugfs_init(struct dpu_core_perf *perf, - struct dentry *parent) -{ - return 0; -} #endif void dpu_core_perf_destroy(struct dpu_core_perf *perf) @@ -608,10 +545,8 @@ void dpu_core_perf_destroy(struct dpu_core_perf *perf) return; } - dpu_core_perf_debugfs_destroy(perf); perf->max_core_clk_rate = 0; perf->core_clk = NULL; - perf->phandle = NULL; perf->catalog = NULL; perf->dev = NULL; } @@ -619,12 +554,10 @@ void dpu_core_perf_destroy(struct dpu_core_perf *perf) int dpu_core_perf_init(struct dpu_core_perf *perf, struct drm_device *dev, struct dpu_mdss_cfg *catalog, - struct dpu_power_handle *phandle, struct dss_clk *core_clk) { perf->dev = dev; perf->catalog = catalog; - perf->phandle = phandle; perf->core_clk = core_clk; perf->max_core_clk_rate = core_clk->max_rate; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h index fbcbe0c7527a..37f518815eb7 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_core_perf.h @@ -19,19 +19,31 @@ #include <drm/drm_crtc.h> #include "dpu_hw_catalog.h" -#include "dpu_power_handle.h" #define DPU_PERF_DEFAULT_MAX_CORE_CLK_RATE 412500000 /** + * enum dpu_core_perf_data_bus_id - data bus identifier + * @DPU_CORE_PERF_DATA_BUS_ID_MNOC: DPU/MNOC data bus + * @DPU_CORE_PERF_DATA_BUS_ID_LLCC: MNOC/LLCC data bus + * @DPU_CORE_PERF_DATA_BUS_ID_EBI: LLCC/EBI data bus + */ +enum dpu_core_perf_data_bus_id { + DPU_CORE_PERF_DATA_BUS_ID_MNOC, + DPU_CORE_PERF_DATA_BUS_ID_LLCC, + DPU_CORE_PERF_DATA_BUS_ID_EBI, + DPU_CORE_PERF_DATA_BUS_ID_MAX, +}; + +/** * struct dpu_core_perf_params - definition of performance parameters * @max_per_pipe_ib: maximum instantaneous bandwidth request * @bw_ctl: arbitrated bandwidth request * @core_clk_rate: core clock rate request */ struct dpu_core_perf_params { - u64 max_per_pipe_ib[DPU_POWER_HANDLE_DBUS_ID_MAX]; - u64 bw_ctl[DPU_POWER_HANDLE_DBUS_ID_MAX]; + u64 max_per_pipe_ib[DPU_CORE_PERF_DATA_BUS_ID_MAX]; + u64 bw_ctl[DPU_CORE_PERF_DATA_BUS_ID_MAX]; u64 core_clk_rate; }; @@ -52,7 +64,6 @@ struct dpu_core_perf_tune { * @dev: Pointer to drm device * @debugfs_root: top level debug folder * @catalog: Pointer to catalog configuration - * @phandle: Pointer to power handler * @core_clk: Pointer to core clock structure * @core_clk_rate: current core clock rate * @max_core_clk_rate: maximum allowable core clock rate @@ -66,7 +77,6 @@ struct dpu_core_perf { struct drm_device *dev; struct dentry *debugfs_root; struct dpu_mdss_cfg *catalog; - struct dpu_power_handle *phandle; struct dss_clk *core_clk; u64 core_clk_rate; u64 max_core_clk_rate; @@ -113,21 +123,20 @@ void dpu_core_perf_destroy(struct dpu_core_perf *perf); * @perf: Pointer to core performance context * @dev: Pointer to drm device * @catalog: Pointer to catalog - * @phandle: Pointer to power handle * @core_clk: pointer to core clock */ int dpu_core_perf_init(struct dpu_core_perf *perf, struct drm_device *dev, struct dpu_mdss_cfg *catalog, - struct dpu_power_handle *phandle, struct dss_clk *core_clk); +struct dpu_kms; + /** * dpu_core_perf_debugfs_init - initialize debugfs for core performance context - * @perf: Pointer to core performance context + * @dpu_kms: Pointer to the dpu_kms struct * @debugfs_parent: Pointer to parent debugfs */ -int dpu_core_perf_debugfs_init(struct dpu_core_perf *perf, - struct dentry *parent); +int dpu_core_perf_debugfs_init(struct dpu_kms *dpu_kms, struct dentry *parent); #endif /* _DPU_CORE_PERF_H_ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c index d4530d60767b..9be7c355debd 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c @@ -33,7 +33,6 @@ #include "dpu_plane.h" #include "dpu_encoder.h" #include "dpu_vbif.h" -#include "dpu_power_handle.h" #include "dpu_core_perf.h" #include "dpu_trace.h" @@ -47,13 +46,7 @@ #define LEFT_MIXER 0 #define RIGHT_MIXER 1 -static inline int _dpu_crtc_get_mixer_width(struct dpu_crtc_state *cstate, - struct drm_display_mode *mode) -{ - return mode->hdisplay / cstate->num_mixers; -} - -static inline struct dpu_kms *_dpu_crtc_get_kms(struct drm_crtc *crtc) +static struct dpu_kms *_dpu_crtc_get_kms(struct drm_crtc *crtc) { struct msm_drm_private *priv = crtc->dev->dev_private; @@ -69,10 +62,7 @@ static void dpu_crtc_destroy(struct drm_crtc *crtc) if (!crtc) return; - dpu_crtc->phandle = NULL; - drm_crtc_cleanup(crtc); - mutex_destroy(&dpu_crtc->crtc_lock); kfree(dpu_crtc); } @@ -287,16 +277,17 @@ enum dpu_intf_mode dpu_crtc_get_intf_mode(struct drm_crtc *crtc) return INTF_MODE_NONE; } - drm_for_each_encoder(encoder, crtc->dev) - if (encoder->crtc == crtc) - return dpu_encoder_get_intf_mode(encoder); + WARN_ON(!drm_modeset_is_locked(&crtc->mutex)); + + /* TODO: Returns the first INTF_MODE, could there be multiple values? */ + drm_for_each_encoder_mask(encoder, crtc->dev, crtc->state->encoder_mask) + return dpu_encoder_get_intf_mode(encoder); return INTF_MODE_NONE; } -static void dpu_crtc_vblank_cb(void *data) +void dpu_crtc_vblank_callback(struct drm_crtc *crtc) { - struct drm_crtc *crtc = (struct drm_crtc *)data; struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); /* keep statistics on vblank callback - with auto reset via debugfs */ @@ -309,6 +300,19 @@ static void dpu_crtc_vblank_cb(void *data) trace_dpu_crtc_vblank_cb(DRMID(crtc)); } +static void dpu_crtc_release_bw_unlocked(struct drm_crtc *crtc) +{ + int ret = 0; + struct drm_modeset_acquire_ctx ctx; + + DRM_MODESET_LOCK_ALL_BEGIN(crtc->dev, ctx, 0, ret); + dpu_core_perf_crtc_release_bw(crtc); + DRM_MODESET_LOCK_ALL_END(ctx, ret); + if (ret) + DRM_ERROR("Failed to acquire modeset locks to release bw, %d\n", + ret); +} + static void dpu_crtc_frame_event_work(struct kthread_work *work) { struct dpu_crtc_frame_event *fevent = container_of(work, @@ -338,7 +342,7 @@ static void dpu_crtc_frame_event_work(struct kthread_work *work) /* release bandwidth and other resources */ trace_dpu_crtc_frame_event_done(DRMID(crtc), fevent->event); - dpu_core_perf_crtc_release_bw(crtc); + dpu_crtc_release_bw_unlocked(crtc); } else { trace_dpu_crtc_frame_event_more_pending(DRMID(crtc), fevent->event); @@ -473,28 +477,21 @@ static void _dpu_crtc_setup_mixer_for_encoder( static void _dpu_crtc_setup_mixers(struct drm_crtc *crtc) { - struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); struct drm_encoder *enc; - mutex_lock(&dpu_crtc->crtc_lock); - /* Check for mixers on all encoders attached to this crtc */ - list_for_each_entry(enc, &crtc->dev->mode_config.encoder_list, head) { - if (enc->crtc != crtc) - continue; + WARN_ON(!drm_modeset_is_locked(&crtc->mutex)); + /* Check for mixers on all encoders attached to this crtc */ + drm_for_each_encoder_mask(enc, crtc->dev, crtc->state->encoder_mask) _dpu_crtc_setup_mixer_for_encoder(crtc, enc); - } - - mutex_unlock(&dpu_crtc->crtc_lock); } static void _dpu_crtc_setup_lm_bounds(struct drm_crtc *crtc, struct drm_crtc_state *state) { - struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); struct dpu_crtc_state *cstate = to_dpu_crtc_state(state); struct drm_display_mode *adj_mode = &state->adjusted_mode; - u32 crtc_split_width = _dpu_crtc_get_mixer_width(cstate, adj_mode); + u32 crtc_split_width = adj_mode->hdisplay / cstate->num_mixers; int i; for (i = 0; i < cstate->num_mixers; i++) { @@ -502,7 +499,7 @@ static void _dpu_crtc_setup_lm_bounds(struct drm_crtc *crtc, r->x1 = crtc_split_width * i; r->y1 = 0; r->x2 = r->x1 + crtc_split_width; - r->y2 = dpu_crtc_get_mixer_height(dpu_crtc, cstate, adj_mode); + r->y2 = adj_mode->vdisplay; trace_dpu_crtc_setup_lm_bounds(DRMID(crtc), i, r); } @@ -552,13 +549,9 @@ static void dpu_crtc_atomic_begin(struct drm_crtc *crtc, spin_unlock_irqrestore(&dev->event_lock, flags); } - list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { - if (encoder->crtc != crtc) - continue; - - /* encoder will trigger pending mask now */ + /* encoder will trigger pending mask now */ + drm_for_each_encoder_mask(encoder, crtc->dev, crtc->state->encoder_mask) dpu_encoder_trigger_kickoff_pending(encoder); - } /* * If no mixers have been allocated in dpu_crtc_atomic_check(), @@ -702,10 +695,9 @@ static int _dpu_crtc_wait_for_frame_done(struct drm_crtc *crtc) return rc; } -void dpu_crtc_commit_kickoff(struct drm_crtc *crtc) +void dpu_crtc_commit_kickoff(struct drm_crtc *crtc, bool async) { struct drm_encoder *encoder; - struct drm_device *dev = crtc->dev; struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); struct dpu_kms *dpu_kms = _dpu_crtc_get_kms(crtc); struct dpu_crtc_state *cstate = to_dpu_crtc_state(crtc->state); @@ -721,127 +713,59 @@ void dpu_crtc_commit_kickoff(struct drm_crtc *crtc) DPU_ATRACE_BEGIN("crtc_commit"); - list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { + /* + * Encoder will flush/start now, unless it has a tx pending. If so, it + * may delay and flush at an irq event (e.g. ppdone) + */ + drm_for_each_encoder_mask(encoder, crtc->dev, + crtc->state->encoder_mask) { struct dpu_encoder_kickoff_params params = { 0 }; - - if (encoder->crtc != crtc) - continue; - - /* - * Encoder will flush/start now, unless it has a tx pending. - * If so, it may delay and flush at an irq event (e.g. ppdone) - */ - dpu_encoder_prepare_for_kickoff(encoder, ¶ms); + dpu_encoder_prepare_for_kickoff(encoder, ¶ms, async); } - /* wait for frame_event_done completion */ - DPU_ATRACE_BEGIN("wait_for_frame_done_event"); - ret = _dpu_crtc_wait_for_frame_done(crtc); - DPU_ATRACE_END("wait_for_frame_done_event"); - if (ret) { - DPU_ERROR("crtc%d wait for frame done failed;frame_pending%d\n", - crtc->base.id, - atomic_read(&dpu_crtc->frame_pending)); - goto end; - } - if (atomic_inc_return(&dpu_crtc->frame_pending) == 1) { - /* acquire bandwidth and other resources */ - DPU_DEBUG("crtc%d first commit\n", crtc->base.id); - } else - DPU_DEBUG("crtc%d commit\n", crtc->base.id); + if (!async) { + /* wait for frame_event_done completion */ + DPU_ATRACE_BEGIN("wait_for_frame_done_event"); + ret = _dpu_crtc_wait_for_frame_done(crtc); + DPU_ATRACE_END("wait_for_frame_done_event"); + if (ret) { + DPU_ERROR("crtc%d wait for frame done failed;frame_pending%d\n", + crtc->base.id, + atomic_read(&dpu_crtc->frame_pending)); + goto end; + } + + if (atomic_inc_return(&dpu_crtc->frame_pending) == 1) { + /* acquire bandwidth and other resources */ + DPU_DEBUG("crtc%d first commit\n", crtc->base.id); + } else + DPU_DEBUG("crtc%d commit\n", crtc->base.id); - dpu_crtc->play_count++; + dpu_crtc->play_count++; + } dpu_vbif_clear_errors(dpu_kms); - list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { - if (encoder->crtc != crtc) - continue; - - dpu_encoder_kickoff(encoder); - } + drm_for_each_encoder_mask(encoder, crtc->dev, crtc->state->encoder_mask) + dpu_encoder_kickoff(encoder, async); end: - reinit_completion(&dpu_crtc->frame_done_comp); + if (!async) + reinit_completion(&dpu_crtc->frame_done_comp); DPU_ATRACE_END("crtc_commit"); } -/** - * _dpu_crtc_vblank_enable_no_lock - update power resource and vblank request - * @dpu_crtc: Pointer to dpu crtc structure - * @enable: Whether to enable/disable vblanks - */ -static void _dpu_crtc_vblank_enable_no_lock( - struct dpu_crtc *dpu_crtc, bool enable) -{ - struct drm_crtc *crtc = &dpu_crtc->base; - struct drm_device *dev = crtc->dev; - struct drm_encoder *enc; - - if (enable) { - /* drop lock since power crtc cb may try to re-acquire lock */ - mutex_unlock(&dpu_crtc->crtc_lock); - pm_runtime_get_sync(dev->dev); - mutex_lock(&dpu_crtc->crtc_lock); - - list_for_each_entry(enc, &dev->mode_config.encoder_list, head) { - if (enc->crtc != crtc) - continue; - - trace_dpu_crtc_vblank_enable(DRMID(&dpu_crtc->base), - DRMID(enc), enable, - dpu_crtc); - - dpu_encoder_register_vblank_callback(enc, - dpu_crtc_vblank_cb, (void *)crtc); - } - } else { - list_for_each_entry(enc, &dev->mode_config.encoder_list, head) { - if (enc->crtc != crtc) - continue; - - trace_dpu_crtc_vblank_enable(DRMID(&dpu_crtc->base), - DRMID(enc), enable, - dpu_crtc); - - dpu_encoder_register_vblank_callback(enc, NULL, NULL); - } - - /* drop lock since power crtc cb may try to re-acquire lock */ - mutex_unlock(&dpu_crtc->crtc_lock); - pm_runtime_put_sync(dev->dev); - mutex_lock(&dpu_crtc->crtc_lock); - } -} - -/** - * _dpu_crtc_set_suspend - notify crtc of suspend enable/disable - * @crtc: Pointer to drm crtc object - * @enable: true to enable suspend, false to indicate resume - */ -static void _dpu_crtc_set_suspend(struct drm_crtc *crtc, bool enable) +static void dpu_crtc_reset(struct drm_crtc *crtc) { - struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); - - DRM_DEBUG_KMS("crtc%d suspend = %d\n", crtc->base.id, enable); - - mutex_lock(&dpu_crtc->crtc_lock); + struct dpu_crtc_state *cstate; - /* - * If the vblank is enabled, release a power reference on suspend - * and take it back during resume (if it is still enabled). - */ - trace_dpu_crtc_set_suspend(DRMID(&dpu_crtc->base), enable, dpu_crtc); - if (dpu_crtc->suspend == enable) - DPU_DEBUG("crtc%d suspend already set to %d, ignoring update\n", - crtc->base.id, enable); - else if (dpu_crtc->enabled && dpu_crtc->vblank_requested) { - _dpu_crtc_vblank_enable_no_lock(dpu_crtc, !enable); - } + if (crtc->state) + dpu_crtc_destroy_state(crtc, crtc->state); - dpu_crtc->suspend = enable; - mutex_unlock(&dpu_crtc->crtc_lock); + crtc->state = kzalloc(sizeof(*cstate), GFP_KERNEL); + if (crtc->state) + crtc->state->crtc = crtc; } /** @@ -873,65 +797,8 @@ static struct drm_crtc_state *dpu_crtc_duplicate_state(struct drm_crtc *crtc) return &cstate->base; } -/** - * dpu_crtc_reset - reset hook for CRTCs - * Resets the atomic state for @crtc by freeing the state pointer (which might - * be NULL, e.g. at driver load time) and allocating a new empty state object. - * @crtc: Pointer to drm crtc structure - */ -static void dpu_crtc_reset(struct drm_crtc *crtc) -{ - struct dpu_crtc *dpu_crtc; - struct dpu_crtc_state *cstate; - - if (!crtc) { - DPU_ERROR("invalid crtc\n"); - return; - } - - /* revert suspend actions, if necessary */ - if (dpu_kms_is_suspend_state(crtc->dev)) - _dpu_crtc_set_suspend(crtc, false); - - /* remove previous state, if present */ - if (crtc->state) { - dpu_crtc_destroy_state(crtc, crtc->state); - crtc->state = 0; - } - - dpu_crtc = to_dpu_crtc(crtc); - cstate = kzalloc(sizeof(*cstate), GFP_KERNEL); - if (!cstate) { - DPU_ERROR("failed to allocate state\n"); - return; - } - - cstate->base.crtc = crtc; - crtc->state = &cstate->base; -} - -static void dpu_crtc_handle_power_event(u32 event_type, void *arg) -{ - struct drm_crtc *crtc = arg; - struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); - struct drm_encoder *encoder; - - mutex_lock(&dpu_crtc->crtc_lock); - - trace_dpu_crtc_handle_power_event(DRMID(crtc), event_type); - - /* restore encoder; crtc will be programmed during commit */ - drm_for_each_encoder(encoder, crtc->dev) { - if (encoder->crtc != crtc) - continue; - - dpu_encoder_virt_restore(encoder); - } - - mutex_unlock(&dpu_crtc->crtc_lock); -} - -static void dpu_crtc_disable(struct drm_crtc *crtc) +static void dpu_crtc_disable(struct drm_crtc *crtc, + struct drm_crtc_state *old_crtc_state) { struct dpu_crtc *dpu_crtc; struct dpu_crtc_state *cstate; @@ -951,13 +818,12 @@ static void dpu_crtc_disable(struct drm_crtc *crtc) DRM_DEBUG_KMS("crtc%d\n", crtc->base.id); - if (dpu_kms_is_suspend_state(crtc->dev)) - _dpu_crtc_set_suspend(crtc, true); - /* Disable/save vblank irq handling */ drm_crtc_vblank_off(crtc); - mutex_lock(&dpu_crtc->crtc_lock); + drm_for_each_encoder_mask(encoder, crtc->dev, + old_crtc_state->encoder_mask) + dpu_encoder_assign_crtc(encoder, NULL); /* wait for frame_event_done completion */ if (_dpu_crtc_wait_for_frame_done(crtc)) @@ -966,10 +832,6 @@ static void dpu_crtc_disable(struct drm_crtc *crtc) atomic_read(&dpu_crtc->frame_pending)); trace_dpu_crtc_disable(DRMID(crtc), false, dpu_crtc); - if (dpu_crtc->enabled && !dpu_crtc->suspend && - dpu_crtc->vblank_requested) { - _dpu_crtc_vblank_enable_no_lock(dpu_crtc, false); - } dpu_crtc->enabled = false; if (atomic_read(&dpu_crtc->frame_pending)) { @@ -981,15 +843,8 @@ static void dpu_crtc_disable(struct drm_crtc *crtc) dpu_core_perf_crtc_update(crtc, 0, true); - drm_for_each_encoder(encoder, crtc->dev) { - if (encoder->crtc != crtc) - continue; + drm_for_each_encoder_mask(encoder, crtc->dev, crtc->state->encoder_mask) dpu_encoder_register_frame_event_callback(encoder, NULL, NULL); - } - - if (dpu_crtc->power_event) - dpu_power_handle_unregister_event(dpu_crtc->phandle, - dpu_crtc->power_event); memset(cstate->mixers, 0, sizeof(cstate->mixers)); cstate->num_mixers = 0; @@ -998,14 +853,14 @@ static void dpu_crtc_disable(struct drm_crtc *crtc) cstate->bw_control = false; cstate->bw_split_vote = false; - mutex_unlock(&dpu_crtc->crtc_lock); - if (crtc->state->event && !crtc->state->active) { spin_lock_irqsave(&crtc->dev->event_lock, flags); drm_crtc_send_vblank_event(crtc, crtc->state->event); crtc->state->event = NULL; spin_unlock_irqrestore(&crtc->dev->event_lock, flags); } + + pm_runtime_put_sync(crtc->dev->dev); } static void dpu_crtc_enable(struct drm_crtc *crtc, @@ -1021,33 +876,23 @@ static void dpu_crtc_enable(struct drm_crtc *crtc, } priv = crtc->dev->dev_private; + pm_runtime_get_sync(crtc->dev->dev); + DRM_DEBUG_KMS("crtc%d\n", crtc->base.id); dpu_crtc = to_dpu_crtc(crtc); - drm_for_each_encoder(encoder, crtc->dev) { - if (encoder->crtc != crtc) - continue; + drm_for_each_encoder_mask(encoder, crtc->dev, crtc->state->encoder_mask) dpu_encoder_register_frame_event_callback(encoder, dpu_crtc_frame_event_cb, (void *)crtc); - } - mutex_lock(&dpu_crtc->crtc_lock); trace_dpu_crtc_enable(DRMID(crtc), true, dpu_crtc); - if (!dpu_crtc->enabled && !dpu_crtc->suspend && - dpu_crtc->vblank_requested) { - _dpu_crtc_vblank_enable_no_lock(dpu_crtc, true); - } dpu_crtc->enabled = true; - mutex_unlock(&dpu_crtc->crtc_lock); + drm_for_each_encoder_mask(encoder, crtc->dev, crtc->state->encoder_mask) + dpu_encoder_assign_crtc(encoder, crtc); /* Enable/restore vblank irq handling */ drm_crtc_vblank_on(crtc); - - dpu_crtc->power_event = dpu_power_handle_register_event( - dpu_crtc->phandle, DPU_POWER_EVENT_ENABLE, - dpu_crtc_handle_power_event, crtc, dpu_crtc->name); - } struct plane_state { @@ -1101,7 +946,7 @@ static int dpu_crtc_atomic_check(struct drm_crtc *crtc, memset(pipe_staged, 0, sizeof(pipe_staged)); - mixer_width = _dpu_crtc_get_mixer_width(cstate, mode); + mixer_width = mode->hdisplay / cstate->num_mixers; _dpu_crtc_setup_lm_bounds(crtc, state); @@ -1289,21 +1134,32 @@ end: int dpu_crtc_vblank(struct drm_crtc *crtc, bool en) { - struct dpu_crtc *dpu_crtc; - - if (!crtc) { - DPU_ERROR("invalid crtc\n"); - return -EINVAL; - } - dpu_crtc = to_dpu_crtc(crtc); + struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); + struct drm_encoder *enc; - mutex_lock(&dpu_crtc->crtc_lock); trace_dpu_crtc_vblank(DRMID(&dpu_crtc->base), en, dpu_crtc); - if (dpu_crtc->enabled && !dpu_crtc->suspend) { - _dpu_crtc_vblank_enable_no_lock(dpu_crtc, en); + + /* + * Normally we would iterate through encoder_mask in crtc state to find + * attached encoders. In this case, we might be disabling vblank _after_ + * encoder_mask has been cleared. + * + * Instead, we "assign" a crtc to the encoder in enable and clear it in + * disable (which is also after encoder_mask is cleared). So instead of + * using encoder mask, we'll ask the encoder to toggle itself iff it's + * currently assigned to our crtc. + * + * Note also that this function cannot be called while crtc is disabled + * since we use drm_crtc_vblank_on/off. So we don't need to worry + * about the assigned crtcs being inconsistent with the current state + * (which means no need to worry about modeset locks). + */ + list_for_each_entry(enc, &crtc->dev->mode_config.encoder_list, head) { + trace_dpu_crtc_vblank_enable(DRMID(crtc), DRMID(enc), en, + dpu_crtc); + + dpu_encoder_toggle_vblank_for_crtc(enc, crtc, en); } - dpu_crtc->vblank_requested = en; - mutex_unlock(&dpu_crtc->crtc_lock); return 0; } @@ -1324,18 +1180,14 @@ static int _dpu_debugfs_status_show(struct seq_file *s, void *data) int i, out_width; - if (!s || !s->private) - return -EINVAL; - dpu_crtc = s->private; crtc = &dpu_crtc->base; drm_modeset_lock_all(crtc->dev); cstate = to_dpu_crtc_state(crtc->state); - mutex_lock(&dpu_crtc->crtc_lock); mode = &crtc->state->adjusted_mode; - out_width = _dpu_crtc_get_mixer_width(cstate, mode); + out_width = mode->hdisplay / cstate->num_mixers; seq_printf(s, "crtc:%d width:%d height:%d\n", crtc->base.id, mode->hdisplay, mode->vdisplay); @@ -1420,9 +1272,6 @@ static int _dpu_debugfs_status_show(struct seq_file *s, void *data) dpu_crtc->vblank_cb_time = ktime_set(0, 0); } - seq_printf(s, "vblank_enable:%d\n", dpu_crtc->vblank_requested); - - mutex_unlock(&dpu_crtc->crtc_lock); drm_modeset_unlock_all(crtc->dev); return 0; @@ -1456,13 +1305,11 @@ static int dpu_crtc_debugfs_state_show(struct seq_file *s, void *v) seq_printf(s, "intf_mode: %d\n", dpu_crtc_get_intf_mode(crtc)); seq_printf(s, "core_clk_rate: %llu\n", dpu_crtc->cur_perf.core_clk_rate); - for (i = DPU_POWER_HANDLE_DBUS_ID_MNOC; - i < DPU_POWER_HANDLE_DBUS_ID_MAX; i++) { - seq_printf(s, "bw_ctl[%s]: %llu\n", - dpu_power_handle_get_dbus_name(i), + for (i = DPU_CORE_PERF_DATA_BUS_ID_MNOC; + i < DPU_CORE_PERF_DATA_BUS_ID_MAX; i++) { + seq_printf(s, "bw_ctl[%d]: %llu\n", i, dpu_crtc->cur_perf.bw_ctl[i]); - seq_printf(s, "max_per_pipe_ib[%s]: %llu\n", - dpu_power_handle_get_dbus_name(i), + seq_printf(s, "max_per_pipe_ib[%d]: %llu\n", i, dpu_crtc->cur_perf.max_per_pipe_ib[i]); } @@ -1472,8 +1319,7 @@ DEFINE_DPU_DEBUGFS_SEQ_FOPS(dpu_crtc_debugfs_state); static int _dpu_crtc_init_debugfs(struct drm_crtc *crtc) { - struct dpu_crtc *dpu_crtc; - struct dpu_kms *dpu_kms; + struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); static const struct file_operations debugfs_status_fops = { .open = _dpu_debugfs_status_open, @@ -1482,12 +1328,6 @@ static int _dpu_crtc_init_debugfs(struct drm_crtc *crtc) .release = single_release, }; - if (!crtc) - return -EINVAL; - dpu_crtc = to_dpu_crtc(crtc); - - dpu_kms = _dpu_crtc_get_kms(crtc); - dpu_crtc->debugfs_root = debugfs_create_dir(dpu_crtc->name, crtc->dev->primary->debugfs_root); if (!dpu_crtc->debugfs_root) @@ -1504,25 +1344,11 @@ static int _dpu_crtc_init_debugfs(struct drm_crtc *crtc) return 0; } - -static void _dpu_crtc_destroy_debugfs(struct drm_crtc *crtc) -{ - struct dpu_crtc *dpu_crtc; - - if (!crtc) - return; - dpu_crtc = to_dpu_crtc(crtc); - debugfs_remove_recursive(dpu_crtc->debugfs_root); -} #else static int _dpu_crtc_init_debugfs(struct drm_crtc *crtc) { return 0; } - -static void _dpu_crtc_destroy_debugfs(struct drm_crtc *crtc) -{ -} #endif /* CONFIG_DEBUG_FS */ static int dpu_crtc_late_register(struct drm_crtc *crtc) @@ -1532,7 +1358,9 @@ static int dpu_crtc_late_register(struct drm_crtc *crtc) static void dpu_crtc_early_unregister(struct drm_crtc *crtc) { - _dpu_crtc_destroy_debugfs(crtc); + struct dpu_crtc *dpu_crtc = to_dpu_crtc(crtc); + + debugfs_remove_recursive(dpu_crtc->debugfs_root); } static const struct drm_crtc_funcs dpu_crtc_funcs = { @@ -1547,7 +1375,7 @@ static const struct drm_crtc_funcs dpu_crtc_funcs = { }; static const struct drm_crtc_helper_funcs dpu_crtc_helper_funcs = { - .disable = dpu_crtc_disable, + .atomic_disable = dpu_crtc_disable, .atomic_enable = dpu_crtc_enable, .atomic_check = dpu_crtc_atomic_check, .atomic_begin = dpu_crtc_atomic_begin, @@ -1574,7 +1402,6 @@ struct drm_crtc *dpu_crtc_init(struct drm_device *dev, struct drm_plane *plane, crtc = &dpu_crtc->base; crtc->dev = dev; - mutex_init(&dpu_crtc->crtc_lock); spin_lock_init(&dpu_crtc->spin_lock); atomic_set(&dpu_crtc->frame_pending, 0); @@ -1594,7 +1421,6 @@ struct drm_crtc *dpu_crtc_init(struct drm_device *dev, struct drm_plane *plane, NULL); drm_crtc_helper_add(crtc, &dpu_crtc_helper_funcs); - plane->crtc = crtc; /* save user friendly CRTC name for later */ snprintf(dpu_crtc->name, DPU_CRTC_NAME_SIZE, "crtc%u", crtc->base.id); @@ -1602,8 +1428,6 @@ struct drm_crtc *dpu_crtc_init(struct drm_device *dev, struct drm_plane *plane, /* initialize event handling */ spin_lock_init(&dpu_crtc->event_lock); - dpu_crtc->phandle = &kms->phandle; - DPU_DEBUG("%s: successfully initialized crtc\n", dpu_crtc->name); return crtc; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.h index 3723b4830335..dbfb38a1986c 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.h @@ -132,8 +132,6 @@ struct dpu_crtc_frame_event { * @vblank_cb_count : count of vblank callback since last reset * @play_count : frame count between crtc enable and disable * @vblank_cb_time : ktime at vblank count reset - * @vblank_requested : whether the user has requested vblank events - * @suspend : whether or not a suspend operation is in progress * @enabled : whether the DPU CRTC is currently enabled. updated in the * commit-thread, not state-swap time which is earlier, so * safe to make decisions on during VBLANK on/off work @@ -142,7 +140,6 @@ struct dpu_crtc_frame_event { * @dirty_list : list of color processing features are dirty * @ad_dirty: list containing ad properties that are dirty * @ad_active: list containing ad properties that are active - * @crtc_lock : crtc lock around create, destroy and access. * @frame_pending : Whether or not an update is pending * @frame_events : static allocation of in-flight frame events * @frame_event_list : available frame event list @@ -152,7 +149,6 @@ struct dpu_crtc_frame_event { * @event_worker : Event worker queue * @event_lock : Spinlock around event handling code * @phandle: Pointer to power handler - * @power_event : registered power event handle * @cur_perf : current performance committed to clock/bandwidth driver */ struct dpu_crtc { @@ -168,8 +164,6 @@ struct dpu_crtc { u32 vblank_cb_count; u64 play_count; ktime_t vblank_cb_time; - bool vblank_requested; - bool suspend; bool enabled; struct list_head feature_list; @@ -178,8 +172,6 @@ struct dpu_crtc { struct list_head ad_dirty; struct list_head ad_active; - struct mutex crtc_lock; - atomic_t frame_pending; struct dpu_crtc_frame_event frame_events[DPU_CRTC_FRAME_EVENT_SIZE]; struct list_head frame_event_list; @@ -189,9 +181,6 @@ struct dpu_crtc { /* for handling internal event thread */ spinlock_t event_lock; - struct dpu_power_handle *phandle; - struct dpu_power_event *power_event; - struct dpu_core_perf_params cur_perf; struct dpu_crtc_smmu_state_data smmu_state; @@ -238,41 +227,12 @@ struct dpu_crtc_state { container_of(x, struct dpu_crtc_state, base) /** - * dpu_crtc_state_is_stereo - Is crtc virtualized with two mixers? - * @cstate: Pointer to dpu crtc state - * @Return: true - has two mixers, false - has one mixer - */ -static inline bool dpu_crtc_state_is_stereo(struct dpu_crtc_state *cstate) -{ - return cstate->num_mixers == CRTC_DUAL_MIXERS; -} - -/** - * dpu_crtc_get_mixer_height - get the mixer height - * Mixer height will be same as panel height - */ -static inline int dpu_crtc_get_mixer_height(struct dpu_crtc *dpu_crtc, - struct dpu_crtc_state *cstate, struct drm_display_mode *mode) -{ - if (!dpu_crtc || !cstate || !mode) - return 0; - - return mode->vdisplay; -} - -/** * dpu_crtc_frame_pending - retun the number of pending frames * @crtc: Pointer to drm crtc object */ static inline int dpu_crtc_frame_pending(struct drm_crtc *crtc) { - struct dpu_crtc *dpu_crtc; - - if (!crtc) - return -EINVAL; - - dpu_crtc = to_dpu_crtc(crtc); - return atomic_read(&dpu_crtc->frame_pending); + return crtc ? atomic_read(&to_dpu_crtc(crtc)->frame_pending) : -EINVAL; } /** @@ -283,10 +243,17 @@ static inline int dpu_crtc_frame_pending(struct drm_crtc *crtc) int dpu_crtc_vblank(struct drm_crtc *crtc, bool en); /** + * dpu_crtc_vblank_callback - called on vblank irq, issues completion events + * @crtc: Pointer to drm crtc object + */ +void dpu_crtc_vblank_callback(struct drm_crtc *crtc); + +/** * dpu_crtc_commit_kickoff - trigger kickoff of the commit for this crtc * @crtc: Pointer to drm crtc object + * @async: true if the commit is asynchronous, false otherwise */ -void dpu_crtc_commit_kickoff(struct drm_crtc *crtc); +void dpu_crtc_commit_kickoff(struct drm_crtc *crtc, bool async); /** * dpu_crtc_complete_commit - callback signalling completion of current commit @@ -329,22 +296,7 @@ enum dpu_intf_mode dpu_crtc_get_intf_mode(struct drm_crtc *crtc); static inline enum dpu_crtc_client_type dpu_crtc_get_client_type( struct drm_crtc *crtc) { - struct dpu_crtc_state *cstate = - crtc ? to_dpu_crtc_state(crtc->state) : NULL; - - if (!cstate) - return NRT_CLIENT; - - return RT_CLIENT; -} - -/** - * dpu_crtc_is_enabled - check if dpu crtc is enabled or not - * @crtc: Pointer to crtc - */ -static inline bool dpu_crtc_is_enabled(struct drm_crtc *crtc) -{ - return crtc ? crtc->enabled : false; + return crtc && crtc->state ? RT_CLIENT : NRT_CLIENT; } #endif /* _DPU_CRTC_H_ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_dbg.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_dbg.c deleted file mode 100644 index ae2aee7ed9e1..000000000000 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_dbg.c +++ /dev/null @@ -1,2393 +0,0 @@ -/* Copyright (c) 2009-2018, The Linux Foundation. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 and - * only version 2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - */ - -#define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ - -#include <linux/delay.h> -#include <linux/spinlock.h> -#include <linux/ktime.h> -#include <linux/debugfs.h> -#include <linux/uaccess.h> -#include <linux/dma-buf.h> -#include <linux/slab.h> -#include <linux/list_sort.h> -#include <linux/pm_runtime.h> - -#include "dpu_dbg.h" -#include "disp/dpu1/dpu_hw_catalog.h" - - -#define DEFAULT_DBGBUS_DPU DPU_DBG_DUMP_IN_MEM -#define DEFAULT_DBGBUS_VBIFRT DPU_DBG_DUMP_IN_MEM -#define REG_BASE_NAME_LEN 80 - -#define DBGBUS_FLAGS_DSPP BIT(0) -#define DBGBUS_DSPP_STATUS 0x34C - -#define DBGBUS_NAME_DPU "dpu" -#define DBGBUS_NAME_VBIF_RT "vbif_rt" - -/* offsets from dpu top address for the debug buses */ -#define DBGBUS_SSPP0 0x188 -#define DBGBUS_AXI_INTF 0x194 -#define DBGBUS_SSPP1 0x298 -#define DBGBUS_DSPP 0x348 -#define DBGBUS_PERIPH 0x418 - -#define TEST_MASK(id, tp) ((id << 4) | (tp << 1) | BIT(0)) - -/* following offsets are with respect to MDP VBIF base for DBG BUS access */ -#define MMSS_VBIF_CLKON 0x4 -#define MMSS_VBIF_TEST_BUS_OUT_CTRL 0x210 -#define MMSS_VBIF_TEST_BUS_OUT 0x230 - -/* Vbif error info */ -#define MMSS_VBIF_PND_ERR 0x190 -#define MMSS_VBIF_SRC_ERR 0x194 -#define MMSS_VBIF_XIN_HALT_CTRL1 0x204 -#define MMSS_VBIF_ERR_INFO 0X1a0 -#define MMSS_VBIF_ERR_INFO_1 0x1a4 -#define MMSS_VBIF_CLIENT_NUM 14 - -/** - * struct dpu_dbg_reg_base - register region base. - * may sub-ranges: sub-ranges are used for dumping - * or may not have sub-ranges: dumping is base -> max_offset - * @reg_base_head: head of this node - * @name: register base name - * @base: base pointer - * @off: cached offset of region for manual register dumping - * @cnt: cached range of region for manual register dumping - * @max_offset: length of region - * @buf: buffer used for manual register dumping - * @buf_len: buffer length used for manual register dumping - * @cb: callback for external dump function, null if not defined - * @cb_ptr: private pointer to callback function - */ -struct dpu_dbg_reg_base { - struct list_head reg_base_head; - char name[REG_BASE_NAME_LEN]; - void __iomem *base; - size_t off; - size_t cnt; - size_t max_offset; - char *buf; - size_t buf_len; - void (*cb)(void *ptr); - void *cb_ptr; -}; - -struct dpu_debug_bus_entry { - u32 wr_addr; - u32 block_id; - u32 test_id; - void (*analyzer)(void __iomem *mem_base, - struct dpu_debug_bus_entry *entry, u32 val); -}; - -struct vbif_debug_bus_entry { - u32 disable_bus_addr; - u32 block_bus_addr; - u32 bit_offset; - u32 block_cnt; - u32 test_pnt_start; - u32 test_pnt_cnt; -}; - -struct dpu_dbg_debug_bus_common { - char *name; - u32 enable_mask; - bool include_in_deferred_work; - u32 flags; - u32 entries_size; - u32 *dumped_content; -}; - -struct dpu_dbg_dpu_debug_bus { - struct dpu_dbg_debug_bus_common cmn; - struct dpu_debug_bus_entry *entries; - u32 top_blk_off; -}; - -struct dpu_dbg_vbif_debug_bus { - struct dpu_dbg_debug_bus_common cmn; - struct vbif_debug_bus_entry *entries; -}; - -/** - * struct dpu_dbg_base - global dpu debug base structure - * @reg_base_list: list of register dumping regions - * @dev: device pointer - * @dump_work: work struct for deferring register dump work to separate thread - * @dbgbus_dpu: debug bus structure for the dpu - * @dbgbus_vbif_rt: debug bus structure for the realtime vbif - */ -static struct dpu_dbg_base { - struct list_head reg_base_list; - struct device *dev; - - struct work_struct dump_work; - - struct dpu_dbg_dpu_debug_bus dbgbus_dpu; - struct dpu_dbg_vbif_debug_bus dbgbus_vbif_rt; -} dpu_dbg_base; - -static void _dpu_debug_bus_xbar_dump(void __iomem *mem_base, - struct dpu_debug_bus_entry *entry, u32 val) -{ - dev_err(dpu_dbg_base.dev, "xbar 0x%x %d %d 0x%x\n", - entry->wr_addr, entry->block_id, entry->test_id, val); -} - -static void _dpu_debug_bus_lm_dump(void __iomem *mem_base, - struct dpu_debug_bus_entry *entry, u32 val) -{ - if (!(val & 0xFFF000)) - return; - - dev_err(dpu_dbg_base.dev, "lm 0x%x %d %d 0x%x\n", - entry->wr_addr, entry->block_id, entry->test_id, val); -} - -static void _dpu_debug_bus_ppb0_dump(void __iomem *mem_base, - struct dpu_debug_bus_entry *entry, u32 val) -{ - if (!(val & BIT(15))) - return; - - dev_err(dpu_dbg_base.dev, "ppb0 0x%x %d %d 0x%x\n", - entry->wr_addr, entry->block_id, entry->test_id, val); -} - -static void _dpu_debug_bus_ppb1_dump(void __iomem *mem_base, - struct dpu_debug_bus_entry *entry, u32 val) -{ - if (!(val & BIT(15))) - return; - - dev_err(dpu_dbg_base.dev, "ppb1 0x%x %d %d 0x%x\n", - entry->wr_addr, entry->block_id, entry->test_id, val); -} - -static struct dpu_debug_bus_entry dbg_bus_dpu_8998[] = { - - /* Unpack 0 sspp 0*/ - { DBGBUS_SSPP0, 50, 2 }, - { DBGBUS_SSPP0, 60, 2 }, - { DBGBUS_SSPP0, 70, 2 }, - { DBGBUS_SSPP0, 85, 2 }, - - /* Upack 0 sspp 1*/ - { DBGBUS_SSPP1, 50, 2 }, - { DBGBUS_SSPP1, 60, 2 }, - { DBGBUS_SSPP1, 70, 2 }, - { DBGBUS_SSPP1, 85, 2 }, - - /* scheduler */ - { DBGBUS_DSPP, 130, 0 }, - { DBGBUS_DSPP, 130, 1 }, - { DBGBUS_DSPP, 130, 2 }, - { DBGBUS_DSPP, 130, 3 }, - { DBGBUS_DSPP, 130, 4 }, - { DBGBUS_DSPP, 130, 5 }, - - /* qseed */ - { DBGBUS_SSPP0, 6, 0}, - { DBGBUS_SSPP0, 6, 1}, - { DBGBUS_SSPP0, 26, 0}, - { DBGBUS_SSPP0, 26, 1}, - { DBGBUS_SSPP1, 6, 0}, - { DBGBUS_SSPP1, 6, 1}, - { DBGBUS_SSPP1, 26, 0}, - { DBGBUS_SSPP1, 26, 1}, - - /* scale */ - { DBGBUS_SSPP0, 16, 0}, - { DBGBUS_SSPP0, 16, 1}, - { DBGBUS_SSPP0, 36, 0}, - { DBGBUS_SSPP0, 36, 1}, - { DBGBUS_SSPP1, 16, 0}, - { DBGBUS_SSPP1, 16, 1}, - { DBGBUS_SSPP1, 36, 0}, - { DBGBUS_SSPP1, 36, 1}, - - /* fetch sspp0 */ - - /* vig 0 */ - { DBGBUS_SSPP0, 0, 0 }, - { DBGBUS_SSPP0, 0, 1 }, - { DBGBUS_SSPP0, 0, 2 }, - { DBGBUS_SSPP0, 0, 3 }, - { DBGBUS_SSPP0, 0, 4 }, - { DBGBUS_SSPP0, 0, 5 }, - { DBGBUS_SSPP0, 0, 6 }, - { DBGBUS_SSPP0, 0, 7 }, - - { DBGBUS_SSPP0, 1, 0 }, - { DBGBUS_SSPP0, 1, 1 }, - { DBGBUS_SSPP0, 1, 2 }, - { DBGBUS_SSPP0, 1, 3 }, - { DBGBUS_SSPP0, 1, 4 }, - { DBGBUS_SSPP0, 1, 5 }, - { DBGBUS_SSPP0, 1, 6 }, - { DBGBUS_SSPP0, 1, 7 }, - - { DBGBUS_SSPP0, 2, 0 }, - { DBGBUS_SSPP0, 2, 1 }, - { DBGBUS_SSPP0, 2, 2 }, - { DBGBUS_SSPP0, 2, 3 }, - { DBGBUS_SSPP0, 2, 4 }, - { DBGBUS_SSPP0, 2, 5 }, - { DBGBUS_SSPP0, 2, 6 }, - { DBGBUS_SSPP0, 2, 7 }, - - { DBGBUS_SSPP0, 4, 0 }, - { DBGBUS_SSPP0, 4, 1 }, - { DBGBUS_SSPP0, 4, 2 }, - { DBGBUS_SSPP0, 4, 3 }, - { DBGBUS_SSPP0, 4, 4 }, - { DBGBUS_SSPP0, 4, 5 }, - { DBGBUS_SSPP0, 4, 6 }, - { DBGBUS_SSPP0, 4, 7 }, - - { DBGBUS_SSPP0, 5, 0 }, - { DBGBUS_SSPP0, 5, 1 }, - { DBGBUS_SSPP0, 5, 2 }, - { DBGBUS_SSPP0, 5, 3 }, - { DBGBUS_SSPP0, 5, 4 }, - { DBGBUS_SSPP0, 5, 5 }, - { DBGBUS_SSPP0, 5, 6 }, - { DBGBUS_SSPP0, 5, 7 }, - - /* vig 2 */ - { DBGBUS_SSPP0, 20, 0 }, - { DBGBUS_SSPP0, 20, 1 }, - { DBGBUS_SSPP0, 20, 2 }, - { DBGBUS_SSPP0, 20, 3 }, - { DBGBUS_SSPP0, 20, 4 }, - { DBGBUS_SSPP0, 20, 5 }, - { DBGBUS_SSPP0, 20, 6 }, - { DBGBUS_SSPP0, 20, 7 }, - - { DBGBUS_SSPP0, 21, 0 }, - { DBGBUS_SSPP0, 21, 1 }, - { DBGBUS_SSPP0, 21, 2 }, - { DBGBUS_SSPP0, 21, 3 }, - { DBGBUS_SSPP0, 21, 4 }, - { DBGBUS_SSPP0, 21, 5 }, - { DBGBUS_SSPP0, 21, 6 }, - { DBGBUS_SSPP0, 21, 7 }, - - { DBGBUS_SSPP0, 22, 0 }, - { DBGBUS_SSPP0, 22, 1 }, - { DBGBUS_SSPP0, 22, 2 }, - { DBGBUS_SSPP0, 22, 3 }, - { DBGBUS_SSPP0, 22, 4 }, - { DBGBUS_SSPP0, 22, 5 }, - { DBGBUS_SSPP0, 22, 6 }, - { DBGBUS_SSPP0, 22, 7 }, - - { DBGBUS_SSPP0, 24, 0 }, - { DBGBUS_SSPP0, 24, 1 }, - { DBGBUS_SSPP0, 24, 2 }, - { DBGBUS_SSPP0, 24, 3 }, - { DBGBUS_SSPP0, 24, 4 }, - { DBGBUS_SSPP0, 24, 5 }, - { DBGBUS_SSPP0, 24, 6 }, - { DBGBUS_SSPP0, 24, 7 }, - - { DBGBUS_SSPP0, 25, 0 }, - { DBGBUS_SSPP0, 25, 1 }, - { DBGBUS_SSPP0, 25, 2 }, - { DBGBUS_SSPP0, 25, 3 }, - { DBGBUS_SSPP0, 25, 4 }, - { DBGBUS_SSPP0, 25, 5 }, - { DBGBUS_SSPP0, 25, 6 }, - { DBGBUS_SSPP0, 25, 7 }, - - /* dma 2 */ - { DBGBUS_SSPP0, 30, 0 }, - { DBGBUS_SSPP0, 30, 1 }, - { DBGBUS_SSPP0, 30, 2 }, - { DBGBUS_SSPP0, 30, 3 }, - { DBGBUS_SSPP0, 30, 4 }, - { DBGBUS_SSPP0, 30, 5 }, - { DBGBUS_SSPP0, 30, 6 }, - { DBGBUS_SSPP0, 30, 7 }, - - { DBGBUS_SSPP0, 31, 0 }, - { DBGBUS_SSPP0, 31, 1 }, - { DBGBUS_SSPP0, 31, 2 }, - { DBGBUS_SSPP0, 31, 3 }, - { DBGBUS_SSPP0, 31, 4 }, - { DBGBUS_SSPP0, 31, 5 }, - { DBGBUS_SSPP0, 31, 6 }, - { DBGBUS_SSPP0, 31, 7 }, - - { DBGBUS_SSPP0, 32, 0 }, - { DBGBUS_SSPP0, 32, 1 }, - { DBGBUS_SSPP0, 32, 2 }, - { DBGBUS_SSPP0, 32, 3 }, - { DBGBUS_SSPP0, 32, 4 }, - { DBGBUS_SSPP0, 32, 5 }, - { DBGBUS_SSPP0, 32, 6 }, - { DBGBUS_SSPP0, 32, 7 }, - - { DBGBUS_SSPP0, 33, 0 }, - { DBGBUS_SSPP0, 33, 1 }, - { DBGBUS_SSPP0, 33, 2 }, - { DBGBUS_SSPP0, 33, 3 }, - { DBGBUS_SSPP0, 33, 4 }, - { DBGBUS_SSPP0, 33, 5 }, - { DBGBUS_SSPP0, 33, 6 }, - { DBGBUS_SSPP0, 33, 7 }, - - { DBGBUS_SSPP0, 34, 0 }, - { DBGBUS_SSPP0, 34, 1 }, - { DBGBUS_SSPP0, 34, 2 }, - { DBGBUS_SSPP0, 34, 3 }, - { DBGBUS_SSPP0, 34, 4 }, - { DBGBUS_SSPP0, 34, 5 }, - { DBGBUS_SSPP0, 34, 6 }, - { DBGBUS_SSPP0, 34, 7 }, - - { DBGBUS_SSPP0, 35, 0 }, - { DBGBUS_SSPP0, 35, 1 }, - { DBGBUS_SSPP0, 35, 2 }, - { DBGBUS_SSPP0, 35, 3 }, - - /* dma 0 */ - { DBGBUS_SSPP0, 40, 0 }, - { DBGBUS_SSPP0, 40, 1 }, - { DBGBUS_SSPP0, 40, 2 }, - { DBGBUS_SSPP0, 40, 3 }, - { DBGBUS_SSPP0, 40, 4 }, - { DBGBUS_SSPP0, 40, 5 }, - { DBGBUS_SSPP0, 40, 6 }, - { DBGBUS_SSPP0, 40, 7 }, - - { DBGBUS_SSPP0, 41, 0 }, - { DBGBUS_SSPP0, 41, 1 }, - { DBGBUS_SSPP0, 41, 2 }, - { DBGBUS_SSPP0, 41, 3 }, - { DBGBUS_SSPP0, 41, 4 }, - { DBGBUS_SSPP0, 41, 5 }, - { DBGBUS_SSPP0, 41, 6 }, - { DBGBUS_SSPP0, 41, 7 }, - - { DBGBUS_SSPP0, 42, 0 }, - { DBGBUS_SSPP0, 42, 1 }, - { DBGBUS_SSPP0, 42, 2 }, - { DBGBUS_SSPP0, 42, 3 }, - { DBGBUS_SSPP0, 42, 4 }, - { DBGBUS_SSPP0, 42, 5 }, - { DBGBUS_SSPP0, 42, 6 }, - { DBGBUS_SSPP0, 42, 7 }, - - { DBGBUS_SSPP0, 44, 0 }, - { DBGBUS_SSPP0, 44, 1 }, - { DBGBUS_SSPP0, 44, 2 }, - { DBGBUS_SSPP0, 44, 3 }, - { DBGBUS_SSPP0, 44, 4 }, - { DBGBUS_SSPP0, 44, 5 }, - { DBGBUS_SSPP0, 44, 6 }, - { DBGBUS_SSPP0, 44, 7 }, - - { DBGBUS_SSPP0, 45, 0 }, - { DBGBUS_SSPP0, 45, 1 }, - { DBGBUS_SSPP0, 45, 2 }, - { DBGBUS_SSPP0, 45, 3 }, - { DBGBUS_SSPP0, 45, 4 }, - { DBGBUS_SSPP0, 45, 5 }, - { DBGBUS_SSPP0, 45, 6 }, - { DBGBUS_SSPP0, 45, 7 }, - - /* fetch sspp1 */ - /* vig 1 */ - { DBGBUS_SSPP1, 0, 0 }, - { DBGBUS_SSPP1, 0, 1 }, - { DBGBUS_SSPP1, 0, 2 }, - { DBGBUS_SSPP1, 0, 3 }, - { DBGBUS_SSPP1, 0, 4 }, - { DBGBUS_SSPP1, 0, 5 }, - { DBGBUS_SSPP1, 0, 6 }, - { DBGBUS_SSPP1, 0, 7 }, - - { DBGBUS_SSPP1, 1, 0 }, - { DBGBUS_SSPP1, 1, 1 }, - { DBGBUS_SSPP1, 1, 2 }, - { DBGBUS_SSPP1, 1, 3 }, - { DBGBUS_SSPP1, 1, 4 }, - { DBGBUS_SSPP1, 1, 5 }, - { DBGBUS_SSPP1, 1, 6 }, - { DBGBUS_SSPP1, 1, 7 }, - - { DBGBUS_SSPP1, 2, 0 }, - { DBGBUS_SSPP1, 2, 1 }, - { DBGBUS_SSPP1, 2, 2 }, - { DBGBUS_SSPP1, 2, 3 }, - { DBGBUS_SSPP1, 2, 4 }, - { DBGBUS_SSPP1, 2, 5 }, - { DBGBUS_SSPP1, 2, 6 }, - { DBGBUS_SSPP1, 2, 7 }, - - { DBGBUS_SSPP1, 4, 0 }, - { DBGBUS_SSPP1, 4, 1 }, - { DBGBUS_SSPP1, 4, 2 }, - { DBGBUS_SSPP1, 4, 3 }, - { DBGBUS_SSPP1, 4, 4 }, - { DBGBUS_SSPP1, 4, 5 }, - { DBGBUS_SSPP1, 4, 6 }, - { DBGBUS_SSPP1, 4, 7 }, - - { DBGBUS_SSPP1, 5, 0 }, - { DBGBUS_SSPP1, 5, 1 }, - { DBGBUS_SSPP1, 5, 2 }, - { DBGBUS_SSPP1, 5, 3 }, - { DBGBUS_SSPP1, 5, 4 }, - { DBGBUS_SSPP1, 5, 5 }, - { DBGBUS_SSPP1, 5, 6 }, - { DBGBUS_SSPP1, 5, 7 }, - - /* vig 3 */ - { DBGBUS_SSPP1, 20, 0 }, - { DBGBUS_SSPP1, 20, 1 }, - { DBGBUS_SSPP1, 20, 2 }, - { DBGBUS_SSPP1, 20, 3 }, - { DBGBUS_SSPP1, 20, 4 }, - { DBGBUS_SSPP1, 20, 5 }, - { DBGBUS_SSPP1, 20, 6 }, - { DBGBUS_SSPP1, 20, 7 }, - - { DBGBUS_SSPP1, 21, 0 }, - { DBGBUS_SSPP1, 21, 1 }, - { DBGBUS_SSPP1, 21, 2 }, - { DBGBUS_SSPP1, 21, 3 }, - { DBGBUS_SSPP1, 21, 4 }, - { DBGBUS_SSPP1, 21, 5 }, - { DBGBUS_SSPP1, 21, 6 }, - { DBGBUS_SSPP1, 21, 7 }, - - { DBGBUS_SSPP1, 22, 0 }, - { DBGBUS_SSPP1, 22, 1 }, - { DBGBUS_SSPP1, 22, 2 }, - { DBGBUS_SSPP1, 22, 3 }, - { DBGBUS_SSPP1, 22, 4 }, - { DBGBUS_SSPP1, 22, 5 }, - { DBGBUS_SSPP1, 22, 6 }, - { DBGBUS_SSPP1, 22, 7 }, - - { DBGBUS_SSPP1, 24, 0 }, - { DBGBUS_SSPP1, 24, 1 }, - { DBGBUS_SSPP1, 24, 2 }, - { DBGBUS_SSPP1, 24, 3 }, - { DBGBUS_SSPP1, 24, 4 }, - { DBGBUS_SSPP1, 24, 5 }, - { DBGBUS_SSPP1, 24, 6 }, - { DBGBUS_SSPP1, 24, 7 }, - - { DBGBUS_SSPP1, 25, 0 }, - { DBGBUS_SSPP1, 25, 1 }, - { DBGBUS_SSPP1, 25, 2 }, - { DBGBUS_SSPP1, 25, 3 }, - { DBGBUS_SSPP1, 25, 4 }, - { DBGBUS_SSPP1, 25, 5 }, - { DBGBUS_SSPP1, 25, 6 }, - { DBGBUS_SSPP1, 25, 7 }, - - /* dma 3 */ - { DBGBUS_SSPP1, 30, 0 }, - { DBGBUS_SSPP1, 30, 1 }, - { DBGBUS_SSPP1, 30, 2 }, - { DBGBUS_SSPP1, 30, 3 }, - { DBGBUS_SSPP1, 30, 4 }, - { DBGBUS_SSPP1, 30, 5 }, - { DBGBUS_SSPP1, 30, 6 }, - { DBGBUS_SSPP1, 30, 7 }, - - { DBGBUS_SSPP1, 31, 0 }, - { DBGBUS_SSPP1, 31, 1 }, - { DBGBUS_SSPP1, 31, 2 }, - { DBGBUS_SSPP1, 31, 3 }, - { DBGBUS_SSPP1, 31, 4 }, - { DBGBUS_SSPP1, 31, 5 }, - { DBGBUS_SSPP1, 31, 6 }, - { DBGBUS_SSPP1, 31, 7 }, - - { DBGBUS_SSPP1, 32, 0 }, - { DBGBUS_SSPP1, 32, 1 }, - { DBGBUS_SSPP1, 32, 2 }, - { DBGBUS_SSPP1, 32, 3 }, - { DBGBUS_SSPP1, 32, 4 }, - { DBGBUS_SSPP1, 32, 5 }, - { DBGBUS_SSPP1, 32, 6 }, - { DBGBUS_SSPP1, 32, 7 }, - - { DBGBUS_SSPP1, 33, 0 }, - { DBGBUS_SSPP1, 33, 1 }, - { DBGBUS_SSPP1, 33, 2 }, - { DBGBUS_SSPP1, 33, 3 }, - { DBGBUS_SSPP1, 33, 4 }, - { DBGBUS_SSPP1, 33, 5 }, - { DBGBUS_SSPP1, 33, 6 }, - { DBGBUS_SSPP1, 33, 7 }, - - { DBGBUS_SSPP1, 34, 0 }, - { DBGBUS_SSPP1, 34, 1 }, - { DBGBUS_SSPP1, 34, 2 }, - { DBGBUS_SSPP1, 34, 3 }, - { DBGBUS_SSPP1, 34, 4 }, - { DBGBUS_SSPP1, 34, 5 }, - { DBGBUS_SSPP1, 34, 6 }, - { DBGBUS_SSPP1, 34, 7 }, - - { DBGBUS_SSPP1, 35, 0 }, - { DBGBUS_SSPP1, 35, 1 }, - { DBGBUS_SSPP1, 35, 2 }, - - /* dma 1 */ - { DBGBUS_SSPP1, 40, 0 }, - { DBGBUS_SSPP1, 40, 1 }, - { DBGBUS_SSPP1, 40, 2 }, - { DBGBUS_SSPP1, 40, 3 }, - { DBGBUS_SSPP1, 40, 4 }, - { DBGBUS_SSPP1, 40, 5 }, - { DBGBUS_SSPP1, 40, 6 }, - { DBGBUS_SSPP1, 40, 7 }, - - { DBGBUS_SSPP1, 41, 0 }, - { DBGBUS_SSPP1, 41, 1 }, - { DBGBUS_SSPP1, 41, 2 }, - { DBGBUS_SSPP1, 41, 3 }, - { DBGBUS_SSPP1, 41, 4 }, - { DBGBUS_SSPP1, 41, 5 }, - { DBGBUS_SSPP1, 41, 6 }, - { DBGBUS_SSPP1, 41, 7 }, - - { DBGBUS_SSPP1, 42, 0 }, - { DBGBUS_SSPP1, 42, 1 }, - { DBGBUS_SSPP1, 42, 2 }, - { DBGBUS_SSPP1, 42, 3 }, - { DBGBUS_SSPP1, 42, 4 }, - { DBGBUS_SSPP1, 42, 5 }, - { DBGBUS_SSPP1, 42, 6 }, - { DBGBUS_SSPP1, 42, 7 }, - - { DBGBUS_SSPP1, 44, 0 }, - { DBGBUS_SSPP1, 44, 1 }, - { DBGBUS_SSPP1, 44, 2 }, - { DBGBUS_SSPP1, 44, 3 }, - { DBGBUS_SSPP1, 44, 4 }, - { DBGBUS_SSPP1, 44, 5 }, - { DBGBUS_SSPP1, 44, 6 }, - { DBGBUS_SSPP1, 44, 7 }, - - { DBGBUS_SSPP1, 45, 0 }, - { DBGBUS_SSPP1, 45, 1 }, - { DBGBUS_SSPP1, 45, 2 }, - { DBGBUS_SSPP1, 45, 3 }, - { DBGBUS_SSPP1, 45, 4 }, - { DBGBUS_SSPP1, 45, 5 }, - { DBGBUS_SSPP1, 45, 6 }, - { DBGBUS_SSPP1, 45, 7 }, - - /* cursor 1 */ - { DBGBUS_SSPP1, 80, 0 }, - { DBGBUS_SSPP1, 80, 1 }, - { DBGBUS_SSPP1, 80, 2 }, - { DBGBUS_SSPP1, 80, 3 }, - { DBGBUS_SSPP1, 80, 4 }, - { DBGBUS_SSPP1, 80, 5 }, - { DBGBUS_SSPP1, 80, 6 }, - { DBGBUS_SSPP1, 80, 7 }, - - { DBGBUS_SSPP1, 81, 0 }, - { DBGBUS_SSPP1, 81, 1 }, - { DBGBUS_SSPP1, 81, 2 }, - { DBGBUS_SSPP1, 81, 3 }, - { DBGBUS_SSPP1, 81, 4 }, - { DBGBUS_SSPP1, 81, 5 }, - { DBGBUS_SSPP1, 81, 6 }, - { DBGBUS_SSPP1, 81, 7 }, - - { DBGBUS_SSPP1, 82, 0 }, - { DBGBUS_SSPP1, 82, 1 }, - { DBGBUS_SSPP1, 82, 2 }, - { DBGBUS_SSPP1, 82, 3 }, - { DBGBUS_SSPP1, 82, 4 }, - { DBGBUS_SSPP1, 82, 5 }, - { DBGBUS_SSPP1, 82, 6 }, - { DBGBUS_SSPP1, 82, 7 }, - - { DBGBUS_SSPP1, 83, 0 }, - { DBGBUS_SSPP1, 83, 1 }, - { DBGBUS_SSPP1, 83, 2 }, - { DBGBUS_SSPP1, 83, 3 }, - { DBGBUS_SSPP1, 83, 4 }, - { DBGBUS_SSPP1, 83, 5 }, - { DBGBUS_SSPP1, 83, 6 }, - { DBGBUS_SSPP1, 83, 7 }, - - { DBGBUS_SSPP1, 84, 0 }, - { DBGBUS_SSPP1, 84, 1 }, - { DBGBUS_SSPP1, 84, 2 }, - { DBGBUS_SSPP1, 84, 3 }, - { DBGBUS_SSPP1, 84, 4 }, - { DBGBUS_SSPP1, 84, 5 }, - { DBGBUS_SSPP1, 84, 6 }, - { DBGBUS_SSPP1, 84, 7 }, - - /* dspp */ - { DBGBUS_DSPP, 13, 0 }, - { DBGBUS_DSPP, 19, 0 }, - { DBGBUS_DSPP, 14, 0 }, - { DBGBUS_DSPP, 14, 1 }, - { DBGBUS_DSPP, 14, 3 }, - { DBGBUS_DSPP, 20, 0 }, - { DBGBUS_DSPP, 20, 1 }, - { DBGBUS_DSPP, 20, 3 }, - - /* ppb_0 */ - { DBGBUS_DSPP, 31, 0, _dpu_debug_bus_ppb0_dump }, - { DBGBUS_DSPP, 33, 0, _dpu_debug_bus_ppb0_dump }, - { DBGBUS_DSPP, 35, 0, _dpu_debug_bus_ppb0_dump }, - { DBGBUS_DSPP, 42, 0, _dpu_debug_bus_ppb0_dump }, - - /* ppb_1 */ - { DBGBUS_DSPP, 32, 0, _dpu_debug_bus_ppb1_dump }, - { DBGBUS_DSPP, 34, 0, _dpu_debug_bus_ppb1_dump }, - { DBGBUS_DSPP, 36, 0, _dpu_debug_bus_ppb1_dump }, - { DBGBUS_DSPP, 43, 0, _dpu_debug_bus_ppb1_dump }, - - /* lm_lut */ - { DBGBUS_DSPP, 109, 0 }, - { DBGBUS_DSPP, 105, 0 }, - { DBGBUS_DSPP, 103, 0 }, - - /* tear-check */ - { DBGBUS_PERIPH, 63, 0 }, - { DBGBUS_PERIPH, 64, 0 }, - { DBGBUS_PERIPH, 65, 0 }, - { DBGBUS_PERIPH, 73, 0 }, - { DBGBUS_PERIPH, 74, 0 }, - - /* crossbar */ - { DBGBUS_DSPP, 0, 0, _dpu_debug_bus_xbar_dump }, - - /* rotator */ - { DBGBUS_DSPP, 9, 0}, - - /* blend */ - /* LM0 */ - { DBGBUS_DSPP, 63, 0}, - { DBGBUS_DSPP, 63, 1}, - { DBGBUS_DSPP, 63, 2}, - { DBGBUS_DSPP, 63, 3}, - { DBGBUS_DSPP, 63, 4}, - { DBGBUS_DSPP, 63, 5}, - { DBGBUS_DSPP, 63, 6}, - { DBGBUS_DSPP, 63, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 64, 0}, - { DBGBUS_DSPP, 64, 1}, - { DBGBUS_DSPP, 64, 2}, - { DBGBUS_DSPP, 64, 3}, - { DBGBUS_DSPP, 64, 4}, - { DBGBUS_DSPP, 64, 5}, - { DBGBUS_DSPP, 64, 6}, - { DBGBUS_DSPP, 64, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 65, 0}, - { DBGBUS_DSPP, 65, 1}, - { DBGBUS_DSPP, 65, 2}, - { DBGBUS_DSPP, 65, 3}, - { DBGBUS_DSPP, 65, 4}, - { DBGBUS_DSPP, 65, 5}, - { DBGBUS_DSPP, 65, 6}, - { DBGBUS_DSPP, 65, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 66, 0}, - { DBGBUS_DSPP, 66, 1}, - { DBGBUS_DSPP, 66, 2}, - { DBGBUS_DSPP, 66, 3}, - { DBGBUS_DSPP, 66, 4}, - { DBGBUS_DSPP, 66, 5}, - { DBGBUS_DSPP, 66, 6}, - { DBGBUS_DSPP, 66, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 67, 0}, - { DBGBUS_DSPP, 67, 1}, - { DBGBUS_DSPP, 67, 2}, - { DBGBUS_DSPP, 67, 3}, - { DBGBUS_DSPP, 67, 4}, - { DBGBUS_DSPP, 67, 5}, - { DBGBUS_DSPP, 67, 6}, - { DBGBUS_DSPP, 67, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 68, 0}, - { DBGBUS_DSPP, 68, 1}, - { DBGBUS_DSPP, 68, 2}, - { DBGBUS_DSPP, 68, 3}, - { DBGBUS_DSPP, 68, 4}, - { DBGBUS_DSPP, 68, 5}, - { DBGBUS_DSPP, 68, 6}, - { DBGBUS_DSPP, 68, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 69, 0}, - { DBGBUS_DSPP, 69, 1}, - { DBGBUS_DSPP, 69, 2}, - { DBGBUS_DSPP, 69, 3}, - { DBGBUS_DSPP, 69, 4}, - { DBGBUS_DSPP, 69, 5}, - { DBGBUS_DSPP, 69, 6}, - { DBGBUS_DSPP, 69, 7, _dpu_debug_bus_lm_dump }, - - /* LM1 */ - { DBGBUS_DSPP, 70, 0}, - { DBGBUS_DSPP, 70, 1}, - { DBGBUS_DSPP, 70, 2}, - { DBGBUS_DSPP, 70, 3}, - { DBGBUS_DSPP, 70, 4}, - { DBGBUS_DSPP, 70, 5}, - { DBGBUS_DSPP, 70, 6}, - { DBGBUS_DSPP, 70, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 71, 0}, - { DBGBUS_DSPP, 71, 1}, - { DBGBUS_DSPP, 71, 2}, - { DBGBUS_DSPP, 71, 3}, - { DBGBUS_DSPP, 71, 4}, - { DBGBUS_DSPP, 71, 5}, - { DBGBUS_DSPP, 71, 6}, - { DBGBUS_DSPP, 71, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 72, 0}, - { DBGBUS_DSPP, 72, 1}, - { DBGBUS_DSPP, 72, 2}, - { DBGBUS_DSPP, 72, 3}, - { DBGBUS_DSPP, 72, 4}, - { DBGBUS_DSPP, 72, 5}, - { DBGBUS_DSPP, 72, 6}, - { DBGBUS_DSPP, 72, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 73, 0}, - { DBGBUS_DSPP, 73, 1}, - { DBGBUS_DSPP, 73, 2}, - { DBGBUS_DSPP, 73, 3}, - { DBGBUS_DSPP, 73, 4}, - { DBGBUS_DSPP, 73, 5}, - { DBGBUS_DSPP, 73, 6}, - { DBGBUS_DSPP, 73, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 74, 0}, - { DBGBUS_DSPP, 74, 1}, - { DBGBUS_DSPP, 74, 2}, - { DBGBUS_DSPP, 74, 3}, - { DBGBUS_DSPP, 74, 4}, - { DBGBUS_DSPP, 74, 5}, - { DBGBUS_DSPP, 74, 6}, - { DBGBUS_DSPP, 74, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 75, 0}, - { DBGBUS_DSPP, 75, 1}, - { DBGBUS_DSPP, 75, 2}, - { DBGBUS_DSPP, 75, 3}, - { DBGBUS_DSPP, 75, 4}, - { DBGBUS_DSPP, 75, 5}, - { DBGBUS_DSPP, 75, 6}, - { DBGBUS_DSPP, 75, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 76, 0}, - { DBGBUS_DSPP, 76, 1}, - { DBGBUS_DSPP, 76, 2}, - { DBGBUS_DSPP, 76, 3}, - { DBGBUS_DSPP, 76, 4}, - { DBGBUS_DSPP, 76, 5}, - { DBGBUS_DSPP, 76, 6}, - { DBGBUS_DSPP, 76, 7, _dpu_debug_bus_lm_dump }, - - /* LM2 */ - { DBGBUS_DSPP, 77, 0}, - { DBGBUS_DSPP, 77, 1}, - { DBGBUS_DSPP, 77, 2}, - { DBGBUS_DSPP, 77, 3}, - { DBGBUS_DSPP, 77, 4}, - { DBGBUS_DSPP, 77, 5}, - { DBGBUS_DSPP, 77, 6}, - { DBGBUS_DSPP, 77, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 78, 0}, - { DBGBUS_DSPP, 78, 1}, - { DBGBUS_DSPP, 78, 2}, - { DBGBUS_DSPP, 78, 3}, - { DBGBUS_DSPP, 78, 4}, - { DBGBUS_DSPP, 78, 5}, - { DBGBUS_DSPP, 78, 6}, - { DBGBUS_DSPP, 78, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 79, 0}, - { DBGBUS_DSPP, 79, 1}, - { DBGBUS_DSPP, 79, 2}, - { DBGBUS_DSPP, 79, 3}, - { DBGBUS_DSPP, 79, 4}, - { DBGBUS_DSPP, 79, 5}, - { DBGBUS_DSPP, 79, 6}, - { DBGBUS_DSPP, 79, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 80, 0}, - { DBGBUS_DSPP, 80, 1}, - { DBGBUS_DSPP, 80, 2}, - { DBGBUS_DSPP, 80, 3}, - { DBGBUS_DSPP, 80, 4}, - { DBGBUS_DSPP, 80, 5}, - { DBGBUS_DSPP, 80, 6}, - { DBGBUS_DSPP, 80, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 81, 0}, - { DBGBUS_DSPP, 81, 1}, - { DBGBUS_DSPP, 81, 2}, - { DBGBUS_DSPP, 81, 3}, - { DBGBUS_DSPP, 81, 4}, - { DBGBUS_DSPP, 81, 5}, - { DBGBUS_DSPP, 81, 6}, - { DBGBUS_DSPP, 81, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 82, 0}, - { DBGBUS_DSPP, 82, 1}, - { DBGBUS_DSPP, 82, 2}, - { DBGBUS_DSPP, 82, 3}, - { DBGBUS_DSPP, 82, 4}, - { DBGBUS_DSPP, 82, 5}, - { DBGBUS_DSPP, 82, 6}, - { DBGBUS_DSPP, 82, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 83, 0}, - { DBGBUS_DSPP, 83, 1}, - { DBGBUS_DSPP, 83, 2}, - { DBGBUS_DSPP, 83, 3}, - { DBGBUS_DSPP, 83, 4}, - { DBGBUS_DSPP, 83, 5}, - { DBGBUS_DSPP, 83, 6}, - { DBGBUS_DSPP, 83, 7, _dpu_debug_bus_lm_dump }, - - /* csc */ - { DBGBUS_SSPP0, 7, 0}, - { DBGBUS_SSPP0, 7, 1}, - { DBGBUS_SSPP0, 27, 0}, - { DBGBUS_SSPP0, 27, 1}, - { DBGBUS_SSPP1, 7, 0}, - { DBGBUS_SSPP1, 7, 1}, - { DBGBUS_SSPP1, 27, 0}, - { DBGBUS_SSPP1, 27, 1}, - - /* pcc */ - { DBGBUS_SSPP0, 3, 3}, - { DBGBUS_SSPP0, 23, 3}, - { DBGBUS_SSPP0, 33, 3}, - { DBGBUS_SSPP0, 43, 3}, - { DBGBUS_SSPP1, 3, 3}, - { DBGBUS_SSPP1, 23, 3}, - { DBGBUS_SSPP1, 33, 3}, - { DBGBUS_SSPP1, 43, 3}, - - /* spa */ - { DBGBUS_SSPP0, 8, 0}, - { DBGBUS_SSPP0, 28, 0}, - { DBGBUS_SSPP1, 8, 0}, - { DBGBUS_SSPP1, 28, 0}, - { DBGBUS_DSPP, 13, 0}, - { DBGBUS_DSPP, 19, 0}, - - /* igc */ - { DBGBUS_SSPP0, 9, 0}, - { DBGBUS_SSPP0, 9, 1}, - { DBGBUS_SSPP0, 9, 3}, - { DBGBUS_SSPP0, 29, 0}, - { DBGBUS_SSPP0, 29, 1}, - { DBGBUS_SSPP0, 29, 3}, - { DBGBUS_SSPP0, 17, 0}, - { DBGBUS_SSPP0, 17, 1}, - { DBGBUS_SSPP0, 17, 3}, - { DBGBUS_SSPP0, 37, 0}, - { DBGBUS_SSPP0, 37, 1}, - { DBGBUS_SSPP0, 37, 3}, - { DBGBUS_SSPP0, 46, 0}, - { DBGBUS_SSPP0, 46, 1}, - { DBGBUS_SSPP0, 46, 3}, - - { DBGBUS_SSPP1, 9, 0}, - { DBGBUS_SSPP1, 9, 1}, - { DBGBUS_SSPP1, 9, 3}, - { DBGBUS_SSPP1, 29, 0}, - { DBGBUS_SSPP1, 29, 1}, - { DBGBUS_SSPP1, 29, 3}, - { DBGBUS_SSPP1, 17, 0}, - { DBGBUS_SSPP1, 17, 1}, - { DBGBUS_SSPP1, 17, 3}, - { DBGBUS_SSPP1, 37, 0}, - { DBGBUS_SSPP1, 37, 1}, - { DBGBUS_SSPP1, 37, 3}, - { DBGBUS_SSPP1, 46, 0}, - { DBGBUS_SSPP1, 46, 1}, - { DBGBUS_SSPP1, 46, 3}, - - { DBGBUS_DSPP, 14, 0}, - { DBGBUS_DSPP, 14, 1}, - { DBGBUS_DSPP, 14, 3}, - { DBGBUS_DSPP, 20, 0}, - { DBGBUS_DSPP, 20, 1}, - { DBGBUS_DSPP, 20, 3}, - - { DBGBUS_PERIPH, 60, 0}, -}; - -static struct dpu_debug_bus_entry dbg_bus_dpu_sdm845[] = { - - /* Unpack 0 sspp 0*/ - { DBGBUS_SSPP0, 50, 2 }, - { DBGBUS_SSPP0, 60, 2 }, - { DBGBUS_SSPP0, 70, 2 }, - - /* Upack 0 sspp 1*/ - { DBGBUS_SSPP1, 50, 2 }, - { DBGBUS_SSPP1, 60, 2 }, - { DBGBUS_SSPP1, 70, 2 }, - - /* scheduler */ - { DBGBUS_DSPP, 130, 0 }, - { DBGBUS_DSPP, 130, 1 }, - { DBGBUS_DSPP, 130, 2 }, - { DBGBUS_DSPP, 130, 3 }, - { DBGBUS_DSPP, 130, 4 }, - { DBGBUS_DSPP, 130, 5 }, - - /* qseed */ - { DBGBUS_SSPP0, 6, 0}, - { DBGBUS_SSPP0, 6, 1}, - { DBGBUS_SSPP0, 26, 0}, - { DBGBUS_SSPP0, 26, 1}, - { DBGBUS_SSPP1, 6, 0}, - { DBGBUS_SSPP1, 6, 1}, - { DBGBUS_SSPP1, 26, 0}, - { DBGBUS_SSPP1, 26, 1}, - - /* scale */ - { DBGBUS_SSPP0, 16, 0}, - { DBGBUS_SSPP0, 16, 1}, - { DBGBUS_SSPP0, 36, 0}, - { DBGBUS_SSPP0, 36, 1}, - { DBGBUS_SSPP1, 16, 0}, - { DBGBUS_SSPP1, 16, 1}, - { DBGBUS_SSPP1, 36, 0}, - { DBGBUS_SSPP1, 36, 1}, - - /* fetch sspp0 */ - - /* vig 0 */ - { DBGBUS_SSPP0, 0, 0 }, - { DBGBUS_SSPP0, 0, 1 }, - { DBGBUS_SSPP0, 0, 2 }, - { DBGBUS_SSPP0, 0, 3 }, - { DBGBUS_SSPP0, 0, 4 }, - { DBGBUS_SSPP0, 0, 5 }, - { DBGBUS_SSPP0, 0, 6 }, - { DBGBUS_SSPP0, 0, 7 }, - - { DBGBUS_SSPP0, 1, 0 }, - { DBGBUS_SSPP0, 1, 1 }, - { DBGBUS_SSPP0, 1, 2 }, - { DBGBUS_SSPP0, 1, 3 }, - { DBGBUS_SSPP0, 1, 4 }, - { DBGBUS_SSPP0, 1, 5 }, - { DBGBUS_SSPP0, 1, 6 }, - { DBGBUS_SSPP0, 1, 7 }, - - { DBGBUS_SSPP0, 2, 0 }, - { DBGBUS_SSPP0, 2, 1 }, - { DBGBUS_SSPP0, 2, 2 }, - { DBGBUS_SSPP0, 2, 3 }, - { DBGBUS_SSPP0, 2, 4 }, - { DBGBUS_SSPP0, 2, 5 }, - { DBGBUS_SSPP0, 2, 6 }, - { DBGBUS_SSPP0, 2, 7 }, - - { DBGBUS_SSPP0, 4, 0 }, - { DBGBUS_SSPP0, 4, 1 }, - { DBGBUS_SSPP0, 4, 2 }, - { DBGBUS_SSPP0, 4, 3 }, - { DBGBUS_SSPP0, 4, 4 }, - { DBGBUS_SSPP0, 4, 5 }, - { DBGBUS_SSPP0, 4, 6 }, - { DBGBUS_SSPP0, 4, 7 }, - - { DBGBUS_SSPP0, 5, 0 }, - { DBGBUS_SSPP0, 5, 1 }, - { DBGBUS_SSPP0, 5, 2 }, - { DBGBUS_SSPP0, 5, 3 }, - { DBGBUS_SSPP0, 5, 4 }, - { DBGBUS_SSPP0, 5, 5 }, - { DBGBUS_SSPP0, 5, 6 }, - { DBGBUS_SSPP0, 5, 7 }, - - /* vig 2 */ - { DBGBUS_SSPP0, 20, 0 }, - { DBGBUS_SSPP0, 20, 1 }, - { DBGBUS_SSPP0, 20, 2 }, - { DBGBUS_SSPP0, 20, 3 }, - { DBGBUS_SSPP0, 20, 4 }, - { DBGBUS_SSPP0, 20, 5 }, - { DBGBUS_SSPP0, 20, 6 }, - { DBGBUS_SSPP0, 20, 7 }, - - { DBGBUS_SSPP0, 21, 0 }, - { DBGBUS_SSPP0, 21, 1 }, - { DBGBUS_SSPP0, 21, 2 }, - { DBGBUS_SSPP0, 21, 3 }, - { DBGBUS_SSPP0, 21, 4 }, - { DBGBUS_SSPP0, 21, 5 }, - { DBGBUS_SSPP0, 21, 6 }, - { DBGBUS_SSPP0, 21, 7 }, - - { DBGBUS_SSPP0, 22, 0 }, - { DBGBUS_SSPP0, 22, 1 }, - { DBGBUS_SSPP0, 22, 2 }, - { DBGBUS_SSPP0, 22, 3 }, - { DBGBUS_SSPP0, 22, 4 }, - { DBGBUS_SSPP0, 22, 5 }, - { DBGBUS_SSPP0, 22, 6 }, - { DBGBUS_SSPP0, 22, 7 }, - - { DBGBUS_SSPP0, 24, 0 }, - { DBGBUS_SSPP0, 24, 1 }, - { DBGBUS_SSPP0, 24, 2 }, - { DBGBUS_SSPP0, 24, 3 }, - { DBGBUS_SSPP0, 24, 4 }, - { DBGBUS_SSPP0, 24, 5 }, - { DBGBUS_SSPP0, 24, 6 }, - { DBGBUS_SSPP0, 24, 7 }, - - { DBGBUS_SSPP0, 25, 0 }, - { DBGBUS_SSPP0, 25, 1 }, - { DBGBUS_SSPP0, 25, 2 }, - { DBGBUS_SSPP0, 25, 3 }, - { DBGBUS_SSPP0, 25, 4 }, - { DBGBUS_SSPP0, 25, 5 }, - { DBGBUS_SSPP0, 25, 6 }, - { DBGBUS_SSPP0, 25, 7 }, - - /* dma 2 */ - { DBGBUS_SSPP0, 30, 0 }, - { DBGBUS_SSPP0, 30, 1 }, - { DBGBUS_SSPP0, 30, 2 }, - { DBGBUS_SSPP0, 30, 3 }, - { DBGBUS_SSPP0, 30, 4 }, - { DBGBUS_SSPP0, 30, 5 }, - { DBGBUS_SSPP0, 30, 6 }, - { DBGBUS_SSPP0, 30, 7 }, - - { DBGBUS_SSPP0, 31, 0 }, - { DBGBUS_SSPP0, 31, 1 }, - { DBGBUS_SSPP0, 31, 2 }, - { DBGBUS_SSPP0, 31, 3 }, - { DBGBUS_SSPP0, 31, 4 }, - { DBGBUS_SSPP0, 31, 5 }, - { DBGBUS_SSPP0, 31, 6 }, - { DBGBUS_SSPP0, 31, 7 }, - - { DBGBUS_SSPP0, 32, 0 }, - { DBGBUS_SSPP0, 32, 1 }, - { DBGBUS_SSPP0, 32, 2 }, - { DBGBUS_SSPP0, 32, 3 }, - { DBGBUS_SSPP0, 32, 4 }, - { DBGBUS_SSPP0, 32, 5 }, - { DBGBUS_SSPP0, 32, 6 }, - { DBGBUS_SSPP0, 32, 7 }, - - { DBGBUS_SSPP0, 33, 0 }, - { DBGBUS_SSPP0, 33, 1 }, - { DBGBUS_SSPP0, 33, 2 }, - { DBGBUS_SSPP0, 33, 3 }, - { DBGBUS_SSPP0, 33, 4 }, - { DBGBUS_SSPP0, 33, 5 }, - { DBGBUS_SSPP0, 33, 6 }, - { DBGBUS_SSPP0, 33, 7 }, - - { DBGBUS_SSPP0, 34, 0 }, - { DBGBUS_SSPP0, 34, 1 }, - { DBGBUS_SSPP0, 34, 2 }, - { DBGBUS_SSPP0, 34, 3 }, - { DBGBUS_SSPP0, 34, 4 }, - { DBGBUS_SSPP0, 34, 5 }, - { DBGBUS_SSPP0, 34, 6 }, - { DBGBUS_SSPP0, 34, 7 }, - - { DBGBUS_SSPP0, 35, 0 }, - { DBGBUS_SSPP0, 35, 1 }, - { DBGBUS_SSPP0, 35, 2 }, - { DBGBUS_SSPP0, 35, 3 }, - - /* dma 0 */ - { DBGBUS_SSPP0, 40, 0 }, - { DBGBUS_SSPP0, 40, 1 }, - { DBGBUS_SSPP0, 40, 2 }, - { DBGBUS_SSPP0, 40, 3 }, - { DBGBUS_SSPP0, 40, 4 }, - { DBGBUS_SSPP0, 40, 5 }, - { DBGBUS_SSPP0, 40, 6 }, - { DBGBUS_SSPP0, 40, 7 }, - - { DBGBUS_SSPP0, 41, 0 }, - { DBGBUS_SSPP0, 41, 1 }, - { DBGBUS_SSPP0, 41, 2 }, - { DBGBUS_SSPP0, 41, 3 }, - { DBGBUS_SSPP0, 41, 4 }, - { DBGBUS_SSPP0, 41, 5 }, - { DBGBUS_SSPP0, 41, 6 }, - { DBGBUS_SSPP0, 41, 7 }, - - { DBGBUS_SSPP0, 42, 0 }, - { DBGBUS_SSPP0, 42, 1 }, - { DBGBUS_SSPP0, 42, 2 }, - { DBGBUS_SSPP0, 42, 3 }, - { DBGBUS_SSPP0, 42, 4 }, - { DBGBUS_SSPP0, 42, 5 }, - { DBGBUS_SSPP0, 42, 6 }, - { DBGBUS_SSPP0, 42, 7 }, - - { DBGBUS_SSPP0, 44, 0 }, - { DBGBUS_SSPP0, 44, 1 }, - { DBGBUS_SSPP0, 44, 2 }, - { DBGBUS_SSPP0, 44, 3 }, - { DBGBUS_SSPP0, 44, 4 }, - { DBGBUS_SSPP0, 44, 5 }, - { DBGBUS_SSPP0, 44, 6 }, - { DBGBUS_SSPP0, 44, 7 }, - - { DBGBUS_SSPP0, 45, 0 }, - { DBGBUS_SSPP0, 45, 1 }, - { DBGBUS_SSPP0, 45, 2 }, - { DBGBUS_SSPP0, 45, 3 }, - { DBGBUS_SSPP0, 45, 4 }, - { DBGBUS_SSPP0, 45, 5 }, - { DBGBUS_SSPP0, 45, 6 }, - { DBGBUS_SSPP0, 45, 7 }, - - /* fetch sspp1 */ - /* vig 1 */ - { DBGBUS_SSPP1, 0, 0 }, - { DBGBUS_SSPP1, 0, 1 }, - { DBGBUS_SSPP1, 0, 2 }, - { DBGBUS_SSPP1, 0, 3 }, - { DBGBUS_SSPP1, 0, 4 }, - { DBGBUS_SSPP1, 0, 5 }, - { DBGBUS_SSPP1, 0, 6 }, - { DBGBUS_SSPP1, 0, 7 }, - - { DBGBUS_SSPP1, 1, 0 }, - { DBGBUS_SSPP1, 1, 1 }, - { DBGBUS_SSPP1, 1, 2 }, - { DBGBUS_SSPP1, 1, 3 }, - { DBGBUS_SSPP1, 1, 4 }, - { DBGBUS_SSPP1, 1, 5 }, - { DBGBUS_SSPP1, 1, 6 }, - { DBGBUS_SSPP1, 1, 7 }, - - { DBGBUS_SSPP1, 2, 0 }, - { DBGBUS_SSPP1, 2, 1 }, - { DBGBUS_SSPP1, 2, 2 }, - { DBGBUS_SSPP1, 2, 3 }, - { DBGBUS_SSPP1, 2, 4 }, - { DBGBUS_SSPP1, 2, 5 }, - { DBGBUS_SSPP1, 2, 6 }, - { DBGBUS_SSPP1, 2, 7 }, - - { DBGBUS_SSPP1, 4, 0 }, - { DBGBUS_SSPP1, 4, 1 }, - { DBGBUS_SSPP1, 4, 2 }, - { DBGBUS_SSPP1, 4, 3 }, - { DBGBUS_SSPP1, 4, 4 }, - { DBGBUS_SSPP1, 4, 5 }, - { DBGBUS_SSPP1, 4, 6 }, - { DBGBUS_SSPP1, 4, 7 }, - - { DBGBUS_SSPP1, 5, 0 }, - { DBGBUS_SSPP1, 5, 1 }, - { DBGBUS_SSPP1, 5, 2 }, - { DBGBUS_SSPP1, 5, 3 }, - { DBGBUS_SSPP1, 5, 4 }, - { DBGBUS_SSPP1, 5, 5 }, - { DBGBUS_SSPP1, 5, 6 }, - { DBGBUS_SSPP1, 5, 7 }, - - /* vig 3 */ - { DBGBUS_SSPP1, 20, 0 }, - { DBGBUS_SSPP1, 20, 1 }, - { DBGBUS_SSPP1, 20, 2 }, - { DBGBUS_SSPP1, 20, 3 }, - { DBGBUS_SSPP1, 20, 4 }, - { DBGBUS_SSPP1, 20, 5 }, - { DBGBUS_SSPP1, 20, 6 }, - { DBGBUS_SSPP1, 20, 7 }, - - { DBGBUS_SSPP1, 21, 0 }, - { DBGBUS_SSPP1, 21, 1 }, - { DBGBUS_SSPP1, 21, 2 }, - { DBGBUS_SSPP1, 21, 3 }, - { DBGBUS_SSPP1, 21, 4 }, - { DBGBUS_SSPP1, 21, 5 }, - { DBGBUS_SSPP1, 21, 6 }, - { DBGBUS_SSPP1, 21, 7 }, - - { DBGBUS_SSPP1, 22, 0 }, - { DBGBUS_SSPP1, 22, 1 }, - { DBGBUS_SSPP1, 22, 2 }, - { DBGBUS_SSPP1, 22, 3 }, - { DBGBUS_SSPP1, 22, 4 }, - { DBGBUS_SSPP1, 22, 5 }, - { DBGBUS_SSPP1, 22, 6 }, - { DBGBUS_SSPP1, 22, 7 }, - - { DBGBUS_SSPP1, 24, 0 }, - { DBGBUS_SSPP1, 24, 1 }, - { DBGBUS_SSPP1, 24, 2 }, - { DBGBUS_SSPP1, 24, 3 }, - { DBGBUS_SSPP1, 24, 4 }, - { DBGBUS_SSPP1, 24, 5 }, - { DBGBUS_SSPP1, 24, 6 }, - { DBGBUS_SSPP1, 24, 7 }, - - { DBGBUS_SSPP1, 25, 0 }, - { DBGBUS_SSPP1, 25, 1 }, - { DBGBUS_SSPP1, 25, 2 }, - { DBGBUS_SSPP1, 25, 3 }, - { DBGBUS_SSPP1, 25, 4 }, - { DBGBUS_SSPP1, 25, 5 }, - { DBGBUS_SSPP1, 25, 6 }, - { DBGBUS_SSPP1, 25, 7 }, - - /* dma 3 */ - { DBGBUS_SSPP1, 30, 0 }, - { DBGBUS_SSPP1, 30, 1 }, - { DBGBUS_SSPP1, 30, 2 }, - { DBGBUS_SSPP1, 30, 3 }, - { DBGBUS_SSPP1, 30, 4 }, - { DBGBUS_SSPP1, 30, 5 }, - { DBGBUS_SSPP1, 30, 6 }, - { DBGBUS_SSPP1, 30, 7 }, - - { DBGBUS_SSPP1, 31, 0 }, - { DBGBUS_SSPP1, 31, 1 }, - { DBGBUS_SSPP1, 31, 2 }, - { DBGBUS_SSPP1, 31, 3 }, - { DBGBUS_SSPP1, 31, 4 }, - { DBGBUS_SSPP1, 31, 5 }, - { DBGBUS_SSPP1, 31, 6 }, - { DBGBUS_SSPP1, 31, 7 }, - - { DBGBUS_SSPP1, 32, 0 }, - { DBGBUS_SSPP1, 32, 1 }, - { DBGBUS_SSPP1, 32, 2 }, - { DBGBUS_SSPP1, 32, 3 }, - { DBGBUS_SSPP1, 32, 4 }, - { DBGBUS_SSPP1, 32, 5 }, - { DBGBUS_SSPP1, 32, 6 }, - { DBGBUS_SSPP1, 32, 7 }, - - { DBGBUS_SSPP1, 33, 0 }, - { DBGBUS_SSPP1, 33, 1 }, - { DBGBUS_SSPP1, 33, 2 }, - { DBGBUS_SSPP1, 33, 3 }, - { DBGBUS_SSPP1, 33, 4 }, - { DBGBUS_SSPP1, 33, 5 }, - { DBGBUS_SSPP1, 33, 6 }, - { DBGBUS_SSPP1, 33, 7 }, - - { DBGBUS_SSPP1, 34, 0 }, - { DBGBUS_SSPP1, 34, 1 }, - { DBGBUS_SSPP1, 34, 2 }, - { DBGBUS_SSPP1, 34, 3 }, - { DBGBUS_SSPP1, 34, 4 }, - { DBGBUS_SSPP1, 34, 5 }, - { DBGBUS_SSPP1, 34, 6 }, - { DBGBUS_SSPP1, 34, 7 }, - - { DBGBUS_SSPP1, 35, 0 }, - { DBGBUS_SSPP1, 35, 1 }, - { DBGBUS_SSPP1, 35, 2 }, - - /* dma 1 */ - { DBGBUS_SSPP1, 40, 0 }, - { DBGBUS_SSPP1, 40, 1 }, - { DBGBUS_SSPP1, 40, 2 }, - { DBGBUS_SSPP1, 40, 3 }, - { DBGBUS_SSPP1, 40, 4 }, - { DBGBUS_SSPP1, 40, 5 }, - { DBGBUS_SSPP1, 40, 6 }, - { DBGBUS_SSPP1, 40, 7 }, - - { DBGBUS_SSPP1, 41, 0 }, - { DBGBUS_SSPP1, 41, 1 }, - { DBGBUS_SSPP1, 41, 2 }, - { DBGBUS_SSPP1, 41, 3 }, - { DBGBUS_SSPP1, 41, 4 }, - { DBGBUS_SSPP1, 41, 5 }, - { DBGBUS_SSPP1, 41, 6 }, - { DBGBUS_SSPP1, 41, 7 }, - - { DBGBUS_SSPP1, 42, 0 }, - { DBGBUS_SSPP1, 42, 1 }, - { DBGBUS_SSPP1, 42, 2 }, - { DBGBUS_SSPP1, 42, 3 }, - { DBGBUS_SSPP1, 42, 4 }, - { DBGBUS_SSPP1, 42, 5 }, - { DBGBUS_SSPP1, 42, 6 }, - { DBGBUS_SSPP1, 42, 7 }, - - { DBGBUS_SSPP1, 44, 0 }, - { DBGBUS_SSPP1, 44, 1 }, - { DBGBUS_SSPP1, 44, 2 }, - { DBGBUS_SSPP1, 44, 3 }, - { DBGBUS_SSPP1, 44, 4 }, - { DBGBUS_SSPP1, 44, 5 }, - { DBGBUS_SSPP1, 44, 6 }, - { DBGBUS_SSPP1, 44, 7 }, - - { DBGBUS_SSPP1, 45, 0 }, - { DBGBUS_SSPP1, 45, 1 }, - { DBGBUS_SSPP1, 45, 2 }, - { DBGBUS_SSPP1, 45, 3 }, - { DBGBUS_SSPP1, 45, 4 }, - { DBGBUS_SSPP1, 45, 5 }, - { DBGBUS_SSPP1, 45, 6 }, - { DBGBUS_SSPP1, 45, 7 }, - - /* dspp */ - { DBGBUS_DSPP, 13, 0 }, - { DBGBUS_DSPP, 19, 0 }, - { DBGBUS_DSPP, 14, 0 }, - { DBGBUS_DSPP, 14, 1 }, - { DBGBUS_DSPP, 14, 3 }, - { DBGBUS_DSPP, 20, 0 }, - { DBGBUS_DSPP, 20, 1 }, - { DBGBUS_DSPP, 20, 3 }, - - /* ppb_0 */ - { DBGBUS_DSPP, 31, 0, _dpu_debug_bus_ppb0_dump }, - { DBGBUS_DSPP, 33, 0, _dpu_debug_bus_ppb0_dump }, - { DBGBUS_DSPP, 35, 0, _dpu_debug_bus_ppb0_dump }, - { DBGBUS_DSPP, 42, 0, _dpu_debug_bus_ppb0_dump }, - - /* ppb_1 */ - { DBGBUS_DSPP, 32, 0, _dpu_debug_bus_ppb1_dump }, - { DBGBUS_DSPP, 34, 0, _dpu_debug_bus_ppb1_dump }, - { DBGBUS_DSPP, 36, 0, _dpu_debug_bus_ppb1_dump }, - { DBGBUS_DSPP, 43, 0, _dpu_debug_bus_ppb1_dump }, - - /* lm_lut */ - { DBGBUS_DSPP, 109, 0 }, - { DBGBUS_DSPP, 105, 0 }, - { DBGBUS_DSPP, 103, 0 }, - - /* crossbar */ - { DBGBUS_DSPP, 0, 0, _dpu_debug_bus_xbar_dump }, - - /* rotator */ - { DBGBUS_DSPP, 9, 0}, - - /* blend */ - /* LM0 */ - { DBGBUS_DSPP, 63, 1}, - { DBGBUS_DSPP, 63, 2}, - { DBGBUS_DSPP, 63, 3}, - { DBGBUS_DSPP, 63, 4}, - { DBGBUS_DSPP, 63, 5}, - { DBGBUS_DSPP, 63, 6}, - { DBGBUS_DSPP, 63, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 64, 1}, - { DBGBUS_DSPP, 64, 2}, - { DBGBUS_DSPP, 64, 3}, - { DBGBUS_DSPP, 64, 4}, - { DBGBUS_DSPP, 64, 5}, - { DBGBUS_DSPP, 64, 6}, - { DBGBUS_DSPP, 64, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 65, 1}, - { DBGBUS_DSPP, 65, 2}, - { DBGBUS_DSPP, 65, 3}, - { DBGBUS_DSPP, 65, 4}, - { DBGBUS_DSPP, 65, 5}, - { DBGBUS_DSPP, 65, 6}, - { DBGBUS_DSPP, 65, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 66, 1}, - { DBGBUS_DSPP, 66, 2}, - { DBGBUS_DSPP, 66, 3}, - { DBGBUS_DSPP, 66, 4}, - { DBGBUS_DSPP, 66, 5}, - { DBGBUS_DSPP, 66, 6}, - { DBGBUS_DSPP, 66, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 67, 1}, - { DBGBUS_DSPP, 67, 2}, - { DBGBUS_DSPP, 67, 3}, - { DBGBUS_DSPP, 67, 4}, - { DBGBUS_DSPP, 67, 5}, - { DBGBUS_DSPP, 67, 6}, - { DBGBUS_DSPP, 67, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 68, 1}, - { DBGBUS_DSPP, 68, 2}, - { DBGBUS_DSPP, 68, 3}, - { DBGBUS_DSPP, 68, 4}, - { DBGBUS_DSPP, 68, 5}, - { DBGBUS_DSPP, 68, 6}, - { DBGBUS_DSPP, 68, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 69, 1}, - { DBGBUS_DSPP, 69, 2}, - { DBGBUS_DSPP, 69, 3}, - { DBGBUS_DSPP, 69, 4}, - { DBGBUS_DSPP, 69, 5}, - { DBGBUS_DSPP, 69, 6}, - { DBGBUS_DSPP, 69, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 84, 1}, - { DBGBUS_DSPP, 84, 2}, - { DBGBUS_DSPP, 84, 3}, - { DBGBUS_DSPP, 84, 4}, - { DBGBUS_DSPP, 84, 5}, - { DBGBUS_DSPP, 84, 6}, - { DBGBUS_DSPP, 84, 7, _dpu_debug_bus_lm_dump }, - - - { DBGBUS_DSPP, 85, 1}, - { DBGBUS_DSPP, 85, 2}, - { DBGBUS_DSPP, 85, 3}, - { DBGBUS_DSPP, 85, 4}, - { DBGBUS_DSPP, 85, 5}, - { DBGBUS_DSPP, 85, 6}, - { DBGBUS_DSPP, 85, 7, _dpu_debug_bus_lm_dump }, - - - { DBGBUS_DSPP, 86, 1}, - { DBGBUS_DSPP, 86, 2}, - { DBGBUS_DSPP, 86, 3}, - { DBGBUS_DSPP, 86, 4}, - { DBGBUS_DSPP, 86, 5}, - { DBGBUS_DSPP, 86, 6}, - { DBGBUS_DSPP, 86, 7, _dpu_debug_bus_lm_dump }, - - - { DBGBUS_DSPP, 87, 1}, - { DBGBUS_DSPP, 87, 2}, - { DBGBUS_DSPP, 87, 3}, - { DBGBUS_DSPP, 87, 4}, - { DBGBUS_DSPP, 87, 5}, - { DBGBUS_DSPP, 87, 6}, - { DBGBUS_DSPP, 87, 7, _dpu_debug_bus_lm_dump }, - - /* LM1 */ - { DBGBUS_DSPP, 70, 1}, - { DBGBUS_DSPP, 70, 2}, - { DBGBUS_DSPP, 70, 3}, - { DBGBUS_DSPP, 70, 4}, - { DBGBUS_DSPP, 70, 5}, - { DBGBUS_DSPP, 70, 6}, - { DBGBUS_DSPP, 70, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 71, 1}, - { DBGBUS_DSPP, 71, 2}, - { DBGBUS_DSPP, 71, 3}, - { DBGBUS_DSPP, 71, 4}, - { DBGBUS_DSPP, 71, 5}, - { DBGBUS_DSPP, 71, 6}, - { DBGBUS_DSPP, 71, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 72, 1}, - { DBGBUS_DSPP, 72, 2}, - { DBGBUS_DSPP, 72, 3}, - { DBGBUS_DSPP, 72, 4}, - { DBGBUS_DSPP, 72, 5}, - { DBGBUS_DSPP, 72, 6}, - { DBGBUS_DSPP, 72, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 73, 1}, - { DBGBUS_DSPP, 73, 2}, - { DBGBUS_DSPP, 73, 3}, - { DBGBUS_DSPP, 73, 4}, - { DBGBUS_DSPP, 73, 5}, - { DBGBUS_DSPP, 73, 6}, - { DBGBUS_DSPP, 73, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 74, 1}, - { DBGBUS_DSPP, 74, 2}, - { DBGBUS_DSPP, 74, 3}, - { DBGBUS_DSPP, 74, 4}, - { DBGBUS_DSPP, 74, 5}, - { DBGBUS_DSPP, 74, 6}, - { DBGBUS_DSPP, 74, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 75, 1}, - { DBGBUS_DSPP, 75, 2}, - { DBGBUS_DSPP, 75, 3}, - { DBGBUS_DSPP, 75, 4}, - { DBGBUS_DSPP, 75, 5}, - { DBGBUS_DSPP, 75, 6}, - { DBGBUS_DSPP, 75, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 76, 1}, - { DBGBUS_DSPP, 76, 2}, - { DBGBUS_DSPP, 76, 3}, - { DBGBUS_DSPP, 76, 4}, - { DBGBUS_DSPP, 76, 5}, - { DBGBUS_DSPP, 76, 6}, - { DBGBUS_DSPP, 76, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 88, 1}, - { DBGBUS_DSPP, 88, 2}, - { DBGBUS_DSPP, 88, 3}, - { DBGBUS_DSPP, 88, 4}, - { DBGBUS_DSPP, 88, 5}, - { DBGBUS_DSPP, 88, 6}, - { DBGBUS_DSPP, 88, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 89, 1}, - { DBGBUS_DSPP, 89, 2}, - { DBGBUS_DSPP, 89, 3}, - { DBGBUS_DSPP, 89, 4}, - { DBGBUS_DSPP, 89, 5}, - { DBGBUS_DSPP, 89, 6}, - { DBGBUS_DSPP, 89, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 90, 1}, - { DBGBUS_DSPP, 90, 2}, - { DBGBUS_DSPP, 90, 3}, - { DBGBUS_DSPP, 90, 4}, - { DBGBUS_DSPP, 90, 5}, - { DBGBUS_DSPP, 90, 6}, - { DBGBUS_DSPP, 90, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 91, 1}, - { DBGBUS_DSPP, 91, 2}, - { DBGBUS_DSPP, 91, 3}, - { DBGBUS_DSPP, 91, 4}, - { DBGBUS_DSPP, 91, 5}, - { DBGBUS_DSPP, 91, 6}, - { DBGBUS_DSPP, 91, 7, _dpu_debug_bus_lm_dump }, - - /* LM2 */ - { DBGBUS_DSPP, 77, 0}, - { DBGBUS_DSPP, 77, 1}, - { DBGBUS_DSPP, 77, 2}, - { DBGBUS_DSPP, 77, 3}, - { DBGBUS_DSPP, 77, 4}, - { DBGBUS_DSPP, 77, 5}, - { DBGBUS_DSPP, 77, 6}, - { DBGBUS_DSPP, 77, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 78, 0}, - { DBGBUS_DSPP, 78, 1}, - { DBGBUS_DSPP, 78, 2}, - { DBGBUS_DSPP, 78, 3}, - { DBGBUS_DSPP, 78, 4}, - { DBGBUS_DSPP, 78, 5}, - { DBGBUS_DSPP, 78, 6}, - { DBGBUS_DSPP, 78, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 79, 0}, - { DBGBUS_DSPP, 79, 1}, - { DBGBUS_DSPP, 79, 2}, - { DBGBUS_DSPP, 79, 3}, - { DBGBUS_DSPP, 79, 4}, - { DBGBUS_DSPP, 79, 5}, - { DBGBUS_DSPP, 79, 6}, - { DBGBUS_DSPP, 79, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 80, 0}, - { DBGBUS_DSPP, 80, 1}, - { DBGBUS_DSPP, 80, 2}, - { DBGBUS_DSPP, 80, 3}, - { DBGBUS_DSPP, 80, 4}, - { DBGBUS_DSPP, 80, 5}, - { DBGBUS_DSPP, 80, 6}, - { DBGBUS_DSPP, 80, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 81, 0}, - { DBGBUS_DSPP, 81, 1}, - { DBGBUS_DSPP, 81, 2}, - { DBGBUS_DSPP, 81, 3}, - { DBGBUS_DSPP, 81, 4}, - { DBGBUS_DSPP, 81, 5}, - { DBGBUS_DSPP, 81, 6}, - { DBGBUS_DSPP, 81, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 82, 0}, - { DBGBUS_DSPP, 82, 1}, - { DBGBUS_DSPP, 82, 2}, - { DBGBUS_DSPP, 82, 3}, - { DBGBUS_DSPP, 82, 4}, - { DBGBUS_DSPP, 82, 5}, - { DBGBUS_DSPP, 82, 6}, - { DBGBUS_DSPP, 82, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 83, 0}, - { DBGBUS_DSPP, 83, 1}, - { DBGBUS_DSPP, 83, 2}, - { DBGBUS_DSPP, 83, 3}, - { DBGBUS_DSPP, 83, 4}, - { DBGBUS_DSPP, 83, 5}, - { DBGBUS_DSPP, 83, 6}, - { DBGBUS_DSPP, 83, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 92, 1}, - { DBGBUS_DSPP, 92, 2}, - { DBGBUS_DSPP, 92, 3}, - { DBGBUS_DSPP, 92, 4}, - { DBGBUS_DSPP, 92, 5}, - { DBGBUS_DSPP, 92, 6}, - { DBGBUS_DSPP, 92, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 93, 1}, - { DBGBUS_DSPP, 93, 2}, - { DBGBUS_DSPP, 93, 3}, - { DBGBUS_DSPP, 93, 4}, - { DBGBUS_DSPP, 93, 5}, - { DBGBUS_DSPP, 93, 6}, - { DBGBUS_DSPP, 93, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 94, 1}, - { DBGBUS_DSPP, 94, 2}, - { DBGBUS_DSPP, 94, 3}, - { DBGBUS_DSPP, 94, 4}, - { DBGBUS_DSPP, 94, 5}, - { DBGBUS_DSPP, 94, 6}, - { DBGBUS_DSPP, 94, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 95, 1}, - { DBGBUS_DSPP, 95, 2}, - { DBGBUS_DSPP, 95, 3}, - { DBGBUS_DSPP, 95, 4}, - { DBGBUS_DSPP, 95, 5}, - { DBGBUS_DSPP, 95, 6}, - { DBGBUS_DSPP, 95, 7, _dpu_debug_bus_lm_dump }, - - /* LM5 */ - { DBGBUS_DSPP, 110, 1}, - { DBGBUS_DSPP, 110, 2}, - { DBGBUS_DSPP, 110, 3}, - { DBGBUS_DSPP, 110, 4}, - { DBGBUS_DSPP, 110, 5}, - { DBGBUS_DSPP, 110, 6}, - { DBGBUS_DSPP, 110, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 111, 1}, - { DBGBUS_DSPP, 111, 2}, - { DBGBUS_DSPP, 111, 3}, - { DBGBUS_DSPP, 111, 4}, - { DBGBUS_DSPP, 111, 5}, - { DBGBUS_DSPP, 111, 6}, - { DBGBUS_DSPP, 111, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 112, 1}, - { DBGBUS_DSPP, 112, 2}, - { DBGBUS_DSPP, 112, 3}, - { DBGBUS_DSPP, 112, 4}, - { DBGBUS_DSPP, 112, 5}, - { DBGBUS_DSPP, 112, 6}, - { DBGBUS_DSPP, 112, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 113, 1}, - { DBGBUS_DSPP, 113, 2}, - { DBGBUS_DSPP, 113, 3}, - { DBGBUS_DSPP, 113, 4}, - { DBGBUS_DSPP, 113, 5}, - { DBGBUS_DSPP, 113, 6}, - { DBGBUS_DSPP, 113, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 114, 1}, - { DBGBUS_DSPP, 114, 2}, - { DBGBUS_DSPP, 114, 3}, - { DBGBUS_DSPP, 114, 4}, - { DBGBUS_DSPP, 114, 5}, - { DBGBUS_DSPP, 114, 6}, - { DBGBUS_DSPP, 114, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 115, 1}, - { DBGBUS_DSPP, 115, 2}, - { DBGBUS_DSPP, 115, 3}, - { DBGBUS_DSPP, 115, 4}, - { DBGBUS_DSPP, 115, 5}, - { DBGBUS_DSPP, 115, 6}, - { DBGBUS_DSPP, 115, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 116, 1}, - { DBGBUS_DSPP, 116, 2}, - { DBGBUS_DSPP, 116, 3}, - { DBGBUS_DSPP, 116, 4}, - { DBGBUS_DSPP, 116, 5}, - { DBGBUS_DSPP, 116, 6}, - { DBGBUS_DSPP, 116, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 117, 1}, - { DBGBUS_DSPP, 117, 2}, - { DBGBUS_DSPP, 117, 3}, - { DBGBUS_DSPP, 117, 4}, - { DBGBUS_DSPP, 117, 5}, - { DBGBUS_DSPP, 117, 6}, - { DBGBUS_DSPP, 117, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 118, 1}, - { DBGBUS_DSPP, 118, 2}, - { DBGBUS_DSPP, 118, 3}, - { DBGBUS_DSPP, 118, 4}, - { DBGBUS_DSPP, 118, 5}, - { DBGBUS_DSPP, 118, 6}, - { DBGBUS_DSPP, 118, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 119, 1}, - { DBGBUS_DSPP, 119, 2}, - { DBGBUS_DSPP, 119, 3}, - { DBGBUS_DSPP, 119, 4}, - { DBGBUS_DSPP, 119, 5}, - { DBGBUS_DSPP, 119, 6}, - { DBGBUS_DSPP, 119, 7, _dpu_debug_bus_lm_dump }, - - { DBGBUS_DSPP, 120, 1}, - { DBGBUS_DSPP, 120, 2}, - { DBGBUS_DSPP, 120, 3}, - { DBGBUS_DSPP, 120, 4}, - { DBGBUS_DSPP, 120, 5}, - { DBGBUS_DSPP, 120, 6}, - { DBGBUS_DSPP, 120, 7, _dpu_debug_bus_lm_dump }, - - /* csc */ - { DBGBUS_SSPP0, 7, 0}, - { DBGBUS_SSPP0, 7, 1}, - { DBGBUS_SSPP0, 27, 0}, - { DBGBUS_SSPP0, 27, 1}, - { DBGBUS_SSPP1, 7, 0}, - { DBGBUS_SSPP1, 7, 1}, - { DBGBUS_SSPP1, 27, 0}, - { DBGBUS_SSPP1, 27, 1}, - - /* pcc */ - { DBGBUS_SSPP0, 3, 3}, - { DBGBUS_SSPP0, 23, 3}, - { DBGBUS_SSPP0, 33, 3}, - { DBGBUS_SSPP0, 43, 3}, - { DBGBUS_SSPP1, 3, 3}, - { DBGBUS_SSPP1, 23, 3}, - { DBGBUS_SSPP1, 33, 3}, - { DBGBUS_SSPP1, 43, 3}, - - /* spa */ - { DBGBUS_SSPP0, 8, 0}, - { DBGBUS_SSPP0, 28, 0}, - { DBGBUS_SSPP1, 8, 0}, - { DBGBUS_SSPP1, 28, 0}, - { DBGBUS_DSPP, 13, 0}, - { DBGBUS_DSPP, 19, 0}, - - /* igc */ - { DBGBUS_SSPP0, 17, 0}, - { DBGBUS_SSPP0, 17, 1}, - { DBGBUS_SSPP0, 17, 3}, - { DBGBUS_SSPP0, 37, 0}, - { DBGBUS_SSPP0, 37, 1}, - { DBGBUS_SSPP0, 37, 3}, - { DBGBUS_SSPP0, 46, 0}, - { DBGBUS_SSPP0, 46, 1}, - { DBGBUS_SSPP0, 46, 3}, - - { DBGBUS_SSPP1, 17, 0}, - { DBGBUS_SSPP1, 17, 1}, - { DBGBUS_SSPP1, 17, 3}, - { DBGBUS_SSPP1, 37, 0}, - { DBGBUS_SSPP1, 37, 1}, - { DBGBUS_SSPP1, 37, 3}, - { DBGBUS_SSPP1, 46, 0}, - { DBGBUS_SSPP1, 46, 1}, - { DBGBUS_SSPP1, 46, 3}, - - { DBGBUS_DSPP, 14, 0}, - { DBGBUS_DSPP, 14, 1}, - { DBGBUS_DSPP, 14, 3}, - { DBGBUS_DSPP, 20, 0}, - { DBGBUS_DSPP, 20, 1}, - { DBGBUS_DSPP, 20, 3}, - - /* intf0-3 */ - { DBGBUS_PERIPH, 0, 0}, - { DBGBUS_PERIPH, 1, 0}, - { DBGBUS_PERIPH, 2, 0}, - { DBGBUS_PERIPH, 3, 0}, - - /* te counter wrapper */ - { DBGBUS_PERIPH, 60, 0}, - - /* dsc0 */ - { DBGBUS_PERIPH, 47, 0}, - { DBGBUS_PERIPH, 47, 1}, - { DBGBUS_PERIPH, 47, 2}, - { DBGBUS_PERIPH, 47, 3}, - { DBGBUS_PERIPH, 47, 4}, - { DBGBUS_PERIPH, 47, 5}, - { DBGBUS_PERIPH, 47, 6}, - { DBGBUS_PERIPH, 47, 7}, - - /* dsc1 */ - { DBGBUS_PERIPH, 48, 0}, - { DBGBUS_PERIPH, 48, 1}, - { DBGBUS_PERIPH, 48, 2}, - { DBGBUS_PERIPH, 48, 3}, - { DBGBUS_PERIPH, 48, 4}, - { DBGBUS_PERIPH, 48, 5}, - { DBGBUS_PERIPH, 48, 6}, - { DBGBUS_PERIPH, 48, 7}, - - /* dsc2 */ - { DBGBUS_PERIPH, 51, 0}, - { DBGBUS_PERIPH, 51, 1}, - { DBGBUS_PERIPH, 51, 2}, - { DBGBUS_PERIPH, 51, 3}, - { DBGBUS_PERIPH, 51, 4}, - { DBGBUS_PERIPH, 51, 5}, - { DBGBUS_PERIPH, 51, 6}, - { DBGBUS_PERIPH, 51, 7}, - - /* dsc3 */ - { DBGBUS_PERIPH, 52, 0}, - { DBGBUS_PERIPH, 52, 1}, - { DBGBUS_PERIPH, 52, 2}, - { DBGBUS_PERIPH, 52, 3}, - { DBGBUS_PERIPH, 52, 4}, - { DBGBUS_PERIPH, 52, 5}, - { DBGBUS_PERIPH, 52, 6}, - { DBGBUS_PERIPH, 52, 7}, - - /* tear-check */ - { DBGBUS_PERIPH, 63, 0 }, - { DBGBUS_PERIPH, 64, 0 }, - { DBGBUS_PERIPH, 65, 0 }, - { DBGBUS_PERIPH, 73, 0 }, - { DBGBUS_PERIPH, 74, 0 }, - - /* cdwn */ - { DBGBUS_PERIPH, 80, 0}, - { DBGBUS_PERIPH, 80, 1}, - { DBGBUS_PERIPH, 80, 2}, - - { DBGBUS_PERIPH, 81, 0}, - { DBGBUS_PERIPH, 81, 1}, - { DBGBUS_PERIPH, 81, 2}, - - { DBGBUS_PERIPH, 82, 0}, - { DBGBUS_PERIPH, 82, 1}, - { DBGBUS_PERIPH, 82, 2}, - { DBGBUS_PERIPH, 82, 3}, - { DBGBUS_PERIPH, 82, 4}, - { DBGBUS_PERIPH, 82, 5}, - { DBGBUS_PERIPH, 82, 6}, - { DBGBUS_PERIPH, 82, 7}, - - /* hdmi */ - { DBGBUS_PERIPH, 68, 0}, - { DBGBUS_PERIPH, 68, 1}, - { DBGBUS_PERIPH, 68, 2}, - { DBGBUS_PERIPH, 68, 3}, - { DBGBUS_PERIPH, 68, 4}, - { DBGBUS_PERIPH, 68, 5}, - - /* edp */ - { DBGBUS_PERIPH, 69, 0}, - { DBGBUS_PERIPH, 69, 1}, - { DBGBUS_PERIPH, 69, 2}, - { DBGBUS_PERIPH, 69, 3}, - { DBGBUS_PERIPH, 69, 4}, - { DBGBUS_PERIPH, 69, 5}, - - /* dsi0 */ - { DBGBUS_PERIPH, 70, 0}, - { DBGBUS_PERIPH, 70, 1}, - { DBGBUS_PERIPH, 70, 2}, - { DBGBUS_PERIPH, 70, 3}, - { DBGBUS_PERIPH, 70, 4}, - { DBGBUS_PERIPH, 70, 5}, - - /* dsi1 */ - { DBGBUS_PERIPH, 71, 0}, - { DBGBUS_PERIPH, 71, 1}, - { DBGBUS_PERIPH, 71, 2}, - { DBGBUS_PERIPH, 71, 3}, - { DBGBUS_PERIPH, 71, 4}, - { DBGBUS_PERIPH, 71, 5}, -}; - -static struct vbif_debug_bus_entry vbif_dbg_bus_msm8998[] = { - {0x214, 0x21c, 16, 2, 0x0, 0xd}, /* arb clients */ - {0x214, 0x21c, 16, 2, 0x80, 0xc0}, /* arb clients */ - {0x214, 0x21c, 16, 2, 0x100, 0x140}, /* arb clients */ - {0x214, 0x21c, 0, 16, 0x0, 0xf}, /* xin blocks - axi side */ - {0x214, 0x21c, 0, 16, 0x80, 0xa4}, /* xin blocks - axi side */ - {0x214, 0x21c, 0, 15, 0x100, 0x124}, /* xin blocks - axi side */ - {0x21c, 0x214, 0, 14, 0, 0xc}, /* xin blocks - clock side */ -}; - -/** - * _dpu_dbg_enable_power - use callback to turn power on for hw register access - * @enable: whether to turn power on or off - */ -static inline void _dpu_dbg_enable_power(int enable) -{ - if (enable) - pm_runtime_get_sync(dpu_dbg_base.dev); - else - pm_runtime_put_sync(dpu_dbg_base.dev); -} - -static void _dpu_dbg_dump_dpu_dbg_bus(struct dpu_dbg_dpu_debug_bus *bus) -{ - bool in_log, in_mem; - u32 **dump_mem = NULL; - u32 *dump_addr = NULL; - u32 status = 0; - struct dpu_debug_bus_entry *head; - phys_addr_t phys = 0; - int list_size; - int i; - u32 offset; - void __iomem *mem_base = NULL; - struct dpu_dbg_reg_base *reg_base; - - if (!bus || !bus->cmn.entries_size) - return; - - list_for_each_entry(reg_base, &dpu_dbg_base.reg_base_list, - reg_base_head) - if (strlen(reg_base->name) && - !strcmp(reg_base->name, bus->cmn.name)) - mem_base = reg_base->base + bus->top_blk_off; - - if (!mem_base) { - pr_err("unable to find mem_base for %s\n", bus->cmn.name); - return; - } - - dump_mem = &bus->cmn.dumped_content; - - /* will keep in memory 4 entries of 4 bytes each */ - list_size = (bus->cmn.entries_size * 4 * 4); - - in_log = (bus->cmn.enable_mask & DPU_DBG_DUMP_IN_LOG); - in_mem = (bus->cmn.enable_mask & DPU_DBG_DUMP_IN_MEM); - - if (!in_log && !in_mem) - return; - - dev_info(dpu_dbg_base.dev, "======== start %s dump =========\n", - bus->cmn.name); - - if (in_mem) { - if (!(*dump_mem)) - *dump_mem = dma_alloc_coherent(dpu_dbg_base.dev, - list_size, &phys, GFP_KERNEL); - - if (*dump_mem) { - dump_addr = *dump_mem; - dev_info(dpu_dbg_base.dev, - "%s: start_addr:0x%pK len:0x%x\n", - __func__, dump_addr, list_size); - } else { - in_mem = false; - pr_err("dump_mem: allocation fails\n"); - } - } - - _dpu_dbg_enable_power(true); - for (i = 0; i < bus->cmn.entries_size; i++) { - head = bus->entries + i; - writel_relaxed(TEST_MASK(head->block_id, head->test_id), - mem_base + head->wr_addr); - wmb(); /* make sure test bits were written */ - - if (bus->cmn.flags & DBGBUS_FLAGS_DSPP) { - offset = DBGBUS_DSPP_STATUS; - /* keep DSPP test point enabled */ - if (head->wr_addr != DBGBUS_DSPP) - writel_relaxed(0xF, mem_base + DBGBUS_DSPP); - } else { - offset = head->wr_addr + 0x4; - } - - status = readl_relaxed(mem_base + offset); - - if (in_log) - dev_info(dpu_dbg_base.dev, - "waddr=0x%x blk=%d tst=%d val=0x%x\n", - head->wr_addr, head->block_id, - head->test_id, status); - - if (dump_addr && in_mem) { - dump_addr[i*4] = head->wr_addr; - dump_addr[i*4 + 1] = head->block_id; - dump_addr[i*4 + 2] = head->test_id; - dump_addr[i*4 + 3] = status; - } - - if (head->analyzer) - head->analyzer(mem_base, head, status); - - /* Disable debug bus once we are done */ - writel_relaxed(0, mem_base + head->wr_addr); - if (bus->cmn.flags & DBGBUS_FLAGS_DSPP && - head->wr_addr != DBGBUS_DSPP) - writel_relaxed(0x0, mem_base + DBGBUS_DSPP); - } - _dpu_dbg_enable_power(false); - - dev_info(dpu_dbg_base.dev, "======== end %s dump =========\n", - bus->cmn.name); -} - -static void _dpu_dbg_dump_vbif_debug_bus_entry( - struct vbif_debug_bus_entry *head, void __iomem *mem_base, - u32 *dump_addr, bool in_log) -{ - int i, j; - u32 val; - - if (!dump_addr && !in_log) - return; - - for (i = 0; i < head->block_cnt; i++) { - writel_relaxed(1 << (i + head->bit_offset), - mem_base + head->block_bus_addr); - /* make sure that current bus blcok enable */ - wmb(); - for (j = head->test_pnt_start; j < head->test_pnt_cnt; j++) { - writel_relaxed(j, mem_base + head->block_bus_addr + 4); - /* make sure that test point is enabled */ - wmb(); - val = readl_relaxed(mem_base + MMSS_VBIF_TEST_BUS_OUT); - if (dump_addr) { - *dump_addr++ = head->block_bus_addr; - *dump_addr++ = i; - *dump_addr++ = j; - *dump_addr++ = val; - } - if (in_log) - dev_info(dpu_dbg_base.dev, - "testpoint:%x arb/xin id=%d index=%d val=0x%x\n", - head->block_bus_addr, i, j, val); - } - } -} - -static void _dpu_dbg_dump_vbif_dbg_bus(struct dpu_dbg_vbif_debug_bus *bus) -{ - bool in_log, in_mem; - u32 **dump_mem = NULL; - u32 *dump_addr = NULL; - u32 value, d0, d1; - unsigned long reg, reg1, reg2; - struct vbif_debug_bus_entry *head; - phys_addr_t phys = 0; - int i, list_size = 0; - void __iomem *mem_base = NULL; - struct vbif_debug_bus_entry *dbg_bus; - u32 bus_size; - struct dpu_dbg_reg_base *reg_base; - - if (!bus || !bus->cmn.entries_size) - return; - - list_for_each_entry(reg_base, &dpu_dbg_base.reg_base_list, - reg_base_head) - if (strlen(reg_base->name) && - !strcmp(reg_base->name, bus->cmn.name)) - mem_base = reg_base->base; - - if (!mem_base) { - pr_err("unable to find mem_base for %s\n", bus->cmn.name); - return; - } - - dbg_bus = bus->entries; - bus_size = bus->cmn.entries_size; - list_size = bus->cmn.entries_size; - dump_mem = &bus->cmn.dumped_content; - - dev_info(dpu_dbg_base.dev, "======== start %s dump =========\n", - bus->cmn.name); - - if (!dump_mem || !dbg_bus || !bus_size || !list_size) - return; - - /* allocate memory for each test point */ - for (i = 0; i < bus_size; i++) { - head = dbg_bus + i; - list_size += (head->block_cnt * head->test_pnt_cnt); - } - - /* 4 bytes * 4 entries for each test point*/ - list_size *= 16; - - in_log = (bus->cmn.enable_mask & DPU_DBG_DUMP_IN_LOG); - in_mem = (bus->cmn.enable_mask & DPU_DBG_DUMP_IN_MEM); - - if (!in_log && !in_mem) - return; - - if (in_mem) { - if (!(*dump_mem)) - *dump_mem = dma_alloc_coherent(dpu_dbg_base.dev, - list_size, &phys, GFP_KERNEL); - - if (*dump_mem) { - dump_addr = *dump_mem; - dev_info(dpu_dbg_base.dev, - "%s: start_addr:0x%pK len:0x%x\n", - __func__, dump_addr, list_size); - } else { - in_mem = false; - pr_err("dump_mem: allocation fails\n"); - } - } - - _dpu_dbg_enable_power(true); - - value = readl_relaxed(mem_base + MMSS_VBIF_CLKON); - writel_relaxed(value | BIT(1), mem_base + MMSS_VBIF_CLKON); - - /* make sure that vbif core is on */ - wmb(); - - /** - * Extract VBIF error info based on XIN halt and error status. - * If the XIN client is not in HALT state, or an error is detected, - * then retrieve the VBIF error info for it. - */ - reg = readl_relaxed(mem_base + MMSS_VBIF_XIN_HALT_CTRL1); - reg1 = readl_relaxed(mem_base + MMSS_VBIF_PND_ERR); - reg2 = readl_relaxed(mem_base + MMSS_VBIF_SRC_ERR); - dev_err(dpu_dbg_base.dev, - "XIN HALT:0x%lX, PND ERR:0x%lX, SRC ERR:0x%lX\n", - reg, reg1, reg2); - reg >>= 16; - reg &= ~(reg1 | reg2); - for (i = 0; i < MMSS_VBIF_CLIENT_NUM; i++) { - if (!test_bit(0, ®)) { - writel_relaxed(i, mem_base + MMSS_VBIF_ERR_INFO); - /* make sure reg write goes through */ - wmb(); - - d0 = readl_relaxed(mem_base + MMSS_VBIF_ERR_INFO); - d1 = readl_relaxed(mem_base + MMSS_VBIF_ERR_INFO_1); - - dev_err(dpu_dbg_base.dev, - "Client:%d, errinfo=0x%X, errinfo1=0x%X\n", - i, d0, d1); - } - reg >>= 1; - } - - for (i = 0; i < bus_size; i++) { - head = dbg_bus + i; - - writel_relaxed(0, mem_base + head->disable_bus_addr); - writel_relaxed(BIT(0), mem_base + MMSS_VBIF_TEST_BUS_OUT_CTRL); - /* make sure that other bus is off */ - wmb(); - - _dpu_dbg_dump_vbif_debug_bus_entry(head, mem_base, dump_addr, - in_log); - if (dump_addr) - dump_addr += (head->block_cnt * head->test_pnt_cnt * 4); - } - - _dpu_dbg_enable_power(false); - - dev_info(dpu_dbg_base.dev, "======== end %s dump =========\n", - bus->cmn.name); -} - -/** - * _dpu_dump_array - dump array of register bases - * @name: string indicating origin of dump - * @dump_dbgbus_dpu: whether to dump the dpu debug bus - * @dump_dbgbus_vbif_rt: whether to dump the vbif rt debug bus - */ -static void _dpu_dump_array(const char *name, bool dump_dbgbus_dpu, - bool dump_dbgbus_vbif_rt) -{ - if (dump_dbgbus_dpu) - _dpu_dbg_dump_dpu_dbg_bus(&dpu_dbg_base.dbgbus_dpu); - - if (dump_dbgbus_vbif_rt) - _dpu_dbg_dump_vbif_dbg_bus(&dpu_dbg_base.dbgbus_vbif_rt); -} - -/** - * _dpu_dump_work - deferred dump work function - * @work: work structure - */ -static void _dpu_dump_work(struct work_struct *work) -{ - _dpu_dump_array("dpudump_workitem", - dpu_dbg_base.dbgbus_dpu.cmn.include_in_deferred_work, - dpu_dbg_base.dbgbus_vbif_rt.cmn.include_in_deferred_work); -} - -void dpu_dbg_dump(bool queue_work, const char *name, bool dump_dbgbus_dpu, - bool dump_dbgbus_vbif_rt) -{ - if (queue_work && work_pending(&dpu_dbg_base.dump_work)) - return; - - if (!queue_work) { - _dpu_dump_array(name, dump_dbgbus_dpu, dump_dbgbus_vbif_rt); - return; - } - - /* schedule work to dump later */ - dpu_dbg_base.dbgbus_dpu.cmn.include_in_deferred_work = dump_dbgbus_dpu; - dpu_dbg_base.dbgbus_vbif_rt.cmn.include_in_deferred_work = - dump_dbgbus_vbif_rt; - schedule_work(&dpu_dbg_base.dump_work); -} - -/* - * dpu_dbg_debugfs_open - debugfs open handler for debug dump - * @inode: debugfs inode - * @file: file handle - */ -static int dpu_dbg_debugfs_open(struct inode *inode, struct file *file) -{ - /* non-seekable */ - file->f_mode &= ~(FMODE_LSEEK | FMODE_PREAD | FMODE_PWRITE); - file->private_data = inode->i_private; - return 0; -} - -/** - * dpu_dbg_dump_write - debugfs write handler for debug dump - * @file: file handler - * @user_buf: user buffer content from debugfs - * @count: size of user buffer - * @ppos: position offset of user buffer - */ -static ssize_t dpu_dbg_dump_write(struct file *file, - const char __user *user_buf, size_t count, loff_t *ppos) -{ - _dpu_dump_array("dump_debugfs", true, true); - return count; -} - -static const struct file_operations dpu_dbg_dump_fops = { - .open = dpu_dbg_debugfs_open, - .write = dpu_dbg_dump_write, -}; - -int dpu_dbg_debugfs_register(struct dentry *debugfs_root) -{ - static struct dpu_dbg_base *dbg = &dpu_dbg_base; - char debug_name[80] = ""; - - if (!debugfs_root) - return -EINVAL; - - debugfs_create_file("dump", 0600, debugfs_root, NULL, - &dpu_dbg_dump_fops); - - if (dbg->dbgbus_dpu.entries) { - dbg->dbgbus_dpu.cmn.name = DBGBUS_NAME_DPU; - snprintf(debug_name, sizeof(debug_name), "%s_dbgbus", - dbg->dbgbus_dpu.cmn.name); - dbg->dbgbus_dpu.cmn.enable_mask = DEFAULT_DBGBUS_DPU; - debugfs_create_u32(debug_name, 0600, debugfs_root, - &dbg->dbgbus_dpu.cmn.enable_mask); - } - - if (dbg->dbgbus_vbif_rt.entries) { - dbg->dbgbus_vbif_rt.cmn.name = DBGBUS_NAME_VBIF_RT; - snprintf(debug_name, sizeof(debug_name), "%s_dbgbus", - dbg->dbgbus_vbif_rt.cmn.name); - dbg->dbgbus_vbif_rt.cmn.enable_mask = DEFAULT_DBGBUS_VBIFRT; - debugfs_create_u32(debug_name, 0600, debugfs_root, - &dbg->dbgbus_vbif_rt.cmn.enable_mask); - } - - return 0; -} - -static void _dpu_dbg_debugfs_destroy(void) -{ -} - -void dpu_dbg_init_dbg_buses(u32 hwversion) -{ - static struct dpu_dbg_base *dbg = &dpu_dbg_base; - - memset(&dbg->dbgbus_dpu, 0, sizeof(dbg->dbgbus_dpu)); - memset(&dbg->dbgbus_vbif_rt, 0, sizeof(dbg->dbgbus_vbif_rt)); - - if (IS_MSM8998_TARGET(hwversion)) { - dbg->dbgbus_dpu.entries = dbg_bus_dpu_8998; - dbg->dbgbus_dpu.cmn.entries_size = ARRAY_SIZE(dbg_bus_dpu_8998); - dbg->dbgbus_dpu.cmn.flags = DBGBUS_FLAGS_DSPP; - - dbg->dbgbus_vbif_rt.entries = vbif_dbg_bus_msm8998; - dbg->dbgbus_vbif_rt.cmn.entries_size = - ARRAY_SIZE(vbif_dbg_bus_msm8998); - } else if (IS_SDM845_TARGET(hwversion) || IS_SDM670_TARGET(hwversion)) { - dbg->dbgbus_dpu.entries = dbg_bus_dpu_sdm845; - dbg->dbgbus_dpu.cmn.entries_size = - ARRAY_SIZE(dbg_bus_dpu_sdm845); - dbg->dbgbus_dpu.cmn.flags = DBGBUS_FLAGS_DSPP; - - /* vbif is unchanged vs 8998 */ - dbg->dbgbus_vbif_rt.entries = vbif_dbg_bus_msm8998; - dbg->dbgbus_vbif_rt.cmn.entries_size = - ARRAY_SIZE(vbif_dbg_bus_msm8998); - } else { - pr_err("unsupported chipset id %X\n", hwversion); - } -} - -int dpu_dbg_init(struct device *dev) -{ - if (!dev) { - pr_err("invalid params\n"); - return -EINVAL; - } - - INIT_LIST_HEAD(&dpu_dbg_base.reg_base_list); - dpu_dbg_base.dev = dev; - - INIT_WORK(&dpu_dbg_base.dump_work, _dpu_dump_work); - - return 0; -} - -/** - * dpu_dbg_destroy - destroy dpu debug facilities - */ -void dpu_dbg_destroy(void) -{ - _dpu_dbg_debugfs_destroy(); -} - -void dpu_dbg_set_dpu_top_offset(u32 blk_off) -{ - dpu_dbg_base.dbgbus_dpu.top_blk_off = blk_off; -} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_dbg.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_dbg.h deleted file mode 100644 index 1e6fa945f98b..000000000000 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_dbg.h +++ /dev/null @@ -1,103 +0,0 @@ -/* Copyright (c) 2016-2018, The Linux Foundation. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 and - * only version 2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - */ - -#ifndef DPU_DBG_H_ -#define DPU_DBG_H_ - -#include <stdarg.h> -#include <linux/debugfs.h> -#include <linux/list.h> - -enum dpu_dbg_dump_flag { - DPU_DBG_DUMP_IN_LOG = BIT(0), - DPU_DBG_DUMP_IN_MEM = BIT(1), -}; - -#if defined(CONFIG_DEBUG_FS) - -/** - * dpu_dbg_init_dbg_buses - initialize debug bus dumping support for the chipset - * @hwversion: Chipset revision - */ -void dpu_dbg_init_dbg_buses(u32 hwversion); - -/** - * dpu_dbg_init - initialize global dpu debug facilities: regdump - * @dev: device handle - * Returns: 0 or -ERROR - */ -int dpu_dbg_init(struct device *dev); - -/** - * dpu_dbg_debugfs_register - register entries at the given debugfs dir - * @debugfs_root: debugfs root in which to create dpu debug entries - * Returns: 0 or -ERROR - */ -int dpu_dbg_debugfs_register(struct dentry *debugfs_root); - -/** - * dpu_dbg_destroy - destroy the global dpu debug facilities - * Returns: none - */ -void dpu_dbg_destroy(void); - -/** - * dpu_dbg_dump - trigger dumping of all dpu_dbg facilities - * @queue_work: whether to queue the dumping work to the work_struct - * @name: string indicating origin of dump - * @dump_dbgbus: dump the dpu debug bus - * @dump_vbif_rt: dump the vbif rt bus - * Returns: none - */ -void dpu_dbg_dump(bool queue_work, const char *name, bool dump_dbgbus_dpu, - bool dump_dbgbus_vbif_rt); - -/** - * dpu_dbg_set_dpu_top_offset - set the target specific offset from mdss base - * address of the top registers. Used for accessing debug bus controls. - * @blk_off: offset from mdss base of the top block - */ -void dpu_dbg_set_dpu_top_offset(u32 blk_off); - -#else - -static inline void dpu_dbg_init_dbg_buses(u32 hwversion) -{ -} - -static inline int dpu_dbg_init(struct device *dev) -{ - return 0; -} - -static inline int dpu_dbg_debugfs_register(struct dentry *debugfs_root) -{ - return 0; -} - -static inline void dpu_dbg_destroy(void) -{ -} - -static inline void dpu_dbg_dump(bool queue_work, const char *name, - bool dump_dbgbus_dpu, bool dump_dbgbus_vbif_rt) -{ -} - -static inline void dpu_dbg_set_dpu_top_offset(u32 blk_off) -{ -} - -#endif /* defined(CONFIG_DEBUG_FS) */ - - -#endif /* DPU_DBG_H_ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c index 96cdf06e7da2..36158b7d99cd 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c @@ -130,8 +130,9 @@ enum dpu_enc_rc_states { * Virtual encoder defers as much as possible to the physical encoders. * Virtual encoder registers itself with the DRM Framework as the encoder. * @base: drm_encoder base class for registration with DRM - * @enc_spin_lock: Virtual-Encoder-Wide Spin Lock for IRQ purposes + * @enc_spinlock: Virtual-Encoder-Wide Spin Lock for IRQ purposes * @bus_scaling_client: Client handle to the bus scaling interface + * @enabled: True if the encoder is active, protected by enc_lock * @num_phys_encs: Actual number of physical encoders contained. * @phys_encs: Container of physical encoders managed. * @cur_master: Pointer to the current master in this mode. Optimization @@ -141,15 +142,17 @@ enum dpu_enc_rc_states { * @intfs_swapped Whether or not the phys_enc interfaces have been swapped * for partial update right-only cases, such as pingpong * split where virtual pingpong does not generate IRQs - * @crtc_vblank_cb: Callback into the upper layer / CRTC for - * notification of the VBLANK - * @crtc_vblank_cb_data: Data from upper layer for VBLANK notification + * @crtc: Pointer to the currently assigned crtc. Normally you + * would use crtc->state->encoder_mask to determine the + * link between encoder/crtc. However in this case we need + * to track crtc in the disable() hook which is called + * _after_ encoder_mask is cleared. * @crtc_kickoff_cb: Callback into CRTC that will flush & start * all CTL paths * @crtc_kickoff_cb_data: Opaque user data given to crtc_kickoff_cb * @debugfs_root: Debug file system root file node - * @enc_lock: Lock around physical encoder create/destroy and - access. + * @enc_lock: Lock around physical encoder + * create/destroy/enable/disable * @frame_busy_mask: Bitmask tracking which phys_enc we are still * busy processing current command. * Bit0 = phys_encs[0] etc. @@ -175,6 +178,8 @@ struct dpu_encoder_virt { spinlock_t enc_spinlock; uint32_t bus_scaling_client; + bool enabled; + unsigned int num_phys_encs; struct dpu_encoder_phys *phys_encs[MAX_PHYS_ENCODERS_PER_VIRTUAL]; struct dpu_encoder_phys *cur_master; @@ -183,8 +188,7 @@ struct dpu_encoder_virt { bool intfs_swapped; - void (*crtc_vblank_cb)(void *); - void *crtc_vblank_cb_data; + struct drm_crtc *crtc; struct dentry *debugfs_root; struct mutex enc_lock; @@ -210,39 +214,6 @@ struct dpu_encoder_virt { }; #define to_dpu_encoder_virt(x) container_of(x, struct dpu_encoder_virt, base) -static inline int _dpu_encoder_power_enable(struct dpu_encoder_virt *dpu_enc, - bool enable) -{ - struct drm_encoder *drm_enc; - struct msm_drm_private *priv; - struct dpu_kms *dpu_kms; - - if (!dpu_enc) { - DPU_ERROR("invalid dpu enc\n"); - return -EINVAL; - } - - drm_enc = &dpu_enc->base; - if (!drm_enc->dev || !drm_enc->dev->dev_private) { - DPU_ERROR("drm device invalid\n"); - return -EINVAL; - } - - priv = drm_enc->dev->dev_private; - if (!priv->kms) { - DPU_ERROR("invalid kms\n"); - return -EINVAL; - } - - dpu_kms = to_dpu_kms(priv->kms); - - if (enable) - pm_runtime_get_sync(&dpu_kms->pdev->dev); - else - pm_runtime_put_sync(&dpu_kms->pdev->dev); - - return 0; -} void dpu_encoder_helper_report_irq_timeout(struct dpu_encoder_phys *phys_enc, enum dpu_intr_idx intr_idx) @@ -488,8 +459,6 @@ static void dpu_encoder_destroy(struct drm_encoder *drm_enc) drm_encoder_cleanup(drm_enc); mutex_destroy(&dpu_enc->enc_lock); - - kfree(dpu_enc); } void dpu_encoder_helper_split_config( @@ -1119,28 +1088,24 @@ static void _dpu_encoder_virt_enable_helper(struct drm_encoder *drm_enc) _dpu_encoder_update_vsync_source(dpu_enc, &dpu_enc->disp_info); } -void dpu_encoder_virt_restore(struct drm_encoder *drm_enc) +void dpu_encoder_virt_runtime_resume(struct drm_encoder *drm_enc) { - struct dpu_encoder_virt *dpu_enc = NULL; - int i; - - if (!drm_enc) { - DPU_ERROR("invalid encoder\n"); - return; - } - dpu_enc = to_dpu_encoder_virt(drm_enc); + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); - for (i = 0; i < dpu_enc->num_phys_encs; i++) { - struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; + mutex_lock(&dpu_enc->enc_lock); - if (phys && (phys != dpu_enc->cur_master) && phys->ops.restore) - phys->ops.restore(phys); - } + if (!dpu_enc->enabled) + goto out; + if (dpu_enc->cur_slave && dpu_enc->cur_slave->ops.restore) + dpu_enc->cur_slave->ops.restore(dpu_enc->cur_slave); if (dpu_enc->cur_master && dpu_enc->cur_master->ops.restore) dpu_enc->cur_master->ops.restore(dpu_enc->cur_master); _dpu_encoder_virt_enable_helper(drm_enc); + +out: + mutex_unlock(&dpu_enc->enc_lock); } static void dpu_encoder_virt_enable(struct drm_encoder *drm_enc) @@ -1154,6 +1119,8 @@ static void dpu_encoder_virt_enable(struct drm_encoder *drm_enc) return; } dpu_enc = to_dpu_encoder_virt(drm_enc); + + mutex_lock(&dpu_enc->enc_lock); cur_mode = &dpu_enc->base.crtc->state->adjusted_mode; trace_dpu_enc_enable(DRMID(drm_enc), cur_mode->hdisplay, @@ -1170,10 +1137,15 @@ static void dpu_encoder_virt_enable(struct drm_encoder *drm_enc) if (ret) { DPU_ERROR_ENC(dpu_enc, "dpu resource control failed: %d\n", ret); - return; + goto out; } _dpu_encoder_virt_enable_helper(drm_enc); + + dpu_enc->enabled = true; + +out: + mutex_unlock(&dpu_enc->enc_lock); } static void dpu_encoder_virt_disable(struct drm_encoder *drm_enc) @@ -1195,11 +1167,14 @@ static void dpu_encoder_virt_disable(struct drm_encoder *drm_enc) return; } - mode = &drm_enc->crtc->state->adjusted_mode; - dpu_enc = to_dpu_encoder_virt(drm_enc); DPU_DEBUG_ENC(dpu_enc, "\n"); + mutex_lock(&dpu_enc->enc_lock); + dpu_enc->enabled = false; + + mode = &drm_enc->crtc->state->adjusted_mode; + priv = drm_enc->dev->dev_private; dpu_kms = to_dpu_kms(priv->kms); @@ -1233,6 +1208,8 @@ static void dpu_encoder_virt_disable(struct drm_encoder *drm_enc) DPU_DEBUG_ENC(dpu_enc, "encoder disabled\n"); dpu_rm_release(&dpu_kms->rm, drm_enc); + + mutex_unlock(&dpu_enc->enc_lock); } static enum dpu_intf dpu_encoder_get_intf(struct dpu_mdss_cfg *catalog, @@ -1263,8 +1240,8 @@ static void dpu_encoder_vblank_callback(struct drm_encoder *drm_enc, dpu_enc = to_dpu_encoder_virt(drm_enc); spin_lock_irqsave(&dpu_enc->enc_spinlock, lock_flags); - if (dpu_enc->crtc_vblank_cb) - dpu_enc->crtc_vblank_cb(dpu_enc->crtc_vblank_cb_data); + if (dpu_enc->crtc) + dpu_crtc_vblank_callback(dpu_enc->crtc); spin_unlock_irqrestore(&dpu_enc->enc_spinlock, lock_flags); atomic_inc(&phy_enc->vsync_cnt); @@ -1284,25 +1261,32 @@ static void dpu_encoder_underrun_callback(struct drm_encoder *drm_enc, DPU_ATRACE_END("encoder_underrun_callback"); } -void dpu_encoder_register_vblank_callback(struct drm_encoder *drm_enc, - void (*vbl_cb)(void *), void *vbl_data) +void dpu_encoder_assign_crtc(struct drm_encoder *drm_enc, struct drm_crtc *crtc) { struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); unsigned long lock_flags; - bool enable; - int i; - enable = vbl_cb ? true : false; + spin_lock_irqsave(&dpu_enc->enc_spinlock, lock_flags); + /* crtc should always be cleared before re-assigning */ + WARN_ON(crtc && dpu_enc->crtc); + dpu_enc->crtc = crtc; + spin_unlock_irqrestore(&dpu_enc->enc_spinlock, lock_flags); +} + +void dpu_encoder_toggle_vblank_for_crtc(struct drm_encoder *drm_enc, + struct drm_crtc *crtc, bool enable) +{ + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); + unsigned long lock_flags; + int i; - if (!drm_enc) { - DPU_ERROR("invalid encoder\n"); - return; - } trace_dpu_enc_vblank_cb(DRMID(drm_enc), enable); spin_lock_irqsave(&dpu_enc->enc_spinlock, lock_flags); - dpu_enc->crtc_vblank_cb = vbl_cb; - dpu_enc->crtc_vblank_cb_data = vbl_data; + if (dpu_enc->crtc != crtc) { + spin_unlock_irqrestore(&dpu_enc->enc_spinlock, lock_flags); + return; + } spin_unlock_irqrestore(&dpu_enc->enc_spinlock, lock_flags); for (i = 0; i < dpu_enc->num_phys_encs; i++) { @@ -1407,8 +1391,9 @@ static void dpu_encoder_off_work(struct kthread_work *work) * phys: Pointer to physical encoder structure * extra_flush_bits: Additional bit mask to include in flush trigger */ -static inline void _dpu_encoder_trigger_flush(struct drm_encoder *drm_enc, - struct dpu_encoder_phys *phys, uint32_t extra_flush_bits) +static void _dpu_encoder_trigger_flush(struct drm_encoder *drm_enc, + struct dpu_encoder_phys *phys, uint32_t extra_flush_bits, + bool async) { struct dpu_hw_ctl *ctl; int pending_kickoff_cnt; @@ -1431,7 +1416,10 @@ static inline void _dpu_encoder_trigger_flush(struct drm_encoder *drm_enc, return; } - pending_kickoff_cnt = dpu_encoder_phys_inc_pending(phys); + if (!async) + pending_kickoff_cnt = dpu_encoder_phys_inc_pending(phys); + else + pending_kickoff_cnt = atomic_read(&phys->pending_kickoff_cnt); if (extra_flush_bits && ctl->ops.update_pending_flush) ctl->ops.update_pending_flush(ctl, extra_flush_bits); @@ -1450,7 +1438,7 @@ static inline void _dpu_encoder_trigger_flush(struct drm_encoder *drm_enc, * _dpu_encoder_trigger_start - trigger start for a physical encoder * phys: Pointer to physical encoder structure */ -static inline void _dpu_encoder_trigger_start(struct dpu_encoder_phys *phys) +static void _dpu_encoder_trigger_start(struct dpu_encoder_phys *phys) { if (!phys) { DPU_ERROR("invalid argument(s)\n"); @@ -1507,7 +1495,7 @@ static int dpu_encoder_helper_wait_event_timeout( return rc; } -void dpu_encoder_helper_hw_reset(struct dpu_encoder_phys *phys_enc) +static void dpu_encoder_helper_hw_reset(struct dpu_encoder_phys *phys_enc) { struct dpu_encoder_virt *dpu_enc; struct dpu_hw_ctl *ctl; @@ -1527,10 +1515,8 @@ void dpu_encoder_helper_hw_reset(struct dpu_encoder_phys *phys_enc) ctl->idx); rc = ctl->ops.reset(ctl); - if (rc) { + if (rc) DPU_ERROR_ENC(dpu_enc, "ctl %d reset failure\n", ctl->idx); - dpu_dbg_dump(false, __func__, true, true); - } phys_enc->enable_state = DPU_ENC_ENABLED; } @@ -1544,7 +1530,8 @@ void dpu_encoder_helper_hw_reset(struct dpu_encoder_phys *phys_enc) * a time. * dpu_enc: Pointer to virtual encoder structure */ -static void _dpu_encoder_kickoff_phys(struct dpu_encoder_virt *dpu_enc) +static void _dpu_encoder_kickoff_phys(struct dpu_encoder_virt *dpu_enc, + bool async) { struct dpu_hw_ctl *ctl; uint32_t i, pending_flush; @@ -1575,7 +1562,8 @@ static void _dpu_encoder_kickoff_phys(struct dpu_encoder_virt *dpu_enc) set_bit(i, dpu_enc->frame_busy_mask); if (!phys->ops.needs_single_flush || !phys->ops.needs_single_flush(phys)) - _dpu_encoder_trigger_flush(&dpu_enc->base, phys, 0x0); + _dpu_encoder_trigger_flush(&dpu_enc->base, phys, 0x0, + async); else if (ctl->ops.get_pending_flush) pending_flush |= ctl->ops.get_pending_flush(ctl); } @@ -1585,7 +1573,7 @@ static void _dpu_encoder_kickoff_phys(struct dpu_encoder_virt *dpu_enc) _dpu_encoder_trigger_flush( &dpu_enc->base, dpu_enc->cur_master, - pending_flush); + pending_flush, async); } _dpu_encoder_trigger_start(dpu_enc->cur_master); @@ -1769,7 +1757,7 @@ static void dpu_encoder_vsync_event_work_handler(struct kthread_work *work) } void dpu_encoder_prepare_for_kickoff(struct drm_encoder *drm_enc, - struct dpu_encoder_kickoff_params *params) + struct dpu_encoder_kickoff_params *params, bool async) { struct dpu_encoder_virt *dpu_enc; struct dpu_encoder_phys *phys; @@ -1803,14 +1791,12 @@ void dpu_encoder_prepare_for_kickoff(struct drm_encoder *drm_enc, if (needs_hw_reset) { trace_dpu_enc_prepare_kickoff_reset(DRMID(drm_enc)); for (i = 0; i < dpu_enc->num_phys_encs; i++) { - phys = dpu_enc->phys_encs[i]; - if (phys && phys->ops.hw_reset) - phys->ops.hw_reset(phys); + dpu_encoder_helper_hw_reset(dpu_enc->phys_encs[i]); } } } -void dpu_encoder_kickoff(struct drm_encoder *drm_enc) +void dpu_encoder_kickoff(struct drm_encoder *drm_enc, bool async) { struct dpu_encoder_virt *dpu_enc; struct dpu_encoder_phys *phys; @@ -1833,7 +1819,7 @@ void dpu_encoder_kickoff(struct drm_encoder *drm_enc) ((atomic_read(&dpu_enc->frame_done_timeout) * HZ) / 1000)); /* All phys encs are ready to go, trigger the kickoff */ - _dpu_encoder_kickoff_phys(dpu_enc); + _dpu_encoder_kickoff_phys(dpu_enc, async); /* allow phys encs to handle any post-kickoff business */ for (i = 0; i < dpu_enc->num_phys_encs; i++) { @@ -1875,14 +1861,9 @@ void dpu_encoder_prepare_commit(struct drm_encoder *drm_enc) #ifdef CONFIG_DEBUG_FS static int _dpu_encoder_status_show(struct seq_file *s, void *data) { - struct dpu_encoder_virt *dpu_enc; + struct dpu_encoder_virt *dpu_enc = s->private; int i; - if (!s || !s->private) - return -EINVAL; - - dpu_enc = s->private; - mutex_lock(&dpu_enc->enc_lock); for (i = 0; i < dpu_enc->num_phys_encs; i++) { struct dpu_encoder_phys *phys = dpu_enc->phys_encs[i]; @@ -1920,7 +1901,7 @@ static int _dpu_encoder_debugfs_status_open(struct inode *inode, static int _dpu_encoder_init_debugfs(struct drm_encoder *drm_enc) { - struct dpu_encoder_virt *dpu_enc; + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(drm_enc); struct msm_drm_private *priv; struct dpu_kms *dpu_kms; int i; @@ -1934,12 +1915,11 @@ static int _dpu_encoder_init_debugfs(struct drm_encoder *drm_enc) char name[DPU_NAME_SIZE]; - if (!drm_enc || !drm_enc->dev || !drm_enc->dev->dev_private) { + if (!drm_enc->dev || !drm_enc->dev->dev_private) { DPU_ERROR("invalid encoder or kms\n"); return -EINVAL; } - dpu_enc = to_dpu_encoder_virt(drm_enc); priv = drm_enc->dev->dev_private; dpu_kms = to_dpu_kms(priv->kms); @@ -1964,26 +1944,11 @@ static int _dpu_encoder_init_debugfs(struct drm_encoder *drm_enc) return 0; } - -static void _dpu_encoder_destroy_debugfs(struct drm_encoder *drm_enc) -{ - struct dpu_encoder_virt *dpu_enc; - - if (!drm_enc) - return; - - dpu_enc = to_dpu_encoder_virt(drm_enc); - debugfs_remove_recursive(dpu_enc->debugfs_root); -} #else static int _dpu_encoder_init_debugfs(struct drm_encoder *drm_enc) { return 0; } - -static void _dpu_encoder_destroy_debugfs(struct drm_encoder *drm_enc) -{ -} #endif static int dpu_encoder_late_register(struct drm_encoder *encoder) @@ -1993,7 +1958,9 @@ static int dpu_encoder_late_register(struct drm_encoder *encoder) static void dpu_encoder_early_unregister(struct drm_encoder *encoder) { - _dpu_encoder_destroy_debugfs(encoder); + struct dpu_encoder_virt *dpu_enc = to_dpu_encoder_virt(encoder); + + debugfs_remove_recursive(dpu_enc->debugfs_root); } static int dpu_encoder_virt_add_phys_encs( @@ -2268,6 +2235,8 @@ struct drm_encoder *dpu_encoder_init(struct drm_device *dev, drm_encoder_helper_add(&dpu_enc->base, &dpu_encoder_helper_funcs); + dpu_enc->enabled = false; + return &dpu_enc->base; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h index 9dbf38f446d9..3f5dafe00580 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h @@ -55,14 +55,22 @@ void dpu_encoder_get_hw_resources(struct drm_encoder *encoder, struct dpu_encoder_hw_resources *hw_res); /** - * dpu_encoder_register_vblank_callback - provide callback to encoder that - * will be called on the next vblank. + * dpu_encoder_assign_crtc - Link the encoder to the crtc it's assigned to * @encoder: encoder pointer - * @cb: callback pointer, provide NULL to deregister and disable IRQs - * @data: user data provided to callback + * @crtc: crtc pointer + */ +void dpu_encoder_assign_crtc(struct drm_encoder *encoder, + struct drm_crtc *crtc); + +/** + * dpu_encoder_toggle_vblank_for_crtc - Toggles vblank interrupts on or off if + * the encoder is assigned to the given crtc + * @encoder: encoder pointer + * @crtc: crtc pointer + * @enable: true if vblank should be enabled */ -void dpu_encoder_register_vblank_callback(struct drm_encoder *encoder, - void (*cb)(void *), void *data); +void dpu_encoder_toggle_vblank_for_crtc(struct drm_encoder *encoder, + struct drm_crtc *crtc, bool enable); /** * dpu_encoder_register_frame_event_callback - provide callback to encoder that @@ -81,9 +89,10 @@ void dpu_encoder_register_frame_event_callback(struct drm_encoder *encoder, * Delayed: Block until next trigger can be issued. * @encoder: encoder pointer * @params: kickoff time parameters + * @async: true if this is an asynchronous commit */ void dpu_encoder_prepare_for_kickoff(struct drm_encoder *encoder, - struct dpu_encoder_kickoff_params *params); + struct dpu_encoder_kickoff_params *params, bool async); /** * dpu_encoder_trigger_kickoff_pending - Clear the flush bits from previous @@ -96,8 +105,9 @@ void dpu_encoder_trigger_kickoff_pending(struct drm_encoder *encoder); * dpu_encoder_kickoff - trigger a double buffer flip of the ctl path * (i.e. ctl flush and start) immediately. * @encoder: encoder pointer + * @async: true if this is an asynchronous commit */ -void dpu_encoder_kickoff(struct drm_encoder *encoder); +void dpu_encoder_kickoff(struct drm_encoder *encoder, bool async); /** * dpu_encoder_wait_for_event - Waits for encoder events @@ -126,10 +136,10 @@ int dpu_encoder_wait_for_event(struct drm_encoder *drm_encoder, enum dpu_intf_mode dpu_encoder_get_intf_mode(struct drm_encoder *encoder); /** - * dpu_encoder_virt_restore - restore the encoder configs + * dpu_encoder_virt_runtime_resume - pm runtime resume the encoder configs * @encoder: encoder pointer */ -void dpu_encoder_virt_restore(struct drm_encoder *encoder); +void dpu_encoder_virt_runtime_resume(struct drm_encoder *encoder); /** * dpu_encoder_init - initialize virtual encoder object diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h index 964efcc757a4..44e6f8b68e70 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h @@ -114,8 +114,6 @@ struct dpu_encoder_virt_ops { * @handle_post_kickoff: Do any work necessary post-kickoff work * @trigger_start: Process start event on physical encoder * @needs_single_flush: Whether encoder slaves need to be flushed - * @hw_reset: Issue HW recovery such as CTL reset and clear - * DPU_ENC_ERR_NEEDS_HW_RESET state * @irq_control: Handler to enable/disable all the encoder IRQs * @prepare_idle_pc: phys encoder can update the vsync_enable status * on idle power collapse prepare @@ -151,7 +149,6 @@ struct dpu_encoder_phys_ops { void (*handle_post_kickoff)(struct dpu_encoder_phys *phys_enc); void (*trigger_start)(struct dpu_encoder_phys *phys_enc); bool (*needs_single_flush)(struct dpu_encoder_phys *phys_enc); - void (*hw_reset)(struct dpu_encoder_phys *phys_enc); void (*irq_control)(struct dpu_encoder_phys *phys, bool enable); void (*prepare_idle_pc)(struct dpu_encoder_phys *phys_enc); void (*restore)(struct dpu_encoder_phys *phys); @@ -342,15 +339,6 @@ struct dpu_encoder_phys *dpu_encoder_phys_cmd_init( */ void dpu_encoder_helper_trigger_start(struct dpu_encoder_phys *phys_enc); -/** - * dpu_encoder_helper_hw_reset - issue ctl hw reset - * This helper function may be optionally specified by physical - * encoders if they require ctl hw reset. If state is currently - * DPU_ENC_ERR_NEEDS_HW_RESET, it is set back to DPU_ENC_ENABLED. - * @phys_enc: Pointer to physical encoder structure - */ -void dpu_encoder_helper_hw_reset(struct dpu_encoder_phys *phys_enc); - static inline enum dpu_3d_blend_mode dpu_encoder_helper_get_3d_blend_mode( struct dpu_encoder_phys *phys_enc) { @@ -362,7 +350,7 @@ static inline enum dpu_3d_blend_mode dpu_encoder_helper_get_3d_blend_mode( dpu_cstate = to_dpu_crtc_state(phys_enc->parent->crtc->state); if (phys_enc->split_role == ENC_ROLE_SOLO && - dpu_crtc_state_is_stereo(dpu_cstate)) + dpu_cstate->num_mixers == CRTC_DUAL_MIXERS) return BLEND_3D_H_ROW_INT; return BLEND_3D_NONE; diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c index b2d7f0ded24c..99ab5ca9bed3 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c @@ -44,14 +44,7 @@ #define DPU_ENC_WR_PTR_START_TIMEOUT_US 20000 -static inline int _dpu_encoder_phys_cmd_get_idle_timeout( - struct dpu_encoder_phys_cmd *cmd_enc) -{ - return KICKOFF_TIMEOUT_MS; -} - -static inline bool dpu_encoder_phys_cmd_is_master( - struct dpu_encoder_phys *phys_enc) +static bool dpu_encoder_phys_cmd_is_master(struct dpu_encoder_phys *phys_enc) { return (phys_enc->split_role != ENC_ROLE_SLAVE) ? true : false; } @@ -243,7 +236,6 @@ static int _dpu_encoder_phys_cmd_handle_ppdone_timeout( atomic_read(&phys_enc->pending_kickoff_cnt)); dpu_encoder_helper_unregister_irq(phys_enc, INTR_IDX_RDPTR); - dpu_dbg_dump(false, __func__, true, true); } atomic_add_unless(&phys_enc->pending_kickoff_cnt, -1, 0); @@ -496,14 +488,11 @@ static void dpu_encoder_phys_cmd_enable_helper( _dpu_encoder_phys_cmd_pingpong_config(phys_enc); if (!dpu_encoder_phys_cmd_is_master(phys_enc)) - goto skip_flush; + return; ctl = phys_enc->hw_ctl; ctl->ops.get_bitmask_intf(ctl, &flush_mask, phys_enc->intf_idx); ctl->ops.update_pending_flush(ctl, flush_mask); - -skip_flush: - return; } static void dpu_encoder_phys_cmd_enable(struct dpu_encoder_phys *phys_enc) @@ -727,7 +716,7 @@ static int dpu_encoder_phys_cmd_wait_for_vblank( wait_info.wq = &cmd_enc->pending_vblank_wq; wait_info.atomic_cnt = &cmd_enc->pending_vblank_cnt; - wait_info.timeout_ms = _dpu_encoder_phys_cmd_get_idle_timeout(cmd_enc); + wait_info.timeout_ms = KICKOFF_TIMEOUT_MS; atomic_inc(&cmd_enc->pending_vblank_cnt); @@ -776,7 +765,6 @@ static void dpu_encoder_phys_cmd_init_ops( ops->wait_for_vblank = dpu_encoder_phys_cmd_wait_for_vblank; ops->trigger_start = dpu_encoder_phys_cmd_trigger_start; ops->needs_single_flush = dpu_encoder_phys_cmd_needs_single_flush; - ops->hw_reset = dpu_encoder_helper_hw_reset; ops->irq_control = dpu_encoder_phys_cmd_irq_control; ops->restore = dpu_encoder_phys_cmd_enable_helper; ops->prepare_idle_pc = dpu_encoder_phys_cmd_prepare_idle_pc; @@ -798,7 +786,7 @@ struct dpu_encoder_phys *dpu_encoder_phys_cmd_init( if (!cmd_enc) { ret = -ENOMEM; DPU_ERROR("failed to allocate\n"); - goto fail; + return ERR_PTR(ret); } phys_enc = &cmd_enc->base; phys_enc->hw_mdptop = p->dpu_kms->hw_mdp; @@ -856,6 +844,5 @@ struct dpu_encoder_phys *dpu_encoder_phys_cmd_init( return phys_enc; -fail: return ERR_PTR(ret); } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c index 84de385a9f62..acdab5b0db18 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c @@ -110,7 +110,7 @@ static void drm_mode_to_intf_timing_params( */ } -static inline u32 get_horizontal_total(const struct intf_timing_params *timing) +static u32 get_horizontal_total(const struct intf_timing_params *timing) { u32 active = timing->xres; u32 inactive = @@ -119,7 +119,7 @@ static inline u32 get_horizontal_total(const struct intf_timing_params *timing) return active + inactive; } -static inline u32 get_vertical_total(const struct intf_timing_params *timing) +static u32 get_vertical_total(const struct intf_timing_params *timing) { u32 active = timing->yres; u32 inactive = @@ -331,7 +331,7 @@ static void dpu_encoder_phys_vid_vblank_irq(void *arg, int irq_idx) if (hw_ctl && hw_ctl->ops.get_flush_register) flush_register = hw_ctl->ops.get_flush_register(hw_ctl); - if (flush_register == 0) + if (!(flush_register & hw_ctl->ops.get_pending_flush(hw_ctl))) new_cnt = atomic_add_unless(&phys_enc->pending_kickoff_cnt, -1, 0); spin_unlock_irqrestore(phys_enc->enc_spinlock, lock_flags); @@ -613,7 +613,6 @@ static void dpu_encoder_phys_vid_prepare_for_kickoff( DPU_ERROR_VIDENC(vid_enc, "ctl %d reset failure: %d\n", ctl->idx, rc); dpu_encoder_helper_unregister_irq(phys_enc, INTR_IDX_VSYNC); - dpu_dbg_dump(false, __func__, true, true); } } @@ -766,7 +765,6 @@ static void dpu_encoder_phys_vid_init_ops(struct dpu_encoder_phys_ops *ops) ops->prepare_for_kickoff = dpu_encoder_phys_vid_prepare_for_kickoff; ops->handle_post_kickoff = dpu_encoder_phys_vid_handle_post_kickoff; ops->needs_single_flush = dpu_encoder_phys_vid_needs_single_flush; - ops->hw_reset = dpu_encoder_helper_hw_reset; ops->get_line_count = dpu_encoder_phys_vid_get_line_count; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c index bfcd165e96df..0874f0a53bf9 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_formats.c @@ -216,7 +216,7 @@ static const struct dpu_format dpu_format_map[] = { INTERLEAVED_RGB_FMT(XBGR8888, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, COLOR_8BIT, C2_R_Cr, C0_G_Y, C1_B_Cb, C3_ALPHA, 4, - true, 4, 0, + false, 4, 0, DPU_FETCH_LINEAR, 1), INTERLEAVED_RGB_FMT(RGBA8888, @@ -921,7 +921,7 @@ static int _dpu_format_populate_addrs_ubwc( + layout->plane_size[2] + layout->plane_size[3]; if (!meta) - goto done; + return 0; /* configure Y metadata plane */ layout->plane_addr[2] = base_addr; @@ -952,12 +952,11 @@ static int _dpu_format_populate_addrs_ubwc( layout->plane_addr[1] = 0; if (!meta) - goto done; + return 0; layout->plane_addr[2] = base_addr; layout->plane_addr[3] = 0; } -done: return 0; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_blk.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_blk.c index 58d29e43faef..92f1c4241b9a 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_blk.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_blk.c @@ -30,16 +30,10 @@ static LIST_HEAD(dpu_hw_blk_list); * @type: hw block type - enum dpu_hw_blk_type * @id: instance id of the hw block * @ops: Pointer to block operations - * return: 0 if success; error code otherwise */ -int dpu_hw_blk_init(struct dpu_hw_blk *hw_blk, u32 type, int id, +void dpu_hw_blk_init(struct dpu_hw_blk *hw_blk, u32 type, int id, struct dpu_hw_blk_ops *ops) { - if (!hw_blk) { - pr_err("invalid parameters\n"); - return -EINVAL; - } - INIT_LIST_HEAD(&hw_blk->list); hw_blk->type = type; hw_blk->id = id; @@ -51,8 +45,6 @@ int dpu_hw_blk_init(struct dpu_hw_blk *hw_blk, u32 type, int id, mutex_lock(&dpu_hw_blk_lock); list_add(&hw_blk->list, &dpu_hw_blk_list); mutex_unlock(&dpu_hw_blk_lock); - - return 0; } /** diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_blk.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_blk.h index 0f4ca8af1ec5..1934c2f7e8fa 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_blk.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_blk.h @@ -44,7 +44,7 @@ struct dpu_hw_blk { struct dpu_hw_blk_ops ops; }; -int dpu_hw_blk_init(struct dpu_hw_blk *hw_blk, u32 type, int id, +void dpu_hw_blk_init(struct dpu_hw_blk *hw_blk, u32 type, int id, struct dpu_hw_blk_ops *ops); void dpu_hw_blk_destroy(struct dpu_hw_blk *hw_blk); diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h index dc060e7358e4..144358a3d0fb 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_catalog.h @@ -736,13 +736,4 @@ struct dpu_mdss_cfg *dpu_hw_catalog_init(u32 hw_rev); */ void dpu_hw_catalog_deinit(struct dpu_mdss_cfg *dpu_cfg); -/** - * dpu_hw_sspp_multirect_enabled - check multirect enabled for the sspp - * @cfg: pointer to sspp cfg - */ -static inline bool dpu_hw_sspp_multirect_enabled(const struct dpu_sspp_cfg *cfg) -{ - return test_bit(DPU_SSPP_SMART_DMA_V1, &cfg->features) || - test_bit(DPU_SSPP_SMART_DMA_V2, &cfg->features); -} #endif /* _DPU_HW_CATALOG_H */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c index eec1051f2afc..1068b4b7940f 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_ctl.c @@ -13,8 +13,8 @@ #include <linux/delay.h> #include "dpu_hwio.h" #include "dpu_hw_ctl.h" -#include "dpu_dbg.h" #include "dpu_kms.h" +#include "dpu_trace.h" #define CTL_LAYER(lm) \ (((lm) == LM_5) ? (0x024) : (((lm) - LM_0) * 0x004)) @@ -72,24 +72,39 @@ static int _mixer_stages(const struct dpu_lm_cfg *mixer, int count, return stages; } +static inline u32 dpu_hw_ctl_get_flush_register(struct dpu_hw_ctl *ctx) +{ + struct dpu_hw_blk_reg_map *c = &ctx->hw; + + return DPU_REG_READ(c, CTL_FLUSH); +} + static inline void dpu_hw_ctl_trigger_start(struct dpu_hw_ctl *ctx) { + trace_dpu_hw_ctl_trigger_start(ctx->pending_flush_mask, + dpu_hw_ctl_get_flush_register(ctx)); DPU_REG_WRITE(&ctx->hw, CTL_START, 0x1); } static inline void dpu_hw_ctl_trigger_pending(struct dpu_hw_ctl *ctx) { + trace_dpu_hw_ctl_trigger_prepare(ctx->pending_flush_mask, + dpu_hw_ctl_get_flush_register(ctx)); DPU_REG_WRITE(&ctx->hw, CTL_PREPARE, 0x1); } static inline void dpu_hw_ctl_clear_pending_flush(struct dpu_hw_ctl *ctx) { + trace_dpu_hw_ctl_clear_pending_flush(ctx->pending_flush_mask, + dpu_hw_ctl_get_flush_register(ctx)); ctx->pending_flush_mask = 0x0; } static inline void dpu_hw_ctl_update_pending_flush(struct dpu_hw_ctl *ctx, u32 flushbits) { + trace_dpu_hw_ctl_update_pending_flush(flushbits, + ctx->pending_flush_mask); ctx->pending_flush_mask |= flushbits; } @@ -103,18 +118,12 @@ static u32 dpu_hw_ctl_get_pending_flush(struct dpu_hw_ctl *ctx) static inline void dpu_hw_ctl_trigger_flush(struct dpu_hw_ctl *ctx) { - + trace_dpu_hw_ctl_trigger_pending_flush(ctx->pending_flush_mask, + dpu_hw_ctl_get_flush_register(ctx)); DPU_REG_WRITE(&ctx->hw, CTL_FLUSH, ctx->pending_flush_mask); } -static inline u32 dpu_hw_ctl_get_flush_register(struct dpu_hw_ctl *ctx) -{ - struct dpu_hw_blk_reg_map *c = &ctx->hw; - - return DPU_REG_READ(c, CTL_FLUSH); -} - -static inline uint32_t dpu_hw_ctl_get_bitmask_sspp(struct dpu_hw_ctl *ctx, +static uint32_t dpu_hw_ctl_get_bitmask_sspp(struct dpu_hw_ctl *ctx, enum dpu_sspp sspp) { uint32_t flushbits = 0; @@ -169,7 +178,7 @@ static inline uint32_t dpu_hw_ctl_get_bitmask_sspp(struct dpu_hw_ctl *ctx, return flushbits; } -static inline uint32_t dpu_hw_ctl_get_bitmask_mixer(struct dpu_hw_ctl *ctx, +static uint32_t dpu_hw_ctl_get_bitmask_mixer(struct dpu_hw_ctl *ctx, enum dpu_lm lm) { uint32_t flushbits = 0; @@ -202,7 +211,7 @@ static inline uint32_t dpu_hw_ctl_get_bitmask_mixer(struct dpu_hw_ctl *ctx, return flushbits; } -static inline int dpu_hw_ctl_get_bitmask_intf(struct dpu_hw_ctl *ctx, +static int dpu_hw_ctl_get_bitmask_intf(struct dpu_hw_ctl *ctx, u32 *flushbits, enum dpu_intf intf) { switch (intf) { @@ -474,10 +483,7 @@ static void _setup_ctl_ops(struct dpu_hw_ctl_ops *ops, ops->get_bitmask_intf = dpu_hw_ctl_get_bitmask_intf; }; -static struct dpu_hw_blk_ops dpu_hw_ops = { - .start = NULL, - .stop = NULL, -}; +static struct dpu_hw_blk_ops dpu_hw_ops; struct dpu_hw_ctl *dpu_hw_ctl_init(enum dpu_ctl idx, void __iomem *addr, @@ -485,7 +491,6 @@ struct dpu_hw_ctl *dpu_hw_ctl_init(enum dpu_ctl idx, { struct dpu_hw_ctl *c; struct dpu_ctl_cfg *cfg; - int rc; c = kzalloc(sizeof(*c), GFP_KERNEL); if (!c) @@ -504,18 +509,9 @@ struct dpu_hw_ctl *dpu_hw_ctl_init(enum dpu_ctl idx, c->mixer_count = m->mixer_count; c->mixer_hw_caps = m->mixer; - rc = dpu_hw_blk_init(&c->base, DPU_HW_BLK_CTL, idx, &dpu_hw_ops); - if (rc) { - DPU_ERROR("failed to init hw blk %d\n", rc); - goto blk_init_error; - } + dpu_hw_blk_init(&c->base, DPU_HW_BLK_CTL, idx, &dpu_hw_ops); return c; - -blk_init_error: - kzfree(c); - - return ERR_PTR(rc); } void dpu_hw_ctl_destroy(struct dpu_hw_ctl *ctx) diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c index 9c6bba0ac7c3..f6a83daa385b 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.c @@ -13,7 +13,6 @@ #include "dpu_hwio.h" #include "dpu_hw_catalog.h" #include "dpu_hw_intf.h" -#include "dpu_dbg.h" #include "dpu_kms.h" #define INTF_TIMING_ENGINE_EN 0x000 @@ -265,10 +264,7 @@ static void _setup_intf_ops(struct dpu_hw_intf_ops *ops, ops->get_line_count = dpu_hw_intf_get_line_count; } -static struct dpu_hw_blk_ops dpu_hw_ops = { - .start = NULL, - .stop = NULL, -}; +static struct dpu_hw_blk_ops dpu_hw_ops; struct dpu_hw_intf *dpu_hw_intf_init(enum dpu_intf idx, void __iomem *addr, @@ -276,7 +272,6 @@ struct dpu_hw_intf *dpu_hw_intf_init(enum dpu_intf idx, { struct dpu_hw_intf *c; struct dpu_intf_cfg *cfg; - int rc; c = kzalloc(sizeof(*c), GFP_KERNEL); if (!c) @@ -297,18 +292,9 @@ struct dpu_hw_intf *dpu_hw_intf_init(enum dpu_intf idx, c->mdss = m; _setup_intf_ops(&c->ops, c->cap->features); - rc = dpu_hw_blk_init(&c->base, DPU_HW_BLK_INTF, idx, &dpu_hw_ops); - if (rc) { - DPU_ERROR("failed to init hw blk %d\n", rc); - goto blk_init_error; - } + dpu_hw_blk_init(&c->base, DPU_HW_BLK_INTF, idx, &dpu_hw_ops); return c; - -blk_init_error: - kzfree(c); - - return ERR_PTR(rc); } void dpu_hw_intf_destroy(struct dpu_hw_intf *intf) diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h index 3b77df460dea..a2b0dbc23058 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_intf.h @@ -92,16 +92,6 @@ struct dpu_hw_intf { }; /** - * to_dpu_hw_intf - convert base object dpu_hw_base to container - * @hw: Pointer to base hardware block - * return: Pointer to hardware block container - */ -static inline struct dpu_hw_intf *to_dpu_hw_intf(struct dpu_hw_blk *hw) -{ - return container_of(hw, struct dpu_hw_intf, base); -} - -/** * dpu_hw_intf_init(): Initializes the intf driver for the passed * interface idx. * @idx: interface index for which driver object is required diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c index acb8dc8acaa5..018df2c3b7ed 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.c @@ -15,7 +15,6 @@ #include "dpu_hwio.h" #include "dpu_hw_lm.h" #include "dpu_hw_mdss.h" -#include "dpu_dbg.h" #include "dpu_kms.h" #define LM_OP_MODE 0x00 @@ -64,16 +63,10 @@ static struct dpu_lm_cfg *_lm_offset(enum dpu_lm mixer, static inline int _stage_offset(struct dpu_hw_mixer *ctx, enum dpu_stage stage) { const struct dpu_lm_sub_blks *sblk = ctx->cap->sblk; - int rc; + if (stage != DPU_STAGE_BASE && stage <= sblk->maxblendstages) + return sblk->blendstage_base[stage - DPU_STAGE_0]; - if (stage == DPU_STAGE_BASE) - rc = -EINVAL; - else if (stage <= sblk->maxblendstages) - rc = sblk->blendstage_base[stage - DPU_STAGE_0]; - else - rc = -EINVAL; - - return rc; + return -EINVAL; } static void dpu_hw_lm_setup_out(struct dpu_hw_mixer *ctx, @@ -163,11 +156,6 @@ static void dpu_hw_lm_setup_color3(struct dpu_hw_mixer *ctx, DPU_REG_WRITE(c, LM_OP_MODE, op_mode); } -static void dpu_hw_lm_gc(struct dpu_hw_mixer *mixer, - void *cfg) -{ -} - static void _setup_mixer_ops(struct dpu_mdss_cfg *m, struct dpu_hw_lm_ops *ops, unsigned long features) @@ -179,13 +167,9 @@ static void _setup_mixer_ops(struct dpu_mdss_cfg *m, ops->setup_blend_config = dpu_hw_lm_setup_blend_config; ops->setup_alpha_out = dpu_hw_lm_setup_color3; ops->setup_border_color = dpu_hw_lm_setup_border_color; - ops->setup_gc = dpu_hw_lm_gc; }; -static struct dpu_hw_blk_ops dpu_hw_ops = { - .start = NULL, - .stop = NULL, -}; +static struct dpu_hw_blk_ops dpu_hw_ops; struct dpu_hw_mixer *dpu_hw_lm_init(enum dpu_lm idx, void __iomem *addr, @@ -193,7 +177,6 @@ struct dpu_hw_mixer *dpu_hw_lm_init(enum dpu_lm idx, { struct dpu_hw_mixer *c; struct dpu_lm_cfg *cfg; - int rc; c = kzalloc(sizeof(*c), GFP_KERNEL); if (!c) @@ -210,18 +193,9 @@ struct dpu_hw_mixer *dpu_hw_lm_init(enum dpu_lm idx, c->cap = cfg; _setup_mixer_ops(m, &c->ops, c->cap->features); - rc = dpu_hw_blk_init(&c->base, DPU_HW_BLK_LM, idx, &dpu_hw_ops); - if (rc) { - DPU_ERROR("failed to init hw blk %d\n", rc); - goto blk_init_error; - } + dpu_hw_blk_init(&c->base, DPU_HW_BLK_LM, idx, &dpu_hw_ops); return c; - -blk_init_error: - kzfree(c); - - return ERR_PTR(rc); } void dpu_hw_lm_destroy(struct dpu_hw_mixer *lm) diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h index 5b036aca8340..6aee839a6a23 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_lm.h @@ -61,11 +61,6 @@ struct dpu_hw_lm_ops { void (*setup_border_color)(struct dpu_hw_mixer *ctx, struct dpu_mdss_color *color, u8 border_en); - /** - * setup_gc : enable/disable gamma correction feature - */ - void (*setup_gc)(struct dpu_hw_mixer *mixer, - void *cfg); }; struct dpu_hw_mixer { diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c index cc3a623903f4..3bdf47ed1845 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.c @@ -16,7 +16,6 @@ #include "dpu_hwio.h" #include "dpu_hw_catalog.h" #include "dpu_hw_pingpong.h" -#include "dpu_dbg.h" #include "dpu_kms.h" #include "dpu_trace.h" @@ -177,7 +176,7 @@ static u32 dpu_hw_pp_get_line_count(struct dpu_hw_pingpong *pp) height = DPU_REG_READ(c, PP_SYNC_CONFIG_HEIGHT) & 0xFFFF; if (height < init) - goto line_count_exit; + return line; line = DPU_REG_READ(c, PP_INT_COUNT_VAL) & 0xFFFF; @@ -186,7 +185,6 @@ static u32 dpu_hw_pp_get_line_count(struct dpu_hw_pingpong *pp) else line -= init; -line_count_exit: return line; } @@ -201,10 +199,7 @@ static void _setup_pingpong_ops(struct dpu_hw_pingpong_ops *ops, ops->get_line_count = dpu_hw_pp_get_line_count; }; -static struct dpu_hw_blk_ops dpu_hw_ops = { - .start = NULL, - .stop = NULL, -}; +static struct dpu_hw_blk_ops dpu_hw_ops; struct dpu_hw_pingpong *dpu_hw_pingpong_init(enum dpu_pingpong idx, void __iomem *addr, @@ -212,7 +207,6 @@ struct dpu_hw_pingpong *dpu_hw_pingpong_init(enum dpu_pingpong idx, { struct dpu_hw_pingpong *c; struct dpu_pingpong_cfg *cfg; - int rc; c = kzalloc(sizeof(*c), GFP_KERNEL); if (!c) @@ -228,18 +222,9 @@ struct dpu_hw_pingpong *dpu_hw_pingpong_init(enum dpu_pingpong idx, c->caps = cfg; _setup_pingpong_ops(&c->ops, c->caps); - rc = dpu_hw_blk_init(&c->base, DPU_HW_BLK_PINGPONG, idx, &dpu_hw_ops); - if (rc) { - DPU_ERROR("failed to init hw blk %d\n", rc); - goto blk_init_error; - } + dpu_hw_blk_init(&c->base, DPU_HW_BLK_PINGPONG, idx, &dpu_hw_ops); return c; - -blk_init_error: - kzfree(c); - - return ERR_PTR(rc); } void dpu_hw_pingpong_destroy(struct dpu_hw_pingpong *pp) diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h index 3caccd7d6a3e..0e02e43cee14 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_pingpong.h @@ -105,16 +105,6 @@ struct dpu_hw_pingpong { }; /** - * dpu_hw_pingpong - convert base object dpu_hw_base to container - * @hw: Pointer to base hardware block - * return: Pointer to hardware block container - */ -static inline struct dpu_hw_pingpong *to_dpu_hw_pingpong(struct dpu_hw_blk *hw) -{ - return container_of(hw, struct dpu_hw_pingpong, base); -} - -/** * dpu_hw_pingpong_init - initializes the pingpong driver for the passed * pingpong idx. * @idx: Pingpong index for which driver object is required diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c index c25b52a6b219..e9132bf5166b 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.c @@ -14,7 +14,6 @@ #include "dpu_hw_catalog.h" #include "dpu_hw_lm.h" #include "dpu_hw_sspp.h" -#include "dpu_dbg.h" #include "dpu_kms.h" #define DPU_FETCH_CONFIG_RESET_VALUE 0x00000087 @@ -141,7 +140,7 @@ /* traffic shaper clock in Hz */ #define TS_CLK 19200000 -static inline int _sspp_subblk_offset(struct dpu_hw_pipe *ctx, +static int _sspp_subblk_offset(struct dpu_hw_pipe *ctx, int s_id, u32 *idx) { @@ -662,7 +661,8 @@ static void _setup_layer_ops(struct dpu_hw_pipe *c, test_bit(DPU_SSPP_CSC_10BIT, &features)) c->ops.setup_csc = dpu_hw_sspp_setup_csc; - if (dpu_hw_sspp_multirect_enabled(c->cap)) + if (test_bit(DPU_SSPP_SMART_DMA_V1, &c->cap->features) || + test_bit(DPU_SSPP_SMART_DMA_V2, &c->cap->features)) c->ops.setup_multirect = dpu_hw_sspp_setup_multirect; if (test_bit(DPU_SSPP_SCALER_QSEED3, &features)) { @@ -697,10 +697,7 @@ static struct dpu_sspp_cfg *_sspp_offset(enum dpu_sspp sspp, return ERR_PTR(-ENOMEM); } -static struct dpu_hw_blk_ops dpu_hw_ops = { - .start = NULL, - .stop = NULL, -}; +static struct dpu_hw_blk_ops dpu_hw_ops; struct dpu_hw_pipe *dpu_hw_sspp_init(enum dpu_sspp idx, void __iomem *addr, struct dpu_mdss_cfg *catalog, @@ -708,7 +705,6 @@ struct dpu_hw_pipe *dpu_hw_sspp_init(enum dpu_sspp idx, { struct dpu_hw_pipe *hw_pipe; struct dpu_sspp_cfg *cfg; - int rc; if (!addr || !catalog) return ERR_PTR(-EINVAL); @@ -730,18 +726,9 @@ struct dpu_hw_pipe *dpu_hw_sspp_init(enum dpu_sspp idx, hw_pipe->cap = cfg; _setup_layer_ops(hw_pipe, hw_pipe->cap->features); - rc = dpu_hw_blk_init(&hw_pipe->base, DPU_HW_BLK_SSPP, idx, &dpu_hw_ops); - if (rc) { - DPU_ERROR("failed to init hw blk %d\n", rc); - goto blk_init_error; - } + dpu_hw_blk_init(&hw_pipe->base, DPU_HW_BLK_SSPP, idx, &dpu_hw_ops); return hw_pipe; - -blk_init_error: - kzfree(hw_pipe); - - return ERR_PTR(rc); } void dpu_hw_sspp_destroy(struct dpu_hw_pipe *ctx) diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h index 4d81e5f5ce1b..119b4e1c16be 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_sspp.h @@ -392,16 +392,6 @@ struct dpu_hw_pipe { }; /** - * dpu_hw_pipe - convert base object dpu_hw_base to container - * @hw: Pointer to base hardware block - * return: Pointer to hardware block container - */ -static inline struct dpu_hw_pipe *to_dpu_hw_pipe(struct dpu_hw_blk *hw) -{ - return container_of(hw, struct dpu_hw_pipe, base); -} - -/** * dpu_hw_sspp_init - initializes the sspp hw driver object. * Should be called once before accessing every pipe. * @idx: Pipe index for which driver object is required diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c index b8781256e21b..a041597bb849 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.c @@ -13,7 +13,6 @@ #include "dpu_hwio.h" #include "dpu_hw_catalog.h" #include "dpu_hw_top.h" -#include "dpu_dbg.h" #include "dpu_kms.h" #define SSPP_SPARE 0x28 @@ -322,10 +321,7 @@ static const struct dpu_mdp_cfg *_top_offset(enum dpu_mdp mdp, return ERR_PTR(-EINVAL); } -static struct dpu_hw_blk_ops dpu_hw_ops = { - .start = NULL, - .stop = NULL, -}; +static struct dpu_hw_blk_ops dpu_hw_ops; struct dpu_hw_mdp *dpu_hw_mdptop_init(enum dpu_mdp idx, void __iomem *addr, @@ -333,7 +329,6 @@ struct dpu_hw_mdp *dpu_hw_mdptop_init(enum dpu_mdp idx, { struct dpu_hw_mdp *mdp; const struct dpu_mdp_cfg *cfg; - int rc; if (!addr || !m) return ERR_PTR(-EINVAL); @@ -355,20 +350,9 @@ struct dpu_hw_mdp *dpu_hw_mdptop_init(enum dpu_mdp idx, mdp->caps = cfg; _setup_mdp_ops(&mdp->ops, mdp->caps->features); - rc = dpu_hw_blk_init(&mdp->base, DPU_HW_BLK_TOP, idx, &dpu_hw_ops); - if (rc) { - DPU_ERROR("failed to init hw blk %d\n", rc); - goto blk_init_error; - } - - dpu_dbg_set_dpu_top_offset(mdp->hw.blk_off); + dpu_hw_blk_init(&mdp->base, DPU_HW_BLK_TOP, idx, &dpu_hw_ops); return mdp; - -blk_init_error: - kzfree(mdp); - - return ERR_PTR(rc); } void dpu_hw_mdp_destroy(struct dpu_hw_mdp *mdp) diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.h index 192e338f20bb..aa21fd834398 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_top.h @@ -161,16 +161,6 @@ struct dpu_hw_mdp { }; /** - * to_dpu_hw_mdp - convert base object dpu_hw_base to container - * @hw: Pointer to base hardware block - * return: Pointer to hardware block container - */ -static inline struct dpu_hw_mdp *to_dpu_hw_mdp(struct dpu_hw_blk *hw) -{ - return container_of(hw, struct dpu_hw_mdp, base); -} - -/** * dpu_hw_mdptop_init - initializes the top driver for the passed idx * @idx: Interface index for which driver object is required * @addr: Mapped register io address of MDP diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c index d43905525f92..38bfd222ed72 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_hw_vbif.c @@ -13,7 +13,6 @@ #include "dpu_hwio.h" #include "dpu_hw_catalog.h" #include "dpu_hw_vbif.h" -#include "dpu_dbg.h" #define VBIF_VERSION 0x0000 #define VBIF_CLK_FORCE_CTRL0 0x0008 diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_io_util.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_io_util.c index b557687b1964..78833c2c27f8 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_io_util.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_io_util.c @@ -16,6 +16,8 @@ #include <linux/err.h> #include <linux/delay.h> +#include <drm/drm_print.h> + #include "dpu_io_util.h" void msm_dss_put_clk(struct dss_clk *clk_arry, int num_clk) @@ -164,7 +166,7 @@ int msm_dss_parse_clock(struct platform_device *pdev, "clock-names", i, &clock_name); if (rc) { - dev_err(&pdev->dev, "Failed to get clock name for %d\n", + DRM_DEV_ERROR(&pdev->dev, "Failed to get clock name for %d\n", i); break; } @@ -176,13 +178,13 @@ int msm_dss_parse_clock(struct platform_device *pdev, rc = msm_dss_get_clk(&pdev->dev, mp->clk_config, num_clk); if (rc) { - dev_err(&pdev->dev, "Failed to get clock refs %d\n", rc); + DRM_DEV_ERROR(&pdev->dev, "Failed to get clock refs %d\n", rc); goto err; } rc = of_clk_set_defaults(pdev->dev.of_node, false); if (rc) { - dev_err(&pdev->dev, "Failed to set clock defaults %d\n", rc); + DRM_DEV_ERROR(&pdev->dev, "Failed to set clock defaults %d\n", rc); goto err; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_irq.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_irq.c deleted file mode 100644 index d5e6ce0140cf..000000000000 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_irq.c +++ /dev/null @@ -1,66 +0,0 @@ -/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 and - * only version 2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - */ - -#define pr_fmt(fmt) "[drm:%s:%d] " fmt, __func__, __LINE__ - -#include <linux/irqdomain.h> -#include <linux/irq.h> -#include <linux/kthread.h> - -#include "dpu_irq.h" -#include "dpu_core_irq.h" - -irqreturn_t dpu_irq(struct msm_kms *kms) -{ - struct dpu_kms *dpu_kms = to_dpu_kms(kms); - - return dpu_core_irq(dpu_kms); -} - -void dpu_irq_preinstall(struct msm_kms *kms) -{ - struct dpu_kms *dpu_kms = to_dpu_kms(kms); - - if (!dpu_kms->dev || !dpu_kms->dev->dev) { - pr_err("invalid device handles\n"); - return; - } - - dpu_core_irq_preinstall(dpu_kms); -} - -int dpu_irq_postinstall(struct msm_kms *kms) -{ - struct dpu_kms *dpu_kms = to_dpu_kms(kms); - int rc; - - if (!kms) { - DPU_ERROR("invalid parameters\n"); - return -EINVAL; - } - - rc = dpu_core_irq_postinstall(dpu_kms); - - return rc; -} - -void dpu_irq_uninstall(struct msm_kms *kms) -{ - struct dpu_kms *dpu_kms = to_dpu_kms(kms); - - if (!kms) { - DPU_ERROR("invalid parameters\n"); - return; - } - - dpu_core_irq_uninstall(dpu_kms); -} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_irq.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_irq.h deleted file mode 100644 index 3e147f7176e2..000000000000 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_irq.h +++ /dev/null @@ -1,59 +0,0 @@ -/* Copyright (c) 2015-2018, The Linux Foundation. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 and - * only version 2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - */ - -#ifndef __DPU_IRQ_H__ -#define __DPU_IRQ_H__ - -#include <linux/kernel.h> -#include <linux/irqdomain.h> - -#include "msm_kms.h" - -/** - * dpu_irq_controller - define MDSS level interrupt controller context - * @enabled_mask: enable status of MDSS level interrupt - * @domain: interrupt domain of this controller - */ -struct dpu_irq_controller { - unsigned long enabled_mask; - struct irq_domain *domain; -}; - -/** - * dpu_irq_preinstall - perform pre-installation of MDSS IRQ handler - * @kms: pointer to kms context - * @return: none - */ -void dpu_irq_preinstall(struct msm_kms *kms); - -/** - * dpu_irq_postinstall - perform post-installation of MDSS IRQ handler - * @kms: pointer to kms context - * @return: 0 if success; error code otherwise - */ -int dpu_irq_postinstall(struct msm_kms *kms); - -/** - * dpu_irq_uninstall - uninstall MDSS IRQ handler - * @drm_dev: pointer to kms context - * @return: none - */ -void dpu_irq_uninstall(struct msm_kms *kms); - -/** - * dpu_irq - MDSS level IRQ handler - * @kms: pointer to kms context - * @return: interrupt handling status - */ -irqreturn_t dpu_irq(struct msm_kms *kms); - -#endif /* __DPU_IRQ_H__ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c index 0a683e65a9f3..4d67b3c96702 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c @@ -81,7 +81,7 @@ static int _dpu_danger_signal_status(struct seq_file *s, struct dpu_danger_safe_status status; int i; - if (!kms || !kms->dev || !kms->dev->dev_private || !kms->hw_mdp) { + if (!kms->dev || !kms->dev->dev_private || !kms->hw_mdp) { DPU_ERROR("invalid arg(s)\n"); return 0; } @@ -138,46 +138,29 @@ static int dpu_debugfs_safe_stats_show(struct seq_file *s, void *v) } DEFINE_DPU_DEBUGFS_SEQ_FOPS(dpu_debugfs_safe_stats); -static void dpu_debugfs_danger_destroy(struct dpu_kms *dpu_kms) -{ - debugfs_remove_recursive(dpu_kms->debugfs_danger); - dpu_kms->debugfs_danger = NULL; -} - -static int dpu_debugfs_danger_init(struct dpu_kms *dpu_kms, +static void dpu_debugfs_danger_init(struct dpu_kms *dpu_kms, struct dentry *parent) { - dpu_kms->debugfs_danger = debugfs_create_dir("danger", - parent); - if (!dpu_kms->debugfs_danger) { - DPU_ERROR("failed to create danger debugfs\n"); - return -EINVAL; - } + struct dentry *entry = debugfs_create_dir("danger", parent); + if (IS_ERR_OR_NULL(entry)) + return; - debugfs_create_file("danger_status", 0600, dpu_kms->debugfs_danger, + debugfs_create_file("danger_status", 0600, entry, dpu_kms, &dpu_debugfs_danger_stats_fops); - debugfs_create_file("safe_status", 0600, dpu_kms->debugfs_danger, + debugfs_create_file("safe_status", 0600, entry, dpu_kms, &dpu_debugfs_safe_stats_fops); - - return 0; } static int _dpu_debugfs_show_regset32(struct seq_file *s, void *data) { - struct dpu_debugfs_regset32 *regset; - struct dpu_kms *dpu_kms; + struct dpu_debugfs_regset32 *regset = s->private; + struct dpu_kms *dpu_kms = regset->dpu_kms; struct drm_device *dev; struct msm_drm_private *priv; void __iomem *base; uint32_t i, addr; - if (!s || !s->private) - return 0; - - regset = s->private; - - dpu_kms = regset->dpu_kms; - if (!dpu_kms || !dpu_kms->mmio) + if (!dpu_kms->mmio) return 0; dev = dpu_kms->dev; @@ -250,57 +233,24 @@ void *dpu_debugfs_create_regset32(const char *name, umode_t mode, static int _dpu_debugfs_init(struct dpu_kms *dpu_kms) { - void *p; - int rc; - - p = dpu_hw_util_get_log_mask_ptr(); + void *p = dpu_hw_util_get_log_mask_ptr(); + struct dentry *entry; - if (!dpu_kms || !p) + if (!p) return -EINVAL; - dpu_kms->debugfs_root = debugfs_create_dir("debug", - dpu_kms->dev->primary->debugfs_root); - if (IS_ERR_OR_NULL(dpu_kms->debugfs_root)) { - DRM_ERROR("debugfs create_dir failed %ld\n", - PTR_ERR(dpu_kms->debugfs_root)); - return PTR_ERR(dpu_kms->debugfs_root); - } - - rc = dpu_dbg_debugfs_register(dpu_kms->debugfs_root); - if (rc) { - DRM_ERROR("failed to reg dpu dbg debugfs: %d\n", rc); - return rc; - } + entry = debugfs_create_dir("debug", dpu_kms->dev->primary->debugfs_root); + if (IS_ERR_OR_NULL(entry)) + return -ENODEV; /* allow root to be NULL */ - debugfs_create_x32(DPU_DEBUGFS_HWMASKNAME, 0600, dpu_kms->debugfs_root, p); - - (void) dpu_debugfs_danger_init(dpu_kms, dpu_kms->debugfs_root); - (void) dpu_debugfs_vbif_init(dpu_kms, dpu_kms->debugfs_root); - (void) dpu_debugfs_core_irq_init(dpu_kms, dpu_kms->debugfs_root); - - rc = dpu_core_perf_debugfs_init(&dpu_kms->perf, dpu_kms->debugfs_root); - if (rc) { - DPU_ERROR("failed to init perf %d\n", rc); - return rc; - } + debugfs_create_x32(DPU_DEBUGFS_HWMASKNAME, 0600, entry, p); - return 0; -} + dpu_debugfs_danger_init(dpu_kms, entry); + dpu_debugfs_vbif_init(dpu_kms, entry); + dpu_debugfs_core_irq_init(dpu_kms, entry); -static void _dpu_debugfs_destroy(struct dpu_kms *dpu_kms) -{ - /* don't need to NULL check debugfs_root */ - if (dpu_kms) { - dpu_debugfs_vbif_destroy(dpu_kms); - dpu_debugfs_danger_destroy(dpu_kms); - dpu_debugfs_core_irq_destroy(dpu_kms); - debugfs_remove_recursive(dpu_kms->debugfs_root); - } -} -#else -static void _dpu_debugfs_destroy(struct dpu_kms *dpu_kms) -{ + return dpu_core_perf_debugfs_init(dpu_kms, entry); } #endif @@ -320,7 +270,10 @@ static void dpu_kms_prepare_commit(struct msm_kms *kms, struct dpu_kms *dpu_kms; struct msm_drm_private *priv; struct drm_device *dev; + struct drm_crtc *crtc; + struct drm_crtc_state *crtc_state; struct drm_encoder *encoder; + int i; if (!kms) return; @@ -332,9 +285,13 @@ static void dpu_kms_prepare_commit(struct msm_kms *kms, priv = dev->dev_private; pm_runtime_get_sync(&dpu_kms->pdev->dev); - list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) - if (encoder->crtc != NULL) + /* Call prepare_commit for all affected encoders */ + for_each_new_crtc_in_state(state, crtc, crtc_state, i) { + drm_for_each_encoder_mask(encoder, crtc->dev, + crtc_state->encoder_mask) { dpu_encoder_prepare_commit(encoder); + } + } } /* @@ -344,15 +301,20 @@ static void dpu_kms_prepare_commit(struct msm_kms *kms, void dpu_kms_encoder_enable(struct drm_encoder *encoder) { const struct drm_encoder_helper_funcs *funcs = encoder->helper_private; - struct drm_crtc *crtc = encoder->crtc; + struct drm_device *dev = encoder->dev; + struct drm_crtc *crtc; /* Forward this enable call to the commit hook */ if (funcs && funcs->commit) funcs->commit(encoder); - if (crtc && crtc->state->active) { + WARN_ON(!drm_modeset_is_locked(&dev->mode_config.connection_mutex)); + drm_for_each_crtc(crtc, dev) { + if (!(crtc->state->encoder_mask & drm_encoder_mask(encoder))) + continue; + trace_dpu_kms_enc_enable(DRMID(crtc)); - dpu_crtc_commit_kickoff(crtc); + dpu_crtc_commit_kickoff(crtc, false); } } @@ -369,7 +331,8 @@ static void dpu_kms_commit(struct msm_kms *kms, struct drm_atomic_state *state) if (crtc->state->active) { trace_dpu_kms_commit(DRMID(crtc)); - dpu_crtc_commit_kickoff(crtc); + dpu_crtc_commit_kickoff(crtc, + state->legacy_cursor_update); } } } @@ -613,22 +576,7 @@ fail: #ifdef CONFIG_DEBUG_FS static int dpu_kms_debugfs_init(struct msm_kms *kms, struct drm_minor *minor) { - struct dpu_kms *dpu_kms = to_dpu_kms(kms); - struct drm_device *dev; - int rc; - - if (!dpu_kms || !dpu_kms->dev || !dpu_kms->dev->dev) { - DPU_ERROR("invalid dpu_kms\n"); - return -EINVAL; - } - - dev = dpu_kms->dev; - - rc = _dpu_debugfs_init(dpu_kms); - if (rc) - DPU_ERROR("dpu_debugfs init failed: %d\n", rc); - - return rc; + return _dpu_debugfs_init(to_dpu_kms(kms)); } #endif @@ -651,12 +599,7 @@ static void _dpu_kms_hw_destroy(struct dpu_kms *dpu_kms) dpu_hw_intr_destroy(dpu_kms->hw_intr); dpu_kms->hw_intr = NULL; - if (dpu_kms->power_event) - dpu_power_handle_unregister_event( - &dpu_kms->phandle, dpu_kms->power_event); - /* safe to call these more than once during shutdown */ - _dpu_debugfs_destroy(dpu_kms); _dpu_kms_mmu_destroy(dpu_kms); if (dpu_kms->catalog) { @@ -676,11 +619,6 @@ static void _dpu_kms_hw_destroy(struct dpu_kms *dpu_kms) dpu_hw_catalog_deinit(dpu_kms->catalog); dpu_kms->catalog = NULL; - if (dpu_kms->core_client) - dpu_power_client_destroy(&dpu_kms->phandle, - dpu_kms->core_client); - dpu_kms->core_client = NULL; - if (dpu_kms->vbif[VBIF_NRT]) devm_iounmap(&dpu_kms->pdev->dev, dpu_kms->vbif[VBIF_NRT]); dpu_kms->vbif[VBIF_NRT] = NULL; @@ -705,131 +643,9 @@ static void dpu_kms_destroy(struct msm_kms *kms) dpu_kms = to_dpu_kms(kms); - dpu_dbg_destroy(); _dpu_kms_hw_destroy(dpu_kms); } -static int dpu_kms_pm_suspend(struct device *dev) -{ - struct drm_device *ddev; - struct drm_modeset_acquire_ctx ctx; - struct drm_atomic_state *state; - struct dpu_kms *dpu_kms; - int ret = 0, num_crtcs = 0; - - if (!dev) - return -EINVAL; - - ddev = dev_get_drvdata(dev); - if (!ddev || !ddev_to_msm_kms(ddev)) - return -EINVAL; - - dpu_kms = to_dpu_kms(ddev_to_msm_kms(ddev)); - - /* disable hot-plug polling */ - drm_kms_helper_poll_disable(ddev); - - /* acquire modeset lock(s) */ - drm_modeset_acquire_init(&ctx, 0); - -retry: - DPU_ATRACE_BEGIN("kms_pm_suspend"); - - ret = drm_modeset_lock_all_ctx(ddev, &ctx); - if (ret) - goto unlock; - - /* save current state for resume */ - if (dpu_kms->suspend_state) - drm_atomic_state_put(dpu_kms->suspend_state); - dpu_kms->suspend_state = drm_atomic_helper_duplicate_state(ddev, &ctx); - if (IS_ERR_OR_NULL(dpu_kms->suspend_state)) { - DRM_ERROR("failed to back up suspend state\n"); - dpu_kms->suspend_state = NULL; - goto unlock; - } - - /* create atomic state to disable all CRTCs */ - state = drm_atomic_state_alloc(ddev); - if (IS_ERR_OR_NULL(state)) { - DRM_ERROR("failed to allocate crtc disable state\n"); - goto unlock; - } - - state->acquire_ctx = &ctx; - - /* check for nothing to do */ - if (num_crtcs == 0) { - DRM_DEBUG("all crtcs are already in the off state\n"); - drm_atomic_state_put(state); - goto suspended; - } - - /* commit the "disable all" state */ - ret = drm_atomic_commit(state); - if (ret < 0) { - DRM_ERROR("failed to disable crtcs, %d\n", ret); - drm_atomic_state_put(state); - goto unlock; - } - -suspended: - dpu_kms->suspend_block = true; - -unlock: - if (ret == -EDEADLK) { - drm_modeset_backoff(&ctx); - goto retry; - } - drm_modeset_drop_locks(&ctx); - drm_modeset_acquire_fini(&ctx); - - DPU_ATRACE_END("kms_pm_suspend"); - return 0; -} - -static int dpu_kms_pm_resume(struct device *dev) -{ - struct drm_device *ddev; - struct dpu_kms *dpu_kms; - int ret; - - if (!dev) - return -EINVAL; - - ddev = dev_get_drvdata(dev); - if (!ddev || !ddev_to_msm_kms(ddev)) - return -EINVAL; - - dpu_kms = to_dpu_kms(ddev_to_msm_kms(ddev)); - - DPU_ATRACE_BEGIN("kms_pm_resume"); - - drm_mode_config_reset(ddev); - - drm_modeset_lock_all(ddev); - - dpu_kms->suspend_block = false; - - if (dpu_kms->suspend_state) { - dpu_kms->suspend_state->acquire_ctx = - ddev->mode_config.acquire_ctx; - ret = drm_atomic_commit(dpu_kms->suspend_state); - if (ret < 0) { - DRM_ERROR("failed to restore state, %d\n", ret); - drm_atomic_state_put(dpu_kms->suspend_state); - } - dpu_kms->suspend_state = NULL; - } - drm_modeset_unlock_all(ddev); - - /* enable hot-plug polling */ - drm_kms_helper_poll_enable(ddev); - - DPU_ATRACE_END("kms_pm_resume"); - return 0; -} - static void _dpu_kms_set_encoder_mode(struct msm_kms *kms, struct drm_encoder *encoder, bool cmd_mode) @@ -858,10 +674,30 @@ static void _dpu_kms_set_encoder_mode(struct msm_kms *kms, encoder->base.id, rc); } +static irqreturn_t dpu_irq(struct msm_kms *kms) +{ + struct dpu_kms *dpu_kms = to_dpu_kms(kms); + + return dpu_core_irq(dpu_kms); +} + +static void dpu_irq_preinstall(struct msm_kms *kms) +{ + struct dpu_kms *dpu_kms = to_dpu_kms(kms); + + dpu_core_irq_preinstall(dpu_kms); +} + +static void dpu_irq_uninstall(struct msm_kms *kms) +{ + struct dpu_kms *dpu_kms = to_dpu_kms(kms); + + dpu_core_irq_uninstall(dpu_kms); +} + static const struct msm_kms_funcs kms_funcs = { .hw_init = dpu_kms_hw_init, .irq_preinstall = dpu_irq_preinstall, - .irq_postinstall = dpu_irq_postinstall, .irq_uninstall = dpu_irq_uninstall, .irq = dpu_irq, .prepare_commit = dpu_kms_prepare_commit, @@ -873,8 +709,6 @@ static const struct msm_kms_funcs kms_funcs = { .check_modified_format = dpu_format_check_modified_format, .get_format = dpu_get_msm_format, .round_pixclk = dpu_kms_round_pixclk, - .pm_suspend = dpu_kms_pm_suspend, - .pm_resume = dpu_kms_pm_resume, .destroy = dpu_kms_destroy, .set_encoder_mode = _dpu_kms_set_encoder_mode, #ifdef CONFIG_DEBUG_FS @@ -882,12 +716,6 @@ static const struct msm_kms_funcs kms_funcs = { #endif }; -/* the caller api needs to turn on clock before calling it */ -static inline void _dpu_kms_core_hw_rev_init(struct dpu_kms *dpu_kms) -{ - dpu_kms->core_rev = readl_relaxed(dpu_kms->mmio + 0x0); -} - static int _dpu_kms_mmu_destroy(struct dpu_kms *dpu_kms) { struct msm_mmu *mmu; @@ -911,6 +739,9 @@ static int _dpu_kms_mmu_init(struct dpu_kms *dpu_kms) if (!domain) return 0; + domain->geometry.aperture_start = 0x1000; + domain->geometry.aperture_end = 0xffffffff; + aspace = msm_gem_address_space_create(dpu_kms->dev->dev, domain, "dpu1"); if (IS_ERR(aspace)) { @@ -960,16 +791,6 @@ u64 dpu_kms_get_clk_rate(struct dpu_kms *dpu_kms, char *clock_name) return clk_get_rate(clk->clk); } -static void dpu_kms_handle_power_event(u32 event_type, void *usr) -{ - struct dpu_kms *dpu_kms = usr; - - if (!dpu_kms) - return; - - dpu_vbif_init_memtypes(dpu_kms); -} - static int dpu_kms_hw_init(struct msm_kms *kms) { struct dpu_kms *dpu_kms; @@ -979,26 +800,20 @@ static int dpu_kms_hw_init(struct msm_kms *kms) if (!kms) { DPU_ERROR("invalid kms\n"); - goto end; + return rc; } dpu_kms = to_dpu_kms(kms); dev = dpu_kms->dev; if (!dev) { DPU_ERROR("invalid device\n"); - goto end; - } - - rc = dpu_dbg_init(&dpu_kms->pdev->dev); - if (rc) { - DRM_ERROR("failed to init dpu dbg: %d\n", rc); - goto end; + return rc; } priv = dev->dev_private; if (!priv) { DPU_ERROR("invalid private data\n"); - goto dbg_destroy; + return rc; } dpu_kms->mmio = msm_ioremap(dpu_kms->pdev, "mdp", "mdp"); @@ -1036,20 +851,9 @@ static int dpu_kms_hw_init(struct msm_kms *kms) dpu_kms->reg_dma_len = dpu_iomap_size(dpu_kms->pdev, "regdma"); } - dpu_kms->core_client = dpu_power_client_create(&dpu_kms->phandle, - "core"); - if (IS_ERR_OR_NULL(dpu_kms->core_client)) { - rc = PTR_ERR(dpu_kms->core_client); - if (!dpu_kms->core_client) - rc = -EINVAL; - DPU_ERROR("dpu power client create failed: %d\n", rc); - dpu_kms->core_client = NULL; - goto error; - } - pm_runtime_get_sync(&dpu_kms->pdev->dev); - _dpu_kms_core_hw_rev_init(dpu_kms); + dpu_kms->core_rev = readl_relaxed(dpu_kms->mmio + 0x0); pr_info("dpu hardware revision:0x%x\n", dpu_kms->core_rev); @@ -1063,8 +867,6 @@ static int dpu_kms_hw_init(struct msm_kms *kms) goto power_error; } - dpu_dbg_init_dbg_buses(dpu_kms->core_rev); - /* * Now we need to read the HW catalog and initialize resources such as * clocks, regulators, GDSC/MMAGIC, ioremap the register ranges etc @@ -1110,7 +912,6 @@ static int dpu_kms_hw_init(struct msm_kms *kms) } rc = dpu_core_perf_init(&dpu_kms->perf, dev, dpu_kms->catalog, - &dpu_kms->phandle, _dpu_kms_get_clk(dpu_kms, "core")); if (rc) { DPU_ERROR("failed to init perf %d\n", rc); @@ -1151,13 +952,7 @@ static int dpu_kms_hw_init(struct msm_kms *kms) */ dev->mode_config.allow_fb_modifiers = true; - /* - * Handle (re)initializations during power enable - */ - dpu_kms_handle_power_event(DPU_POWER_EVENT_ENABLE, dpu_kms); - dpu_kms->power_event = dpu_power_handle_register_event( - &dpu_kms->phandle, DPU_POWER_EVENT_ENABLE, - dpu_kms_handle_power_event, dpu_kms, "kms"); + dpu_vbif_init_memtypes(dpu_kms); pm_runtime_put_sync(&dpu_kms->pdev->dev); @@ -1171,9 +966,7 @@ power_error: pm_runtime_put_sync(&dpu_kms->pdev->dev); error: _dpu_kms_hw_destroy(dpu_kms); -dbg_destroy: - dpu_dbg_destroy(); -end: + return rc; } @@ -1221,8 +1014,6 @@ static int dpu_bind(struct device *dev, struct device *master, void *data) return ret; } - dpu_power_resource_init(pdev, &dpu_kms->phandle); - platform_set_drvdata(pdev, dpu_kms); msm_kms_init(&dpu_kms->base, &kms_funcs); @@ -1242,7 +1033,6 @@ static void dpu_unbind(struct device *dev, struct device *master, void *data) struct dpu_kms *dpu_kms = platform_get_drvdata(pdev); struct dss_module_power *mp = &dpu_kms->mp; - dpu_power_resource_deinit(pdev, &dpu_kms->phandle); msm_dss_put_clk(mp->clk_config, mp->num_clk); devm_kfree(&pdev->dev, mp->clk_config); mp->num_clk = 0; @@ -1278,19 +1068,13 @@ static int __maybe_unused dpu_runtime_suspend(struct device *dev) ddev = dpu_kms->dev; if (!ddev) { DPU_ERROR("invalid drm_device\n"); - goto exit; + return rc; } - rc = dpu_power_resource_enable(&dpu_kms->phandle, - dpu_kms->core_client, false); - if (rc) - DPU_ERROR("resource disable failed: %d\n", rc); - rc = msm_dss_enable_clk(mp->clk_config, mp->num_clk, false); if (rc) DPU_ERROR("clock disable failed rc:%d\n", rc); -exit: return rc; } @@ -1299,27 +1083,27 @@ static int __maybe_unused dpu_runtime_resume(struct device *dev) int rc = -1; struct platform_device *pdev = to_platform_device(dev); struct dpu_kms *dpu_kms = platform_get_drvdata(pdev); + struct drm_encoder *encoder; struct drm_device *ddev; struct dss_module_power *mp = &dpu_kms->mp; ddev = dpu_kms->dev; if (!ddev) { DPU_ERROR("invalid drm_device\n"); - goto exit; + return rc; } rc = msm_dss_enable_clk(mp->clk_config, mp->num_clk, true); if (rc) { DPU_ERROR("clock enable failed rc:%d\n", rc); - goto exit; + return rc; } - rc = dpu_power_resource_enable(&dpu_kms->phandle, - dpu_kms->core_client, true); - if (rc) - DPU_ERROR("resource enable failed: %d\n", rc); + dpu_vbif_init_memtypes(dpu_kms); + + drm_for_each_encoder(encoder, ddev) + dpu_encoder_virt_runtime_resume(encoder); -exit: return rc; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h index 66d466628e2b..ac75cfc267f4 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h @@ -23,15 +23,13 @@ #include "msm_kms.h" #include "msm_mmu.h" #include "msm_gem.h" -#include "dpu_dbg.h" #include "dpu_hw_catalog.h" #include "dpu_hw_ctl.h" #include "dpu_hw_lm.h" #include "dpu_hw_interrupts.h" #include "dpu_hw_top.h" +#include "dpu_io_util.h" #include "dpu_rm.h" -#include "dpu_power_handle.h" -#include "dpu_irq.h" #include "dpu_core_perf.h" #define DRMID(x) ((x) ? (x)->base.id : -1) @@ -104,7 +102,6 @@ struct dpu_irq { atomic_t *enable_counts; atomic_t *irq_counts; spinlock_t cb_lock; - struct dentry *debugfs_file; }; struct dpu_kms { @@ -113,15 +110,6 @@ struct dpu_kms { int core_rev; struct dpu_mdss_cfg *catalog; - struct dpu_power_handle phandle; - struct dpu_power_client *core_client; - struct dpu_power_event *power_event; - - /* directory entry for debugfs */ - struct dentry *debugfs_root; - struct dentry *debugfs_danger; - struct dentry *debugfs_vbif; - /* io/register spaces: */ void __iomem *mmio, *vbif[VBIF_MAX], *reg_dma; unsigned long mmio_len, vbif_len[VBIF_MAX], reg_dma_len; @@ -135,10 +123,6 @@ struct dpu_kms { struct dpu_core_perf perf; - /* saved atomic state during system suspend */ - struct drm_atomic_state *suspend_state; - bool suspend_block; - struct dpu_rm rm; bool rm_init; @@ -164,33 +148,6 @@ struct vsync_info { ((struct msm_drm_private *)((D)->dev_private))->kms : NULL) /** - * dpu_kms_is_suspend_state - whether or not the system is pm suspended - * @dev: Pointer to drm device - * Return: Suspend status - */ -static inline bool dpu_kms_is_suspend_state(struct drm_device *dev) -{ - if (!ddev_to_msm_kms(dev)) - return false; - - return to_dpu_kms(ddev_to_msm_kms(dev))->suspend_state != NULL; -} - -/** - * dpu_kms_is_suspend_blocked - whether or not commits are blocked due to pm - * suspend status - * @dev: Pointer to drm device - * Return: True if commits should be rejected due to pm suspend - */ -static inline bool dpu_kms_is_suspend_blocked(struct drm_device *dev) -{ - if (!dpu_kms_is_suspend_state(dev)) - return false; - - return to_dpu_kms(ddev_to_msm_kms(dev))->suspend_block; -} - -/** * Debugfs functions - extra helper functions for debugfs support * * Main debugfs documentation is located at, diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c index 2235ef8129f4..cb307a2abf06 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c @@ -9,6 +9,11 @@ #define HW_INTR_STATUS 0x0010 +struct dpu_irq_controller { + unsigned long enabled_mask; + struct irq_domain *domain; +}; + struct dpu_mdss { struct msm_mdss base; void __iomem *mmio; @@ -115,13 +120,12 @@ static int _dpu_mdss_irq_domain_add(struct dpu_mdss *dpu_mdss) return 0; } -static int _dpu_mdss_irq_domain_fini(struct dpu_mdss *dpu_mdss) +static void _dpu_mdss_irq_domain_fini(struct dpu_mdss *dpu_mdss) { if (dpu_mdss->irq_controller.domain) { irq_domain_remove(dpu_mdss->irq_controller.domain); dpu_mdss->irq_controller.domain = NULL; } - return 0; } static int dpu_mdss_enable(struct msm_mdss *mdss) { @@ -156,18 +160,16 @@ static void dpu_mdss_destroy(struct drm_device *dev) struct dpu_mdss *dpu_mdss = to_dpu_mdss(priv->mdss); struct dss_module_power *mp = &dpu_mdss->mp; + pm_runtime_suspend(dev->dev); + pm_runtime_disable(dev->dev); _dpu_mdss_irq_domain_fini(dpu_mdss); - free_irq(platform_get_irq(pdev, 0), dpu_mdss); - msm_dss_put_clk(mp->clk_config, mp->num_clk); devm_kfree(&pdev->dev, mp->clk_config); if (dpu_mdss->mmio) devm_iounmap(&pdev->dev, dpu_mdss->mmio); dpu_mdss->mmio = NULL; - - pm_runtime_disable(dev->dev); priv->mdss = NULL; } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c index d77a8cb15404..fd75870eb17f 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c @@ -137,7 +137,7 @@ static struct dpu_kms *_dpu_plane_get_kms(struct drm_plane *plane) * @src_wdith: width of source buffer * Return: fill level corresponding to the source buffer/format or 0 if error */ -static inline int _dpu_plane_calc_fill_level(struct drm_plane *plane, +static int _dpu_plane_calc_fill_level(struct drm_plane *plane, const struct dpu_format *fmt, u32 src_width) { struct dpu_plane *pdpu, *tmp; @@ -430,24 +430,14 @@ static void _dpu_plane_set_qos_remap(struct drm_plane *plane) dpu_vbif_set_qos_remap(dpu_kms, &qos_params); } -/** - * _dpu_plane_get_aspace: gets the address space - */ -static inline struct msm_gem_address_space *_dpu_plane_get_aspace( - struct dpu_plane *pdpu) -{ - struct dpu_kms *kms = _dpu_plane_get_kms(&pdpu->base); - - return kms->base.aspace; -} - -static inline void _dpu_plane_set_scanout(struct drm_plane *plane, +static void _dpu_plane_set_scanout(struct drm_plane *plane, struct dpu_plane_state *pstate, struct dpu_hw_pipe_cfg *pipe_cfg, struct drm_framebuffer *fb) { struct dpu_plane *pdpu = to_dpu_plane(plane); - struct msm_gem_address_space *aspace = _dpu_plane_get_aspace(pdpu); + struct dpu_kms *kms = _dpu_plane_get_kms(&pdpu->base); + struct msm_gem_address_space *aspace = kms->base.aspace; int ret; ret = dpu_format_populate_layout(aspace, fb, &pipe_cfg->layout); @@ -525,7 +515,7 @@ static void _dpu_plane_setup_scaler3(struct dpu_plane *pdpu, scale_cfg->enable = 1; } -static inline void _dpu_plane_setup_csc(struct dpu_plane *pdpu) +static void _dpu_plane_setup_csc(struct dpu_plane *pdpu) { static const struct dpu_csc_cfg dpu_csc_YUV2RGB_601L = { { @@ -801,7 +791,7 @@ static int dpu_plane_prepare_fb(struct drm_plane *plane, struct drm_gem_object *obj; struct msm_gem_object *msm_obj; struct dma_fence *fence; - struct msm_gem_address_space *aspace = _dpu_plane_get_aspace(pdpu); + struct dpu_kms *kms = _dpu_plane_get_kms(&pdpu->base); int ret; if (!new_state->fb) @@ -810,7 +800,7 @@ static int dpu_plane_prepare_fb(struct drm_plane *plane, DPU_DEBUG_PLANE(pdpu, "FB[%u]\n", fb->base.id); /* cache aspace */ - pstate->aspace = aspace; + pstate->aspace = kms->base.aspace; /* * TODO: Need to sort out the msm_framebuffer_prepare() call below so @@ -1191,19 +1181,8 @@ static void dpu_plane_destroy(struct drm_plane *plane) static void dpu_plane_destroy_state(struct drm_plane *plane, struct drm_plane_state *state) { - struct dpu_plane_state *pstate; - - if (!plane || !state) { - DPU_ERROR("invalid arg(s), plane %d state %d\n", - plane != 0, state != 0); - return; - } - - pstate = to_dpu_plane_state(state); - __drm_atomic_helper_plane_destroy_state(state); - - kfree(pstate); + kfree(to_dpu_plane_state(state)); } static struct drm_plane_state * @@ -1273,26 +1252,12 @@ static ssize_t _dpu_plane_danger_read(struct file *file, char __user *buff, size_t count, loff_t *ppos) { struct dpu_kms *kms = file->private_data; - struct dpu_mdss_cfg *cfg = kms->catalog; - int len = 0; - char buf[40] = {'\0'}; - - if (!cfg) - return -ENODEV; + int len; + char buf[40]; - if (*ppos) - return 0; /* the end */ - - len = snprintf(buf, sizeof(buf), "%d\n", !kms->has_danger_ctrl); - if (len < 0 || len >= sizeof(buf)) - return 0; + len = scnprintf(buf, sizeof(buf), "%d\n", !kms->has_danger_ctrl); - if ((count < sizeof(buf)) || copy_to_user(buff, buf, len)) - return -EFAULT; - - *ppos += len; /* increase offset */ - - return len; + return simple_read_from_buffer(buff, count, ppos, buf, len); } static void _dpu_plane_set_danger_state(struct dpu_kms *kms, bool enable) @@ -1322,23 +1287,12 @@ static ssize_t _dpu_plane_danger_write(struct file *file, const char __user *user_buf, size_t count, loff_t *ppos) { struct dpu_kms *kms = file->private_data; - struct dpu_mdss_cfg *cfg = kms->catalog; int disable_panic; - char buf[10]; - - if (!cfg) - return -EFAULT; - - if (count >= sizeof(buf)) - return -EFAULT; - - if (copy_from_user(buf, user_buf, count)) - return -EFAULT; - - buf[count] = 0; /* end of string */ + int ret; - if (kstrtoint(buf, 0, &disable_panic)) - return -EFAULT; + ret = kstrtouint_from_user(user_buf, count, 0, &disable_panic); + if (ret) + return ret; if (disable_panic) { /* Disable panic signal for all active pipes */ @@ -1363,33 +1317,10 @@ static const struct file_operations dpu_plane_danger_enable = { static int _dpu_plane_init_debugfs(struct drm_plane *plane) { - struct dpu_plane *pdpu; - struct dpu_kms *kms; - struct msm_drm_private *priv; - const struct dpu_sspp_sub_blks *sblk = 0; - const struct dpu_sspp_cfg *cfg = 0; - - if (!plane || !plane->dev) { - DPU_ERROR("invalid arguments\n"); - return -EINVAL; - } - - priv = plane->dev->dev_private; - if (!priv || !priv->kms) { - DPU_ERROR("invalid KMS reference\n"); - return -EINVAL; - } - - kms = to_dpu_kms(priv->kms); - pdpu = to_dpu_plane(plane); - - if (pdpu && pdpu->pipe_hw) - cfg = pdpu->pipe_hw->cap; - if (cfg) - sblk = cfg->sblk; - - if (!sblk) - return 0; + struct dpu_plane *pdpu = to_dpu_plane(plane); + struct dpu_kms *kms = _dpu_plane_get_kms(plane); + const struct dpu_sspp_cfg *cfg = pdpu->pipe_hw->cap; + const struct dpu_sspp_sub_blks *sblk = cfg->sblk; /* create overall sub-directory for the pipe */ pdpu->debugfs_root = @@ -1460,25 +1391,11 @@ static int _dpu_plane_init_debugfs(struct drm_plane *plane) return 0; } - -static void _dpu_plane_destroy_debugfs(struct drm_plane *plane) -{ - struct dpu_plane *pdpu; - - if (!plane) - return; - pdpu = to_dpu_plane(plane); - - debugfs_remove_recursive(pdpu->debugfs_root); -} #else static int _dpu_plane_init_debugfs(struct drm_plane *plane) { return 0; } -static void _dpu_plane_destroy_debugfs(struct drm_plane *plane) -{ -} #endif static int dpu_plane_late_register(struct drm_plane *plane) @@ -1488,7 +1405,9 @@ static int dpu_plane_late_register(struct drm_plane *plane) static void dpu_plane_early_unregister(struct drm_plane *plane) { - _dpu_plane_destroy_debugfs(plane); + struct dpu_plane *pdpu = to_dpu_plane(plane); + + debugfs_remove_recursive(pdpu->debugfs_root); } static const struct drm_plane_funcs dpu_plane_funcs = { @@ -1537,7 +1456,7 @@ struct drm_plane *dpu_plane_init(struct drm_device *dev, if (!pdpu) { DPU_ERROR("[%u]failed to allocate local plane struct\n", pipe); ret = -ENOMEM; - goto exit; + return ERR_PTR(ret); } /* cache local stuff for later */ @@ -1623,6 +1542,5 @@ clean_sspp: dpu_hw_sspp_destroy(pdpu->pipe_hw); clean_plane: kfree(pdpu); -exit: return ERR_PTR(ret); } diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_power_handle.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_power_handle.c deleted file mode 100644 index fc14116789f2..000000000000 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_power_handle.c +++ /dev/null @@ -1,240 +0,0 @@ -/* Copyright (c) 2014-2018, The Linux Foundation. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 and - * only version 2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - */ - -#define pr_fmt(fmt) "[drm:%s:%d]: " fmt, __func__, __LINE__ - -#include <linux/kernel.h> -#include <linux/of.h> -#include <linux/string.h> -#include <linux/of_address.h> -#include <linux/slab.h> -#include <linux/mutex.h> -#include <linux/of_platform.h> - -#include "dpu_power_handle.h" -#include "dpu_trace.h" - -static const char *data_bus_name[DPU_POWER_HANDLE_DBUS_ID_MAX] = { - [DPU_POWER_HANDLE_DBUS_ID_MNOC] = "qcom,dpu-data-bus", - [DPU_POWER_HANDLE_DBUS_ID_LLCC] = "qcom,dpu-llcc-bus", - [DPU_POWER_HANDLE_DBUS_ID_EBI] = "qcom,dpu-ebi-bus", -}; - -const char *dpu_power_handle_get_dbus_name(u32 bus_id) -{ - if (bus_id < DPU_POWER_HANDLE_DBUS_ID_MAX) - return data_bus_name[bus_id]; - - return NULL; -} - -static void dpu_power_event_trigger_locked(struct dpu_power_handle *phandle, - u32 event_type) -{ - struct dpu_power_event *event; - - list_for_each_entry(event, &phandle->event_list, list) { - if (event->event_type & event_type) - event->cb_fnc(event_type, event->usr); - } -} - -struct dpu_power_client *dpu_power_client_create( - struct dpu_power_handle *phandle, char *client_name) -{ - struct dpu_power_client *client; - static u32 id; - - if (!client_name || !phandle) { - pr_err("client name is null or invalid power data\n"); - return ERR_PTR(-EINVAL); - } - - client = kzalloc(sizeof(struct dpu_power_client), GFP_KERNEL); - if (!client) - return ERR_PTR(-ENOMEM); - - mutex_lock(&phandle->phandle_lock); - strlcpy(client->name, client_name, MAX_CLIENT_NAME_LEN); - client->usecase_ndx = VOTE_INDEX_DISABLE; - client->id = id; - client->active = true; - pr_debug("client %s created:%pK id :%d\n", client_name, - client, id); - id++; - list_add(&client->list, &phandle->power_client_clist); - mutex_unlock(&phandle->phandle_lock); - - return client; -} - -void dpu_power_client_destroy(struct dpu_power_handle *phandle, - struct dpu_power_client *client) -{ - if (!client || !phandle) { - pr_err("reg bus vote: invalid client handle\n"); - } else if (!client->active) { - pr_err("dpu power deinit already done\n"); - kfree(client); - } else { - pr_debug("bus vote client %s destroyed:%pK id:%u\n", - client->name, client, client->id); - mutex_lock(&phandle->phandle_lock); - list_del_init(&client->list); - mutex_unlock(&phandle->phandle_lock); - kfree(client); - } -} - -void dpu_power_resource_init(struct platform_device *pdev, - struct dpu_power_handle *phandle) -{ - phandle->dev = &pdev->dev; - - INIT_LIST_HEAD(&phandle->power_client_clist); - INIT_LIST_HEAD(&phandle->event_list); - - mutex_init(&phandle->phandle_lock); -} - -void dpu_power_resource_deinit(struct platform_device *pdev, - struct dpu_power_handle *phandle) -{ - struct dpu_power_client *curr_client, *next_client; - struct dpu_power_event *curr_event, *next_event; - - if (!phandle || !pdev) { - pr_err("invalid input param\n"); - return; - } - - mutex_lock(&phandle->phandle_lock); - list_for_each_entry_safe(curr_client, next_client, - &phandle->power_client_clist, list) { - pr_err("client:%s-%d still registered with refcount:%d\n", - curr_client->name, curr_client->id, - curr_client->refcount); - curr_client->active = false; - list_del(&curr_client->list); - } - - list_for_each_entry_safe(curr_event, next_event, - &phandle->event_list, list) { - pr_err("event:%d, client:%s still registered\n", - curr_event->event_type, - curr_event->client_name); - curr_event->active = false; - list_del(&curr_event->list); - } - mutex_unlock(&phandle->phandle_lock); -} - -int dpu_power_resource_enable(struct dpu_power_handle *phandle, - struct dpu_power_client *pclient, bool enable) -{ - bool changed = false; - u32 max_usecase_ndx = VOTE_INDEX_DISABLE, prev_usecase_ndx; - struct dpu_power_client *client; - u32 event_type; - - if (!phandle || !pclient) { - pr_err("invalid input argument\n"); - return -EINVAL; - } - - mutex_lock(&phandle->phandle_lock); - if (enable) - pclient->refcount++; - else if (pclient->refcount) - pclient->refcount--; - - if (pclient->refcount) - pclient->usecase_ndx = VOTE_INDEX_LOW; - else - pclient->usecase_ndx = VOTE_INDEX_DISABLE; - - list_for_each_entry(client, &phandle->power_client_clist, list) { - if (client->usecase_ndx < VOTE_INDEX_MAX && - client->usecase_ndx > max_usecase_ndx) - max_usecase_ndx = client->usecase_ndx; - } - - if (phandle->current_usecase_ndx != max_usecase_ndx) { - changed = true; - prev_usecase_ndx = phandle->current_usecase_ndx; - phandle->current_usecase_ndx = max_usecase_ndx; - } - - pr_debug("%pS: changed=%d current idx=%d request client %s id:%u enable:%d refcount:%d\n", - __builtin_return_address(0), changed, max_usecase_ndx, - pclient->name, pclient->id, enable, pclient->refcount); - - if (!changed) - goto end; - - event_type = enable ? DPU_POWER_EVENT_ENABLE : DPU_POWER_EVENT_DISABLE; - - dpu_power_event_trigger_locked(phandle, event_type); -end: - mutex_unlock(&phandle->phandle_lock); - return 0; -} - -struct dpu_power_event *dpu_power_handle_register_event( - struct dpu_power_handle *phandle, - u32 event_type, void (*cb_fnc)(u32 event_type, void *usr), - void *usr, char *client_name) -{ - struct dpu_power_event *event; - - if (!phandle) { - pr_err("invalid power handle\n"); - return ERR_PTR(-EINVAL); - } else if (!cb_fnc || !event_type) { - pr_err("no callback fnc or event type\n"); - return ERR_PTR(-EINVAL); - } - - event = kzalloc(sizeof(struct dpu_power_event), GFP_KERNEL); - if (!event) - return ERR_PTR(-ENOMEM); - - event->event_type = event_type; - event->cb_fnc = cb_fnc; - event->usr = usr; - strlcpy(event->client_name, client_name, MAX_CLIENT_NAME_LEN); - event->active = true; - - mutex_lock(&phandle->phandle_lock); - list_add(&event->list, &phandle->event_list); - mutex_unlock(&phandle->phandle_lock); - - return event; -} - -void dpu_power_handle_unregister_event( - struct dpu_power_handle *phandle, - struct dpu_power_event *event) -{ - if (!phandle || !event) { - pr_err("invalid phandle or event\n"); - } else if (!event->active) { - pr_err("power handle deinit already done\n"); - kfree(event); - } else { - mutex_lock(&phandle->phandle_lock); - list_del_init(&event->list); - mutex_unlock(&phandle->phandle_lock); - kfree(event); - } -} diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_power_handle.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_power_handle.h deleted file mode 100644 index a65b7a297f21..000000000000 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_power_handle.h +++ /dev/null @@ -1,217 +0,0 @@ -/* Copyright (c) 2016-2018, The Linux Foundation. All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 and - * only version 2 as published by the Free Software Foundation. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - */ - -#ifndef _DPU_POWER_HANDLE_H_ -#define _DPU_POWER_HANDLE_H_ - -#define MAX_CLIENT_NAME_LEN 128 - -#define DPU_POWER_HANDLE_ENABLE_BUS_AB_QUOTA 0 -#define DPU_POWER_HANDLE_DISABLE_BUS_AB_QUOTA 0 -#define DPU_POWER_HANDLE_ENABLE_BUS_IB_QUOTA 1600000000 -#define DPU_POWER_HANDLE_DISABLE_BUS_IB_QUOTA 0 - -#include "dpu_io_util.h" - -/* events will be triggered on power handler enable/disable */ -#define DPU_POWER_EVENT_DISABLE BIT(0) -#define DPU_POWER_EVENT_ENABLE BIT(1) - -/** - * mdss_bus_vote_type: register bus vote type - * VOTE_INDEX_DISABLE: removes the client vote - * VOTE_INDEX_LOW: keeps the lowest vote for register bus - * VOTE_INDEX_MAX: invalid - */ -enum mdss_bus_vote_type { - VOTE_INDEX_DISABLE, - VOTE_INDEX_LOW, - VOTE_INDEX_MAX, -}; - -/** - * enum dpu_power_handle_data_bus_client - type of axi bus clients - * @DPU_POWER_HANDLE_DATA_BUS_CLIENT_RT: core real-time bus client - * @DPU_POWER_HANDLE_DATA_BUS_CLIENT_NRT: core non-real-time bus client - * @DPU_POWER_HANDLE_DATA_BUS_CLIENT_MAX: maximum number of bus client type - */ -enum dpu_power_handle_data_bus_client { - DPU_POWER_HANDLE_DATA_BUS_CLIENT_RT, - DPU_POWER_HANDLE_DATA_BUS_CLIENT_NRT, - DPU_POWER_HANDLE_DATA_BUS_CLIENT_MAX -}; - -/** - * enum DPU_POWER_HANDLE_DBUS_ID - data bus identifier - * @DPU_POWER_HANDLE_DBUS_ID_MNOC: DPU/MNOC data bus - * @DPU_POWER_HANDLE_DBUS_ID_LLCC: MNOC/LLCC data bus - * @DPU_POWER_HANDLE_DBUS_ID_EBI: LLCC/EBI data bus - */ -enum DPU_POWER_HANDLE_DBUS_ID { - DPU_POWER_HANDLE_DBUS_ID_MNOC, - DPU_POWER_HANDLE_DBUS_ID_LLCC, - DPU_POWER_HANDLE_DBUS_ID_EBI, - DPU_POWER_HANDLE_DBUS_ID_MAX, -}; - -/** - * struct dpu_power_client: stores the power client for dpu driver - * @name: name of the client - * @usecase_ndx: current regs bus vote type - * @refcount: current refcount if multiple modules are using same - * same client for enable/disable. Power module will - * aggregate the refcount and vote accordingly for this - * client. - * @id: assigned during create. helps for debugging. - * @list: list to attach power handle master list - * @ab: arbitrated bandwidth for each bus client - * @ib: instantaneous bandwidth for each bus client - * @active: inidcates the state of dpu power handle - */ -struct dpu_power_client { - char name[MAX_CLIENT_NAME_LEN]; - short usecase_ndx; - short refcount; - u32 id; - struct list_head list; - u64 ab[DPU_POWER_HANDLE_DATA_BUS_CLIENT_MAX]; - u64 ib[DPU_POWER_HANDLE_DATA_BUS_CLIENT_MAX]; - bool active; -}; - -/* - * struct dpu_power_event - local event registration structure - * @client_name: name of the client registering - * @cb_fnc: pointer to desired callback function - * @usr: user pointer to pass to callback event trigger - * @event: refer to DPU_POWER_HANDLE_EVENT_* - * @list: list to attach event master list - * @active: indicates the state of dpu power handle - */ -struct dpu_power_event { - char client_name[MAX_CLIENT_NAME_LEN]; - void (*cb_fnc)(u32 event_type, void *usr); - void *usr; - u32 event_type; - struct list_head list; - bool active; -}; - -/** - * struct dpu_power_handle: power handle main struct - * @client_clist: master list to store all clients - * @phandle_lock: lock to synchronize the enable/disable - * @dev: pointer to device structure - * @usecase_ndx: current usecase index - * @event_list: current power handle event list - */ -struct dpu_power_handle { - struct list_head power_client_clist; - struct mutex phandle_lock; - struct device *dev; - u32 current_usecase_ndx; - struct list_head event_list; -}; - -/** - * dpu_power_resource_init() - initializes the dpu power handle - * @pdev: platform device to search the power resources - * @pdata: power handle to store the power resources - */ -void dpu_power_resource_init(struct platform_device *pdev, - struct dpu_power_handle *pdata); - -/** - * dpu_power_resource_deinit() - release the dpu power handle - * @pdev: platform device for power resources - * @pdata: power handle containing the resources - * - * Return: error code. - */ -void dpu_power_resource_deinit(struct platform_device *pdev, - struct dpu_power_handle *pdata); - -/** - * dpu_power_client_create() - create the client on power handle - * @pdata: power handle containing the resources - * @client_name: new client name for registration - * - * Return: error code. - */ -struct dpu_power_client *dpu_power_client_create(struct dpu_power_handle *pdata, - char *client_name); - -/** - * dpu_power_client_destroy() - destroy the client on power handle - * @pdata: power handle containing the resources - * @client_name: new client name for registration - * - * Return: none - */ -void dpu_power_client_destroy(struct dpu_power_handle *phandle, - struct dpu_power_client *client); - -/** - * dpu_power_resource_enable() - enable/disable the power resources - * @pdata: power handle containing the resources - * @client: client information to enable/disable its vote - * @enable: boolean request for enable/disable - * - * Return: error code. - */ -int dpu_power_resource_enable(struct dpu_power_handle *pdata, - struct dpu_power_client *pclient, bool enable); - -/** - * dpu_power_data_bus_bandwidth_ctrl() - control data bus bandwidth enable - * @phandle: power handle containing the resources - * @client: client information to bandwidth control - * @enable: true to enable bandwidth for data base - * - * Return: none - */ -void dpu_power_data_bus_bandwidth_ctrl(struct dpu_power_handle *phandle, - struct dpu_power_client *pclient, int enable); - -/** - * dpu_power_handle_register_event - register a callback function for an event. - * Clients can register for multiple events with a single register. - * Any block with access to phandle can register for the event - * notification. - * @phandle: power handle containing the resources - * @event_type: event type to register; refer DPU_POWER_HANDLE_EVENT_* - * @cb_fnc: pointer to desired callback function - * @usr: user pointer to pass to callback on event trigger - * - * Return: event pointer if success, or error code otherwise - */ -struct dpu_power_event *dpu_power_handle_register_event( - struct dpu_power_handle *phandle, - u32 event_type, void (*cb_fnc)(u32 event_type, void *usr), - void *usr, char *client_name); -/** - * dpu_power_handle_unregister_event - unregister callback for event(s) - * @phandle: power handle containing the resources - * @event: event pointer returned after power handle register - */ -void dpu_power_handle_unregister_event(struct dpu_power_handle *phandle, - struct dpu_power_event *event); - -/** - * dpu_power_handle_get_dbus_name - get name of given data bus identifier - * @bus_id: data bus identifier - * Return: Pointer to name string if success; NULL otherwise - */ -const char *dpu_power_handle_get_dbus_name(u32 bus_id); - -#endif /* _DPU_POWER_HANDLE_H_ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h index e12c4cefb742..c78b521ceda1 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_trace.h @@ -99,27 +99,6 @@ TRACE_EVENT(dpu_perf_set_ot, __entry->vbif_idx) ) -TRACE_EVENT(dpu_perf_update_bus, - TP_PROTO(int client, unsigned long long ab_quota, - unsigned long long ib_quota), - TP_ARGS(client, ab_quota, ib_quota), - TP_STRUCT__entry( - __field(int, client) - __field(u64, ab_quota) - __field(u64, ib_quota) - ), - TP_fast_assign( - __entry->client = client; - __entry->ab_quota = ab_quota; - __entry->ib_quota = ib_quota; - ), - TP_printk("Request client:%d ab=%llu ib=%llu", - __entry->client, - __entry->ab_quota, - __entry->ib_quota) -) - - TRACE_EVENT(dpu_cmd_release_bw, TP_PROTO(u32 crtc_id), TP_ARGS(crtc_id), @@ -319,6 +298,10 @@ DEFINE_EVENT(dpu_drm_obj_template, dpu_kms_wait_for_commit_done, TP_PROTO(uint32_t drm_id), TP_ARGS(drm_id) ); +DEFINE_EVENT(dpu_drm_obj_template, dpu_crtc_runtime_resume, + TP_PROTO(uint32_t drm_id), + TP_ARGS(drm_id) +); TRACE_EVENT(dpu_enc_enable, TP_PROTO(uint32_t drm_id, int hdisplay, int vdisplay), @@ -539,10 +522,6 @@ DEFINE_EVENT(dpu_id_event_template, dpu_crtc_frame_event_cb, TP_PROTO(uint32_t drm_id, u32 event), TP_ARGS(drm_id, event) ); -DEFINE_EVENT(dpu_id_event_template, dpu_crtc_handle_power_event, - TP_PROTO(uint32_t drm_id, u32 event), - TP_ARGS(drm_id, event) -); DEFINE_EVENT(dpu_id_event_template, dpu_crtc_frame_event_done, TP_PROTO(uint32_t drm_id, u32 event), TP_ARGS(drm_id, event) @@ -749,24 +728,17 @@ TRACE_EVENT(dpu_crtc_vblank_enable, __field( uint32_t, enc_id ) __field( bool, enable ) __field( bool, enabled ) - __field( bool, suspend ) - __field( bool, vblank_requested ) ), TP_fast_assign( __entry->drm_id = drm_id; __entry->enc_id = enc_id; __entry->enable = enable; __entry->enabled = crtc->enabled; - __entry->suspend = crtc->suspend; - __entry->vblank_requested = crtc->vblank_requested; ), - TP_printk("id:%u encoder:%u enable:%s state{enabled:%s suspend:%s " - "vblank_req:%s}", + TP_printk("id:%u encoder:%u enable:%s state{enabled:%s}", __entry->drm_id, __entry->enc_id, __entry->enable ? "true" : "false", - __entry->enabled ? "true" : "false", - __entry->suspend ? "true" : "false", - __entry->vblank_requested ? "true" : "false") + __entry->enabled ? "true" : "false") ); DECLARE_EVENT_CLASS(dpu_crtc_enable_template, @@ -776,25 +748,15 @@ DECLARE_EVENT_CLASS(dpu_crtc_enable_template, __field( uint32_t, drm_id ) __field( bool, enable ) __field( bool, enabled ) - __field( bool, suspend ) - __field( bool, vblank_requested ) ), TP_fast_assign( __entry->drm_id = drm_id; __entry->enable = enable; __entry->enabled = crtc->enabled; - __entry->suspend = crtc->suspend; - __entry->vblank_requested = crtc->vblank_requested; ), - TP_printk("id:%u enable:%s state{enabled:%s suspend:%s vblank_req:%s}", + TP_printk("id:%u enable:%s state{enabled:%s}", __entry->drm_id, __entry->enable ? "true" : "false", - __entry->enabled ? "true" : "false", - __entry->suspend ? "true" : "false", - __entry->vblank_requested ? "true" : "false") -); -DEFINE_EVENT(dpu_crtc_enable_template, dpu_crtc_set_suspend, - TP_PROTO(uint32_t drm_id, bool enable, struct dpu_crtc *crtc), - TP_ARGS(drm_id, enable, crtc) + __entry->enabled ? "true" : "false") ); DEFINE_EVENT(dpu_crtc_enable_template, dpu_crtc_enable, TP_PROTO(uint32_t drm_id, bool enable, struct dpu_crtc *crtc), @@ -1004,6 +966,53 @@ TRACE_EVENT(dpu_core_perf_update_clk, __entry->stop_req ? "true" : "false", __entry->clk_rate) ); +TRACE_EVENT(dpu_hw_ctl_update_pending_flush, + TP_PROTO(u32 new_bits, u32 pending_mask), + TP_ARGS(new_bits, pending_mask), + TP_STRUCT__entry( + __field( u32, new_bits ) + __field( u32, pending_mask ) + ), + TP_fast_assign( + __entry->new_bits = new_bits; + __entry->pending_mask = pending_mask; + ), + TP_printk("new=%x existing=%x", __entry->new_bits, + __entry->pending_mask) +); + +DECLARE_EVENT_CLASS(dpu_hw_ctl_pending_flush_template, + TP_PROTO(u32 pending_mask, u32 ctl_flush), + TP_ARGS(pending_mask, ctl_flush), + TP_STRUCT__entry( + __field( u32, pending_mask ) + __field( u32, ctl_flush ) + ), + TP_fast_assign( + __entry->pending_mask = pending_mask; + __entry->ctl_flush = ctl_flush; + ), + TP_printk("pending_mask=%x CTL_FLUSH=%x", __entry->pending_mask, + __entry->ctl_flush) +); +DEFINE_EVENT(dpu_hw_ctl_pending_flush_template, dpu_hw_ctl_clear_pending_flush, + TP_PROTO(u32 pending_mask, u32 ctl_flush), + TP_ARGS(pending_mask, ctl_flush) +); +DEFINE_EVENT(dpu_hw_ctl_pending_flush_template, + dpu_hw_ctl_trigger_pending_flush, + TP_PROTO(u32 pending_mask, u32 ctl_flush), + TP_ARGS(pending_mask, ctl_flush) +); +DEFINE_EVENT(dpu_hw_ctl_pending_flush_template, dpu_hw_ctl_trigger_prepare, + TP_PROTO(u32 pending_mask, u32 ctl_flush), + TP_ARGS(pending_mask, ctl_flush) +); +DEFINE_EVENT(dpu_hw_ctl_pending_flush_template, dpu_hw_ctl_trigger_start, + TP_PROTO(u32 pending_mask, u32 ctl_flush), + TP_ARGS(pending_mask, ctl_flush) +); + #define DPU_ATRACE_END(name) trace_tracing_mark_write(current->tgid, name, 0) #define DPU_ATRACE_BEGIN(name) trace_tracing_mark_write(current->tgid, name, 1) #define DPU_ATRACE_FUNC() DPU_ATRACE_BEGIN(__func__) diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c index 295528292296..ef753ea9c499 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.c @@ -191,7 +191,7 @@ void dpu_vbif_set_ot_limit(struct dpu_kms *dpu_kms, ot_lim = _dpu_vbif_get_ot_limit(vbif, params) & 0xFF; if (ot_lim == 0) - goto exit; + return; trace_dpu_perf_set_ot(params->num, params->xin_id, ot_lim, params->vbif_idx); @@ -210,8 +210,6 @@ void dpu_vbif_set_ot_limit(struct dpu_kms *dpu_kms, if (forced_on) mdp->ops.setup_clk_force_ctrl(mdp, params->clk_ctrl, false); -exit: - return; } void dpu_vbif_set_qos_remap(struct dpu_kms *dpu_kms, @@ -312,31 +310,25 @@ void dpu_vbif_init_memtypes(struct dpu_kms *dpu_kms) } #ifdef CONFIG_DEBUG_FS -void dpu_debugfs_vbif_destroy(struct dpu_kms *dpu_kms) -{ - debugfs_remove_recursive(dpu_kms->debugfs_vbif); - dpu_kms->debugfs_vbif = NULL; -} -int dpu_debugfs_vbif_init(struct dpu_kms *dpu_kms, struct dentry *debugfs_root) +void dpu_debugfs_vbif_init(struct dpu_kms *dpu_kms, struct dentry *debugfs_root) { char vbif_name[32]; - struct dentry *debugfs_vbif; + struct dentry *entry, *debugfs_vbif; int i, j; - dpu_kms->debugfs_vbif = debugfs_create_dir("vbif", debugfs_root); - if (!dpu_kms->debugfs_vbif) { - DPU_ERROR("failed to create vbif debugfs\n"); - return -EINVAL; - } + entry = debugfs_create_dir("vbif", debugfs_root); + if (IS_ERR_OR_NULL(entry)) + return; for (i = 0; i < dpu_kms->catalog->vbif_count; i++) { struct dpu_vbif_cfg *vbif = &dpu_kms->catalog->vbif[i]; snprintf(vbif_name, sizeof(vbif_name), "%d", vbif->id); - debugfs_vbif = debugfs_create_dir(vbif_name, - dpu_kms->debugfs_vbif); + debugfs_vbif = debugfs_create_dir(vbif_name, entry); + if (IS_ERR_OR_NULL(debugfs_vbif)) + continue; debugfs_create_u32("features", 0600, debugfs_vbif, (u32 *)&vbif->features); @@ -378,7 +370,5 @@ int dpu_debugfs_vbif_init(struct dpu_kms *dpu_kms, struct dentry *debugfs_root) (u32 *)&cfg->ot_limit); } } - - return 0; } #endif diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.h b/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.h index f17af52dbbd5..6356876d7a66 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.h +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_vbif.h @@ -78,17 +78,6 @@ void dpu_vbif_clear_errors(struct dpu_kms *dpu_kms); */ void dpu_vbif_init_memtypes(struct dpu_kms *dpu_kms); -#ifdef CONFIG_DEBUG_FS -int dpu_debugfs_vbif_init(struct dpu_kms *dpu_kms, struct dentry *debugfs_root); -void dpu_debugfs_vbif_destroy(struct dpu_kms *dpu_kms); -#else -static inline int dpu_debugfs_vbif_init(struct dpu_kms *dpu_kms, - struct dentry *debugfs_root) -{ - return 0; -} -static inline void dpu_debugfs_vbif_destroy(struct dpu_kms *dpu_kms) -{ -} -#endif +void dpu_debugfs_vbif_init(struct dpu_kms *dpu_kms, struct dentry *debugfs_root); + #endif /* __DPU_VBIF_H__ */ diff --git a/drivers/gpu/drm/msm/disp/dpu1/msm_media_info.h b/drivers/gpu/drm/msm/disp/dpu1/msm_media_info.h index 4f12e5c534c8..9fc9dbde8a27 100644 --- a/drivers/gpu/drm/msm/disp/dpu1/msm_media_info.h +++ b/drivers/gpu/drm/msm/disp/dpu1/msm_media_info.h @@ -813,18 +813,6 @@ enum color_fmts { #define COLOR_FMT_P010_UBWC COLOR_FMT_P010_UBWC #define COLOR_FMT_P010 COLOR_FMT_P010 -static inline unsigned int VENUS_EXTRADATA_SIZE(int width, int height) -{ - (void)height; - (void)width; - - /* - * In the future, calculate the size based on the w/h but just - * hardcode it for now since 16K satisfies all current usecases. - */ - return 16 * 1024; -} - /* * Function arguments: * @color_fmt @@ -832,38 +820,32 @@ static inline unsigned int VENUS_EXTRADATA_SIZE(int width, int height) * Progressive: width * Interlaced: width */ -static inline unsigned int VENUS_Y_STRIDE(int color_fmt, int width) +static unsigned int VENUS_Y_STRIDE(int color_fmt, int width) { - unsigned int alignment, stride = 0; + unsigned int stride = 0; if (!width) - goto invalid_input; + return 0; switch (color_fmt) { case COLOR_FMT_NV21: case COLOR_FMT_NV12: case COLOR_FMT_NV12_MVTB: case COLOR_FMT_NV12_UBWC: - alignment = 128; - stride = MSM_MEDIA_ALIGN(width, alignment); + stride = MSM_MEDIA_ALIGN(width, 128); break; case COLOR_FMT_NV12_BPP10_UBWC: - alignment = 256; stride = MSM_MEDIA_ALIGN(width, 192); - stride = MSM_MEDIA_ALIGN(stride * 4/3, alignment); + stride = MSM_MEDIA_ALIGN(stride * 4 / 3, 256); break; case COLOR_FMT_P010_UBWC: - alignment = 256; - stride = MSM_MEDIA_ALIGN(width * 2, alignment); + stride = MSM_MEDIA_ALIGN(width * 2, 256); break; case COLOR_FMT_P010: - alignment = 128; - stride = MSM_MEDIA_ALIGN(width*2, alignment); - break; - default: + stride = MSM_MEDIA_ALIGN(width * 2, 128); break; } -invalid_input: + return stride; } @@ -874,38 +856,32 @@ invalid_input: * Progressive: width * Interlaced: width */ -static inline unsigned int VENUS_UV_STRIDE(int color_fmt, int width) +static unsigned int VENUS_UV_STRIDE(int color_fmt, int width) { - unsigned int alignment, stride = 0; + unsigned int stride = 0; if (!width) - goto invalid_input; + return 0; switch (color_fmt) { case COLOR_FMT_NV21: case COLOR_FMT_NV12: case COLOR_FMT_NV12_MVTB: case COLOR_FMT_NV12_UBWC: - alignment = 128; - stride = MSM_MEDIA_ALIGN(width, alignment); + stride = MSM_MEDIA_ALIGN(width, 128); break; case COLOR_FMT_NV12_BPP10_UBWC: - alignment = 256; stride = MSM_MEDIA_ALIGN(width, 192); - stride = MSM_MEDIA_ALIGN(stride * 4/3, alignment); + stride = MSM_MEDIA_ALIGN(stride * 4 / 3, 256); break; case COLOR_FMT_P010_UBWC: - alignment = 256; - stride = MSM_MEDIA_ALIGN(width * 2, alignment); + stride = MSM_MEDIA_ALIGN(width * 2, 256); break; case COLOR_FMT_P010: - alignment = 128; - stride = MSM_MEDIA_ALIGN(width*2, alignment); - break; - default: + stride = MSM_MEDIA_ALIGN(width * 2, 128); break; } -invalid_input: + return stride; } @@ -916,12 +892,12 @@ invalid_input: * Progressive: height * Interlaced: (height+1)>>1 */ -static inline unsigned int VENUS_Y_SCANLINES(int color_fmt, int height) +static unsigned int VENUS_Y_SCANLINES(int color_fmt, int height) { - unsigned int alignment, sclines = 0; + unsigned int sclines = 0; if (!height) - goto invalid_input; + return 0; switch (color_fmt) { case COLOR_FMT_NV21: @@ -929,17 +905,14 @@ static inline unsigned int VENUS_Y_SCANLINES(int color_fmt, int height) case COLOR_FMT_NV12_MVTB: case COLOR_FMT_NV12_UBWC: case COLOR_FMT_P010: - alignment = 32; + sclines = MSM_MEDIA_ALIGN(height, 32); break; case COLOR_FMT_NV12_BPP10_UBWC: case COLOR_FMT_P010_UBWC: - alignment = 16; + sclines = MSM_MEDIA_ALIGN(height, 16); break; - default: - return 0; } - sclines = MSM_MEDIA_ALIGN(height, alignment); -invalid_input: + return sclines; } @@ -950,12 +923,12 @@ invalid_input: * Progressive: height * Interlaced: (height+1)>>1 */ -static inline unsigned int VENUS_UV_SCANLINES(int color_fmt, int height) +static unsigned int VENUS_UV_SCANLINES(int color_fmt, int height) { - unsigned int alignment, sclines = 0; + unsigned int sclines = 0; if (!height) - goto invalid_input; + return 0; switch (color_fmt) { case COLOR_FMT_NV21: @@ -964,18 +937,13 @@ static inline unsigned int VENUS_UV_SCANLINES(int color_fmt, int height) case COLOR_FMT_NV12_BPP10_UBWC: case COLOR_FMT_P010_UBWC: case COLOR_FMT_P010: - alignment = 16; + sclines = MSM_MEDIA_ALIGN((height + 1) >> 1, 16); break; case COLOR_FMT_NV12_UBWC: - alignment = 32; + sclines = MSM_MEDIA_ALIGN((height + 1) >> 1, 32); break; - default: - goto invalid_input; } - sclines = MSM_MEDIA_ALIGN((height+1)>>1, alignment); - -invalid_input: return sclines; } @@ -986,12 +954,12 @@ invalid_input: * Progressive: width * Interlaced: width */ -static inline unsigned int VENUS_Y_META_STRIDE(int color_fmt, int width) +static unsigned int VENUS_Y_META_STRIDE(int color_fmt, int width) { - int y_tile_width = 0, y_meta_stride = 0; + int y_tile_width = 0, y_meta_stride; if (!width) - goto invalid_input; + return 0; switch (color_fmt) { case COLOR_FMT_NV12_UBWC: @@ -1002,14 +970,11 @@ static inline unsigned int VENUS_Y_META_STRIDE(int color_fmt, int width) y_tile_width = 48; break; default: - goto invalid_input; + return 0; } y_meta_stride = MSM_MEDIA_ROUNDUP(width, y_tile_width); - y_meta_stride = MSM_MEDIA_ALIGN(y_meta_stride, 64); - -invalid_input: - return y_meta_stride; + return MSM_MEDIA_ALIGN(y_meta_stride, 64); } /* @@ -1019,12 +984,12 @@ invalid_input: * Progressive: height * Interlaced: (height+1)>>1 */ -static inline unsigned int VENUS_Y_META_SCANLINES(int color_fmt, int height) +static unsigned int VENUS_Y_META_SCANLINES(int color_fmt, int height) { - int y_tile_height = 0, y_meta_scanlines = 0; + int y_tile_height = 0, y_meta_scanlines; if (!height) - goto invalid_input; + return 0; switch (color_fmt) { case COLOR_FMT_NV12_UBWC: @@ -1035,14 +1000,11 @@ static inline unsigned int VENUS_Y_META_SCANLINES(int color_fmt, int height) y_tile_height = 4; break; default: - goto invalid_input; + return 0; } y_meta_scanlines = MSM_MEDIA_ROUNDUP(height, y_tile_height); - y_meta_scanlines = MSM_MEDIA_ALIGN(y_meta_scanlines, 16); - -invalid_input: - return y_meta_scanlines; + return MSM_MEDIA_ALIGN(y_meta_scanlines, 16); } /* @@ -1052,12 +1014,12 @@ invalid_input: * Progressive: width * Interlaced: width */ -static inline unsigned int VENUS_UV_META_STRIDE(int color_fmt, int width) +static unsigned int VENUS_UV_META_STRIDE(int color_fmt, int width) { - int uv_tile_width = 0, uv_meta_stride = 0; + int uv_tile_width = 0, uv_meta_stride; if (!width) - goto invalid_input; + return 0; switch (color_fmt) { case COLOR_FMT_NV12_UBWC: @@ -1068,14 +1030,11 @@ static inline unsigned int VENUS_UV_META_STRIDE(int color_fmt, int width) uv_tile_width = 24; break; default: - goto invalid_input; + return 0; } uv_meta_stride = MSM_MEDIA_ROUNDUP((width+1)>>1, uv_tile_width); - uv_meta_stride = MSM_MEDIA_ALIGN(uv_meta_stride, 64); - -invalid_input: - return uv_meta_stride; + return MSM_MEDIA_ALIGN(uv_meta_stride, 64); } /* @@ -1085,12 +1044,12 @@ invalid_input: * Progressive: height * Interlaced: (height+1)>>1 */ -static inline unsigned int VENUS_UV_META_SCANLINES(int color_fmt, int height) +static unsigned int VENUS_UV_META_SCANLINES(int color_fmt, int height) { - int uv_tile_height = 0, uv_meta_scanlines = 0; + int uv_tile_height = 0, uv_meta_scanlines; if (!height) - goto invalid_input; + return 0; switch (color_fmt) { case COLOR_FMT_NV12_UBWC: @@ -1101,22 +1060,19 @@ static inline unsigned int VENUS_UV_META_SCANLINES(int color_fmt, int height) uv_tile_height = 4; break; default: - goto invalid_input; + return 0; } uv_meta_scanlines = MSM_MEDIA_ROUNDUP((height+1)>>1, uv_tile_height); - uv_meta_scanlines = MSM_MEDIA_ALIGN(uv_meta_scanlines, 16); - -invalid_input: - return uv_meta_scanlines; + return MSM_MEDIA_ALIGN(uv_meta_scanlines, 16); } -static inline unsigned int VENUS_RGB_STRIDE(int color_fmt, int width) +static unsigned int VENUS_RGB_STRIDE(int color_fmt, int width) { - unsigned int alignment = 0, stride = 0, bpp = 4; + unsigned int alignment = 0, bpp = 4; if (!width) - goto invalid_input; + return 0; switch (color_fmt) { case COLOR_FMT_RGBA8888: @@ -1131,21 +1087,18 @@ static inline unsigned int VENUS_RGB_STRIDE(int color_fmt, int width) alignment = 256; break; default: - goto invalid_input; + return 0; } - stride = MSM_MEDIA_ALIGN(width * bpp, alignment); - -invalid_input: - return stride; + return MSM_MEDIA_ALIGN(width * bpp, alignment); } -static inline unsigned int VENUS_RGB_SCANLINES(int color_fmt, int height) +static unsigned int VENUS_RGB_SCANLINES(int color_fmt, int height) { - unsigned int alignment = 0, scanlines = 0; + unsigned int alignment = 0; if (!height) - goto invalid_input; + return 0; switch (color_fmt) { case COLOR_FMT_RGBA8888: @@ -1157,220 +1110,46 @@ static inline unsigned int VENUS_RGB_SCANLINES(int color_fmt, int height) alignment = 16; break; default: - goto invalid_input; + return 0; } - scanlines = MSM_MEDIA_ALIGN(height, alignment); - -invalid_input: - return scanlines; + return MSM_MEDIA_ALIGN(height, alignment); } -static inline unsigned int VENUS_RGB_META_STRIDE(int color_fmt, int width) +static unsigned int VENUS_RGB_META_STRIDE(int color_fmt, int width) { - int rgb_tile_width = 0, rgb_meta_stride = 0; + int rgb_meta_stride; if (!width) - goto invalid_input; + return 0; switch (color_fmt) { case COLOR_FMT_RGBA8888_UBWC: case COLOR_FMT_RGBA1010102_UBWC: case COLOR_FMT_RGB565_UBWC: - rgb_tile_width = 16; - break; - default: - goto invalid_input; + rgb_meta_stride = MSM_MEDIA_ROUNDUP(width, 16); + return MSM_MEDIA_ALIGN(rgb_meta_stride, 64); } - rgb_meta_stride = MSM_MEDIA_ROUNDUP(width, rgb_tile_width); - rgb_meta_stride = MSM_MEDIA_ALIGN(rgb_meta_stride, 64); - -invalid_input: - return rgb_meta_stride; + return 0; } -static inline unsigned int VENUS_RGB_META_SCANLINES(int color_fmt, int height) +static unsigned int VENUS_RGB_META_SCANLINES(int color_fmt, int height) { - int rgb_tile_height = 0, rgb_meta_scanlines = 0; + int rgb_meta_scanlines; if (!height) - goto invalid_input; + return 0; switch (color_fmt) { case COLOR_FMT_RGBA8888_UBWC: case COLOR_FMT_RGBA1010102_UBWC: case COLOR_FMT_RGB565_UBWC: - rgb_tile_height = 4; - break; - default: - goto invalid_input; + rgb_meta_scanlines = MSM_MEDIA_ROUNDUP(height, 4); + return MSM_MEDIA_ALIGN(rgb_meta_scanlines, 16); } - rgb_meta_scanlines = MSM_MEDIA_ROUNDUP(height, rgb_tile_height); - rgb_meta_scanlines = MSM_MEDIA_ALIGN(rgb_meta_scanlines, 16); - -invalid_input: - return rgb_meta_scanlines; -} - -/* - * Function arguments: - * @color_fmt - * @width - * Progressive: width - * Interlaced: width - * @height - * Progressive: height - * Interlaced: height - */ -static inline unsigned int VENUS_BUFFER_SIZE( - int color_fmt, int width, int height) -{ - const unsigned int extra_size = VENUS_EXTRADATA_SIZE(width, height); - unsigned int uv_alignment = 0, size = 0; - unsigned int y_plane, uv_plane, y_stride, - uv_stride, y_sclines, uv_sclines; - unsigned int y_ubwc_plane = 0, uv_ubwc_plane = 0; - unsigned int y_meta_stride = 0, y_meta_scanlines = 0; - unsigned int uv_meta_stride = 0, uv_meta_scanlines = 0; - unsigned int y_meta_plane = 0, uv_meta_plane = 0; - unsigned int rgb_stride = 0, rgb_scanlines = 0; - unsigned int rgb_plane = 0, rgb_ubwc_plane = 0, rgb_meta_plane = 0; - unsigned int rgb_meta_stride = 0, rgb_meta_scanlines = 0; - - if (!width || !height) - goto invalid_input; - - y_stride = VENUS_Y_STRIDE(color_fmt, width); - uv_stride = VENUS_UV_STRIDE(color_fmt, width); - y_sclines = VENUS_Y_SCANLINES(color_fmt, height); - uv_sclines = VENUS_UV_SCANLINES(color_fmt, height); - rgb_stride = VENUS_RGB_STRIDE(color_fmt, width); - rgb_scanlines = VENUS_RGB_SCANLINES(color_fmt, height); - - switch (color_fmt) { - case COLOR_FMT_NV21: - case COLOR_FMT_NV12: - case COLOR_FMT_P010: - uv_alignment = 4096; - y_plane = y_stride * y_sclines; - uv_plane = uv_stride * uv_sclines + uv_alignment; - size = y_plane + uv_plane + - MSM_MEDIA_MAX(extra_size, 8 * y_stride); - size = MSM_MEDIA_ALIGN(size, 4096); - break; - case COLOR_FMT_NV12_MVTB: - uv_alignment = 4096; - y_plane = y_stride * y_sclines; - uv_plane = uv_stride * uv_sclines + uv_alignment; - size = y_plane + uv_plane; - size = 2 * size + extra_size; - size = MSM_MEDIA_ALIGN(size, 4096); - break; - case COLOR_FMT_NV12_UBWC: - y_sclines = VENUS_Y_SCANLINES(color_fmt, (height+1)>>1); - y_ubwc_plane = MSM_MEDIA_ALIGN(y_stride * y_sclines, 4096); - uv_sclines = VENUS_UV_SCANLINES(color_fmt, (height+1)>>1); - uv_ubwc_plane = MSM_MEDIA_ALIGN(uv_stride * uv_sclines, 4096); - y_meta_stride = VENUS_Y_META_STRIDE(color_fmt, width); - y_meta_scanlines = - VENUS_Y_META_SCANLINES(color_fmt, (height+1)>>1); - y_meta_plane = MSM_MEDIA_ALIGN( - y_meta_stride * y_meta_scanlines, 4096); - uv_meta_stride = VENUS_UV_META_STRIDE(color_fmt, width); - uv_meta_scanlines = - VENUS_UV_META_SCANLINES(color_fmt, (height+1)>>1); - uv_meta_plane = MSM_MEDIA_ALIGN(uv_meta_stride * - uv_meta_scanlines, 4096); - - size = (y_ubwc_plane + uv_ubwc_plane + y_meta_plane + - uv_meta_plane)*2 + - MSM_MEDIA_MAX(extra_size + 8192, 48 * y_stride); - size = MSM_MEDIA_ALIGN(size, 4096); - break; - case COLOR_FMT_NV12_BPP10_UBWC: - y_ubwc_plane = MSM_MEDIA_ALIGN(y_stride * y_sclines, 4096); - uv_ubwc_plane = MSM_MEDIA_ALIGN(uv_stride * uv_sclines, 4096); - y_meta_stride = VENUS_Y_META_STRIDE(color_fmt, width); - y_meta_scanlines = VENUS_Y_META_SCANLINES(color_fmt, height); - y_meta_plane = MSM_MEDIA_ALIGN( - y_meta_stride * y_meta_scanlines, 4096); - uv_meta_stride = VENUS_UV_META_STRIDE(color_fmt, width); - uv_meta_scanlines = VENUS_UV_META_SCANLINES(color_fmt, height); - uv_meta_plane = MSM_MEDIA_ALIGN(uv_meta_stride * - uv_meta_scanlines, 4096); - - size = y_ubwc_plane + uv_ubwc_plane + y_meta_plane + - uv_meta_plane + - MSM_MEDIA_MAX(extra_size + 8192, 48 * y_stride); - size = MSM_MEDIA_ALIGN(size, 4096); - break; - case COLOR_FMT_P010_UBWC: - y_ubwc_plane = MSM_MEDIA_ALIGN(y_stride * y_sclines, 4096); - uv_ubwc_plane = MSM_MEDIA_ALIGN(uv_stride * uv_sclines, 4096); - y_meta_stride = VENUS_Y_META_STRIDE(color_fmt, width); - y_meta_scanlines = VENUS_Y_META_SCANLINES(color_fmt, height); - y_meta_plane = MSM_MEDIA_ALIGN( - y_meta_stride * y_meta_scanlines, 4096); - uv_meta_stride = VENUS_UV_META_STRIDE(color_fmt, width); - uv_meta_scanlines = VENUS_UV_META_SCANLINES(color_fmt, height); - uv_meta_plane = MSM_MEDIA_ALIGN(uv_meta_stride * - uv_meta_scanlines, 4096); - - size = y_ubwc_plane + uv_ubwc_plane + y_meta_plane + - uv_meta_plane; - size = MSM_MEDIA_ALIGN(size, 4096); - break; - case COLOR_FMT_RGBA8888: - rgb_plane = MSM_MEDIA_ALIGN(rgb_stride * rgb_scanlines, 4096); - size = rgb_plane; - size = MSM_MEDIA_ALIGN(size, 4096); - break; - case COLOR_FMT_RGBA8888_UBWC: - case COLOR_FMT_RGBA1010102_UBWC: - case COLOR_FMT_RGB565_UBWC: - rgb_ubwc_plane = MSM_MEDIA_ALIGN(rgb_stride * rgb_scanlines, - 4096); - rgb_meta_stride = VENUS_RGB_META_STRIDE(color_fmt, width); - rgb_meta_scanlines = VENUS_RGB_META_SCANLINES(color_fmt, - height); - rgb_meta_plane = MSM_MEDIA_ALIGN(rgb_meta_stride * - rgb_meta_scanlines, 4096); - size = rgb_ubwc_plane + rgb_meta_plane; - size = MSM_MEDIA_ALIGN(size, 4096); - break; - default: - break; - } -invalid_input: - return size; -} - -static inline unsigned int VENUS_VIEW2_OFFSET( - int color_fmt, int width, int height) -{ - unsigned int offset = 0; - unsigned int y_plane, uv_plane, y_stride, - uv_stride, y_sclines, uv_sclines; - if (!width || !height) - goto invalid_input; - - y_stride = VENUS_Y_STRIDE(color_fmt, width); - uv_stride = VENUS_UV_STRIDE(color_fmt, width); - y_sclines = VENUS_Y_SCANLINES(color_fmt, height); - uv_sclines = VENUS_UV_SCANLINES(color_fmt, height); - switch (color_fmt) { - case COLOR_FMT_NV12_MVTB: - y_plane = y_stride * y_sclines; - uv_plane = uv_stride * uv_sclines; - offset = y_plane + uv_plane; - break; - default: - break; - } -invalid_input: - return offset; + return 0; } #endif diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c index 7b028f778960..cc32ea5f4289 100644 --- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_crtc.c @@ -128,7 +128,7 @@ static void unref_cursor_worker(struct drm_flip_work *work, void *val) struct mdp4_kms *mdp4_kms = get_kms(&mdp4_crtc->base); struct msm_kms *kms = &mdp4_kms->base.base; - msm_gem_put_iova(val, kms->aspace); + msm_gem_unpin_iova(val, kms->aspace); drm_gem_object_put_unlocked(val); } @@ -378,7 +378,7 @@ static void update_cursor(struct drm_crtc *crtc) if (next_bo) { /* take a obj ref + iova ref when we start scanning out: */ drm_gem_object_get(next_bo); - msm_gem_get_iova(next_bo, kms->aspace, &iova); + msm_gem_get_and_pin_iova(next_bo, kms->aspace, &iova); /* enable cursor: */ mdp4_write(mdp4_kms, REG_MDP4_DMA_CURSOR_SIZE(dma), @@ -423,7 +423,7 @@ static int mdp4_crtc_cursor_set(struct drm_crtc *crtc, int ret; if ((width > CURSOR_WIDTH) || (height > CURSOR_HEIGHT)) { - dev_err(dev->dev, "bad cursor size: %dx%d\n", width, height); + DRM_DEV_ERROR(dev->dev, "bad cursor size: %dx%d\n", width, height); return -EINVAL; } @@ -436,7 +436,7 @@ static int mdp4_crtc_cursor_set(struct drm_crtc *crtc, } if (cursor_bo) { - ret = msm_gem_get_iova(cursor_bo, kms->aspace, &iova); + ret = msm_gem_get_and_pin_iova(cursor_bo, kms->aspace, &iova); if (ret) goto fail; } else { diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_dtv_encoder.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_dtv_encoder.c index f6bc86a35d8d..ff8f2da160c0 100644 --- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_dtv_encoder.c +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_dtv_encoder.c @@ -45,7 +45,7 @@ static void bs_init(struct mdp4_dtv_encoder *mdp4_dtv_encoder) struct lcdc_platform_data *dtv_pdata = mdp4_find_pdata("dtv.0"); if (!dtv_pdata) { - dev_err(dev->dev, "could not find dtv pdata\n"); + DRM_DEV_ERROR(dev->dev, "could not find dtv pdata\n"); return; } @@ -202,16 +202,16 @@ static void mdp4_dtv_encoder_enable(struct drm_encoder *encoder) ret = clk_set_rate(mdp4_dtv_encoder->mdp_clk, pc); if (ret) - dev_err(dev->dev, "failed to set mdp_clk to %lu: %d\n", + DRM_DEV_ERROR(dev->dev, "failed to set mdp_clk to %lu: %d\n", pc, ret); ret = clk_prepare_enable(mdp4_dtv_encoder->mdp_clk); if (ret) - dev_err(dev->dev, "failed to enabled mdp_clk: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to enabled mdp_clk: %d\n", ret); ret = clk_prepare_enable(mdp4_dtv_encoder->hdmi_clk); if (ret) - dev_err(dev->dev, "failed to enable hdmi_clk: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to enable hdmi_clk: %d\n", ret); mdp4_write(mdp4_kms, REG_MDP4_DTV_ENABLE, 1); @@ -251,14 +251,14 @@ struct drm_encoder *mdp4_dtv_encoder_init(struct drm_device *dev) mdp4_dtv_encoder->hdmi_clk = devm_clk_get(dev->dev, "hdmi_clk"); if (IS_ERR(mdp4_dtv_encoder->hdmi_clk)) { - dev_err(dev->dev, "failed to get hdmi_clk\n"); + DRM_DEV_ERROR(dev->dev, "failed to get hdmi_clk\n"); ret = PTR_ERR(mdp4_dtv_encoder->hdmi_clk); goto fail; } mdp4_dtv_encoder->mdp_clk = devm_clk_get(dev->dev, "tv_clk"); if (IS_ERR(mdp4_dtv_encoder->mdp_clk)) { - dev_err(dev->dev, "failed to get tv_clk\n"); + DRM_DEV_ERROR(dev->dev, "failed to get tv_clk\n"); ret = PTR_ERR(mdp4_dtv_encoder->mdp_clk); goto fail; } diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c index 44d1cda56974..e437aa806f7b 100644 --- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_kms.c @@ -43,7 +43,7 @@ static int mdp4_hw_init(struct msm_kms *kms) DBG("found MDP4 version v%d.%d", major, minor); if (major != 4) { - dev_err(dev->dev, "unexpected MDP version: v%d.%d\n", + DRM_DEV_ERROR(dev->dev, "unexpected MDP version: v%d.%d\n", major, minor); ret = -ENXIO; goto out; @@ -165,7 +165,7 @@ static void mdp4_destroy(struct msm_kms *kms) struct msm_gem_address_space *aspace = kms->aspace; if (mdp4_kms->blank_cursor_iova) - msm_gem_put_iova(mdp4_kms->blank_cursor_bo, kms->aspace); + msm_gem_unpin_iova(mdp4_kms->blank_cursor_bo, kms->aspace); drm_gem_object_put_unlocked(mdp4_kms->blank_cursor_bo); if (aspace) { @@ -206,7 +206,8 @@ int mdp4_disable(struct mdp4_kms *mdp4_kms) clk_disable_unprepare(mdp4_kms->clk); if (mdp4_kms->pclk) clk_disable_unprepare(mdp4_kms->pclk); - clk_disable_unprepare(mdp4_kms->lut_clk); + if (mdp4_kms->lut_clk) + clk_disable_unprepare(mdp4_kms->lut_clk); if (mdp4_kms->axi_clk) clk_disable_unprepare(mdp4_kms->axi_clk); @@ -220,7 +221,8 @@ int mdp4_enable(struct mdp4_kms *mdp4_kms) clk_prepare_enable(mdp4_kms->clk); if (mdp4_kms->pclk) clk_prepare_enable(mdp4_kms->pclk); - clk_prepare_enable(mdp4_kms->lut_clk); + if (mdp4_kms->lut_clk) + clk_prepare_enable(mdp4_kms->lut_clk); if (mdp4_kms->axi_clk) clk_prepare_enable(mdp4_kms->axi_clk); @@ -251,7 +253,7 @@ static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms, encoder = mdp4_lcdc_encoder_init(dev, panel_node); if (IS_ERR(encoder)) { - dev_err(dev->dev, "failed to construct LCDC encoder\n"); + DRM_DEV_ERROR(dev->dev, "failed to construct LCDC encoder\n"); return PTR_ERR(encoder); } @@ -260,7 +262,7 @@ static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms, connector = mdp4_lvds_connector_init(dev, panel_node, encoder); if (IS_ERR(connector)) { - dev_err(dev->dev, "failed to initialize LVDS connector\n"); + DRM_DEV_ERROR(dev->dev, "failed to initialize LVDS connector\n"); return PTR_ERR(connector); } @@ -271,7 +273,7 @@ static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms, case DRM_MODE_ENCODER_TMDS: encoder = mdp4_dtv_encoder_init(dev); if (IS_ERR(encoder)) { - dev_err(dev->dev, "failed to construct DTV encoder\n"); + DRM_DEV_ERROR(dev->dev, "failed to construct DTV encoder\n"); return PTR_ERR(encoder); } @@ -282,7 +284,7 @@ static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms, /* Construct bridge/connector for HDMI: */ ret = msm_hdmi_modeset_init(priv->hdmi, dev, encoder); if (ret) { - dev_err(dev->dev, "failed to initialize HDMI: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to initialize HDMI: %d\n", ret); return ret; } } @@ -300,7 +302,7 @@ static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms, encoder = mdp4_dsi_encoder_init(dev); if (IS_ERR(encoder)) { ret = PTR_ERR(encoder); - dev_err(dev->dev, + DRM_DEV_ERROR(dev->dev, "failed to construct DSI encoder: %d\n", ret); return ret; } @@ -311,14 +313,14 @@ static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms, ret = msm_dsi_modeset_init(priv->dsi[dsi_id], dev, encoder); if (ret) { - dev_err(dev->dev, "failed to initialize DSI: %d\n", + DRM_DEV_ERROR(dev->dev, "failed to initialize DSI: %d\n", ret); return ret; } break; default: - dev_err(dev->dev, "Invalid or unsupported interface\n"); + DRM_DEV_ERROR(dev->dev, "Invalid or unsupported interface\n"); return -EINVAL; } @@ -354,7 +356,7 @@ static int modeset_init(struct mdp4_kms *mdp4_kms) for (i = 0; i < ARRAY_SIZE(vg_planes); i++) { plane = mdp4_plane_init(dev, vg_planes[i], false); if (IS_ERR(plane)) { - dev_err(dev->dev, + DRM_DEV_ERROR(dev->dev, "failed to construct plane for VG%d\n", i + 1); ret = PTR_ERR(plane); goto fail; @@ -365,7 +367,7 @@ static int modeset_init(struct mdp4_kms *mdp4_kms) for (i = 0; i < ARRAY_SIZE(mdp4_crtcs); i++) { plane = mdp4_plane_init(dev, rgb_planes[i], true); if (IS_ERR(plane)) { - dev_err(dev->dev, + DRM_DEV_ERROR(dev->dev, "failed to construct plane for RGB%d\n", i + 1); ret = PTR_ERR(plane); goto fail; @@ -374,7 +376,7 @@ static int modeset_init(struct mdp4_kms *mdp4_kms) crtc = mdp4_crtc_init(dev, plane, priv->num_crtcs, i, mdp4_crtcs[i]); if (IS_ERR(crtc)) { - dev_err(dev->dev, "failed to construct crtc for %s\n", + DRM_DEV_ERROR(dev->dev, "failed to construct crtc for %s\n", mdp4_crtc_names[i]); ret = PTR_ERR(crtc); goto fail; @@ -396,7 +398,7 @@ static int modeset_init(struct mdp4_kms *mdp4_kms) for (i = 0; i < ARRAY_SIZE(mdp4_intfs); i++) { ret = mdp4_modeset_init_intf(mdp4_kms, mdp4_intfs[i]); if (ret) { - dev_err(dev->dev, "failed to initialize intf: %d, %d\n", + DRM_DEV_ERROR(dev->dev, "failed to initialize intf: %d, %d\n", i, ret); goto fail; } @@ -419,7 +421,7 @@ struct msm_kms *mdp4_kms_init(struct drm_device *dev) mdp4_kms = kzalloc(sizeof(*mdp4_kms), GFP_KERNEL); if (!mdp4_kms) { - dev_err(dev->dev, "failed to allocate kms\n"); + DRM_DEV_ERROR(dev->dev, "failed to allocate kms\n"); ret = -ENOMEM; goto fail; } @@ -439,7 +441,7 @@ struct msm_kms *mdp4_kms_init(struct drm_device *dev) irq = platform_get_irq(pdev, 0); if (irq < 0) { ret = irq; - dev_err(dev->dev, "failed to get irq: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to get irq: %d\n", ret); goto fail; } @@ -456,14 +458,14 @@ struct msm_kms *mdp4_kms_init(struct drm_device *dev) if (mdp4_kms->vdd) { ret = regulator_enable(mdp4_kms->vdd); if (ret) { - dev_err(dev->dev, "failed to enable regulator vdd: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to enable regulator vdd: %d\n", ret); goto fail; } } mdp4_kms->clk = devm_clk_get(&pdev->dev, "core_clk"); if (IS_ERR(mdp4_kms->clk)) { - dev_err(dev->dev, "failed to get core_clk\n"); + DRM_DEV_ERROR(dev->dev, "failed to get core_clk\n"); ret = PTR_ERR(mdp4_kms->clk); goto fail; } @@ -472,23 +474,25 @@ struct msm_kms *mdp4_kms_init(struct drm_device *dev) if (IS_ERR(mdp4_kms->pclk)) mdp4_kms->pclk = NULL; - // XXX if (rev >= MDP_REV_42) { ??? - mdp4_kms->lut_clk = devm_clk_get(&pdev->dev, "lut_clk"); - if (IS_ERR(mdp4_kms->lut_clk)) { - dev_err(dev->dev, "failed to get lut_clk\n"); - ret = PTR_ERR(mdp4_kms->lut_clk); - goto fail; + if (mdp4_kms->rev >= 2) { + mdp4_kms->lut_clk = devm_clk_get(&pdev->dev, "lut_clk"); + if (IS_ERR(mdp4_kms->lut_clk)) { + DRM_DEV_ERROR(dev->dev, "failed to get lut_clk\n"); + ret = PTR_ERR(mdp4_kms->lut_clk); + goto fail; + } } mdp4_kms->axi_clk = devm_clk_get(&pdev->dev, "bus_clk"); if (IS_ERR(mdp4_kms->axi_clk)) { - dev_err(dev->dev, "failed to get axi_clk\n"); + DRM_DEV_ERROR(dev->dev, "failed to get axi_clk\n"); ret = PTR_ERR(mdp4_kms->axi_clk); goto fail; } clk_set_rate(mdp4_kms->clk, config->max_clk); - clk_set_rate(mdp4_kms->lut_clk, config->max_clk); + if (mdp4_kms->lut_clk) + clk_set_rate(mdp4_kms->lut_clk, config->max_clk); pm_runtime_enable(dev->dev); mdp4_kms->rpm_enabled = true; @@ -519,29 +523,29 @@ struct msm_kms *mdp4_kms_init(struct drm_device *dev) if (ret) goto fail; } else { - dev_info(dev->dev, "no iommu, fallback to phys " + DRM_DEV_INFO(dev->dev, "no iommu, fallback to phys " "contig buffers for scanout\n"); aspace = NULL; } ret = modeset_init(mdp4_kms); if (ret) { - dev_err(dev->dev, "modeset_init failed: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "modeset_init failed: %d\n", ret); goto fail; } - mdp4_kms->blank_cursor_bo = msm_gem_new(dev, SZ_16K, MSM_BO_WC); + mdp4_kms->blank_cursor_bo = msm_gem_new(dev, SZ_16K, MSM_BO_WC | MSM_BO_SCANOUT); if (IS_ERR(mdp4_kms->blank_cursor_bo)) { ret = PTR_ERR(mdp4_kms->blank_cursor_bo); - dev_err(dev->dev, "could not allocate blank-cursor bo: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "could not allocate blank-cursor bo: %d\n", ret); mdp4_kms->blank_cursor_bo = NULL; goto fail; } - ret = msm_gem_get_iova(mdp4_kms->blank_cursor_bo, kms->aspace, + ret = msm_gem_get_and_pin_iova(mdp4_kms->blank_cursor_bo, kms->aspace, &mdp4_kms->blank_cursor_iova); if (ret) { - dev_err(dev->dev, "could not pin blank-cursor bo: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "could not pin blank-cursor bo: %d\n", ret); goto fail; } diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c index d47b8f4af991..fff77a4b12c2 100644 --- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_lcdc_encoder.c @@ -47,7 +47,7 @@ static void bs_init(struct mdp4_lcdc_encoder *mdp4_lcdc_encoder) struct lcdc_platform_data *lcdc_pdata = mdp4_find_pdata("lvds.0"); if (!lcdc_pdata) { - dev_err(dev->dev, "could not find lvds pdata\n"); + DRM_DEV_ERROR(dev->dev, "could not find lvds pdata\n"); return; } @@ -224,7 +224,7 @@ static void setup_phy(struct drm_encoder *encoder) break; default: - dev_err(dev->dev, "unknown bpp: %d\n", bpp); + DRM_DEV_ERROR(dev->dev, "unknown bpp: %d\n", bpp); return; } @@ -241,7 +241,7 @@ static void setup_phy(struct drm_encoder *encoder) MDP4_LCDC_LVDS_INTF_CTL_CH1_CLK_LANE_EN; break; default: - dev_err(dev->dev, "unknown # of channels: %d\n", nchan); + DRM_DEV_ERROR(dev->dev, "unknown # of channels: %d\n", nchan); return; } @@ -354,7 +354,7 @@ static void mdp4_lcdc_encoder_disable(struct drm_encoder *encoder) for (i = 0; i < ARRAY_SIZE(mdp4_lcdc_encoder->regs); i++) { ret = regulator_disable(mdp4_lcdc_encoder->regs[i]); if (ret) - dev_err(dev->dev, "failed to disable regulator: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to disable regulator: %d\n", ret); } bs_set(mdp4_lcdc_encoder, 0); @@ -370,20 +370,25 @@ static void mdp4_lcdc_encoder_enable(struct drm_encoder *encoder) unsigned long pc = mdp4_lcdc_encoder->pixclock; struct mdp4_kms *mdp4_kms = get_kms(encoder); struct drm_panel *panel; + uint32_t config; int i, ret; if (WARN_ON(mdp4_lcdc_encoder->enabled)) return; /* TODO: hard-coded for 18bpp: */ - mdp4_crtc_set_config(encoder->crtc, - MDP4_DMA_CONFIG_R_BPC(BPC6) | - MDP4_DMA_CONFIG_G_BPC(BPC6) | - MDP4_DMA_CONFIG_B_BPC(BPC6) | - MDP4_DMA_CONFIG_PACK_ALIGN_MSB | - MDP4_DMA_CONFIG_PACK(0x21) | - MDP4_DMA_CONFIG_DEFLKR_EN | - MDP4_DMA_CONFIG_DITHER_EN); + config = + MDP4_DMA_CONFIG_R_BPC(BPC6) | + MDP4_DMA_CONFIG_G_BPC(BPC6) | + MDP4_DMA_CONFIG_B_BPC(BPC6) | + MDP4_DMA_CONFIG_PACK(0x21) | + MDP4_DMA_CONFIG_DEFLKR_EN | + MDP4_DMA_CONFIG_DITHER_EN; + + if (!of_property_read_bool(dev->dev->of_node, "qcom,lcdc-align-lsb")) + config |= MDP4_DMA_CONFIG_PACK_ALIGN_MSB; + + mdp4_crtc_set_config(encoder->crtc, config); mdp4_crtc_set_intf(encoder->crtc, INTF_LCDC_DTV, 0); bs_set(mdp4_lcdc_encoder, 1); @@ -391,16 +396,16 @@ static void mdp4_lcdc_encoder_enable(struct drm_encoder *encoder) for (i = 0; i < ARRAY_SIZE(mdp4_lcdc_encoder->regs); i++) { ret = regulator_enable(mdp4_lcdc_encoder->regs[i]); if (ret) - dev_err(dev->dev, "failed to enable regulator: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to enable regulator: %d\n", ret); } DBG("setting lcdc_clk=%lu", pc); ret = clk_set_rate(mdp4_lcdc_encoder->lcdc_clk, pc); if (ret) - dev_err(dev->dev, "failed to configure lcdc_clk: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to configure lcdc_clk: %d\n", ret); ret = clk_prepare_enable(mdp4_lcdc_encoder->lcdc_clk); if (ret) - dev_err(dev->dev, "failed to enable lcdc_clk: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to enable lcdc_clk: %d\n", ret); panel = of_drm_find_panel(mdp4_lcdc_encoder->panel_node); if (!IS_ERR(panel)) { @@ -454,7 +459,7 @@ struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev, /* TODO: do we need different pll in other cases? */ mdp4_lcdc_encoder->lcdc_clk = mpd4_lvds_pll_init(dev); if (IS_ERR(mdp4_lcdc_encoder->lcdc_clk)) { - dev_err(dev->dev, "failed to get lvds_clk\n"); + DRM_DEV_ERROR(dev->dev, "failed to get lvds_clk\n"); ret = PTR_ERR(mdp4_lcdc_encoder->lcdc_clk); goto fail; } @@ -463,7 +468,7 @@ struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev, reg = devm_regulator_get(dev->dev, "lvds-vccs-3p3v"); if (IS_ERR(reg)) { ret = PTR_ERR(reg); - dev_err(dev->dev, "failed to get lvds-vccs-3p3v: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to get lvds-vccs-3p3v: %d\n", ret); goto fail; } mdp4_lcdc_encoder->regs[0] = reg; @@ -471,7 +476,7 @@ struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev, reg = devm_regulator_get(dev->dev, "lvds-pll-vdda"); if (IS_ERR(reg)) { ret = PTR_ERR(reg); - dev_err(dev->dev, "failed to get lvds-pll-vdda: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to get lvds-pll-vdda: %d\n", ret); goto fail; } mdp4_lcdc_encoder->regs[1] = reg; @@ -479,7 +484,7 @@ struct drm_encoder *mdp4_lcdc_encoder_init(struct drm_device *dev, reg = devm_regulator_get(dev->dev, "lvds-vdda"); if (IS_ERR(reg)) { ret = PTR_ERR(reg); - dev_err(dev->dev, "failed to get lvds-vdda: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to get lvds-vdda: %d\n", ret); goto fail; } mdp4_lcdc_encoder->regs[2] = reg; diff --git a/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c b/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c index 7a499731ce93..005066f7154d 100644 --- a/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c +++ b/drivers/gpu/drm/msm/disp/mdp4/mdp4_plane.c @@ -234,22 +234,22 @@ static int mdp4_plane_mode_set(struct drm_plane *plane, format = to_mdp_format(msm_framebuffer_format(fb)); if (src_w > (crtc_w * DOWN_SCALE_MAX)) { - dev_err(dev->dev, "Width down scaling exceeds limits!\n"); + DRM_DEV_ERROR(dev->dev, "Width down scaling exceeds limits!\n"); return -ERANGE; } if (src_h > (crtc_h * DOWN_SCALE_MAX)) { - dev_err(dev->dev, "Height down scaling exceeds limits!\n"); + DRM_DEV_ERROR(dev->dev, "Height down scaling exceeds limits!\n"); return -ERANGE; } if (crtc_w > (src_w * UP_SCALE_MAX)) { - dev_err(dev->dev, "Width up scaling exceeds limits!\n"); + DRM_DEV_ERROR(dev->dev, "Width up scaling exceeds limits!\n"); return -ERANGE; } if (crtc_h > (src_h * UP_SCALE_MAX)) { - dev_err(dev->dev, "Height up scaling exceeds limits!\n"); + DRM_DEV_ERROR(dev->dev, "Height up scaling exceeds limits!\n"); return -ERANGE; } diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c index 824067d2d427..ea8f7d7daf7f 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c @@ -553,6 +553,91 @@ const struct mdp5_cfg_hw msm8x96_config = { .max_clk = 412500000, }; +const struct mdp5_cfg_hw msm8917_config = { + .name = "msm8917", + .mdp = { + .count = 1, + .caps = MDP_CAP_CDM, + }, + .ctl = { + .count = 3, + .base = { 0x01000, 0x01200, 0x01400 }, + .flush_hw_mask = 0xffffffff, + }, + .pipe_vig = { + .count = 1, + .base = { 0x04000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SCALE | + MDP_PIPE_CAP_CSC | + MDP_PIPE_CAP_DECIMATION | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_rgb = { + .count = 2, + .base = { 0x14000, 0x16000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_DECIMATION | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_dma = { + .count = 1, + .base = { 0x24000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SW_PIX_EXT | + 0, + }, + .pipe_cursor = { + .count = 1, + .base = { 0x34000 }, + .caps = MDP_PIPE_CAP_HFLIP | + MDP_PIPE_CAP_VFLIP | + MDP_PIPE_CAP_SW_PIX_EXT | + MDP_PIPE_CAP_CURSOR | + 0, + }, + + .lm = { + .count = 2, + .base = { 0x44000, 0x45000 }, + .instances = { + { .id = 0, .pp = 0, .dspp = 0, + .caps = MDP_LM_CAP_DISPLAY, }, + { .id = 1, .pp = -1, .dspp = -1, + .caps = MDP_LM_CAP_WB }, + }, + .nb_stages = 8, + .max_width = 2048, + .max_height = 0xFFFF, + }, + .dspp = { + .count = 1, + .base = { 0x54000 }, + + }, + .pp = { + .count = 1, + .base = { 0x70000 }, + }, + .cdm = { + .count = 1, + .base = { 0x79200 }, + }, + .intf = { + .base = { 0x6a000, 0x6a800 }, + .connect = { + [0] = INTF_DISABLED, + [1] = INTF_DSI, + }, + }, + .max_clk = 320000000, +}; + static const struct mdp5_cfg_handler cfg_handlers[] = { { .revision = 0, .config = { .hw = &msm8x74v1_config } }, { .revision = 2, .config = { .hw = &msm8x74v2_config } }, @@ -560,6 +645,7 @@ static const struct mdp5_cfg_handler cfg_handlers[] = { { .revision = 6, .config = { .hw = &msm8x16_config } }, { .revision = 9, .config = { .hw = &msm8x94_config } }, { .revision = 7, .config = { .hw = &msm8x96_config } }, + { .revision = 15, .config = { .hw = &msm8917_config } }, }; static struct mdp5_cfg_platform *mdp5_get_config(struct platform_device *dev); @@ -600,7 +686,7 @@ struct mdp5_cfg_handler *mdp5_cfg_init(struct mdp5_kms *mdp5_kms, } if (major != 1) { - dev_err(dev->dev, "unexpected MDP major version: v%d.%d\n", + DRM_DEV_ERROR(dev->dev, "unexpected MDP major version: v%d.%d\n", major, minor); ret = -ENXIO; goto fail; @@ -615,7 +701,7 @@ struct mdp5_cfg_handler *mdp5_cfg_init(struct mdp5_kms *mdp5_kms, break; } if (unlikely(!mdp5_cfg)) { - dev_err(dev->dev, "unexpected MDP minor revision: v%d.%d\n", + DRM_DEV_ERROR(dev->dev, "unexpected MDP minor revision: v%d.%d\n", major, minor); ret = -ENXIO; goto fail; diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c index 33972c1816ed..976585d8bfd6 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cmd_encoder.c @@ -55,20 +55,20 @@ static int pingpong_tearcheck_setup(struct drm_encoder *encoder, int pp_id = mixer->pp; if (IS_ERR_OR_NULL(mdp5_kms->vsync_clk)) { - dev_err(dev, "vsync_clk is not initialized\n"); + DRM_DEV_ERROR(dev, "vsync_clk is not initialized\n"); return -EINVAL; } total_lines_x100 = mode->vtotal * mode->vrefresh; if (!total_lines_x100) { - dev_err(dev, "%s: vtotal(%d) or vrefresh(%d) is 0\n", + DRM_DEV_ERROR(dev, "%s: vtotal(%d) or vrefresh(%d) is 0\n", __func__, mode->vtotal, mode->vrefresh); return -EINVAL; } vsync_clk_speed = clk_round_rate(mdp5_kms->vsync_clk, VSYNC_CLK_RATE); if (vsync_clk_speed <= 0) { - dev_err(dev, "vsync_clk round rate failed %ld\n", + DRM_DEV_ERROR(dev, "vsync_clk round rate failed %ld\n", vsync_clk_speed); return -EINVAL; } @@ -102,13 +102,13 @@ static int pingpong_tearcheck_enable(struct drm_encoder *encoder) ret = clk_set_rate(mdp5_kms->vsync_clk, clk_round_rate(mdp5_kms->vsync_clk, VSYNC_CLK_RATE)); if (ret) { - dev_err(encoder->dev->dev, + DRM_DEV_ERROR(encoder->dev->dev, "vsync_clk clk_set_rate failed, %d\n", ret); return ret; } ret = clk_prepare_enable(mdp5_kms->vsync_clk); if (ret) { - dev_err(encoder->dev->dev, + DRM_DEV_ERROR(encoder->dev->dev, "vsync_clk clk_prepare_enable failed, %d\n", ret); return ret; } diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c index 4878b81c96fb..2f95e6525589 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_crtc.c @@ -173,7 +173,7 @@ static void unref_cursor_worker(struct drm_flip_work *work, void *val) struct mdp5_kms *mdp5_kms = get_kms(&mdp5_crtc->base); struct msm_kms *kms = &mdp5_kms->base.base; - msm_gem_put_iova(val, kms->aspace); + msm_gem_unpin_iova(val, kms->aspace); drm_gem_object_put_unlocked(val); } @@ -655,7 +655,7 @@ static int mdp5_crtc_atomic_check(struct drm_crtc *crtc, ret = mdp5_crtc_setup_pipeline(crtc, state, need_right_mixer); if (ret) { - dev_err(dev->dev, "couldn't assign mixers %d\n", ret); + DRM_DEV_ERROR(dev->dev, "couldn't assign mixers %d\n", ret); return ret; } @@ -672,7 +672,7 @@ static int mdp5_crtc_atomic_check(struct drm_crtc *crtc, * and that we don't have conflicting mixer stages: */ if ((cnt + start - 1) >= hw_cfg->lm.nb_stages) { - dev_err(dev->dev, "too many planes! cnt=%d, start stage=%d\n", + DRM_DEV_ERROR(dev->dev, "too many planes! cnt=%d, start stage=%d\n", cnt, start); return -EINVAL; } @@ -872,7 +872,7 @@ static int mdp5_crtc_cursor_set(struct drm_crtc *crtc, } if ((width > CURSOR_WIDTH) || (height > CURSOR_HEIGHT)) { - dev_err(dev->dev, "bad cursor size: %dx%d\n", width, height); + DRM_DEV_ERROR(dev->dev, "bad cursor size: %dx%d\n", width, height); return -EINVAL; } @@ -896,7 +896,7 @@ static int mdp5_crtc_cursor_set(struct drm_crtc *crtc, if (!cursor_bo) return -ENOENT; - ret = msm_gem_get_iova(cursor_bo, kms->aspace, + ret = msm_gem_get_and_pin_iova(cursor_bo, kms->aspace, &mdp5_crtc->cursor.iova); if (ret) return -EINVAL; @@ -917,7 +917,7 @@ static int mdp5_crtc_cursor_set(struct drm_crtc *crtc, set_cursor: ret = mdp5_ctl_set_cursor(ctl, pipeline, 0, cursor_enable); if (ret) { - dev_err(dev->dev, "failed to %sable cursor: %d\n", + DRM_DEV_ERROR(dev->dev, "failed to %sable cursor: %d\n", cursor_enable ? "en" : "dis", ret); goto end; } diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_ctl.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_ctl.c index f93d5681267c..65a871f9f0d9 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_ctl.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_ctl.c @@ -262,13 +262,13 @@ int mdp5_ctl_set_cursor(struct mdp5_ctl *ctl, struct mdp5_pipeline *pipeline, struct mdp5_hw_mixer *mixer = pipeline->mixer; if (unlikely(WARN_ON(!mixer))) { - dev_err(ctl_mgr->dev->dev, "CTL %d cannot find LM", + DRM_DEV_ERROR(ctl_mgr->dev->dev, "CTL %d cannot find LM", ctl->id); return -EINVAL; } if (pipeline->r_mixer) { - dev_err(ctl_mgr->dev->dev, "unsupported configuration"); + DRM_DEV_ERROR(ctl_mgr->dev->dev, "unsupported configuration"); return -EINVAL; } @@ -604,10 +604,10 @@ int mdp5_ctl_pair(struct mdp5_ctl *ctlx, struct mdp5_ctl *ctly, bool enable) mdp5_write(mdp5_kms, REG_MDP5_SPARE_0, 0); return 0; } else if ((ctlx->pair != NULL) || (ctly->pair != NULL)) { - dev_err(ctl_mgr->dev->dev, "CTLs already paired\n"); + DRM_DEV_ERROR(ctl_mgr->dev->dev, "CTLs already paired\n"); return -EINVAL; } else if (!(ctlx->status & ctly->status & CTL_STAT_BOOKED)) { - dev_err(ctl_mgr->dev->dev, "Only pair booked CTLs\n"); + DRM_DEV_ERROR(ctl_mgr->dev->dev, "Only pair booked CTLs\n"); return -EINVAL; } @@ -652,7 +652,7 @@ struct mdp5_ctl *mdp5_ctlm_request(struct mdp5_ctl_manager *ctl_mgr, if ((ctl_mgr->ctls[c].status & checkm) == match) goto found; - dev_err(ctl_mgr->dev->dev, "No more CTL available!"); + DRM_DEV_ERROR(ctl_mgr->dev->dev, "No more CTL available!"); goto unlock; found: @@ -698,13 +698,13 @@ struct mdp5_ctl_manager *mdp5_ctlm_init(struct drm_device *dev, ctl_mgr = kzalloc(sizeof(*ctl_mgr), GFP_KERNEL); if (!ctl_mgr) { - dev_err(dev->dev, "failed to allocate CTL manager\n"); + DRM_DEV_ERROR(dev->dev, "failed to allocate CTL manager\n"); ret = -ENOMEM; goto fail; } if (unlikely(WARN_ON(ctl_cfg->count > MAX_CTL))) { - dev_err(dev->dev, "Increase static pool size to at least %d\n", + DRM_DEV_ERROR(dev->dev, "Increase static pool size to at least %d\n", ctl_cfg->count); ret = -ENOSPC; goto fail; @@ -723,7 +723,7 @@ struct mdp5_ctl_manager *mdp5_ctlm_init(struct drm_device *dev, struct mdp5_ctl *ctl = &ctl_mgr->ctls[c]; if (WARN_ON(!ctl_cfg->base[c])) { - dev_err(dev->dev, "CTL_%d: base is null!\n", c); + DRM_DEV_ERROR(dev->dev, "CTL_%d: base is null!\n", c); ret = -EINVAL; spin_unlock_irqrestore(&ctl_mgr->pool_lock, flags); goto fail; diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c index f71d8cf2261b..97179bec8902 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_kms.c @@ -264,7 +264,7 @@ static int mdp5_kms_debugfs_init(struct msm_kms *kms, struct drm_minor *minor) minor->debugfs_root, minor); if (ret) { - dev_err(dev->dev, "could not install mdp5_debugfs_list\n"); + DRM_DEV_ERROR(dev->dev, "could not install mdp5_debugfs_list\n"); return ret; } @@ -337,7 +337,7 @@ static struct drm_encoder *construct_encoder(struct mdp5_kms *mdp5_kms, encoder = mdp5_encoder_init(dev, intf, ctl); if (IS_ERR(encoder)) { - dev_err(dev->dev, "failed to construct encoder\n"); + DRM_DEV_ERROR(dev->dev, "failed to construct encoder\n"); return encoder; } @@ -418,7 +418,7 @@ static int modeset_init_intf(struct mdp5_kms *mdp5_kms, int dsi_id = get_dsi_id_from_intf(hw_cfg, intf->num); if ((dsi_id >= ARRAY_SIZE(priv->dsi)) || (dsi_id < 0)) { - dev_err(dev->dev, "failed to find dsi from intf %d\n", + DRM_DEV_ERROR(dev->dev, "failed to find dsi from intf %d\n", intf->num); ret = -EINVAL; break; @@ -443,7 +443,7 @@ static int modeset_init_intf(struct mdp5_kms *mdp5_kms, break; } default: - dev_err(dev->dev, "unknown intf: %d\n", intf->type); + DRM_DEV_ERROR(dev->dev, "unknown intf: %d\n", intf->type); ret = -EINVAL; break; } @@ -500,7 +500,7 @@ static int modeset_init(struct mdp5_kms *mdp5_kms) plane = mdp5_plane_init(dev, type); if (IS_ERR(plane)) { ret = PTR_ERR(plane); - dev_err(dev->dev, "failed to construct plane %d (%d)\n", i, ret); + DRM_DEV_ERROR(dev->dev, "failed to construct plane %d (%d)\n", i, ret); goto fail; } priv->planes[priv->num_planes++] = plane; @@ -517,7 +517,7 @@ static int modeset_init(struct mdp5_kms *mdp5_kms) crtc = mdp5_crtc_init(dev, primary[i], cursor[i], i); if (IS_ERR(crtc)) { ret = PTR_ERR(crtc); - dev_err(dev->dev, "failed to construct crtc %d (%d)\n", i, ret); + DRM_DEV_ERROR(dev->dev, "failed to construct crtc %d (%d)\n", i, ret); goto fail; } priv->crtcs[priv->num_crtcs++] = crtc; @@ -552,7 +552,7 @@ static void read_mdp_hw_revision(struct mdp5_kms *mdp5_kms, *major = FIELD(version, MDP5_HW_VERSION_MAJOR); *minor = FIELD(version, MDP5_HW_VERSION_MINOR); - dev_info(dev, "MDP5 version v%d.%d", *major, *minor); + DRM_DEV_INFO(dev, "MDP5 version v%d.%d", *major, *minor); } static int get_clk(struct platform_device *pdev, struct clk **clkp, @@ -561,7 +561,7 @@ static int get_clk(struct platform_device *pdev, struct clk **clkp, struct device *dev = &pdev->dev; struct clk *clk = msm_clk_get(pdev, name); if (IS_ERR(clk) && mandatory) { - dev_err(dev, "failed to get %s (%ld)\n", name, PTR_ERR(clk)); + DRM_DEV_ERROR(dev, "failed to get %s (%ld)\n", name, PTR_ERR(clk)); return PTR_ERR(clk); } if (IS_ERR(clk)) @@ -688,7 +688,7 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev) irq = irq_of_parse_and_map(pdev->dev.of_node, 0); if (irq < 0) { ret = irq; - dev_err(&pdev->dev, "failed to get irq: %d\n", ret); + DRM_DEV_ERROR(&pdev->dev, "failed to get irq: %d\n", ret); goto fail; } @@ -724,12 +724,12 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev) ret = aspace->mmu->funcs->attach(aspace->mmu, iommu_ports, ARRAY_SIZE(iommu_ports)); if (ret) { - dev_err(&pdev->dev, "failed to attach iommu: %d\n", + DRM_DEV_ERROR(&pdev->dev, "failed to attach iommu: %d\n", ret); goto fail; } } else { - dev_info(&pdev->dev, + DRM_DEV_INFO(&pdev->dev, "no iommu, fallback to phys contig buffers for scanout\n"); aspace = NULL; } @@ -738,7 +738,7 @@ struct msm_kms *mdp5_kms_init(struct drm_device *dev) ret = modeset_init(mdp5_kms); if (ret) { - dev_err(&pdev->dev, "modeset_init failed: %d\n", ret); + DRM_DEV_ERROR(&pdev->dev, "modeset_init failed: %d\n", ret); goto fail; } @@ -795,7 +795,7 @@ static int construct_pipes(struct mdp5_kms *mdp5_kms, int cnt, hwpipe = mdp5_pipe_init(pipes[i], offsets[i], caps); if (IS_ERR(hwpipe)) { ret = PTR_ERR(hwpipe); - dev_err(dev->dev, "failed to construct pipe for %s (%d)\n", + DRM_DEV_ERROR(dev->dev, "failed to construct pipe for %s (%d)\n", pipe2name(pipes[i]), ret); return ret; } @@ -867,7 +867,7 @@ static int hwmixer_init(struct mdp5_kms *mdp5_kms) mixer = mdp5_mixer_init(&hw_cfg->lm.instances[i]); if (IS_ERR(mixer)) { ret = PTR_ERR(mixer); - dev_err(dev->dev, "failed to construct LM%d (%d)\n", + DRM_DEV_ERROR(dev->dev, "failed to construct LM%d (%d)\n", i, ret); return ret; } @@ -897,7 +897,7 @@ static int interface_init(struct mdp5_kms *mdp5_kms) intf = kzalloc(sizeof(*intf), GFP_KERNEL); if (!intf) { - dev_err(dev->dev, "failed to construct INTF%d\n", i); + DRM_DEV_ERROR(dev->dev, "failed to construct INTF%d\n", i); return -ENOMEM; } diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_mdss.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_mdss.c index 1cc4e57f0226..889c2940692c 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_mdss.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_mdss.c @@ -132,7 +132,7 @@ static int mdss_irq_domain_init(struct mdp5_mdss *mdp5_mdss) d = irq_domain_add_linear(dev->of_node, 32, &mdss_hw_irqdomain_ops, mdp5_mdss); if (!d) { - dev_err(dev, "mdss irq domain add failed\n"); + DRM_DEV_ERROR(dev, "mdss irq domain add failed\n"); return -ENXIO; } @@ -246,7 +246,7 @@ int mdp5_mdss_init(struct drm_device *dev) ret = msm_mdss_get_clocks(mdp5_mdss); if (ret) { - dev_err(dev->dev, "failed to get clocks: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to get clocks: %d\n", ret); goto fail; } @@ -259,7 +259,7 @@ int mdp5_mdss_init(struct drm_device *dev) ret = regulator_enable(mdp5_mdss->vdd); if (ret) { - dev_err(dev->dev, "failed to enable regulator vdd: %d\n", + DRM_DEV_ERROR(dev->dev, "failed to enable regulator vdd: %d\n", ret); goto fail; } @@ -267,13 +267,13 @@ int mdp5_mdss_init(struct drm_device *dev) ret = devm_request_irq(dev->dev, platform_get_irq(pdev, 0), mdss_irq, 0, "mdss_isr", mdp5_mdss); if (ret) { - dev_err(dev->dev, "failed to init irq: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to init irq: %d\n", ret); goto fail_irq; } ret = mdss_irq_domain_init(mdp5_mdss); if (ret) { - dev_err(dev->dev, "failed to init sub-block irqs: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to init sub-block irqs: %d\n", ret); goto fail_irq; } diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c index 310459541e48..be13140967b4 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c @@ -125,7 +125,7 @@ static int mdp5_plane_atomic_set_property(struct drm_plane *plane, SET_PROPERTY(zpos, ZPOS, uint8_t); - dev_err(dev->dev, "Invalid property\n"); + DRM_DEV_ERROR(dev->dev, "Invalid property\n"); ret = -EINVAL; done: return ret; @@ -153,7 +153,7 @@ static int mdp5_plane_atomic_get_property(struct drm_plane *plane, GET_PROPERTY(zpos, ZPOS, uint8_t); - dev_err(dev->dev, "Invalid property\n"); + DRM_DEV_ERROR(dev->dev, "Invalid property\n"); ret = -EINVAL; done: return ret; @@ -658,7 +658,7 @@ static int calc_scalex_steps(struct drm_plane *plane, ret = calc_phase_step(src, dest, &phasex_step); if (ret) { - dev_err(dev, "X scaling (%d->%d) failed: %d\n", src, dest, ret); + DRM_DEV_ERROR(dev, "X scaling (%d->%d) failed: %d\n", src, dest, ret); return ret; } @@ -683,7 +683,7 @@ static int calc_scaley_steps(struct drm_plane *plane, ret = calc_phase_step(src, dest, &phasey_step); if (ret) { - dev_err(dev, "Y scaling (%d->%d) failed: %d\n", src, dest, ret); + DRM_DEV_ERROR(dev, "Y scaling (%d->%d) failed: %d\n", src, dest, ret); return ret; } diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.c index 96c2b828dba4..7cebcb2b3a37 100644 --- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.c +++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_smp.c @@ -88,7 +88,7 @@ static int smp_request_block(struct mdp5_smp *smp, avail = cnt - bitmap_weight(state->state, cnt); if (nblks > avail) { - dev_err(smp->dev->dev, "out of blks (req=%d > avail=%d)\n", + DRM_DEV_ERROR(smp->dev->dev, "out of blks (req=%d > avail=%d)\n", nblks, avail); return -ENOSPC; } @@ -188,7 +188,7 @@ int mdp5_smp_assign(struct mdp5_smp *smp, struct mdp5_smp_state *state, DBG("%s[%d]: request %d SMP blocks", pipe2name(pipe), i, n); ret = smp_request_block(smp, state, cid, n); if (ret) { - dev_err(dev->dev, "Cannot allocate %d SMP blocks: %d\n", + DRM_DEV_ERROR(dev->dev, "Cannot allocate %d SMP blocks: %d\n", n, ret); return ret; } diff --git a/drivers/gpu/drm/msm/dsi/dsi.c b/drivers/gpu/drm/msm/dsi/dsi.c index a9768f823290..7b2a1e6a8810 100644 --- a/drivers/gpu/drm/msm/dsi/dsi.c +++ b/drivers/gpu/drm/msm/dsi/dsi.c @@ -29,7 +29,7 @@ static int dsi_get_phy(struct msm_dsi *msm_dsi) phy_node = of_parse_phandle(pdev->dev.of_node, "phys", 0); if (!phy_node) { - dev_err(&pdev->dev, "cannot find phy device\n"); + DRM_DEV_ERROR(&pdev->dev, "cannot find phy device\n"); return -ENXIO; } @@ -40,7 +40,7 @@ static int dsi_get_phy(struct msm_dsi *msm_dsi) of_node_put(phy_node); if (!phy_pdev || !msm_dsi->phy) { - dev_err(&pdev->dev, "%s: phy driver is not ready\n", __func__); + DRM_DEV_ERROR(&pdev->dev, "%s: phy driver is not ready\n", __func__); return -EPROBE_DEFER; } @@ -210,7 +210,7 @@ int msm_dsi_modeset_init(struct msm_dsi *msm_dsi, struct drm_device *dev, ret = msm_dsi_host_modeset_init(msm_dsi->host, dev); if (ret) { - dev_err(dev->dev, "failed to modeset init host: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to modeset init host: %d\n", ret); goto fail; } @@ -222,7 +222,7 @@ int msm_dsi_modeset_init(struct msm_dsi *msm_dsi, struct drm_device *dev, msm_dsi->bridge = msm_dsi_manager_bridge_init(msm_dsi->id); if (IS_ERR(msm_dsi->bridge)) { ret = PTR_ERR(msm_dsi->bridge); - dev_err(dev->dev, "failed to create dsi bridge: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to create dsi bridge: %d\n", ret); msm_dsi->bridge = NULL; goto fail; } @@ -244,7 +244,7 @@ int msm_dsi_modeset_init(struct msm_dsi *msm_dsi, struct drm_device *dev, if (IS_ERR(msm_dsi->connector)) { ret = PTR_ERR(msm_dsi->connector); - dev_err(dev->dev, + DRM_DEV_ERROR(dev->dev, "failed to create dsi connector: %d\n", ret); msm_dsi->connector = NULL; goto fail; diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c index 9c6c523eacdc..38e481d2d606 100644 --- a/drivers/gpu/drm/msm/dsi/dsi_host.c +++ b/drivers/gpu/drm/msm/dsi/dsi_host.c @@ -1050,7 +1050,7 @@ static void dsi_wait4video_done(struct msm_dsi_host *msm_host) msecs_to_jiffies(70)); if (ret <= 0) - dev_err(dev, "wait for video done timed out\n"); + DRM_DEV_ERROR(dev, "wait for video done timed out\n"); dsi_intr_ctrl(msm_host, DSI_IRQ_MASK_VIDEO_DONE, 0); } @@ -1083,6 +1083,8 @@ int dsi_tx_buf_alloc_6g(struct msm_dsi_host *msm_host, int size) return PTR_ERR(data); } + msm_gem_object_set_name(msm_host->tx_gem_obj, "tx_gem"); + msm_host->tx_size = msm_host->tx_gem_obj->size; return 0; @@ -1118,7 +1120,7 @@ static void dsi_tx_buf_free(struct msm_dsi_host *msm_host) priv = dev->dev_private; if (msm_host->tx_gem_obj) { - msm_gem_put_iova(msm_host->tx_gem_obj, priv->kms->aspace); + msm_gem_unpin_iova(msm_host->tx_gem_obj, priv->kms->aspace); drm_gem_object_put_unlocked(msm_host->tx_gem_obj); msm_host->tx_gem_obj = NULL; } @@ -1248,7 +1250,7 @@ int dsi_dma_base_get_6g(struct msm_dsi_host *msm_host, uint64_t *dma_base) if (!dma_base) return -EINVAL; - return msm_gem_get_iova(msm_host->tx_gem_obj, + return msm_gem_get_and_pin_iova(msm_host->tx_gem_obj, priv->kms->aspace, dma_base); } @@ -1673,7 +1675,7 @@ static int dsi_host_parse_lane_data(struct msm_dsi_host *msm_host, prop = of_find_property(ep, "data-lanes", &len); if (!prop) { - dev_dbg(dev, + DRM_DEV_DEBUG(dev, "failed to find data lane mapping, using default\n"); return 0; } @@ -1681,7 +1683,7 @@ static int dsi_host_parse_lane_data(struct msm_dsi_host *msm_host, num_lanes = len / sizeof(u32); if (num_lanes < 1 || num_lanes > 4) { - dev_err(dev, "bad number of data lanes\n"); + DRM_DEV_ERROR(dev, "bad number of data lanes\n"); return -EINVAL; } @@ -1690,7 +1692,7 @@ static int dsi_host_parse_lane_data(struct msm_dsi_host *msm_host, ret = of_property_read_u32_array(ep, "data-lanes", lane_map, num_lanes); if (ret) { - dev_err(dev, "failed to read lane data\n"); + DRM_DEV_ERROR(dev, "failed to read lane data\n"); return ret; } @@ -1711,7 +1713,7 @@ static int dsi_host_parse_lane_data(struct msm_dsi_host *msm_host, */ for (j = 0; j < num_lanes; j++) { if (lane_map[j] < 0 || lane_map[j] > 3) - dev_err(dev, "bad physical lane entry %u\n", + DRM_DEV_ERROR(dev, "bad physical lane entry %u\n", lane_map[j]); if (swap[lane_map[j]] != j) @@ -1742,13 +1744,13 @@ static int dsi_host_parse_dt(struct msm_dsi_host *msm_host) */ endpoint = of_graph_get_endpoint_by_regs(np, 1, -1); if (!endpoint) { - dev_dbg(dev, "%s: no endpoint\n", __func__); + DRM_DEV_DEBUG(dev, "%s: no endpoint\n", __func__); return 0; } ret = dsi_host_parse_lane_data(msm_host, endpoint); if (ret) { - dev_err(dev, "%s: invalid lane configuration %d\n", + DRM_DEV_ERROR(dev, "%s: invalid lane configuration %d\n", __func__, ret); ret = -EINVAL; goto err; @@ -1757,7 +1759,7 @@ static int dsi_host_parse_dt(struct msm_dsi_host *msm_host) /* Get panel node from the output port's endpoint data */ device_node = of_graph_get_remote_node(np, 1, 0); if (!device_node) { - dev_dbg(dev, "%s: no valid device\n", __func__); + DRM_DEV_DEBUG(dev, "%s: no valid device\n", __func__); ret = -ENODEV; goto err; } @@ -1768,7 +1770,7 @@ static int dsi_host_parse_dt(struct msm_dsi_host *msm_host) msm_host->sfpb = syscon_regmap_lookup_by_phandle(np, "syscon-sfpb"); if (IS_ERR(msm_host->sfpb)) { - dev_err(dev, "%s: failed to get sfpb regmap\n", + DRM_DEV_ERROR(dev, "%s: failed to get sfpb regmap\n", __func__); ret = PTR_ERR(msm_host->sfpb); } @@ -1918,7 +1920,7 @@ int msm_dsi_host_modeset_init(struct mipi_dsi_host *host, msm_host->irq = irq_of_parse_and_map(pdev->dev.of_node, 0); if (msm_host->irq < 0) { ret = msm_host->irq; - dev_err(dev->dev, "failed to get irq: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to get irq: %d\n", ret); return ret; } @@ -1926,7 +1928,7 @@ int msm_dsi_host_modeset_init(struct mipi_dsi_host *host, dsi_host_irq, IRQF_TRIGGER_HIGH | IRQF_ONESHOT, "dsi_isr", msm_host); if (ret < 0) { - dev_err(&pdev->dev, "failed to request IRQ%u: %d\n", + DRM_DEV_ERROR(&pdev->dev, "failed to request IRQ%u: %d\n", msm_host->irq, ret); return ret; } diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c index 9a9fa0c75a13..1760483b247e 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy.c @@ -404,7 +404,7 @@ static int dsi_phy_regulator_init(struct msm_dsi_phy *phy) ret = devm_regulator_bulk_get(dev, num, s); if (ret < 0) { - dev_err(dev, "%s: failed to init regulator, ret=%d\n", + DRM_DEV_ERROR(dev, "%s: failed to init regulator, ret=%d\n", __func__, ret); return ret; } @@ -441,7 +441,7 @@ static int dsi_phy_regulator_enable(struct msm_dsi_phy *phy) ret = regulator_set_load(s[i].consumer, regs[i].enable_load); if (ret < 0) { - dev_err(dev, + DRM_DEV_ERROR(dev, "regulator %d set op mode failed, %d\n", i, ret); goto fail; @@ -451,7 +451,7 @@ static int dsi_phy_regulator_enable(struct msm_dsi_phy *phy) ret = regulator_bulk_enable(num, s); if (ret < 0) { - dev_err(dev, "regulator enable failed, %d\n", ret); + DRM_DEV_ERROR(dev, "regulator enable failed, %d\n", ret); goto fail; } @@ -472,7 +472,7 @@ static int dsi_phy_enable_resource(struct msm_dsi_phy *phy) ret = clk_prepare_enable(phy->ahb_clk); if (ret) { - dev_err(dev, "%s: can't enable ahb clk, %d\n", __func__, ret); + DRM_DEV_ERROR(dev, "%s: can't enable ahb clk, %d\n", __func__, ret); pm_runtime_put_sync(dev); } @@ -543,7 +543,7 @@ int msm_dsi_phy_init_common(struct msm_dsi_phy *phy) phy->reg_base = msm_ioremap(pdev, "dsi_phy_regulator", "DSI_PHY_REG"); if (IS_ERR(phy->reg_base)) { - dev_err(&pdev->dev, "%s: failed to map phy regulator base\n", + DRM_DEV_ERROR(&pdev->dev, "%s: failed to map phy regulator base\n", __func__); ret = -ENOMEM; goto fail; @@ -574,7 +574,7 @@ static int dsi_phy_driver_probe(struct platform_device *pdev) phy->id = dsi_phy_get_id(phy); if (phy->id < 0) { ret = phy->id; - dev_err(dev, "%s: couldn't identify PHY index, %d\n", + DRM_DEV_ERROR(dev, "%s: couldn't identify PHY index, %d\n", __func__, ret); goto fail; } @@ -584,20 +584,20 @@ static int dsi_phy_driver_probe(struct platform_device *pdev) phy->base = msm_ioremap(pdev, "dsi_phy", "DSI_PHY"); if (IS_ERR(phy->base)) { - dev_err(dev, "%s: failed to map phy base\n", __func__); + DRM_DEV_ERROR(dev, "%s: failed to map phy base\n", __func__); ret = -ENOMEM; goto fail; } ret = dsi_phy_regulator_init(phy); if (ret) { - dev_err(dev, "%s: failed to init regulator\n", __func__); + DRM_DEV_ERROR(dev, "%s: failed to init regulator\n", __func__); goto fail; } phy->ahb_clk = msm_clk_get(pdev, "iface"); if (IS_ERR(phy->ahb_clk)) { - dev_err(dev, "%s: Unable to get ahb clk\n", __func__); + DRM_DEV_ERROR(dev, "%s: Unable to get ahb clk\n", __func__); ret = PTR_ERR(phy->ahb_clk); goto fail; } @@ -617,7 +617,7 @@ static int dsi_phy_driver_probe(struct platform_device *pdev) phy->pll = msm_dsi_pll_init(pdev, phy->cfg->type, phy->id); if (IS_ERR_OR_NULL(phy->pll)) - dev_info(dev, + DRM_DEV_INFO(dev, "%s: pll init failed: %ld, need separate pll clk driver\n", __func__, PTR_ERR(phy->pll)); @@ -675,21 +675,21 @@ int msm_dsi_phy_enable(struct msm_dsi_phy *phy, int src_pll_id, ret = dsi_phy_enable_resource(phy); if (ret) { - dev_err(dev, "%s: resource enable failed, %d\n", + DRM_DEV_ERROR(dev, "%s: resource enable failed, %d\n", __func__, ret); goto res_en_fail; } ret = dsi_phy_regulator_enable(phy); if (ret) { - dev_err(dev, "%s: regulator enable failed, %d\n", + DRM_DEV_ERROR(dev, "%s: regulator enable failed, %d\n", __func__, ret); goto reg_en_fail; } ret = phy->cfg->ops.enable(phy, src_pll_id, clk_req); if (ret) { - dev_err(dev, "%s: phy enable failed, %d\n", __func__, ret); + DRM_DEV_ERROR(dev, "%s: phy enable failed, %d\n", __func__, ret); goto phy_en_fail; } @@ -702,7 +702,7 @@ int msm_dsi_phy_enable(struct msm_dsi_phy *phy, int src_pll_id, if (phy->usecase != MSM_DSI_PHY_SLAVE) { ret = msm_dsi_pll_restore_state(phy->pll); if (ret) { - dev_err(dev, "%s: failed to restore pll state, %d\n", + DRM_DEV_ERROR(dev, "%s: failed to restore pll state, %d\n", __func__, ret); goto pll_restor_fail; } diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c index b3fffc8dbb2a..44959e79ce28 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c @@ -93,7 +93,7 @@ static int dsi_10nm_phy_enable(struct msm_dsi_phy *phy, int src_pll_id, DBG(""); if (msm_dsi_dphy_timing_calc_v3(timing, clk_req)) { - dev_err(&phy->pdev->dev, + DRM_DEV_ERROR(&phy->pdev->dev, "%s: D-PHY timing calculation failed\n", __func__); return -EINVAL; } @@ -172,7 +172,7 @@ static int dsi_10nm_phy_enable(struct msm_dsi_phy *phy, int src_pll_id, ret = msm_dsi_pll_set_usecase(phy->pll, phy->usecase); if (ret) { - dev_err(&phy->pdev->dev, "%s: set pll usecase failed, %d\n", + DRM_DEV_ERROR(&phy->pdev->dev, "%s: set pll usecase failed, %d\n", __func__, ret); return ret; } @@ -196,7 +196,7 @@ static int dsi_10nm_phy_init(struct msm_dsi_phy *phy) phy->lane_base = msm_ioremap(pdev, "dsi_phy_lane", "DSI_PHY_LANE"); if (IS_ERR(phy->lane_base)) { - dev_err(&pdev->dev, "%s: failed to map phy lane base\n", + DRM_DEV_ERROR(&pdev->dev, "%s: failed to map phy lane base\n", __func__); return -ENOMEM; } diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c index 513f4234adc1..a172c667e8bc 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_14nm.c @@ -64,7 +64,7 @@ static int dsi_14nm_phy_enable(struct msm_dsi_phy *phy, int src_pll_id, void __iomem *lane_base = phy->lane_base; if (msm_dsi_dphy_timing_calc_v2(timing, clk_req)) { - dev_err(&phy->pdev->dev, + DRM_DEV_ERROR(&phy->pdev->dev, "%s: D-PHY timing calculation failed\n", __func__); return -EINVAL; } @@ -115,7 +115,7 @@ static int dsi_14nm_phy_enable(struct msm_dsi_phy *phy, int src_pll_id, ret = msm_dsi_pll_set_usecase(phy->pll, phy->usecase); if (ret) { - dev_err(&phy->pdev->dev, "%s: set pll usecase failed, %d\n", + DRM_DEV_ERROR(&phy->pdev->dev, "%s: set pll usecase failed, %d\n", __func__, ret); return ret; } @@ -142,7 +142,7 @@ static int dsi_14nm_phy_init(struct msm_dsi_phy *phy) phy->lane_base = msm_ioremap(pdev, "dsi_phy_lane", "DSI_PHY_LANE"); if (IS_ERR(phy->lane_base)) { - dev_err(&pdev->dev, "%s: failed to map phy lane base\n", + DRM_DEV_ERROR(&pdev->dev, "%s: failed to map phy lane base\n", __func__); return -ENOMEM; } diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_20nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_20nm.c index 1ca6c69516f5..9ea9478d3707 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_20nm.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_20nm.c @@ -82,7 +82,7 @@ static int dsi_20nm_phy_enable(struct msm_dsi_phy *phy, int src_pll_id, DBG(""); if (msm_dsi_dphy_timing_calc(timing, clk_req)) { - dev_err(&phy->pdev->dev, + DRM_DEV_ERROR(&phy->pdev->dev, "%s: D-PHY timing calculation failed\n", __func__); return -EINVAL; } diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c index 4972b52cbe44..c79505d97fe8 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm.c @@ -76,7 +76,7 @@ static int dsi_28nm_phy_enable(struct msm_dsi_phy *phy, int src_pll_id, DBG(""); if (msm_dsi_dphy_timing_calc(timing, clk_req)) { - dev_err(&phy->pdev->dev, + DRM_DEV_ERROR(&phy->pdev->dev, "%s: D-PHY timing calculation failed\n", __func__); return -EINVAL; } diff --git a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c index 398004463498..98790b44da48 100644 --- a/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c +++ b/drivers/gpu/drm/msm/dsi/phy/dsi_phy_28nm_8960.c @@ -132,7 +132,7 @@ static int dsi_28nm_phy_enable(struct msm_dsi_phy *phy, int src_pll_id, DBG(""); if (msm_dsi_dphy_timing_calc(timing, clk_req)) { - dev_err(&phy->pdev->dev, + DRM_DEV_ERROR(&phy->pdev->dev, "%s: D-PHY timing calculation failed\n", __func__); return -EINVAL; } diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll.c b/drivers/gpu/drm/msm/dsi/pll/dsi_pll.c index 613e206fa4fc..7a1fb4da2ad3 100644 --- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll.c +++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll.c @@ -175,7 +175,7 @@ struct msm_dsi_pll *msm_dsi_pll_init(struct platform_device *pdev, } if (IS_ERR(pll)) { - dev_err(dev, "%s: failed to init DSI PLL\n", __func__); + DRM_DEV_ERROR(dev, "%s: failed to init DSI PLL\n", __func__); return pll; } diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c index 4c03f0b7343e..aabab6311043 100644 --- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c +++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_10nm.c @@ -17,7 +17,7 @@ * | | * | | * +---------+ | +----------+ | +----+ - * dsi0vco_clk ---| out_div |--o--| divl_3_0 |--o--| /8 |-- dsi0pllbyte + * dsi0vco_clk ---| out_div |--o--| divl_3_0 |--o--| /8 |-- dsi0_phy_pll_out_byteclk * +---------+ | +----------+ | +----+ * | | * | | dsi0_pll_by_2_bit_clk @@ -25,7 +25,7 @@ * | | +----+ | |\ dsi0_pclk_mux * | |--| /2 |--o--| \ | * | | +----+ | \ | +---------+ - * | --------------| |--o--| div_7_4 |-- dsi0pll + * | --------------| |--o--| div_7_4 |-- dsi0_phy_pll_out_dsiclk * |------------------------------| / +---------+ * | +-----+ | / * -----------| /4? |--o----------|/ @@ -39,6 +39,8 @@ #define DSI_PIXEL_PLL_CLK 1 #define NUM_PROVIDED_CLKS 2 +#define VCO_REF_CLK_RATE 19200000 + struct dsi_pll_regs { u32 pll_prop_gain_rate; u32 pll_lockdet_rate; @@ -316,7 +318,7 @@ static int dsi_pll_10nm_vco_set_rate(struct clk_hw *hw, unsigned long rate, parent_rate); pll_10nm->vco_current_rate = rate; - pll_10nm->vco_ref_clk_rate = parent_rate; + pll_10nm->vco_ref_clk_rate = VCO_REF_CLK_RATE; dsi_pll_setup_config(pll_10nm); @@ -688,7 +690,7 @@ static int pll_10nm_register(struct dsi_pll_10nm *pll_10nm) hws[num++] = hw; - snprintf(clk_name, 32, "dsi%dpllbyte", pll_10nm->id); + snprintf(clk_name, 32, "dsi%d_phy_pll_out_byteclk", pll_10nm->id); snprintf(parent, 32, "dsi%d_pll_bit_clk", pll_10nm->id); /* DSI Byte clock = VCO_CLK / OUT_DIV / BIT_DIV / 8 */ @@ -737,7 +739,7 @@ static int pll_10nm_register(struct dsi_pll_10nm *pll_10nm) hws[num++] = hw; - snprintf(clk_name, 32, "dsi%dpll", pll_10nm->id); + snprintf(clk_name, 32, "dsi%d_phy_pll_out_dsiclk", pll_10nm->id); snprintf(parent, 32, "dsi%d_pclk_mux", pll_10nm->id); /* PIX CLK DIV : DIV_CTRL_7_4*/ @@ -760,7 +762,7 @@ static int pll_10nm_register(struct dsi_pll_10nm *pll_10nm) ret = of_clk_add_hw_provider(dev->of_node, of_clk_hw_onecell_get, pll_10nm->hw_data); if (ret) { - dev_err(dev, "failed to register clk provider: %d\n", ret); + DRM_DEV_ERROR(dev, "failed to register clk provider: %d\n", ret); return ret; } @@ -788,13 +790,13 @@ struct msm_dsi_pll *msm_dsi_pll_10nm_init(struct platform_device *pdev, int id) pll_10nm->phy_cmn_mmio = msm_ioremap(pdev, "dsi_phy", "DSI_PHY"); if (IS_ERR_OR_NULL(pll_10nm->phy_cmn_mmio)) { - dev_err(&pdev->dev, "failed to map CMN PHY base\n"); + DRM_DEV_ERROR(&pdev->dev, "failed to map CMN PHY base\n"); return ERR_PTR(-ENOMEM); } pll_10nm->mmio = msm_ioremap(pdev, "dsi_pll", "DSI_PLL"); if (IS_ERR_OR_NULL(pll_10nm->mmio)) { - dev_err(&pdev->dev, "failed to map PLL base\n"); + DRM_DEV_ERROR(&pdev->dev, "failed to map PLL base\n"); return ERR_PTR(-ENOMEM); } @@ -813,7 +815,7 @@ struct msm_dsi_pll *msm_dsi_pll_10nm_init(struct platform_device *pdev, int id) ret = pll_10nm_register(pll_10nm); if (ret) { - dev_err(&pdev->dev, "failed to register PLL: %d\n", ret); + DRM_DEV_ERROR(&pdev->dev, "failed to register PLL: %d\n", ret); return ERR_PTR(ret); } diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_14nm.c b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_14nm.c index 71fe60e5f01f..0e18cddd6f22 100644 --- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_14nm.c +++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_14nm.c @@ -783,7 +783,7 @@ static int dsi_pll_14nm_enable_seq(struct msm_dsi_pll *pll) POLL_TIMEOUT_US); if (unlikely(!locked)) - dev_err(&pll_14nm->pdev->dev, "DSI PLL lock failed\n"); + DRM_DEV_ERROR(&pll_14nm->pdev->dev, "DSI PLL lock failed\n"); else DBG("DSI PLL lock success"); @@ -829,7 +829,7 @@ static int dsi_pll_14nm_restore_state(struct msm_dsi_pll *pll) ret = dsi_pll_14nm_vco_set_rate(&pll->clk_hw, cached_state->vco_rate, 0); if (ret) { - dev_err(&pll_14nm->pdev->dev, + DRM_DEV_ERROR(&pll_14nm->pdev->dev, "restore vco rate failed. ret=%d\n", ret); return ret; } @@ -1039,7 +1039,7 @@ static int pll_14nm_register(struct dsi_pll_14nm *pll_14nm) ret = of_clk_add_hw_provider(dev->of_node, of_clk_hw_onecell_get, pll_14nm->hw_data); if (ret) { - dev_err(dev, "failed to register clk provider: %d\n", ret); + DRM_DEV_ERROR(dev, "failed to register clk provider: %d\n", ret); return ret; } @@ -1067,13 +1067,13 @@ struct msm_dsi_pll *msm_dsi_pll_14nm_init(struct platform_device *pdev, int id) pll_14nm->phy_cmn_mmio = msm_ioremap(pdev, "dsi_phy", "DSI_PHY"); if (IS_ERR_OR_NULL(pll_14nm->phy_cmn_mmio)) { - dev_err(&pdev->dev, "failed to map CMN PHY base\n"); + DRM_DEV_ERROR(&pdev->dev, "failed to map CMN PHY base\n"); return ERR_PTR(-ENOMEM); } pll_14nm->mmio = msm_ioremap(pdev, "dsi_pll", "DSI_PLL"); if (IS_ERR_OR_NULL(pll_14nm->mmio)) { - dev_err(&pdev->dev, "failed to map PLL base\n"); + DRM_DEV_ERROR(&pdev->dev, "failed to map PLL base\n"); return ERR_PTR(-ENOMEM); } @@ -1096,7 +1096,7 @@ struct msm_dsi_pll *msm_dsi_pll_14nm_init(struct platform_device *pdev, int id) ret = pll_14nm_register(pll_14nm); if (ret) { - dev_err(&pdev->dev, "failed to register PLL: %d\n", ret); + DRM_DEV_ERROR(&pdev->dev, "failed to register PLL: %d\n", ret); return ERR_PTR(ret); } diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_28nm.c b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_28nm.c index 26e3a01a99c2..dcbbaeb1b1fb 100644 --- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_28nm.c +++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_28nm.c @@ -156,7 +156,7 @@ static int dsi_pll_28nm_clk_set_rate(struct clk_hw *hw, unsigned long rate, if (rate <= lpfr_lut[i].vco_rate) break; if (i == LPFR_LUT_SIZE) { - dev_err(dev, "unable to get loop filter resistance. vco=%lu\n", + DRM_DEV_ERROR(dev, "unable to get loop filter resistance. vco=%lu\n", rate); return -EINVAL; } @@ -386,7 +386,7 @@ static int dsi_pll_28nm_enable_seq_hpm(struct msm_dsi_pll *pll) } if (unlikely(!locked)) - dev_err(dev, "DSI PLL lock failed\n"); + DRM_DEV_ERROR(dev, "DSI PLL lock failed\n"); else DBG("DSI PLL Lock success"); @@ -429,7 +429,7 @@ static int dsi_pll_28nm_enable_seq_lp(struct msm_dsi_pll *pll) locked = pll_28nm_poll_for_ready(pll_28nm, max_reads, timeout_us); if (unlikely(!locked)) - dev_err(dev, "DSI PLL lock failed\n"); + DRM_DEV_ERROR(dev, "DSI PLL lock failed\n"); else DBG("DSI PLL lock success"); @@ -468,7 +468,7 @@ static int dsi_pll_28nm_restore_state(struct msm_dsi_pll *pll) ret = dsi_pll_28nm_clk_set_rate(&pll->clk_hw, cached_state->vco_rate, 0); if (ret) { - dev_err(&pll_28nm->pdev->dev, + DRM_DEV_ERROR(&pll_28nm->pdev->dev, "restore vco rate failed. ret=%d\n", ret); return ret; } @@ -581,7 +581,7 @@ static int pll_28nm_register(struct dsi_pll_28nm *pll_28nm) ret = of_clk_add_provider(dev->of_node, of_clk_src_onecell_get, &pll_28nm->clk_data); if (ret) { - dev_err(dev, "failed to register clk provider: %d\n", ret); + DRM_DEV_ERROR(dev, "failed to register clk provider: %d\n", ret); return ret; } @@ -607,7 +607,7 @@ struct msm_dsi_pll *msm_dsi_pll_28nm_init(struct platform_device *pdev, pll_28nm->mmio = msm_ioremap(pdev, "dsi_pll", "DSI_PLL"); if (IS_ERR_OR_NULL(pll_28nm->mmio)) { - dev_err(&pdev->dev, "%s: failed to map pll base\n", __func__); + DRM_DEV_ERROR(&pdev->dev, "%s: failed to map pll base\n", __func__); return ERR_PTR(-ENOMEM); } @@ -633,13 +633,13 @@ struct msm_dsi_pll *msm_dsi_pll_28nm_init(struct platform_device *pdev, pll->en_seq_cnt = 1; pll->enable_seqs[0] = dsi_pll_28nm_enable_seq_lp; } else { - dev_err(&pdev->dev, "phy type (%d) is not 28nm\n", type); + DRM_DEV_ERROR(&pdev->dev, "phy type (%d) is not 28nm\n", type); return ERR_PTR(-EINVAL); } ret = pll_28nm_register(pll_28nm); if (ret) { - dev_err(&pdev->dev, "failed to register PLL: %d\n", ret); + DRM_DEV_ERROR(&pdev->dev, "failed to register PLL: %d\n", ret); return ERR_PTR(ret); } diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_28nm_8960.c b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_28nm_8960.c index 49008451085b..d6897464755f 100644 --- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll_28nm_8960.c +++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll_28nm_8960.c @@ -327,7 +327,7 @@ static int dsi_pll_28nm_enable_seq(struct msm_dsi_pll *pll) locked = pll_28nm_poll_for_ready(pll_28nm, max_reads, timeout_us); if (unlikely(!locked)) - dev_err(dev, "DSI PLL lock failed\n"); + DRM_DEV_ERROR(dev, "DSI PLL lock failed\n"); else DBG("DSI PLL lock success"); @@ -368,7 +368,7 @@ static int dsi_pll_28nm_restore_state(struct msm_dsi_pll *pll) ret = dsi_pll_28nm_clk_set_rate(&pll->clk_hw, cached_state->vco_rate, 0); if (ret) { - dev_err(&pll_28nm->pdev->dev, + DRM_DEV_ERROR(&pll_28nm->pdev->dev, "restore vco rate failed. ret=%d\n", ret); return ret; } @@ -482,7 +482,7 @@ static int pll_28nm_register(struct dsi_pll_28nm *pll_28nm) ret = of_clk_add_provider(dev->of_node, of_clk_src_onecell_get, &pll_28nm->clk_data); if (ret) { - dev_err(dev, "failed to register clk provider: %d\n", ret); + DRM_DEV_ERROR(dev, "failed to register clk provider: %d\n", ret); return ret; } @@ -508,7 +508,7 @@ struct msm_dsi_pll *msm_dsi_pll_28nm_8960_init(struct platform_device *pdev, pll_28nm->mmio = msm_ioremap(pdev, "dsi_pll", "DSI_PLL"); if (IS_ERR_OR_NULL(pll_28nm->mmio)) { - dev_err(&pdev->dev, "%s: failed to map pll base\n", __func__); + DRM_DEV_ERROR(&pdev->dev, "%s: failed to map pll base\n", __func__); return ERR_PTR(-ENOMEM); } @@ -526,7 +526,7 @@ struct msm_dsi_pll *msm_dsi_pll_28nm_8960_init(struct platform_device *pdev, ret = pll_28nm_register(pll_28nm); if (ret) { - dev_err(&pdev->dev, "failed to register PLL: %d\n", ret); + DRM_DEV_ERROR(&pdev->dev, "failed to register PLL: %d\n", ret); return ERR_PTR(ret); } diff --git a/drivers/gpu/drm/msm/edp/edp.c b/drivers/gpu/drm/msm/edp/edp.c index 0940e84b2821..6a63aba98a30 100644 --- a/drivers/gpu/drm/msm/edp/edp.c +++ b/drivers/gpu/drm/msm/edp/edp.c @@ -157,7 +157,7 @@ int msm_edp_modeset_init(struct msm_edp *edp, struct drm_device *dev, edp->bridge = msm_edp_bridge_init(edp); if (IS_ERR(edp->bridge)) { ret = PTR_ERR(edp->bridge); - dev_err(dev->dev, "failed to create eDP bridge: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to create eDP bridge: %d\n", ret); edp->bridge = NULL; goto fail; } @@ -165,7 +165,7 @@ int msm_edp_modeset_init(struct msm_edp *edp, struct drm_device *dev, edp->connector = msm_edp_connector_init(edp); if (IS_ERR(edp->connector)) { ret = PTR_ERR(edp->connector); - dev_err(dev->dev, "failed to create eDP connector: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to create eDP connector: %d\n", ret); edp->connector = NULL; goto fail; } @@ -173,7 +173,7 @@ int msm_edp_modeset_init(struct msm_edp *edp, struct drm_device *dev, edp->irq = irq_of_parse_and_map(pdev->dev.of_node, 0); if (edp->irq < 0) { ret = edp->irq; - dev_err(dev->dev, "failed to get IRQ: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to get IRQ: %d\n", ret); goto fail; } @@ -181,7 +181,7 @@ int msm_edp_modeset_init(struct msm_edp *edp, struct drm_device *dev, edp_irq, IRQF_TRIGGER_HIGH | IRQF_ONESHOT, "edp_isr", edp); if (ret < 0) { - dev_err(dev->dev, "failed to request IRQ%u: %d\n", + DRM_DEV_ERROR(dev->dev, "failed to request IRQ%u: %d\n", edp->irq, ret); goto fail; } diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c index 23670907a29d..e247d6942a49 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi.c @@ -98,7 +98,7 @@ static int msm_hdmi_get_phy(struct hdmi *hdmi) phy_node = of_parse_phandle(pdev->dev.of_node, "phys", 0); if (!phy_node) { - dev_err(&pdev->dev, "cannot find phy device\n"); + DRM_DEV_ERROR(&pdev->dev, "cannot find phy device\n"); return -ENXIO; } @@ -109,7 +109,7 @@ static int msm_hdmi_get_phy(struct hdmi *hdmi) of_node_put(phy_node); if (!phy_pdev || !hdmi->phy) { - dev_err(&pdev->dev, "phy driver is not ready\n"); + DRM_DEV_ERROR(&pdev->dev, "phy driver is not ready\n"); return -EPROBE_DEFER; } @@ -153,7 +153,7 @@ static struct hdmi *msm_hdmi_init(struct platform_device *pdev) hdmi->qfprom_mmio = msm_ioremap(pdev, config->qfprom_mmio_name, "HDMI_QFPROM"); if (IS_ERR(hdmi->qfprom_mmio)) { - dev_info(&pdev->dev, "can't find qfprom resource\n"); + DRM_DEV_INFO(&pdev->dev, "can't find qfprom resource\n"); hdmi->qfprom_mmio = NULL; } @@ -172,7 +172,7 @@ static struct hdmi *msm_hdmi_init(struct platform_device *pdev) config->hpd_reg_names[i]); if (IS_ERR(reg)) { ret = PTR_ERR(reg); - dev_err(&pdev->dev, "failed to get hpd regulator: %s (%d)\n", + DRM_DEV_ERROR(&pdev->dev, "failed to get hpd regulator: %s (%d)\n", config->hpd_reg_names[i], ret); goto fail; } @@ -195,7 +195,7 @@ static struct hdmi *msm_hdmi_init(struct platform_device *pdev) config->pwr_reg_names[i]); if (IS_ERR(reg)) { ret = PTR_ERR(reg); - dev_err(&pdev->dev, "failed to get pwr regulator: %s (%d)\n", + DRM_DEV_ERROR(&pdev->dev, "failed to get pwr regulator: %s (%d)\n", config->pwr_reg_names[i], ret); goto fail; } @@ -217,7 +217,7 @@ static struct hdmi *msm_hdmi_init(struct platform_device *pdev) clk = msm_clk_get(pdev, config->hpd_clk_names[i]); if (IS_ERR(clk)) { ret = PTR_ERR(clk); - dev_err(&pdev->dev, "failed to get hpd clk: %s (%d)\n", + DRM_DEV_ERROR(&pdev->dev, "failed to get hpd clk: %s (%d)\n", config->hpd_clk_names[i], ret); goto fail; } @@ -239,7 +239,7 @@ static struct hdmi *msm_hdmi_init(struct platform_device *pdev) clk = msm_clk_get(pdev, config->pwr_clk_names[i]); if (IS_ERR(clk)) { ret = PTR_ERR(clk); - dev_err(&pdev->dev, "failed to get pwr clk: %s (%d)\n", + DRM_DEV_ERROR(&pdev->dev, "failed to get pwr clk: %s (%d)\n", config->pwr_clk_names[i], ret); goto fail; } @@ -254,14 +254,14 @@ static struct hdmi *msm_hdmi_init(struct platform_device *pdev) hdmi->i2c = msm_hdmi_i2c_init(hdmi); if (IS_ERR(hdmi->i2c)) { ret = PTR_ERR(hdmi->i2c); - dev_err(&pdev->dev, "failed to get i2c: %d\n", ret); + DRM_DEV_ERROR(&pdev->dev, "failed to get i2c: %d\n", ret); hdmi->i2c = NULL; goto fail; } ret = msm_hdmi_get_phy(hdmi); if (ret) { - dev_err(&pdev->dev, "failed to get phy\n"); + DRM_DEV_ERROR(&pdev->dev, "failed to get phy\n"); goto fail; } @@ -303,7 +303,7 @@ int msm_hdmi_modeset_init(struct hdmi *hdmi, hdmi->bridge = msm_hdmi_bridge_init(hdmi); if (IS_ERR(hdmi->bridge)) { ret = PTR_ERR(hdmi->bridge); - dev_err(dev->dev, "failed to create HDMI bridge: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to create HDMI bridge: %d\n", ret); hdmi->bridge = NULL; goto fail; } @@ -311,7 +311,7 @@ int msm_hdmi_modeset_init(struct hdmi *hdmi, hdmi->connector = msm_hdmi_connector_init(hdmi); if (IS_ERR(hdmi->connector)) { ret = PTR_ERR(hdmi->connector); - dev_err(dev->dev, "failed to create HDMI connector: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to create HDMI connector: %d\n", ret); hdmi->connector = NULL; goto fail; } @@ -319,7 +319,7 @@ int msm_hdmi_modeset_init(struct hdmi *hdmi, hdmi->irq = irq_of_parse_and_map(pdev->dev.of_node, 0); if (hdmi->irq < 0) { ret = hdmi->irq; - dev_err(dev->dev, "failed to get irq: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to get irq: %d\n", ret); goto fail; } @@ -327,11 +327,17 @@ int msm_hdmi_modeset_init(struct hdmi *hdmi, msm_hdmi_irq, IRQF_TRIGGER_HIGH | IRQF_ONESHOT, "hdmi_isr", hdmi); if (ret < 0) { - dev_err(dev->dev, "failed to request IRQ%u: %d\n", + DRM_DEV_ERROR(dev->dev, "failed to request IRQ%u: %d\n", hdmi->irq, ret); goto fail; } + ret = msm_hdmi_hpd_enable(hdmi->connector); + if (ret < 0) { + DRM_DEV_ERROR(&hdmi->pdev->dev, "failed to enable HPD: %d\n", ret); + goto fail; + } + encoder->bridge = hdmi->bridge; priv->bridges[priv->num_bridges++] = hdmi->bridge; @@ -476,7 +482,7 @@ static int msm_hdmi_audio_hw_params(struct device *dev, void *data, unsigned int level_shift = 0; /* 0dB */ bool down_mix = false; - dev_dbg(dev, "%u Hz, %d bit, %d channels\n", params->sample_rate, + DRM_DEV_DEBUG(dev, "%u Hz, %d bit, %d channels\n", params->sample_rate, params->sample_width, params->cea.channels); switch (params->cea.channels) { @@ -527,7 +533,7 @@ static int msm_hdmi_audio_hw_params(struct device *dev, void *data, rate = HDMI_SAMPLE_RATE_192KHZ; break; default: - dev_err(dev, "rate[%d] not supported!\n", + DRM_DEV_ERROR(dev, "rate[%d] not supported!\n", params->sample_rate); return -EINVAL; } @@ -571,7 +577,7 @@ static int msm_hdmi_bind(struct device *dev, struct device *master, void *data) { struct drm_device *drm = dev_get_drvdata(master); struct msm_drm_private *priv = drm->dev_private; - static struct hdmi_platform_config *hdmi_cfg; + struct hdmi_platform_config *hdmi_cfg; struct hdmi *hdmi; struct device_node *of_node = dev->of_node; int i, err; @@ -579,7 +585,7 @@ static int msm_hdmi_bind(struct device *dev, struct device *master, void *data) hdmi_cfg = (struct hdmi_platform_config *) of_device_get_match_data(dev); if (!hdmi_cfg) { - dev_err(dev, "unknown hdmi_cfg: %pOFn\n", of_node); + DRM_DEV_ERROR(dev, "unknown hdmi_cfg: %pOFn\n", of_node); return -ENXIO; } diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.h b/drivers/gpu/drm/msm/hdmi/hdmi.h index accc9a61611d..5c5df6ab2a57 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi.h +++ b/drivers/gpu/drm/msm/hdmi/hdmi.h @@ -245,6 +245,7 @@ void msm_hdmi_bridge_destroy(struct drm_bridge *bridge); void msm_hdmi_connector_irq(struct drm_connector *connector); struct drm_connector *msm_hdmi_connector_init(struct hdmi *hdmi); +int msm_hdmi_hpd_enable(struct drm_connector *connector); /* * i2c adapter for ddc: diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c b/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c index 5ed4cab2819f..7ba6f52ed72c 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c @@ -40,7 +40,7 @@ static void msm_hdmi_power_on(struct drm_bridge *bridge) for (i = 0; i < config->pwr_reg_cnt; i++) { ret = regulator_enable(hdmi->pwr_regs[i]); if (ret) { - dev_err(dev->dev, "failed to enable pwr regulator: %s (%d)\n", + DRM_DEV_ERROR(dev->dev, "failed to enable pwr regulator: %s (%d)\n", config->pwr_reg_names[i], ret); } } @@ -49,7 +49,7 @@ static void msm_hdmi_power_on(struct drm_bridge *bridge) DBG("pixclock: %lu", hdmi->pixclock); ret = clk_set_rate(hdmi->pwr_clks[0], hdmi->pixclock); if (ret) { - dev_err(dev->dev, "failed to set pixel clk: %s (%d)\n", + DRM_DEV_ERROR(dev->dev, "failed to set pixel clk: %s (%d)\n", config->pwr_clk_names[0], ret); } } @@ -57,7 +57,7 @@ static void msm_hdmi_power_on(struct drm_bridge *bridge) for (i = 0; i < config->pwr_clk_cnt; i++) { ret = clk_prepare_enable(hdmi->pwr_clks[i]); if (ret) { - dev_err(dev->dev, "failed to enable pwr clk: %s (%d)\n", + DRM_DEV_ERROR(dev->dev, "failed to enable pwr clk: %s (%d)\n", config->pwr_clk_names[i], ret); } } @@ -82,7 +82,7 @@ static void power_off(struct drm_bridge *bridge) for (i = 0; i < config->pwr_reg_cnt; i++) { ret = regulator_disable(hdmi->pwr_regs[i]); if (ret) { - dev_err(dev->dev, "failed to disable pwr regulator: %s (%d)\n", + DRM_DEV_ERROR(dev->dev, "failed to disable pwr regulator: %s (%d)\n", config->pwr_reg_names[i], ret); } } @@ -106,7 +106,7 @@ static void msm_hdmi_config_avi_infoframe(struct hdmi *hdmi) len = hdmi_infoframe_pack(&frame, buffer, sizeof(buffer)); if (len < 0) { - dev_err(&hdmi->pdev->dev, + DRM_DEV_ERROR(&hdmi->pdev->dev, "failed to configure avi infoframe\n"); return; } diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_connector.c b/drivers/gpu/drm/msm/hdmi/hdmi_connector.c index e9c9a0af508e..a6eeab2c4dc3 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi_connector.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi_connector.c @@ -90,7 +90,7 @@ static int gpio_config(struct hdmi *hdmi, bool on) if (gpio.num != -1) { ret = gpio_request(gpio.num, gpio.label); if (ret) { - dev_err(dev, + DRM_DEV_ERROR(dev, "'%s'(%d) gpio_request failed: %d\n", gpio.label, gpio.num, ret); goto err; @@ -156,7 +156,7 @@ static void enable_hpd_clocks(struct hdmi *hdmi, bool enable) ret = clk_prepare_enable(hdmi->hpd_clks[i]); if (ret) { - dev_err(dev, + DRM_DEV_ERROR(dev, "failed to enable hpd clk: %s (%d)\n", config->hpd_clk_names[i], ret); } @@ -167,8 +167,9 @@ static void enable_hpd_clocks(struct hdmi *hdmi, bool enable) } } -static int hpd_enable(struct hdmi_connector *hdmi_connector) +int msm_hdmi_hpd_enable(struct drm_connector *connector) { + struct hdmi_connector *hdmi_connector = to_hdmi_connector(connector); struct hdmi *hdmi = hdmi_connector->hdmi; const struct hdmi_platform_config *config = hdmi->config; struct device *dev = &hdmi->pdev->dev; @@ -179,7 +180,7 @@ static int hpd_enable(struct hdmi_connector *hdmi_connector) for (i = 0; i < config->hpd_reg_cnt; i++) { ret = regulator_enable(hdmi->hpd_regs[i]); if (ret) { - dev_err(dev, "failed to enable hpd regulator: %s (%d)\n", + DRM_DEV_ERROR(dev, "failed to enable hpd regulator: %s (%d)\n", config->hpd_reg_names[i], ret); goto fail; } @@ -187,13 +188,13 @@ static int hpd_enable(struct hdmi_connector *hdmi_connector) ret = pinctrl_pm_select_default_state(dev); if (ret) { - dev_err(dev, "pinctrl state chg failed: %d\n", ret); + DRM_DEV_ERROR(dev, "pinctrl state chg failed: %d\n", ret); goto fail; } ret = gpio_config(hdmi, true); if (ret) { - dev_err(dev, "failed to configure GPIOs: %d\n", ret); + DRM_DEV_ERROR(dev, "failed to configure GPIOs: %d\n", ret); goto fail; } @@ -450,7 +451,6 @@ struct drm_connector *msm_hdmi_connector_init(struct hdmi *hdmi) { struct drm_connector *connector = NULL; struct hdmi_connector *hdmi_connector; - int ret; hdmi_connector = kzalloc(sizeof(*hdmi_connector), GFP_KERNEL); if (!hdmi_connector) @@ -471,12 +471,6 @@ struct drm_connector *msm_hdmi_connector_init(struct hdmi *hdmi) connector->interlace_allowed = 0; connector->doublescan_allowed = 0; - ret = hpd_enable(hdmi_connector); - if (ret) { - dev_err(&hdmi->pdev->dev, "failed to enable HPD: %d\n", ret); - return ERR_PTR(ret); - } - drm_connector_attach_encoder(connector, hdmi->encoder); return connector; diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_i2c.c b/drivers/gpu/drm/msm/hdmi/hdmi_i2c.c index 73e20219d431..25d2fe2c60e8 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi_i2c.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi_i2c.c @@ -66,7 +66,7 @@ static int ddc_clear_irq(struct hdmi_i2c_adapter *hdmi_i2c) } while ((ddc_int_ctrl & HDMI_DDC_INT_CTRL_SW_DONE_INT) && retry); if (!retry) { - dev_err(dev->dev, "timeout waiting for DDC\n"); + DRM_DEV_ERROR(dev->dev, "timeout waiting for DDC\n"); return -ETIMEDOUT; } diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_phy.c b/drivers/gpu/drm/msm/hdmi/hdmi_phy.c index 4157722d6b4d..1f4331ed69bd 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi_phy.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi_phy.c @@ -37,7 +37,7 @@ static int msm_hdmi_phy_resource_init(struct hdmi_phy *phy) reg = devm_regulator_get(dev, cfg->reg_names[i]); if (IS_ERR(reg)) { ret = PTR_ERR(reg); - dev_err(dev, "failed to get phy regulator: %s (%d)\n", + DRM_DEV_ERROR(dev, "failed to get phy regulator: %s (%d)\n", cfg->reg_names[i], ret); return ret; } @@ -51,7 +51,7 @@ static int msm_hdmi_phy_resource_init(struct hdmi_phy *phy) clk = msm_clk_get(phy->pdev, cfg->clk_names[i]); if (IS_ERR(clk)) { ret = PTR_ERR(clk); - dev_err(dev, "failed to get phy clock: %s (%d)\n", + DRM_DEV_ERROR(dev, "failed to get phy clock: %s (%d)\n", cfg->clk_names[i], ret); return ret; } @@ -73,14 +73,14 @@ int msm_hdmi_phy_resource_enable(struct hdmi_phy *phy) for (i = 0; i < cfg->num_regs; i++) { ret = regulator_enable(phy->regs[i]); if (ret) - dev_err(dev, "failed to enable regulator: %s (%d)\n", + DRM_DEV_ERROR(dev, "failed to enable regulator: %s (%d)\n", cfg->reg_names[i], ret); } for (i = 0; i < cfg->num_clks; i++) { ret = clk_prepare_enable(phy->clks[i]); if (ret) - dev_err(dev, "failed to enable clock: %s (%d)\n", + DRM_DEV_ERROR(dev, "failed to enable clock: %s (%d)\n", cfg->clk_names[i], ret); } @@ -159,7 +159,7 @@ static int msm_hdmi_phy_probe(struct platform_device *pdev) phy->mmio = msm_ioremap(pdev, "hdmi_phy", "HDMI_PHY"); if (IS_ERR(phy->mmio)) { - dev_err(dev, "%s: failed to map phy base\n", __func__); + DRM_DEV_ERROR(dev, "%s: failed to map phy base\n", __func__); return -ENOMEM; } @@ -177,7 +177,7 @@ static int msm_hdmi_phy_probe(struct platform_device *pdev) ret = msm_hdmi_phy_pll_init(pdev, phy->cfg->type); if (ret) { - dev_err(dev, "couldn't init PLL\n"); + DRM_DEV_ERROR(dev, "couldn't init PLL\n"); msm_hdmi_phy_resource_disable(phy); return ret; } diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_phy_8996.c b/drivers/gpu/drm/msm/hdmi/hdmi_phy_8996.c index 0df504c61833..318708f26731 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi_phy_8996.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi_phy_8996.c @@ -725,7 +725,7 @@ int msm_hdmi_pll_8996_init(struct platform_device *pdev) pll->mmio_qserdes_com = msm_ioremap(pdev, "hdmi_pll", "HDMI_PLL"); if (IS_ERR(pll->mmio_qserdes_com)) { - dev_err(dev, "failed to map pll base\n"); + DRM_DEV_ERROR(dev, "failed to map pll base\n"); return -ENOMEM; } @@ -737,7 +737,7 @@ int msm_hdmi_pll_8996_init(struct platform_device *pdev) pll->mmio_qserdes_tx[i] = msm_ioremap(pdev, name, label); if (IS_ERR(pll->mmio_qserdes_tx[i])) { - dev_err(dev, "failed to map pll base\n"); + DRM_DEV_ERROR(dev, "failed to map pll base\n"); return -ENOMEM; } } @@ -745,7 +745,7 @@ int msm_hdmi_pll_8996_init(struct platform_device *pdev) clk = devm_clk_register(dev, &pll->clk_hw); if (IS_ERR(clk)) { - dev_err(dev, "failed to register pll clock\n"); + DRM_DEV_ERROR(dev, "failed to register pll clock\n"); return -EINVAL; } diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_pll_8960.c b/drivers/gpu/drm/msm/hdmi/hdmi_pll_8960.c index 99590758c68b..c6dae6e437f9 100644 --- a/drivers/gpu/drm/msm/hdmi/hdmi_pll_8960.c +++ b/drivers/gpu/drm/msm/hdmi/hdmi_pll_8960.c @@ -445,7 +445,7 @@ int msm_hdmi_pll_8960_init(struct platform_device *pdev) pll->mmio = msm_ioremap(pdev, "hdmi_pll", "HDMI_PLL"); if (IS_ERR(pll->mmio)) { - dev_err(dev, "failed to map pll base\n"); + DRM_DEV_ERROR(dev, "failed to map pll base\n"); return -ENOMEM; } @@ -454,7 +454,7 @@ int msm_hdmi_pll_8960_init(struct platform_device *pdev) clk = devm_clk_register(dev, &pll->clk_hw); if (IS_ERR(clk)) { - dev_err(dev, "failed to register pll clock\n"); + DRM_DEV_ERROR(dev, "failed to register pll clock\n"); return -EINVAL; } diff --git a/drivers/gpu/drm/msm/msm_atomic.c b/drivers/gpu/drm/msm/msm_atomic.c index 4bcdeca7479d..f5b1256e32b6 100644 --- a/drivers/gpu/drm/msm/msm_atomic.c +++ b/drivers/gpu/drm/msm/msm_atomic.c @@ -34,7 +34,12 @@ static void msm_atomic_wait_for_commit_done(struct drm_device *dev, if (!new_crtc_state->active) continue; + if (drm_crtc_vblank_get(crtc)) + continue; + kms->funcs->wait_for_crtc_commit_done(kms, crtc); + + drm_crtc_vblank_put(crtc); } } @@ -78,7 +83,8 @@ void msm_atomic_commit_tail(struct drm_atomic_state *state) kms->funcs->commit(kms, state); } - msm_atomic_wait_for_commit_done(dev, state); + if (!state->legacy_cursor_update) + msm_atomic_wait_for_commit_done(dev, state); kms->funcs->complete_commit(kms, state); diff --git a/drivers/gpu/drm/msm/msm_debugfs.c b/drivers/gpu/drm/msm/msm_debugfs.c index f0da0d3c8a80..fb423d309e91 100644 --- a/drivers/gpu/drm/msm/msm_debugfs.c +++ b/drivers/gpu/drm/msm/msm_debugfs.c @@ -84,7 +84,7 @@ static int msm_gpu_open(struct inode *inode, struct file *file) ret = mutex_lock_interruptible(&dev->struct_mutex); if (ret) - return ret; + goto free_priv; pm_runtime_get_sync(&gpu->pdev->dev); show_priv->state = gpu->funcs->gpu_state_get(gpu); @@ -94,13 +94,20 @@ static int msm_gpu_open(struct inode *inode, struct file *file) if (IS_ERR(show_priv->state)) { ret = PTR_ERR(show_priv->state); - kfree(show_priv); - return ret; + goto free_priv; } show_priv->dev = dev; - return single_open(file, msm_gpu_show, show_priv); + ret = single_open(file, msm_gpu_show, show_priv); + if (ret) + goto free_priv; + + return 0; + +free_priv: + kfree(show_priv); + return ret; } static const struct file_operations msm_gpu_fops = { @@ -194,13 +201,13 @@ static int late_init_minor(struct drm_minor *minor) ret = msm_rd_debugfs_init(minor); if (ret) { - dev_err(minor->dev->dev, "could not install rd debugfs\n"); + DRM_DEV_ERROR(minor->dev->dev, "could not install rd debugfs\n"); return ret; } ret = msm_perf_debugfs_init(minor); if (ret) { - dev_err(minor->dev->dev, "could not install perf debugfs\n"); + DRM_DEV_ERROR(minor->dev->dev, "could not install perf debugfs\n"); return ret; } @@ -228,14 +235,14 @@ int msm_debugfs_init(struct drm_minor *minor) minor->debugfs_root, minor); if (ret) { - dev_err(dev->dev, "could not install msm_debugfs_list\n"); + DRM_DEV_ERROR(dev->dev, "could not install msm_debugfs_list\n"); return ret; } debugfs_create_file("gpu", S_IRUSR, minor->debugfs_root, dev, &msm_gpu_fops); - if (priv->kms->funcs->debugfs_init) { + if (priv->kms && priv->kms->funcs->debugfs_init) { ret = priv->kms->funcs->debugfs_init(priv->kms, minor); if (ret) return ret; diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c index 5e758d95751a..d2cdc7b553fe 100644 --- a/drivers/gpu/drm/msm/msm_drv.c +++ b/drivers/gpu/drm/msm/msm_drv.c @@ -23,8 +23,10 @@ #include "msm_drv.h" #include "msm_debugfs.h" #include "msm_fence.h" +#include "msm_gem.h" #include "msm_gpu.h" #include "msm_kms.h" +#include "adreno/adreno_gpu.h" /* @@ -35,9 +37,11 @@ * - 1.3.0 - adds GMEM_BASE + NR_RINGS params, SUBMITQUEUE_NEW + * SUBMITQUEUE_CLOSE ioctls, and MSM_INFO_IOVA flag for * MSM_GEM_INFO ioctl. + * - 1.4.0 - softpin, MSM_RELOC_BO_DUMP, and GEM_INFO support to set/get + * GEM object's debug name */ #define MSM_VERSION_MAJOR 1 -#define MSM_VERSION_MINOR 3 +#define MSM_VERSION_MINOR 4 #define MSM_VERSION_PATCHLEVEL 0 static const struct drm_mode_config_funcs mode_config_funcs = { @@ -170,7 +174,7 @@ void __iomem *msm_ioremap(struct platform_device *pdev, const char *name, res = platform_get_resource(pdev, IORESOURCE_MEM, 0); if (!res) { - dev_err(&pdev->dev, "failed to get memory resource: %s\n", name); + DRM_DEV_ERROR(&pdev->dev, "failed to get memory resource: %s\n", name); return ERR_PTR(-EINVAL); } @@ -178,7 +182,7 @@ void __iomem *msm_ioremap(struct platform_device *pdev, const char *name, ptr = devm_ioremap_nocache(&pdev->dev, res->start, size); if (!ptr) { - dev_err(&pdev->dev, "failed to ioremap: %s\n", name); + DRM_DEV_ERROR(&pdev->dev, "failed to ioremap: %s\n", name); return ERR_PTR(-ENOMEM); } @@ -358,6 +362,14 @@ static int get_mdp_ver(struct platform_device *pdev) #include <linux/of_address.h> +bool msm_use_mmu(struct drm_device *dev) +{ + struct msm_drm_private *priv = dev->dev_private; + + /* a2xx comes with its own MMU */ + return priv->is_a2xx || iommu_present(&platform_bus_type); +} + static int msm_init_vram(struct drm_device *dev) { struct msm_drm_private *priv = dev->dev_private; @@ -396,7 +408,7 @@ static int msm_init_vram(struct drm_device *dev) * Grab the entire CMA chunk carved out in early startup in * mach-msm: */ - } else if (!iommu_present(&platform_bus_type)) { + } else if (!msm_use_mmu(dev)) { DRM_INFO("using %s VRAM carveout\n", vram); size = memparse(vram, NULL); } @@ -419,12 +431,12 @@ static int msm_init_vram(struct drm_device *dev) p = dma_alloc_attrs(dev->dev, size, &priv->vram.paddr, GFP_KERNEL, attrs); if (!p) { - dev_err(dev->dev, "failed to allocate VRAM\n"); + DRM_DEV_ERROR(dev->dev, "failed to allocate VRAM\n"); priv->vram.paddr = 0; return -ENOMEM; } - dev_info(dev->dev, "VRAM: %08x->%08x\n", + DRM_DEV_INFO(dev->dev, "VRAM: %08x->%08x\n", (uint32_t)priv->vram.paddr, (uint32_t)(priv->vram.paddr + size)); } @@ -444,7 +456,7 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) ddev = drm_dev_alloc(drv, dev); if (IS_ERR(ddev)) { - dev_err(dev, "failed to allocate drm_device\n"); + DRM_DEV_ERROR(dev, "failed to allocate drm_device\n"); return PTR_ERR(ddev); } @@ -508,19 +520,16 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) priv->kms = kms; break; default: - kms = ERR_PTR(-ENODEV); + /* valid only for the dummy headless case, where of_node=NULL */ + WARN_ON(dev->of_node); + kms = NULL; break; } if (IS_ERR(kms)) { - /* - * NOTE: once we have GPU support, having no kms should not - * be considered fatal.. ideally we would still support gpu - * and (for example) use dmabuf/prime to share buffers with - * imx drm driver on iMX5 - */ - dev_err(dev, "failed to load kms\n"); + DRM_DEV_ERROR(dev, "failed to load kms\n"); ret = PTR_ERR(kms); + priv->kms = NULL; goto err_msm_uninit; } @@ -530,7 +539,7 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) if (kms) { ret = kms->funcs->hw_init(kms); if (ret) { - dev_err(dev, "kms hw init failed: %d\n", ret); + DRM_DEV_ERROR(dev, "kms hw init failed: %d\n", ret); goto err_msm_uninit; } } @@ -554,17 +563,18 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) kthread_run(kthread_worker_fn, &priv->disp_thread[i].worker, "crtc_commit:%d", priv->disp_thread[i].crtc_id); - ret = sched_setscheduler(priv->disp_thread[i].thread, - SCHED_FIFO, ¶m); - if (ret) - pr_warn("display thread priority update failed: %d\n", - ret); - if (IS_ERR(priv->disp_thread[i].thread)) { - dev_err(dev, "failed to create crtc_commit kthread\n"); + DRM_DEV_ERROR(dev, "failed to create crtc_commit kthread\n"); priv->disp_thread[i].thread = NULL; + goto err_msm_uninit; } + ret = sched_setscheduler(priv->disp_thread[i].thread, + SCHED_FIFO, ¶m); + if (ret) + dev_warn(dev, "disp_thread set priority failed: %d\n", + ret); + /* initialize event thread */ priv->event_thread[i].crtc_id = priv->crtcs[i]->base.id; kthread_init_worker(&priv->event_thread[i].worker); @@ -573,6 +583,12 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) kthread_run(kthread_worker_fn, &priv->event_thread[i].worker, "crtc_event:%d", priv->event_thread[i].crtc_id); + if (IS_ERR(priv->event_thread[i].thread)) { + DRM_DEV_ERROR(dev, "failed to create crtc_event kthread\n"); + priv->event_thread[i].thread = NULL; + goto err_msm_uninit; + } + /** * event thread should also run at same priority as disp_thread * because it is handling frame_done events. A lower priority @@ -581,39 +597,15 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) * failure at crtc commit level. */ ret = sched_setscheduler(priv->event_thread[i].thread, - SCHED_FIFO, ¶m); + SCHED_FIFO, ¶m); if (ret) - pr_warn("display event thread priority update failed: %d\n", - ret); - - if (IS_ERR(priv->event_thread[i].thread)) { - dev_err(dev, "failed to create crtc_event kthread\n"); - priv->event_thread[i].thread = NULL; - } - - if ((!priv->disp_thread[i].thread) || - !priv->event_thread[i].thread) { - /* clean up previously created threads if any */ - for ( ; i >= 0; i--) { - if (priv->disp_thread[i].thread) { - kthread_stop( - priv->disp_thread[i].thread); - priv->disp_thread[i].thread = NULL; - } - - if (priv->event_thread[i].thread) { - kthread_stop( - priv->event_thread[i].thread); - priv->event_thread[i].thread = NULL; - } - } - goto err_msm_uninit; - } + dev_warn(dev, "event_thread set priority failed:%d\n", + ret); } ret = drm_vblank_init(ddev, priv->num_crtcs); if (ret < 0) { - dev_err(dev, "failed to initialize vblank\n"); + DRM_DEV_ERROR(dev, "failed to initialize vblank\n"); goto err_msm_uninit; } @@ -622,7 +614,7 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) ret = drm_irq_install(ddev, kms->irq); pm_runtime_put_sync(dev); if (ret < 0) { - dev_err(dev, "failed to install IRQ handler\n"); + DRM_DEV_ERROR(dev, "failed to install IRQ handler\n"); goto err_msm_uninit; } } @@ -634,7 +626,7 @@ static int msm_drm_init(struct device *dev, struct drm_driver *drv) drm_mode_config_reset(ddev); #ifdef CONFIG_DRM_FBDEV_EMULATION - if (fbdev) + if (kms && fbdev) priv->fbdev = msm_fbdev_init(ddev); #endif @@ -742,7 +734,11 @@ static int msm_irq_postinstall(struct drm_device *dev) struct msm_drm_private *priv = dev->dev_private; struct msm_kms *kms = priv->kms; BUG_ON(!kms); - return kms->funcs->irq_postinstall(kms); + + if (kms->funcs->irq_postinstall) + return kms->funcs->irq_postinstall(kms); + + return 0; } static void msm_irq_uninstall(struct drm_device *dev) @@ -809,7 +805,7 @@ static int msm_ioctl_gem_new(struct drm_device *dev, void *data, } return msm_gem_new_handle(dev, file, args->size, - args->flags, &args->handle); + args->flags, &args->handle, NULL); } static inline ktime_t to_ktime(struct drm_msm_timespec timeout) @@ -867,6 +863,10 @@ static int msm_ioctl_gem_info_iova(struct drm_device *dev, if (!priv->gpu) return -EINVAL; + /* + * Don't pin the memory here - just get an address so that userspace can + * be productive + */ return msm_gem_get_iova(obj, priv->gpu->aspace, iova); } @@ -875,23 +875,66 @@ static int msm_ioctl_gem_info(struct drm_device *dev, void *data, { struct drm_msm_gem_info *args = data; struct drm_gem_object *obj; - int ret = 0; + struct msm_gem_object *msm_obj; + int i, ret = 0; + + if (args->pad) + return -EINVAL; - if (args->flags & ~MSM_INFO_FLAGS) + switch (args->info) { + case MSM_INFO_GET_OFFSET: + case MSM_INFO_GET_IOVA: + /* value returned as immediate, not pointer, so len==0: */ + if (args->len) + return -EINVAL; + break; + case MSM_INFO_SET_NAME: + case MSM_INFO_GET_NAME: + break; + default: return -EINVAL; + } obj = drm_gem_object_lookup(file, args->handle); if (!obj) return -ENOENT; - if (args->flags & MSM_INFO_IOVA) { - uint64_t iova; + msm_obj = to_msm_bo(obj); - ret = msm_ioctl_gem_info_iova(dev, obj, &iova); - if (!ret) - args->offset = iova; - } else { - args->offset = msm_gem_mmap_offset(obj); + switch (args->info) { + case MSM_INFO_GET_OFFSET: + args->value = msm_gem_mmap_offset(obj); + break; + case MSM_INFO_GET_IOVA: + ret = msm_ioctl_gem_info_iova(dev, obj, &args->value); + break; + case MSM_INFO_SET_NAME: + /* length check should leave room for terminating null: */ + if (args->len >= sizeof(msm_obj->name)) { + ret = -EINVAL; + break; + } + ret = copy_from_user(msm_obj->name, + u64_to_user_ptr(args->value), args->len); + msm_obj->name[args->len] = '\0'; + for (i = 0; i < args->len; i++) { + if (!isprint(msm_obj->name[i])) { + msm_obj->name[i] = '\0'; + break; + } + } + break; + case MSM_INFO_GET_NAME: + if (args->value && (args->len < strlen(msm_obj->name))) { + ret = -EINVAL; + break; + } + args->len = strlen(msm_obj->name); + if (args->value) { + ret = copy_to_user(u64_to_user_ptr(args->value), + msm_obj->name, args->len); + } + break; } drm_gem_object_put_unlocked(obj); @@ -1070,18 +1113,15 @@ static int msm_pm_suspend(struct device *dev) { struct drm_device *ddev = dev_get_drvdata(dev); struct msm_drm_private *priv = ddev->dev_private; - struct msm_kms *kms = priv->kms; - /* TODO: Use atomic helper suspend/resume */ - if (kms && kms->funcs && kms->funcs->pm_suspend) - return kms->funcs->pm_suspend(dev); - - drm_kms_helper_poll_disable(ddev); + if (WARN_ON(priv->pm_state)) + drm_atomic_state_put(priv->pm_state); priv->pm_state = drm_atomic_helper_suspend(ddev); if (IS_ERR(priv->pm_state)) { - drm_kms_helper_poll_enable(ddev); - return PTR_ERR(priv->pm_state); + int ret = PTR_ERR(priv->pm_state); + DRM_ERROR("Failed to suspend dpu, %d\n", ret); + return ret; } return 0; @@ -1091,16 +1131,16 @@ static int msm_pm_resume(struct device *dev) { struct drm_device *ddev = dev_get_drvdata(dev); struct msm_drm_private *priv = ddev->dev_private; - struct msm_kms *kms = priv->kms; + int ret; - /* TODO: Use atomic helper suspend/resume */ - if (kms && kms->funcs && kms->funcs->pm_resume) - return kms->funcs->pm_resume(dev); + if (WARN_ON(!priv->pm_state)) + return -ENOENT; - drm_atomic_helper_resume(ddev, priv->pm_state); - drm_kms_helper_poll_enable(ddev); + ret = drm_atomic_helper_resume(ddev, priv->pm_state); + if (!ret) + priv->pm_state = NULL; - return 0; + return ret; } #endif @@ -1185,7 +1225,7 @@ static int add_components_mdp(struct device *mdp_dev, ret = of_graph_parse_endpoint(ep_node, &ep); if (ret) { - dev_err(mdp_dev, "unable to parse port endpoint\n"); + DRM_DEV_ERROR(mdp_dev, "unable to parse port endpoint\n"); of_node_put(ep_node); return ret; } @@ -1207,8 +1247,10 @@ static int add_components_mdp(struct device *mdp_dev, if (!intf) continue; - drm_of_component_match_add(master_dev, matchptr, compare_of, - intf); + if (of_device_is_available(intf)) + drm_of_component_match_add(master_dev, matchptr, + compare_of, intf); + of_node_put(intf); } @@ -1236,13 +1278,13 @@ static int add_display_components(struct device *dev, of_device_is_compatible(dev->of_node, "qcom,sdm845-mdss")) { ret = of_platform_populate(dev->of_node, NULL, NULL, dev); if (ret) { - dev_err(dev, "failed to populate children devices\n"); + DRM_DEV_ERROR(dev, "failed to populate children devices\n"); return ret; } mdp_dev = device_find_child(dev, NULL, compare_name_mdp); if (!mdp_dev) { - dev_err(dev, "failed to find MDSS MDP node\n"); + DRM_DEV_ERROR(dev, "failed to find MDSS MDP node\n"); of_platform_depopulate(dev); return -ENODEV; } @@ -1272,6 +1314,7 @@ static int add_display_components(struct device *dev, static const struct of_device_id msm_gpu_match[] = { { .compatible = "qcom,adreno" }, { .compatible = "qcom,adreno-3xx" }, + { .compatible = "amd,imageon" }, { .compatible = "qcom,kgsl-3d0" }, { }, }; @@ -1316,9 +1359,11 @@ static int msm_pdev_probe(struct platform_device *pdev) struct component_match *match = NULL; int ret; - ret = add_display_components(&pdev->dev, &match); - if (ret) - return ret; + if (get_mdp_ver(pdev)) { + ret = add_display_components(&pdev->dev, &match); + if (ret) + return ret; + } ret = add_gpu_components(&pdev->dev, &match); if (ret) diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h index 9d11f321f5a9..9cd6a96c6bf2 100644 --- a/drivers/gpu/drm/msm/msm_drv.h +++ b/drivers/gpu/drm/msm/msm_drv.h @@ -179,6 +179,8 @@ struct msm_drm_private { /* when we have more than one 'msm_gpu' these need to be an array: */ struct msm_gpu *gpu; struct msm_file_private *lastctx; + /* gpu is only set on open(), but we need this info earlier */ + bool is_a2xx; struct drm_fb_helper *fbdev; @@ -241,10 +243,16 @@ struct drm_atomic_state *msm_atomic_state_alloc(struct drm_device *dev); void msm_atomic_state_clear(struct drm_atomic_state *state); void msm_atomic_state_free(struct drm_atomic_state *state); +int msm_gem_init_vma(struct msm_gem_address_space *aspace, + struct msm_gem_vma *vma, int npages); +void msm_gem_purge_vma(struct msm_gem_address_space *aspace, + struct msm_gem_vma *vma); void msm_gem_unmap_vma(struct msm_gem_address_space *aspace, - struct msm_gem_vma *vma, struct sg_table *sgt); + struct msm_gem_vma *vma); int msm_gem_map_vma(struct msm_gem_address_space *aspace, struct msm_gem_vma *vma, struct sg_table *sgt, int npages); +void msm_gem_close_vma(struct msm_gem_address_space *aspace, + struct msm_gem_vma *vma); void msm_gem_address_space_put(struct msm_gem_address_space *aspace); @@ -252,9 +260,15 @@ struct msm_gem_address_space * msm_gem_address_space_create(struct device *dev, struct iommu_domain *domain, const char *name); +struct msm_gem_address_space * +msm_gem_address_space_create_a2xx(struct device *dev, struct msm_gpu *gpu, + const char *name, uint64_t va_start, uint64_t va_end); + int msm_register_mmu(struct drm_device *dev, struct msm_mmu *mmu); void msm_unregister_mmu(struct drm_device *dev, struct msm_mmu *mmu); +bool msm_use_mmu(struct drm_device *dev); + void msm_gem_submit_free(struct msm_gem_submit *submit); int msm_ioctl_gem_submit(struct drm_device *dev, void *data, struct drm_file *file); @@ -269,12 +283,14 @@ vm_fault_t msm_gem_fault(struct vm_fault *vmf); uint64_t msm_gem_mmap_offset(struct drm_gem_object *obj); int msm_gem_get_iova(struct drm_gem_object *obj, struct msm_gem_address_space *aspace, uint64_t *iova); +int msm_gem_get_and_pin_iova(struct drm_gem_object *obj, + struct msm_gem_address_space *aspace, uint64_t *iova); uint64_t msm_gem_iova(struct drm_gem_object *obj, struct msm_gem_address_space *aspace); +void msm_gem_unpin_iova(struct drm_gem_object *obj, + struct msm_gem_address_space *aspace); struct page **msm_gem_get_pages(struct drm_gem_object *obj); void msm_gem_put_pages(struct drm_gem_object *obj); -void msm_gem_put_iova(struct drm_gem_object *obj, - struct msm_gem_address_space *aspace); int msm_gem_dumb_create(struct drm_file *file, struct drm_device *dev, struct drm_mode_create_dumb *args); int msm_gem_dumb_map_offset(struct drm_file *file, struct drm_device *dev, @@ -301,7 +317,7 @@ int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op, ktime_t *timeout); int msm_gem_cpu_fini(struct drm_gem_object *obj); void msm_gem_free_object(struct drm_gem_object *obj); int msm_gem_new_handle(struct drm_device *dev, struct drm_file *file, - uint32_t size, uint32_t flags, uint32_t *handle); + uint32_t size, uint32_t flags, uint32_t *handle, char *name); struct drm_gem_object *msm_gem_new(struct drm_device *dev, uint32_t size, uint32_t flags); struct drm_gem_object *msm_gem_new_locked(struct drm_device *dev, @@ -312,9 +328,13 @@ void *msm_gem_kernel_new(struct drm_device *dev, uint32_t size, void *msm_gem_kernel_new_locked(struct drm_device *dev, uint32_t size, uint32_t flags, struct msm_gem_address_space *aspace, struct drm_gem_object **bo, uint64_t *iova); +void msm_gem_kernel_put(struct drm_gem_object *bo, + struct msm_gem_address_space *aspace, bool locked); struct drm_gem_object *msm_gem_import(struct drm_device *dev, struct dma_buf *dmabuf, struct sg_table *sgt); +void msm_gem_object_set_name(struct drm_gem_object *bo, const char *fmt, ...); + int msm_framebuffer_prepare(struct drm_framebuffer *fb, struct msm_gem_address_space *aspace); void msm_framebuffer_cleanup(struct drm_framebuffer *fb, diff --git a/drivers/gpu/drm/msm/msm_fb.c b/drivers/gpu/drm/msm/msm_fb.c index 2a7348aeb38d..67dfd8d3dc12 100644 --- a/drivers/gpu/drm/msm/msm_fb.c +++ b/drivers/gpu/drm/msm/msm_fb.c @@ -66,7 +66,7 @@ int msm_framebuffer_prepare(struct drm_framebuffer *fb, uint64_t iova; for (i = 0; i < n; i++) { - ret = msm_gem_get_iova(fb->obj[i], aspace, &iova); + ret = msm_gem_get_and_pin_iova(fb->obj[i], aspace, &iova); DBG("FB[%u]: iova[%d]: %08llx (%d)", fb->base.id, i, iova, ret); if (ret) return ret; @@ -81,7 +81,7 @@ void msm_framebuffer_cleanup(struct drm_framebuffer *fb, int i, n = fb->format->num_planes; for (i = 0; i < n; i++) - msm_gem_put_iova(fb->obj[i], aspace); + msm_gem_unpin_iova(fb->obj[i], aspace); } uint32_t msm_framebuffer_iova(struct drm_framebuffer *fb, @@ -154,7 +154,7 @@ static struct drm_framebuffer *msm_framebuffer_init(struct drm_device *dev, format = kms->funcs->get_format(kms, mode_cmd->pixel_format, mode_cmd->modifier[0]); if (!format) { - dev_err(dev->dev, "unsupported pixel format: %4.4s\n", + DRM_DEV_ERROR(dev->dev, "unsupported pixel format: %4.4s\n", (char *)&mode_cmd->pixel_format); ret = -EINVAL; goto fail; @@ -196,7 +196,7 @@ static struct drm_framebuffer *msm_framebuffer_init(struct drm_device *dev, ret = drm_framebuffer_init(dev, fb, &msm_framebuffer_funcs); if (ret) { - dev_err(dev->dev, "framebuffer init failed: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "framebuffer init failed: %d\n", ret); goto fail; } @@ -233,13 +233,15 @@ msm_alloc_stolen_fb(struct drm_device *dev, int w, int h, int p, uint32_t format bo = msm_gem_new(dev, size, MSM_BO_SCANOUT | MSM_BO_WC); } if (IS_ERR(bo)) { - dev_err(dev->dev, "failed to allocate buffer object\n"); + DRM_DEV_ERROR(dev->dev, "failed to allocate buffer object\n"); return ERR_CAST(bo); } + msm_gem_object_set_name(bo, "stolenfb"); + fb = msm_framebuffer_init(dev, &mode_cmd, &bo); if (IS_ERR(fb)) { - dev_err(dev->dev, "failed to allocate fb\n"); + DRM_DEV_ERROR(dev->dev, "failed to allocate fb\n"); /* note: if fb creation failed, we can't rely on fb destroy * to unref the bo: */ diff --git a/drivers/gpu/drm/msm/msm_fbdev.c b/drivers/gpu/drm/msm/msm_fbdev.c index 456622b46335..c03e860ba737 100644 --- a/drivers/gpu/drm/msm/msm_fbdev.c +++ b/drivers/gpu/drm/msm/msm_fbdev.c @@ -91,7 +91,7 @@ static int msm_fbdev_create(struct drm_fb_helper *helper, sizes->surface_height, pitch, format); if (IS_ERR(fb)) { - dev_err(dev->dev, "failed to allocate fb\n"); + DRM_DEV_ERROR(dev->dev, "failed to allocate fb\n"); return PTR_ERR(fb); } @@ -104,15 +104,15 @@ static int msm_fbdev_create(struct drm_fb_helper *helper, * in panic (ie. lock-safe, etc) we could avoid pinning the * buffer now: */ - ret = msm_gem_get_iova(bo, priv->kms->aspace, &paddr); + ret = msm_gem_get_and_pin_iova(bo, priv->kms->aspace, &paddr); if (ret) { - dev_err(dev->dev, "failed to get buffer obj iova: %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to get buffer obj iova: %d\n", ret); goto fail_unlock; } fbi = drm_fb_helper_alloc_fbi(helper); if (IS_ERR(fbi)) { - dev_err(dev->dev, "failed to allocate fb info\n"); + DRM_DEV_ERROR(dev->dev, "failed to allocate fb info\n"); ret = PTR_ERR(fbi); goto fail_unlock; } @@ -176,7 +176,7 @@ struct drm_fb_helper *msm_fbdev_init(struct drm_device *dev) ret = drm_fb_helper_init(dev, helper, priv->num_connectors); if (ret) { - dev_err(dev->dev, "could not init fbdev: ret=%d\n", ret); + DRM_DEV_ERROR(dev->dev, "could not init fbdev: ret=%d\n", ret); goto fail; } diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c index f59ca27a4a35..20c979a7fc9c 100644 --- a/drivers/gpu/drm/msm/msm_gem.c +++ b/drivers/gpu/drm/msm/msm_gem.c @@ -88,7 +88,7 @@ static struct page **get_pages(struct drm_gem_object *obj) p = get_pages_vram(obj, npages); if (IS_ERR(p)) { - dev_err(dev->dev, "could not get pages: %ld\n", + DRM_DEV_ERROR(dev->dev, "could not get pages: %ld\n", PTR_ERR(p)); return p; } @@ -99,7 +99,7 @@ static struct page **get_pages(struct drm_gem_object *obj) if (IS_ERR(msm_obj->sgt)) { void *ptr = ERR_CAST(msm_obj->sgt); - dev_err(dev->dev, "failed to allocate sgt\n"); + DRM_DEV_ERROR(dev->dev, "failed to allocate sgt\n"); msm_obj->sgt = NULL; return ptr; } @@ -280,7 +280,7 @@ static uint64_t mmap_offset(struct drm_gem_object *obj) ret = drm_gem_create_mmap_offset(obj); if (ret) { - dev_err(dev->dev, "could not allocate mmap offset\n"); + DRM_DEV_ERROR(dev->dev, "could not allocate mmap offset\n"); return 0; } @@ -352,63 +352,104 @@ put_iova(struct drm_gem_object *obj) WARN_ON(!mutex_is_locked(&msm_obj->lock)); list_for_each_entry_safe(vma, tmp, &msm_obj->vmas, list) { - msm_gem_unmap_vma(vma->aspace, vma, msm_obj->sgt); + msm_gem_purge_vma(vma->aspace, vma); + msm_gem_close_vma(vma->aspace, vma); del_vma(vma); } } -/* get iova, taking a reference. Should have a matching put */ -int msm_gem_get_iova(struct drm_gem_object *obj, +static int msm_gem_get_iova_locked(struct drm_gem_object *obj, struct msm_gem_address_space *aspace, uint64_t *iova) { struct msm_gem_object *msm_obj = to_msm_bo(obj); struct msm_gem_vma *vma; int ret = 0; - mutex_lock(&msm_obj->lock); - - if (WARN_ON(msm_obj->madv != MSM_MADV_WILLNEED)) { - mutex_unlock(&msm_obj->lock); - return -EBUSY; - } + WARN_ON(!mutex_is_locked(&msm_obj->lock)); vma = lookup_vma(obj, aspace); if (!vma) { - struct page **pages; - vma = add_vma(obj, aspace); - if (IS_ERR(vma)) { - ret = PTR_ERR(vma); - goto unlock; - } + if (IS_ERR(vma)) + return PTR_ERR(vma); - pages = get_pages(obj); - if (IS_ERR(pages)) { - ret = PTR_ERR(pages); - goto fail; + ret = msm_gem_init_vma(aspace, vma, obj->size >> PAGE_SHIFT); + if (ret) { + del_vma(vma); + return ret; } - - ret = msm_gem_map_vma(aspace, vma, msm_obj->sgt, - obj->size >> PAGE_SHIFT); - if (ret) - goto fail; } *iova = vma->iova; + return 0; +} + +static int msm_gem_pin_iova(struct drm_gem_object *obj, + struct msm_gem_address_space *aspace) +{ + struct msm_gem_object *msm_obj = to_msm_bo(obj); + struct msm_gem_vma *vma; + struct page **pages; + + WARN_ON(!mutex_is_locked(&msm_obj->lock)); + + if (WARN_ON(msm_obj->madv != MSM_MADV_WILLNEED)) + return -EBUSY; + + vma = lookup_vma(obj, aspace); + if (WARN_ON(!vma)) + return -EINVAL; + + pages = get_pages(obj); + if (IS_ERR(pages)) + return PTR_ERR(pages); + + return msm_gem_map_vma(aspace, vma, msm_obj->sgt, + obj->size >> PAGE_SHIFT); +} + +/* get iova and pin it. Should have a matching put */ +int msm_gem_get_and_pin_iova(struct drm_gem_object *obj, + struct msm_gem_address_space *aspace, uint64_t *iova) +{ + struct msm_gem_object *msm_obj = to_msm_bo(obj); + u64 local; + int ret; + + mutex_lock(&msm_obj->lock); + + ret = msm_gem_get_iova_locked(obj, aspace, &local); + + if (!ret) + ret = msm_gem_pin_iova(obj, aspace); + + if (!ret) + *iova = local; mutex_unlock(&msm_obj->lock); - return 0; + return ret; +} -fail: - del_vma(vma); -unlock: +/* + * Get an iova but don't pin it. Doesn't need a put because iovas are currently + * valid for the life of the object + */ +int msm_gem_get_iova(struct drm_gem_object *obj, + struct msm_gem_address_space *aspace, uint64_t *iova) +{ + struct msm_gem_object *msm_obj = to_msm_bo(obj); + int ret; + + mutex_lock(&msm_obj->lock); + ret = msm_gem_get_iova_locked(obj, aspace, iova); mutex_unlock(&msm_obj->lock); + return ret; } /* get iova without taking a reference, used in places where you have - * already done a 'msm_gem_get_iova()'. + * already done a 'msm_gem_get_and_pin_iova' or 'msm_gem_get_iova' */ uint64_t msm_gem_iova(struct drm_gem_object *obj, struct msm_gem_address_space *aspace) @@ -424,15 +465,24 @@ uint64_t msm_gem_iova(struct drm_gem_object *obj, return vma ? vma->iova : 0; } -void msm_gem_put_iova(struct drm_gem_object *obj, +/* + * Unpin a iova by updating the reference counts. The memory isn't actually + * purged until something else (shrinker, mm_notifier, destroy, etc) decides + * to get rid of it + */ +void msm_gem_unpin_iova(struct drm_gem_object *obj, struct msm_gem_address_space *aspace) { - // XXX TODO .. - // NOTE: probably don't need a _locked() version.. we wouldn't - // normally unmap here, but instead just mark that it could be - // unmapped (if the iova refcnt drops to zero), but then later - // if another _get_iova_locked() fails we can start unmapping - // things that are no longer needed.. + struct msm_gem_object *msm_obj = to_msm_bo(obj); + struct msm_gem_vma *vma; + + mutex_lock(&msm_obj->lock); + vma = lookup_vma(obj, aspace); + + if (!WARN_ON(!vma)) + msm_gem_unmap_vma(aspace, vma); + + mutex_unlock(&msm_obj->lock); } int msm_gem_dumb_create(struct drm_file *file, struct drm_device *dev, @@ -441,7 +491,7 @@ int msm_gem_dumb_create(struct drm_file *file, struct drm_device *dev, args->pitch = align_pitch(args->width, args->bpp); args->size = PAGE_ALIGN(args->pitch * args->height); return msm_gem_new_handle(dev, file, args->size, - MSM_BO_SCANOUT | MSM_BO_WC, &args->handle); + MSM_BO_SCANOUT | MSM_BO_WC, &args->handle, "dumb"); } int msm_gem_dumb_map_offset(struct drm_file *file, struct drm_device *dev, @@ -473,7 +523,7 @@ static void *get_vaddr(struct drm_gem_object *obj, unsigned madv) mutex_lock(&msm_obj->lock); if (WARN_ON(msm_obj->madv > madv)) { - dev_err(obj->dev->dev, "Invalid madv state: %u vs %u\n", + DRM_DEV_ERROR(obj->dev->dev, "Invalid madv state: %u vs %u\n", msm_obj->madv, madv); mutex_unlock(&msm_obj->lock); return ERR_PTR(-EBUSY); @@ -708,7 +758,7 @@ static void describe_fence(struct dma_fence *fence, const char *type, struct seq_file *m) { if (!dma_fence_is_signaled(fence)) - seq_printf(m, "\t%9s: %s %s seq %u\n", type, + seq_printf(m, "\t%9s: %s %s seq %llu\n", type, fence->ops->get_driver_name(fence), fence->ops->get_timeline_name(fence), fence->seqno); @@ -739,16 +789,24 @@ void msm_gem_describe(struct drm_gem_object *obj, struct seq_file *m) break; } - seq_printf(m, "%08x: %c %2d (%2d) %08llx %p\t", + seq_printf(m, "%08x: %c %2d (%2d) %08llx %p", msm_obj->flags, is_active(msm_obj) ? 'A' : 'I', obj->name, kref_read(&obj->refcount), off, msm_obj->vaddr); - /* FIXME: we need to print the address space here too */ - list_for_each_entry(vma, &msm_obj->vmas, list) - seq_printf(m, " %08llx", vma->iova); + seq_printf(m, " %08zu %9s %-32s\n", obj->size, madv, msm_obj->name); + + if (!list_empty(&msm_obj->vmas)) { + + seq_puts(m, " vmas:"); - seq_printf(m, " %zu%s\n", obj->size, madv); + list_for_each_entry(vma, &msm_obj->vmas, list) + seq_printf(m, " [%s: %08llx,%s,inuse=%d]", vma->aspace->name, + vma->iova, vma->mapped ? "mapped" : "unmapped", + vma->inuse); + + seq_puts(m, "\n"); + } rcu_read_lock(); fobj = rcu_dereference(robj->fence); @@ -775,9 +833,10 @@ void msm_gem_describe_objects(struct list_head *list, struct seq_file *m) int count = 0; size_t size = 0; + seq_puts(m, " flags id ref offset kaddr size madv name\n"); list_for_each_entry(msm_obj, list, mm_list) { struct drm_gem_object *obj = &msm_obj->base; - seq_printf(m, " "); + seq_puts(m, " "); msm_gem_describe(obj, m); count++; size += obj->size; @@ -831,7 +890,8 @@ void msm_gem_free_object(struct drm_gem_object *obj) /* convenience method to construct a GEM buffer object, and userspace handle */ int msm_gem_new_handle(struct drm_device *dev, struct drm_file *file, - uint32_t size, uint32_t flags, uint32_t *handle) + uint32_t size, uint32_t flags, uint32_t *handle, + char *name) { struct drm_gem_object *obj; int ret; @@ -841,6 +901,9 @@ int msm_gem_new_handle(struct drm_device *dev, struct drm_file *file, if (IS_ERR(obj)) return PTR_ERR(obj); + if (name) + msm_gem_object_set_name(obj, "%s", name); + ret = drm_gem_handle_create(file, obj, handle); /* drop reference from allocate - handle holds it now */ @@ -864,7 +927,7 @@ static int msm_gem_new_impl(struct drm_device *dev, case MSM_BO_WC: break; default: - dev_err(dev->dev, "invalid cache flag: %x\n", + DRM_DEV_ERROR(dev->dev, "invalid cache flag: %x\n", (flags & MSM_BO_CACHE_MASK)); return -EINVAL; } @@ -912,9 +975,9 @@ static struct drm_gem_object *_msm_gem_new(struct drm_device *dev, size = PAGE_ALIGN(size); - if (!iommu_present(&platform_bus_type)) + if (!msm_use_mmu(dev)) use_vram = true; - else if ((flags & MSM_BO_STOLEN) && priv->vram.size) + else if ((flags & (MSM_BO_STOLEN | MSM_BO_SCANOUT)) && priv->vram.size) use_vram = true; if (WARN_ON(use_vram && !priv->vram.size)) @@ -989,8 +1052,8 @@ struct drm_gem_object *msm_gem_import(struct drm_device *dev, int ret, npages; /* if we don't have IOMMU, don't bother pretending we can import: */ - if (!iommu_present(&platform_bus_type)) { - dev_err(dev->dev, "cannot import without IOMMU\n"); + if (!msm_use_mmu(dev)) { + DRM_DEV_ERROR(dev->dev, "cannot import without IOMMU\n"); return ERR_PTR(-EINVAL); } @@ -1040,24 +1103,30 @@ static void *_msm_gem_kernel_new(struct drm_device *dev, uint32_t size, return ERR_CAST(obj); if (iova) { - ret = msm_gem_get_iova(obj, aspace, iova); - if (ret) { - drm_gem_object_put(obj); - return ERR_PTR(ret); - } + ret = msm_gem_get_and_pin_iova(obj, aspace, iova); + if (ret) + goto err; } vaddr = msm_gem_get_vaddr(obj); if (IS_ERR(vaddr)) { - msm_gem_put_iova(obj, aspace); - drm_gem_object_put(obj); - return ERR_CAST(vaddr); + msm_gem_unpin_iova(obj, aspace); + ret = PTR_ERR(vaddr); + goto err; } if (bo) *bo = obj; return vaddr; +err: + if (locked) + drm_gem_object_put(obj); + else + drm_gem_object_put_unlocked(obj); + + return ERR_PTR(ret); + } void *msm_gem_kernel_new(struct drm_device *dev, uint32_t size, @@ -1073,3 +1142,31 @@ void *msm_gem_kernel_new_locked(struct drm_device *dev, uint32_t size, { return _msm_gem_kernel_new(dev, size, flags, aspace, bo, iova, true); } + +void msm_gem_kernel_put(struct drm_gem_object *bo, + struct msm_gem_address_space *aspace, bool locked) +{ + if (IS_ERR_OR_NULL(bo)) + return; + + msm_gem_put_vaddr(bo); + msm_gem_unpin_iova(bo, aspace); + + if (locked) + drm_gem_object_put(bo); + else + drm_gem_object_put_unlocked(bo); +} + +void msm_gem_object_set_name(struct drm_gem_object *bo, const char *fmt, ...) +{ + struct msm_gem_object *msm_obj = to_msm_bo(bo); + va_list ap; + + if (!fmt) + return; + + va_start(ap, fmt); + vsnprintf(msm_obj->name, sizeof(msm_obj->name), fmt, ap); + va_end(ap); +} diff --git a/drivers/gpu/drm/msm/msm_gem.h b/drivers/gpu/drm/msm/msm_gem.h index c5d9bd3e47a8..2064fac871b8 100644 --- a/drivers/gpu/drm/msm/msm_gem.h +++ b/drivers/gpu/drm/msm/msm_gem.h @@ -41,6 +41,8 @@ struct msm_gem_vma { uint64_t iova; struct msm_gem_address_space *aspace; struct list_head list; /* node in msm_gem_object::vmas */ + bool mapped; + int inuse; }; struct msm_gem_object { @@ -91,6 +93,8 @@ struct msm_gem_object { */ struct drm_mm_node *vram_node; struct mutex lock; /* Protects resources associated with bo */ + + char name[32]; /* Identifier to print for the debugfs files */ }; #define to_msm_bo(x) container_of(x, struct msm_gem_object, base) @@ -150,6 +154,7 @@ struct msm_gem_submit { struct msm_ringbuffer *ring; unsigned int nr_cmds; unsigned int nr_bos; + u32 ident; /* A "identifier" for the submit for logging */ struct { uint32_t type; uint32_t size; /* in dwords */ diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c index a90aedd6883a..12b983fc0b56 100644 --- a/drivers/gpu/drm/msm/msm_gem_submit.c +++ b/drivers/gpu/drm/msm/msm_gem_submit.c @@ -20,6 +20,7 @@ #include "msm_drv.h" #include "msm_gpu.h" #include "msm_gem.h" +#include "msm_gpu_trace.h" /* * Cmdstream submission: @@ -48,7 +49,6 @@ static struct msm_gem_submit *submit_create(struct drm_device *dev, submit->dev = dev; submit->gpu = gpu; submit->fence = NULL; - submit->pid = get_pid(task_pid(current)); submit->cmd = (void *)&submit->bos[nr_bos]; submit->queue = queue; submit->ring = gpu->rb[queue->prio]; @@ -77,7 +77,7 @@ void msm_gem_submit_free(struct msm_gem_submit *submit) static inline unsigned long __must_check copy_from_user_inatomic(void *to, const void __user *from, unsigned long n) { - if (access_ok(VERIFY_READ, from, n)) + if (access_ok(from, n)) return __copy_from_user_inatomic(to, from, n); return -EFAULT; } @@ -114,8 +114,11 @@ static int submit_lookup_objects(struct msm_gem_submit *submit, pagefault_disable(); } +/* at least one of READ and/or WRITE flags should be set: */ +#define MANDATORY_FLAGS (MSM_SUBMIT_BO_READ | MSM_SUBMIT_BO_WRITE) + if ((submit_bo.flags & ~MSM_SUBMIT_BO_FLAGS) || - !(submit_bo.flags & MSM_SUBMIT_BO_FLAGS)) { + !(submit_bo.flags & MANDATORY_FLAGS)) { DRM_ERROR("invalid flags: %x\n", submit_bo.flags); ret = -EINVAL; goto out_unlock; @@ -167,7 +170,7 @@ static void submit_unlock_unpin_bo(struct msm_gem_submit *submit, struct msm_gem_object *msm_obj = submit->bos[i].obj; if (submit->bos[i].flags & BO_PINNED) - msm_gem_put_iova(&msm_obj->base, submit->gpu->aspace); + msm_gem_unpin_iova(&msm_obj->base, submit->gpu->aspace); if (submit->bos[i].flags & BO_LOCKED) ww_mutex_unlock(&msm_obj->resv->lock); @@ -270,7 +273,7 @@ static int submit_pin_objects(struct msm_gem_submit *submit) uint64_t iova; /* if locking succeeded, pin bo: */ - ret = msm_gem_get_iova(&msm_obj->base, + ret = msm_gem_get_and_pin_iova(&msm_obj->base, submit->gpu->aspace, &iova); if (ret) @@ -318,6 +321,9 @@ static int submit_reloc(struct msm_gem_submit *submit, struct msm_gem_object *ob uint32_t *ptr; int ret = 0; + if (!nr_relocs) + return 0; + if (offset % 4) { DRM_ERROR("non-aligned cmdstream buffer: %u\n", offset); return -EINVAL; @@ -406,19 +412,19 @@ static void submit_cleanup(struct msm_gem_submit *submit) int msm_ioctl_gem_submit(struct drm_device *dev, void *data, struct drm_file *file) { + static atomic_t ident = ATOMIC_INIT(0); struct msm_drm_private *priv = dev->dev_private; struct drm_msm_gem_submit *args = data; struct msm_file_private *ctx = file->driver_priv; struct msm_gem_submit *submit; struct msm_gpu *gpu = priv->gpu; - struct dma_fence *in_fence = NULL; struct sync_file *sync_file = NULL; struct msm_gpu_submitqueue *queue; struct msm_ringbuffer *ring; int out_fence_fd = -1; + struct pid *pid = get_pid(task_pid(current)); unsigned i; - int ret; - + int ret, submitid; if (!gpu) return -ENXIO; @@ -441,9 +447,16 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, if (!queue) return -ENOENT; + /* Get a unique identifier for the submission for logging purposes */ + submitid = atomic_inc_return(&ident) - 1; + ring = gpu->rb[queue->prio]; + trace_msm_gpu_submit(pid_nr(pid), ring->id, submitid, + args->nr_bos, args->nr_cmds); if (args->flags & MSM_SUBMIT_FENCE_FD_IN) { + struct dma_fence *in_fence; + in_fence = sync_file_get_fence(args->fence_fd); if (!in_fence) @@ -453,11 +466,13 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, * Wait if the fence is from a foreign context, or if the fence * array contains any fence from a foreign context. */ - if (!dma_fence_match_context(in_fence, ring->fctx->context)) { + ret = 0; + if (!dma_fence_match_context(in_fence, ring->fctx->context)) ret = dma_fence_wait(in_fence, true); - if (ret) - return ret; - } + + dma_fence_put(in_fence); + if (ret) + return ret; } ret = mutex_lock_interruptible(&dev->struct_mutex); @@ -478,6 +493,9 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, goto out_unlock; } + submit->pid = pid; + submit->ident = submitid; + if (args->flags & MSM_SUBMIT_SUDO) submit->in_rb = true; @@ -583,8 +601,6 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data, } out: - if (in_fence) - dma_fence_put(in_fence); submit_cleanup(submit); if (ret) msm_gem_submit_free(submit); diff --git a/drivers/gpu/drm/msm/msm_gem_vma.c b/drivers/gpu/drm/msm/msm_gem_vma.c index ffbec224551b..557360788084 100644 --- a/drivers/gpu/drm/msm/msm_gem_vma.c +++ b/drivers/gpu/drm/msm/msm_gem_vma.c @@ -38,20 +38,72 @@ void msm_gem_address_space_put(struct msm_gem_address_space *aspace) kref_put(&aspace->kref, msm_gem_address_space_destroy); } -void -msm_gem_unmap_vma(struct msm_gem_address_space *aspace, - struct msm_gem_vma *vma, struct sg_table *sgt) +/* Actually unmap memory for the vma */ +void msm_gem_purge_vma(struct msm_gem_address_space *aspace, + struct msm_gem_vma *vma) { - if (!aspace || !vma->iova) + unsigned size = vma->node.size << PAGE_SHIFT; + + /* Print a message if we try to purge a vma in use */ + if (WARN_ON(vma->inuse > 0)) + return; + + /* Don't do anything if the memory isn't mapped */ + if (!vma->mapped) return; - if (aspace->mmu) { - unsigned size = vma->node.size << PAGE_SHIFT; - aspace->mmu->funcs->unmap(aspace->mmu, vma->iova, sgt, size); - } + if (aspace->mmu) + aspace->mmu->funcs->unmap(aspace->mmu, vma->iova, size); + + vma->mapped = false; +} + +/* Remove reference counts for the mapping */ +void msm_gem_unmap_vma(struct msm_gem_address_space *aspace, + struct msm_gem_vma *vma) +{ + if (!WARN_ON(!vma->iova)) + vma->inuse--; +} + +int +msm_gem_map_vma(struct msm_gem_address_space *aspace, + struct msm_gem_vma *vma, struct sg_table *sgt, int npages) +{ + unsigned size = npages << PAGE_SHIFT; + int ret = 0; + + if (WARN_ON(!vma->iova)) + return -EINVAL; + + /* Increase the usage counter */ + vma->inuse++; + + if (vma->mapped) + return 0; + + vma->mapped = true; + + if (aspace->mmu) + ret = aspace->mmu->funcs->map(aspace->mmu, vma->iova, sgt, + size, IOMMU_READ | IOMMU_WRITE); + + if (ret) + vma->mapped = false; + + return ret; +} + +/* Close an iova. Warn if it is still in use */ +void msm_gem_close_vma(struct msm_gem_address_space *aspace, + struct msm_gem_vma *vma) +{ + if (WARN_ON(vma->inuse > 0 || vma->mapped)) + return; spin_lock(&aspace->lock); - drm_mm_remove_node(&vma->node); + if (vma->iova) + drm_mm_remove_node(&vma->node); spin_unlock(&aspace->lock); vma->iova = 0; @@ -59,18 +111,16 @@ msm_gem_unmap_vma(struct msm_gem_address_space *aspace, msm_gem_address_space_put(aspace); } -int -msm_gem_map_vma(struct msm_gem_address_space *aspace, - struct msm_gem_vma *vma, struct sg_table *sgt, int npages) +/* Initialize a new vma and allocate an iova for it */ +int msm_gem_init_vma(struct msm_gem_address_space *aspace, + struct msm_gem_vma *vma, int npages) { int ret; - spin_lock(&aspace->lock); - if (WARN_ON(drm_mm_node_allocated(&vma->node))) { - spin_unlock(&aspace->lock); - return 0; - } + if (WARN_ON(vma->iova)) + return -EBUSY; + spin_lock(&aspace->lock); ret = drm_mm_insert_node(&aspace->mm, &vma->node, npages); spin_unlock(&aspace->lock); @@ -78,19 +128,14 @@ msm_gem_map_vma(struct msm_gem_address_space *aspace, return ret; vma->iova = vma->node.start << PAGE_SHIFT; + vma->mapped = false; - if (aspace->mmu) { - unsigned size = npages << PAGE_SHIFT; - ret = aspace->mmu->funcs->map(aspace->mmu, vma->iova, sgt, - size, IOMMU_READ | IOMMU_WRITE); - } - - /* Get a reference to the aspace to keep it around */ kref_get(&aspace->kref); - return ret; + return 0; } + struct msm_gem_address_space * msm_gem_address_space_create(struct device *dev, struct iommu_domain *domain, const char *name) @@ -114,3 +159,26 @@ msm_gem_address_space_create(struct device *dev, struct iommu_domain *domain, return aspace; } + +struct msm_gem_address_space * +msm_gem_address_space_create_a2xx(struct device *dev, struct msm_gpu *gpu, + const char *name, uint64_t va_start, uint64_t va_end) +{ + struct msm_gem_address_space *aspace; + u64 size = va_end - va_start; + + aspace = kzalloc(sizeof(*aspace), GFP_KERNEL); + if (!aspace) + return ERR_PTR(-ENOMEM); + + spin_lock_init(&aspace->lock); + aspace->name = name; + aspace->mmu = msm_gpummu_new(dev, gpu); + + drm_mm_init(&aspace->mm, (va_start >> PAGE_SHIFT), + size >> PAGE_SHIFT); + + kref_init(&aspace->kref); + + return aspace; +} diff --git a/drivers/gpu/drm/msm/msm_gpu.c b/drivers/gpu/drm/msm/msm_gpu.c index 11aac8337066..5f3eff304355 100644 --- a/drivers/gpu/drm/msm/msm_gpu.c +++ b/drivers/gpu/drm/msm/msm_gpu.c @@ -19,6 +19,8 @@ #include "msm_gem.h" #include "msm_mmu.h" #include "msm_fence.h" +#include "msm_gpu_trace.h" +#include "adreno/adreno_gpu.h" #include <generated/utsrelease.h> #include <linux/string_helpers.h> @@ -107,7 +109,7 @@ static void msm_devfreq_init(struct msm_gpu *gpu) &msm_devfreq_profile, "simple_ondemand", NULL); if (IS_ERR(gpu->devfreq.devfreq)) { - dev_err(&gpu->pdev->dev, "Couldn't initialize GPU devfreq\n"); + DRM_DEV_ERROR(&gpu->pdev->dev, "Couldn't initialize GPU devfreq\n"); gpu->devfreq.devfreq = NULL; } @@ -122,7 +124,7 @@ static int enable_pwrrail(struct msm_gpu *gpu) if (gpu->gpu_reg) { ret = regulator_enable(gpu->gpu_reg); if (ret) { - dev_err(dev->dev, "failed to enable 'gpu_reg': %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to enable 'gpu_reg': %d\n", ret); return ret; } } @@ -130,7 +132,7 @@ static int enable_pwrrail(struct msm_gpu *gpu) if (gpu->gpu_cx) { ret = regulator_enable(gpu->gpu_cx); if (ret) { - dev_err(dev->dev, "failed to enable 'gpu_cx': %d\n", ret); + DRM_DEV_ERROR(dev->dev, "failed to enable 'gpu_cx': %d\n", ret); return ret; } } @@ -315,28 +317,28 @@ static void msm_gpu_crashstate_get_bo(struct msm_gpu_state *state, struct msm_gpu_state_bo *state_bo = &state->bos[state->nr_bos]; /* Don't record write only objects */ - state_bo->size = obj->base.size; state_bo->iova = iova; - /* Only store the data for buffer objects marked for read */ - if ((flags & MSM_SUBMIT_BO_READ)) { + /* Only store data for non imported buffer objects marked for read */ + if ((flags & MSM_SUBMIT_BO_READ) && !obj->base.import_attach) { void *ptr; state_bo->data = kvmalloc(obj->base.size, GFP_KERNEL); if (!state_bo->data) - return; + goto out; ptr = msm_gem_get_vaddr_active(&obj->base); if (IS_ERR(ptr)) { kvfree(state_bo->data); - return; + state_bo->data = NULL; + goto out; } memcpy(state_bo->data, ptr, obj->base.size); msm_gem_put_vaddr(&obj->base); } - +out: state->nr_bos++; } @@ -345,6 +347,10 @@ static void msm_gpu_crashstate_capture(struct msm_gpu *gpu, { struct msm_gpu_state *state; + /* Check if the target supports capturing crash state */ + if (!gpu->funcs->gpu_state_get) + return; + /* Only save one crash state at a time */ if (gpu->crashstate) return; @@ -360,12 +366,15 @@ static void msm_gpu_crashstate_capture(struct msm_gpu *gpu, if (submit) { int i; - state->bos = kcalloc(submit->nr_bos, + state->bos = kcalloc(submit->nr_cmds, sizeof(struct msm_gpu_state_bo), GFP_KERNEL); - for (i = 0; state->bos && i < submit->nr_bos; i++) - msm_gpu_crashstate_get_bo(state, submit->bos[i].obj, - submit->bos[i].iova, submit->bos[i].flags); + for (i = 0; state->bos && i < submit->nr_cmds; i++) { + int idx = submit->cmd[i].idx; + + msm_gpu_crashstate_get_bo(state, submit->bos[idx].obj, + submit->bos[idx].iova, submit->bos[idx].flags); + } } /* Set the active crash state to be dumped on failure */ @@ -428,16 +437,15 @@ static void recover_worker(struct work_struct *work) mutex_lock(&dev->struct_mutex); - dev_err(dev->dev, "%s: hangcheck recover!\n", gpu->name); + DRM_DEV_ERROR(dev->dev, "%s: hangcheck recover!\n", gpu->name); submit = find_submit(cur_ring, cur_ring->memptrs->fence + 1); if (submit) { struct task_struct *task; - rcu_read_lock(); - task = pid_task(submit->pid, PIDTYPE_PID); + task = get_pid_task(submit->pid, PIDTYPE_PID); if (task) { - comm = kstrdup(task->comm, GFP_ATOMIC); + comm = kstrdup(task->comm, GFP_KERNEL); /* * So slightly annoying, in other paths like @@ -450,13 +458,13 @@ static void recover_worker(struct work_struct *work) * about the submit going away. */ mutex_unlock(&dev->struct_mutex); - cmd = kstrdup_quotable_cmdline(task, GFP_ATOMIC); + cmd = kstrdup_quotable_cmdline(task, GFP_KERNEL); + put_task_struct(task); mutex_lock(&dev->struct_mutex); } - rcu_read_unlock(); if (comm && cmd) { - dev_err(dev->dev, "%s: offending task: %s (%s)\n", + DRM_DEV_ERROR(dev->dev, "%s: offending task: %s (%s)\n", gpu->name, comm, cmd); msm_rd_dump_submit(priv->hangrd, submit, @@ -539,11 +547,11 @@ static void hangcheck_handler(struct timer_list *t) } else if (fence < ring->seqno) { /* no progress and not done.. hung! */ ring->hangcheck_fence = fence; - dev_err(dev->dev, "%s: hangcheck detected gpu lockup rb %d!\n", + DRM_DEV_ERROR(dev->dev, "%s: hangcheck detected gpu lockup rb %d!\n", gpu->name, ring->id); - dev_err(dev->dev, "%s: completed fence: %u\n", + DRM_DEV_ERROR(dev->dev, "%s: completed fence: %u\n", gpu->name, fence); - dev_err(dev->dev, "%s: submitted fence: %u\n", + DRM_DEV_ERROR(dev->dev, "%s: submitted fence: %u\n", gpu->name, ring->seqno); queue_work(priv->wq, &gpu->recover_work); @@ -659,15 +667,33 @@ out: * Cmdstream submission/retirement: */ -static void retire_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit) +static void retire_submit(struct msm_gpu *gpu, struct msm_ringbuffer *ring, + struct msm_gem_submit *submit) { + int index = submit->seqno % MSM_GPU_SUBMIT_STATS_COUNT; + volatile struct msm_gpu_submit_stats *stats; + u64 elapsed, clock = 0; int i; + stats = &ring->memptrs->stats[index]; + /* Convert 19.2Mhz alwayson ticks to nanoseconds for elapsed time */ + elapsed = (stats->alwayson_end - stats->alwayson_start) * 10000; + do_div(elapsed, 192); + + /* Calculate the clock frequency from the number of CP cycles */ + if (elapsed) { + clock = (stats->cpcycles_end - stats->cpcycles_start) * 1000; + do_div(clock, elapsed); + } + + trace_msm_gpu_submit_retired(submit, elapsed, clock, + stats->alwayson_start, stats->alwayson_end); + for (i = 0; i < submit->nr_bos; i++) { struct msm_gem_object *msm_obj = submit->bos[i].obj; /* move to inactive: */ msm_gem_move_to_inactive(&msm_obj->base); - msm_gem_put_iova(&msm_obj->base, gpu->aspace); + msm_gem_unpin_iova(&msm_obj->base, gpu->aspace); drm_gem_object_put(&msm_obj->base); } @@ -690,7 +716,7 @@ static void retire_submits(struct msm_gpu *gpu) list_for_each_entry_safe(submit, tmp, &ring->submits, node) { if (dma_fence_is_signaled(submit->fence)) - retire_submit(gpu, submit); + retire_submit(gpu, ring, submit); } } } @@ -751,7 +777,7 @@ void msm_gpu_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit, /* submit takes a reference to the bo and iova until retired: */ drm_gem_object_get(&msm_obj->base); - msm_gem_get_iova(&msm_obj->base, + msm_gem_get_and_pin_iova(&msm_obj->base, submit->gpu->aspace, &iova); if (submit->bos[i].flags & MSM_SUBMIT_BO_WRITE) @@ -800,7 +826,6 @@ static struct msm_gem_address_space * msm_gpu_create_address_space(struct msm_gpu *gpu, struct platform_device *pdev, uint64_t va_start, uint64_t va_end) { - struct iommu_domain *iommu; struct msm_gem_address_space *aspace; int ret; @@ -809,20 +834,27 @@ msm_gpu_create_address_space(struct msm_gpu *gpu, struct platform_device *pdev, * and have separate page tables per context. For now, to keep things * simple and to get something working, just use a single address space: */ - iommu = iommu_domain_alloc(&platform_bus_type); - if (!iommu) - return NULL; - - iommu->geometry.aperture_start = va_start; - iommu->geometry.aperture_end = va_end; - - dev_info(gpu->dev->dev, "%s: using IOMMU\n", gpu->name); + if (!adreno_is_a2xx(to_adreno_gpu(gpu))) { + struct iommu_domain *iommu = iommu_domain_alloc(&platform_bus_type); + if (!iommu) + return NULL; + + iommu->geometry.aperture_start = va_start; + iommu->geometry.aperture_end = va_end; + + DRM_DEV_INFO(gpu->dev->dev, "%s: using IOMMU\n", gpu->name); + + aspace = msm_gem_address_space_create(&pdev->dev, iommu, "gpu"); + if (IS_ERR(aspace)) + iommu_domain_free(iommu); + } else { + aspace = msm_gem_address_space_create_a2xx(&pdev->dev, gpu, "gpu", + va_start, va_end); + } - aspace = msm_gem_address_space_create(&pdev->dev, iommu, "gpu"); if (IS_ERR(aspace)) { - dev_err(gpu->dev->dev, "failed to init iommu: %ld\n", + DRM_DEV_ERROR(gpu->dev->dev, "failed to init mmu: %ld\n", PTR_ERR(aspace)); - iommu_domain_free(iommu); return ERR_CAST(aspace); } @@ -871,14 +903,14 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev, gpu->irq = platform_get_irq_byname(pdev, config->irqname); if (gpu->irq < 0) { ret = gpu->irq; - dev_err(drm->dev, "failed to get irq: %d\n", ret); + DRM_DEV_ERROR(drm->dev, "failed to get irq: %d\n", ret); goto fail; } ret = devm_request_irq(&pdev->dev, gpu->irq, irq_handler, IRQF_TRIGGER_HIGH, gpu->name, gpu); if (ret) { - dev_err(drm->dev, "failed to request IRQ%u: %d\n", gpu->irq, ret); + DRM_DEV_ERROR(drm->dev, "failed to request IRQ%u: %d\n", gpu->irq, ret); goto fail; } @@ -911,22 +943,25 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev, config->va_start, config->va_end); if (gpu->aspace == NULL) - dev_info(drm->dev, "%s: no IOMMU, fallback to VRAM carveout!\n", name); + DRM_DEV_INFO(drm->dev, "%s: no IOMMU, fallback to VRAM carveout!\n", name); else if (IS_ERR(gpu->aspace)) { ret = PTR_ERR(gpu->aspace); goto fail; } - memptrs = msm_gem_kernel_new(drm, sizeof(*gpu->memptrs_bo), + memptrs = msm_gem_kernel_new(drm, + sizeof(struct msm_rbmemptrs) * nr_rings, MSM_BO_UNCACHED, gpu->aspace, &gpu->memptrs_bo, &memptrs_iova); if (IS_ERR(memptrs)) { ret = PTR_ERR(memptrs); - dev_err(drm->dev, "could not allocate memptrs: %d\n", ret); + DRM_DEV_ERROR(drm->dev, "could not allocate memptrs: %d\n", ret); goto fail; } + msm_gem_object_set_name(gpu->memptrs_bo, "memptrs"); + if (nr_rings > ARRAY_SIZE(gpu->rb)) { DRM_DEV_INFO_ONCE(drm->dev, "Only creating %zu ringbuffers\n", ARRAY_SIZE(gpu->rb)); @@ -939,7 +974,7 @@ int msm_gpu_init(struct drm_device *drm, struct platform_device *pdev, if (IS_ERR(gpu->rb[i])) { ret = PTR_ERR(gpu->rb[i]); - dev_err(drm->dev, + DRM_DEV_ERROR(drm->dev, "could not create ringbuffer %d: %d\n", i, ret); goto fail; } @@ -958,11 +993,7 @@ fail: gpu->rb[i] = NULL; } - if (gpu->memptrs_bo) { - msm_gem_put_vaddr(gpu->memptrs_bo); - msm_gem_put_iova(gpu->memptrs_bo, gpu->aspace); - drm_gem_object_put_unlocked(gpu->memptrs_bo); - } + msm_gem_kernel_put(gpu->memptrs_bo, gpu->aspace, false); platform_set_drvdata(pdev, NULL); return ret; @@ -981,11 +1012,7 @@ void msm_gpu_cleanup(struct msm_gpu *gpu) gpu->rb[i] = NULL; } - if (gpu->memptrs_bo) { - msm_gem_put_vaddr(gpu->memptrs_bo); - msm_gem_put_iova(gpu->memptrs_bo, gpu->aspace); - drm_gem_object_put_unlocked(gpu->memptrs_bo); - } + msm_gem_kernel_put(gpu->memptrs_bo, gpu->aspace, false); if (!IS_ERR_OR_NULL(gpu->aspace)) { gpu->aspace->mmu->funcs->detach(gpu->aspace->mmu, diff --git a/drivers/gpu/drm/msm/msm_gpu.h b/drivers/gpu/drm/msm/msm_gpu.h index f82bac086666..efb49bb64191 100644 --- a/drivers/gpu/drm/msm/msm_gpu.h +++ b/drivers/gpu/drm/msm/msm_gpu.h @@ -187,6 +187,7 @@ struct msm_gpu_state_bo { u64 iova; size_t size; void *data; + bool encoded; }; struct msm_gpu_state { @@ -201,6 +202,7 @@ struct msm_gpu_state { u32 wptr; void *data; int data_size; + bool encoded; } ring[MSM_GPU_MAX_RINGS]; int nr_registers; diff --git a/drivers/gpu/drm/msm/msm_gpu_trace.h b/drivers/gpu/drm/msm/msm_gpu_trace.h new file mode 100644 index 000000000000..1155118a27a1 --- /dev/null +++ b/drivers/gpu/drm/msm/msm_gpu_trace.h @@ -0,0 +1,90 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#if !defined(_MSM_GPU_TRACE_H_) || defined(TRACE_HEADER_MULTI_READ) +#define _MSM_GPU_TRACE_H_ + +#include <linux/tracepoint.h> + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM drm_msm +#define TRACE_INCLUDE_FILE msm_gpu_trace + +TRACE_EVENT(msm_gpu_submit, + TP_PROTO(pid_t pid, u32 ringid, u32 id, u32 nr_bos, u32 nr_cmds), + TP_ARGS(pid, ringid, id, nr_bos, nr_cmds), + TP_STRUCT__entry( + __field(pid_t, pid) + __field(u32, id) + __field(u32, ringid) + __field(u32, nr_cmds) + __field(u32, nr_bos) + ), + TP_fast_assign( + __entry->pid = pid; + __entry->id = id; + __entry->ringid = ringid; + __entry->nr_bos = nr_bos; + __entry->nr_cmds = nr_cmds + ), + TP_printk("id=%d pid=%d ring=%d bos=%d cmds=%d", + __entry->id, __entry->pid, __entry->ringid, + __entry->nr_bos, __entry->nr_cmds) +); + +TRACE_EVENT(msm_gpu_submit_flush, + TP_PROTO(struct msm_gem_submit *submit, u64 ticks), + TP_ARGS(submit, ticks), + TP_STRUCT__entry( + __field(pid_t, pid) + __field(u32, id) + __field(u32, ringid) + __field(u32, seqno) + __field(u64, ticks) + ), + TP_fast_assign( + __entry->pid = pid_nr(submit->pid); + __entry->id = submit->ident; + __entry->ringid = submit->ring->id; + __entry->seqno = submit->seqno; + __entry->ticks = ticks; + ), + TP_printk("id=%d pid=%d ring=%d:%d ticks=%lld", + __entry->id, __entry->pid, __entry->ringid, __entry->seqno, + __entry->ticks) +); + + +TRACE_EVENT(msm_gpu_submit_retired, + TP_PROTO(struct msm_gem_submit *submit, u64 elapsed, u64 clock, + u64 start, u64 end), + TP_ARGS(submit, elapsed, clock, start, end), + TP_STRUCT__entry( + __field(pid_t, pid) + __field(u32, id) + __field(u32, ringid) + __field(u32, seqno) + __field(u64, elapsed) + __field(u64, clock) + __field(u64, start_ticks) + __field(u64, end_ticks) + ), + TP_fast_assign( + __entry->pid = pid_nr(submit->pid); + __entry->id = submit->ident; + __entry->ringid = submit->ring->id; + __entry->seqno = submit->seqno; + __entry->elapsed = elapsed; + __entry->clock = clock; + __entry->start_ticks = start; + __entry->end_ticks = end; + ), + TP_printk("id=%d pid=%d ring=%d:%d elapsed=%lld ns mhz=%lld start=%lld end=%lld", + __entry->id, __entry->pid, __entry->ringid, __entry->seqno, + __entry->elapsed, __entry->clock, + __entry->start_ticks, __entry->end_ticks) +); + +#endif + +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH ../../drivers/gpu/drm/msm +#include <trace/define_trace.h> diff --git a/drivers/gpu/drm/msm/msm_gpu_tracepoints.c b/drivers/gpu/drm/msm/msm_gpu_tracepoints.c new file mode 100644 index 000000000000..72c074f8c4f8 --- /dev/null +++ b/drivers/gpu/drm/msm/msm_gpu_tracepoints.c @@ -0,0 +1,6 @@ +// SPDX-License-Identifier: GPL-2.0 +#include "msm_gem.h" +#include "msm_ringbuffer.h" + +#define CREATE_TRACE_POINTS +#include "msm_gpu_trace.h" diff --git a/drivers/gpu/drm/msm/msm_gpummu.c b/drivers/gpu/drm/msm/msm_gpummu.c new file mode 100644 index 000000000000..27312b553dd8 --- /dev/null +++ b/drivers/gpu/drm/msm/msm_gpummu.c @@ -0,0 +1,123 @@ +// SPDX-License-Identifier: GPL-2.0 +/* Copyright (c) 2018 The Linux Foundation. All rights reserved. */ + +#include "msm_drv.h" +#include "msm_mmu.h" +#include "adreno/adreno_gpu.h" +#include "adreno/a2xx.xml.h" + +struct msm_gpummu { + struct msm_mmu base; + struct msm_gpu *gpu; + dma_addr_t pt_base; + uint32_t *table; +}; +#define to_msm_gpummu(x) container_of(x, struct msm_gpummu, base) + +#define GPUMMU_VA_START SZ_16M +#define GPUMMU_VA_RANGE (0xfff * SZ_64K) +#define GPUMMU_PAGE_SIZE SZ_4K +#define TABLE_SIZE (sizeof(uint32_t) * GPUMMU_VA_RANGE / GPUMMU_PAGE_SIZE) + +static int msm_gpummu_attach(struct msm_mmu *mmu, const char * const *names, + int cnt) +{ + return 0; +} + +static void msm_gpummu_detach(struct msm_mmu *mmu, const char * const *names, + int cnt) +{ +} + +static int msm_gpummu_map(struct msm_mmu *mmu, uint64_t iova, + struct sg_table *sgt, unsigned len, int prot) +{ + struct msm_gpummu *gpummu = to_msm_gpummu(mmu); + unsigned idx = (iova - GPUMMU_VA_START) / GPUMMU_PAGE_SIZE; + struct scatterlist *sg; + unsigned prot_bits = 0; + unsigned i, j; + + if (prot & IOMMU_WRITE) + prot_bits |= 1; + if (prot & IOMMU_READ) + prot_bits |= 2; + + for_each_sg(sgt->sgl, sg, sgt->nents, i) { + dma_addr_t addr = sg->dma_address; + for (j = 0; j < sg->length / GPUMMU_PAGE_SIZE; j++, idx++) { + gpummu->table[idx] = addr | prot_bits; + addr += GPUMMU_PAGE_SIZE; + } + } + + /* we can improve by deferring flush for multiple map() */ + gpu_write(gpummu->gpu, REG_A2XX_MH_MMU_INVALIDATE, + A2XX_MH_MMU_INVALIDATE_INVALIDATE_ALL | + A2XX_MH_MMU_INVALIDATE_INVALIDATE_TC); + return 0; +} + +static int msm_gpummu_unmap(struct msm_mmu *mmu, uint64_t iova, unsigned len) +{ + struct msm_gpummu *gpummu = to_msm_gpummu(mmu); + unsigned idx = (iova - GPUMMU_VA_START) / GPUMMU_PAGE_SIZE; + unsigned i; + + for (i = 0; i < len / GPUMMU_PAGE_SIZE; i++, idx++) + gpummu->table[idx] = 0; + + gpu_write(gpummu->gpu, REG_A2XX_MH_MMU_INVALIDATE, + A2XX_MH_MMU_INVALIDATE_INVALIDATE_ALL | + A2XX_MH_MMU_INVALIDATE_INVALIDATE_TC); + return 0; +} + +static void msm_gpummu_destroy(struct msm_mmu *mmu) +{ + struct msm_gpummu *gpummu = to_msm_gpummu(mmu); + + dma_free_attrs(mmu->dev, TABLE_SIZE, gpummu->table, gpummu->pt_base, + DMA_ATTR_FORCE_CONTIGUOUS); + + kfree(gpummu); +} + +static const struct msm_mmu_funcs funcs = { + .attach = msm_gpummu_attach, + .detach = msm_gpummu_detach, + .map = msm_gpummu_map, + .unmap = msm_gpummu_unmap, + .destroy = msm_gpummu_destroy, +}; + +struct msm_mmu *msm_gpummu_new(struct device *dev, struct msm_gpu *gpu) +{ + struct msm_gpummu *gpummu; + + gpummu = kzalloc(sizeof(*gpummu), GFP_KERNEL); + if (!gpummu) + return ERR_PTR(-ENOMEM); + + gpummu->table = dma_alloc_attrs(dev, TABLE_SIZE + 32, &gpummu->pt_base, + GFP_KERNEL | __GFP_ZERO, DMA_ATTR_FORCE_CONTIGUOUS); + if (!gpummu->table) { + kfree(gpummu); + return ERR_PTR(-ENOMEM); + } + + gpummu->gpu = gpu; + msm_mmu_init(&gpummu->base, dev, &funcs); + + return &gpummu->base; +} + +void msm_gpummu_params(struct msm_mmu *mmu, dma_addr_t *pt_base, + dma_addr_t *tran_error) +{ + dma_addr_t base = to_msm_gpummu(mmu)->pt_base; + + *pt_base = base; + *tran_error = base + TABLE_SIZE; /* 32-byte aligned */ +} diff --git a/drivers/gpu/drm/msm/msm_iommu.c b/drivers/gpu/drm/msm/msm_iommu.c index b23d33622f37..4d62790cd425 100644 --- a/drivers/gpu/drm/msm/msm_iommu.c +++ b/drivers/gpu/drm/msm/msm_iommu.c @@ -66,13 +66,12 @@ static int msm_iommu_map(struct msm_mmu *mmu, uint64_t iova, // pm_runtime_get_sync(mmu->dev); ret = iommu_map_sg(iommu->domain, iova, sgt->sgl, sgt->nents, prot); // pm_runtime_put_sync(mmu->dev); - WARN_ON(ret < 0); + WARN_ON(!ret); return (ret == len) ? 0 : -EINVAL; } -static int msm_iommu_unmap(struct msm_mmu *mmu, uint64_t iova, - struct sg_table *sgt, unsigned len) +static int msm_iommu_unmap(struct msm_mmu *mmu, uint64_t iova, unsigned len) { struct msm_iommu *iommu = to_msm_iommu(mmu); diff --git a/drivers/gpu/drm/msm/msm_kms.h b/drivers/gpu/drm/msm/msm_kms.h index fd88cebb6adb..2b81b43a4bab 100644 --- a/drivers/gpu/drm/msm/msm_kms.h +++ b/drivers/gpu/drm/msm/msm_kms.h @@ -67,9 +67,6 @@ struct msm_kms_funcs { void (*set_encoder_mode)(struct msm_kms *kms, struct drm_encoder *encoder, bool cmd_mode); - /* pm suspend/resume hooks */ - int (*pm_suspend)(struct device *dev); - int (*pm_resume)(struct device *dev); /* cleanup: */ void (*destroy)(struct msm_kms *kms); #ifdef CONFIG_DEBUG_FS diff --git a/drivers/gpu/drm/msm/msm_mmu.h b/drivers/gpu/drm/msm/msm_mmu.h index aa2c5d4580c8..d21b26604d0b 100644 --- a/drivers/gpu/drm/msm/msm_mmu.h +++ b/drivers/gpu/drm/msm/msm_mmu.h @@ -25,8 +25,7 @@ struct msm_mmu_funcs { void (*detach)(struct msm_mmu *mmu, const char * const *names, int cnt); int (*map)(struct msm_mmu *mmu, uint64_t iova, struct sg_table *sgt, unsigned len, int prot); - int (*unmap)(struct msm_mmu *mmu, uint64_t iova, struct sg_table *sgt, - unsigned len); + int (*unmap)(struct msm_mmu *mmu, uint64_t iova, unsigned len); void (*destroy)(struct msm_mmu *mmu); }; @@ -54,4 +53,7 @@ static inline void msm_mmu_set_fault_handler(struct msm_mmu *mmu, void *arg, mmu->handler = handler; } +void msm_gpummu_params(struct msm_mmu *mmu, dma_addr_t *pt_base, + dma_addr_t *tran_error); + #endif /* __MSM_MMU_H__ */ diff --git a/drivers/gpu/drm/msm/msm_rd.c b/drivers/gpu/drm/msm/msm_rd.c index cca933458439..90e9d0a48dc0 100644 --- a/drivers/gpu/drm/msm/msm_rd.c +++ b/drivers/gpu/drm/msm/msm_rd.c @@ -316,10 +316,11 @@ static void snapshot_buf(struct msm_rd_state *rd, uint64_t iova, uint32_t size) { struct msm_gem_object *obj = submit->bos[idx].obj; + unsigned offset = 0; const char *buf; if (iova) { - buf += iova - submit->bos[idx].iova; + offset = iova - submit->bos[idx].iova; } else { iova = submit->bos[idx].iova; size = obj->base.size; @@ -340,11 +341,19 @@ static void snapshot_buf(struct msm_rd_state *rd, if (IS_ERR(buf)) return; + buf += offset; + rd_write_section(rd, RD_BUFFER_CONTENTS, buf, size); msm_gem_put_vaddr(&obj->base); } +static bool +should_dump(struct msm_gem_submit *submit, int idx) +{ + return rd_full || (submit->bos[idx].flags & MSM_SUBMIT_BO_DUMP); +} + /* called under struct_mutex */ void msm_rd_dump_submit(struct msm_rd_state *rd, struct msm_gem_submit *submit, const char *fmt, ...) @@ -386,15 +395,16 @@ void msm_rd_dump_submit(struct msm_rd_state *rd, struct msm_gem_submit *submit, rd_write_section(rd, RD_CMD, msg, ALIGN(n, 4)); - for (i = 0; rd_full && i < submit->nr_bos; i++) - snapshot_buf(rd, submit, i, 0, 0); + for (i = 0; i < submit->nr_bos; i++) + if (should_dump(submit, i)) + snapshot_buf(rd, submit, i, 0, 0); for (i = 0; i < submit->nr_cmds; i++) { uint64_t iova = submit->cmd[i].iova; uint32_t szd = submit->cmd[i].size; /* in dwords */ /* snapshot cmdstream bo's (if we haven't already): */ - if (!rd_full) { + if (!should_dump(submit, i)) { snapshot_buf(rd, submit, submit->cmd[i].idx, submit->cmd[i].iova, szd * 4); } diff --git a/drivers/gpu/drm/msm/msm_ringbuffer.c b/drivers/gpu/drm/msm/msm_ringbuffer.c index 6f5295b3f2f6..20a96fe69dcd 100644 --- a/drivers/gpu/drm/msm/msm_ringbuffer.c +++ b/drivers/gpu/drm/msm/msm_ringbuffer.c @@ -36,15 +36,18 @@ struct msm_ringbuffer *msm_ringbuffer_new(struct msm_gpu *gpu, int id, ring->gpu = gpu; ring->id = id; - /* Pass NULL for the iova pointer - we will map it later */ + ring->start = msm_gem_kernel_new(gpu->dev, MSM_GPU_RINGBUFFER_SZ, - MSM_BO_WC, gpu->aspace, &ring->bo, NULL); + MSM_BO_WC, gpu->aspace, &ring->bo, &ring->iova); if (IS_ERR(ring->start)) { ret = PTR_ERR(ring->start); ring->start = 0; goto fail; } + + msm_gem_object_set_name(ring->bo, "ring%d", id); + ring->end = ring->start + (MSM_GPU_RINGBUFFER_SZ >> 2); ring->next = ring->start; ring->cur = ring->start; @@ -73,10 +76,7 @@ void msm_ringbuffer_destroy(struct msm_ringbuffer *ring) msm_fence_context_free(ring->fctx); - if (ring->bo) { - msm_gem_put_iova(ring->bo, ring->gpu->aspace); - msm_gem_put_vaddr(ring->bo); - drm_gem_object_put_unlocked(ring->bo); - } + msm_gem_kernel_put(ring->bo, ring->gpu->aspace, false); + kfree(ring); } diff --git a/drivers/gpu/drm/msm/msm_ringbuffer.h b/drivers/gpu/drm/msm/msm_ringbuffer.h index cffce094aecb..6434ebb13136 100644 --- a/drivers/gpu/drm/msm/msm_ringbuffer.h +++ b/drivers/gpu/drm/msm/msm_ringbuffer.h @@ -23,9 +23,25 @@ #define rbmemptr(ring, member) \ ((ring)->memptrs_iova + offsetof(struct msm_rbmemptrs, member)) +#define rbmemptr_stats(ring, index, member) \ + (rbmemptr((ring), stats) + \ + ((index) * sizeof(struct msm_gpu_submit_stats)) + \ + offsetof(struct msm_gpu_submit_stats, member)) + +struct msm_gpu_submit_stats { + u64 cpcycles_start; + u64 cpcycles_end; + u64 alwayson_start; + u64 alwayson_end; +}; + +#define MSM_GPU_SUBMIT_STATS_COUNT 64 + struct msm_rbmemptrs { volatile uint32_t rptr; volatile uint32_t fence; + + volatile struct msm_gpu_submit_stats stats[MSM_GPU_SUBMIT_STATS_COUNT]; }; struct msm_ringbuffer { diff --git a/drivers/gpu/drm/nouveau/Kconfig b/drivers/gpu/drm/nouveau/Kconfig index 4b75ad40dd80..432c440223bb 100644 --- a/drivers/gpu/drm/nouveau/Kconfig +++ b/drivers/gpu/drm/nouveau/Kconfig @@ -4,7 +4,8 @@ config DRM_NOUVEAU select FW_LOADER select DRM_KMS_HELPER select DRM_TTM - select FB_BACKLIGHT if DRM_NOUVEAU_BACKLIGHT + select BACKLIGHT_CLASS_DEVICE if DRM_NOUVEAU_BACKLIGHT + select BACKLIGHT_LCD_SUPPORT if DRM_NOUVEAU_BACKLIGHT select ACPI_VIDEO if ACPI && X86 && BACKLIGHT_CLASS_DEVICE && INPUT select X86_PLATFORM_DEVICES if ACPI && X86 select ACPI_WMI if ACPI && X86 diff --git a/drivers/gpu/drm/nouveau/dispnv04/disp.c b/drivers/gpu/drm/nouveau/dispnv04/disp.c index 70dce544984e..1727d399833c 100644 --- a/drivers/gpu/drm/nouveau/dispnv04/disp.c +++ b/drivers/gpu/drm/nouveau/dispnv04/disp.c @@ -67,7 +67,7 @@ nv04_display_create(struct drm_device *dev) for (i = 0; i < dcb->entries; i++) { struct dcb_output *dcbent = &dcb->entry[i]; - connector = nouveau_connector_create(dev, dcbent->connector); + connector = nouveau_connector_create(dev, dcbent); if (IS_ERR(connector)) continue; diff --git a/drivers/gpu/drm/nouveau/dispnv50/Kbuild b/drivers/gpu/drm/nouveau/dispnv50/Kbuild index 849b0f45afb8..3d074aa31173 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/Kbuild +++ b/drivers/gpu/drm/nouveau/dispnv50/Kbuild @@ -7,6 +7,7 @@ nouveau-y += dispnv50/core827d.o nouveau-y += dispnv50/core907d.o nouveau-y += dispnv50/core917d.o nouveau-y += dispnv50/corec37d.o +nouveau-y += dispnv50/corec57d.o nouveau-y += dispnv50/dac507d.o nouveau-y += dispnv50/dac907d.o @@ -23,12 +24,14 @@ nouveau-y += dispnv50/head827d.o nouveau-y += dispnv50/head907d.o nouveau-y += dispnv50/head917d.o nouveau-y += dispnv50/headc37d.o +nouveau-y += dispnv50/headc57d.o nouveau-y += dispnv50/wimm.o nouveau-y += dispnv50/wimmc37b.o nouveau-y += dispnv50/wndw.o nouveau-y += dispnv50/wndwc37e.o +nouveau-y += dispnv50/wndwc57e.o nouveau-y += dispnv50/base.o nouveau-y += dispnv50/base507c.o diff --git a/drivers/gpu/drm/nouveau/dispnv50/atom.h b/drivers/gpu/drm/nouveau/dispnv50/atom.h index 908feb1fc60f..a194990d2b0d 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/atom.h +++ b/drivers/gpu/drm/nouveau/dispnv50/atom.h @@ -54,9 +54,10 @@ struct nv50_head_atom { u64 offset:40; u8 buffer:1; u8 mode:4; - u8 size:2; + u16 size:11; u8 range:2; u8 output_mode:2; + void (*load)(struct drm_color_lut *, int size, void __iomem *); } olut; struct { @@ -169,9 +170,11 @@ struct nv50_wndw_atom { u8 buffer:1; u8 enable:2; u8 mode:4; - u8 size:2; + u16 size:11; u8 range:2; u8 output_mode:2; + void (*load)(struct drm_color_lut *, int size, + void __iomem *); } i; } xlut; diff --git a/drivers/gpu/drm/nouveau/dispnv50/base907c.c b/drivers/gpu/drm/nouveau/dispnv50/base907c.c index a562fc94ce59..049ce6da321c 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/base907c.c +++ b/drivers/gpu/drm/nouveau/dispnv50/base907c.c @@ -80,6 +80,7 @@ base907c_ilut(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) { asyw->xlut.i.mode = 7; asyw->xlut.i.enable = 2; + asyw->xlut.i.load = head907d_olut_load; } const struct nv50_wndw_func diff --git a/drivers/gpu/drm/nouveau/dispnv50/core.c b/drivers/gpu/drm/nouveau/dispnv50/core.c index f3c49adb1bdb..c25e0ebe3c92 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/core.c +++ b/drivers/gpu/drm/nouveau/dispnv50/core.c @@ -42,6 +42,7 @@ nv50_core_new(struct nouveau_drm *drm, struct nv50_core **pcore) int version; int (*new)(struct nouveau_drm *, s32, struct nv50_core **); } cores[] = { + { TU104_DISP_CORE_CHANNEL_DMA, 0, corec57d_new }, { GV100_DISP_CORE_CHANNEL_DMA, 0, corec37d_new }, { GP102_DISP_CORE_CHANNEL_DMA, 0, core917d_new }, { GP100_DISP_CORE_CHANNEL_DMA, 0, core917d_new }, diff --git a/drivers/gpu/drm/nouveau/dispnv50/core.h b/drivers/gpu/drm/nouveau/dispnv50/core.h index 8470df9dd13d..df8336b593f7 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/core.h +++ b/drivers/gpu/drm/nouveau/dispnv50/core.h @@ -46,5 +46,9 @@ extern const struct nv50_outp_func sor907d; int core917d_new(struct nouveau_drm *, s32, struct nv50_core **); int corec37d_new(struct nouveau_drm *, s32, struct nv50_core **); +int corec37d_ntfy_wait_done(struct nouveau_bo *, u32, struct nvif_device *); +void corec37d_update(struct nv50_core *, u32 *, bool); extern const struct nv50_outp_func sorc37d; + +int corec57d_new(struct nouveau_drm *, s32, struct nv50_core **); #endif diff --git a/drivers/gpu/drm/nouveau/dispnv50/corec37d.c b/drivers/gpu/drm/nouveau/dispnv50/corec37d.c index b5c17c948918..7860774b65bc 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/corec37d.c +++ b/drivers/gpu/drm/nouveau/dispnv50/corec37d.c @@ -24,7 +24,7 @@ #include <nouveau_bo.h> -static void +void corec37d_update(struct nv50_core *core, u32 *interlock, bool ntfy) { u32 *push; @@ -71,7 +71,7 @@ corec37d_ntfy_init(struct nouveau_bo *bo, u32 offset) nouveau_bo_wr32(bo, offset / 4 + 3, 0x00000000); } -void +static void corec37d_init(struct nv50_core *core) { const u32 windows = 8; /*XXX*/ diff --git a/drivers/gpu/drm/nouveau/dispnv50/corec57d.c b/drivers/gpu/drm/nouveau/dispnv50/corec57d.c new file mode 100644 index 000000000000..b606d68cda10 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/corec57d.c @@ -0,0 +1,61 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "core.h" +#include "head.h" + +static void +corec57d_init(struct nv50_core *core) +{ + const u32 windows = 8; /*XXX*/ + u32 *push, i; + if ((push = evo_wait(&core->chan, 2 + 6 * windows + 2))) { + evo_mthd(push, 0x0208, 1); + evo_data(push, core->chan.sync.handle); + for (i = 0; i < windows; i++) { + evo_mthd(push, 0x1000 + (i * 0x080), 3); + evo_data(push, i >> 1); + evo_data(push, 0x0000000f); + evo_data(push, 0x00000000); + evo_mthd(push, 0x1010 + (i * 0x080), 1); + evo_data(push, 0x00117fff); + } + evo_mthd(push, 0x0200, 1); + evo_data(push, 0x00000001); + evo_kick(push, &core->chan); + } +} + +static const struct nv50_core_func +corec57d = { + .init = corec57d_init, + .ntfy_init = corec37d_ntfy_init, + .ntfy_wait_done = corec37d_ntfy_wait_done, + .update = corec37d_update, + .head = &headc57d, + .sor = &sorc37d, +}; + +int +corec57d_new(struct nouveau_drm *drm, s32 oclass, struct nv50_core **pcore) +{ + return core507d_new_(&corec57d, drm, oclass, pcore); +} diff --git a/drivers/gpu/drm/nouveau/dispnv50/curs.c b/drivers/gpu/drm/nouveau/dispnv50/curs.c index f592087338c4..cb6e4d2b1b45 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/curs.c +++ b/drivers/gpu/drm/nouveau/dispnv50/curs.c @@ -31,6 +31,7 @@ nv50_curs_new(struct nouveau_drm *drm, int head, struct nv50_wndw **pwndw) int version; int (*new)(struct nouveau_drm *, int, s32, struct nv50_wndw **); } curses[] = { + { TU104_DISP_CURSOR, 0, cursc37a_new }, { GV100_DISP_CURSOR, 0, cursc37a_new }, { GK104_DISP_CURSOR, 0, curs907a_new }, { GF110_DISP_CURSOR, 0, curs907a_new }, diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.c b/drivers/gpu/drm/nouveau/dispnv50/disp.c index 500bf3e0f6f2..67107f0b1299 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/disp.c +++ b/drivers/gpu/drm/nouveau/dispnv50/disp.c @@ -198,6 +198,22 @@ nv50_dmac_create(struct nvif_device *device, struct nvif_object *disp, /****************************************************************************** * EVO channel helpers *****************************************************************************/ +static void +evo_flush(struct nv50_dmac *dmac) +{ + /* Push buffer fetches are not coherent with BAR1, we need to ensure + * writes have been flushed right through to VRAM before writing PUT. + */ + if (dmac->push.type & NVIF_MEM_VRAM) { + struct nvif_device *device = dmac->base.device; + nvif_wr32(&device->object, 0x070000, 0x00000001); + nvif_msec(device, 2000, + if (!(nvif_rd32(&device->object, 0x070000) & 0x00000002)) + break; + ); + } +} + u32 * evo_wait(struct nv50_dmac *evoc, int nr) { @@ -208,6 +224,7 @@ evo_wait(struct nv50_dmac *evoc, int nr) mutex_lock(&dmac->lock); if (put + nr >= (PAGE_SIZE / 4) - 8) { dmac->ptr[put] = 0x20000000; + evo_flush(dmac); nvif_wr32(&dmac->base.user, 0x0000, 0x00000000); if (nvif_msec(device, 2000, @@ -230,17 +247,7 @@ evo_kick(u32 *push, struct nv50_dmac *evoc) { struct nv50_dmac *dmac = evoc; - /* Push buffer fetches are not coherent with BAR1, we need to ensure - * writes have been flushed right through to VRAM before writing PUT. - */ - if (dmac->push.type & NVIF_MEM_VRAM) { - struct nvif_device *device = dmac->base.device; - nvif_wr32(&device->object, 0x070000, 0x00000001); - nvif_msec(device, 2000, - if (!(nvif_rd32(&device->object, 0x070000) & 0x00000002)) - break; - ); - } + evo_flush(dmac); nvif_wr32(&dmac->base.user, 0x0000, (push - dmac->ptr) << 2); mutex_unlock(&dmac->lock); @@ -1288,8 +1295,16 @@ nv50_mstm_fini(struct nv50_mstm *mstm) static void nv50_mstm_init(struct nv50_mstm *mstm) { - if (mstm && mstm->mgr.mst_state) - drm_dp_mst_topology_mgr_resume(&mstm->mgr); + int ret; + + if (!mstm || !mstm->mgr.mst_state) + return; + + ret = drm_dp_mst_topology_mgr_resume(&mstm->mgr); + if (ret == -1) { + drm_dp_mst_topology_mgr_set_mst(&mstm->mgr, false); + drm_kms_helper_hotplug_event(mstm->mgr.dev); + } } static void @@ -1297,6 +1312,7 @@ nv50_mstm_del(struct nv50_mstm **pmstm) { struct nv50_mstm *mstm = *pmstm; if (mstm) { + drm_dp_mst_topology_mgr_destroy(&mstm->mgr); kfree(*pmstm); *pmstm = NULL; } @@ -2330,7 +2346,7 @@ nv50_display_create(struct drm_device *dev) /* create encoder/connector objects based on VBIOS DCB table */ for (i = 0, dcbe = &dcb->entry[0]; i < dcb->entries; i++, dcbe++) { - connector = nouveau_connector_create(dev, dcbe->connector); + connector = nouveau_connector_create(dev, dcbe); if (IS_ERR(connector)) continue; diff --git a/drivers/gpu/drm/nouveau/dispnv50/disp.h b/drivers/gpu/drm/nouveau/dispnv50/disp.h index e48c5eb35b49..2216c58620c2 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/disp.h +++ b/drivers/gpu/drm/nouveau/dispnv50/disp.h @@ -45,6 +45,8 @@ struct nv50_disp_interlock { void corec37d_ntfy_init(struct nouveau_bo *, u32); +void head907d_olut_load(struct drm_color_lut *, int size, void __iomem *); + struct nv50_chan { struct nvif_object user; struct nvif_device *device; diff --git a/drivers/gpu/drm/nouveau/dispnv50/head.c b/drivers/gpu/drm/nouveau/dispnv50/head.c index 4f57e5379796..ac97ebce5b35 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/head.c +++ b/drivers/gpu/drm/nouveau/dispnv50/head.c @@ -50,9 +50,9 @@ nv50_head_flush_set(struct nv50_head *head, struct nv50_head_atom *asyh) if (asyh->set.core ) head->func->core_set(head, asyh); if (asyh->set.olut ) { asyh->olut.offset = nv50_lut_load(&head->olut, - asyh->olut.mode <= 1, asyh->olut.buffer, - asyh->state.gamma_lut); + asyh->state.gamma_lut, + asyh->olut.load); head->func->olut_set(head, asyh); } if (asyh->set.curs ) head->func->curs_set(head, asyh); @@ -210,7 +210,7 @@ nv50_head_atomic_check_lut(struct nv50_head *head, } } - if (!olut) { + if (!olut && !head->func->olut_identity) { asyh->olut.handle = 0; return 0; } diff --git a/drivers/gpu/drm/nouveau/dispnv50/head.h b/drivers/gpu/drm/nouveau/dispnv50/head.h index 37b3248c6dae..d1c002f534d4 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/head.h +++ b/drivers/gpu/drm/nouveau/dispnv50/head.h @@ -21,6 +21,7 @@ struct nv50_head_func { void (*view)(struct nv50_head *, struct nv50_head_atom *); void (*mode)(struct nv50_head *, struct nv50_head_atom *); void (*olut)(struct nv50_head *, struct nv50_head_atom *); + bool olut_identity; void (*olut_set)(struct nv50_head *, struct nv50_head_atom *); void (*olut_clr)(struct nv50_head *); void (*core_calc)(struct nv50_head *, struct nv50_head_atom *); @@ -75,4 +76,14 @@ int head917d_curs_layout(struct nv50_head *, struct nv50_wndw_atom *, struct nv50_head_atom *); extern const struct nv50_head_func headc37d; +void headc37d_view(struct nv50_head *, struct nv50_head_atom *); +void headc37d_core_set(struct nv50_head *, struct nv50_head_atom *); +void headc37d_core_clr(struct nv50_head *); +int headc37d_curs_format(struct nv50_head *, struct nv50_wndw_atom *, + struct nv50_head_atom *); +void headc37d_curs_set(struct nv50_head *, struct nv50_head_atom *); +void headc37d_curs_clr(struct nv50_head *); +void headc37d_dither(struct nv50_head *, struct nv50_head_atom *); + +extern const struct nv50_head_func headc57d; #endif diff --git a/drivers/gpu/drm/nouveau/dispnv50/head507d.c b/drivers/gpu/drm/nouveau/dispnv50/head507d.c index 51bc5996fd37..7561be5ca707 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/head507d.c +++ b/drivers/gpu/drm/nouveau/dispnv50/head507d.c @@ -254,6 +254,23 @@ head507d_olut_set(struct nv50_head *head, struct nv50_head_atom *asyh) } } +static void +head507d_olut_load(struct drm_color_lut *in, int size, void __iomem *mem) +{ + for (; size--; in++, mem += 8) { + writew(drm_color_lut_extract(in-> red, 11) << 3, mem + 0); + writew(drm_color_lut_extract(in->green, 11) << 3, mem + 2); + writew(drm_color_lut_extract(in-> blue, 11) << 3, mem + 4); + } + + /* INTERPOLATE modes require a "next" entry to interpolate with, + * so we replicate the last entry to deal with this for now. + */ + writew(readw(mem - 8), mem + 0); + writew(readw(mem - 6), mem + 2); + writew(readw(mem - 4), mem + 4); +} + void head507d_olut(struct nv50_head *head, struct nv50_head_atom *asyh) { @@ -261,6 +278,8 @@ head507d_olut(struct nv50_head *head, struct nv50_head_atom *asyh) asyh->olut.mode = 0; else asyh->olut.mode = 1; + + asyh->olut.load = head507d_olut_load; } void diff --git a/drivers/gpu/drm/nouveau/dispnv50/head907d.c b/drivers/gpu/drm/nouveau/dispnv50/head907d.c index 633907163eb1..c2d09dd97b1f 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/head907d.c +++ b/drivers/gpu/drm/nouveau/dispnv50/head907d.c @@ -214,9 +214,27 @@ head907d_olut_set(struct nv50_head *head, struct nv50_head_atom *asyh) } void +head907d_olut_load(struct drm_color_lut *in, int size, void __iomem *mem) +{ + for (; size--; in++, mem += 8) { + writew(drm_color_lut_extract(in-> red, 14) + 0x6000, mem + 0); + writew(drm_color_lut_extract(in->green, 14) + 0x6000, mem + 2); + writew(drm_color_lut_extract(in-> blue, 14) + 0x6000, mem + 4); + } + + /* INTERPOLATE modes require a "next" entry to interpolate with, + * so we replicate the last entry to deal with this for now. + */ + writew(readw(mem - 8), mem + 0); + writew(readw(mem - 6), mem + 2); + writew(readw(mem - 4), mem + 4); +} + +void head907d_olut(struct nv50_head *head, struct nv50_head_atom *asyh) { asyh->olut.mode = 7; + asyh->olut.load = head907d_olut_load; } void diff --git a/drivers/gpu/drm/nouveau/dispnv50/headc37d.c b/drivers/gpu/drm/nouveau/dispnv50/headc37d.c index 989c14083066..ef6a99d95a9c 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/headc37d.c +++ b/drivers/gpu/drm/nouveau/dispnv50/headc37d.c @@ -65,7 +65,7 @@ headc37d_procamp(struct nv50_head *head, struct nv50_head_atom *asyh) } } -static void +void headc37d_dither(struct nv50_head *head, struct nv50_head_atom *asyh) { struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; @@ -79,7 +79,7 @@ headc37d_dither(struct nv50_head *head, struct nv50_head_atom *asyh) } } -static void +void headc37d_curs_clr(struct nv50_head *head) { struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; @@ -93,7 +93,7 @@ headc37d_curs_clr(struct nv50_head *head) } } -static void +void headc37d_curs_set(struct nv50_head *head, struct nv50_head_atom *asyh) { struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; @@ -112,7 +112,7 @@ headc37d_curs_set(struct nv50_head *head, struct nv50_head_atom *asyh) } } -static int +int headc37d_curs_format(struct nv50_head *head, struct nv50_wndw_atom *asyw, struct nv50_head_atom *asyh) { @@ -155,6 +155,7 @@ headc37d_olut(struct nv50_head *head, struct nv50_head_atom *asyh) asyh->olut.size = 0; asyh->olut.range = 0; asyh->olut.output_mode = 1; + asyh->olut.load = head907d_olut_load; } static void @@ -181,7 +182,7 @@ headc37d_mode(struct nv50_head *head, struct nv50_head_atom *asyh) } } -static void +void headc37d_view(struct nv50_head *head, struct nv50_head_atom *asyh) { struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; diff --git a/drivers/gpu/drm/nouveau/dispnv50/headc57d.c b/drivers/gpu/drm/nouveau/dispnv50/headc57d.c new file mode 100644 index 000000000000..32a7f9e85fb0 --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/headc57d.c @@ -0,0 +1,206 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "head.h" +#include "atom.h" +#include "core.h" + +static void +headc57d_or(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 2))) { + /*XXX: This is a dirty hack until OR depth handling is + * improved later for deep colour etc. + */ + switch (asyh->or.depth) { + case 6: asyh->or.depth = 5; break; + case 5: asyh->or.depth = 4; break; + case 2: asyh->or.depth = 1; break; + case 0: asyh->or.depth = 4; break; + default: + WARN_ON(1); + break; + } + + evo_mthd(push, 0x2004 + (head->base.index * 0x400), 1); + evo_data(push, 0xfc000001 | + asyh->or.depth << 4 | + asyh->or.nvsync << 3 | + asyh->or.nhsync << 2); + evo_kick(push, core); + } +} + +static void +headc57d_procamp(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x2000 + (head->base.index * 0x400), 1); +#if 0 + evo_data(push, 0x80000000 | + asyh->procamp.sat.sin << 16 | + asyh->procamp.sat.cos << 4); +#else + evo_data(push, 0); +#endif + evo_kick(push, core); + } +} + +void +headc57d_olut_clr(struct nv50_head *head) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 2))) { + evo_mthd(push, 0x2288 + (head->base.index * 0x400), 1); + evo_data(push, 0x00000000); + evo_kick(push, core); + } +} + +void +headc57d_olut_set(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + u32 *push; + if ((push = evo_wait(core, 4))) { + evo_mthd(push, 0x2280 + (head->base.index * 0x400), 4); + evo_data(push, asyh->olut.size << 8 | + asyh->olut.mode << 2 | + asyh->olut.output_mode); + evo_data(push, 0xffffffff); /* FP_NORM_SCALE. */ + evo_data(push, asyh->olut.handle); + evo_data(push, asyh->olut.offset >> 8); + evo_kick(push, core); + } +} + +static void +headc57d_olut_load_8(struct drm_color_lut *in, int size, void __iomem *mem) +{ + memset_io(mem, 0x00, 0x20); /* VSS header. */ + mem += 0x20; + + while (size--) { + u16 r = drm_color_lut_extract(in-> red + 0, 16); + u16 g = drm_color_lut_extract(in->green + 0, 16); + u16 b = drm_color_lut_extract(in-> blue + 0, 16); + u16 ri = 0, gi = 0, bi = 0, i; + + if (in++, size) { + ri = (drm_color_lut_extract(in-> red, 16) - r) / 4; + gi = (drm_color_lut_extract(in->green, 16) - g) / 4; + bi = (drm_color_lut_extract(in-> blue, 16) - b) / 4; + } + + for (i = 0; i < 4; i++, mem += 8) { + writew(r + ri * i, mem + 0); + writew(g + gi * i, mem + 2); + writew(b + bi * i, mem + 4); + } + } + + /* INTERPOLATE modes require a "next" entry to interpolate with, + * so we replicate the last entry to deal with this for now. + */ + writew(readw(mem - 8), mem + 0); + writew(readw(mem - 6), mem + 2); + writew(readw(mem - 4), mem + 4); +} + +static void +headc57d_olut_load(struct drm_color_lut *in, int size, void __iomem *mem) +{ + memset_io(mem, 0x00, 0x20); /* VSS header. */ + mem += 0x20; + + for (; size--; in++, mem += 0x08) { + writew(drm_color_lut_extract(in-> red, 16), mem + 0); + writew(drm_color_lut_extract(in->green, 16), mem + 2); + writew(drm_color_lut_extract(in-> blue, 16), mem + 4); + } + + /* INTERPOLATE modes require a "next" entry to interpolate with, + * so we replicate the last entry to deal with this for now. + */ + writew(readw(mem - 8), mem + 0); + writew(readw(mem - 6), mem + 2); + writew(readw(mem - 4), mem + 4); +} + +void +headc57d_olut(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + asyh->olut.mode = 2; /* DIRECT10 */ + asyh->olut.size = 4 /* VSS header. */ + 1024 + 1 /* Entries. */; + asyh->olut.output_mode = 1; /* INTERPOLATE_ENABLE. */ + if (asyh->state.gamma_lut && + asyh->state.gamma_lut->length / sizeof(struct drm_color_lut) == 256) + asyh->olut.load = headc57d_olut_load_8; + else + asyh->olut.load = headc57d_olut_load; +} + +static void +headc57d_mode(struct nv50_head *head, struct nv50_head_atom *asyh) +{ + struct nv50_dmac *core = &nv50_disp(head->base.base.dev)->core->chan; + struct nv50_head_mode *m = &asyh->mode; + u32 *push; + if ((push = evo_wait(core, 12))) { + evo_mthd(push, 0x2064 + (head->base.index * 0x400), 5); + evo_data(push, (m->v.active << 16) | m->h.active ); + evo_data(push, (m->v.synce << 16) | m->h.synce ); + evo_data(push, (m->v.blanke << 16) | m->h.blanke ); + evo_data(push, (m->v.blanks << 16) | m->h.blanks ); + evo_data(push, (m->v.blank2e << 16) | m->v.blank2s); + evo_mthd(push, 0x200c + (head->base.index * 0x400), 1); + evo_data(push, m->clock * 1000); + evo_mthd(push, 0x2028 + (head->base.index * 0x400), 1); + evo_data(push, m->clock * 1000); + /*XXX: HEAD_USAGE_BOUNDS, doesn't belong here. */ + evo_mthd(push, 0x2030 + (head->base.index * 0x400), 1); + evo_data(push, 0x00001014); + evo_kick(push, core); + } +} + +const struct nv50_head_func +headc57d = { + .view = headc37d_view, + .mode = headc57d_mode, + .olut = headc57d_olut, + .olut_identity = true, + .olut_set = headc57d_olut_set, + .olut_clr = headc57d_olut_clr, + .curs_layout = head917d_curs_layout, + .curs_format = headc37d_curs_format, + .curs_set = headc37d_curs_set, + .curs_clr = headc37d_curs_clr, + .dither = headc37d_dither, + .procamp = headc57d_procamp, + .or = headc57d_or, +}; diff --git a/drivers/gpu/drm/nouveau/dispnv50/lut.c b/drivers/gpu/drm/nouveau/dispnv50/lut.c index a6b96ae2a22f..994def4fd51a 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/lut.c +++ b/drivers/gpu/drm/nouveau/dispnv50/lut.c @@ -29,45 +29,29 @@ #include <nvif/class.h> u32 -nv50_lut_load(struct nv50_lut *lut, bool legacy, int buffer, - struct drm_property_blob *blob) +nv50_lut_load(struct nv50_lut *lut, int buffer, struct drm_property_blob *blob, + void (*load)(struct drm_color_lut *, int, void __iomem *)) { - struct drm_color_lut *in = (struct drm_color_lut *)blob->data; + struct drm_color_lut *in = blob ? blob->data : NULL; void __iomem *mem = lut->mem[buffer].object.map.ptr; - const int size = blob->length / sizeof(*in); - int bits, shift, i; - u16 zero, r, g, b; - u32 addr = lut->mem[buffer].addr; - - /* This can't happen.. But it shuts the compiler up. */ - if (WARN_ON(size != 256)) - return 0; + const u32 addr = lut->mem[buffer].addr; + int i; - if (legacy) { - bits = 11; - shift = 3; - zero = 0x0000; + if (!in) { + in = kvmalloc_array(1024, sizeof(*in), GFP_KERNEL); + if (!WARN_ON(!in)) { + for (i = 0; i < 1024; i++) { + in[i].red = + in[i].green = + in[i].blue = (i << 16) >> 10; + } + load(in, 1024, mem); + kvfree(in); + } } else { - bits = 14; - shift = 0; - zero = 0x6000; - } - - for (i = 0; i < size; i++) { - r = (drm_color_lut_extract(in[i]. red, bits) + zero) << shift; - g = (drm_color_lut_extract(in[i].green, bits) + zero) << shift; - b = (drm_color_lut_extract(in[i]. blue, bits) + zero) << shift; - writew(r, mem + (i * 0x08) + 0); - writew(g, mem + (i * 0x08) + 2); - writew(b, mem + (i * 0x08) + 4); + load(in, blob->length / sizeof(*in), mem); } - /* INTERPOLATE modes require a "next" entry to interpolate with, - * so we replicate the last entry to deal with this for now. - */ - writew(r, mem + (i * 0x08) + 0); - writew(g, mem + (i * 0x08) + 2); - writew(b, mem + (i * 0x08) + 4); return addr; } diff --git a/drivers/gpu/drm/nouveau/dispnv50/lut.h b/drivers/gpu/drm/nouveau/dispnv50/lut.h index 6d7b8352e4cb..b3b9040cfe9a 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/lut.h +++ b/drivers/gpu/drm/nouveau/dispnv50/lut.h @@ -2,6 +2,7 @@ #define __NV50_KMS_LUT_H__ #include <nvif/mem.h> struct drm_property_blob; +struct drm_color_lut; struct nv50_disp; struct nv50_lut { @@ -10,6 +11,6 @@ struct nv50_lut { int nv50_lut_init(struct nv50_disp *, struct nvif_mmu *, struct nv50_lut *); void nv50_lut_fini(struct nv50_lut *); -u32 nv50_lut_load(struct nv50_lut *, bool legacy, int buffer, - struct drm_property_blob *); +u32 nv50_lut_load(struct nv50_lut *, int buffer, struct drm_property_blob *, + void (*)(struct drm_color_lut *, int size, void __iomem *)); #endif diff --git a/drivers/gpu/drm/nouveau/dispnv50/wimm.c b/drivers/gpu/drm/nouveau/dispnv50/wimm.c index fc36e0696407..bc9eeaf212ae 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/wimm.c +++ b/drivers/gpu/drm/nouveau/dispnv50/wimm.c @@ -31,6 +31,7 @@ nv50_wimm_init(struct nouveau_drm *drm, struct nv50_wndw *wndw) int version; int (*init)(struct nouveau_drm *, s32, struct nv50_wndw *); } wimms[] = { + { TU104_DISP_WINDOW_IMM_CHANNEL_DMA, 0, wimmc37b_init }, { GV100_DISP_WINDOW_IMM_CHANNEL_DMA, 0, wimmc37b_init }, {} }; diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndw.c b/drivers/gpu/drm/nouveau/dispnv50/wndw.c index 2187922e8dc2..ba9eea2ff16b 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/wndw.c +++ b/drivers/gpu/drm/nouveau/dispnv50/wndw.c @@ -139,10 +139,8 @@ nv50_wndw_flush_set(struct nv50_wndw *wndw, u32 *interlock, if (asyw->set.xlut ) { if (asyw->ilut) { asyw->xlut.i.offset = - nv50_lut_load(&wndw->ilut, - asyw->xlut.i.mode <= 1, - asyw->xlut.i.buffer, - asyw->ilut); + nv50_lut_load(&wndw->ilut, asyw->xlut.i.buffer, + asyw->ilut, asyw->xlut.i.load); } wndw->func->xlut_set(wndw, asyw); } @@ -322,6 +320,11 @@ nv50_wndw_atomic_check_lut(struct nv50_wndw *wndw, asyh->wndw.olut &= ~BIT(wndw->id); } + if (!ilut && wndw->func->ilut_identity) { + static struct drm_property_blob dummy = {}; + ilut = &dummy; + } + /* Recalculate LUT state. */ memset(&asyw->xlut, 0x00, sizeof(asyw->xlut)); if ((asyw->ilut = wndw->func->ilut ? ilut : NULL)) { @@ -623,6 +626,7 @@ nv50_wndw_new(struct nouveau_drm *drm, enum drm_plane_type type, int index, int (*new)(struct nouveau_drm *, enum drm_plane_type, int, s32, struct nv50_wndw **); } wndws[] = { + { TU104_DISP_WINDOW_CHANNEL_DMA, 0, wndwc57e_new }, { GV100_DISP_WINDOW_CHANNEL_DMA, 0, wndwc37e_new }, {} }; diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndw.h b/drivers/gpu/drm/nouveau/dispnv50/wndw.h index b0b6428034b0..03f3d8dc235a 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/wndw.h +++ b/drivers/gpu/drm/nouveau/dispnv50/wndw.h @@ -65,6 +65,7 @@ struct nv50_wndw_func { int (*ntfy_wait_begun)(struct nouveau_bo *, u32 offset, struct nvif_device *); void (*ilut)(struct nv50_wndw *, struct nv50_wndw_atom *); + bool ilut_identity; bool olut_core; void (*xlut_set)(struct nv50_wndw *, struct nv50_wndw_atom *); void (*xlut_clr)(struct nv50_wndw *); @@ -90,6 +91,23 @@ extern const struct nv50_wimm_func curs507a; int wndwc37e_new(struct nouveau_drm *, enum drm_plane_type, int, s32, struct nv50_wndw **); +int wndwc37e_new_(const struct nv50_wndw_func *, struct nouveau_drm *, + enum drm_plane_type type, int index, s32 oclass, u32 heads, + struct nv50_wndw **); +int wndwc37e_acquire(struct nv50_wndw *, struct nv50_wndw_atom *, + struct nv50_head_atom *); +void wndwc37e_release(struct nv50_wndw *, struct nv50_wndw_atom *, + struct nv50_head_atom *); +void wndwc37e_sema_set(struct nv50_wndw *, struct nv50_wndw_atom *); +void wndwc37e_sema_clr(struct nv50_wndw *); +void wndwc37e_ntfy_set(struct nv50_wndw *, struct nv50_wndw_atom *); +void wndwc37e_ntfy_clr(struct nv50_wndw *); +void wndwc37e_image_set(struct nv50_wndw *, struct nv50_wndw_atom *); +void wndwc37e_image_clr(struct nv50_wndw *); +void wndwc37e_update(struct nv50_wndw *, u32 *); + +int wndwc57e_new(struct nouveau_drm *, enum drm_plane_type, int, s32, + struct nv50_wndw **); int nv50_wndw_new(struct nouveau_drm *, enum drm_plane_type, int index, struct nv50_wndw **); diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndwc37e.c b/drivers/gpu/drm/nouveau/dispnv50/wndwc37e.c index 44afb0f069a5..e52a85c83f7a 100644 --- a/drivers/gpu/drm/nouveau/dispnv50/wndwc37e.c +++ b/drivers/gpu/drm/nouveau/dispnv50/wndwc37e.c @@ -61,9 +61,10 @@ wndwc37e_ilut(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) asyw->xlut.i.size = 0; asyw->xlut.i.range = 0; asyw->xlut.i.output_mode = 1; + asyw->xlut.i.load = head907d_olut_load; } -static void +void wndwc37e_image_clr(struct nv50_wndw *wndw) { u32 *push; @@ -76,7 +77,7 @@ wndwc37e_image_clr(struct nv50_wndw *wndw) } } -static void +void wndwc37e_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) { u32 *push; @@ -117,7 +118,7 @@ wndwc37e_image_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) evo_kick(push, &wndw->wndw); } -static void +void wndwc37e_ntfy_clr(struct nv50_wndw *wndw) { u32 *push; @@ -128,7 +129,7 @@ wndwc37e_ntfy_clr(struct nv50_wndw *wndw) } } -static void +void wndwc37e_ntfy_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) { u32 *push; @@ -140,7 +141,7 @@ wndwc37e_ntfy_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) } } -static void +void wndwc37e_sema_clr(struct nv50_wndw *wndw) { u32 *push; @@ -151,7 +152,7 @@ wndwc37e_sema_clr(struct nv50_wndw *wndw) } } -static void +void wndwc37e_sema_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) { u32 *push; @@ -165,7 +166,7 @@ wndwc37e_sema_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) } } -static void +void wndwc37e_update(struct nv50_wndw *wndw, u32 *interlock) { u32 *push; @@ -183,13 +184,13 @@ wndwc37e_update(struct nv50_wndw *wndw, u32 *interlock) } } -static void +void wndwc37e_release(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw, struct nv50_head_atom *asyh) { } -static int +int wndwc37e_acquire(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw, struct nv50_head_atom *asyh) { @@ -236,7 +237,7 @@ wndwc37e = { .update = wndwc37e_update, }; -static int +int wndwc37e_new_(const struct nv50_wndw_func *func, struct nouveau_drm *drm, enum drm_plane_type type, int index, s32 oclass, u32 heads, struct nv50_wndw **pwndw) diff --git a/drivers/gpu/drm/nouveau/dispnv50/wndwc57e.c b/drivers/gpu/drm/nouveau/dispnv50/wndwc57e.c new file mode 100644 index 000000000000..ba89f1a5fcfa --- /dev/null +++ b/drivers/gpu/drm/nouveau/dispnv50/wndwc57e.c @@ -0,0 +1,133 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "wndw.h" +#include "atom.h" + +#include <drm/drm_atomic_helper.h> +#include <drm/drm_plane_helper.h> +#include <nouveau_bo.h> + +#include <nvif/clc37e.h> + +static void +wndwc57e_ilut_clr(struct nv50_wndw *wndw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 2))) { + evo_mthd(push, 0x0444, 1); + evo_data(push, 0x00000000); + evo_kick(push, &wndw->wndw); + } +} + +static void +wndwc57e_ilut_set(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u32 *push; + if ((push = evo_wait(&wndw->wndw, 4))) { + evo_mthd(push, 0x0440, 3); + evo_data(push, asyw->xlut.i.size << 8 | + asyw->xlut.i.mode << 2 | + asyw->xlut.i.output_mode); + evo_data(push, asyw->xlut.handle); + evo_data(push, asyw->xlut.i.offset >> 8); + evo_kick(push, &wndw->wndw); + } +} + +static u16 +fixedU0_16_FP16(u16 fixed) +{ + int sign = 0, exp = 0, man = 0; + if (fixed) { + while (--exp && !(fixed & 0x8000)) + fixed <<= 1; + man = ((fixed << 1) & 0xffc0) >> 6; + exp += 15; + } + return (sign << 15) | (exp << 10) | man; +} + +static void +wndwc57e_ilut_load(struct drm_color_lut *in, int size, void __iomem *mem) +{ + memset_io(mem, 0x00, 0x20); /* VSS header. */ + mem += 0x20; + + for (; size--; in++, mem += 0x08) { + u16 r = fixedU0_16_FP16(drm_color_lut_extract(in-> red, 16)); + u16 g = fixedU0_16_FP16(drm_color_lut_extract(in->green, 16)); + u16 b = fixedU0_16_FP16(drm_color_lut_extract(in-> blue, 16)); + writew(r, mem + 0); + writew(g, mem + 2); + writew(b, mem + 4); + } + + /* INTERPOLATE modes require a "next" entry to interpolate with, + * so we replicate the last entry to deal with this for now. + */ + writew(readw(mem - 8), mem + 0); + writew(readw(mem - 6), mem + 2); + writew(readw(mem - 4), mem + 4); +} + +static void +wndwc57e_ilut(struct nv50_wndw *wndw, struct nv50_wndw_atom *asyw) +{ + u16 size = asyw->ilut->length / sizeof(struct drm_color_lut); + if (size == 256) { + asyw->xlut.i.mode = 1; /* DIRECT8. */ + } else { + asyw->xlut.i.mode = 2; /* DIRECT10. */ + size = 1024; + } + asyw->xlut.i.size = 4 /* VSS header. */ + size + 1 /* Entries. */; + asyw->xlut.i.output_mode = 0; /* INTERPOLATE_DISABLE. */ + asyw->xlut.i.load = wndwc57e_ilut_load; +} + +static const struct nv50_wndw_func +wndwc57e = { + .acquire = wndwc37e_acquire, + .release = wndwc37e_release, + .sema_set = wndwc37e_sema_set, + .sema_clr = wndwc37e_sema_clr, + .ntfy_set = wndwc37e_ntfy_set, + .ntfy_clr = wndwc37e_ntfy_clr, + .ntfy_reset = corec37d_ntfy_init, + .ntfy_wait_begun = base507c_ntfy_wait_begun, + .ilut = wndwc57e_ilut, + .ilut_identity = true, + .xlut_set = wndwc57e_ilut_set, + .xlut_clr = wndwc57e_ilut_clr, + .image_set = wndwc37e_image_set, + .image_clr = wndwc37e_image_clr, + .update = wndwc37e_update, +}; + +int +wndwc57e_new(struct nouveau_drm *drm, enum drm_plane_type type, int index, + s32 oclass, struct nv50_wndw **pwndw) +{ + return wndwc37e_new_(&wndwc57e, drm, type, index, oclass, + BIT(index >> 1), pwndw); +} diff --git a/drivers/gpu/drm/nouveau/include/nvif/cl0080.h b/drivers/gpu/drm/nouveau/include/nvif/cl0080.h index 4f5233107f5f..4cbed0329367 100644 --- a/drivers/gpu/drm/nouveau/include/nvif/cl0080.h +++ b/drivers/gpu/drm/nouveau/include/nvif/cl0080.h @@ -32,6 +32,7 @@ struct nv_device_info_v0 { #define NV_DEVICE_INFO_V0_MAXWELL 0x09 #define NV_DEVICE_INFO_V0_PASCAL 0x0a #define NV_DEVICE_INFO_V0_VOLTA 0x0b +#define NV_DEVICE_INFO_V0_TURING 0x0c __u8 family; __u8 pad06[2]; __u64 ram_size; diff --git a/drivers/gpu/drm/nouveau/include/nvif/cla06f.h b/drivers/gpu/drm/nouveau/include/nvif/cla06f.h index fbfcffc5feb2..81401eb970ea 100644 --- a/drivers/gpu/drm/nouveau/include/nvif/cla06f.h +++ b/drivers/gpu/drm/nouveau/include/nvif/cla06f.h @@ -4,12 +4,13 @@ struct kepler_channel_gpfifo_a_v0 { __u8 version; - __u8 pad01[1]; + __u8 priv; __u16 chid; __u32 ilength; __u64 ioffset; __u64 runlist; __u64 vmm; + __u64 inst; }; #define NVA06F_V0_NTFY_NON_STALL_INTERRUPT 0x00 diff --git a/drivers/gpu/drm/nouveau/include/nvif/class.h b/drivers/gpu/drm/nouveau/include/nvif/class.h index 6db56bd7d67e..1d82cbf70cf4 100644 --- a/drivers/gpu/drm/nouveau/include/nvif/class.h +++ b/drivers/gpu/drm/nouveau/include/nvif/class.h @@ -68,7 +68,8 @@ #define KEPLER_CHANNEL_GPFIFO_B /* cla06f.h */ 0x0000a16f #define MAXWELL_CHANNEL_GPFIFO_A /* cla06f.h */ 0x0000b06f #define PASCAL_CHANNEL_GPFIFO_A /* cla06f.h */ 0x0000c06f -#define VOLTA_CHANNEL_GPFIFO_A /* cla06f.h */ 0x0000c36f +#define VOLTA_CHANNEL_GPFIFO_A /* clc36f.h */ 0x0000c36f +#define TURING_CHANNEL_GPFIFO_A /* clc36f.h */ 0x0000c46f #define NV50_DISP /* cl5070.h */ 0x00005070 #define G82_DISP /* cl5070.h */ 0x00008270 @@ -83,6 +84,7 @@ #define GP100_DISP /* cl5070.h */ 0x00009770 #define GP102_DISP /* cl5070.h */ 0x00009870 #define GV100_DISP /* cl5070.h */ 0x0000c370 +#define TU104_DISP /* cl5070.h */ 0x0000c570 #define NV31_MPEG 0x00003174 #define G82_MPEG 0x00008274 @@ -95,6 +97,7 @@ #define GF110_DISP_CURSOR /* cl507a.h */ 0x0000907a #define GK104_DISP_CURSOR /* cl507a.h */ 0x0000917a #define GV100_DISP_CURSOR /* cl507a.h */ 0x0000c37a +#define TU104_DISP_CURSOR /* cl507a.h */ 0x0000c57a #define NV50_DISP_OVERLAY /* cl507b.h */ 0x0000507b #define G82_DISP_OVERLAY /* cl507b.h */ 0x0000827b @@ -103,6 +106,7 @@ #define GK104_DISP_OVERLAY /* cl507b.h */ 0x0000917b #define GV100_DISP_WINDOW_IMM_CHANNEL_DMA /* clc37b.h */ 0x0000c37b +#define TU104_DISP_WINDOW_IMM_CHANNEL_DMA /* clc37b.h */ 0x0000c57b #define NV50_DISP_BASE_CHANNEL_DMA /* cl507c.h */ 0x0000507c #define G82_DISP_BASE_CHANNEL_DMA /* cl507c.h */ 0x0000827c @@ -125,6 +129,7 @@ #define GP100_DISP_CORE_CHANNEL_DMA /* cl507d.h */ 0x0000977d #define GP102_DISP_CORE_CHANNEL_DMA /* cl507d.h */ 0x0000987d #define GV100_DISP_CORE_CHANNEL_DMA /* cl507d.h */ 0x0000c37d +#define TU104_DISP_CORE_CHANNEL_DMA /* cl507d.h */ 0x0000c57d #define NV50_DISP_OVERLAY_CHANNEL_DMA /* cl507e.h */ 0x0000507e #define G82_DISP_OVERLAY_CHANNEL_DMA /* cl507e.h */ 0x0000827e @@ -134,6 +139,7 @@ #define GK104_DISP_OVERLAY_CONTROL_DMA /* cl507e.h */ 0x0000917e #define GV100_DISP_WINDOW_CHANNEL_DMA /* clc37e.h */ 0x0000c37e +#define TU104_DISP_WINDOW_CHANNEL_DMA /* clc37e.h */ 0x0000c57e #define NV50_TESLA 0x00005097 #define G82_TESLA 0x00008297 @@ -183,6 +189,7 @@ #define PASCAL_DMA_COPY_A 0x0000c0b5 #define PASCAL_DMA_COPY_B 0x0000c1b5 #define VOLTA_DMA_COPY_A 0x0000c3b5 +#define TURING_DMA_COPY_A 0x0000c5b5 #define FERMI_DECOMPRESS 0x000090b8 diff --git a/drivers/gpu/drm/nouveau/include/nvif/clc36f.h b/drivers/gpu/drm/nouveau/include/nvif/clc36f.h new file mode 100644 index 000000000000..6b14d7e3f6bb --- /dev/null +++ b/drivers/gpu/drm/nouveau/include/nvif/clc36f.h @@ -0,0 +1,19 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __NVIF_CLC36F_H__ +#define __NVIF_CLC36F_H__ + +struct volta_channel_gpfifo_a_v0 { + __u8 version; + __u8 priv; + __u16 chid; + __u32 ilength; + __u64 ioffset; + __u64 runlist; + __u64 vmm; + __u64 inst; + __u32 token; +}; + +#define NVC36F_V0_NTFY_NON_STALL_INTERRUPT 0x00 +#define NVC36F_V0_NTFY_KILLED 0x01 +#endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/core/device.h b/drivers/gpu/drm/nouveau/include/nvkm/core/device.h index d83d834b7452..72e4dc1f0236 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/core/device.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/core/device.h @@ -61,7 +61,11 @@ enum nvkm_devidx { NVKM_ENGINE_NVENC2, NVKM_ENGINE_NVENC_LAST = NVKM_ENGINE_NVENC2, - NVKM_ENGINE_NVDEC, + NVKM_ENGINE_NVDEC0, + NVKM_ENGINE_NVDEC1, + NVKM_ENGINE_NVDEC2, + NVKM_ENGINE_NVDEC_LAST = NVKM_ENGINE_NVDEC2, + NVKM_ENGINE_PM, NVKM_ENGINE_SEC, NVKM_ENGINE_SEC2, @@ -114,6 +118,7 @@ struct nvkm_device { GM100 = 0x110, GP100 = 0x130, GV100 = 0x140, + TU100 = 0x160, } card_type; u32 chipset; u8 chiprev; @@ -163,7 +168,7 @@ struct nvkm_device { struct nvkm_engine *msppp; struct nvkm_engine *msvld; struct nvkm_engine *nvenc[3]; - struct nvkm_nvdec *nvdec; + struct nvkm_nvdec *nvdec[3]; struct nvkm_pm *pm; struct nvkm_engine *sec; struct nvkm_sec2 *sec2; @@ -235,7 +240,7 @@ struct nvkm_device_chip { int (*msppp )(struct nvkm_device *, int idx, struct nvkm_engine **); int (*msvld )(struct nvkm_device *, int idx, struct nvkm_engine **); int (*nvenc[3])(struct nvkm_device *, int idx, struct nvkm_engine **); - int (*nvdec )(struct nvkm_device *, int idx, struct nvkm_nvdec **); + int (*nvdec[3])(struct nvkm_device *, int idx, struct nvkm_nvdec **); int (*pm )(struct nvkm_device *, int idx, struct nvkm_pm **); int (*sec )(struct nvkm_device *, int idx, struct nvkm_engine **); int (*sec2 )(struct nvkm_device *, int idx, struct nvkm_sec2 **); diff --git a/drivers/gpu/drm/nouveau/include/nvkm/core/memory.h b/drivers/gpu/drm/nouveau/include/nvkm/core/memory.h index 05f505de0075..f34c80310861 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/core/memory.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/core/memory.h @@ -29,6 +29,7 @@ struct nvkm_memory_func { void *(*dtor)(struct nvkm_memory *); enum nvkm_memory_target (*target)(struct nvkm_memory *); u8 (*page)(struct nvkm_memory *); + u64 (*bar2)(struct nvkm_memory *); u64 (*addr)(struct nvkm_memory *); u64 (*size)(struct nvkm_memory *); void (*boot)(struct nvkm_memory *, struct nvkm_vmm *); @@ -56,6 +57,7 @@ void nvkm_memory_tags_put(struct nvkm_memory *, struct nvkm_device *, #define nvkm_memory_target(p) (p)->func->target(p) #define nvkm_memory_page(p) (p)->func->page(p) +#define nvkm_memory_bar2(p) (p)->func->bar2(p) #define nvkm_memory_addr(p) (p)->func->addr(p) #define nvkm_memory_size(p) (p)->func->size(p) #define nvkm_memory_boot(p,v) (p)->func->boot((p),(v)) diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/ce.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/ce.h index fc295e1faa19..86abe76023c2 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/engine/ce.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/ce.h @@ -11,4 +11,5 @@ int gm200_ce_new(struct nvkm_device *, int, struct nvkm_engine **); int gp100_ce_new(struct nvkm_device *, int, struct nvkm_engine **); int gp102_ce_new(struct nvkm_device *, int, struct nvkm_engine **); int gv100_ce_new(struct nvkm_device *, int, struct nvkm_engine **); +int tu104_ce_new(struct nvkm_device *, int, struct nvkm_engine **); #endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h index ef7dc0844d26..5ca86e178bb9 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h @@ -36,4 +36,5 @@ int gm200_disp_new(struct nvkm_device *, int, struct nvkm_disp **); int gp100_disp_new(struct nvkm_device *, int, struct nvkm_disp **); int gp102_disp_new(struct nvkm_device *, int, struct nvkm_disp **); int gv100_disp_new(struct nvkm_device *, int, struct nvkm_disp **); +int tu104_disp_new(struct nvkm_device *, int, struct nvkm_disp **); #endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h index 7e39fbed2519..3b2b685778eb 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h @@ -74,4 +74,5 @@ int gm20b_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **); int gp100_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **); int gp10b_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **); int gv100_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **); +int tu104_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **); #endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bar.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bar.h index f6bd94c7e0f7..fd9d713b611c 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bar.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bar.h @@ -16,8 +16,10 @@ struct nvkm_bar { }; struct nvkm_vmm *nvkm_bar_bar1_vmm(struct nvkm_device *); +void nvkm_bar_bar1_reset(struct nvkm_device *); void nvkm_bar_bar2_init(struct nvkm_device *); void nvkm_bar_bar2_fini(struct nvkm_device *); +void nvkm_bar_bar2_reset(struct nvkm_device *); struct nvkm_vmm *nvkm_bar_bar2_vmm(struct nvkm_device *); void nvkm_bar_flush(struct nvkm_bar *); @@ -27,4 +29,5 @@ int gf100_bar_new(struct nvkm_device *, int, struct nvkm_bar **); int gk20a_bar_new(struct nvkm_device *, int, struct nvkm_bar **); int gm107_bar_new(struct nvkm_device *, int, struct nvkm_bar **); int gm20b_bar_new(struct nvkm_device *, int, struct nvkm_bar **); +int tu104_bar_new(struct nvkm_device *, int, struct nvkm_bar **); #endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/M0203.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/M0203.h index 703a5b524b96..425ccc47e3b7 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/M0203.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/M0203.h @@ -12,11 +12,14 @@ u32 nvbios_M0203Tp(struct nvkm_bios *, u8 *ver, u8 *hdr, u8 *cnt, u8 *len, struct nvbios_M0203T *); struct nvbios_M0203E { -#define M0203E_TYPE_DDR2 0x0 -#define M0203E_TYPE_DDR3 0x1 -#define M0203E_TYPE_GDDR3 0x2 -#define M0203E_TYPE_GDDR5 0x3 -#define M0203E_TYPE_SKIP 0xf +#define M0203E_TYPE_DDR2 0x0 +#define M0203E_TYPE_DDR3 0x1 +#define M0203E_TYPE_GDDR3 0x2 +#define M0203E_TYPE_GDDR5 0x3 +#define M0203E_TYPE_HBM2 0x6 +#define M0203E_TYPE_GDDR5X 0x8 +#define M0203E_TYPE_GDDR6 0x9 +#define M0203E_TYPE_SKIP 0xf u8 type; u8 strap; u8 group; diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/conn.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/conn.h index ed9e0a6a0011..8463b421d345 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/conn.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/conn.h @@ -20,6 +20,7 @@ enum dcb_connector_type { DCB_CONNECTOR_DMS59_DP0 = 0x64, DCB_CONNECTOR_DMS59_DP1 = 0x65, DCB_CONNECTOR_WFD = 0x70, + DCB_CONNECTOR_USB_C = 0x71, DCB_CONNECTOR_NONE = 0xff }; diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/devinit.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/devinit.h index 486e7635c29d..1b71812a790b 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/devinit.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/devinit.h @@ -31,4 +31,5 @@ int gf100_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **); int gm107_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **); int gm200_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **); int gv100_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **); +int tu104_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **); #endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/fault.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/fault.h index 5a77498fe6a0..127f48066026 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/fault.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/fault.h @@ -30,4 +30,5 @@ struct nvkm_fault_data { int gp100_fault_new(struct nvkm_device *, int, struct nvkm_fault **); int gv100_fault_new(struct nvkm_device *, int, struct nvkm_fault **); +int tu104_fault_new(struct nvkm_device *, int, struct nvkm_fault **); #endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/fb.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/fb.h index 96ccc624ee81..27298f8b7ead 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/fb.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/fb.h @@ -105,7 +105,10 @@ enum nvkm_ram_type { NVKM_RAM_TYPE_GDDR2, NVKM_RAM_TYPE_GDDR3, NVKM_RAM_TYPE_GDDR4, - NVKM_RAM_TYPE_GDDR5 + NVKM_RAM_TYPE_GDDR5, + NVKM_RAM_TYPE_GDDR5X, + NVKM_RAM_TYPE_GDDR6, + NVKM_RAM_TYPE_HBM2, }; struct nvkm_ram { diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/mc.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/mc.h index 61c93c86e2e2..b66dedd8abb6 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/mc.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/mc.h @@ -31,4 +31,5 @@ int gk104_mc_new(struct nvkm_device *, int, struct nvkm_mc **); int gk20a_mc_new(struct nvkm_device *, int, struct nvkm_mc **); int gp100_mc_new(struct nvkm_device *, int, struct nvkm_mc **); int gp10b_mc_new(struct nvkm_device *, int, struct nvkm_mc **); +int tu104_mc_new(struct nvkm_device *, int, struct nvkm_mc **); #endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/mmu.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/mmu.h index 688595545e21..0a0e064f22e5 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/mmu.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/mmu.h @@ -130,4 +130,5 @@ int gm20b_mmu_new(struct nvkm_device *, int, struct nvkm_mmu **); int gp100_mmu_new(struct nvkm_device *, int, struct nvkm_mmu **); int gp10b_mmu_new(struct nvkm_device *, int, struct nvkm_mmu **); int gv100_mmu_new(struct nvkm_device *, int, struct nvkm_mmu **); +int tu104_mmu_new(struct nvkm_device *, int, struct nvkm_mmu **); #endif diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/timer.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/timer.h index e9b0746826ca..3693ebf371b6 100644 --- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/timer.h +++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/timer.h @@ -28,6 +28,18 @@ struct nvkm_timer { u64 nvkm_timer_read(struct nvkm_timer *); void nvkm_timer_alarm(struct nvkm_timer *, u32 nsec, struct nvkm_alarm *); +struct nvkm_timer_wait { + struct nvkm_timer *tmr; + u64 limit; + u64 time0; + u64 time1; + int reads; +}; + +void nvkm_timer_wait_init(struct nvkm_device *, u64 nsec, + struct nvkm_timer_wait *); +s64 nvkm_timer_wait_test(struct nvkm_timer_wait *); + /* Delay based on GPU time (ie. PTIMER). * * Will return -ETIMEDOUT unless the loop was terminated with 'break', @@ -38,21 +50,17 @@ void nvkm_timer_alarm(struct nvkm_timer *, u32 nsec, struct nvkm_alarm *); */ #define NVKM_DELAY _warn = false; #define nvkm_nsec(d,n,cond...) ({ \ - struct nvkm_device *_device = (d); \ - struct nvkm_timer *_tmr = _device->timer; \ - u64 _nsecs = (n), _time0 = nvkm_timer_read(_tmr); \ - s64 _taken = 0; \ + struct nvkm_timer_wait _wait; \ bool _warn = true; \ + s64 _taken = 0; \ \ + nvkm_timer_wait_init((d), (n), &_wait); \ do { \ cond \ - } while (_taken = nvkm_timer_read(_tmr) - _time0, _taken < _nsecs); \ + } while ((_taken = nvkm_timer_wait_test(&_wait)) >= 0); \ \ - if (_taken >= _nsecs) { \ - if (_warn) \ - dev_WARN(_device->dev, "timeout\n"); \ - _taken = -ETIMEDOUT; \ - } \ + if (_warn && _taken < 0) \ + dev_WARN(_wait.tmr->subdev.device->dev, "timeout\n"); \ _taken; \ }) #define nvkm_usec(d,u,cond...) nvkm_nsec((d), (u) * 1000, ##cond) diff --git a/drivers/gpu/drm/nouveau/nouveau_abi16.c b/drivers/gpu/drm/nouveau/nouveau_abi16.c index e67a471331b5..b06cdac8f3a2 100644 --- a/drivers/gpu/drm/nouveau/nouveau_abi16.c +++ b/drivers/gpu/drm/nouveau/nouveau_abi16.c @@ -306,7 +306,7 @@ nouveau_abi16_ioctl_channel_alloc(ABI16_IOCTL_ARGS) /* create channel object and initialise dma and fence management */ ret = nouveau_channel_new(drm, device, init->fb_ctxdma_handle, - init->tt_ctxdma_handle, &chan->chan); + init->tt_ctxdma_handle, false, &chan->chan); if (ret) goto done; diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c index 7214022dfb91..73eff52036d2 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.c +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c @@ -1141,6 +1141,8 @@ nouveau_bo_move_init(struct nouveau_drm *drm) struct ttm_mem_reg *, struct ttm_mem_reg *); int (*init)(struct nouveau_channel *, u32 handle); } _methods[] = { + { "COPY", 4, 0xc5b5, nve0_bo_move_copy, nve0_bo_move_init }, + { "GRCE", 0, 0xc5b5, nve0_bo_move_copy, nvc0_bo_move_init }, { "COPY", 4, 0xc3b5, nve0_bo_move_copy, nve0_bo_move_init }, { "GRCE", 0, 0xc3b5, nve0_bo_move_copy, nvc0_bo_move_init }, { "COPY", 4, 0xc1b5, nve0_bo_move_copy, nve0_bo_move_init }, diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.c b/drivers/gpu/drm/nouveau/nouveau_chan.c index 92d3115f96b5..668afbc29c3e 100644 --- a/drivers/gpu/drm/nouveau/nouveau_chan.c +++ b/drivers/gpu/drm/nouveau/nouveau_chan.c @@ -29,6 +29,7 @@ #include <nvif/cl506f.h> #include <nvif/cl906f.h> #include <nvif/cla06f.h> +#include <nvif/clc36f.h> #include <nvif/ioctl.h> /*XXX*/ @@ -217,10 +218,11 @@ nouveau_channel_prep(struct nouveau_drm *drm, struct nvif_device *device, static int nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device, - u64 runlist, struct nouveau_channel **pchan) + u64 runlist, bool priv, struct nouveau_channel **pchan) { struct nouveau_cli *cli = (void *)device->object.client; - static const u16 oclasses[] = { VOLTA_CHANNEL_GPFIFO_A, + static const u16 oclasses[] = { TURING_CHANNEL_GPFIFO_A, + VOLTA_CHANNEL_GPFIFO_A, PASCAL_CHANNEL_GPFIFO_A, MAXWELL_CHANNEL_GPFIFO_A, KEPLER_CHANNEL_GPFIFO_B, @@ -234,6 +236,7 @@ nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device, struct nv50_channel_gpfifo_v0 nv50; struct fermi_channel_gpfifo_v0 fermi; struct kepler_channel_gpfifo_a_v0 kepler; + struct volta_channel_gpfifo_a_v0 volta; } args; struct nouveau_channel *chan; u32 size; @@ -247,12 +250,22 @@ nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device, /* create channel object */ do { + if (oclass[0] >= VOLTA_CHANNEL_GPFIFO_A) { + args.volta.version = 0; + args.volta.ilength = 0x02000; + args.volta.ioffset = 0x10000 + chan->push.addr; + args.volta.runlist = runlist; + args.volta.vmm = nvif_handle(&cli->vmm.vmm.object); + args.volta.priv = priv; + size = sizeof(args.volta); + } else if (oclass[0] >= KEPLER_CHANNEL_GPFIFO_A) { args.kepler.version = 0; args.kepler.ilength = 0x02000; args.kepler.ioffset = 0x10000 + chan->push.addr; args.kepler.runlist = runlist; args.kepler.vmm = nvif_handle(&cli->vmm.vmm.object); + args.kepler.priv = priv; size = sizeof(args.kepler); } else if (oclass[0] >= FERMI_CHANNEL_GPFIFO) { @@ -273,13 +286,20 @@ nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device, ret = nvif_object_init(&device->object, 0, *oclass++, &args, size, &chan->user); if (ret == 0) { - if (chan->user.oclass >= KEPLER_CHANNEL_GPFIFO_A) + if (chan->user.oclass >= VOLTA_CHANNEL_GPFIFO_A) { + chan->chid = args.volta.chid; + chan->inst = args.volta.inst; + chan->token = args.volta.token; + } else + if (chan->user.oclass >= KEPLER_CHANNEL_GPFIFO_A) { chan->chid = args.kepler.chid; - else - if (chan->user.oclass >= FERMI_CHANNEL_GPFIFO) + chan->inst = args.kepler.inst; + } else + if (chan->user.oclass >= FERMI_CHANNEL_GPFIFO) { chan->chid = args.fermi.chid; - else + } else { chan->chid = args.nv50.chid; + } return ret; } } while (*oclass); @@ -448,7 +468,8 @@ nouveau_channel_init(struct nouveau_channel *chan, u32 vram, u32 gart) int nouveau_channel_new(struct nouveau_drm *drm, struct nvif_device *device, - u32 arg0, u32 arg1, struct nouveau_channel **pchan) + u32 arg0, u32 arg1, bool priv, + struct nouveau_channel **pchan) { struct nouveau_cli *cli = (void *)device->object.client; bool super; @@ -458,7 +479,7 @@ nouveau_channel_new(struct nouveau_drm *drm, struct nvif_device *device, super = cli->base.super; cli->base.super = true; - ret = nouveau_channel_ind(drm, device, arg0, pchan); + ret = nouveau_channel_ind(drm, device, arg0, priv, pchan); if (ret) { NV_PRINTK(dbg, cli, "ib channel create, %d\n", ret); ret = nouveau_channel_dma(drm, device, pchan); diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.h b/drivers/gpu/drm/nouveau/nouveau_chan.h index 64454c2ebd90..28418f4e5748 100644 --- a/drivers/gpu/drm/nouveau/nouveau_chan.h +++ b/drivers/gpu/drm/nouveau/nouveau_chan.h @@ -10,6 +10,8 @@ struct nouveau_channel { struct nouveau_drm *drm; int chid; + u64 inst; + u32 token; struct nvif_object vram; struct nvif_object gart; @@ -48,7 +50,8 @@ struct nouveau_channel { int nouveau_channels_init(struct nouveau_drm *); int nouveau_channel_new(struct nouveau_drm *, struct nvif_device *, - u32 arg0, u32 arg1, struct nouveau_channel **); + u32 arg0, u32 arg1, bool priv, + struct nouveau_channel **); void nouveau_channel_del(struct nouveau_channel **); int nouveau_channel_idle(struct nouveau_channel *); diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c index fd80661dff92..3f463c91314a 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.c +++ b/drivers/gpu/drm/nouveau/nouveau_connector.c @@ -403,6 +403,7 @@ nouveau_connector_destroy(struct drm_connector *connector) if (nv_connector->aux.transfer) { drm_dp_cec_unregister_connector(&nv_connector->aux); drm_dp_aux_unregister(&nv_connector->aux); + kfree(nv_connector->aux.name); } kfree(connector); } @@ -1218,7 +1219,8 @@ drm_conntype_from_dcb(enum dcb_connector_type dcb) case DCB_CONNECTOR_LVDS_SPWG: return DRM_MODE_CONNECTOR_LVDS; case DCB_CONNECTOR_DMS59_DP0: case DCB_CONNECTOR_DMS59_DP1: - case DCB_CONNECTOR_DP : return DRM_MODE_CONNECTOR_DisplayPort; + case DCB_CONNECTOR_DP : + case DCB_CONNECTOR_USB_C : return DRM_MODE_CONNECTOR_DisplayPort; case DCB_CONNECTOR_eDP : return DRM_MODE_CONNECTOR_eDP; case DCB_CONNECTOR_HDMI_0 : case DCB_CONNECTOR_HDMI_1 : @@ -1232,7 +1234,8 @@ drm_conntype_from_dcb(enum dcb_connector_type dcb) } struct drm_connector * -nouveau_connector_create(struct drm_device *dev, int index) +nouveau_connector_create(struct drm_device *dev, + const struct dcb_output *dcbe) { const struct drm_connector_funcs *funcs = &nouveau_connector_funcs; struct nouveau_drm *drm = nouveau_drm(dev); @@ -1240,6 +1243,8 @@ nouveau_connector_create(struct drm_device *dev, int index) struct nouveau_connector *nv_connector = NULL; struct drm_connector *connector; struct drm_connector_list_iter conn_iter; + char aux_name[48] = {0}; + int index = dcbe->connector; int type, ret = 0; bool dummy; @@ -1342,6 +1347,9 @@ nouveau_connector_create(struct drm_device *dev, int index) case DRM_MODE_CONNECTOR_eDP: nv_connector->aux.dev = dev->dev; nv_connector->aux.transfer = nouveau_connector_aux_xfer; + snprintf(aux_name, sizeof(aux_name), "sor-%04x-%04x", + dcbe->hasht, dcbe->hashm); + nv_connector->aux.name = kstrdup(aux_name, GFP_KERNEL); ret = drm_dp_aux_register(&nv_connector->aux); if (ret) { NV_ERROR(drm, "failed to register aux channel\n"); diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.h b/drivers/gpu/drm/nouveau/nouveau_connector.h index f57ef35b1e5e..f43a8d63aef8 100644 --- a/drivers/gpu/drm/nouveau/nouveau_connector.h +++ b/drivers/gpu/drm/nouveau/nouveau_connector.h @@ -38,6 +38,7 @@ #include "nouveau_encoder.h" struct nvkm_i2c_port; +struct dcb_output; #ifdef CONFIG_DRM_NOUVEAU_BACKLIGHT struct nouveau_backlight; @@ -113,7 +114,7 @@ nouveau_crtc_connector_get(struct nouveau_crtc *nv_crtc) } struct drm_connector * -nouveau_connector_create(struct drm_device *, int index); +nouveau_connector_create(struct drm_device *, const struct dcb_output *); extern int nouveau_tv_disable; extern int nouveau_ignorelid; diff --git a/drivers/gpu/drm/nouveau/nouveau_debugfs.c b/drivers/gpu/drm/nouveau/nouveau_debugfs.c index 9109b69cd052..88a52f6b39fe 100644 --- a/drivers/gpu/drm/nouveau/nouveau_debugfs.c +++ b/drivers/gpu/drm/nouveau/nouveau_debugfs.c @@ -47,6 +47,26 @@ nouveau_debugfs_vbios_image(struct seq_file *m, void *data) } static int +nouveau_debugfs_strap_peek(struct seq_file *m, void *data) +{ + struct drm_info_node *node = m->private; + struct nouveau_drm *drm = nouveau_drm(node->minor->dev); + int ret; + + ret = pm_runtime_get_sync(drm->dev->dev); + if (ret < 0 && ret != -EACCES) + return ret; + + seq_printf(m, "0x%08x\n", + nvif_rd32(&drm->client.device.object, 0x101000)); + + pm_runtime_mark_last_busy(drm->dev->dev); + pm_runtime_put_autosuspend(drm->dev->dev); + + return 0; +} + +static int nouveau_debugfs_pstate_get(struct seq_file *m, void *data) { struct drm_device *drm = m->private; @@ -185,7 +205,8 @@ static const struct file_operations nouveau_pstate_fops = { }; static struct drm_info_list nouveau_debugfs_list[] = { - { "vbios.rom", nouveau_debugfs_vbios_image, 0, NULL }, + { "vbios.rom", nouveau_debugfs_vbios_image, 0, NULL }, + { "strap_peek", nouveau_debugfs_strap_peek, 0, NULL }, }; #define NOUVEAU_DEBUGFS_ENTRIES ARRAY_SIZE(nouveau_debugfs_list) @@ -199,8 +220,9 @@ static const struct nouveau_debugfs_files { int nouveau_drm_debugfs_init(struct drm_minor *minor) { + struct nouveau_drm *drm = nouveau_drm(minor->dev); struct dentry *dentry; - int i; + int i, ret; for (i = 0; i < ARRAY_SIZE(nouveau_debugfs_files); i++) { dentry = debugfs_create_file(nouveau_debugfs_files[i].name, @@ -211,9 +233,23 @@ nouveau_drm_debugfs_init(struct drm_minor *minor) return -ENOMEM; } - return drm_debugfs_create_files(nouveau_debugfs_list, - NOUVEAU_DEBUGFS_ENTRIES, - minor->debugfs_root, minor); + ret = drm_debugfs_create_files(nouveau_debugfs_list, + NOUVEAU_DEBUGFS_ENTRIES, + minor->debugfs_root, minor); + if (ret) + return ret; + + /* Set the size of the vbios since we know it, and it's confusing to + * userspace if it wants to seek() but the file has a length of 0 + */ + dentry = debugfs_lookup("vbios.rom", minor->debugfs_root); + if (!dentry) + return 0; + + d_inode(dentry)->i_size = drm->vbios.length; + dput(dentry); + + return 0; } int diff --git a/drivers/gpu/drm/nouveau/nouveau_dma.c b/drivers/gpu/drm/nouveau/nouveau_dma.c index 945afd34138e..078f65d849ce 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dma.c +++ b/drivers/gpu/drm/nouveau/nouveau_dma.c @@ -101,7 +101,7 @@ nv50_dma_push(struct nouveau_channel *chan, u64 offset, int length) nvif_wr32(&chan->user, 0x8c, chan->dma.ib_put); if (user->func && user->func->doorbell) - user->func->doorbell(user, chan->chid); + user->func->doorbell(user, chan->token); chan->dma.ib_free--; } diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c index 2b2baf6e0e0d..f900e94592f8 100644 --- a/drivers/gpu/drm/nouveau/nouveau_drm.c +++ b/drivers/gpu/drm/nouveau/nouveau_drm.c @@ -353,6 +353,7 @@ nouveau_accel_init(struct nouveau_drm *drm) case MAXWELL_CHANNEL_GPFIFO_A: case PASCAL_CHANNEL_GPFIFO_A: case VOLTA_CHANNEL_GPFIFO_A: + case TURING_CHANNEL_GPFIFO_A: ret = nvc0_fence_create(drm); break; default: @@ -370,7 +371,7 @@ nouveau_accel_init(struct nouveau_drm *drm) if (device->info.family >= NV_DEVICE_INFO_V0_KEPLER) { ret = nouveau_channel_new(drm, &drm->client.device, nvif_fifo_runlist_ce(device), 0, - &drm->cechan); + true, &drm->cechan); if (ret) NV_ERROR(drm, "failed to create ce channel, %d\n", ret); @@ -381,7 +382,8 @@ nouveau_accel_init(struct nouveau_drm *drm) device->info.chipset != 0xaa && device->info.chipset != 0xac) { ret = nouveau_channel_new(drm, &drm->client.device, - NvDmaFB, NvDmaTT, &drm->cechan); + NvDmaFB, NvDmaTT, false, + &drm->cechan); if (ret) NV_ERROR(drm, "failed to create ce channel, %d\n", ret); @@ -393,7 +395,7 @@ nouveau_accel_init(struct nouveau_drm *drm) } ret = nouveau_channel_new(drm, &drm->client.device, - arg0, arg1, &drm->channel); + arg0, arg1, false, &drm->channel); if (ret) { NV_ERROR(drm, "failed to create kernel channel, %d\n", ret); nouveau_accel_fini(drm); @@ -1171,10 +1173,16 @@ nouveau_platform_device_create(const struct nvkm_device_tegra_func *func, goto err_free; } + err = nouveau_drm_device_init(drm); + if (err) + goto err_put; + platform_set_drvdata(pdev, drm); return drm; +err_put: + drm_dev_put(drm); err_free: nvkm_device_del(pdevice); diff --git a/drivers/gpu/drm/nouveau/nouveau_vmm.h b/drivers/gpu/drm/nouveau/nouveau_vmm.h index 7e3b118cf7c4..ede872f6f668 100644 --- a/drivers/gpu/drm/nouveau/nouveau_vmm.h +++ b/drivers/gpu/drm/nouveau/nouveau_vmm.h @@ -25,7 +25,6 @@ void nouveau_vma_unmap(struct nouveau_vma *); struct nouveau_vmm { struct nouveau_cli *cli; struct nvif_vmm vmm; - struct nvkm_vm *vm; }; int nouveau_vmm_init(struct nouveau_cli *, s32 oclass, struct nouveau_vmm *); diff --git a/drivers/gpu/drm/nouveau/nvif/disp.c b/drivers/gpu/drm/nouveau/nvif/disp.c index 18c7d064f75c..ef97dd223a32 100644 --- a/drivers/gpu/drm/nouveau/nvif/disp.c +++ b/drivers/gpu/drm/nouveau/nvif/disp.c @@ -34,6 +34,7 @@ int nvif_disp_ctor(struct nvif_device *device, s32 oclass, struct nvif_disp *disp) { static const struct nvif_mclass disps[] = { + { TU104_DISP, -1 }, { GV100_DISP, -1 }, { GP102_DISP, -1 }, { GP100_DISP, -1 }, diff --git a/drivers/gpu/drm/nouveau/nvkm/core/subdev.c b/drivers/gpu/drm/nouveau/nvkm/core/subdev.c index 03f676c18aad..c61b467cf45e 100644 --- a/drivers/gpu/drm/nouveau/nvkm/core/subdev.c +++ b/drivers/gpu/drm/nouveau/nvkm/core/subdev.c @@ -79,7 +79,9 @@ nvkm_subdev_name[NVKM_SUBDEV_NR] = { [NVKM_ENGINE_NVENC0 ] = "nvenc0", [NVKM_ENGINE_NVENC1 ] = "nvenc1", [NVKM_ENGINE_NVENC2 ] = "nvenc2", - [NVKM_ENGINE_NVDEC ] = "nvdec", + [NVKM_ENGINE_NVDEC0 ] = "nvdec0", + [NVKM_ENGINE_NVDEC1 ] = "nvdec1", + [NVKM_ENGINE_NVDEC2 ] = "nvdec2", [NVKM_ENGINE_PM ] = "pm", [NVKM_ENGINE_SEC ] = "sec", [NVKM_ENGINE_SEC2 ] = "sec2", diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild index 80d784441904..177a23301d6a 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild @@ -6,3 +6,4 @@ nvkm-y += nvkm/engine/ce/gm200.o nvkm-y += nvkm/engine/ce/gp100.o nvkm-y += nvkm/engine/ce/gp102.o nvkm-y += nvkm/engine/ce/gv100.o +nvkm-y += nvkm/engine/ce/tu104.o diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ce/tu104.c b/drivers/gpu/drm/nouveau/nvkm/engine/ce/tu104.c new file mode 100644 index 000000000000..3c25043bbb33 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/ce/tu104.c @@ -0,0 +1,40 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "priv.h" + +#include <nvif/class.h> + +static const struct nvkm_engine_func +tu104_ce = { + .intr = gp100_ce_intr, + .sclass = { + { -1, -1, TURING_DMA_COPY_A }, + {} + } +}; + +int +tu104_ce_new(struct nvkm_device *device, int index, + struct nvkm_engine **pengine) +{ + return nvkm_engine_new_(&tu104_ce, device, index, true, pengine); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c index e294013426ce..bfbc9341e0c2 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c @@ -2221,7 +2221,7 @@ nv132_chipset = { .dma = gf119_dma_new, .fifo = gp100_fifo_new, .gr = gp102_gr_new, - .nvdec = gp102_nvdec_new, + .nvdec[0] = gp102_nvdec_new, .sec2 = gp102_sec2_new, .sw = gf100_sw_new, }; @@ -2257,7 +2257,7 @@ nv134_chipset = { .dma = gf119_dma_new, .fifo = gp100_fifo_new, .gr = gp104_gr_new, - .nvdec = gp102_nvdec_new, + .nvdec[0] = gp102_nvdec_new, .sec2 = gp102_sec2_new, .sw = gf100_sw_new, }; @@ -2293,7 +2293,7 @@ nv136_chipset = { .dma = gf119_dma_new, .fifo = gp100_fifo_new, .gr = gp104_gr_new, - .nvdec = gp102_nvdec_new, + .nvdec[0] = gp102_nvdec_new, .sec2 = gp102_sec2_new, .sw = gf100_sw_new, }; @@ -2329,7 +2329,7 @@ nv137_chipset = { .dma = gf119_dma_new, .fifo = gp100_fifo_new, .gr = gp107_gr_new, - .nvdec = gp102_nvdec_new, + .nvdec[0] = gp102_nvdec_new, .sec2 = gp102_sec2_new, .sw = gf100_sw_new, }; @@ -2365,7 +2365,7 @@ nv138_chipset = { .dma = gf119_dma_new, .fifo = gp100_fifo_new, .gr = gp107_gr_new, - .nvdec = gp102_nvdec_new, + .nvdec[0] = gp102_nvdec_new, .sec2 = gp102_sec2_new, .sw = gf100_sw_new, }; @@ -2430,10 +2430,74 @@ nv140_chipset = { .dma = gv100_dma_new, .fifo = gv100_fifo_new, .gr = gv100_gr_new, - .nvdec = gp102_nvdec_new, + .nvdec[0] = gp102_nvdec_new, .sec2 = gp102_sec2_new, }; +static const struct nvkm_device_chip +nv164_chipset = { + .name = "TU104", + .bar = tu104_bar_new, + .bios = nvkm_bios_new, + .bus = gf100_bus_new, + .devinit = tu104_devinit_new, + .fault = tu104_fault_new, + .fb = gv100_fb_new, + .fuse = gm107_fuse_new, + .gpio = gk104_gpio_new, + .i2c = gm200_i2c_new, + .ibus = gm200_ibus_new, + .imem = nv50_instmem_new, + .ltc = gp102_ltc_new, + .mc = tu104_mc_new, + .mmu = tu104_mmu_new, + .pci = gp100_pci_new, + .pmu = gp102_pmu_new, + .therm = gp100_therm_new, + .timer = gk20a_timer_new, + .top = gk104_top_new, + .ce[0] = tu104_ce_new, + .ce[1] = tu104_ce_new, + .ce[2] = tu104_ce_new, + .ce[3] = tu104_ce_new, + .ce[4] = tu104_ce_new, + .disp = tu104_disp_new, + .dma = gv100_dma_new, + .fifo = tu104_fifo_new, +}; + +static const struct nvkm_device_chip +nv166_chipset = { + .name = "TU106", + .bar = tu104_bar_new, + .bios = nvkm_bios_new, + .bus = gf100_bus_new, + .devinit = tu104_devinit_new, + .fault = tu104_fault_new, + .fb = gv100_fb_new, + .fuse = gm107_fuse_new, + .gpio = gk104_gpio_new, + .i2c = gm200_i2c_new, + .ibus = gm200_ibus_new, + .imem = nv50_instmem_new, + .ltc = gp102_ltc_new, + .mc = tu104_mc_new, + .mmu = tu104_mmu_new, + .pci = gp100_pci_new, + .pmu = gp102_pmu_new, + .therm = gp100_therm_new, + .timer = gk20a_timer_new, + .top = gk104_top_new, + .ce[0] = tu104_ce_new, + .ce[1] = tu104_ce_new, + .ce[2] = tu104_ce_new, + .ce[3] = tu104_ce_new, + .ce[4] = tu104_ce_new, + .disp = tu104_disp_new, + .dma = gv100_dma_new, + .fifo = tu104_fifo_new, +}; + static int nvkm_device_event_ctor(struct nvkm_object *object, void *data, u32 size, struct nvkm_notify *notify) @@ -2529,7 +2593,9 @@ nvkm_device_engine(struct nvkm_device *device, int index) _(NVENC0 , device->nvenc[0], device->nvenc[0]); _(NVENC1 , device->nvenc[1], device->nvenc[1]); _(NVENC2 , device->nvenc[2], device->nvenc[2]); - _(NVDEC , device->nvdec , &device->nvdec->engine); + _(NVDEC0 , device->nvdec[0], &device->nvdec[0]->engine); + _(NVDEC1 , device->nvdec[1], &device->nvdec[1]->engine); + _(NVDEC2 , device->nvdec[2], &device->nvdec[2]->engine); _(PM , device->pm , &device->pm->engine); _(SEC , device->sec , device->sec); _(SEC2 , device->sec2 , &device->sec2->engine); @@ -2791,6 +2857,7 @@ nvkm_device_ctor(const struct nvkm_device_func *func, case 0x120: device->card_type = GM100; break; case 0x130: device->card_type = GP100; break; case 0x140: device->card_type = GV100; break; + case 0x160: device->card_type = TU100; break; default: break; } @@ -2883,6 +2950,8 @@ nvkm_device_ctor(const struct nvkm_device_func *func, case 0x138: device->chip = &nv138_chipset; break; case 0x13b: device->chip = &nv13b_chipset; break; case 0x140: device->chip = &nv140_chipset; break; + case 0x164: device->chip = &nv164_chipset; break; + case 0x166: device->chip = &nv166_chipset; break; default: nvdev_error(device, "unknown chipset (%08x)\n", boot0); goto done; @@ -2988,7 +3057,9 @@ nvkm_device_ctor(const struct nvkm_device_func *func, _(NVKM_ENGINE_NVENC0 , nvenc[0]); _(NVKM_ENGINE_NVENC1 , nvenc[1]); _(NVKM_ENGINE_NVENC2 , nvenc[2]); - _(NVKM_ENGINE_NVDEC , nvdec); + _(NVKM_ENGINE_NVDEC0 , nvdec[0]); + _(NVKM_ENGINE_NVDEC1 , nvdec[1]); + _(NVKM_ENGINE_NVDEC2 , nvdec[2]); _(NVKM_ENGINE_PM , pm); _(NVKM_ENGINE_SEC , sec); _(NVKM_ENGINE_SEC2 , sec2); diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/user.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/user.c index dde6bbafa709..092ddc4ffefa 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/device/user.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/user.c @@ -91,7 +91,7 @@ nvkm_udevice_info_v1(struct nvkm_device *device, case ENGINE_A(MSENC ); break; case ENGINE_A(VIC ); break; case ENGINE_A(SEC2 ); break; - case ENGINE_A(NVDEC ); break; + case ENGINE_B(NVDEC ); break; case ENGINE_B(NVENC ); break; default: args->mthd = NV_DEVICE_INFO_INVALID; @@ -175,6 +175,7 @@ nvkm_udevice_info(struct nvkm_udevice *udev, void *data, u32 size) case GM100: args->v0.family = NV_DEVICE_INFO_V0_MAXWELL; break; case GP100: args->v0.family = NV_DEVICE_INFO_V0_PASCAL; break; case GV100: args->v0.family = NV_DEVICE_INFO_V0_VOLTA; break; + case TU100: args->v0.family = NV_DEVICE_INFO_V0_TURING; break; default: args->v0.family = 0; break; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild index 8089ac9a12e2..c6a257ba4347 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild @@ -15,6 +15,7 @@ nvkm-y += nvkm/engine/disp/gm200.o nvkm-y += nvkm/engine/disp/gp100.o nvkm-y += nvkm/engine/disp/gp102.o nvkm-y += nvkm/engine/disp/gv100.o +nvkm-y += nvkm/engine/disp/tu104.o nvkm-y += nvkm/engine/disp/vga.o nvkm-y += nvkm/engine/disp/head.o @@ -38,6 +39,7 @@ nvkm-y += nvkm/engine/disp/sorgk104.o nvkm-y += nvkm/engine/disp/sorgm107.o nvkm-y += nvkm/engine/disp/sorgm200.o nvkm-y += nvkm/engine/disp/sorgv100.o +nvkm-y += nvkm/engine/disp/sortu104.o nvkm-y += nvkm/engine/disp/outp.o nvkm-y += nvkm/engine/disp/dp.o @@ -69,6 +71,7 @@ nvkm-y += nvkm/engine/disp/rootgm200.o nvkm-y += nvkm/engine/disp/rootgp100.o nvkm-y += nvkm/engine/disp/rootgp102.o nvkm-y += nvkm/engine/disp/rootgv100.o +nvkm-y += nvkm/engine/disp/roottu104.o nvkm-y += nvkm/engine/disp/channv50.o nvkm-y += nvkm/engine/disp/changf119.o diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gv100.c index d0a7e3456da1..47be0ba4aebe 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gv100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gv100.c @@ -28,7 +28,7 @@ #include <core/gpuobj.h> #include <subdev/timer.h> -static int +int gv100_disp_wndw_cnt(struct nvkm_disp *disp, unsigned long *pmask) { struct nvkm_device *device = disp->engine.subdev.device; @@ -36,7 +36,7 @@ gv100_disp_wndw_cnt(struct nvkm_disp *disp, unsigned long *pmask) return (nvkm_rd32(device, 0x610074) & 0x03f00000) >> 20; } -static void +void gv100_disp_super(struct work_struct *work) { struct nv50_disp *disp = @@ -257,7 +257,7 @@ gv100_disp_intr_head_timing(struct nv50_disp *disp, int head) } } -static void +void gv100_disp_intr(struct nv50_disp *disp) { struct nvkm_subdev *subdev = &disp->base.engine.subdev; @@ -297,7 +297,7 @@ gv100_disp_intr(struct nv50_disp *disp) nvkm_warn(subdev, "intr %08x\n", stat); } -static void +void gv100_disp_fini(struct nv50_disp *disp) { struct nvkm_device *device = disp->base.engine.subdev.device; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ior.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ior.h index 0f0c86c32ec3..790e42f460fd 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/ior.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/ior.h @@ -144,6 +144,11 @@ void gm200_sor_route_set(struct nvkm_outp *, struct nvkm_ior *); int gm200_sor_route_get(struct nvkm_outp *, int *); void gm200_sor_dp_drive(struct nvkm_ior *, int, int, int, int, int); +void gv100_sor_state(struct nvkm_ior *, struct nvkm_ior_state *); +void gv100_sor_dp_audio(struct nvkm_ior *, int, bool); +void gv100_sor_dp_audio_sym(struct nvkm_ior *, int, u16, u32); +void gv100_sor_dp_watermark(struct nvkm_ior *, int, u8); + void g84_hdmi_ctrl(struct nvkm_ior *, int, bool, u8, u8, u8 *, u8 , u8 *, u8); void gt215_hdmi_ctrl(struct nvkm_ior *, int, bool, u8, u8, u8 *, u8 , u8 *, u8); void gf119_hdmi_ctrl(struct nvkm_ior *, int, bool, u8, u8, u8 *, u8 , u8 *, u8); @@ -195,4 +200,6 @@ int gm200_sor_new(struct nvkm_disp *, int); int gv100_sor_cnt(struct nvkm_disp *, unsigned long *); int gv100_sor_new(struct nvkm_disp *, int); + +int tu104_sor_new(struct nvkm_disp *, int); #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.h index 8580382ab248..c36a8a7cafa1 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/nv50.h @@ -78,6 +78,11 @@ void gf119_disp_intr(struct nv50_disp *); void gf119_disp_super(struct work_struct *); void gf119_disp_intr_error(struct nv50_disp *, int); +void gv100_disp_fini(struct nv50_disp *); +void gv100_disp_intr(struct nv50_disp *); +void gv100_disp_super(struct work_struct *); +int gv100_disp_wndw_cnt(struct nvkm_disp *, unsigned long *); + void nv50_disp_dptmds_war_2(struct nv50_disp *, struct dcb_output *); void nv50_disp_dptmds_war_3(struct nv50_disp *, struct dcb_output *); void nv50_disp_update_sppll1(struct nv50_disp *); diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.h index 6ca4f9184b51..97de928cbde1 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.h @@ -37,4 +37,5 @@ extern const struct nvkm_disp_oclass gm200_disp_root_oclass; extern const struct nvkm_disp_oclass gp100_disp_root_oclass; extern const struct nvkm_disp_oclass gp102_disp_root_oclass; extern const struct nvkm_disp_oclass gv100_disp_root_oclass; +extern const struct nvkm_disp_oclass tu104_disp_root_oclass; #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/roottu104.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/roottu104.c new file mode 100644 index 000000000000..ad438c62f66c --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/roottu104.c @@ -0,0 +1,52 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "rootnv50.h" +#include "channv50.h" + +#include <nvif/class.h> + +static const struct nv50_disp_root_func +tu104_disp_root = { + .user = { + {{0,0,TU104_DISP_CURSOR }, gv100_disp_curs_new }, + {{0,0,TU104_DISP_WINDOW_IMM_CHANNEL_DMA}, gv100_disp_wimm_new }, + {{0,0,TU104_DISP_CORE_CHANNEL_DMA }, gv100_disp_core_new }, + {{0,0,TU104_DISP_WINDOW_CHANNEL_DMA }, gv100_disp_wndw_new }, + {} + }, +}; + +static int +tu104_disp_root_new(struct nvkm_disp *disp, const struct nvkm_oclass *oclass, + void *data, u32 size, struct nvkm_object **pobject) +{ + return nv50_disp_root_new_(&tu104_disp_root, disp, oclass, + data, size, pobject); +} + +const struct nvkm_disp_oclass +tu104_disp_root_oclass = { + .base.oclass = TU104_DISP, + .base.minver = -1, + .base.maxver = -1, + .ctor = tu104_disp_root_new, +}; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgv100.c index 8ba881a729ee..b0597ff9a714 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgv100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgv100.c @@ -23,7 +23,7 @@ #include <subdev/timer.h> -static void +void gv100_sor_dp_watermark(struct nvkm_ior *sor, int head, u8 watermark) { struct nvkm_device *device = sor->disp->engine.subdev.device; @@ -31,7 +31,7 @@ gv100_sor_dp_watermark(struct nvkm_ior *sor, int head, u8 watermark) nvkm_mask(device, 0x616550 + hoff, 0x0c00003f, 0x08000000 | watermark); } -static void +void gv100_sor_dp_audio_sym(struct nvkm_ior *sor, int head, u16 h, u32 v) { struct nvkm_device *device = sor->disp->engine.subdev.device; @@ -40,7 +40,7 @@ gv100_sor_dp_audio_sym(struct nvkm_ior *sor, int head, u16 h, u32 v) nvkm_mask(device, 0x61656c + hoff, 0x00ffffff, v); } -static void +void gv100_sor_dp_audio(struct nvkm_ior *sor, int head, bool enable) { struct nvkm_device *device = sor->disp->engine.subdev.device; @@ -54,7 +54,7 @@ gv100_sor_dp_audio(struct nvkm_ior *sor, int head, bool enable) ); } -static void +void gv100_sor_state(struct nvkm_ior *sor, struct nvkm_ior_state *state) { struct nvkm_device *device = sor->disp->engine.subdev.device; diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sortu104.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sortu104.c new file mode 100644 index 000000000000..df026a525ef1 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sortu104.c @@ -0,0 +1,97 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "ior.h" + +#include <subdev/timer.h> + +static void +tu104_sor_dp_vcpi(struct nvkm_ior *sor, int head, + u8 slot, u8 slot_nr, u16 pbn, u16 aligned) +{ + struct nvkm_device *device = sor->disp->engine.subdev.device; + const u32 hoff = head * 0x800; + + nvkm_mask(device, 0x61657c + hoff, 0xffffffff, (aligned << 16) | pbn); + nvkm_mask(device, 0x616578 + hoff, 0x00003f3f, (slot_nr << 8) | slot); +} + +static int +tu104_sor_dp_links(struct nvkm_ior *sor, struct nvkm_i2c_aux *aux) +{ + struct nvkm_device *device = sor->disp->engine.subdev.device; + const u32 soff = nv50_ior_base(sor); + const u32 loff = nv50_sor_link(sor); + u32 dpctrl = 0x00000000; + u32 clksor = 0x00000000; + + clksor |= sor->dp.bw << 18; + dpctrl |= ((1 << sor->dp.nr) - 1) << 16; + if (sor->dp.mst) + dpctrl |= 0x40000000; + if (sor->dp.ef) + dpctrl |= 0x00004000; + + nvkm_mask(device, 0x612300 + soff, 0x007c0000, clksor); + + /*XXX*/ + nvkm_msec(device, 40, NVKM_DELAY); + nvkm_mask(device, 0x612300 + soff, 0x00030000, 0x00010000); + nvkm_mask(device, 0x61c10c + loff, 0x00000003, 0x00000001); + + nvkm_mask(device, 0x61c10c + loff, 0x401f4000, dpctrl); + return 0; +} + +static const struct nvkm_ior_func +tu104_sor = { + .route = { + .get = gm200_sor_route_get, + .set = gm200_sor_route_set, + }, + .state = gv100_sor_state, + .power = nv50_sor_power, + .clock = gf119_sor_clock, + .hdmi = { + .ctrl = gv100_hdmi_ctrl, + }, + .dp = { + .lanes = { 0, 1, 2, 3 }, + .links = tu104_sor_dp_links, + .power = g94_sor_dp_power, + .pattern = gm107_sor_dp_pattern, + .drive = gm200_sor_dp_drive, + .vcpi = tu104_sor_dp_vcpi, + .audio = gv100_sor_dp_audio, + .audio_sym = gv100_sor_dp_audio_sym, + .watermark = gv100_sor_dp_watermark, + }, + .hda = { + .hpd = gf119_hda_hpd, + .eld = gf119_hda_eld, + }, +}; + +int +tu104_sor_new(struct nvkm_disp *disp, int id) +{ + return nvkm_ior_new_(&tu104_sor, disp, SOR, id); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/tu104.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/tu104.c new file mode 100644 index 000000000000..13fa21459d38 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/tu104.c @@ -0,0 +1,152 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "nv50.h" +#include "head.h" +#include "ior.h" +#include "channv50.h" +#include "rootnv50.h" + +#include <core/gpuobj.h> +#include <subdev/timer.h> + +static int +tu104_disp_init(struct nv50_disp *disp) +{ + struct nvkm_device *device = disp->base.engine.subdev.device; + struct nvkm_head *head; + int i, j; + u32 tmp; + + /* Claim ownership of display. */ + if (nvkm_rd32(device, 0x6254e8) & 0x00000002) { + nvkm_mask(device, 0x6254e8, 0x00000001, 0x00000000); + if (nvkm_msec(device, 2000, + if (!(nvkm_rd32(device, 0x6254e8) & 0x00000002)) + break; + ) < 0) + return -EBUSY; + } + + /* Lock pin capabilities. */ + tmp = 0x00000021; /*XXX*/ + nvkm_wr32(device, 0x640008, tmp); + + /* SOR capabilities. */ + for (i = 0; i < disp->sor.nr; i++) { + tmp = nvkm_rd32(device, 0x61c000 + (i * 0x800)); + nvkm_mask(device, 0x640000, 0x00000100 << i, 0x00000100 << i); + nvkm_wr32(device, 0x640144 + (i * 0x08), tmp); + } + + /* Head capabilities. */ + list_for_each_entry(head, &disp->base.head, head) { + const int id = head->id; + + /* RG. */ + tmp = nvkm_rd32(device, 0x616300 + (id * 0x800)); + nvkm_wr32(device, 0x640048 + (id * 0x020), tmp); + + /* POSTCOMP. */ + for (j = 0; j < 5 * 4; j += 4) { + tmp = nvkm_rd32(device, 0x616140 + (id * 0x800) + j); + nvkm_wr32(device, 0x640680 + (id * 0x20) + j, tmp); + } + } + + /* Window capabilities. */ + for (i = 0; i < disp->wndw.nr; i++) { + nvkm_mask(device, 0x640004, 1 << i, 1 << i); + for (j = 0; j < 6 * 4; j += 4) { + tmp = nvkm_rd32(device, 0x630100 + (i * 0x800) + j); + nvkm_mask(device, 0x640780 + (i * 0x20) + j, 0xffffffff, tmp); + } + nvkm_mask(device, 0x64000c, 0x00000100, 0x00000100); + } + + /* IHUB capabilities. */ + for (i = 0; i < 3; i++) { + tmp = nvkm_rd32(device, 0x62e000 + (i * 0x04)); + nvkm_wr32(device, 0x640010 + (i * 0x04), tmp); + } + + nvkm_mask(device, 0x610078, 0x00000001, 0x00000001); + + /* Setup instance memory. */ + switch (nvkm_memory_target(disp->inst->memory)) { + case NVKM_MEM_TARGET_VRAM: tmp = 0x00000001; break; + case NVKM_MEM_TARGET_NCOH: tmp = 0x00000002; break; + case NVKM_MEM_TARGET_HOST: tmp = 0x00000003; break; + default: + break; + } + nvkm_wr32(device, 0x610010, 0x00000008 | tmp); + nvkm_wr32(device, 0x610014, disp->inst->addr >> 16); + + /* CTRL_DISP: AWAKEN, ERROR, SUPERVISOR[1-3]. */ + nvkm_wr32(device, 0x611cf0, 0x00000187); /* MSK. */ + nvkm_wr32(device, 0x611db0, 0x00000187); /* EN. */ + + /* EXC_OTHER: CURSn, CORE. */ + nvkm_wr32(device, 0x611cec, disp->head.mask << 16 | + 0x00000001); /* MSK. */ + nvkm_wr32(device, 0x611dac, 0x00000000); /* EN. */ + + /* EXC_WINIM. */ + nvkm_wr32(device, 0x611ce8, disp->wndw.mask); /* MSK. */ + nvkm_wr32(device, 0x611da8, 0x00000000); /* EN. */ + + /* EXC_WIN. */ + nvkm_wr32(device, 0x611ce4, disp->wndw.mask); /* MSK. */ + nvkm_wr32(device, 0x611da4, 0x00000000); /* EN. */ + + /* HEAD_TIMING(n): VBLANK. */ + list_for_each_entry(head, &disp->base.head, head) { + const u32 hoff = head->id * 4; + nvkm_wr32(device, 0x611cc0 + hoff, 0x00000004); /* MSK. */ + nvkm_wr32(device, 0x611d80 + hoff, 0x00000000); /* EN. */ + } + + /* OR. */ + nvkm_wr32(device, 0x611cf4, 0x00000000); /* MSK. */ + nvkm_wr32(device, 0x611db4, 0x00000000); /* EN. */ + return 0; +} + +static const struct nv50_disp_func +tu104_disp = { + .init = tu104_disp_init, + .fini = gv100_disp_fini, + .intr = gv100_disp_intr, + .uevent = &gv100_disp_chan_uevent, + .super = gv100_disp_super, + .root = &tu104_disp_root_oclass, + .wndw = { .cnt = gv100_disp_wndw_cnt }, + .head = { .cnt = gv100_head_cnt, .new = gv100_head_new }, + .sor = { .cnt = gv100_sor_cnt, .new = tu104_sor_new }, + .ramht_size = 0x2000, +}; + +int +tu104_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp) +{ + return nv50_disp_new_(&tu104_disp, device, index, pdisp); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/wndwgv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/wndwgv100.c index 98911805aabf..5d3b641dbb14 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/wndwgv100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/wndwgv100.c @@ -118,7 +118,7 @@ gv100_disp_wndw_mthd_base = { const struct nv50_disp_chan_mthd gv100_disp_wndw_mthd = { - .name = "Base", + .name = "Window", .addr = 0x001000, .prev = 0x000800, .data = { diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild index f00408577a6a..87d8e054e40a 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild @@ -16,6 +16,7 @@ nvkm-y += nvkm/engine/fifo/gm20b.o nvkm-y += nvkm/engine/fifo/gp100.o nvkm-y += nvkm/engine/fifo/gp10b.o nvkm-y += nvkm/engine/fifo/gv100.o +nvkm-y += nvkm/engine/fifo/tu104.o nvkm-y += nvkm/engine/fifo/chan.o nvkm-y += nvkm/engine/fifo/channv50.o @@ -33,5 +34,7 @@ nvkm-y += nvkm/engine/fifo/gpfifog84.o nvkm-y += nvkm/engine/fifo/gpfifogf100.o nvkm-y += nvkm/engine/fifo/gpfifogk104.o nvkm-y += nvkm/engine/fifo/gpfifogv100.o +nvkm-y += nvkm/engine/fifo/gpfifotu104.o nvkm-y += nvkm/engine/fifo/usergv100.o +nvkm-y += nvkm/engine/fifo/usertu104.o diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.h b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.h index 3ffef236189e..2c7c5afc1ea5 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.h @@ -17,6 +17,7 @@ struct nvkm_fifo_chan_func { bool suspend); int (*object_ctor)(struct nvkm_fifo_chan *, struct nvkm_object *); void (*object_dtor)(struct nvkm_fifo_chan *, int); + u32 (*submit_token)(struct nvkm_fifo_chan *); }; int nvkm_fifo_chan_ctor(const struct nvkm_fifo_chan_func *, struct nvkm_fifo *, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/changk104.h b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/changk104.h index 8e28ba6b2307..a14545d871d8 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/changk104.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/changk104.h @@ -14,6 +14,8 @@ struct gk104_fifo_chan { struct list_head head; bool killed; + struct nvkm_memory *mthd; + struct { struct nvkm_gpuobj *inst; struct nvkm_vma *vma; @@ -36,4 +38,15 @@ int gk104_fifo_gpfifo_kick_locked(struct gk104_fifo_chan *); int gv100_fifo_gpfifo_new(struct gk104_fifo *, const struct nvkm_oclass *, void *data, u32 size, struct nvkm_object **); +int gv100_fifo_gpfifo_new_(const struct nvkm_fifo_chan_func *, + struct gk104_fifo *, u64 *, u16 *, u64, u64, u64, + u64 *, bool, u32 *, const struct nvkm_oclass *, + struct nvkm_object **); +int gv100_fifo_gpfifo_engine_init(struct nvkm_fifo_chan *, + struct nvkm_engine *); +int gv100_fifo_gpfifo_engine_fini(struct nvkm_fifo_chan *, + struct nvkm_engine *, bool); + +int tu104_fifo_gpfifo_new(struct gk104_fifo *, const struct nvkm_oclass *, + void *data, u32 size, struct nvkm_object **); #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gf100.c index f69576868164..10a2e7039a75 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gf100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gf100.c @@ -346,10 +346,10 @@ gf100_fifo_intr_fault(struct gf100_fifo *fifo, int unit) if (eu && eu->data2) { switch (eu->data2) { case NVKM_SUBDEV_BAR: - nvkm_mask(device, 0x001704, 0x00000000, 0x00000000); + nvkm_bar_bar1_reset(device); break; case NVKM_SUBDEV_INSTMEM: - nvkm_mask(device, 0x001714, 0x00000000, 0x00000000); + nvkm_bar_bar2_reset(device); break; case NVKM_ENGINE_IFB: nvkm_mask(device, 0x001718, 0x00000000, 0x00000000); diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c index afccf9721cf0..1053fe796466 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c @@ -149,16 +149,41 @@ gk104_fifo_uevent_init(struct nvkm_fifo *fifo) } void -gk104_fifo_runlist_commit(struct gk104_fifo *fifo, int runl) +gk104_fifo_runlist_commit(struct gk104_fifo *fifo, int runl, + struct nvkm_memory *mem, int nr) +{ + struct nvkm_subdev *subdev = &fifo->base.engine.subdev; + struct nvkm_device *device = subdev->device; + int target; + + switch (nvkm_memory_target(mem)) { + case NVKM_MEM_TARGET_VRAM: target = 0; break; + case NVKM_MEM_TARGET_NCOH: target = 3; break; + default: + WARN_ON(1); + return; + } + + nvkm_wr32(device, 0x002270, (nvkm_memory_addr(mem) >> 12) | + (target << 28)); + nvkm_wr32(device, 0x002274, (runl << 20) | nr); + + if (nvkm_msec(device, 2000, + if (!(nvkm_rd32(device, 0x002284 + (runl * 0x08)) & 0x00100000)) + break; + ) < 0) + nvkm_error(subdev, "runlist %d update timeout\n", runl); +} + +void +gk104_fifo_runlist_update(struct gk104_fifo *fifo, int runl) { const struct gk104_fifo_runlist_func *func = fifo->func->runlist; struct gk104_fifo_chan *chan; struct nvkm_subdev *subdev = &fifo->base.engine.subdev; - struct nvkm_device *device = subdev->device; struct nvkm_memory *mem; struct nvkm_fifo_cgrp *cgrp; int nr = 0; - int target; mutex_lock(&subdev->mutex); mem = fifo->runlist[runl].mem[fifo->runlist[runl].next]; @@ -177,24 +202,7 @@ gk104_fifo_runlist_commit(struct gk104_fifo *fifo, int runl) } nvkm_done(mem); - switch (nvkm_memory_target(mem)) { - case NVKM_MEM_TARGET_VRAM: target = 0; break; - case NVKM_MEM_TARGET_NCOH: target = 3; break; - default: - WARN_ON(1); - goto unlock; - } - - nvkm_wr32(device, 0x002270, (nvkm_memory_addr(mem) >> 12) | - (target << 28)); - nvkm_wr32(device, 0x002274, (runl << 20) | nr); - - if (nvkm_msec(device, 2000, - if (!(nvkm_rd32(device, 0x002284 + (runl * 0x08)) & 0x00100000)) - break; - ) < 0) - nvkm_error(subdev, "runlist %d update timeout\n", runl); -unlock: + func->commit(fifo, runl, mem, nr); mutex_unlock(&subdev->mutex); } @@ -238,6 +246,29 @@ const struct gk104_fifo_runlist_func gk104_fifo_runlist = { .size = 8, .chan = gk104_fifo_runlist_chan, + .commit = gk104_fifo_runlist_commit, +}; + +void +gk104_fifo_pbdma_init(struct gk104_fifo *fifo) +{ + struct nvkm_device *device = fifo->base.engine.subdev.device; + nvkm_wr32(device, 0x000204, (1 << fifo->pbdma_nr) - 1); +} + +int +gk104_fifo_pbdma_nr(struct gk104_fifo *fifo) +{ + struct nvkm_device *device = fifo->base.engine.subdev.device; + /* Determine number of PBDMAs by checking valid enable bits. */ + nvkm_wr32(device, 0x000204, 0xffffffff); + return hweight32(nvkm_rd32(device, 0x000204)); +} + +const struct gk104_fifo_pbdma_func +gk104_fifo_pbdma = { + .nr = gk104_fifo_pbdma_nr, + .init = gk104_fifo_pbdma_init, }; static void @@ -267,7 +298,7 @@ gk104_fifo_recover_work(struct work_struct *w) } for (todo = runm; runl = __ffs(todo), todo; todo &= ~BIT(runl)) - gk104_fifo_runlist_commit(fifo, runl); + gk104_fifo_runlist_update(fifo, runl); nvkm_wr32(device, 0x00262c, runm); nvkm_mask(device, 0x002630, runm, 0x00000000); @@ -456,10 +487,10 @@ gk104_fifo_fault(struct nvkm_fifo *base, struct nvkm_fault_data *info) if (ee && ee->data2) { switch (ee->data2) { case NVKM_SUBDEV_BAR: - nvkm_mask(device, 0x001704, 0x00000000, 0x00000000); + nvkm_bar_bar1_reset(device); break; case NVKM_SUBDEV_INSTMEM: - nvkm_mask(device, 0x001714, 0x00000000, 0x00000000); + nvkm_bar_bar2_reset(device); break; case NVKM_ENGINE_IFB: nvkm_mask(device, 0x001718, 0x00000000, 0x00000000); @@ -904,9 +935,7 @@ gk104_fifo_oneinit(struct nvkm_fifo *base) enum nvkm_devidx engidx; u32 *map; - /* Determine number of PBDMAs by checking valid enable bits. */ - nvkm_wr32(device, 0x000204, 0xffffffff); - fifo->pbdma_nr = hweight32(nvkm_rd32(device, 0x000204)); + fifo->pbdma_nr = fifo->func->pbdma->nr(fifo); nvkm_debug(subdev, "%d PBDMA(s)\n", fifo->pbdma_nr); /* Read PBDMA->runlist(s) mapping from HW. */ @@ -978,7 +1007,7 @@ gk104_fifo_init(struct nvkm_fifo *base) int i; /* Enable PBDMAs. */ - nvkm_wr32(device, 0x000204, (1 << fifo->pbdma_nr) - 1); + fifo->func->pbdma->init(fifo); /* PBDMA[n] */ for (i = 0; i < fifo->pbdma_nr; i++) { @@ -995,8 +1024,8 @@ gk104_fifo_init(struct nvkm_fifo *base) nvkm_wr32(device, 0x002254, 0x10000000 | fifo->user.bar->addr >> 12); - if (fifo->func->init_pbdma_timeout) - fifo->func->init_pbdma_timeout(fifo); + if (fifo->func->pbdma->init_timeout) + fifo->func->pbdma->init_timeout(fifo); nvkm_wr32(device, 0x002100, 0xffffffff); nvkm_wr32(device, 0x002140, 0x7fffffff); @@ -1175,6 +1204,7 @@ gk104_fifo_fault_gpcclient[] = { static const struct gk104_fifo_func gk104_fifo = { + .pbdma = &gk104_fifo_pbdma, .fault.access = gk104_fifo_fault_access, .fault.engine = gk104_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.h b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.h index d295b81e18d6..d4e565658f46 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.h @@ -45,7 +45,11 @@ struct gk104_fifo { }; struct gk104_fifo_func { - void (*init_pbdma_timeout)(struct gk104_fifo *); + const struct gk104_fifo_pbdma_func { + int (*nr)(struct gk104_fifo *); + void (*init)(struct gk104_fifo *); + void (*init_timeout)(struct gk104_fifo *); + } *pbdma; struct { const struct nvkm_enum *access; @@ -61,6 +65,8 @@ struct gk104_fifo_func { struct nvkm_memory *, u32 offset); void (*chan)(struct gk104_fifo_chan *, struct nvkm_memory *, u32 offset); + void (*commit)(struct gk104_fifo *, int runl, + struct nvkm_memory *, int entries); } *runlist; struct gk104_fifo_user_user { @@ -81,8 +87,11 @@ int gk104_fifo_new_(const struct gk104_fifo_func *, struct nvkm_device *, int index, int nr, struct nvkm_fifo **); void gk104_fifo_runlist_insert(struct gk104_fifo *, struct gk104_fifo_chan *); void gk104_fifo_runlist_remove(struct gk104_fifo *, struct gk104_fifo_chan *); -void gk104_fifo_runlist_commit(struct gk104_fifo *, int runl); +void gk104_fifo_runlist_update(struct gk104_fifo *, int runl); +extern const struct gk104_fifo_pbdma_func gk104_fifo_pbdma; +int gk104_fifo_pbdma_nr(struct gk104_fifo *); +void gk104_fifo_pbdma_init(struct gk104_fifo *); extern const struct nvkm_enum gk104_fifo_fault_access[]; extern const struct nvkm_enum gk104_fifo_fault_engine[]; extern const struct nvkm_enum gk104_fifo_fault_reason[]; @@ -91,15 +100,30 @@ extern const struct nvkm_enum gk104_fifo_fault_gpcclient[]; extern const struct gk104_fifo_runlist_func gk104_fifo_runlist; void gk104_fifo_runlist_chan(struct gk104_fifo_chan *, struct nvkm_memory *, u32); +void gk104_fifo_runlist_commit(struct gk104_fifo *, int runl, + struct nvkm_memory *, int); extern const struct gk104_fifo_runlist_func gk110_fifo_runlist; void gk110_fifo_runlist_cgrp(struct nvkm_fifo_cgrp *, struct nvkm_memory *, u32); -void gk208_fifo_init_pbdma_timeout(struct gk104_fifo *); +extern const struct gk104_fifo_pbdma_func gk208_fifo_pbdma; +void gk208_fifo_pbdma_init_timeout(struct gk104_fifo *); extern const struct nvkm_enum gm107_fifo_fault_engine[]; extern const struct gk104_fifo_runlist_func gm107_fifo_runlist; +extern const struct gk104_fifo_pbdma_func gm200_fifo_pbdma; +int gm200_fifo_pbdma_nr(struct gk104_fifo *); + extern const struct nvkm_enum gp100_fifo_fault_engine[]; + +extern const struct nvkm_enum gv100_fifo_fault_access[]; +extern const struct nvkm_enum gv100_fifo_fault_reason[]; +extern const struct nvkm_enum gv100_fifo_fault_hubclient[]; +extern const struct nvkm_enum gv100_fifo_fault_gpcclient[]; +void gv100_fifo_runlist_cgrp(struct nvkm_fifo_cgrp *, + struct nvkm_memory *, u32); +void gv100_fifo_runlist_chan(struct gk104_fifo_chan *, + struct nvkm_memory *, u32); #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk110.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk110.c index ac7655a130fb..8adfa6b182cb 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk110.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk110.c @@ -43,10 +43,12 @@ gk110_fifo_runlist = { .size = 8, .cgrp = gk110_fifo_runlist_cgrp, .chan = gk104_fifo_runlist_chan, + .commit = gk104_fifo_runlist_commit, }; static const struct gk104_fifo_func gk110_fifo = { + .pbdma = &gk104_fifo_pbdma, .fault.access = gk104_fifo_fault_access, .fault.engine = gk104_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk208.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk208.c index 5ea7e452cc66..9553fb4af601 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk208.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk208.c @@ -27,7 +27,7 @@ #include <nvif/class.h> void -gk208_fifo_init_pbdma_timeout(struct gk104_fifo *fifo) +gk208_fifo_pbdma_init_timeout(struct gk104_fifo *fifo) { struct nvkm_device *device = fifo->base.engine.subdev.device; int i; @@ -36,9 +36,16 @@ gk208_fifo_init_pbdma_timeout(struct gk104_fifo *fifo) nvkm_wr32(device, 0x04012c + (i * 0x2000), 0x0000ffff); } +const struct gk104_fifo_pbdma_func +gk208_fifo_pbdma = { + .nr = gk104_fifo_pbdma_nr, + .init = gk104_fifo_pbdma_init, + .init_timeout = gk208_fifo_pbdma_init_timeout, +}; + static const struct gk104_fifo_func gk208_fifo = { - .init_pbdma_timeout = gk208_fifo_init_pbdma_timeout, + .pbdma = &gk208_fifo_pbdma, .fault.access = gk104_fifo_fault_access, .fault.engine = gk104_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk20a.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk20a.c index 535a0eb67a5f..a4c6ac3cd6c7 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk20a.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk20a.c @@ -26,7 +26,7 @@ static const struct gk104_fifo_func gk20a_fifo = { - .init_pbdma_timeout = gk208_fifo_init_pbdma_timeout, + .pbdma = &gk208_fifo_pbdma, .fault.access = gk104_fifo_fault_access, .fault.engine = gk104_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm107.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm107.c index 79ae19b1db67..acf230764cb0 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm107.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm107.c @@ -41,6 +41,7 @@ gm107_fifo_runlist = { .size = 8, .cgrp = gk110_fifo_runlist_cgrp, .chan = gm107_fifo_runlist_chan, + .commit = gk104_fifo_runlist_commit, }; const struct nvkm_enum @@ -68,7 +69,7 @@ gm107_fifo_fault_engine[] = { static const struct gk104_fifo_func gm107_fifo = { - .init_pbdma_timeout = gk208_fifo_init_pbdma_timeout, + .pbdma = &gk208_fifo_pbdma, .fault.access = gk104_fifo_fault_access, .fault.engine = gm107_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm200.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm200.c index 49565faa854d..b96c1c5d6577 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm200.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm200.c @@ -26,9 +26,23 @@ #include <nvif/class.h> +int +gm200_fifo_pbdma_nr(struct gk104_fifo *fifo) +{ + struct nvkm_device *device = fifo->base.engine.subdev.device; + return nvkm_rd32(device, 0x002004) & 0x000000ff; +} + +const struct gk104_fifo_pbdma_func +gm200_fifo_pbdma = { + .nr = gm200_fifo_pbdma_nr, + .init = gk104_fifo_pbdma_init, + .init_timeout = gk208_fifo_pbdma_init_timeout, +}; + static const struct gk104_fifo_func gm200_fifo = { - .init_pbdma_timeout = gk208_fifo_init_pbdma_timeout, + .pbdma = &gm200_fifo_pbdma, .fault.access = gk104_fifo_fault_access, .fault.engine = gm107_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm20b.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm20b.c index 46736513bd11..a49539b9e4ec 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm20b.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm20b.c @@ -26,7 +26,7 @@ static const struct gk104_fifo_func gm20b_fifo = { - .init_pbdma_timeout = gk208_fifo_init_pbdma_timeout, + .pbdma = &gm200_fifo_pbdma, .fault.access = gk104_fifo_fault_access, .fault.engine = gm107_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gp100.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gp100.c index e2f8f9087d7c..54377e0f6a88 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gp100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gp100.c @@ -52,7 +52,7 @@ gp100_fifo_fault_engine[] = { static const struct gk104_fifo_func gp100_fifo = { - .init_pbdma_timeout = gk208_fifo_init_pbdma_timeout, + .pbdma = &gm200_fifo_pbdma, .fault.access = gk104_fifo_fault_access, .fault.engine = gp100_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gp10b.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gp10b.c index 7733bf7c6545..778ba7e46fb3 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gp10b.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gp10b.c @@ -26,7 +26,7 @@ static const struct gk104_fifo_func gp10b_fifo = { - .init_pbdma_timeout = gk208_fifo_init_pbdma_timeout, + .pbdma = &gm200_fifo_pbdma, .fault.access = gk104_fifo_fault_access, .fault.engine = gp100_fifo_fault_engine, .fault.reason = gk104_fifo_fault_reason, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk104.c index 118b37aea318..728a1edbf98c 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk104.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk104.c @@ -85,7 +85,7 @@ gk104_fifo_gpfifo_engine_addr(struct nvkm_engine *engine) case NVKM_ENGINE_MSVLD : return 0x0270; case NVKM_ENGINE_VIC : return 0x0280; case NVKM_ENGINE_MSENC : return 0x0290; - case NVKM_ENGINE_NVDEC : return 0x02100270; + case NVKM_ENGINE_NVDEC0: return 0x02100270; case NVKM_ENGINE_NVENC0: return 0x02100290; case NVKM_ENGINE_NVENC1: return 0x0210; default: @@ -192,7 +192,7 @@ gk104_fifo_gpfifo_fini(struct nvkm_fifo_chan *base) gk104_fifo_runlist_remove(fifo, chan); nvkm_mask(device, 0x800004 + coff, 0x00000800, 0x00000800); gk104_fifo_gpfifo_kick(chan); - gk104_fifo_runlist_commit(fifo, chan->runl); + gk104_fifo_runlist_update(fifo, chan->runl); } nvkm_wr32(device, 0x800000 + coff, 0x00000000); @@ -213,7 +213,7 @@ gk104_fifo_gpfifo_init(struct nvkm_fifo_chan *base) if (list_empty(&chan->head) && !chan->killed) { gk104_fifo_runlist_insert(fifo, chan); nvkm_mask(device, 0x800004 + coff, 0x00000400, 0x00000400); - gk104_fifo_runlist_commit(fifo, chan->runl); + gk104_fifo_runlist_update(fifo, chan->runl); nvkm_mask(device, 0x800004 + coff, 0x00000400, 0x00000400); } } @@ -222,6 +222,7 @@ void * gk104_fifo_gpfifo_dtor(struct nvkm_fifo_chan *base) { struct gk104_fifo_chan *chan = gk104_fifo_chan(base); + nvkm_memory_unref(&chan->mthd); kfree(chan->cgrp); return chan; } @@ -240,7 +241,7 @@ gk104_fifo_gpfifo_func = { static int gk104_fifo_gpfifo_new_(struct gk104_fifo *fifo, u64 *runlists, u16 *chid, - u64 vmm, u64 ioffset, u64 ilength, + u64 vmm, u64 ioffset, u64 ilength, u64 *inst, bool priv, const struct nvkm_oclass *oclass, struct nvkm_object **pobject) { @@ -279,6 +280,7 @@ gk104_fifo_gpfifo_new_(struct gk104_fifo *fifo, u64 *runlists, u16 *chid, return ret; *chid = chan->base.chid; + *inst = chan->base.inst->addr; /* Hack to support GPUs where even individual channels should be * part of a channel group. @@ -315,6 +317,7 @@ gk104_fifo_gpfifo_new_(struct gk104_fifo *fifo, u64 *runlists, u16 *chid, nvkm_wo32(chan->base.inst, 0x94, 0x30000001); nvkm_wo32(chan->base.inst, 0x9c, 0x00000100); nvkm_wo32(chan->base.inst, 0xac, 0x0000001f); + nvkm_wo32(chan->base.inst, 0xe4, priv ? 0x00000020 : 0x00000000); nvkm_wo32(chan->base.inst, 0xe8, chan->base.chid); nvkm_wo32(chan->base.inst, 0xb8, 0xf8000000); nvkm_wo32(chan->base.inst, 0xf8, 0x10003080); /* 0x002310 */ @@ -337,15 +340,19 @@ gk104_fifo_gpfifo_new(struct gk104_fifo *fifo, const struct nvkm_oclass *oclass, if (!(ret = nvif_unpack(ret, &data, &size, args->v0, 0, 0, false))) { nvif_ioctl(parent, "create channel gpfifo vers %d vmm %llx " "ioffset %016llx ilength %08x " - "runlist %016llx\n", + "runlist %016llx priv %d\n", args->v0.version, args->v0.vmm, args->v0.ioffset, - args->v0.ilength, args->v0.runlist); + args->v0.ilength, args->v0.runlist, args->v0.priv); + if (args->v0.priv && !oclass->client->super) + return -EINVAL; return gk104_fifo_gpfifo_new_(fifo, &args->v0.runlist, &args->v0.chid, args->v0.vmm, args->v0.ioffset, args->v0.ilength, + &args->v0.inst, + args->v0.priv, oclass, pobject); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogv100.c index 9598853ced56..a7462cf59d65 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogv100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogv100.c @@ -25,9 +25,15 @@ #include <core/client.h> #include <core/gpuobj.h> -#include <nvif/cla06f.h> +#include <nvif/clc36f.h> #include <nvif/unpack.h> +static u32 +gv100_fifo_gpfifo_submit_token(struct nvkm_fifo_chan *chan) +{ + return chan->chid; +} + static int gv100_fifo_gpfifo_engine_valid(struct gk104_fifo_chan *chan, bool ce, bool valid) { @@ -56,7 +62,7 @@ gv100_fifo_gpfifo_engine_valid(struct gk104_fifo_chan *chan, bool ce, bool valid return ret; } -static int +int gv100_fifo_gpfifo_engine_fini(struct nvkm_fifo_chan *base, struct nvkm_engine *engine, bool suspend) { @@ -79,7 +85,7 @@ gv100_fifo_gpfifo_engine_fini(struct nvkm_fifo_chan *base, return ret; } -static int +int gv100_fifo_gpfifo_engine_init(struct nvkm_fifo_chan *base, struct nvkm_engine *engine) { @@ -100,8 +106,8 @@ gv100_fifo_gpfifo_engine_init(struct nvkm_fifo_chan *base, return gv100_fifo_gpfifo_engine_valid(chan, false, true); } -const struct nvkm_fifo_chan_func -gv100_fifo_gpfifo_func = { +static const struct nvkm_fifo_chan_func +gv100_fifo_gpfifo = { .dtor = gk104_fifo_gpfifo_dtor, .init = gk104_fifo_gpfifo_init, .fini = gk104_fifo_gpfifo_fini, @@ -110,19 +116,23 @@ gv100_fifo_gpfifo_func = { .engine_dtor = gk104_fifo_gpfifo_engine_dtor, .engine_init = gv100_fifo_gpfifo_engine_init, .engine_fini = gv100_fifo_gpfifo_engine_fini, + .submit_token = gv100_fifo_gpfifo_submit_token, }; -static int -gv100_fifo_gpfifo_new_(struct gk104_fifo *fifo, u64 *runlists, u16 *chid, - u64 vmm, u64 ioffset, u64 ilength, - const struct nvkm_oclass *oclass, +int +gv100_fifo_gpfifo_new_(const struct nvkm_fifo_chan_func *func, + struct gk104_fifo *fifo, u64 *runlists, u16 *chid, + u64 vmm, u64 ioffset, u64 ilength, u64 *inst, bool priv, + u32 *token, const struct nvkm_oclass *oclass, struct nvkm_object **pobject) { + struct nvkm_device *device = fifo->base.engine.subdev.device; struct gk104_fifo_chan *chan; int runlist = ffs(*runlists) -1, ret, i; unsigned long engm; u64 subdevs = 0; - u64 usermem; + u64 usermem, mthd; + u32 size; if (!vmm || runlist < 0 || runlist >= fifo->runlist_nr) return -EINVAL; @@ -142,14 +152,15 @@ gv100_fifo_gpfifo_new_(struct gk104_fifo *fifo, u64 *runlists, u16 *chid, chan->runl = runlist; INIT_LIST_HEAD(&chan->head); - ret = nvkm_fifo_chan_ctor(&gv100_fifo_gpfifo_func, &fifo->base, - 0x1000, 0x1000, true, vmm, 0, subdevs, - 1, fifo->user.bar->addr, 0x200, + ret = nvkm_fifo_chan_ctor(func, &fifo->base, 0x1000, 0x1000, true, vmm, + 0, subdevs, 1, fifo->user.bar->addr, 0x200, oclass, &chan->base); if (ret) return ret; *chid = chan->base.chid; + *inst = chan->base.inst->addr; + *token = chan->base.func->submit_token(&chan->base); /* Hack to support GPUs where even individual channels should be * part of a channel group. @@ -173,6 +184,20 @@ gv100_fifo_gpfifo_new_(struct gk104_fifo *fifo, u64 *runlists, u16 *chid, nvkm_done(fifo->user.mem); usermem = nvkm_memory_addr(fifo->user.mem) + usermem; + /* Allocate fault method buffer (magics come from nvgpu). */ + size = nvkm_rd32(device, 0x104028); /* NV_PCE_PCE_MAP */ + size = 27 * 5 * (((9 + 1 + 3) * hweight32(size)) + 2); + size = roundup(size, PAGE_SIZE); + + ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST, size, 0x1000, true, + &chan->mthd); + if (ret) + return ret; + + mthd = nvkm_memory_bar2(chan->mthd); + if (mthd == ~0ULL) + return -EFAULT; + /* RAMFC */ nvkm_kmap(chan->base.inst); nvkm_wo32(chan->base.inst, 0x008, lower_32_bits(usermem)); @@ -184,13 +209,13 @@ gv100_fifo_gpfifo_new_(struct gk104_fifo *fifo, u64 *runlists, u16 *chid, (ilength << 16)); nvkm_wo32(chan->base.inst, 0x084, 0x20400000); nvkm_wo32(chan->base.inst, 0x094, 0x30000001); - nvkm_wo32(chan->base.inst, 0x0e4, 0x00000020); + nvkm_wo32(chan->base.inst, 0x0e4, priv ? 0x00000020 : 0x00000000); nvkm_wo32(chan->base.inst, 0x0e8, chan->base.chid); - nvkm_wo32(chan->base.inst, 0x0f4, 0x00001100); + nvkm_wo32(chan->base.inst, 0x0f4, 0x00001000); nvkm_wo32(chan->base.inst, 0x0f8, 0x10003080); nvkm_mo32(chan->base.inst, 0x218, 0x00000000, 0x00000000); - nvkm_wo32(chan->base.inst, 0x220, 0x020a1000); - nvkm_wo32(chan->base.inst, 0x224, 0x00000000); + nvkm_wo32(chan->base.inst, 0x220, lower_32_bits(mthd)); + nvkm_wo32(chan->base.inst, 0x224, upper_32_bits(mthd)); nvkm_done(chan->base.inst); return gv100_fifo_gpfifo_engine_valid(chan, true, true); } @@ -201,7 +226,7 @@ gv100_fifo_gpfifo_new(struct gk104_fifo *fifo, const struct nvkm_oclass *oclass, { struct nvkm_object *parent = oclass->parent; union { - struct kepler_channel_gpfifo_a_v0 v0; + struct volta_channel_gpfifo_a_v0 v0; } *args = data; int ret = -ENOSYS; @@ -209,15 +234,20 @@ gv100_fifo_gpfifo_new(struct gk104_fifo *fifo, const struct nvkm_oclass *oclass, if (!(ret = nvif_unpack(ret, &data, &size, args->v0, 0, 0, false))) { nvif_ioctl(parent, "create channel gpfifo vers %d vmm %llx " "ioffset %016llx ilength %08x " - "runlist %016llx\n", + "runlist %016llx priv %d\n", args->v0.version, args->v0.vmm, args->v0.ioffset, - args->v0.ilength, args->v0.runlist); - return gv100_fifo_gpfifo_new_(fifo, + args->v0.ilength, args->v0.runlist, args->v0.priv); + if (args->v0.priv && !oclass->client->super) + return -EINVAL; + return gv100_fifo_gpfifo_new_(&gv100_fifo_gpfifo, fifo, &args->v0.runlist, &args->v0.chid, args->v0.vmm, args->v0.ioffset, args->v0.ilength, + &args->v0.inst, + args->v0.priv, + &args->v0.token, oclass, pobject); } diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifotu104.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifotu104.c new file mode 100644 index 000000000000..ff70484dd01a --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifotu104.c @@ -0,0 +1,83 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "changk104.h" +#include "cgrp.h" + +#include <core/client.h> +#include <core/gpuobj.h> + +#include <nvif/clc36f.h> +#include <nvif/unpack.h> + +static u32 +tu104_fifo_gpfifo_submit_token(struct nvkm_fifo_chan *base) +{ + struct gk104_fifo_chan *chan = gk104_fifo_chan(base); + return (chan->runl << 16) | chan->base.chid; +} + +static const struct nvkm_fifo_chan_func +tu104_fifo_gpfifo = { + .dtor = gk104_fifo_gpfifo_dtor, + .init = gk104_fifo_gpfifo_init, + .fini = gk104_fifo_gpfifo_fini, + .ntfy = gf100_fifo_chan_ntfy, + .engine_ctor = gk104_fifo_gpfifo_engine_ctor, + .engine_dtor = gk104_fifo_gpfifo_engine_dtor, + .engine_init = gv100_fifo_gpfifo_engine_init, + .engine_fini = gv100_fifo_gpfifo_engine_fini, + .submit_token = tu104_fifo_gpfifo_submit_token, +}; + +int +tu104_fifo_gpfifo_new(struct gk104_fifo *fifo, const struct nvkm_oclass *oclass, + void *data, u32 size, struct nvkm_object **pobject) +{ + struct nvkm_object *parent = oclass->parent; + union { + struct volta_channel_gpfifo_a_v0 v0; + } *args = data; + int ret = -ENOSYS; + + nvif_ioctl(parent, "create channel gpfifo size %d\n", size); + if (!(ret = nvif_unpack(ret, &data, &size, args->v0, 0, 0, false))) { + nvif_ioctl(parent, "create channel gpfifo vers %d vmm %llx " + "ioffset %016llx ilength %08x " + "runlist %016llx priv %d\n", + args->v0.version, args->v0.vmm, args->v0.ioffset, + args->v0.ilength, args->v0.runlist, args->v0.priv); + if (args->v0.priv && !oclass->client->super) + return -EINVAL; + return gv100_fifo_gpfifo_new_(&tu104_fifo_gpfifo, fifo, + &args->v0.runlist, + &args->v0.chid, + args->v0.vmm, + args->v0.ioffset, + args->v0.ilength, + &args->v0.inst, + args->v0.priv, + &args->v0.token, + oclass, pobject); + } + + return ret; +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gv100.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gv100.c index 4e1d159c0ae7..6ee1bb32a071 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gv100.c +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gv100.c @@ -28,7 +28,7 @@ #include <nvif/class.h> -static void +void gv100_fifo_runlist_chan(struct gk104_fifo_chan *chan, struct nvkm_memory *memory, u32 offset) { @@ -42,7 +42,7 @@ gv100_fifo_runlist_chan(struct gk104_fifo_chan *chan, nvkm_wo32(memory, offset + 0xc, upper_32_bits(inst)); } -static void +void gv100_fifo_runlist_cgrp(struct nvkm_fifo_cgrp *cgrp, struct nvkm_memory *memory, u32 offset) { @@ -57,9 +57,10 @@ gv100_fifo_runlist = { .size = 16, .cgrp = gv100_fifo_runlist_cgrp, .chan = gv100_fifo_runlist_chan, + .commit = gk104_fifo_runlist_commit, }; -static const struct nvkm_enum +const struct nvkm_enum gv100_fifo_fault_gpcclient[] = { { 0x00, "T1_0" }, { 0x01, "T1_1" }, @@ -161,7 +162,7 @@ gv100_fifo_fault_gpcclient[] = { {} }; -static const struct nvkm_enum +const struct nvkm_enum gv100_fifo_fault_hubclient[] = { { 0x00, "VIP" }, { 0x01, "CE0" }, @@ -223,7 +224,7 @@ gv100_fifo_fault_hubclient[] = { {} }; -static const struct nvkm_enum +const struct nvkm_enum gv100_fifo_fault_reason[] = { { 0x00, "PDE" }, { 0x01, "PDE_SIZE" }, @@ -271,7 +272,7 @@ gv100_fifo_fault_engine[] = { {} }; -static const struct nvkm_enum +const struct nvkm_enum gv100_fifo_fault_access[] = { { 0x0, "VIRT_READ" }, { 0x1, "VIRT_WRITE" }, @@ -287,7 +288,7 @@ gv100_fifo_fault_access[] = { static const struct gk104_fifo_func gv100_fifo = { - .init_pbdma_timeout = gk208_fifo_init_pbdma_timeout, + .pbdma = &gm200_fifo_pbdma, .fault.access = gv100_fifo_fault_access, .fault.engine = gv100_fifo_fault_engine, .fault.reason = gv100_fifo_fault_reason, diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/tu104.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/tu104.c new file mode 100644 index 000000000000..98c80705bc61 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/tu104.c @@ -0,0 +1,116 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "gk104.h" +#include "cgrp.h" +#include "changk104.h" +#include "user.h" + +#include <core/gpuobj.h> + +#include <nvif/class.h> + +static void +tu104_fifo_runlist_commit(struct gk104_fifo *fifo, int runl, + struct nvkm_memory *mem, int nr) +{ + struct nvkm_device *device = fifo->base.engine.subdev.device; + u64 addr = nvkm_memory_addr(mem); + /*XXX: target? */ + + nvkm_wr32(device, 0x002b00 + (runl * 0x10), lower_32_bits(addr)); + nvkm_wr32(device, 0x002b04 + (runl * 0x10), upper_32_bits(addr)); + nvkm_wr32(device, 0x002b08 + (runl * 0x10), nr); + + /*XXX: how to wait? can you even wait? */ +} + +const struct gk104_fifo_runlist_func +tu104_fifo_runlist = { + .size = 16, + .cgrp = gv100_fifo_runlist_cgrp, + .chan = gv100_fifo_runlist_chan, + .commit = tu104_fifo_runlist_commit, +}; + +static const struct nvkm_enum +tu104_fifo_fault_engine[] = { + { 0x01, "DISPLAY" }, + { 0x03, "PTP" }, + { 0x06, "PWR_PMU" }, + { 0x08, "IFB", NULL, NVKM_ENGINE_IFB }, + { 0x09, "PERF" }, + { 0x1f, "PHYSICAL" }, + { 0x20, "HOST0" }, + { 0x21, "HOST1" }, + { 0x22, "HOST2" }, + { 0x23, "HOST3" }, + { 0x24, "HOST4" }, + { 0x25, "HOST5" }, + { 0x26, "HOST6" }, + { 0x27, "HOST7" }, + { 0x28, "HOST8" }, + { 0x29, "HOST9" }, + { 0x2a, "HOST10" }, + { 0x2b, "HOST11" }, + { 0x2c, "HOST12" }, + { 0x2d, "HOST13" }, + { 0x2e, "HOST14" }, + { 0x80, "BAR1", NULL, NVKM_SUBDEV_BAR }, + { 0xc0, "BAR2", NULL, NVKM_SUBDEV_INSTMEM }, + {} +}; + +static void +tu104_fifo_pbdma_init(struct gk104_fifo *fifo) +{ + struct nvkm_device *device = fifo->base.engine.subdev.device; + const u32 mask = (1 << fifo->pbdma_nr) - 1; + /*XXX: this is a bit of a guess at this point in time. */ + nvkm_mask(device, 0xb65000, 0x80000fff, 0x80000000 | mask); +} + +static const struct gk104_fifo_pbdma_func +tu104_fifo_pbdma = { + .nr = gm200_fifo_pbdma_nr, + .init = tu104_fifo_pbdma_init, + .init_timeout = gk208_fifo_pbdma_init_timeout, +}; + +static const struct gk104_fifo_func +tu104_fifo = { + .pbdma = &tu104_fifo_pbdma, + .fault.access = gv100_fifo_fault_access, + .fault.engine = tu104_fifo_fault_engine, + .fault.reason = gv100_fifo_fault_reason, + .fault.hubclient = gv100_fifo_fault_hubclient, + .fault.gpcclient = gv100_fifo_fault_gpcclient, + .runlist = &tu104_fifo_runlist, + .user = {{-1,-1,VOLTA_USERMODE_A }, tu104_fifo_user_new }, + .chan = {{ 0, 0,TURING_CHANNEL_GPFIFO_A}, tu104_fifo_gpfifo_new }, + .cgrp_force = true, +}; + +int +tu104_fifo_new(struct nvkm_device *device, int index, struct nvkm_fifo **pfifo) +{ + return gk104_fifo_new_(&tu104_fifo, device, index, 4096, pfifo); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/user.h b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/user.h index ed840921ebe8..14b0c6bde8eb 100644 --- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/user.h +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/user.h @@ -3,4 +3,6 @@ #include "priv.h" int gv100_fifo_user_new(const struct nvkm_oclass *, void *, u32, struct nvkm_object **); +int tu104_fifo_user_new(const struct nvkm_oclass *, void *, u32, + struct nvkm_object **); #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/usertu104.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/usertu104.c new file mode 100644 index 000000000000..8f98548a21f6 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/usertu104.c @@ -0,0 +1,45 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "user.h" + +static int +tu104_fifo_user_map(struct nvkm_object *object, void *argv, u32 argc, + enum nvkm_object_map *type, u64 *addr, u64 *size) +{ + struct nvkm_device *device = object->engine->subdev.device; + *addr = 0xbb0000 + device->func->resource_addr(device, 0); + *size = 0x010000; + *type = NVKM_OBJECT_MAP_IO; + return 0; +} + +static const struct nvkm_object_func +tu104_fifo_user = { + .map = tu104_fifo_user_map, +}; + +int +tu104_fifo_user_new(const struct nvkm_oclass *oclass, void *argv, u32 argc, + struct nvkm_object **pobject) +{ + return nvkm_object_new_(&tu104_fifo_user, oclass, argv, argc, pobject); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/falcon/base.c b/drivers/gpu/drm/nouveau/nvkm/falcon/base.c index 14be41f24155..427340153640 100644 --- a/drivers/gpu/drm/nouveau/nvkm/falcon/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/falcon/base.c @@ -197,7 +197,7 @@ nvkm_falcon_ctor(const struct nvkm_falcon_func *func, case NVKM_SUBDEV_PMU: debug_reg = 0xc08; break; - case NVKM_ENGINE_NVDEC: + case NVKM_ENGINE_NVDEC0: debug_reg = 0xd00; break; case NVKM_ENGINE_SEC2: diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/Kbuild index e5830453813d..ab0282dc0736 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/Kbuild @@ -5,3 +5,4 @@ nvkm-y += nvkm/subdev/bar/gf100.o nvkm-y += nvkm/subdev/bar/gk20a.o nvkm-y += nvkm/subdev/bar/gm107.o nvkm-y += nvkm/subdev/bar/gm20b.o +nvkm-y += nvkm/subdev/bar/tu104.o diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/base.c index 243f0a5c8a62..209a6a40834a 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/base.c @@ -36,6 +36,16 @@ nvkm_bar_bar1_vmm(struct nvkm_device *device) return device->bar->func->bar1.vmm(device->bar); } +void +nvkm_bar_bar1_reset(struct nvkm_device *device) +{ + struct nvkm_bar *bar = device->bar; + if (bar) { + bar->func->bar1.init(bar); + bar->func->bar1.wait(bar); + } +} + struct nvkm_vmm * nvkm_bar_bar2_vmm(struct nvkm_device *device) { @@ -49,6 +59,16 @@ nvkm_bar_bar2_vmm(struct nvkm_device *device) } void +nvkm_bar_bar2_reset(struct nvkm_device *device) +{ + struct nvkm_bar *bar = device->bar; + if (bar && bar->bar2) { + bar->func->bar2.init(bar); + bar->func->bar2.wait(bar); + } +} + +void nvkm_bar_bar2_fini(struct nvkm_device *device) { struct nvkm_bar *bar = device->bar; diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bar/tu104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/tu104.c new file mode 100644 index 000000000000..ecaead156e9b --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bar/tu104.c @@ -0,0 +1,98 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "gf100.h" + +#include <core/memory.h> +#include <subdev/timer.h> + +static void +tu104_bar_bar2_wait(struct nvkm_bar *bar) +{ + struct nvkm_device *device = bar->subdev.device; + nvkm_msec(device, 2000, + if (!(nvkm_rd32(device, 0xb80f50) & 0x0000000c)) + break; + ); +} + +static void +tu104_bar_bar2_fini(struct nvkm_bar *bar) +{ + nvkm_mask(bar->subdev.device, 0xb80f48, 0x80000000, 0x00000000); +} + +static void +tu104_bar_bar2_init(struct nvkm_bar *base) +{ + struct nvkm_device *device = base->subdev.device; + struct gf100_bar *bar = gf100_bar(base); + u32 addr = nvkm_memory_addr(bar->bar[0].inst) >> 12; + if (bar->bar2_halve) + addr |= 0x40000000; + nvkm_wr32(device, 0xb80f48, 0x80000000 | addr); +} + +static void +tu104_bar_bar1_wait(struct nvkm_bar *bar) +{ + struct nvkm_device *device = bar->subdev.device; + nvkm_msec(device, 2000, + if (!(nvkm_rd32(device, 0xb80f50) & 0x00000003)) + break; + ); +} + +static void +tu104_bar_bar1_fini(struct nvkm_bar *bar) +{ + nvkm_mask(bar->subdev.device, 0xb80f40, 0x80000000, 0x00000000); +} + +static void +tu104_bar_bar1_init(struct nvkm_bar *base) +{ + struct nvkm_device *device = base->subdev.device; + struct gf100_bar *bar = gf100_bar(base); + const u32 addr = nvkm_memory_addr(bar->bar[1].inst) >> 12; + nvkm_wr32(device, 0xb80f40, 0x80000000 | addr); +} + +static const struct nvkm_bar_func +tu104_bar = { + .dtor = gf100_bar_dtor, + .oneinit = gf100_bar_oneinit, + .bar1.init = tu104_bar_bar1_init, + .bar1.fini = tu104_bar_bar1_fini, + .bar1.wait = tu104_bar_bar1_wait, + .bar1.vmm = gf100_bar_bar1_vmm, + .bar2.init = tu104_bar_bar2_init, + .bar2.fini = tu104_bar_bar2_fini, + .bar2.wait = tu104_bar_bar2_wait, + .bar2.vmm = gf100_bar_bar2_vmm, + .flush = g84_bar_flush, +}; + +int +tu104_bar_new(struct nvkm_device *device, int index, struct nvkm_bar **pbar) +{ + return gf100_bar_new_(&tu104_bar, device, index, pbar); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/Kbuild index 50a436926484..3ef505a5c01b 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/Kbuild @@ -13,3 +13,4 @@ nvkm-y += nvkm/subdev/devinit/gf100.o nvkm-y += nvkm/subdev/devinit/gm107.o nvkm-y += nvkm/subdev/devinit/gm200.o nvkm-y += nvkm/subdev/devinit/gv100.o +nvkm-y += nvkm/subdev/devinit/tu104.o diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gm200.c b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gm200.c index 17235e940ca9..59940dacc2ba 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gm200.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gm200.c @@ -105,6 +105,15 @@ pmu_load(struct nv50_devinit *init, u8 type, bool post, return pmu_exec(init, pmu.init_addr_pmu), 0; } +void +gm200_devinit_preos(struct nv50_devinit *init, bool post) +{ + /* Optional: Execute PRE_OS application on PMU, which should at + * least take care of fans until a full PMU has been loaded. + */ + pmu_load(init, 0x01, post, NULL, NULL); +} + int gm200_devinit_post(struct nvkm_devinit *base, bool post) { @@ -156,10 +165,7 @@ gm200_devinit_post(struct nvkm_devinit *base, bool post) return -ETIMEDOUT; } - /* Optional: Execute PRE_OS application on PMU, which should at - * least take care of fans until a full PMU has been loaded. - */ - pmu_load(init, 0x01, post, NULL, NULL); + gm200_devinit_preos(init, post); return 0; } diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/nv50.h b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/nv50.h index 9b9f0dc1e192..72d130bb7f7c 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/nv50.h +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/nv50.h @@ -26,4 +26,5 @@ void gf100_devinit_preinit(struct nvkm_devinit *); u64 gm107_devinit_disable(struct nvkm_devinit *); int gm200_devinit_post(struct nvkm_devinit *, bool); +void gm200_devinit_preos(struct nv50_devinit *, bool); #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/tu104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/tu104.c new file mode 100644 index 000000000000..aae87b3fc429 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/tu104.c @@ -0,0 +1,89 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "nv50.h" + +#include <subdev/bios.h> +#include <subdev/bios/pll.h> +#include <subdev/clk/pll.h> + +static int +tu104_devinit_pll_set(struct nvkm_devinit *init, u32 type, u32 freq) +{ + struct nvkm_subdev *subdev = &init->subdev; + struct nvkm_device *device = subdev->device; + struct nvbios_pll info; + int head = type - PLL_VPLL0; + int N, fN, M, P; + int ret; + + ret = nvbios_pll_parse(device->bios, type, &info); + if (ret) + return ret; + + ret = gt215_pll_calc(subdev, &info, freq, &N, &fN, &M, &P); + if (ret < 0) + return ret; + + switch (info.type) { + case PLL_VPLL0: + case PLL_VPLL1: + case PLL_VPLL2: + case PLL_VPLL3: + nvkm_wr32(device, 0x00ef10 + (head * 0x40), fN << 16); + nvkm_wr32(device, 0x00ef04 + (head * 0x40), (P << 16) | + (N << 8) | + (M << 0)); + /*XXX*/ + nvkm_wr32(device, 0x00ef0c + (head * 0x40), 0x00000900); + nvkm_wr32(device, 0x00ef00 + (head * 0x40), 0x02000014); + break; + default: + nvkm_warn(subdev, "%08x/%dKhz unimplemented\n", type, freq); + ret = -EINVAL; + break; + } + + return ret; +} + +static int +tu104_devinit_post(struct nvkm_devinit *base, bool post) +{ + struct nv50_devinit *init = nv50_devinit(base); + gm200_devinit_preos(init, post); + return 0; +} + +static const struct nvkm_devinit_func +tu104_devinit = { + .init = nv50_devinit_init, + .post = tu104_devinit_post, + .pll_set = tu104_devinit_pll_set, + .disable = gm107_devinit_disable, +}; + +int +tu104_devinit_new(struct nvkm_device *device, int index, + struct nvkm_devinit **pinit) +{ + return nv50_devinit_new_(&tu104_devinit, device, index, pinit); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/Kbuild index 45bb46fb0929..794eb1745b2f 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/Kbuild @@ -1,3 +1,4 @@ nvkm-y += nvkm/subdev/fault/base.o nvkm-y += nvkm/subdev/fault/gp100.o nvkm-y += nvkm/subdev/fault/gv100.o +nvkm-y += nvkm/subdev/fault/tu104.o diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/base.c index 16ad91c91a7b..4ba1e21e8fda 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/base.c @@ -23,21 +23,19 @@ #include <core/memory.h> #include <core/notify.h> -#include <subdev/bar.h> -#include <subdev/mmu.h> static void nvkm_fault_ntfy_fini(struct nvkm_event *event, int type, int index) { struct nvkm_fault *fault = container_of(event, typeof(*fault), event); - fault->func->buffer.fini(fault->buffer[index]); + fault->func->buffer.intr(fault->buffer[index], false); } static void nvkm_fault_ntfy_init(struct nvkm_event *event, int type, int index) { struct nvkm_fault *fault = container_of(event, typeof(*fault), event); - fault->func->buffer.init(fault->buffer[index]); + fault->func->buffer.intr(fault->buffer[index], true); } static int @@ -91,7 +89,6 @@ nvkm_fault_oneinit_buffer(struct nvkm_fault *fault, int id) { struct nvkm_subdev *subdev = &fault->subdev; struct nvkm_device *device = subdev->device; - struct nvkm_vmm *bar2 = nvkm_bar_bar2_vmm(device); struct nvkm_fault_buffer *buffer; int ret; @@ -99,7 +96,7 @@ nvkm_fault_oneinit_buffer(struct nvkm_fault *fault, int id) return -ENOMEM; buffer->fault = fault; buffer->id = id; - buffer->entries = fault->func->buffer.entries(buffer); + fault->func->buffer.info(buffer); fault->buffer[id] = buffer; nvkm_debug(subdev, "buffer %d: %d entries\n", id, buffer->entries); @@ -110,12 +107,12 @@ nvkm_fault_oneinit_buffer(struct nvkm_fault *fault, int id) if (ret) return ret; - ret = nvkm_vmm_get(bar2, 12, nvkm_memory_size(buffer->mem), - &buffer->vma); - if (ret) - return ret; + /* Pin fault buffer in BAR2. */ + buffer->addr = nvkm_memory_bar2(buffer->mem); + if (buffer->addr == ~0ULL) + return -EFAULT; - return nvkm_memory_map(buffer->mem, 0, bar2, buffer->vma, NULL, 0); + return 0; } static int @@ -146,7 +143,6 @@ nvkm_fault_oneinit(struct nvkm_subdev *subdev) static void * nvkm_fault_dtor(struct nvkm_subdev *subdev) { - struct nvkm_vmm *bar2 = nvkm_bar_bar2_vmm(subdev->device); struct nvkm_fault *fault = nvkm_fault(subdev); int i; @@ -154,7 +150,6 @@ nvkm_fault_dtor(struct nvkm_subdev *subdev) for (i = 0; i < fault->buffer_nr; i++) { if (fault->buffer[i]) { - nvkm_vmm_put(bar2, &fault->buffer[i]->vma); nvkm_memory_unref(&fault->buffer[i]->mem); kfree(fault->buffer[i]); } diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gp100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gp100.c index 5e71db2e8d75..8fb96fe614f9 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gp100.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gp100.c @@ -21,7 +21,14 @@ */ #include "priv.h" -#include <subdev/mmu.h> +#include <subdev/mc.h> + +static void +gp100_fault_buffer_intr(struct nvkm_fault_buffer *buffer, bool enable) +{ + struct nvkm_device *device = buffer->fault->subdev.device; + nvkm_mc_intr_mask(device, NVKM_SUBDEV_FAULT, enable); +} static void gp100_fault_buffer_fini(struct nvkm_fault_buffer *buffer) @@ -34,15 +41,17 @@ static void gp100_fault_buffer_init(struct nvkm_fault_buffer *buffer) { struct nvkm_device *device = buffer->fault->subdev.device; - nvkm_wr32(device, 0x002a74, upper_32_bits(buffer->vma->addr)); - nvkm_wr32(device, 0x002a70, lower_32_bits(buffer->vma->addr)); + nvkm_wr32(device, 0x002a74, upper_32_bits(buffer->addr)); + nvkm_wr32(device, 0x002a70, lower_32_bits(buffer->addr)); nvkm_mask(device, 0x002a70, 0x00000001, 0x00000001); } -static u32 -gp100_fault_buffer_entries(struct nvkm_fault_buffer *buffer) +static void +gp100_fault_buffer_info(struct nvkm_fault_buffer *buffer) { - return nvkm_rd32(buffer->fault->subdev.device, 0x002a78); + buffer->entries = nvkm_rd32(buffer->fault->subdev.device, 0x002a78); + buffer->get = 0x002a7c; + buffer->put = 0x002a80; } static void @@ -56,9 +65,10 @@ gp100_fault = { .intr = gp100_fault_intr, .buffer.nr = 1, .buffer.entry_size = 32, - .buffer.entries = gp100_fault_buffer_entries, + .buffer.info = gp100_fault_buffer_info, .buffer.init = gp100_fault_buffer_init, .buffer.fini = gp100_fault_buffer_fini, + .buffer.intr = gp100_fault_buffer_intr, }; int diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gv100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gv100.c index 3cd610d7deb5..6fc54e17c935 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gv100.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/gv100.c @@ -30,9 +30,8 @@ gv100_fault_buffer_process(struct nvkm_fault_buffer *buffer) { struct nvkm_device *device = buffer->fault->subdev.device; struct nvkm_memory *mem = buffer->mem; - const u32 foff = buffer->id * 0x14; - u32 get = nvkm_rd32(device, 0x100e2c + foff); - u32 put = nvkm_rd32(device, 0x100e30 + foff); + u32 get = nvkm_rd32(device, buffer->get); + u32 put = nvkm_rd32(device, buffer->put); if (put == get) return; @@ -51,7 +50,7 @@ gv100_fault_buffer_process(struct nvkm_fault_buffer *buffer) if (++get == buffer->entries) get = 0; - nvkm_wr32(device, 0x100e2c + foff, get); + nvkm_wr32(device, buffer->get, get); info.addr = ((u64)addrhi << 32) | addrlo; info.inst = ((u64)insthi << 32) | instlo; @@ -70,13 +69,21 @@ gv100_fault_buffer_process(struct nvkm_fault_buffer *buffer) } static void -gv100_fault_buffer_fini(struct nvkm_fault_buffer *buffer) +gv100_fault_buffer_intr(struct nvkm_fault_buffer *buffer, bool enable) { struct nvkm_device *device = buffer->fault->subdev.device; const u32 intr = buffer->id ? 0x08000000 : 0x20000000; - const u32 foff = buffer->id * 0x14; + if (enable) + nvkm_mask(device, 0x100a2c, intr, intr); + else + nvkm_mask(device, 0x100a34, intr, intr); +} - nvkm_mask(device, 0x100a34, intr, intr); +static void +gv100_fault_buffer_fini(struct nvkm_fault_buffer *buffer) +{ + struct nvkm_device *device = buffer->fault->subdev.device; + const u32 foff = buffer->id * 0x14; nvkm_mask(device, 0x100e34 + foff, 0x80000000, 0x00000000); } @@ -84,23 +91,25 @@ static void gv100_fault_buffer_init(struct nvkm_fault_buffer *buffer) { struct nvkm_device *device = buffer->fault->subdev.device; - const u32 intr = buffer->id ? 0x08000000 : 0x20000000; const u32 foff = buffer->id * 0x14; nvkm_mask(device, 0x100e34 + foff, 0xc0000000, 0x40000000); - nvkm_wr32(device, 0x100e28 + foff, upper_32_bits(buffer->vma->addr)); - nvkm_wr32(device, 0x100e24 + foff, lower_32_bits(buffer->vma->addr)); + nvkm_wr32(device, 0x100e28 + foff, upper_32_bits(buffer->addr)); + nvkm_wr32(device, 0x100e24 + foff, lower_32_bits(buffer->addr)); nvkm_mask(device, 0x100e34 + foff, 0x80000000, 0x80000000); - nvkm_mask(device, 0x100a2c, intr, intr); } -static u32 -gv100_fault_buffer_entries(struct nvkm_fault_buffer *buffer) +static void +gv100_fault_buffer_info(struct nvkm_fault_buffer *buffer) { struct nvkm_device *device = buffer->fault->subdev.device; const u32 foff = buffer->id * 0x14; + nvkm_mask(device, 0x100e34 + foff, 0x40000000, 0x40000000); - return nvkm_rd32(device, 0x100e34 + foff) & 0x000fffff; + + buffer->entries = nvkm_rd32(device, 0x100e34 + foff) & 0x000fffff; + buffer->get = 0x100e2c + foff; + buffer->put = 0x100e30 + foff; } static int @@ -166,6 +175,8 @@ static void gv100_fault_fini(struct nvkm_fault *fault) { nvkm_notify_put(&fault->nrpfb); + if (fault->buffer[0]) + fault->func->buffer.fini(fault->buffer[0]); nvkm_mask(fault->subdev.device, 0x100a34, 0x80000000, 0x80000000); } @@ -173,14 +184,15 @@ static void gv100_fault_init(struct nvkm_fault *fault) { nvkm_mask(fault->subdev.device, 0x100a2c, 0x80000000, 0x80000000); + fault->func->buffer.init(fault->buffer[0]); nvkm_notify_get(&fault->nrpfb); } -static int +int gv100_fault_oneinit(struct nvkm_fault *fault) { return nvkm_notify_init(&fault->buffer[0]->object, &fault->event, - gv100_fault_ntfy_nrpfb, false, NULL, 0, 0, + gv100_fault_ntfy_nrpfb, true, NULL, 0, 0, &fault->nrpfb); } @@ -192,9 +204,10 @@ gv100_fault = { .intr = gv100_fault_intr, .buffer.nr = 2, .buffer.entry_size = 32, - .buffer.entries = gv100_fault_buffer_entries, + .buffer.info = gv100_fault_buffer_info, .buffer.init = gv100_fault_buffer_init, .buffer.fini = gv100_fault_buffer_fini, + .buffer.intr = gv100_fault_buffer_intr, }; int diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/priv.h index e4d2f5234fd1..8ca8b2876dad 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/priv.h +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/priv.h @@ -12,8 +12,10 @@ struct nvkm_fault_buffer { struct nvkm_fault *fault; int id; int entries; + u32 get; + u32 put; struct nvkm_memory *mem; - struct nvkm_vma *vma; + u64 addr; }; int nvkm_fault_new_(const struct nvkm_fault_func *, struct nvkm_device *, @@ -27,9 +29,12 @@ struct nvkm_fault_func { struct { int nr; u32 entry_size; - u32 (*entries)(struct nvkm_fault_buffer *); + void (*info)(struct nvkm_fault_buffer *); void (*init)(struct nvkm_fault_buffer *); void (*fini)(struct nvkm_fault_buffer *); + void (*intr)(struct nvkm_fault_buffer *, bool enable); } buffer; }; + +int gv100_fault_oneinit(struct nvkm_fault *); #endif diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fault/tu104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/tu104.c new file mode 100644 index 000000000000..9c8a3adf99d7 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fault/tu104.c @@ -0,0 +1,167 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "priv.h" + +#include <core/memory.h> +#include <subdev/mmu.h> +#include <engine/fifo.h> + +#include <nvif/class.h> + +static void +tu104_fault_buffer_intr(struct nvkm_fault_buffer *buffer, bool enable) +{ + /*XXX: Earlier versions of RM touched the old regs on Turing, + * which don't appear to actually work anymore, but newer + * versions of RM don't appear to touch anything at all.. + */ +} + +static void +tu104_fault_buffer_fini(struct nvkm_fault_buffer *buffer) +{ + struct nvkm_device *device = buffer->fault->subdev.device; + const u32 foff = buffer->id * 0x20; + nvkm_mask(device, 0xb83010 + foff, 0x80000000, 0x00000000); +} + +static void +tu104_fault_buffer_init(struct nvkm_fault_buffer *buffer) +{ + struct nvkm_device *device = buffer->fault->subdev.device; + const u32 foff = buffer->id * 0x20; + + nvkm_mask(device, 0xb83010 + foff, 0xc0000000, 0x40000000); + nvkm_wr32(device, 0xb83004 + foff, upper_32_bits(buffer->addr)); + nvkm_wr32(device, 0xb83000 + foff, lower_32_bits(buffer->addr)); + nvkm_mask(device, 0xb83010 + foff, 0x80000000, 0x80000000); +} + +static void +tu104_fault_buffer_info(struct nvkm_fault_buffer *buffer) +{ + struct nvkm_device *device = buffer->fault->subdev.device; + const u32 foff = buffer->id * 0x20; + + nvkm_mask(device, 0xb83010 + foff, 0x40000000, 0x40000000); + + buffer->entries = nvkm_rd32(device, 0xb83010 + foff) & 0x000fffff; + buffer->get = 0xb83008 + foff; + buffer->put = 0xb8300c + foff; +} + +static void +tu104_fault_intr_fault(struct nvkm_fault *fault) +{ + struct nvkm_subdev *subdev = &fault->subdev; + struct nvkm_device *device = subdev->device; + struct nvkm_fault_data info; + const u32 addrlo = nvkm_rd32(device, 0xb83080); + const u32 addrhi = nvkm_rd32(device, 0xb83084); + const u32 info0 = nvkm_rd32(device, 0xb83088); + const u32 insthi = nvkm_rd32(device, 0xb8308c); + const u32 info1 = nvkm_rd32(device, 0xb83090); + + info.addr = ((u64)addrhi << 32) | addrlo; + info.inst = ((u64)insthi << 32) | (info0 & 0xfffff000); + info.time = 0; + info.engine = (info0 & 0x000000ff); + info.valid = (info1 & 0x80000000) >> 31; + info.gpc = (info1 & 0x1f000000) >> 24; + info.hub = (info1 & 0x00100000) >> 20; + info.access = (info1 & 0x000f0000) >> 16; + info.client = (info1 & 0x00007f00) >> 8; + info.reason = (info1 & 0x0000001f); + + nvkm_fifo_fault(device->fifo, &info); +} + +static void +tu104_fault_intr(struct nvkm_fault *fault) +{ + struct nvkm_subdev *subdev = &fault->subdev; + struct nvkm_device *device = subdev->device; + u32 stat = nvkm_rd32(device, 0xb83094); + + if (stat & 0x80000000) { + tu104_fault_intr_fault(fault); + nvkm_wr32(device, 0xb83094, 0x80000000); + stat &= ~0x80000000; + } + + if (stat & 0x00000200) { + if (fault->buffer[0]) { + nvkm_event_send(&fault->event, 1, 0, NULL, 0); + stat &= ~0x00000200; + } + } + + /*XXX: guess, can't confirm until we get fw... */ + if (stat & 0x00000100) { + if (fault->buffer[1]) { + nvkm_event_send(&fault->event, 1, 1, NULL, 0); + stat &= ~0x00000100; + } + } + + if (stat) { + nvkm_debug(subdev, "intr %08x\n", stat); + } +} + +static void +tu104_fault_fini(struct nvkm_fault *fault) +{ + nvkm_notify_put(&fault->nrpfb); + if (fault->buffer[0]) + fault->func->buffer.fini(fault->buffer[0]); + /*XXX: disable priv faults */ +} + +static void +tu104_fault_init(struct nvkm_fault *fault) +{ + /*XXX: enable priv faults */ + fault->func->buffer.init(fault->buffer[0]); + nvkm_notify_get(&fault->nrpfb); +} + +static const struct nvkm_fault_func +tu104_fault = { + .oneinit = gv100_fault_oneinit, + .init = tu104_fault_init, + .fini = tu104_fault_fini, + .intr = tu104_fault_intr, + .buffer.nr = 2, + .buffer.entry_size = 32, + .buffer.info = tu104_fault_buffer_info, + .buffer.init = tu104_fault_buffer_init, + .buffer.fini = tu104_fault_buffer_fini, + .buffer.intr = tu104_fault_buffer_intr, +}; + +int +tu104_fault_new(struct nvkm_device *device, int index, + struct nvkm_fault **pfault) +{ + return nvkm_fault_new_(&tu104_fault, device, index, pfault); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/base.c index 434d2fc5bb1c..b2bb5a3ccb02 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/base.c @@ -68,10 +68,13 @@ nvkm_fb_bios_memtype(struct nvkm_bios *bios) if (nvbios_M0203Em(bios, ramcfg, &ver, &hdr, &M0203E)) { switch (M0203E.type) { - case M0203E_TYPE_DDR2 : return NVKM_RAM_TYPE_DDR2; - case M0203E_TYPE_DDR3 : return NVKM_RAM_TYPE_DDR3; - case M0203E_TYPE_GDDR3: return NVKM_RAM_TYPE_GDDR3; - case M0203E_TYPE_GDDR5: return NVKM_RAM_TYPE_GDDR5; + case M0203E_TYPE_DDR2 : return NVKM_RAM_TYPE_DDR2; + case M0203E_TYPE_DDR3 : return NVKM_RAM_TYPE_DDR3; + case M0203E_TYPE_GDDR3 : return NVKM_RAM_TYPE_GDDR3; + case M0203E_TYPE_GDDR5 : return NVKM_RAM_TYPE_GDDR5; + case M0203E_TYPE_GDDR5X: return NVKM_RAM_TYPE_GDDR5X; + case M0203E_TYPE_GDDR6 : return NVKM_RAM_TYPE_GDDR6; + case M0203E_TYPE_HBM2 : return NVKM_RAM_TYPE_HBM2; default: nvkm_warn(subdev, "M0203E type %02x\n", M0203E.type); return NVKM_RAM_TYPE_UNKNOWN; diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ram.c b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ram.c index 24c7bd505731..b11867f682cb 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ram.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/fb/ram.c @@ -184,6 +184,9 @@ nvkm_ram_ctor(const struct nvkm_ram_func *func, struct nvkm_fb *fb, [NVKM_RAM_TYPE_GDDR3 ] = "GDDR3", [NVKM_RAM_TYPE_GDDR4 ] = "GDDR4", [NVKM_RAM_TYPE_GDDR5 ] = "GDDR5", + [NVKM_RAM_TYPE_GDDR5X ] = "GDDR5X", + [NVKM_RAM_TYPE_GDDR6 ] = "GDDR6", + [NVKM_RAM_TYPE_HBM2 ] = "HBM2", }; struct nvkm_subdev *subdev = &fb->subdev; int ret; diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv50.c b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv50.c index db48a1daca0c..02c4eb28cef4 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv50.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv50.c @@ -288,6 +288,19 @@ nv50_instobj_addr(struct nvkm_memory *memory) return nvkm_memory_addr(nv50_instobj(memory)->ram); } +static u64 +nv50_instobj_bar2(struct nvkm_memory *memory) +{ + struct nv50_instobj *iobj = nv50_instobj(memory); + u64 addr = ~0ULL; + if (nv50_instobj_acquire(&iobj->base.memory)) { + iobj->lru.next = NULL; /* Exclude from eviction. */ + addr = iobj->bar->addr; + } + nv50_instobj_release(&iobj->base.memory); + return addr; +} + static enum nvkm_memory_target nv50_instobj_target(struct nvkm_memory *memory) { @@ -325,8 +338,9 @@ static const struct nvkm_memory_func nv50_instobj_func = { .dtor = nv50_instobj_dtor, .target = nv50_instobj_target, - .size = nv50_instobj_size, + .bar2 = nv50_instobj_bar2, .addr = nv50_instobj_addr, + .size = nv50_instobj_size, .boot = nv50_instobj_boot, .acquire = nv50_instobj_acquire, .release = nv50_instobj_release, diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/Kbuild index 2befbe36dc28..f3b06329c338 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/Kbuild @@ -12,3 +12,4 @@ nvkm-y += nvkm/subdev/mc/gk104.o nvkm-y += nvkm/subdev/mc/gk20a.o nvkm-y += nvkm/subdev/mc/gp100.o nvkm-y += nvkm/subdev/mc/gp10b.o +nvkm-y += nvkm/subdev/mc/tu104.o diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/base.c index 09f669ac6630..0e57ab2a709f 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/base.c @@ -108,6 +108,9 @@ nvkm_mc_intr(struct nvkm_device *device, bool *handled) if (stat) nvkm_error(&mc->subdev, "intr %08x\n", stat); *handled = intr != 0; + + if (mc->func->intr_hack) + mc->func->intr_hack(mc, handled); } static u32 diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/priv.h index d9e3691d45b7..eb91a4cf452b 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/priv.h +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/priv.h @@ -26,6 +26,7 @@ struct nvkm_mc_func { void (*intr_mask)(struct nvkm_mc *, u32 mask, u32 stat); /* retrieve pending interrupt mask (NV_PMC_INTR) */ u32 (*intr_stat)(struct nvkm_mc *); + void (*intr_hack)(struct nvkm_mc *, bool *handled); const struct nvkm_mc_map *reset; void (*unk260)(struct nvkm_mc *, u32); }; diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mc/tu104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/tu104.c new file mode 100644 index 000000000000..b7165bd18999 --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mc/tu104.c @@ -0,0 +1,55 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "priv.h" + +static void +tu104_mc_intr_hack(struct nvkm_mc *mc, bool *handled) +{ + struct nvkm_device *device = mc->subdev.device; + u32 stat = nvkm_rd32(device, 0xb81010); + if (stat & 0x00000050) { + struct nvkm_subdev *subdev = + nvkm_device_subdev(device, NVKM_SUBDEV_FAULT); + nvkm_wr32(device, 0xb81010, stat & 0x00000050); + if (subdev) + nvkm_subdev_intr(subdev); + *handled = true; + } +} + +static const struct nvkm_mc_func +tu104_mc = { + .init = nv50_mc_init, + .intr = gp100_mc_intr, + .intr_unarm = gp100_mc_intr_unarm, + .intr_rearm = gp100_mc_intr_rearm, + .intr_mask = gp100_mc_intr_mask, + .intr_stat = gf100_mc_intr_stat, + .intr_hack = tu104_mc_intr_hack, + .reset = gk104_mc_reset, +}; + +int +tu104_mc_new(struct nvkm_device *device, int index, struct nvkm_mc **pmc) +{ + return gp100_mc_new_(&tu104_mc, device, index, pmc); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/Kbuild index 58a24e3a0598..8966180b36cc 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/Kbuild +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/Kbuild @@ -13,6 +13,7 @@ nvkm-y += nvkm/subdev/mmu/gm20b.o nvkm-y += nvkm/subdev/mmu/gp100.o nvkm-y += nvkm/subdev/mmu/gp10b.o nvkm-y += nvkm/subdev/mmu/gv100.o +nvkm-y += nvkm/subdev/mmu/tu104.o nvkm-y += nvkm/subdev/mmu/mem.o nvkm-y += nvkm/subdev/mmu/memnv04.o @@ -33,6 +34,7 @@ nvkm-y += nvkm/subdev/mmu/vmmgm20b.o nvkm-y += nvkm/subdev/mmu/vmmgp100.o nvkm-y += nvkm/subdev/mmu/vmmgp10b.o nvkm-y += nvkm/subdev/mmu/vmmgv100.o +nvkm-y += nvkm/subdev/mmu/vmmtu104.o nvkm-y += nvkm/subdev/mmu/umem.o nvkm-y += nvkm/subdev/mmu/ummu.o diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/tu104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/tu104.c new file mode 100644 index 000000000000..8e6f4096170d --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/tu104.c @@ -0,0 +1,43 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "mem.h" +#include "vmm.h" + +#include <core/option.h> + +#include <nvif/class.h> + +static const struct nvkm_mmu_func +tu104_mmu = { + .dma_bits = 47, + .mmu = {{ -1, -1, NVIF_CLASS_MMU_GF100}}, + .mem = {{ -1, 0, NVIF_CLASS_MEM_GF100}, gf100_mem_new, gf100_mem_map }, + .vmm = {{ -1, 0, NVIF_CLASS_VMM_GP100}, tu104_vmm_new }, + .kind = gm200_mmu_kind, + .kind_sys = true, +}; + +int +tu104_mmu_new(struct nvkm_device *device, int index, struct nvkm_mmu **pmmu) +{ + return nvkm_mmu_new_(&tu104_mmu, device, index, pmmu); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/uvmm.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/uvmm.c index 37b201b95f15..6889076097ec 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/uvmm.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/uvmm.c @@ -134,23 +134,10 @@ nvkm_uvmm_mthd_map(struct nvkm_uvmm *uvmm, void *argv, u32 argc) goto fail; } - if (vma->addr != addr) { - const u64 tail = vma->size + vma->addr - addr; - if (ret = -ENOMEM, !(vma = nvkm_vma_tail(vma, tail))) - goto fail; - vma->part = true; - nvkm_vmm_node_insert(vmm, vma); - } - - if (vma->size != size) { - const u64 tail = vma->size - size; - struct nvkm_vma *tmp; - if (ret = -ENOMEM, !(tmp = nvkm_vma_tail(vma, tail))) { - nvkm_vmm_unmap_region(vmm, vma); - goto fail; - } - tmp->part = true; - nvkm_vmm_node_insert(vmm, tmp); + vma = nvkm_vmm_node_split(vmm, vma, addr, size); + if (!vma) { + ret = -ENOMEM; + goto fail; } } vma->busy = true; diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c index 7459def78d50..6b87fff014b3 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c @@ -767,6 +767,20 @@ nvkm_vma_tail(struct nvkm_vma *vma, u64 tail) return new; } +static inline void +nvkm_vmm_free_remove(struct nvkm_vmm *vmm, struct nvkm_vma *vma) +{ + rb_erase(&vma->tree, &vmm->free); +} + +static inline void +nvkm_vmm_free_delete(struct nvkm_vmm *vmm, struct nvkm_vma *vma) +{ + nvkm_vmm_free_remove(vmm, vma); + list_del(&vma->head); + kfree(vma); +} + static void nvkm_vmm_free_insert(struct nvkm_vmm *vmm, struct nvkm_vma *vma) { @@ -795,7 +809,21 @@ nvkm_vmm_free_insert(struct nvkm_vmm *vmm, struct nvkm_vma *vma) rb_insert_color(&vma->tree, &vmm->free); } -void +static inline void +nvkm_vmm_node_remove(struct nvkm_vmm *vmm, struct nvkm_vma *vma) +{ + rb_erase(&vma->tree, &vmm->root); +} + +static inline void +nvkm_vmm_node_delete(struct nvkm_vmm *vmm, struct nvkm_vma *vma) +{ + nvkm_vmm_node_remove(vmm, vma); + list_del(&vma->head); + kfree(vma); +} + +static void nvkm_vmm_node_insert(struct nvkm_vmm *vmm, struct nvkm_vma *vma) { struct rb_node **ptr = &vmm->root.rb_node; @@ -834,6 +862,78 @@ nvkm_vmm_node_search(struct nvkm_vmm *vmm, u64 addr) return NULL; } +#define node(root, dir) (((root)->head.dir == &vmm->list) ? NULL : \ + list_entry((root)->head.dir, struct nvkm_vma, head)) + +static struct nvkm_vma * +nvkm_vmm_node_merge(struct nvkm_vmm *vmm, struct nvkm_vma *prev, + struct nvkm_vma *vma, struct nvkm_vma *next, u64 size) +{ + if (next) { + if (vma->size == size) { + vma->size += next->size; + nvkm_vmm_node_delete(vmm, next); + if (prev) { + prev->size += vma->size; + nvkm_vmm_node_delete(vmm, vma); + return prev; + } + return vma; + } + BUG_ON(prev); + + nvkm_vmm_node_remove(vmm, next); + vma->size -= size; + next->addr -= size; + next->size += size; + nvkm_vmm_node_insert(vmm, next); + return next; + } + + if (prev) { + if (vma->size != size) { + nvkm_vmm_node_remove(vmm, vma); + prev->size += size; + vma->addr += size; + vma->size -= size; + nvkm_vmm_node_insert(vmm, vma); + } else { + prev->size += vma->size; + nvkm_vmm_node_delete(vmm, vma); + } + return prev; + } + + return vma; +} + +struct nvkm_vma * +nvkm_vmm_node_split(struct nvkm_vmm *vmm, + struct nvkm_vma *vma, u64 addr, u64 size) +{ + struct nvkm_vma *prev = NULL; + + if (vma->addr != addr) { + prev = vma; + if (!(vma = nvkm_vma_tail(vma, vma->size + vma->addr - addr))) + return NULL; + vma->part = true; + nvkm_vmm_node_insert(vmm, vma); + } + + if (vma->size != size) { + struct nvkm_vma *tmp; + if (!(tmp = nvkm_vma_tail(vma, vma->size - size))) { + nvkm_vmm_node_merge(vmm, prev, vma, NULL, vma->size); + return NULL; + } + tmp->part = true; + nvkm_vmm_node_insert(vmm, tmp); + } + + return vma; +} + static void nvkm_vmm_dtor(struct nvkm_vmm *vmm) { @@ -954,37 +1054,20 @@ nvkm_vmm_new_(const struct nvkm_vmm_func *func, struct nvkm_mmu *mmu, return nvkm_vmm_ctor(func, mmu, hdr, addr, size, key, name, *pvmm); } -#define node(root, dir) ((root)->head.dir == &vmm->list) ? NULL : \ - list_entry((root)->head.dir, struct nvkm_vma, head) - void nvkm_vmm_unmap_region(struct nvkm_vmm *vmm, struct nvkm_vma *vma) { - struct nvkm_vma *next; + struct nvkm_vma *next = node(vma, next); + struct nvkm_vma *prev = NULL; nvkm_memory_tags_put(vma->memory, vmm->mmu->subdev.device, &vma->tags); nvkm_memory_unref(&vma->memory); - if (vma->part) { - struct nvkm_vma *prev = node(vma, prev); - if (!prev->memory) { - prev->size += vma->size; - rb_erase(&vma->tree, &vmm->root); - list_del(&vma->head); - kfree(vma); - vma = prev; - } - } - - next = node(vma, next); - if (next && next->part) { - if (!next->memory) { - vma->size += next->size; - rb_erase(&next->tree, &vmm->root); - list_del(&next->head); - kfree(next); - } - } + if (!vma->part || ((prev = node(vma, prev)), prev->memory)) + prev = NULL; + if (!next->part || next->memory) + next = NULL; + nvkm_vmm_node_merge(vmm, prev, vma, next, vma->size); } void @@ -1163,18 +1246,14 @@ nvkm_vmm_put_region(struct nvkm_vmm *vmm, struct nvkm_vma *vma) struct nvkm_vma *prev, *next; if ((prev = node(vma, prev)) && !prev->used) { - rb_erase(&prev->tree, &vmm->free); - list_del(&prev->head); vma->addr = prev->addr; vma->size += prev->size; - kfree(prev); + nvkm_vmm_free_delete(vmm, prev); } if ((next = node(vma, next)) && !next->used) { - rb_erase(&next->tree, &vmm->free); - list_del(&next->head); vma->size += next->size; - kfree(next); + nvkm_vmm_free_delete(vmm, next); } nvkm_vmm_free_insert(vmm, vma); @@ -1250,7 +1329,7 @@ nvkm_vmm_put_locked(struct nvkm_vmm *vmm, struct nvkm_vma *vma) } /* Remove VMA from the list of allocated nodes. */ - rb_erase(&vma->tree, &vmm->root); + nvkm_vmm_node_remove(vmm, vma); /* Merge VMA back into the free list. */ vma->page = NVKM_VMA_PAGE_NONE; @@ -1357,7 +1436,7 @@ nvkm_vmm_get_locked(struct nvkm_vmm *vmm, bool getref, bool mapref, bool sparse, tail = ALIGN_DOWN(tail, vmm->func->page_block); if (addr <= tail && tail - addr >= size) { - rb_erase(&this->tree, &vmm->free); + nvkm_vmm_free_remove(vmm, this); vma = this; break; } diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h index 1a3b0a3724ca..42ad326521a3 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h @@ -157,6 +157,8 @@ int nvkm_vmm_ctor(const struct nvkm_vmm_func *, struct nvkm_mmu *, u32 pd_header, u64 addr, u64 size, struct lock_class_key *, const char *name, struct nvkm_vmm *); struct nvkm_vma *nvkm_vmm_node_search(struct nvkm_vmm *, u64 addr); +struct nvkm_vma *nvkm_vmm_node_split(struct nvkm_vmm *, struct nvkm_vma *, + u64 addr, u64 size); int nvkm_vmm_get_locked(struct nvkm_vmm *, bool getref, bool mapref, bool sparse, u8 page, u8 align, u64 size, struct nvkm_vma **pvma); @@ -165,7 +167,6 @@ void nvkm_vmm_unmap_locked(struct nvkm_vmm *, struct nvkm_vma *); void nvkm_vmm_unmap_region(struct nvkm_vmm *vmm, struct nvkm_vma *vma); struct nvkm_vma *nvkm_vma_tail(struct nvkm_vma *, u64 tail); -void nvkm_vmm_node_insert(struct nvkm_vmm *, struct nvkm_vma *); int nv04_vmm_new_(const struct nvkm_vmm_func *, struct nvkm_mmu *, u32, u64, u64, void *, u32, struct lock_class_key *, @@ -200,6 +201,8 @@ int gp100_vmm_join(struct nvkm_vmm *, struct nvkm_memory *); int gp100_vmm_valid(struct nvkm_vmm *, void *, u32, struct nvkm_vmm_map *); void gp100_vmm_flush(struct nvkm_vmm *, int); +int gv100_vmm_join(struct nvkm_vmm *, struct nvkm_memory *); + int nv04_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32, struct lock_class_key *, const char *, struct nvkm_vmm **); int nv41_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32, @@ -239,6 +242,9 @@ int gp10b_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32, int gv100_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32, struct lock_class_key *, const char *, struct nvkm_vmm **); +int tu104_vmm_new(struct nvkm_mmu *, u64, u64, void *, u32, + struct lock_class_key *, const char *, + struct nvkm_vmm **); #define VMM_PRINT(l,v,p,f,a...) do { \ struct nvkm_vmm *_vmm = (v); \ diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmtu104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmtu104.c new file mode 100644 index 000000000000..adaadd92110f --- /dev/null +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmmtu104.c @@ -0,0 +1,77 @@ +/* + * Copyright 2018 Red Hat Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + */ +#include "vmm.h" + +#include <subdev/timer.h> + +static void +tu104_vmm_flush(struct nvkm_vmm *vmm, int depth) +{ + struct nvkm_subdev *subdev = &vmm->mmu->subdev; + struct nvkm_device *device = subdev->device; + u32 type = depth << 24; /*XXX: not confirmed */ + + type = 0x00000001; /* PAGE_ALL */ + if (atomic_read(&vmm->engref[NVKM_SUBDEV_BAR])) + type |= 0x00000004; /* HUB_ONLY */ + + mutex_lock(&subdev->mutex); + + nvkm_wr32(device, 0xb830a0, vmm->pd->pt[0]->addr >> 8); + nvkm_wr32(device, 0xb830a4, 0x00000000); + nvkm_wr32(device, 0x100e68, 0x00000000); + nvkm_wr32(device, 0xb830b0, 0x80000000 | type); + + nvkm_msec(device, 2000, + if (!(nvkm_rd32(device, 0xb830b0) & 0x80000000)) + break; + ); + + mutex_unlock(&subdev->mutex); +} + +static const struct nvkm_vmm_func +tu104_vmm = { + .join = gv100_vmm_join, + .part = gf100_vmm_part, + .aper = gf100_vmm_aper, + .valid = gp100_vmm_valid, + .flush = tu104_vmm_flush, + .page = { + { 47, &gp100_vmm_desc_16[4], NVKM_VMM_PAGE_Sxxx }, + { 38, &gp100_vmm_desc_16[3], NVKM_VMM_PAGE_Sxxx }, + { 29, &gp100_vmm_desc_16[2], NVKM_VMM_PAGE_Sxxx }, + { 21, &gp100_vmm_desc_16[1], NVKM_VMM_PAGE_SVxC }, + { 16, &gp100_vmm_desc_16[0], NVKM_VMM_PAGE_SVxC }, + { 12, &gp100_vmm_desc_12[0], NVKM_VMM_PAGE_SVHx }, + {} + } +}; + +int +tu104_vmm_new(struct nvkm_mmu *mmu, u64 addr, u64 size, + void *argv, u32 argc, struct lock_class_key *key, + const char *name, struct nvkm_vmm **pvmm) +{ + return nv04_vmm_new_(&tu104_vmm, mmu, 0, addr, size, + argv, argc, key, name, pvmm); +} diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/gp102.c b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/gp102.c index 1f7a3c1a7f50..84a2f243ed9b 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/gp102.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/gp102.c @@ -59,10 +59,10 @@ gp102_run_secure_scrub(struct nvkm_secboot *sb) nvkm_debug(subdev, "running VPR scrubber binary on NVDEC...\n"); - engine = nvkm_engine_ref(&device->nvdec->engine); + engine = nvkm_engine_ref(&device->nvdec[0]->engine); if (IS_ERR(engine)) return PTR_ERR(engine); - falcon = device->nvdec->falcon; + falcon = device->nvdec[0]->falcon; nvkm_falcon_get(falcon, &sb->subdev); diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/timer/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/timer/base.c index 36de23d12ae4..dd922033628c 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/timer/base.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/timer/base.c @@ -23,6 +23,42 @@ */ #include "priv.h" +s64 +nvkm_timer_wait_test(struct nvkm_timer_wait *wait) +{ + struct nvkm_subdev *subdev = &wait->tmr->subdev; + u64 time = nvkm_timer_read(wait->tmr); + + if (wait->reads == 0) { + wait->time0 = time; + wait->time1 = time; + } + + if (wait->time1 == time) { + if (wait->reads++ == 16) { + nvkm_fatal(subdev, "stalled at %016llx\n", time); + return -ETIMEDOUT; + } + } else { + wait->time1 = time; + wait->reads = 1; + } + + if (wait->time1 - wait->time0 > wait->limit) + return -ETIMEDOUT; + + return wait->time1 - wait->time0; +} + +void +nvkm_timer_wait_init(struct nvkm_device *device, u64 nsec, + struct nvkm_timer_wait *wait) +{ + wait->tmr = device->timer; + wait->limit = nsec; + wait->reads = 0; +} + u64 nvkm_timer_read(struct nvkm_timer *tmr) { diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/top/gk104.c b/drivers/gpu/drm/nouveau/nvkm/subdev/top/gk104.c index 4f1f3e890650..39081eadfd84 100644 --- a/drivers/gpu/drm/nouveau/nvkm/subdev/top/gk104.c +++ b/drivers/gpu/drm/nouveau/nvkm/subdev/top/gk104.c @@ -86,7 +86,7 @@ gk104_top_oneinit(struct nvkm_top *top) case 0x0000000d: A_(SEC2 ); break; case 0x0000000e: B_(NVENC ); break; case 0x0000000f: A_(NVENC1); break; - case 0x00000010: A_(NVDEC ); break; + case 0x00000010: B_(NVDEC ); break; case 0x00000013: B_(CE ); break; break; default: diff --git a/drivers/gpu/drm/omapdrm/displays/panel-dpi.c b/drivers/gpu/drm/omapdrm/displays/panel-dpi.c index 1f8161b041be..465120809eb3 100644 --- a/drivers/gpu/drm/omapdrm/displays/panel-dpi.c +++ b/drivers/gpu/drm/omapdrm/displays/panel-dpi.c @@ -177,6 +177,7 @@ static int panel_dpi_probe(struct platform_device *pdev) dssdev->type = OMAP_DISPLAY_TYPE_DPI; dssdev->owner = THIS_MODULE; dssdev->of_ports = BIT(0); + drm_bus_flags_from_videomode(&ddata->vm, &dssdev->bus_flags); omapdss_display_init(dssdev); omapdss_device_register(dssdev); diff --git a/drivers/gpu/drm/omapdrm/dss/dsi.c b/drivers/gpu/drm/omapdrm/dss/dsi.c index 0a485c5b982e..00a9c2ab9e6c 100644 --- a/drivers/gpu/drm/omapdrm/dss/dsi.c +++ b/drivers/gpu/drm/omapdrm/dss/dsi.c @@ -5418,9 +5418,15 @@ static int dsi_probe(struct platform_device *pdev) dsi->num_lanes_supported = 3; } + r = of_platform_populate(dev->of_node, NULL, NULL, dev); + if (r) { + DSSERR("Failed to populate DSI child devices: %d\n", r); + goto err_pm_disable; + } + r = dsi_init_output(dsi); if (r) - goto err_pm_disable; + goto err_of_depopulate; r = dsi_probe_of(dsi); if (r) { @@ -5428,22 +5434,16 @@ static int dsi_probe(struct platform_device *pdev) goto err_uninit_output; } - r = of_platform_populate(dev->of_node, NULL, NULL, dev); - if (r) { - DSSERR("Failed to populate DSI child devices: %d\n", r); - goto err_uninit_output; - } - r = component_add(&pdev->dev, &dsi_component_ops); if (r) - goto err_of_depopulate; + goto err_uninit_output; return 0; -err_of_depopulate: - of_platform_depopulate(dev); err_uninit_output: dsi_uninit_output(dsi); +err_of_depopulate: + of_platform_depopulate(dev); err_pm_disable: pm_runtime_disable(dev); return r; diff --git a/drivers/gpu/drm/omapdrm/dss/omapdss.h b/drivers/gpu/drm/omapdrm/dss/omapdss.h index 1f698a95a94a..33e15cb77efa 100644 --- a/drivers/gpu/drm/omapdrm/dss/omapdss.h +++ b/drivers/gpu/drm/omapdrm/dss/omapdss.h @@ -432,7 +432,7 @@ struct omap_dss_device { const struct omap_dss_driver *driver; const struct omap_dss_device_ops *ops; unsigned long ops_flags; - unsigned long bus_flags; + u32 bus_flags; /* helper variable for driver suspend/resume */ bool activate_after_resume; diff --git a/drivers/gpu/drm/omapdrm/omap_encoder.c b/drivers/gpu/drm/omapdrm/omap_encoder.c index 4566e0a75cb8..148b6b20274f 100644 --- a/drivers/gpu/drm/omapdrm/omap_encoder.c +++ b/drivers/gpu/drm/omapdrm/omap_encoder.c @@ -52,17 +52,44 @@ static const struct drm_encoder_funcs omap_encoder_funcs = { .destroy = omap_encoder_destroy, }; +static void omap_encoder_hdmi_mode_set(struct drm_encoder *encoder, + struct drm_display_mode *adjusted_mode) +{ + struct drm_device *dev = encoder->dev; + struct omap_encoder *omap_encoder = to_omap_encoder(encoder); + struct omap_dss_device *dssdev = omap_encoder->output; + struct drm_connector *connector; + bool hdmi_mode; + + hdmi_mode = false; + list_for_each_entry(connector, &dev->mode_config.connector_list, head) { + if (connector->encoder == encoder) { + hdmi_mode = omap_connector_get_hdmi_mode(connector); + break; + } + } + + if (dssdev->ops->hdmi.set_hdmi_mode) + dssdev->ops->hdmi.set_hdmi_mode(dssdev, hdmi_mode); + + if (hdmi_mode && dssdev->ops->hdmi.set_infoframe) { + struct hdmi_avi_infoframe avi; + int r; + + r = drm_hdmi_avi_infoframe_from_display_mode(&avi, connector, + adjusted_mode); + if (r == 0) + dssdev->ops->hdmi.set_infoframe(dssdev, &avi); + } +} + static void omap_encoder_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) { - struct drm_device *dev = encoder->dev; struct omap_encoder *omap_encoder = to_omap_encoder(encoder); - struct drm_connector *connector; struct omap_dss_device *dssdev; struct videomode vm = { 0 }; - bool hdmi_mode; - int r; drm_display_mode_to_videomode(adjusted_mode, &vm); @@ -112,27 +139,8 @@ static void omap_encoder_mode_set(struct drm_encoder *encoder, } /* Set the HDMI mode and HDMI infoframe if applicable. */ - hdmi_mode = false; - list_for_each_entry(connector, &dev->mode_config.connector_list, head) { - if (connector->encoder == encoder) { - hdmi_mode = omap_connector_get_hdmi_mode(connector); - break; - } - } - - dssdev = omap_encoder->output; - - if (dssdev->ops->hdmi.set_hdmi_mode) - dssdev->ops->hdmi.set_hdmi_mode(dssdev, hdmi_mode); - - if (hdmi_mode && dssdev->ops->hdmi.set_infoframe) { - struct hdmi_avi_infoframe avi; - - r = drm_hdmi_avi_infoframe_from_display_mode(&avi, connector, - adjusted_mode); - if (r == 0) - dssdev->ops->hdmi.set_infoframe(dssdev, &avi); - } + if (omap_encoder->output->output_type == OMAP_DISPLAY_TYPE_HDMI) + omap_encoder_hdmi_mode_set(encoder, adjusted_mode); } static void omap_encoder_disable(struct drm_encoder *encoder) diff --git a/drivers/gpu/drm/qxl/qxl_ioctl.c b/drivers/gpu/drm/qxl/qxl_ioctl.c index 6e828158bcb0..d410e2925162 100644 --- a/drivers/gpu/drm/qxl/qxl_ioctl.c +++ b/drivers/gpu/drm/qxl/qxl_ioctl.c @@ -163,8 +163,7 @@ static int qxl_process_single_command(struct qxl_device *qdev, if (cmd->command_size > PAGE_SIZE - sizeof(union qxl_release_info)) return -EINVAL; - if (!access_ok(VERIFY_READ, - u64_to_user_ptr(cmd->command), + if (!access_ok(u64_to_user_ptr(cmd->command), cmd->command_size)) return -EFAULT; diff --git a/drivers/gpu/drm/qxl/qxl_release.c b/drivers/gpu/drm/qxl/qxl_release.c index 0a693fede05e..30f85f0130cb 100644 --- a/drivers/gpu/drm/qxl/qxl_release.c +++ b/drivers/gpu/drm/qxl/qxl_release.c @@ -217,7 +217,7 @@ int qxl_release_list_add(struct qxl_release *release, struct qxl_bo *bo) qxl_bo_ref(bo); entry->tv.bo = &bo->tbo; - entry->tv.shared = false; + entry->tv.num_shared = 0; list_add_tail(&entry->tv.head, &release->bos); return 0; } diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c index 1ae31dbc61c6..f43305329939 100644 --- a/drivers/gpu/drm/radeon/radeon_cs.c +++ b/drivers/gpu/drm/radeon/radeon_cs.c @@ -178,7 +178,7 @@ static int radeon_cs_parser_relocs(struct radeon_cs_parser *p) } p->relocs[i].tv.bo = &p->relocs[i].robj->tbo; - p->relocs[i].tv.shared = !r->write_domain; + p->relocs[i].tv.num_shared = !r->write_domain; radeon_cs_buckets_add(&buckets, &p->relocs[i].tv.head, priority); @@ -253,7 +253,7 @@ static int radeon_cs_sync_rings(struct radeon_cs_parser *p) resv = reloc->robj->tbo.resv; r = radeon_sync_resv(p->rdev, &p->ib.sync, resv, - reloc->tv.shared); + reloc->tv.num_shared); if (r) return r; } diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c index 27d8e7dd2d06..44617dec8183 100644 --- a/drivers/gpu/drm/radeon/radeon_gem.c +++ b/drivers/gpu/drm/radeon/radeon_gem.c @@ -552,7 +552,7 @@ static void radeon_gem_va_update_vm(struct radeon_device *rdev, INIT_LIST_HEAD(&list); tv.bo = &bo_va->bo->tbo; - tv.shared = true; + tv.num_shared = 1; list_add(&tv.head, &list); vm_bos = radeon_vm_get_bos(rdev, bo_va->vm, &list); diff --git a/drivers/gpu/drm/radeon/radeon_mn.c b/drivers/gpu/drm/radeon/radeon_mn.c index f8b35df44c60..b3019505065a 100644 --- a/drivers/gpu/drm/radeon/radeon_mn.c +++ b/drivers/gpu/drm/radeon/radeon_mn.c @@ -119,40 +119,38 @@ static void radeon_mn_release(struct mmu_notifier *mn, * unmap them by move them into system domain again. */ static int radeon_mn_invalidate_range_start(struct mmu_notifier *mn, - struct mm_struct *mm, - unsigned long start, - unsigned long end, - bool blockable) + const struct mmu_notifier_range *range) { struct radeon_mn *rmn = container_of(mn, struct radeon_mn, mn); struct ttm_operation_ctx ctx = { false, false }; struct interval_tree_node *it; + unsigned long end; int ret = 0; /* notification is exclusive, but interval is inclusive */ - end -= 1; + end = range->end - 1; /* TODO we should be able to split locking for interval tree and * the tear down. */ - if (blockable) + if (range->blockable) mutex_lock(&rmn->lock); else if (!mutex_trylock(&rmn->lock)) return -EAGAIN; - it = interval_tree_iter_first(&rmn->objects, start, end); + it = interval_tree_iter_first(&rmn->objects, range->start, end); while (it) { struct radeon_mn_node *node; struct radeon_bo *bo; long r; - if (!blockable) { + if (!range->blockable) { ret = -EAGAIN; goto out_unlock; } node = container_of(it, struct radeon_mn_node, it); - it = interval_tree_iter_next(it, start, end); + it = interval_tree_iter_next(it, range->start, end); list_for_each_entry(bo, &node->bos, mn_list) { diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c index fed11ece0de6..0d374211661c 100644 --- a/drivers/gpu/drm/radeon/radeon_vm.c +++ b/drivers/gpu/drm/radeon/radeon_vm.c @@ -142,7 +142,7 @@ struct radeon_bo_list *radeon_vm_get_bos(struct radeon_device *rdev, list[0].preferred_domains = RADEON_GEM_DOMAIN_VRAM; list[0].allowed_domains = RADEON_GEM_DOMAIN_VRAM; list[0].tv.bo = &vm->page_directory->tbo; - list[0].tv.shared = true; + list[0].tv.num_shared = 1; list[0].tiling_flags = 0; list_add(&list[0].tv.head, head); @@ -154,7 +154,7 @@ struct radeon_bo_list *radeon_vm_get_bos(struct radeon_device *rdev, list[idx].preferred_domains = RADEON_GEM_DOMAIN_VRAM; list[idx].allowed_domains = RADEON_GEM_DOMAIN_VRAM; list[idx].tv.bo = &list[idx].robj->tbo; - list[idx].tv.shared = true; + list[idx].tv.num_shared = 1; list[idx].tiling_flags = 0; list_add(&list[idx++].tv.head, head); } @@ -946,7 +946,7 @@ int radeon_vm_bo_update(struct radeon_device *rdev, bo_va->flags &= ~RADEON_VM_PAGE_WRITEABLE; if (mem) { - addr = mem->start << PAGE_SHIFT; + addr = (u64)mem->start << PAGE_SHIFT; if (mem->mem_type != TTM_PL_SYSTEM) { bo_va->flags |= RADEON_VM_PAGE_VALID; } diff --git a/drivers/gpu/drm/rcar-du/rcar_du_group.c b/drivers/gpu/drm/rcar-du/rcar_du_group.c index d85f0a1c1581..cebf313c6e1f 100644 --- a/drivers/gpu/drm/rcar-du/rcar_du_group.c +++ b/drivers/gpu/drm/rcar-du/rcar_du_group.c @@ -202,10 +202,25 @@ void rcar_du_group_put(struct rcar_du_group *rgrp) static void __rcar_du_group_start_stop(struct rcar_du_group *rgrp, bool start) { - struct rcar_du_crtc *rcrtc = &rgrp->dev->crtcs[rgrp->index * 2]; + struct rcar_du_device *rcdu = rgrp->dev; + + /* + * Group start/stop is controlled by the DRES and DEN bits of DSYSR0 + * for the first group and DSYSR2 for the second group. On most DU + * instances, this maps to the first CRTC of the group, and we can just + * use rcar_du_crtc_dsysr_clr_set() to access the correct DSYSR. On + * M3-N, however, DU2 doesn't exist, but DSYSR2 does. We thus need to + * access the register directly using group read/write. + */ + if (rcdu->info->channels_mask & BIT(rgrp->index * 2)) { + struct rcar_du_crtc *rcrtc = &rgrp->dev->crtcs[rgrp->index * 2]; - rcar_du_crtc_dsysr_clr_set(rcrtc, DSYSR_DRES | DSYSR_DEN, - start ? DSYSR_DEN : DSYSR_DRES); + rcar_du_crtc_dsysr_clr_set(rcrtc, DSYSR_DRES | DSYSR_DEN, + start ? DSYSR_DEN : DSYSR_DRES); + } else { + rcar_du_group_write(rgrp, DSYSR, + start ? DSYSR_DEN : DSYSR_DRES); + } } void rcar_du_group_start_stop(struct rcar_du_group *rgrp, bool start) diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c index 37f9a3b651ab..be6c2573039a 100644 --- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c +++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c @@ -448,11 +448,6 @@ static int rockchip_drm_platform_remove(struct platform_device *pdev) return 0; } -static void rockchip_drm_platform_shutdown(struct platform_device *pdev) -{ - rockchip_drm_platform_remove(pdev); -} - static const struct of_device_id rockchip_drm_dt_ids[] = { { .compatible = "rockchip,display-subsystem", }, { /* sentinel */ }, @@ -462,7 +457,6 @@ MODULE_DEVICE_TABLE(of, rockchip_drm_dt_ids); static struct platform_driver rockchip_drm_platform_driver = { .probe = rockchip_drm_platform_probe, .remove = rockchip_drm_platform_remove, - .shutdown = rockchip_drm_platform_shutdown, .driver = { .name = "rockchip-drm", .of_match_table = rockchip_drm_dt_ids, diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c index 18ebbb05762e..dbb69063b3d5 100644 --- a/drivers/gpu/drm/scheduler/sched_main.c +++ b/drivers/gpu/drm/scheduler/sched_main.c @@ -60,6 +60,8 @@ static void drm_sched_process_job(struct dma_fence *f, struct dma_fence_cb *cb); +static void drm_sched_expel_job_unlocked(struct drm_sched_job *s_job); + /** * drm_sched_rq_init - initialize a given run queue struct * @@ -209,6 +211,62 @@ void drm_sched_fault(struct drm_gpu_scheduler *sched) } EXPORT_SYMBOL(drm_sched_fault); +/** + * drm_sched_suspend_timeout - Suspend scheduler job timeout + * + * @sched: scheduler instance for which to suspend the timeout + * + * Suspend the delayed work timeout for the scheduler. This is done by + * modifying the delayed work timeout to an arbitrary large value, + * MAX_SCHEDULE_TIMEOUT in this case. Note that this function can be + * called from an IRQ context. + * + * Returns the timeout remaining + * + */ +unsigned long drm_sched_suspend_timeout(struct drm_gpu_scheduler *sched) +{ + unsigned long sched_timeout, now = jiffies; + + sched_timeout = sched->work_tdr.timer.expires; + + /* + * Modify the timeout to an arbitrarily large value. This also prevents + * the timeout to be restarted when new submissions arrive + */ + if (mod_delayed_work(system_wq, &sched->work_tdr, MAX_SCHEDULE_TIMEOUT) + && time_after(sched_timeout, now)) + return sched_timeout - now; + else + return sched->timeout; +} +EXPORT_SYMBOL(drm_sched_suspend_timeout); + +/** + * drm_sched_resume_timeout - Resume scheduler job timeout + * + * @sched: scheduler instance for which to resume the timeout + * @remaining: remaining timeout + * + * Resume the delayed work timeout for the scheduler. Note that + * this function can be called from an IRQ context. + */ +void drm_sched_resume_timeout(struct drm_gpu_scheduler *sched, + unsigned long remaining) +{ + unsigned long flags; + + spin_lock_irqsave(&sched->job_list_lock, flags); + + if (list_empty(&sched->ring_mirror_list)) + cancel_delayed_work(&sched->work_tdr); + else + mod_delayed_work(system_wq, &sched->work_tdr, remaining); + + spin_unlock_irqrestore(&sched->job_list_lock, flags); +} +EXPORT_SYMBOL(drm_sched_resume_timeout); + /* job_finish is called after hw fence signaled */ static void drm_sched_job_finish(struct work_struct *work) @@ -216,6 +274,7 @@ static void drm_sched_job_finish(struct work_struct *work) struct drm_sched_job *s_job = container_of(work, struct drm_sched_job, finish_work); struct drm_gpu_scheduler *sched = s_job->sched; + unsigned long flags; /* * Canceling the timeout without removing our job from the ring mirror @@ -226,12 +285,12 @@ static void drm_sched_job_finish(struct work_struct *work) */ cancel_delayed_work_sync(&sched->work_tdr); - spin_lock(&sched->job_list_lock); + spin_lock_irqsave(&sched->job_list_lock, flags); /* remove job from ring_mirror_list */ - list_del(&s_job->node); + list_del_init(&s_job->node); /* queue TDR for next job */ drm_sched_start_timeout(sched); - spin_unlock(&sched->job_list_lock); + spin_unlock_irqrestore(&sched->job_list_lock, flags); sched->ops->free_job(s_job); } @@ -247,56 +306,33 @@ static void drm_sched_job_finish_cb(struct dma_fence *f, static void drm_sched_job_begin(struct drm_sched_job *s_job) { struct drm_gpu_scheduler *sched = s_job->sched; + unsigned long flags; dma_fence_add_callback(&s_job->s_fence->finished, &s_job->finish_cb, drm_sched_job_finish_cb); - spin_lock(&sched->job_list_lock); + spin_lock_irqsave(&sched->job_list_lock, flags); list_add_tail(&s_job->node, &sched->ring_mirror_list); drm_sched_start_timeout(sched); - spin_unlock(&sched->job_list_lock); + spin_unlock_irqrestore(&sched->job_list_lock, flags); } static void drm_sched_job_timedout(struct work_struct *work) { struct drm_gpu_scheduler *sched; struct drm_sched_job *job; - int r; + unsigned long flags; sched = container_of(work, struct drm_gpu_scheduler, work_tdr.work); - - spin_lock(&sched->job_list_lock); - list_for_each_entry_reverse(job, &sched->ring_mirror_list, node) { - struct drm_sched_fence *fence = job->s_fence; - - if (!dma_fence_remove_callback(fence->parent, &fence->cb)) - goto already_signaled; - } - job = list_first_entry_or_null(&sched->ring_mirror_list, struct drm_sched_job, node); - spin_unlock(&sched->job_list_lock); if (job) - sched->ops->timedout_job(job); - - spin_lock(&sched->job_list_lock); - list_for_each_entry(job, &sched->ring_mirror_list, node) { - struct drm_sched_fence *fence = job->s_fence; + job->sched->ops->timedout_job(job); - if (!fence->parent || !list_empty(&fence->cb.node)) - continue; - - r = dma_fence_add_callback(fence->parent, &fence->cb, - drm_sched_process_job); - if (r) - drm_sched_process_job(fence->parent, &fence->cb); - -already_signaled: - ; - } + spin_lock_irqsave(&sched->job_list_lock, flags); drm_sched_start_timeout(sched); - spin_unlock(&sched->job_list_lock); + spin_unlock_irqrestore(&sched->job_list_lock, flags); } /** @@ -310,9 +346,10 @@ void drm_sched_hw_job_reset(struct drm_gpu_scheduler *sched, struct drm_sched_jo { struct drm_sched_job *s_job; struct drm_sched_entity *entity, *tmp; + unsigned long flags; int i; - spin_lock(&sched->job_list_lock); + spin_lock_irqsave(&sched->job_list_lock, flags); list_for_each_entry_reverse(s_job, &sched->ring_mirror_list, node) { if (s_job->s_fence->parent && dma_fence_remove_callback(s_job->s_fence->parent, @@ -322,7 +359,7 @@ void drm_sched_hw_job_reset(struct drm_gpu_scheduler *sched, struct drm_sched_jo atomic_dec(&sched->hw_rq_count); } } - spin_unlock(&sched->job_list_lock); + spin_unlock_irqrestore(&sched->job_list_lock, flags); if (bad && bad->s_priority != DRM_SCHED_PRIORITY_KERNEL) { atomic_inc(&bad->karma); @@ -360,9 +397,10 @@ void drm_sched_job_recovery(struct drm_gpu_scheduler *sched) { struct drm_sched_job *s_job, *tmp; bool found_guilty = false; + unsigned long flags; int r; - spin_lock(&sched->job_list_lock); + spin_lock_irqsave(&sched->job_list_lock, flags); list_for_each_entry_safe(s_job, tmp, &sched->ring_mirror_list, node) { struct drm_sched_fence *s_fence = s_job->s_fence; struct dma_fence *fence; @@ -376,7 +414,7 @@ void drm_sched_job_recovery(struct drm_gpu_scheduler *sched) if (found_guilty && s_job->s_fence->scheduled.context == guilty_context) dma_fence_set_error(&s_fence->finished, -ECANCELED); - spin_unlock(&sched->job_list_lock); + spin_unlock_irqrestore(&sched->job_list_lock, flags); fence = sched->ops->run_job(s_job); atomic_inc(&sched->hw_rq_count); @@ -391,12 +429,14 @@ void drm_sched_job_recovery(struct drm_gpu_scheduler *sched) r); dma_fence_put(fence); } else { + if (s_fence->finished.error < 0) + drm_sched_expel_job_unlocked(s_job); drm_sched_process_job(NULL, &s_fence->cb); } - spin_lock(&sched->job_list_lock); + spin_lock_irqsave(&sched->job_list_lock, flags); } drm_sched_start_timeout(sched); - spin_unlock(&sched->job_list_lock); + spin_unlock_irqrestore(&sched->job_list_lock, flags); } EXPORT_SYMBOL(drm_sched_job_recovery); @@ -595,6 +635,8 @@ static int drm_sched_main(void *param) r); dma_fence_put(fence); } else { + if (s_fence->finished.error < 0) + drm_sched_expel_job_unlocked(sched_job); drm_sched_process_job(NULL, &s_fence->cb); } @@ -603,6 +645,15 @@ static int drm_sched_main(void *param) return 0; } +static void drm_sched_expel_job_unlocked(struct drm_sched_job *s_job) +{ + struct drm_gpu_scheduler *sched = s_job->sched; + + spin_lock(&sched->job_list_lock); + list_del_init(&s_job->node); + spin_unlock(&sched->job_list_lock); +} + /** * drm_sched_init - Init a gpu scheduler instance * @@ -622,7 +673,7 @@ int drm_sched_init(struct drm_gpu_scheduler *sched, long timeout, const char *name) { - int i; + int i, ret; sched->ops = ops; sched->hw_submission_limit = hw_submission; sched->name = name; @@ -643,8 +694,10 @@ int drm_sched_init(struct drm_gpu_scheduler *sched, /* Each scheduler will run on a seperate kernel thread */ sched->thread = kthread_run(drm_sched_main, sched, sched->name); if (IS_ERR(sched->thread)) { + ret = PTR_ERR(sched->thread); + sched->thread = NULL; DRM_ERROR("Failed to create scheduler for %s.\n", name); - return PTR_ERR(sched->thread); + return ret; } sched->ready = true; diff --git a/drivers/gpu/drm/selftests/Makefile b/drivers/gpu/drm/selftests/Makefile index 383d8d6c5847..1bb73dc4c88c 100644 --- a/drivers/gpu/drm/selftests/Makefile +++ b/drivers/gpu/drm/selftests/Makefile @@ -1,4 +1,5 @@ test-drm_modeset-y := test-drm_modeset_common.o test-drm_plane_helper.o \ - test-drm_format.o test-drm_framebuffer.o + test-drm_format.o test-drm_framebuffer.o \ + test-drm_damage_helper.o obj-$(CONFIG_DRM_DEBUG_SELFTEST) += test-drm_mm.o test-drm_modeset.o diff --git a/drivers/gpu/drm/selftests/drm_modeset_selftests.h b/drivers/gpu/drm/selftests/drm_modeset_selftests.h index 92601defd8f6..464753746013 100644 --- a/drivers/gpu/drm/selftests/drm_modeset_selftests.h +++ b/drivers/gpu/drm/selftests/drm_modeset_selftests.h @@ -11,3 +11,24 @@ selftest(check_drm_format_block_width, igt_check_drm_format_block_width) selftest(check_drm_format_block_height, igt_check_drm_format_block_height) selftest(check_drm_format_min_pitch, igt_check_drm_format_min_pitch) selftest(check_drm_framebuffer_create, igt_check_drm_framebuffer_create) +selftest(damage_iter_no_damage, igt_damage_iter_no_damage) +selftest(damage_iter_no_damage_fractional_src, igt_damage_iter_no_damage_fractional_src) +selftest(damage_iter_no_damage_src_moved, igt_damage_iter_no_damage_src_moved) +selftest(damage_iter_no_damage_fractional_src_moved, igt_damage_iter_no_damage_fractional_src_moved) +selftest(damage_iter_no_damage_not_visible, igt_damage_iter_no_damage_not_visible) +selftest(damage_iter_no_damage_no_crtc, igt_damage_iter_no_damage_no_crtc) +selftest(damage_iter_no_damage_no_fb, igt_damage_iter_no_damage_no_fb) +selftest(damage_iter_simple_damage, igt_damage_iter_simple_damage) +selftest(damage_iter_single_damage, igt_damage_iter_single_damage) +selftest(damage_iter_single_damage_intersect_src, igt_damage_iter_single_damage_intersect_src) +selftest(damage_iter_single_damage_outside_src, igt_damage_iter_single_damage_outside_src) +selftest(damage_iter_single_damage_fractional_src, igt_damage_iter_single_damage_fractional_src) +selftest(damage_iter_single_damage_intersect_fractional_src, igt_damage_iter_single_damage_intersect_fractional_src) +selftest(damage_iter_single_damage_outside_fractional_src, igt_damage_iter_single_damage_outside_fractional_src) +selftest(damage_iter_single_damage_src_moved, igt_damage_iter_single_damage_src_moved) +selftest(damage_iter_single_damage_fractional_src_moved, igt_damage_iter_single_damage_fractional_src_moved) +selftest(damage_iter_damage, igt_damage_iter_damage) +selftest(damage_iter_damage_one_intersect, igt_damage_iter_damage_one_intersect) +selftest(damage_iter_damage_one_outside, igt_damage_iter_damage_one_outside) +selftest(damage_iter_damage_src_moved, igt_damage_iter_damage_src_moved) +selftest(damage_iter_damage_not_visible, igt_damage_iter_damage_not_visible) diff --git a/drivers/gpu/drm/selftests/test-drm_damage_helper.c b/drivers/gpu/drm/selftests/test-drm_damage_helper.c new file mode 100644 index 000000000000..9d2bcdf8bc29 --- /dev/null +++ b/drivers/gpu/drm/selftests/test-drm_damage_helper.c @@ -0,0 +1,811 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Test case for drm_damage_helper functions + */ + +#define pr_fmt(fmt) "drm_damage_helper: " fmt + +#include <drm/drm_damage_helper.h> + +#include "test-drm_modeset_common.h" + +static void set_plane_src(struct drm_plane_state *state, int x1, int y1, int x2, + int y2) +{ + state->src.x1 = x1; + state->src.y1 = y1; + state->src.x2 = x2; + state->src.y2 = y2; +} + +static void set_damage_clip(struct drm_mode_rect *r, int x1, int y1, int x2, + int y2) +{ + r->x1 = x1; + r->y1 = y1; + r->x2 = x2; + r->y2 = y2; +} + +static void set_damage_blob(struct drm_property_blob *damage_blob, + struct drm_mode_rect *r, uint32_t size) +{ + damage_blob->length = size; + damage_blob->data = r; +} + +static void set_plane_damage(struct drm_plane_state *state, + struct drm_property_blob *damage_blob) +{ + state->fb_damage_clips = damage_blob; +} + +static bool check_damage_clip(struct drm_plane_state *state, struct drm_rect *r, + int x1, int y1, int x2, int y2) +{ + /* + * Round down x1/y1 and round up x2/y2. This is because damage is not in + * 16.16 fixed point so to catch all pixels. + */ + int src_x1 = state->src.x1 >> 16; + int src_y1 = state->src.y1 >> 16; + int src_x2 = (state->src.x2 >> 16) + !!(state->src.x2 & 0xFFFF); + int src_y2 = (state->src.y2 >> 16) + !!(state->src.y2 & 0xFFFF); + + if (x1 >= x2 || y1 >= y2) { + pr_err("Cannot have damage clip with no dimension.\n"); + return false; + } + + if (x1 < src_x1 || y1 < src_y1 || x2 > src_x2 || y2 > src_y2) { + pr_err("Damage cannot be outside rounded plane src.\n"); + return false; + } + + if (r->x1 != x1 || r->y1 != y1 || r->x2 != x2 || r->y2 != y2) { + pr_err("Damage = %d %d %d %d\n", r->x1, r->y1, r->x2, r->y2); + return false; + } + + return true; +} + +int igt_damage_iter_no_damage(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + /* Plane src same as fb size. */ + set_plane_src(&old_state, 0, 0, fb.width << 16, fb.height << 16); + set_plane_src(&state, 0, 0, fb.width << 16, fb.height << 16); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 1, "Should return plane src as damage."); + FAIL_ON(!check_damage_clip(&state, &clip, 0, 0, 2048, 2048)); + + return 0; +} + +int igt_damage_iter_no_damage_fractional_src(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + /* Plane src has fractional part. */ + set_plane_src(&old_state, 0x3fffe, 0x3fffe, + 0x3fffe + (1024 << 16), 0x3fffe + (768 << 16)); + set_plane_src(&state, 0x3fffe, 0x3fffe, + 0x3fffe + (1024 << 16), 0x3fffe + (768 << 16)); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 1, "Should return rounded off plane src as damage."); + FAIL_ON(!check_damage_clip(&state, &clip, 3, 3, 1028, 772)); + + return 0; +} + +int igt_damage_iter_no_damage_src_moved(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + /* Plane src moved since old plane state. */ + set_plane_src(&old_state, 0, 0, 1024 << 16, 768 << 16); + set_plane_src(&state, 10 << 16, 10 << 16, + (10 + 1024) << 16, (10 + 768) << 16); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 1, "Should return plane src as damage."); + FAIL_ON(!check_damage_clip(&state, &clip, 10, 10, 1034, 778)); + + return 0; +} + +int igt_damage_iter_no_damage_fractional_src_moved(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + /* Plane src has fractional part and it moved since old plane state. */ + set_plane_src(&old_state, 0x3fffe, 0x3fffe, + 0x3fffe + (1024 << 16), 0x3fffe + (768 << 16)); + set_plane_src(&state, 0x40002, 0x40002, + 0x40002 + (1024 << 16), 0x40002 + (768 << 16)); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 1, "Should return plane src as damage."); + FAIL_ON(!check_damage_clip(&state, &clip, 4, 4, 1029, 773)); + + return 0; +} + +int igt_damage_iter_no_damage_not_visible(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = false, + }; + + set_plane_src(&old_state, 0, 0, 1024 << 16, 768 << 16); + set_plane_src(&state, 0, 0, 1024 << 16, 768 << 16); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 0, "Should have no damage."); + + return 0; +} + +int igt_damage_iter_no_damage_no_crtc(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = 0, + .fb = &fb, + }; + + set_plane_src(&old_state, 0, 0, 1024 << 16, 768 << 16); + set_plane_src(&state, 0, 0, 1024 << 16, 768 << 16); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 0, "Should have no damage."); + + return 0; +} + +int igt_damage_iter_no_damage_no_fb(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = 0, + }; + + set_plane_src(&old_state, 0, 0, 1024 << 16, 768 << 16); + set_plane_src(&state, 0, 0, 1024 << 16, 768 << 16); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 0, "Should have no damage."); + + return 0; +} + +int igt_damage_iter_simple_damage(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_property_blob damage_blob; + struct drm_mode_rect damage; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + set_plane_src(&old_state, 0, 0, 1024 << 16, 768 << 16); + set_plane_src(&state, 0, 0, 1024 << 16, 768 << 16); + /* Damage set to plane src */ + set_damage_clip(&damage, 0, 0, 1024, 768); + set_damage_blob(&damage_blob, &damage, sizeof(damage)); + set_plane_damage(&state, &damage_blob); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 1, "Should return damage when set."); + FAIL_ON(!check_damage_clip(&state, &clip, 0, 0, 1024, 768)); + + return 0; +} + +int igt_damage_iter_single_damage(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_property_blob damage_blob; + struct drm_mode_rect damage; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + set_plane_src(&old_state, 0, 0, 1024 << 16, 768 << 16); + set_plane_src(&state, 0, 0, 1024 << 16, 768 << 16); + set_damage_clip(&damage, 256, 192, 768, 576); + set_damage_blob(&damage_blob, &damage, sizeof(damage)); + set_plane_damage(&state, &damage_blob); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 1, "Should return damage when set."); + FAIL_ON(!check_damage_clip(&state, &clip, 256, 192, 768, 576)); + + return 0; +} + +int igt_damage_iter_single_damage_intersect_src(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_property_blob damage_blob; + struct drm_mode_rect damage; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + set_plane_src(&old_state, 0, 0, 1024 << 16, 768 << 16); + set_plane_src(&state, 0, 0, 1024 << 16, 768 << 16); + /* Damage intersect with plane src. */ + set_damage_clip(&damage, 256, 192, 1360, 768); + set_damage_blob(&damage_blob, &damage, sizeof(damage)); + set_plane_damage(&state, &damage_blob); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 1, "Should return damage clipped to src."); + FAIL_ON(!check_damage_clip(&state, &clip, 256, 192, 1024, 768)); + + return 0; +} + +int igt_damage_iter_single_damage_outside_src(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_property_blob damage_blob; + struct drm_mode_rect damage; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + set_plane_src(&old_state, 0, 0, 1024 << 16, 768 << 16); + set_plane_src(&state, 0, 0, 1024 << 16, 768 << 16); + /* Damage clip outside plane src */ + set_damage_clip(&damage, 1360, 1360, 1380, 1380); + set_damage_blob(&damage_blob, &damage, sizeof(damage)); + set_plane_damage(&state, &damage_blob); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 0, "Should have no damage."); + + return 0; +} + +int igt_damage_iter_single_damage_fractional_src(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_property_blob damage_blob; + struct drm_mode_rect damage; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + /* Plane src has fractional part. */ + set_plane_src(&old_state, 0x40002, 0x40002, + 0x40002 + (1024 << 16), 0x40002 + (768 << 16)); + set_plane_src(&state, 0x40002, 0x40002, + 0x40002 + (1024 << 16), 0x40002 + (768 << 16)); + set_damage_clip(&damage, 10, 10, 256, 330); + set_damage_blob(&damage_blob, &damage, sizeof(damage)); + set_plane_damage(&state, &damage_blob); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 1, "Should return damage when set."); + FAIL_ON(!check_damage_clip(&state, &clip, 10, 10, 256, 330)); + + return 0; +} + +int igt_damage_iter_single_damage_intersect_fractional_src(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_property_blob damage_blob; + struct drm_mode_rect damage; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + /* Plane src has fractional part. */ + set_plane_src(&old_state, 0x40002, 0x40002, + 0x40002 + (1024 << 16), 0x40002 + (768 << 16)); + set_plane_src(&state, 0x40002, 0x40002, + 0x40002 + (1024 << 16), 0x40002 + (768 << 16)); + /* Damage intersect with plane src. */ + set_damage_clip(&damage, 10, 1, 1360, 330); + set_damage_blob(&damage_blob, &damage, sizeof(damage)); + set_plane_damage(&state, &damage_blob); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 1, "Should return damage clipped to rounded off src."); + FAIL_ON(!check_damage_clip(&state, &clip, 10, 4, 1029, 330)); + + return 0; +} + +int igt_damage_iter_single_damage_outside_fractional_src(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_property_blob damage_blob; + struct drm_mode_rect damage; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + /* Plane src has fractional part. */ + set_plane_src(&old_state, 0x40002, 0x40002, + 0x40002 + (1024 << 16), 0x40002 + (768 << 16)); + set_plane_src(&state, 0x40002, 0x40002, + 0x40002 + (1024 << 16), 0x40002 + (768 << 16)); + /* Damage clip outside plane src */ + set_damage_clip(&damage, 1360, 1360, 1380, 1380); + set_damage_blob(&damage_blob, &damage, sizeof(damage)); + set_plane_damage(&state, &damage_blob); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 0, "Should have no damage."); + + return 0; +} + +int igt_damage_iter_single_damage_src_moved(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_property_blob damage_blob; + struct drm_mode_rect damage; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + /* Plane src moved since old plane state. */ + set_plane_src(&old_state, 0, 0, 1024 << 16, 768 << 16); + set_plane_src(&state, 10 << 16, 10 << 16, + (10 + 1024) << 16, (10 + 768) << 16); + set_damage_clip(&damage, 20, 30, 256, 256); + set_damage_blob(&damage_blob, &damage, sizeof(damage)); + set_plane_damage(&state, &damage_blob); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 1, "Should return plane src as damage."); + FAIL_ON(!check_damage_clip(&state, &clip, 10, 10, 1034, 778)); + + return 0; +} + +int igt_damage_iter_single_damage_fractional_src_moved(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_property_blob damage_blob; + struct drm_mode_rect damage; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + /* Plane src with fractional part moved since old plane state. */ + set_plane_src(&old_state, 0x3fffe, 0x3fffe, + 0x3fffe + (1024 << 16), 0x3fffe + (768 << 16)); + set_plane_src(&state, 0x40002, 0x40002, + 0x40002 + (1024 << 16), 0x40002 + (768 << 16)); + /* Damage intersect with plane src. */ + set_damage_clip(&damage, 20, 30, 1360, 256); + set_damage_blob(&damage_blob, &damage, sizeof(damage)); + set_plane_damage(&state, &damage_blob); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 1, "Should return rounded off plane src as damage."); + FAIL_ON(!check_damage_clip(&state, &clip, 4, 4, 1029, 773)); + + return 0; +} + +int igt_damage_iter_damage(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_property_blob damage_blob; + struct drm_mode_rect damage[2]; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + set_plane_src(&old_state, 0, 0, 1024 << 16, 768 << 16); + set_plane_src(&state, 0, 0, 1024 << 16, 768 << 16); + /* 2 damage clips. */ + set_damage_clip(&damage[0], 20, 30, 200, 180); + set_damage_clip(&damage[1], 240, 200, 280, 250); + set_damage_blob(&damage_blob, &damage[0], sizeof(damage)); + set_plane_damage(&state, &damage_blob); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) { + if (num_hits == 0) + FAIL_ON(!check_damage_clip(&state, &clip, 20, 30, 200, 180)); + if (num_hits == 1) + FAIL_ON(!check_damage_clip(&state, &clip, 240, 200, 280, 250)); + num_hits++; + } + + FAIL(num_hits != 2, "Should return damage when set."); + + return 0; +} + +int igt_damage_iter_damage_one_intersect(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_property_blob damage_blob; + struct drm_mode_rect damage[2]; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + set_plane_src(&old_state, 0x40002, 0x40002, + 0x40002 + (1024 << 16), 0x40002 + (768 << 16)); + set_plane_src(&state, 0x40002, 0x40002, + 0x40002 + (1024 << 16), 0x40002 + (768 << 16)); + /* 2 damage clips, one intersect plane src. */ + set_damage_clip(&damage[0], 20, 30, 200, 180); + set_damage_clip(&damage[1], 2, 2, 1360, 1360); + set_damage_blob(&damage_blob, &damage[0], sizeof(damage)); + set_plane_damage(&state, &damage_blob); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) { + if (num_hits == 0) + FAIL_ON(!check_damage_clip(&state, &clip, 20, 30, 200, 180)); + if (num_hits == 1) + FAIL_ON(!check_damage_clip(&state, &clip, 4, 4, 1029, 773)); + num_hits++; + } + + FAIL(num_hits != 2, "Should return damage when set."); + + return 0; +} + +int igt_damage_iter_damage_one_outside(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_property_blob damage_blob; + struct drm_mode_rect damage[2]; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + set_plane_src(&old_state, 0, 0, 1024 << 16, 768 << 16); + set_plane_src(&state, 0, 0, 1024 << 16, 768 << 16); + /* 2 damage clips, one outside plane src. */ + set_damage_clip(&damage[0], 1360, 1360, 1380, 1380); + set_damage_clip(&damage[1], 240, 200, 280, 250); + set_damage_blob(&damage_blob, &damage[0], sizeof(damage)); + set_plane_damage(&state, &damage_blob); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 1, "Should return damage when set."); + FAIL_ON(!check_damage_clip(&state, &clip, 240, 200, 280, 250)); + + return 0; +} + +int igt_damage_iter_damage_src_moved(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_property_blob damage_blob; + struct drm_mode_rect damage[2]; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = true, + }; + + set_plane_src(&old_state, 0x40002, 0x40002, + 0x40002 + (1024 << 16), 0x40002 + (768 << 16)); + set_plane_src(&state, 0x3fffe, 0x3fffe, + 0x3fffe + (1024 << 16), 0x3fffe + (768 << 16)); + /* 2 damage clips, one outside plane src. */ + set_damage_clip(&damage[0], 1360, 1360, 1380, 1380); + set_damage_clip(&damage[1], 240, 200, 280, 250); + set_damage_blob(&damage_blob, &damage[0], sizeof(damage)); + set_plane_damage(&state, &damage_blob); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 1, "Should return round off plane src as damage."); + FAIL_ON(!check_damage_clip(&state, &clip, 3, 3, 1028, 772)); + + return 0; +} + +int igt_damage_iter_damage_not_visible(void *ignored) +{ + struct drm_atomic_helper_damage_iter iter; + struct drm_plane_state old_state; + struct drm_property_blob damage_blob; + struct drm_mode_rect damage[2]; + struct drm_rect clip; + uint32_t num_hits = 0; + + struct drm_framebuffer fb = { + .width = 2048, + .height = 2048 + }; + + struct drm_plane_state state = { + .crtc = ZERO_SIZE_PTR, + .fb = &fb, + .visible = false, + }; + + set_plane_src(&old_state, 0x40002, 0x40002, + 0x40002 + (1024 << 16), 0x40002 + (768 << 16)); + set_plane_src(&state, 0x3fffe, 0x3fffe, + 0x3fffe + (1024 << 16), 0x3fffe + (768 << 16)); + /* 2 damage clips, one outside plane src. */ + set_damage_clip(&damage[0], 1360, 1360, 1380, 1380); + set_damage_clip(&damage[1], 240, 200, 280, 250); + set_damage_blob(&damage_blob, &damage[0], sizeof(damage)); + set_plane_damage(&state, &damage_blob); + drm_atomic_helper_damage_iter_init(&iter, &old_state, &state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + FAIL(num_hits != 0, "Should not return any damage."); + + return 0; +} diff --git a/drivers/gpu/drm/selftests/test-drm_modeset_common.h b/drivers/gpu/drm/selftests/test-drm_modeset_common.h index d5df5bd51b05..8c76f09c12d1 100644 --- a/drivers/gpu/drm/selftests/test-drm_modeset_common.h +++ b/drivers/gpu/drm/selftests/test-drm_modeset_common.h @@ -18,5 +18,26 @@ int igt_check_drm_format_block_width(void *ignored); int igt_check_drm_format_block_height(void *ignored); int igt_check_drm_format_min_pitch(void *ignored); int igt_check_drm_framebuffer_create(void *ignored); +int igt_damage_iter_no_damage(void *ignored); +int igt_damage_iter_no_damage_fractional_src(void *ignored); +int igt_damage_iter_no_damage_src_moved(void *ignored); +int igt_damage_iter_no_damage_fractional_src_moved(void *ignored); +int igt_damage_iter_no_damage_not_visible(void *ignored); +int igt_damage_iter_no_damage_no_crtc(void *ignored); +int igt_damage_iter_no_damage_no_fb(void *ignored); +int igt_damage_iter_simple_damage(void *ignored); +int igt_damage_iter_single_damage(void *ignored); +int igt_damage_iter_single_damage_intersect_src(void *ignored); +int igt_damage_iter_single_damage_outside_src(void *ignored); +int igt_damage_iter_single_damage_fractional_src(void *ignored); +int igt_damage_iter_single_damage_intersect_fractional_src(void *ignored); +int igt_damage_iter_single_damage_outside_fractional_src(void *ignored); +int igt_damage_iter_single_damage_src_moved(void *ignored); +int igt_damage_iter_single_damage_fractional_src_moved(void *ignored); +int igt_damage_iter_damage(void *ignored); +int igt_damage_iter_damage_one_intersect(void *ignored); +int igt_damage_iter_damage_one_outside(void *ignored); +int igt_damage_iter_damage_src_moved(void *ignored); +int igt_damage_iter_damage_not_visible(void *ignored); #endif diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c index f80e82e16475..607a6ea17ecc 100644 --- a/drivers/gpu/drm/tegra/dc.c +++ b/drivers/gpu/drm/tegra/dc.c @@ -1978,6 +1978,23 @@ static irqreturn_t tegra_dc_irq(int irq, void *data) return IRQ_HANDLED; } +static bool tegra_dc_has_window_groups(struct tegra_dc *dc) +{ + unsigned int i; + + if (!dc->soc->wgrps) + return true; + + for (i = 0; i < dc->soc->num_wgrps; i++) { + const struct tegra_windowgroup_soc *wgrp = &dc->soc->wgrps[i]; + + if (wgrp->dc == dc->pipe && wgrp->num_windows > 0) + return true; + } + + return false; +} + static int tegra_dc_init(struct host1x_client *client) { struct drm_device *drm = dev_get_drvdata(client->parent); @@ -1993,22 +2010,8 @@ static int tegra_dc_init(struct host1x_client *client) * assign a primary plane to them, which in turn will cause KMS to * crash. */ - if (dc->soc->wgrps) { - bool has_wgrps = false; - unsigned int i; - - for (i = 0; i < dc->soc->num_wgrps; i++) { - const struct tegra_windowgroup_soc *wgrp = &dc->soc->wgrps[i]; - - if (wgrp->dc == dc->pipe && wgrp->num_windows > 0) { - has_wgrps = true; - break; - } - } - - if (!has_wgrps) - return 0; - } + if (!tegra_dc_has_window_groups(dc)) + return 0; dc->syncpt = host1x_syncpt_request(client, flags); if (!dc->syncpt) @@ -2094,6 +2097,9 @@ static int tegra_dc_exit(struct host1x_client *client) struct tegra_dc *dc = host1x_client_to_dc(client); int err; + if (!tegra_dc_has_window_groups(dc)) + return 0; + devm_free_irq(dc->dev, dc->irq, dc); err = tegra_dc_rgb_exit(dc); diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c index 65ea4988b332..4b70ce664c41 100644 --- a/drivers/gpu/drm/tegra/drm.c +++ b/drivers/gpu/drm/tegra/drm.c @@ -1274,6 +1274,7 @@ static const struct of_device_id host1x_drm_subdevs[] = { { .compatible = "nvidia,tegra194-display", }, { .compatible = "nvidia,tegra194-dc", }, { .compatible = "nvidia,tegra194-sor", }, + { .compatible = "nvidia,tegra194-vic", }, { /* sentinel */ } }; diff --git a/drivers/gpu/drm/tegra/falcon.c b/drivers/gpu/drm/tegra/falcon.c index f685e72949d1..352d05feabb0 100644 --- a/drivers/gpu/drm/tegra/falcon.c +++ b/drivers/gpu/drm/tegra/falcon.c @@ -141,9 +141,9 @@ int falcon_load_firmware(struct falcon *falcon) /* allocate iova space for the firmware */ falcon->firmware.vaddr = falcon->ops->alloc(falcon, firmware->size, &falcon->firmware.paddr); - if (!falcon->firmware.vaddr) { - dev_err(falcon->dev, "dma memory mapping failed\n"); - return -ENOMEM; + if (IS_ERR(falcon->firmware.vaddr)) { + dev_err(falcon->dev, "DMA memory mapping failed\n"); + return PTR_ERR(falcon->firmware.vaddr); } /* copy firmware image into local area. this also ensures endianness */ @@ -197,11 +197,19 @@ void falcon_exit(struct falcon *falcon) int falcon_boot(struct falcon *falcon) { unsigned long offset; + u32 value; int err; if (!falcon->firmware.vaddr) return -EINVAL; + err = readl_poll_timeout(falcon->regs + FALCON_DMACTL, value, + (value & (FALCON_DMACTL_IMEM_SCRUBBING | + FALCON_DMACTL_DMEM_SCRUBBING)) == 0, + 10, 10000); + if (err < 0) + return err; + falcon_writel(falcon, 0, FALCON_DMACTL); /* setup the address of the binary data so Falcon can access it later */ diff --git a/drivers/gpu/drm/tegra/hub.c b/drivers/gpu/drm/tegra/hub.c index 3e26be5359cf..71cc3cf60066 100644 --- a/drivers/gpu/drm/tegra/hub.c +++ b/drivers/gpu/drm/tegra/hub.c @@ -742,7 +742,9 @@ static const struct host1x_client_ops tegra_display_hub_ops = { static int tegra_display_hub_probe(struct platform_device *pdev) { + struct device_node *child = NULL; struct tegra_display_hub *hub; + struct clk *clk; unsigned int i; int err; @@ -801,6 +803,34 @@ static int tegra_display_hub_probe(struct platform_device *pdev) return err; } + hub->num_heads = of_get_child_count(pdev->dev.of_node); + + hub->clk_heads = devm_kcalloc(&pdev->dev, hub->num_heads, sizeof(clk), + GFP_KERNEL); + if (!hub->clk_heads) + return -ENOMEM; + + for (i = 0; i < hub->num_heads; i++) { + child = of_get_next_child(pdev->dev.of_node, child); + if (!child) { + dev_err(&pdev->dev, "failed to find node for head %u\n", + i); + return -ENODEV; + } + + clk = devm_get_clk_from_child(&pdev->dev, child, "dc"); + if (IS_ERR(clk)) { + dev_err(&pdev->dev, "failed to get clock for head %u\n", + i); + of_node_put(child); + return PTR_ERR(clk); + } + + hub->clk_heads[i] = clk; + } + + of_node_put(child); + /* XXX: enable clock across reset? */ err = reset_control_assert(hub->rst); if (err < 0) @@ -840,12 +870,16 @@ static int tegra_display_hub_remove(struct platform_device *pdev) static int __maybe_unused tegra_display_hub_suspend(struct device *dev) { struct tegra_display_hub *hub = dev_get_drvdata(dev); + unsigned int i = hub->num_heads; int err; err = reset_control_assert(hub->rst); if (err < 0) return err; + while (i--) + clk_disable_unprepare(hub->clk_heads[i]); + clk_disable_unprepare(hub->clk_hub); clk_disable_unprepare(hub->clk_dsc); clk_disable_unprepare(hub->clk_disp); @@ -856,6 +890,7 @@ static int __maybe_unused tegra_display_hub_suspend(struct device *dev) static int __maybe_unused tegra_display_hub_resume(struct device *dev) { struct tegra_display_hub *hub = dev_get_drvdata(dev); + unsigned int i; int err; err = clk_prepare_enable(hub->clk_disp); @@ -870,13 +905,22 @@ static int __maybe_unused tegra_display_hub_resume(struct device *dev) if (err < 0) goto disable_dsc; + for (i = 0; i < hub->num_heads; i++) { + err = clk_prepare_enable(hub->clk_heads[i]); + if (err < 0) + goto disable_heads; + } + err = reset_control_deassert(hub->rst); if (err < 0) - goto disable_hub; + goto disable_heads; return 0; -disable_hub: +disable_heads: + while (i--) + clk_disable_unprepare(hub->clk_heads[i]); + clk_disable_unprepare(hub->clk_hub); disable_dsc: clk_disable_unprepare(hub->clk_dsc); diff --git a/drivers/gpu/drm/tegra/hub.h b/drivers/gpu/drm/tegra/hub.h index 6696a85fc1f2..479087c0705a 100644 --- a/drivers/gpu/drm/tegra/hub.h +++ b/drivers/gpu/drm/tegra/hub.h @@ -49,6 +49,9 @@ struct tegra_display_hub { struct clk *clk_hub; struct reset_control *rst; + unsigned int num_heads; + struct clk **clk_heads; + const struct tegra_display_hub_soc *soc; struct tegra_windowgroup *wgrps; }; diff --git a/drivers/gpu/drm/tegra/sor.c b/drivers/gpu/drm/tegra/sor.c index d90bf5f6a67a..44feac2a0359 100644 --- a/drivers/gpu/drm/tegra/sor.c +++ b/drivers/gpu/drm/tegra/sor.c @@ -19,6 +19,8 @@ #include <soc/tegra/pmc.h> +#include <sound/hda_verbs.h> + #include <drm/drm_atomic_helper.h> #include <drm/drm_dp_helper.h> #include <drm/drm_panel.h> @@ -29,14 +31,6 @@ #include "sor.h" #include "trace.h" -/* - * XXX Remove this after the commit adding it to soc/tegra/pmc.h has been - * merged. Having this around after the commit is merged should be safe since - * the preprocessor will effectively replace all occurrences and therefore no - * duplicate will be defined. - */ -#define TEGRA_IO_PAD_HDMI_DP0 26 - #define SOR_REKEY 0x38 struct tegra_sor_hdmi_settings { @@ -407,6 +401,7 @@ struct tegra_sor { const struct tegra_sor_soc *soc; void __iomem *regs; unsigned int index; + unsigned int irq; struct reset_control *rst; struct clk *clk_parent; @@ -433,6 +428,11 @@ struct tegra_sor { struct delayed_work scdc; bool scdc_enabled; + + struct { + unsigned int sample_rate; + unsigned int channels; + } audio; }; struct tegra_sor_state { @@ -2140,6 +2140,144 @@ tegra_sor_hdmi_setup_avi_infoframe(struct tegra_sor *sor, return 0; } +static void tegra_sor_write_eld(struct tegra_sor *sor) +{ + size_t length = drm_eld_size(sor->output.connector.eld), i; + + for (i = 0; i < length; i++) + tegra_sor_writel(sor, i << 8 | sor->output.connector.eld[i], + SOR_AUDIO_HDA_ELD_BUFWR); + + /* + * The HDA codec will always report an ELD buffer size of 96 bytes and + * the HDA codec driver will check that each byte read from the buffer + * is valid. Therefore every byte must be written, even if no 96 bytes + * were parsed from EDID. + */ + for (i = length; i < 96; i++) + tegra_sor_writel(sor, i << 8 | 0, SOR_AUDIO_HDA_ELD_BUFWR); +} + +static void tegra_sor_audio_prepare(struct tegra_sor *sor) +{ + u32 value; + + tegra_sor_write_eld(sor); + + value = SOR_AUDIO_HDA_PRESENSE_ELDV | SOR_AUDIO_HDA_PRESENSE_PD; + tegra_sor_writel(sor, value, SOR_AUDIO_HDA_PRESENSE); +} + +static void tegra_sor_audio_unprepare(struct tegra_sor *sor) +{ + tegra_sor_writel(sor, 0, SOR_AUDIO_HDA_PRESENSE); +} + +static int tegra_sor_hdmi_enable_audio_infoframe(struct tegra_sor *sor) +{ + u8 buffer[HDMI_INFOFRAME_SIZE(AUDIO)]; + struct hdmi_audio_infoframe frame; + u32 value; + int err; + + err = hdmi_audio_infoframe_init(&frame); + if (err < 0) { + dev_err(sor->dev, "failed to setup audio infoframe: %d\n", err); + return err; + } + + frame.channels = sor->audio.channels; + + err = hdmi_audio_infoframe_pack(&frame, buffer, sizeof(buffer)); + if (err < 0) { + dev_err(sor->dev, "failed to pack audio infoframe: %d\n", err); + return err; + } + + tegra_sor_hdmi_write_infopack(sor, buffer, err); + + value = tegra_sor_readl(sor, SOR_HDMI_AUDIO_INFOFRAME_CTRL); + value |= INFOFRAME_CTRL_CHECKSUM_ENABLE; + value |= INFOFRAME_CTRL_ENABLE; + tegra_sor_writel(sor, value, SOR_HDMI_AUDIO_INFOFRAME_CTRL); + + return 0; +} + +static void tegra_sor_hdmi_audio_enable(struct tegra_sor *sor) +{ + u32 value; + + value = tegra_sor_readl(sor, SOR_AUDIO_CNTRL); + + /* select HDA audio input */ + value &= ~SOR_AUDIO_CNTRL_SOURCE_SELECT(SOURCE_SELECT_MASK); + value |= SOR_AUDIO_CNTRL_SOURCE_SELECT(SOURCE_SELECT_HDA); + + /* inject null samples */ + if (sor->audio.channels != 2) + value &= ~SOR_AUDIO_CNTRL_INJECT_NULLSMPL; + else + value |= SOR_AUDIO_CNTRL_INJECT_NULLSMPL; + + value |= SOR_AUDIO_CNTRL_AFIFO_FLUSH; + + tegra_sor_writel(sor, value, SOR_AUDIO_CNTRL); + + /* enable advertising HBR capability */ + tegra_sor_writel(sor, SOR_AUDIO_SPARE_HBR_ENABLE, SOR_AUDIO_SPARE); + + tegra_sor_writel(sor, 0, SOR_HDMI_ACR_CTRL); + + value = SOR_HDMI_SPARE_ACR_PRIORITY_HIGH | + SOR_HDMI_SPARE_CTS_RESET(1) | + SOR_HDMI_SPARE_HW_CTS_ENABLE; + tegra_sor_writel(sor, value, SOR_HDMI_SPARE); + + /* enable HW CTS */ + value = SOR_HDMI_ACR_SUBPACK_LOW_SB1(0); + tegra_sor_writel(sor, value, SOR_HDMI_ACR_0441_SUBPACK_LOW); + + /* allow packet to be sent */ + value = SOR_HDMI_ACR_SUBPACK_HIGH_ENABLE; + tegra_sor_writel(sor, value, SOR_HDMI_ACR_0441_SUBPACK_HIGH); + + /* reset N counter and enable lookup */ + value = SOR_HDMI_AUDIO_N_RESET | SOR_HDMI_AUDIO_N_LOOKUP; + tegra_sor_writel(sor, value, SOR_HDMI_AUDIO_N); + + value = (24000 * 4096) / (128 * sor->audio.sample_rate / 1000); + tegra_sor_writel(sor, value, SOR_AUDIO_AVAL_0320); + tegra_sor_writel(sor, 4096, SOR_AUDIO_NVAL_0320); + + tegra_sor_writel(sor, 20000, SOR_AUDIO_AVAL_0441); + tegra_sor_writel(sor, 4704, SOR_AUDIO_NVAL_0441); + + tegra_sor_writel(sor, 20000, SOR_AUDIO_AVAL_0882); + tegra_sor_writel(sor, 9408, SOR_AUDIO_NVAL_0882); + + tegra_sor_writel(sor, 20000, SOR_AUDIO_AVAL_1764); + tegra_sor_writel(sor, 18816, SOR_AUDIO_NVAL_1764); + + value = (24000 * 6144) / (128 * sor->audio.sample_rate / 1000); + tegra_sor_writel(sor, value, SOR_AUDIO_AVAL_0480); + tegra_sor_writel(sor, 6144, SOR_AUDIO_NVAL_0480); + + value = (24000 * 12288) / (128 * sor->audio.sample_rate / 1000); + tegra_sor_writel(sor, value, SOR_AUDIO_AVAL_0960); + tegra_sor_writel(sor, 12288, SOR_AUDIO_NVAL_0960); + + value = (24000 * 24576) / (128 * sor->audio.sample_rate / 1000); + tegra_sor_writel(sor, value, SOR_AUDIO_AVAL_1920); + tegra_sor_writel(sor, 24576, SOR_AUDIO_NVAL_1920); + + value = tegra_sor_readl(sor, SOR_HDMI_AUDIO_N); + value &= ~SOR_HDMI_AUDIO_N_RESET; + tegra_sor_writel(sor, value, SOR_HDMI_AUDIO_N); + + tegra_sor_hdmi_enable_audio_infoframe(sor); +} + static void tegra_sor_hdmi_disable_audio_infoframe(struct tegra_sor *sor) { u32 value; @@ -2149,6 +2287,11 @@ static void tegra_sor_hdmi_disable_audio_infoframe(struct tegra_sor *sor) tegra_sor_writel(sor, value, SOR_HDMI_AUDIO_INFOFRAME_CTRL); } +static void tegra_sor_hdmi_audio_disable(struct tegra_sor *sor) +{ + tegra_sor_hdmi_disable_audio_infoframe(sor); +} + static struct tegra_sor_hdmi_settings * tegra_sor_hdmi_find_settings(struct tegra_sor *sor, unsigned long frequency) { @@ -2244,6 +2387,7 @@ static void tegra_sor_hdmi_disable(struct drm_encoder *encoder) u32 value; int err; + tegra_sor_audio_unprepare(sor); tegra_sor_hdmi_scdc_stop(sor); err = tegra_sor_detach(sor); @@ -2652,6 +2796,7 @@ static void tegra_sor_hdmi_enable(struct drm_encoder *encoder) dev_err(sor->dev, "failed to wakeup SOR: %d\n", err); tegra_sor_hdmi_scdc_start(sor); + tegra_sor_audio_prepare(sor); } static const struct drm_encoder_helper_funcs tegra_sor_hdmi_helpers = { @@ -2667,6 +2812,7 @@ static int tegra_sor_init(struct host1x_client *client) struct tegra_sor *sor = host1x_client_to_sor(client); int connector = DRM_MODE_CONNECTOR_Unknown; int encoder = DRM_MODE_ENCODER_NONE; + u32 value; int err; if (!sor->aux) { @@ -2760,6 +2906,15 @@ static int tegra_sor_init(struct host1x_client *client) if (err < 0) return err; + /* + * Enable and unmask the HDA codec SCRATCH0 register interrupt. This + * is used for interoperability between the HDA codec driver and the + * HDMI/DP driver. + */ + value = SOR_INT_CODEC_SCRATCH1 | SOR_INT_CODEC_SCRATCH0; + tegra_sor_writel(sor, value, SOR_INT_ENABLE); + tegra_sor_writel(sor, value, SOR_INT_MASK); + return 0; } @@ -2768,6 +2923,9 @@ static int tegra_sor_exit(struct host1x_client *client) struct tegra_sor *sor = host1x_client_to_sor(client); int err; + tegra_sor_writel(sor, 0, SOR_INT_MASK); + tegra_sor_writel(sor, 0, SOR_INT_ENABLE); + tegra_output_exit(&sor->output); if (sor->aux) { @@ -3038,6 +3196,54 @@ static int tegra_sor_parse_dt(struct tegra_sor *sor) return 0; } +static void tegra_hda_parse_format(unsigned int format, unsigned int *rate, + unsigned int *channels) +{ + unsigned int mul, div; + + if (format & AC_FMT_BASE_44K) + *rate = 44100; + else + *rate = 48000; + + mul = (format & AC_FMT_MULT_MASK) >> AC_FMT_MULT_SHIFT; + div = (format & AC_FMT_DIV_MASK) >> AC_FMT_DIV_SHIFT; + + *rate = *rate * (mul + 1) / (div + 1); + + *channels = (format & AC_FMT_CHAN_MASK) >> AC_FMT_CHAN_SHIFT; +} + +static irqreturn_t tegra_sor_irq(int irq, void *data) +{ + struct tegra_sor *sor = data; + u32 value; + + value = tegra_sor_readl(sor, SOR_INT_STATUS); + tegra_sor_writel(sor, value, SOR_INT_STATUS); + + if (value & SOR_INT_CODEC_SCRATCH0) { + value = tegra_sor_readl(sor, SOR_AUDIO_HDA_CODEC_SCRATCH0); + + if (value & SOR_AUDIO_HDA_CODEC_SCRATCH0_VALID) { + unsigned int format, sample_rate, channels; + + format = value & SOR_AUDIO_HDA_CODEC_SCRATCH0_FMT_MASK; + + tegra_hda_parse_format(format, &sample_rate, &channels); + + sor->audio.sample_rate = sample_rate; + sor->audio.channels = channels; + + tegra_sor_hdmi_audio_enable(sor); + } else { + tegra_sor_hdmi_audio_disable(sor); + } + } + + return IRQ_HANDLED; +} + static int tegra_sor_probe(struct platform_device *pdev) { struct device_node *np; @@ -3120,14 +3326,38 @@ static int tegra_sor_probe(struct platform_device *pdev) goto remove; } - if (!pdev->dev.pm_domain) { - sor->rst = devm_reset_control_get(&pdev->dev, "sor"); - if (IS_ERR(sor->rst)) { - err = PTR_ERR(sor->rst); + err = platform_get_irq(pdev, 0); + if (err < 0) { + dev_err(&pdev->dev, "failed to get IRQ: %d\n", err); + goto remove; + } + + sor->irq = err; + + err = devm_request_irq(sor->dev, sor->irq, tegra_sor_irq, 0, + dev_name(sor->dev), sor); + if (err < 0) { + dev_err(&pdev->dev, "failed to request IRQ: %d\n", err); + goto remove; + } + + sor->rst = devm_reset_control_get(&pdev->dev, "sor"); + if (IS_ERR(sor->rst)) { + err = PTR_ERR(sor->rst); + + if (err != -EBUSY || WARN_ON(!pdev->dev.pm_domain)) { dev_err(&pdev->dev, "failed to get reset control: %d\n", err); goto remove; } + + /* + * At this point, the reset control is most likely being used + * by the generic power domain implementation. With any luck + * the power domain will have taken care of resetting the SOR + * and we don't have to do anything. + */ + sor->rst = NULL; } sor->clk = devm_clk_get(&pdev->dev, NULL); diff --git a/drivers/gpu/drm/tegra/sor.h b/drivers/gpu/drm/tegra/sor.h index fb0854d92a27..13f7e68bec42 100644 --- a/drivers/gpu/drm/tegra/sor.h +++ b/drivers/gpu/drm/tegra/sor.h @@ -364,12 +364,28 @@ #define INFOFRAME_HEADER_VERSION(x) (((x) & 0xff) << 8) #define INFOFRAME_HEADER_TYPE(x) (((x) & 0xff) << 0) +#define SOR_HDMI_ACR_CTRL 0xb1 + +#define SOR_HDMI_ACR_0320_SUBPACK_LOW 0xb2 +#define SOR_HDMI_ACR_SUBPACK_LOW_SB1(x) (((x) & 0xff) << 24) + +#define SOR_HDMI_ACR_0320_SUBPACK_HIGH 0xb3 +#define SOR_HDMI_ACR_SUBPACK_HIGH_ENABLE (1 << 31) + +#define SOR_HDMI_ACR_0441_SUBPACK_LOW 0xb4 +#define SOR_HDMI_ACR_0441_SUBPACK_HIGH 0xb5 + #define SOR_HDMI_CTRL 0xc0 #define SOR_HDMI_CTRL_ENABLE (1 << 30) #define SOR_HDMI_CTRL_MAX_AC_PACKET(x) (((x) & 0x1f) << 16) #define SOR_HDMI_CTRL_AUDIO_LAYOUT (1 << 10) #define SOR_HDMI_CTRL_REKEY(x) (((x) & 0x7f) << 0) +#define SOR_HDMI_SPARE 0xcb +#define SOR_HDMI_SPARE_ACR_PRIORITY_HIGH (1 << 31) +#define SOR_HDMI_SPARE_CTS_RESET(x) (((x) & 0x7) << 16) +#define SOR_HDMI_SPARE_HW_CTS_ENABLE (1 << 0) + #define SOR_REFCLK 0xe6 #define SOR_REFCLK_DIV_INT(x) ((((x) >> 2) & 0xff) << 8) #define SOR_REFCLK_DIV_FRAC(x) (((x) & 0x3) << 6) @@ -378,10 +394,62 @@ #define SOR_INPUT_CONTROL_ARM_VIDEO_RANGE_LIMITED (1 << 1) #define SOR_INPUT_CONTROL_HDMI_SRC_SELECT(x) (((x) & 0x1) << 0) +#define SOR_AUDIO_CNTRL 0xfc +#define SOR_AUDIO_CNTRL_INJECT_NULLSMPL (1 << 29) +#define SOR_AUDIO_CNTRL_SOURCE_SELECT(x) (((x) & 0x3) << 20) +#define SOURCE_SELECT_MASK 0x3 +#define SOURCE_SELECT_HDA 0x2 +#define SOURCE_SELECT_SPDIF 0x1 +#define SOURCE_SELECT_AUTO 0x0 +#define SOR_AUDIO_CNTRL_AFIFO_FLUSH (1 << 12) + +#define SOR_AUDIO_SPARE 0xfe +#define SOR_AUDIO_SPARE_HBR_ENABLE (1 << 27) + +#define SOR_AUDIO_NVAL_0320 0xff +#define SOR_AUDIO_NVAL_0441 0x100 +#define SOR_AUDIO_NVAL_0882 0x101 +#define SOR_AUDIO_NVAL_1764 0x102 +#define SOR_AUDIO_NVAL_0480 0x103 +#define SOR_AUDIO_NVAL_0960 0x104 +#define SOR_AUDIO_NVAL_1920 0x105 + +#define SOR_AUDIO_HDA_CODEC_SCRATCH0 0x10a +#define SOR_AUDIO_HDA_CODEC_SCRATCH0_VALID (1 << 30) +#define SOR_AUDIO_HDA_CODEC_SCRATCH0_FMT_MASK 0xffff + +#define SOR_AUDIO_HDA_ELD_BUFWR 0x10c +#define SOR_AUDIO_HDA_ELD_BUFWR_INDEX(x) (((x) & 0xff) << 8) +#define SOR_AUDIO_HDA_ELD_BUFWR_DATA(x) (((x) & 0xff) << 0) + +#define SOR_AUDIO_HDA_PRESENSE 0x10d +#define SOR_AUDIO_HDA_PRESENSE_ELDV (1 << 1) +#define SOR_AUDIO_HDA_PRESENSE_PD (1 << 0) + +#define SOR_AUDIO_AVAL_0320 0x10f +#define SOR_AUDIO_AVAL_0441 0x110 +#define SOR_AUDIO_AVAL_0882 0x111 +#define SOR_AUDIO_AVAL_1764 0x112 +#define SOR_AUDIO_AVAL_0480 0x113 +#define SOR_AUDIO_AVAL_0960 0x114 +#define SOR_AUDIO_AVAL_1920 0x115 + +#define SOR_INT_STATUS 0x11c +#define SOR_INT_CODEC_CP_REQUEST (1 << 2) +#define SOR_INT_CODEC_SCRATCH1 (1 << 1) +#define SOR_INT_CODEC_SCRATCH0 (1 << 0) + +#define SOR_INT_MASK 0x11d +#define SOR_INT_ENABLE 0x11e + #define SOR_HDMI_VSI_INFOFRAME_CTRL 0x123 #define SOR_HDMI_VSI_INFOFRAME_STATUS 0x124 #define SOR_HDMI_VSI_INFOFRAME_HEADER 0x125 +#define SOR_HDMI_AUDIO_N 0x13c +#define SOR_HDMI_AUDIO_N_LOOKUP (1 << 28) +#define SOR_HDMI_AUDIO_N_RESET (1 << 20) + #define SOR_HDMI2_CTRL 0x13e #define SOR_HDMI2_CTRL_CLOCK_MODE_DIV_BY_4 (1 << 1) #define SOR_HDMI2_CTRL_SCRAMBLE (1 << 0) diff --git a/drivers/gpu/drm/tegra/vic.c b/drivers/gpu/drm/tegra/vic.c index 9f657a63b0bb..d47983deb1cf 100644 --- a/drivers/gpu/drm/tegra/vic.c +++ b/drivers/gpu/drm/tegra/vic.c @@ -38,6 +38,7 @@ struct vic { struct iommu_domain *domain; struct device *dev; struct clk *clk; + struct reset_control *rst; /* Platform configuration */ const struct vic_config *config; @@ -56,13 +57,37 @@ static void vic_writel(struct vic *vic, u32 value, unsigned int offset) static int vic_runtime_resume(struct device *dev) { struct vic *vic = dev_get_drvdata(dev); + int err; + + err = clk_prepare_enable(vic->clk); + if (err < 0) + return err; + + usleep_range(10, 20); + + err = reset_control_deassert(vic->rst); + if (err < 0) + goto disable; + + usleep_range(10, 20); + + return 0; - return clk_prepare_enable(vic->clk); +disable: + clk_disable_unprepare(vic->clk); + return err; } static int vic_runtime_suspend(struct device *dev) { struct vic *vic = dev_get_drvdata(dev); + int err; + + err = reset_control_assert(vic->rst); + if (err < 0) + return err; + + usleep_range(2000, 4000); clk_disable_unprepare(vic->clk); @@ -282,10 +307,18 @@ static const struct vic_config vic_t186_config = { .version = 0x18, }; +#define NVIDIA_TEGRA_194_VIC_FIRMWARE "nvidia/tegra194/vic.bin" + +static const struct vic_config vic_t194_config = { + .firmware = NVIDIA_TEGRA_194_VIC_FIRMWARE, + .version = 0x19, +}; + static const struct of_device_id vic_match[] = { { .compatible = "nvidia,tegra124-vic", .data = &vic_t124_config }, { .compatible = "nvidia,tegra210-vic", .data = &vic_t210_config }, { .compatible = "nvidia,tegra186-vic", .data = &vic_t186_config }, + { .compatible = "nvidia,tegra194-vic", .data = &vic_t194_config }, { }, }; @@ -323,6 +356,14 @@ static int vic_probe(struct platform_device *pdev) return PTR_ERR(vic->clk); } + if (!dev->pm_domain) { + vic->rst = devm_reset_control_get(dev, "vic"); + if (IS_ERR(vic->rst)) { + dev_err(&pdev->dev, "failed to get reset\n"); + return PTR_ERR(vic->rst); + } + } + vic->falcon.dev = dev; vic->falcon.regs = vic->regs; vic->falcon.ops = &vic_falcon_ops; @@ -418,3 +459,6 @@ MODULE_FIRMWARE(NVIDIA_TEGRA_210_VIC_FIRMWARE); #if IS_ENABLED(CONFIG_ARCH_TEGRA_186_SOC) MODULE_FIRMWARE(NVIDIA_TEGRA_186_VIC_FIRMWARE); #endif +#if IS_ENABLED(CONFIG_ARCH_TEGRA_194_SOC) +MODULE_FIRMWARE(NVIDIA_TEGRA_194_VIC_FIRMWARE); +#endif diff --git a/drivers/gpu/drm/tilcdc/tilcdc_drv.c b/drivers/gpu/drm/tilcdc/tilcdc_drv.c index 3dac08b24140..337e86a1d5ea 100644 --- a/drivers/gpu/drm/tilcdc/tilcdc_drv.c +++ b/drivers/gpu/drm/tilcdc/tilcdc_drv.c @@ -183,6 +183,12 @@ static void tilcdc_fini(struct drm_device *dev) { struct tilcdc_drm_private *priv = dev->dev_private; +#ifdef CONFIG_CPU_FREQ + if (priv->freq_transition.notifier_call) + cpufreq_unregister_notifier(&priv->freq_transition, + CPUFREQ_TRANSITION_NOTIFIER); +#endif + if (priv->crtc) tilcdc_crtc_shutdown(priv->crtc); @@ -194,12 +200,6 @@ static void tilcdc_fini(struct drm_device *dev) drm_mode_config_cleanup(dev); tilcdc_remove_external_device(dev); -#ifdef CONFIG_CPU_FREQ - if (priv->freq_transition.notifier_call) - cpufreq_unregister_notifier(&priv->freq_transition, - CPUFREQ_TRANSITION_NOTIFIER); -#endif - if (priv->clk) clk_put(priv->clk); @@ -270,17 +270,6 @@ static int tilcdc_init(struct drm_driver *ddrv, struct device *dev) goto init_failed; } -#ifdef CONFIG_CPU_FREQ - priv->freq_transition.notifier_call = cpufreq_transition; - ret = cpufreq_register_notifier(&priv->freq_transition, - CPUFREQ_TRANSITION_NOTIFIER); - if (ret) { - dev_err(dev, "failed to register cpufreq notifier\n"); - priv->freq_transition.notifier_call = NULL; - goto init_failed; - } -#endif - if (of_property_read_u32(node, "max-bandwidth", &priv->max_bandwidth)) priv->max_bandwidth = TILCDC_DEFAULT_MAX_BANDWIDTH; @@ -357,6 +346,17 @@ static int tilcdc_init(struct drm_driver *ddrv, struct device *dev) } modeset_init(ddev); +#ifdef CONFIG_CPU_FREQ + priv->freq_transition.notifier_call = cpufreq_transition; + ret = cpufreq_register_notifier(&priv->freq_transition, + CPUFREQ_TRANSITION_NOTIFIER); + if (ret) { + dev_err(dev, "failed to register cpufreq notifier\n"); + priv->freq_transition.notifier_call = NULL; + goto init_failed; + } +#endif + if (priv->is_componentized) { ret = component_bind_all(dev, ddev); if (ret < 0) diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c index d87935bf8e30..0ec08394e17a 100644 --- a/drivers/gpu/drm/ttm/ttm_bo.c +++ b/drivers/gpu/drm/ttm/ttm_bo.c @@ -77,38 +77,39 @@ static inline int ttm_mem_type_from_place(const struct ttm_place *place, return 0; } -static void ttm_mem_type_debug(struct ttm_bo_device *bdev, int mem_type) +static void ttm_mem_type_debug(struct ttm_bo_device *bdev, struct drm_printer *p, + int mem_type) { struct ttm_mem_type_manager *man = &bdev->man[mem_type]; - struct drm_printer p = drm_debug_printer(TTM_PFX); - pr_err(" has_type: %d\n", man->has_type); - pr_err(" use_type: %d\n", man->use_type); - pr_err(" flags: 0x%08X\n", man->flags); - pr_err(" gpu_offset: 0x%08llX\n", man->gpu_offset); - pr_err(" size: %llu\n", man->size); - pr_err(" available_caching: 0x%08X\n", man->available_caching); - pr_err(" default_caching: 0x%08X\n", man->default_caching); + drm_printf(p, " has_type: %d\n", man->has_type); + drm_printf(p, " use_type: %d\n", man->use_type); + drm_printf(p, " flags: 0x%08X\n", man->flags); + drm_printf(p, " gpu_offset: 0x%08llX\n", man->gpu_offset); + drm_printf(p, " size: %llu\n", man->size); + drm_printf(p, " available_caching: 0x%08X\n", man->available_caching); + drm_printf(p, " default_caching: 0x%08X\n", man->default_caching); if (mem_type != TTM_PL_SYSTEM) - (*man->func->debug)(man, &p); + (*man->func->debug)(man, p); } static void ttm_bo_mem_space_debug(struct ttm_buffer_object *bo, struct ttm_placement *placement) { + struct drm_printer p = drm_debug_printer(TTM_PFX); int i, ret, mem_type; - pr_err("No space for %p (%lu pages, %luK, %luM)\n", - bo, bo->mem.num_pages, bo->mem.size >> 10, - bo->mem.size >> 20); + drm_printf(&p, "No space for %p (%lu pages, %luK, %luM)\n", + bo, bo->mem.num_pages, bo->mem.size >> 10, + bo->mem.size >> 20); for (i = 0; i < placement->num_placement; i++) { ret = ttm_mem_type_from_place(&placement->placement[i], &mem_type); if (ret) return; - pr_err(" placement[%d]=0x%08X (%d)\n", - i, placement->placement[i].flags, mem_type); - ttm_mem_type_debug(bo->bdev, mem_type); + drm_printf(&p, " placement[%d]=0x%08X (%d)\n", + i, placement->placement[i].flags, mem_type); + ttm_mem_type_debug(bo->bdev, &p, mem_type); } } diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index ba80150d1052..895d77d799e4 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -492,8 +492,10 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo, if (!fbo) return -ENOMEM; - ttm_bo_get(bo); fbo->base = *bo; + fbo->base.mem.placement |= TTM_PL_FLAG_NO_EVICT; + + ttm_bo_get(bo); fbo->bo = bo; /** diff --git a/drivers/gpu/drm/ttm/ttm_execbuf_util.c b/drivers/gpu/drm/ttm/ttm_execbuf_util.c index efa005a1c1b7..93860346c426 100644 --- a/drivers/gpu/drm/ttm/ttm_execbuf_util.c +++ b/drivers/gpu/drm/ttm/ttm_execbuf_util.c @@ -126,10 +126,11 @@ int ttm_eu_reserve_buffers(struct ww_acquire_ctx *ticket, } if (!ret) { - if (!entry->shared) + if (!entry->num_shared) continue; - ret = reservation_object_reserve_shared(bo->resv, 1); + ret = reservation_object_reserve_shared(bo->resv, + entry->num_shared); if (!ret) continue; } @@ -150,8 +151,9 @@ int ttm_eu_reserve_buffers(struct ww_acquire_ctx *ticket, } } - if (!ret && entry->shared) - ret = reservation_object_reserve_shared(bo->resv, 1); + if (!ret && entry->num_shared) + ret = reservation_object_reserve_shared(bo->resv, + entry->num_shared); if (unlikely(ret != 0)) { if (ret == -EINTR) @@ -199,7 +201,7 @@ void ttm_eu_fence_buffer_objects(struct ww_acquire_ctx *ticket, list_for_each_entry(entry, list, head) { bo = entry->bo; - if (entry->shared) + if (entry->num_shared) reservation_object_add_shared_fence(bo->resv, fence); else reservation_object_add_excl_fence(bo->resv, fence); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index b9c078860a7c..25afb1d594e3 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -34,7 +34,7 @@ #include <drm/ttm/ttm_placement.h> #include <drm/ttm/ttm_bo_driver.h> #include <drm/ttm/ttm_module.h> -#include <linux/dma_remapping.h> +#include <linux/intel-iommu.h> #define VMWGFX_DRIVER_DESC "Linux drm driver for VMware graphics devices" #define VMWGFX_CHIP_SVGAII 0 @@ -49,6 +49,8 @@ #define VMWGFX_REPO "In Tree" +#define VMWGFX_VALIDATION_MEM_GRAN (16*PAGE_SIZE) + /** * Fully encoded drm commands. Might move to vmw_drm.h @@ -581,7 +583,7 @@ static int vmw_dma_select_mode(struct vmw_private *dev_priv) dev_priv->map_mode = vmw_dma_map_populate; - if (dma_ops->sync_single_for_cpu) + if (dma_ops && dma_ops->sync_single_for_cpu) dev_priv->map_mode = vmw_dma_alloc_coherent; #ifdef CONFIG_SWIOTLB if (swiotlb_nr_tbl() == 0) @@ -665,7 +667,6 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) mutex_init(&dev_priv->cmdbuf_mutex); mutex_init(&dev_priv->release_mutex); mutex_init(&dev_priv->binding_mutex); - mutex_init(&dev_priv->requested_layout_mutex); mutex_init(&dev_priv->global_kms_state_mutex); ttm_lock_init(&dev_priv->reservation_sem); spin_lock_init(&dev_priv->resource_lock); @@ -912,7 +913,7 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) spin_unlock(&dev_priv->cap_lock); } - + vmw_validation_mem_init_ttm(dev_priv, VMWGFX_VALIDATION_MEM_GRAN); ret = vmw_kms_init(dev_priv); if (unlikely(ret != 0)) goto out_no_kms; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index 28df788da44e..cd607ba9c2fe 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -466,15 +466,6 @@ struct vmw_private { uint32_t num_displays; /* - * Currently requested_layout_mutex is used to protect the gui - * positionig state in display unit. With that use case currently this - * mutex is only taken during layout ioctl and atomic check_modeset. - * Other display unit state can be protected with this mutex but that - * needs careful consideration. - */ - struct mutex requested_layout_mutex; - - /* * Framebuffer info. */ @@ -484,8 +475,6 @@ struct vmw_private { struct vmw_overlay *overlay_priv; struct drm_property *hotplug_mode_update_property; struct drm_property *implicit_placement_property; - unsigned num_implicit; - struct vmw_framebuffer *implicit_fb; struct mutex global_kms_state_mutex; spinlock_t cursor_lock; struct drm_atomic_state *suspend_state; @@ -604,6 +593,9 @@ struct vmw_private { struct vmw_cmdbuf_man *cman; DECLARE_BITMAP(irqthread_pending, VMW_IRQTHREAD_MAX); + + /* Validation memory reservation */ + struct vmw_validation_mem vvm; }; static inline struct vmw_surface *vmw_res_to_srf(struct vmw_resource *res) @@ -842,6 +834,8 @@ extern int vmw_fifo_flush(struct vmw_private *dev_priv, extern int vmw_mmap(struct file *filp, struct vm_area_struct *vma); +extern void vmw_validation_mem_init_ttm(struct vmw_private *dev_priv, + size_t gran); /** * TTM buffer object driver - vmwgfx_ttm_buffer.c */ diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index 5a6b70ba137a..f2d13a72c05d 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -1738,7 +1738,6 @@ static int vmw_cmd_check_define_gmrfb(struct vmw_private *dev_priv, void *buf) { struct vmw_buffer_object *vmw_bo; - int ret; struct { uint32_t header; @@ -1748,7 +1747,6 @@ static int vmw_cmd_check_define_gmrfb(struct vmw_private *dev_priv, return vmw_translate_guest_ptr(dev_priv, sw_context, &cmd->body.ptr, &vmw_bo); - return ret; } @@ -3837,6 +3835,8 @@ int vmw_execbuf_process(struct drm_file *file_priv, struct sync_file *sync_file = NULL; DECLARE_VAL_CONTEXT(val_ctx, &sw_context->res_ht, 1); + vmw_validation_set_val_mem(&val_ctx, &dev_priv->vvm); + if (flags & DRM_VMW_EXECBUF_FLAG_EXPORT_FENCE_FD) { out_fence_fd = get_unused_fd_flags(O_CLOEXEC); if (out_fence_fd < 0) { diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c index f87261545f2c..301260e23e52 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_fence.c @@ -906,13 +906,10 @@ static void vmw_event_fence_action_seq_passed(struct vmw_fence_action *action) container_of(action, struct vmw_event_fence_action, action); struct drm_device *dev = eaction->dev; struct drm_pending_event *event = eaction->event; - struct drm_file *file_priv; - if (unlikely(event == NULL)) return; - file_priv = event->file_priv; spin_lock_irq(&dev->event_lock); if (likely(eaction->tv_sec != NULL)) { diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c index e6b11f6ae2e4..b351fb5214d3 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c @@ -30,6 +30,7 @@ #include <drm/drm_atomic.h> #include <drm/drm_atomic_helper.h> #include <drm/drm_rect.h> +#include <drm/drm_damage_helper.h> /* Might need a hrtimer here? */ #define VMWGFX_PRESENT_RATE ((HZ / 60 > 0) ? HZ / 60 : 1) @@ -456,21 +457,8 @@ int vmw_du_primary_plane_atomic_check(struct drm_plane *plane, struct drm_crtc *crtc = state->crtc; struct vmw_connector_state *vcs; struct vmw_display_unit *du = vmw_crtc_to_du(crtc); - struct vmw_private *dev_priv = vmw_priv(crtc->dev); - struct vmw_framebuffer *vfb = vmw_framebuffer_to_vfb(new_fb); vcs = vmw_connector_state_to_vcs(du->connector.state); - - /* Only one active implicit framebuffer at a time. */ - mutex_lock(&dev_priv->global_kms_state_mutex); - if (vcs->is_implicit && dev_priv->implicit_fb && - !(dev_priv->num_implicit == 1 && du->active_implicit) - && dev_priv->implicit_fb != vfb) { - DRM_ERROR("Multiple implicit framebuffers " - "not supported.\n"); - ret = -EINVAL; - } - mutex_unlock(&dev_priv->global_kms_state_mutex); } @@ -846,58 +834,6 @@ static void vmw_framebuffer_surface_destroy(struct drm_framebuffer *framebuffer) kfree(vfbs); } -static int vmw_framebuffer_surface_dirty(struct drm_framebuffer *framebuffer, - struct drm_file *file_priv, - unsigned flags, unsigned color, - struct drm_clip_rect *clips, - unsigned num_clips) -{ - struct vmw_private *dev_priv = vmw_priv(framebuffer->dev); - struct vmw_framebuffer_surface *vfbs = - vmw_framebuffer_to_vfbs(framebuffer); - struct drm_clip_rect norect; - int ret, inc = 1; - - /* Legacy Display Unit does not support 3D */ - if (dev_priv->active_display_unit == vmw_du_legacy) - return -EINVAL; - - drm_modeset_lock_all(dev_priv->dev); - - ret = ttm_read_lock(&dev_priv->reservation_sem, true); - if (unlikely(ret != 0)) { - drm_modeset_unlock_all(dev_priv->dev); - return ret; - } - - if (!num_clips) { - num_clips = 1; - clips = &norect; - norect.x1 = norect.y1 = 0; - norect.x2 = framebuffer->width; - norect.y2 = framebuffer->height; - } else if (flags & DRM_MODE_FB_DIRTY_ANNOTATE_COPY) { - num_clips /= 2; - inc = 2; /* skip source rects */ - } - - if (dev_priv->active_display_unit == vmw_du_screen_object) - ret = vmw_kms_sou_do_surface_dirty(dev_priv, &vfbs->base, - clips, NULL, NULL, 0, 0, - num_clips, inc, NULL, NULL); - else - ret = vmw_kms_stdu_surface_dirty(dev_priv, &vfbs->base, - clips, NULL, NULL, 0, 0, - num_clips, inc, NULL, NULL); - - vmw_fifo_flush(dev_priv, false); - ttm_read_unlock(&dev_priv->reservation_sem); - - drm_modeset_unlock_all(dev_priv->dev); - - return 0; -} - /** * vmw_kms_readback - Perform a readback from the screen system to * a buffer-object backed framebuffer. @@ -941,7 +877,7 @@ int vmw_kms_readback(struct vmw_private *dev_priv, static const struct drm_framebuffer_funcs vmw_framebuffer_surface_funcs = { .destroy = vmw_framebuffer_surface_destroy, - .dirty = vmw_framebuffer_surface_dirty, + .dirty = drm_atomic_helper_dirtyfb, }; static int vmw_kms_new_framebuffer_surface(struct vmw_private *dev_priv, @@ -1084,16 +1020,6 @@ static int vmw_framebuffer_bo_dirty(struct drm_framebuffer *framebuffer, } switch (dev_priv->active_display_unit) { - case vmw_du_screen_target: - ret = vmw_kms_stdu_dma(dev_priv, NULL, &vfbd->base, NULL, - clips, NULL, num_clips, increment, - true, true, NULL); - break; - case vmw_du_screen_object: - ret = vmw_kms_sou_do_bo_dirty(dev_priv, &vfbd->base, - clips, NULL, num_clips, - increment, true, NULL, NULL); - break; case vmw_du_legacy: ret = vmw_kms_ldu_do_bo_dirty(dev_priv, &vfbd->base, 0, 0, clips, num_clips, increment); @@ -1112,9 +1038,25 @@ static int vmw_framebuffer_bo_dirty(struct drm_framebuffer *framebuffer, return ret; } +static int vmw_framebuffer_bo_dirty_ext(struct drm_framebuffer *framebuffer, + struct drm_file *file_priv, + unsigned int flags, unsigned int color, + struct drm_clip_rect *clips, + unsigned int num_clips) +{ + struct vmw_private *dev_priv = vmw_priv(framebuffer->dev); + + if (dev_priv->active_display_unit == vmw_du_legacy) + return vmw_framebuffer_bo_dirty(framebuffer, file_priv, flags, + color, clips, num_clips); + + return drm_atomic_helper_dirtyfb(framebuffer, file_priv, flags, color, + clips, num_clips); +} + static const struct drm_framebuffer_funcs vmw_framebuffer_bo_funcs = { .destroy = vmw_framebuffer_bo_destroy, - .dirty = vmw_framebuffer_bo_dirty, + .dirty = vmw_framebuffer_bo_dirty_ext, }; /** @@ -1565,6 +1507,88 @@ static int vmw_kms_check_display_memory(struct drm_device *dev, } /** + * vmw_crtc_state_and_lock - Return new or current crtc state with locked + * crtc mutex + * @state: The atomic state pointer containing the new atomic state + * @crtc: The crtc + * + * This function returns the new crtc state if it's part of the state update. + * Otherwise returns the current crtc state. It also makes sure that the + * crtc mutex is locked. + * + * Returns: A valid crtc state pointer or NULL. It may also return a + * pointer error, in particular -EDEADLK if locking needs to be rerun. + */ +static struct drm_crtc_state * +vmw_crtc_state_and_lock(struct drm_atomic_state *state, struct drm_crtc *crtc) +{ + struct drm_crtc_state *crtc_state; + + crtc_state = drm_atomic_get_new_crtc_state(state, crtc); + if (crtc_state) { + lockdep_assert_held(&crtc->mutex.mutex.base); + } else { + int ret = drm_modeset_lock(&crtc->mutex, state->acquire_ctx); + + if (ret != 0 && ret != -EALREADY) + return ERR_PTR(ret); + + crtc_state = crtc->state; + } + + return crtc_state; +} + +/** + * vmw_kms_check_implicit - Verify that all implicit display units scan out + * from the same fb after the new state is committed. + * @dev: The drm_device. + * @state: The new state to be checked. + * + * Returns: + * Zero on success, + * -EINVAL on invalid state, + * -EDEADLK if modeset locking needs to be rerun. + */ +static int vmw_kms_check_implicit(struct drm_device *dev, + struct drm_atomic_state *state) +{ + struct drm_framebuffer *implicit_fb = NULL; + struct drm_crtc *crtc; + struct drm_crtc_state *crtc_state; + struct drm_plane_state *plane_state; + + drm_for_each_crtc(crtc, dev) { + struct vmw_display_unit *du = vmw_crtc_to_du(crtc); + + if (!du->is_implicit) + continue; + + crtc_state = vmw_crtc_state_and_lock(state, crtc); + if (IS_ERR(crtc_state)) + return PTR_ERR(crtc_state); + + if (!crtc_state || !crtc_state->enable) + continue; + + /* + * Can't move primary planes across crtcs, so this is OK. + * It also means we don't need to take the plane mutex. + */ + plane_state = du->primary.state; + if (plane_state->crtc != crtc) + continue; + + if (!implicit_fb) + implicit_fb = plane_state->fb; + else if (implicit_fb != plane_state->fb) + return -EINVAL; + } + + return 0; +} + +/** * vmw_kms_check_topology - Validates topology in drm_atomic_state * @dev: DRM device * @state: the driver state object @@ -1575,7 +1599,6 @@ static int vmw_kms_check_display_memory(struct drm_device *dev, static int vmw_kms_check_topology(struct drm_device *dev, struct drm_atomic_state *state) { - struct vmw_private *dev_priv = vmw_priv(dev); struct drm_crtc_state *old_crtc_state, *new_crtc_state; struct drm_rect *rects; struct drm_crtc *crtc; @@ -1587,19 +1610,31 @@ static int vmw_kms_check_topology(struct drm_device *dev, if (!rects) return -ENOMEM; - mutex_lock(&dev_priv->requested_layout_mutex); - drm_for_each_crtc(crtc, dev) { struct vmw_display_unit *du = vmw_crtc_to_du(crtc); - struct drm_crtc_state *crtc_state = crtc->state; + struct drm_crtc_state *crtc_state; i = drm_crtc_index(crtc); - if (crtc_state && crtc_state->enable) { + crtc_state = vmw_crtc_state_and_lock(state, crtc); + if (IS_ERR(crtc_state)) { + ret = PTR_ERR(crtc_state); + goto clean; + } + + if (!crtc_state) + continue; + + if (crtc_state->enable) { rects[i].x1 = du->gui_x; rects[i].y1 = du->gui_y; rects[i].x2 = du->gui_x + crtc_state->mode.hdisplay; rects[i].y2 = du->gui_y + crtc_state->mode.vdisplay; + } else { + rects[i].x1 = 0; + rects[i].y1 = 0; + rects[i].x2 = 0; + rects[i].y2 = 0; } } @@ -1611,14 +1646,6 @@ static int vmw_kms_check_topology(struct drm_device *dev, struct drm_connector_state *conn_state; struct vmw_connector_state *vmw_conn_state; - if (!new_crtc_state->enable) { - rects[i].x1 = 0; - rects[i].y1 = 0; - rects[i].x2 = 0; - rects[i].y2 = 0; - continue; - } - if (!du->pref_active) { ret = -EINVAL; goto clean; @@ -1639,18 +1666,12 @@ static int vmw_kms_check_topology(struct drm_device *dev, vmw_conn_state = vmw_connector_state_to_vcs(conn_state); vmw_conn_state->gui_x = du->gui_x; vmw_conn_state->gui_y = du->gui_y; - - rects[i].x1 = du->gui_x; - rects[i].y1 = du->gui_y; - rects[i].x2 = du->gui_x + new_crtc_state->mode.hdisplay; - rects[i].y2 = du->gui_y + new_crtc_state->mode.vdisplay; } ret = vmw_kms_check_display_memory(dev, dev->mode_config.num_crtc, rects); clean: - mutex_unlock(&dev_priv->requested_layout_mutex); kfree(rects); return ret; } @@ -1681,6 +1702,10 @@ vmw_kms_atomic_check_modeset(struct drm_device *dev, if (ret) return ret; + ret = vmw_kms_check_implicit(dev, state); + if (ret) + return ret; + if (!state->allow_modeset) return ret; @@ -2003,11 +2028,25 @@ static int vmw_du_update_layout(struct vmw_private *dev_priv, struct vmw_display_unit *du; struct drm_connector *con; struct drm_connector_list_iter conn_iter; + struct drm_modeset_acquire_ctx ctx; + struct drm_crtc *crtc; + int ret; + + /* Currently gui_x/y is protected with the crtc mutex */ + mutex_lock(&dev->mode_config.mutex); + drm_modeset_acquire_init(&ctx, 0); +retry: + drm_for_each_crtc(crtc, dev) { + ret = drm_modeset_lock(&crtc->mutex, &ctx); + if (ret < 0) { + if (ret == -EDEADLK) { + drm_modeset_backoff(&ctx); + goto retry; + } + goto out_fini; + } + } - /* - * Currently only gui_x/y is protected with requested_layout_mutex. - */ - mutex_lock(&dev_priv->requested_layout_mutex); drm_connector_list_iter_begin(dev, &conn_iter); drm_for_each_connector_iter(con, &conn_iter) { du = vmw_connector_to_du(con); @@ -2026,9 +2065,7 @@ static int vmw_du_update_layout(struct vmw_private *dev_priv, } } drm_connector_list_iter_end(&conn_iter); - mutex_unlock(&dev_priv->requested_layout_mutex); - mutex_lock(&dev->mode_config.mutex); list_for_each_entry(con, &dev->mode_config.connector_list, head) { du = vmw_connector_to_du(con); if (num_rects > du->unit) { @@ -2048,10 +2085,13 @@ static int vmw_du_update_layout(struct vmw_private *dev_priv, } con->status = vmw_du_connector_detect(con, true); } - mutex_unlock(&dev->mode_config.mutex); drm_sysfs_hotplug_event(dev); - +out_fini: + drm_modeset_drop_locks(&ctx); + drm_modeset_acquire_fini(&ctx); + mutex_unlock(&dev->mode_config.mutex); + return 0; } @@ -2275,84 +2315,6 @@ int vmw_du_connector_fill_modes(struct drm_connector *connector, return 1; } -int vmw_du_connector_set_property(struct drm_connector *connector, - struct drm_property *property, - uint64_t val) -{ - struct vmw_display_unit *du = vmw_connector_to_du(connector); - struct vmw_private *dev_priv = vmw_priv(connector->dev); - - if (property == dev_priv->implicit_placement_property) - du->is_implicit = val; - - return 0; -} - - - -/** - * vmw_du_connector_atomic_set_property - Atomic version of get property - * - * @crtc - crtc the property is associated with - * - * Returns: - * Zero on success, negative errno on failure. - */ -int -vmw_du_connector_atomic_set_property(struct drm_connector *connector, - struct drm_connector_state *state, - struct drm_property *property, - uint64_t val) -{ - struct vmw_private *dev_priv = vmw_priv(connector->dev); - struct vmw_connector_state *vcs = vmw_connector_state_to_vcs(state); - struct vmw_display_unit *du = vmw_connector_to_du(connector); - - - if (property == dev_priv->implicit_placement_property) { - vcs->is_implicit = val; - - /* - * We should really be doing a drm_atomic_commit() to - * commit the new state, but since this doesn't cause - * an immedate state change, this is probably ok - */ - du->is_implicit = vcs->is_implicit; - } else { - return -EINVAL; - } - - return 0; -} - - -/** - * vmw_du_connector_atomic_get_property - Atomic version of get property - * - * @connector - connector the property is associated with - * - * Returns: - * Zero on success, negative errno on failure. - */ -int -vmw_du_connector_atomic_get_property(struct drm_connector *connector, - const struct drm_connector_state *state, - struct drm_property *property, - uint64_t *val) -{ - struct vmw_private *dev_priv = vmw_priv(connector->dev); - struct vmw_connector_state *vcs = vmw_connector_state_to_vcs(state); - - if (property == dev_priv->implicit_placement_property) - *val = vcs->is_implicit; - else { - DRM_ERROR("Invalid Property %s\n", property->name); - return -EINVAL; - } - - return 0; -} - /** * vmw_kms_update_layout_ioctl - Handler for DRM_VMW_UPDATE_LAYOUT ioctl * @dev: drm device for the ioctl @@ -2742,143 +2704,25 @@ int vmw_kms_fbdev_init_data(struct vmw_private *dev_priv, } /** - * vmw_kms_del_active - unregister a crtc binding to the implicit framebuffer - * - * @dev_priv: Pointer to a device private struct. - * @du: The display unit of the crtc. - */ -void vmw_kms_del_active(struct vmw_private *dev_priv, - struct vmw_display_unit *du) -{ - mutex_lock(&dev_priv->global_kms_state_mutex); - if (du->active_implicit) { - if (--(dev_priv->num_implicit) == 0) - dev_priv->implicit_fb = NULL; - du->active_implicit = false; - } - mutex_unlock(&dev_priv->global_kms_state_mutex); -} - -/** - * vmw_kms_add_active - register a crtc binding to an implicit framebuffer - * - * @vmw_priv: Pointer to a device private struct. - * @du: The display unit of the crtc. - * @vfb: The implicit framebuffer - * - * Registers a binding to an implicit framebuffer. - */ -void vmw_kms_add_active(struct vmw_private *dev_priv, - struct vmw_display_unit *du, - struct vmw_framebuffer *vfb) -{ - mutex_lock(&dev_priv->global_kms_state_mutex); - WARN_ON_ONCE(!dev_priv->num_implicit && dev_priv->implicit_fb); - - if (!du->active_implicit && du->is_implicit) { - dev_priv->implicit_fb = vfb; - du->active_implicit = true; - dev_priv->num_implicit++; - } - mutex_unlock(&dev_priv->global_kms_state_mutex); -} - -/** - * vmw_kms_screen_object_flippable - Check whether we can page-flip a crtc. - * - * @dev_priv: Pointer to device-private struct. - * @crtc: The crtc we want to flip. - * - * Returns true or false depending whether it's OK to flip this crtc - * based on the criterion that we must not have more than one implicit - * frame-buffer at any one time. - */ -bool vmw_kms_crtc_flippable(struct vmw_private *dev_priv, - struct drm_crtc *crtc) -{ - struct vmw_display_unit *du = vmw_crtc_to_du(crtc); - bool ret; - - mutex_lock(&dev_priv->global_kms_state_mutex); - ret = !du->is_implicit || dev_priv->num_implicit == 1; - mutex_unlock(&dev_priv->global_kms_state_mutex); - - return ret; -} - -/** - * vmw_kms_update_implicit_fb - Update the implicit fb. - * - * @dev_priv: Pointer to device-private struct. - * @crtc: The crtc the new implicit frame-buffer is bound to. - */ -void vmw_kms_update_implicit_fb(struct vmw_private *dev_priv, - struct drm_crtc *crtc) -{ - struct vmw_display_unit *du = vmw_crtc_to_du(crtc); - struct drm_plane *plane = crtc->primary; - struct vmw_framebuffer *vfb; - - mutex_lock(&dev_priv->global_kms_state_mutex); - - if (!du->is_implicit) - goto out_unlock; - - vfb = vmw_framebuffer_to_vfb(plane->state->fb); - WARN_ON_ONCE(dev_priv->num_implicit != 1 && - dev_priv->implicit_fb != vfb); - - dev_priv->implicit_fb = vfb; -out_unlock: - mutex_unlock(&dev_priv->global_kms_state_mutex); -} - -/** * vmw_kms_create_implicit_placement_proparty - Set up the implicit placement * property. * * @dev_priv: Pointer to a device private struct. - * @immutable: Whether the property is immutable. * * Sets up the implicit placement property unless it's already set up. */ void -vmw_kms_create_implicit_placement_property(struct vmw_private *dev_priv, - bool immutable) +vmw_kms_create_implicit_placement_property(struct vmw_private *dev_priv) { if (dev_priv->implicit_placement_property) return; dev_priv->implicit_placement_property = drm_property_create_range(dev_priv->dev, - immutable ? - DRM_MODE_PROP_IMMUTABLE : 0, + DRM_MODE_PROP_IMMUTABLE, "implicit_placement", 0, 1); - -} - - -/** - * vmw_kms_set_config - Wrapper around drm_atomic_helper_set_config - * - * @set: The configuration to set. - * - * The vmwgfx Xorg driver doesn't assign the mode::type member, which - * when drm_mode_set_crtcinfo is called as part of the configuration setting - * causes it to return incorrect crtc dimensions causing severe problems in - * the vmwgfx modesetting. So explicitly clear that member before calling - * into drm_atomic_helper_set_config. - */ -int vmw_kms_set_config(struct drm_mode_set *set, - struct drm_modeset_acquire_ctx *ctx) -{ - if (set && set->mode) - set->mode->type = 0; - - return drm_atomic_helper_set_config(set, ctx); } - /** * vmw_kms_suspend - Save modesetting state and turn modesetting off. * @@ -2935,3 +2779,124 @@ void vmw_kms_lost_device(struct drm_device *dev) { drm_atomic_helper_shutdown(dev); } + +/** + * vmw_du_helper_plane_update - Helper to do plane update on a display unit. + * @update: The closure structure. + * + * Call this helper after setting callbacks in &vmw_du_update_plane to do plane + * update on display unit. + * + * Return: 0 on success or a negative error code on failure. + */ +int vmw_du_helper_plane_update(struct vmw_du_update_plane *update) +{ + struct drm_plane_state *state = update->plane->state; + struct drm_plane_state *old_state = update->old_state; + struct drm_atomic_helper_damage_iter iter; + struct drm_rect clip; + struct drm_rect bb; + DECLARE_VAL_CONTEXT(val_ctx, NULL, 0); + uint32_t reserved_size = 0; + uint32_t submit_size = 0; + uint32_t curr_size = 0; + uint32_t num_hits = 0; + void *cmd_start; + char *cmd_next; + int ret; + + /* + * Iterate in advance to check if really need plane update and find the + * number of clips that actually are in plane src for fifo allocation. + */ + drm_atomic_helper_damage_iter_init(&iter, old_state, state); + drm_atomic_for_each_plane_damage(&iter, &clip) + num_hits++; + + if (num_hits == 0) + return 0; + + if (update->vfb->bo) { + struct vmw_framebuffer_bo *vfbbo = + container_of(update->vfb, typeof(*vfbbo), base); + + ret = vmw_validation_add_bo(&val_ctx, vfbbo->buffer, false, + update->cpu_blit); + } else { + struct vmw_framebuffer_surface *vfbs = + container_of(update->vfb, typeof(*vfbs), base); + + ret = vmw_validation_add_resource(&val_ctx, &vfbs->surface->res, + 0, NULL, NULL); + } + + if (ret) + return ret; + + ret = vmw_validation_prepare(&val_ctx, update->mutex, update->intr); + if (ret) + goto out_unref; + + reserved_size = update->calc_fifo_size(update, num_hits); + cmd_start = vmw_fifo_reserve(update->dev_priv, reserved_size); + if (!cmd_start) { + ret = -ENOMEM; + goto out_revert; + } + + cmd_next = cmd_start; + + if (update->post_prepare) { + curr_size = update->post_prepare(update, cmd_next); + cmd_next += curr_size; + submit_size += curr_size; + } + + if (update->pre_clip) { + curr_size = update->pre_clip(update, cmd_next, num_hits); + cmd_next += curr_size; + submit_size += curr_size; + } + + bb.x1 = INT_MAX; + bb.y1 = INT_MAX; + bb.x2 = INT_MIN; + bb.y2 = INT_MIN; + + drm_atomic_helper_damage_iter_init(&iter, old_state, state); + drm_atomic_for_each_plane_damage(&iter, &clip) { + uint32_t fb_x = clip.x1; + uint32_t fb_y = clip.y1; + + vmw_du_translate_to_crtc(state, &clip); + if (update->clip) { + curr_size = update->clip(update, cmd_next, &clip, fb_x, + fb_y); + cmd_next += curr_size; + submit_size += curr_size; + } + bb.x1 = min_t(int, bb.x1, clip.x1); + bb.y1 = min_t(int, bb.y1, clip.y1); + bb.x2 = max_t(int, bb.x2, clip.x2); + bb.y2 = max_t(int, bb.y2, clip.y2); + } + + curr_size = update->post_clip(update, cmd_next, &bb); + submit_size += curr_size; + + if (reserved_size < submit_size) + submit_size = 0; + + vmw_fifo_commit(update->dev_priv, submit_size); + + vmw_kms_helper_validation_finish(update->dev_priv, NULL, &val_ctx, + update->out_fence, NULL); + return ret; + +out_revert: + vmw_validation_revert(&val_ctx); + +out_unref: + vmw_validation_unref_lists(&val_ctx); + return ret; +} diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h index 76ec570c0684..655abbcd4058 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h @@ -33,7 +33,123 @@ #include <drm/drm_encoder.h> #include "vmwgfx_drv.h" +/** + * struct vmw_du_update_plane - Closure structure for vmw_du_helper_plane_update + * @plane: Plane which is being updated. + * @old_state: Old state of plane. + * @dev_priv: Device private. + * @du: Display unit on which to update the plane. + * @vfb: Framebuffer which is blitted to display unit. + * @out_fence: Out fence for resource finish. + * @mutex: The mutex used to protect resource reservation. + * @cpu_blit: True if need cpu blit. + * @intr: Whether to perform waits interruptible if possible. + * + * This structure loosely represent the set of operations needed to perform a + * plane update on a display unit. Implementer will define that functionality + * according to the function callbacks for this structure. In brief it involves + * surface/buffer object validation, populate FIFO commands and command + * submission to the device. + */ +struct vmw_du_update_plane { + /** + * @calc_fifo_size: Calculate fifo size. + * + * Determine fifo size for the commands needed for update. The number of + * damage clips on display unit @num_hits will be passed to allocate + * sufficient fifo space. + * + * Return: Fifo size needed + */ + uint32_t (*calc_fifo_size)(struct vmw_du_update_plane *update, + uint32_t num_hits); + + /** + * @post_prepare: Populate fifo for resource preparation. + * + * Some surface resource or buffer object need some extra cmd submission + * like update GB image for proxy surface and define a GMRFB for screen + * object. That should should be done here as this callback will be + * called after FIFO allocation with the address of command buufer. + * + * This callback is optional. + * + * Return: Size of commands populated to command buffer. + */ + uint32_t (*post_prepare)(struct vmw_du_update_plane *update, void *cmd); + + /** + * @pre_clip: Populate fifo before clip. + * + * This is where pre clip related command should be populated like + * surface copy/DMA, etc. + * + * This callback is optional. + * + * Return: Size of commands populated to command buffer. + */ + uint32_t (*pre_clip)(struct vmw_du_update_plane *update, void *cmd, + uint32_t num_hits); + + /** + * @clip: Populate fifo for clip. + * + * This is where to populate clips for surface copy/dma or blit commands + * if needed. This will be called times have damage in display unit, + * which is one if doing full update. @clip is the damage in destination + * coordinates which is crtc/DU and @src_x, @src_y is damage clip src in + * framebuffer coordinate. + * + * This callback is optional. + * + * Return: Size of commands populated to command buffer. + */ + uint32_t (*clip)(struct vmw_du_update_plane *update, void *cmd, + struct drm_rect *clip, uint32_t src_x, uint32_t src_y); + + /** + * @post_clip: Populate fifo after clip. + * + * This is where to populate display unit update commands or blit + * commands. + * + * Return: Size of commands populated to command buffer. + */ + uint32_t (*post_clip)(struct vmw_du_update_plane *update, void *cmd, + struct drm_rect *bb); + + struct drm_plane *plane; + struct drm_plane_state *old_state; + struct vmw_private *dev_priv; + struct vmw_display_unit *du; + struct vmw_framebuffer *vfb; + struct vmw_fence_obj **out_fence; + struct mutex *mutex; + bool cpu_blit; + bool intr; +}; + +/** + * struct vmw_du_update_plane_surface - closure structure for surface + * @base: base closure structure. + * @cmd_start: FIFO command start address (used by SOU only). + */ +struct vmw_du_update_plane_surface { + struct vmw_du_update_plane base; + /* This member is to handle special case SOU surface update */ + void *cmd_start; +}; +/** + * struct vmw_du_update_plane_buffer - Closure structure for buffer object + * @base: Base closure structure. + * @fb_left: x1 for fb damage bounding box. + * @fb_top: y1 for fb damage bounding box. + */ +struct vmw_du_update_plane_buffer { + struct vmw_du_update_plane base; + int fb_left, fb_top; +}; /** * struct vmw_kms_dirty - closure structure for the vmw_kms_helper_dirty @@ -191,8 +307,6 @@ struct vmw_plane_state { struct vmw_connector_state { struct drm_connector_state base; - bool is_implicit; - /** * @gui_x: * @@ -254,7 +368,6 @@ struct vmw_display_unit { int gui_x; int gui_y; bool is_implicit; - bool active_implicit; int set_gui_x; int set_gui_y; }; @@ -334,17 +447,8 @@ int vmw_kms_fbdev_init_data(struct vmw_private *dev_priv, struct drm_crtc **p_crtc, struct drm_display_mode **p_mode); void vmw_guess_mode_timing(struct drm_display_mode *mode); -void vmw_kms_del_active(struct vmw_private *dev_priv, - struct vmw_display_unit *du); -void vmw_kms_add_active(struct vmw_private *dev_priv, - struct vmw_display_unit *du, - struct vmw_framebuffer *vfb); -bool vmw_kms_crtc_flippable(struct vmw_private *dev_priv, - struct drm_crtc *crtc); -void vmw_kms_update_implicit_fb(struct vmw_private *dev_priv, - struct drm_crtc *crtc); -void vmw_kms_create_implicit_placement_property(struct vmw_private *dev_priv, - bool immutable); +void vmw_kms_update_implicit_fb(struct vmw_private *dev_priv); +void vmw_kms_create_implicit_placement_property(struct vmw_private *dev_priv); /* Universal Plane Helpers */ void vmw_du_primary_plane_destroy(struct drm_plane *plane); @@ -456,6 +560,20 @@ int vmw_kms_stdu_dma(struct vmw_private *dev_priv, bool interruptible, struct drm_crtc *crtc); -int vmw_kms_set_config(struct drm_mode_set *set, - struct drm_modeset_acquire_ctx *ctx); +int vmw_du_helper_plane_update(struct vmw_du_update_plane *update); + +/** + * vmw_du_translate_to_crtc - Translate a rect from framebuffer to crtc + * @state: Plane state. + * @r: Rectangle to translate. + */ +static inline void vmw_du_translate_to_crtc(struct drm_plane_state *state, + struct drm_rect *r) +{ + int translate_crtc_x = -((state->src_x >> 16) - state->crtc_x); + int translate_crtc_y = -((state->src_y >> 16) - state->crtc_y); + + drm_rect_translate(r, translate_crtc_x, translate_crtc_y); +} + #endif diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c index 4b5378495eea..16be515c4c0f 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c @@ -233,7 +233,7 @@ static const struct drm_crtc_funcs vmw_legacy_crtc_funcs = { .reset = vmw_du_crtc_reset, .atomic_duplicate_state = vmw_du_crtc_duplicate_state, .atomic_destroy_state = vmw_du_crtc_destroy_state, - .set_config = vmw_kms_set_config, + .set_config = drm_atomic_helper_set_config, }; @@ -263,13 +263,10 @@ static const struct drm_connector_funcs vmw_legacy_connector_funcs = { .dpms = vmw_du_connector_dpms, .detect = vmw_du_connector_detect, .fill_modes = vmw_du_connector_fill_modes, - .set_property = vmw_du_connector_set_property, .destroy = vmw_ldu_connector_destroy, .reset = vmw_du_connector_reset, .atomic_duplicate_state = vmw_du_connector_duplicate_state, .atomic_destroy_state = vmw_du_connector_destroy_state, - .atomic_set_property = vmw_du_connector_atomic_set_property, - .atomic_get_property = vmw_du_connector_atomic_get_property, }; static const struct @@ -416,7 +413,6 @@ static int vmw_ldu_init(struct vmw_private *dev_priv, unsigned unit) drm_plane_helper_add(cursor, &vmw_ldu_cursor_plane_helper_funcs); - vmw_du_connector_reset(connector); ret = drm_connector_init(dev, connector, &vmw_legacy_connector_funcs, DRM_MODE_CONNECTOR_VIRTUAL); @@ -427,8 +423,6 @@ static int vmw_ldu_init(struct vmw_private *dev_priv, unsigned unit) drm_connector_helper_add(connector, &vmw_ldu_connector_helper_funcs); connector->status = vmw_du_connector_detect(connector, true); - vmw_connector_state_to_vcs(connector->state)->is_implicit = true; - ret = drm_encoder_init(dev, encoder, &vmw_legacy_encoder_funcs, DRM_MODE_ENCODER_VIRTUAL, NULL); @@ -447,7 +441,6 @@ static int vmw_ldu_init(struct vmw_private *dev_priv, unsigned unit) goto err_free_encoder; } - vmw_du_crtc_reset(crtc); ret = drm_crtc_init_with_planes(dev, crtc, &ldu->base.primary, &ldu->base.cursor, @@ -513,7 +506,7 @@ int vmw_kms_ldu_init_display(struct vmw_private *dev_priv) if (ret != 0) goto err_free; - vmw_kms_create_implicit_placement_property(dev_priv, true); + vmw_kms_create_implicit_placement_property(dev_priv); if (dev_priv->capabilities & SVGA_CAP_MULTIMON) for (i = 0; i < VMWGFX_NUM_DISPLAY_UNITS; ++i) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c index 8a029bade32a..3025bfc001a1 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c @@ -85,7 +85,7 @@ static void vmw_resource_release(struct kref *kref) struct ttm_validate_buffer val_buf; val_buf.bo = bo; - val_buf.shared = false; + val_buf.num_shared = 0; res->func->unbind(res, false, &val_buf); } res->backup_dirty = false; @@ -462,7 +462,7 @@ vmw_resource_check_buffer(struct ww_acquire_ctx *ticket, INIT_LIST_HEAD(&val_list); val_buf->bo = ttm_bo_reference(&res->backup->base); - val_buf->shared = false; + val_buf->num_shared = 0; list_add_tail(&val_buf->head, &val_list); ret = ttm_eu_reserve_buffers(ticket, &val_list, interruptible, NULL); if (unlikely(ret != 0)) @@ -565,7 +565,7 @@ static int vmw_resource_do_evict(struct ww_acquire_ctx *ticket, BUG_ON(!func->may_evict); val_buf.bo = NULL; - val_buf.shared = false; + val_buf.num_shared = 0; ret = vmw_resource_check_buffer(ticket, res, interruptible, &val_buf); if (unlikely(ret != 0)) return ret; @@ -614,7 +614,7 @@ int vmw_resource_validate(struct vmw_resource *res, bool intr) return 0; val_buf.bo = NULL; - val_buf.shared = false; + val_buf.num_shared = 0; if (res->backup) val_buf.bo = &res->backup->base; do { @@ -685,7 +685,7 @@ void vmw_resource_unbind_list(struct vmw_buffer_object *vbo) struct vmw_resource *res, *next; struct ttm_validate_buffer val_buf = { .bo = &vbo->base, - .shared = false + .num_shared = 0 }; lockdep_assert_held(&vbo->base.resv->lock.base); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c index 333418dc259f..cd586c52af7e 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c @@ -29,6 +29,7 @@ #include <drm/drm_plane_helper.h> #include <drm/drm_atomic.h> #include <drm/drm_atomic_helper.h> +#include <drm/drm_damage_helper.h> #define vmw_crtc_to_sou(x) \ @@ -76,6 +77,11 @@ struct vmw_kms_sou_dirty_cmd { SVGA3dCmdBlitSurfaceToScreen body; }; +struct vmw_kms_sou_define_gmrfb { + uint32_t header; + SVGAFifoCmdDefineGMRFB body; +}; + /** * Display unit using screen objects. */ @@ -241,28 +247,20 @@ static void vmw_sou_crtc_mode_set_nofb(struct drm_crtc *crtc) sou->buffer = vps->bo; sou->buffer_size = vps->bo_size; - if (sou->base.is_implicit) { - x = crtc->x; - y = crtc->y; - } else { - conn_state = sou->base.connector.state; - vmw_conn_state = vmw_connector_state_to_vcs(conn_state); + conn_state = sou->base.connector.state; + vmw_conn_state = vmw_connector_state_to_vcs(conn_state); - x = vmw_conn_state->gui_x; - y = vmw_conn_state->gui_y; - } + x = vmw_conn_state->gui_x; + y = vmw_conn_state->gui_y; ret = vmw_sou_fifo_create(dev_priv, sou, x, y, &crtc->mode); if (ret) DRM_ERROR("Failed to define Screen Object %dx%d\n", crtc->x, crtc->y); - vmw_kms_add_active(dev_priv, &sou->base, vfb); } else { sou->buffer = NULL; sou->buffer_size = 0; - - vmw_kms_del_active(dev_priv, &sou->base); } } @@ -317,38 +315,14 @@ static void vmw_sou_crtc_atomic_disable(struct drm_crtc *crtc, } } -static int vmw_sou_crtc_page_flip(struct drm_crtc *crtc, - struct drm_framebuffer *new_fb, - struct drm_pending_vblank_event *event, - uint32_t flags, - struct drm_modeset_acquire_ctx *ctx) -{ - struct vmw_private *dev_priv = vmw_priv(crtc->dev); - int ret; - - if (!vmw_kms_crtc_flippable(dev_priv, crtc)) - return -EINVAL; - - ret = drm_atomic_helper_page_flip(crtc, new_fb, event, flags, ctx); - if (ret) { - DRM_ERROR("Page flip error %d.\n", ret); - return ret; - } - - if (vmw_crtc_to_du(crtc)->is_implicit) - vmw_kms_update_implicit_fb(dev_priv, crtc); - - return ret; -} - static const struct drm_crtc_funcs vmw_screen_object_crtc_funcs = { .gamma_set = vmw_du_crtc_gamma_set, .destroy = vmw_sou_crtc_destroy, .reset = vmw_du_crtc_reset, .atomic_duplicate_state = vmw_du_crtc_duplicate_state, .atomic_destroy_state = vmw_du_crtc_destroy_state, - .set_config = vmw_kms_set_config, - .page_flip = vmw_sou_crtc_page_flip, + .set_config = drm_atomic_helper_set_config, + .page_flip = drm_atomic_helper_page_flip, }; /* @@ -377,13 +351,10 @@ static const struct drm_connector_funcs vmw_sou_connector_funcs = { .dpms = vmw_du_connector_dpms, .detect = vmw_du_connector_detect, .fill_modes = vmw_du_connector_fill_modes, - .set_property = vmw_du_connector_set_property, .destroy = vmw_sou_connector_destroy, .reset = vmw_du_connector_reset, .atomic_duplicate_state = vmw_du_connector_duplicate_state, .atomic_destroy_state = vmw_du_connector_destroy_state, - .atomic_set_property = vmw_du_connector_atomic_set_property, - .atomic_get_property = vmw_du_connector_atomic_get_property, }; @@ -498,6 +469,263 @@ vmw_sou_primary_plane_prepare_fb(struct drm_plane *plane, return vmw_bo_pin_in_vram(dev_priv, vps->bo, true); } +static uint32_t vmw_sou_bo_fifo_size(struct vmw_du_update_plane *update, + uint32_t num_hits) +{ + return sizeof(struct vmw_kms_sou_define_gmrfb) + + sizeof(struct vmw_kms_sou_bo_blit) * num_hits; +} + +static uint32_t vmw_sou_bo_define_gmrfb(struct vmw_du_update_plane *update, + void *cmd) +{ + struct vmw_framebuffer_bo *vfbbo = + container_of(update->vfb, typeof(*vfbbo), base); + struct vmw_kms_sou_define_gmrfb *gmr = cmd; + int depth = update->vfb->base.format->depth; + + /* Emulate RGBA support, contrary to svga_reg.h this is not + * supported by hosts. This is only a problem if we are reading + * this value later and expecting what we uploaded back. + */ + if (depth == 32) + depth = 24; + + gmr->header = SVGA_CMD_DEFINE_GMRFB; + + gmr->body.format.bitsPerPixel = update->vfb->base.format->cpp[0] * 8; + gmr->body.format.colorDepth = depth; + gmr->body.format.reserved = 0; + gmr->body.bytesPerLine = update->vfb->base.pitches[0]; + vmw_bo_get_guest_ptr(&vfbbo->buffer->base, &gmr->body.ptr); + + return sizeof(*gmr); +} + +static uint32_t vmw_sou_bo_populate_clip(struct vmw_du_update_plane *update, + void *cmd, struct drm_rect *clip, + uint32_t fb_x, uint32_t fb_y) +{ + struct vmw_kms_sou_bo_blit *blit = cmd; + + blit->header = SVGA_CMD_BLIT_GMRFB_TO_SCREEN; + blit->body.destScreenId = update->du->unit; + blit->body.srcOrigin.x = fb_x; + blit->body.srcOrigin.y = fb_y; + blit->body.destRect.left = clip->x1; + blit->body.destRect.top = clip->y1; + blit->body.destRect.right = clip->x2; + blit->body.destRect.bottom = clip->y2; + + return sizeof(*blit); +} + +static uint32_t vmw_stud_bo_post_clip(struct vmw_du_update_plane *update, + void *cmd, struct drm_rect *bb) +{ + return 0; +} + +/** + * vmw_sou_plane_update_bo - Update display unit for bo backed fb. + * @dev_priv: Device private. + * @plane: Plane state. + * @old_state: Old plane state. + * @vfb: Framebuffer which is blitted to display unit. + * @out_fence: If non-NULL, will return a ref-counted pointer to vmw_fence_obj. + * The returned fence pointer may be NULL in which case the device + * has already synchronized. + * + * Return: 0 on success or a negative error code on failure. + */ +static int vmw_sou_plane_update_bo(struct vmw_private *dev_priv, + struct drm_plane *plane, + struct drm_plane_state *old_state, + struct vmw_framebuffer *vfb, + struct vmw_fence_obj **out_fence) +{ + struct vmw_du_update_plane_buffer bo_update; + + memset(&bo_update, 0, sizeof(struct vmw_du_update_plane_buffer)); + bo_update.base.plane = plane; + bo_update.base.old_state = old_state; + bo_update.base.dev_priv = dev_priv; + bo_update.base.du = vmw_crtc_to_du(plane->state->crtc); + bo_update.base.vfb = vfb; + bo_update.base.out_fence = out_fence; + bo_update.base.mutex = NULL; + bo_update.base.cpu_blit = false; + bo_update.base.intr = true; + + bo_update.base.calc_fifo_size = vmw_sou_bo_fifo_size; + bo_update.base.post_prepare = vmw_sou_bo_define_gmrfb; + bo_update.base.clip = vmw_sou_bo_populate_clip; + bo_update.base.post_clip = vmw_stud_bo_post_clip; + + return vmw_du_helper_plane_update(&bo_update.base); +} + +static uint32_t vmw_sou_surface_fifo_size(struct vmw_du_update_plane *update, + uint32_t num_hits) +{ + return sizeof(struct vmw_kms_sou_dirty_cmd) + sizeof(SVGASignedRect) * + num_hits; +} + +static uint32_t vmw_sou_surface_post_prepare(struct vmw_du_update_plane *update, + void *cmd) +{ + struct vmw_du_update_plane_surface *srf_update; + + srf_update = container_of(update, typeof(*srf_update), base); + + /* + * SOU SVGA_3D_CMD_BLIT_SURFACE_TO_SCREEN is special in the sense that + * its bounding box is filled before iterating over all the clips. So + * store the FIFO start address and revisit to fill the details. + */ + srf_update->cmd_start = cmd; + + return 0; +} + +static uint32_t vmw_sou_surface_pre_clip(struct vmw_du_update_plane *update, + void *cmd, uint32_t num_hits) +{ + struct vmw_kms_sou_dirty_cmd *blit = cmd; + struct vmw_framebuffer_surface *vfbs; + + vfbs = container_of(update->vfb, typeof(*vfbs), base); + + blit->header.id = SVGA_3D_CMD_BLIT_SURFACE_TO_SCREEN; + blit->header.size = sizeof(blit->body) + sizeof(SVGASignedRect) * + num_hits; + + blit->body.srcImage.sid = vfbs->surface->res.id; + blit->body.destScreenId = update->du->unit; + + /* Update the source and destination bounding box later in post_clip */ + blit->body.srcRect.left = 0; + blit->body.srcRect.top = 0; + blit->body.srcRect.right = 0; + blit->body.srcRect.bottom = 0; + + blit->body.destRect.left = 0; + blit->body.destRect.top = 0; + blit->body.destRect.right = 0; + blit->body.destRect.bottom = 0; + + return sizeof(*blit); +} + +static uint32_t vmw_sou_surface_clip_rect(struct vmw_du_update_plane *update, + void *cmd, struct drm_rect *clip, + uint32_t src_x, uint32_t src_y) +{ + SVGASignedRect *rect = cmd; + + /* + * rects are relative to dest bounding box rect on screen object, so + * translate to it later in post_clip + */ + rect->left = clip->x1; + rect->top = clip->y1; + rect->right = clip->x2; + rect->bottom = clip->y2; + + return sizeof(*rect); +} + +static uint32_t vmw_sou_surface_post_clip(struct vmw_du_update_plane *update, + void *cmd, struct drm_rect *bb) +{ + struct vmw_du_update_plane_surface *srf_update; + struct drm_plane_state *state = update->plane->state; + struct drm_rect src_bb; + struct vmw_kms_sou_dirty_cmd *blit; + SVGASignedRect *rect; + uint32_t num_hits; + int translate_src_x; + int translate_src_y; + int i; + + srf_update = container_of(update, typeof(*srf_update), base); + + blit = srf_update->cmd_start; + rect = (SVGASignedRect *)&blit[1]; + + num_hits = (blit->header.size - sizeof(blit->body))/ + sizeof(SVGASignedRect); + + src_bb = *bb; + + /* To translate bb back to fb src coord */ + translate_src_x = (state->src_x >> 16) - state->crtc_x; + translate_src_y = (state->src_y >> 16) - state->crtc_y; + + drm_rect_translate(&src_bb, translate_src_x, translate_src_y); + + blit->body.srcRect.left = src_bb.x1; + blit->body.srcRect.top = src_bb.y1; + blit->body.srcRect.right = src_bb.x2; + blit->body.srcRect.bottom = src_bb.y2; + + blit->body.destRect.left = bb->x1; + blit->body.destRect.top = bb->y1; + blit->body.destRect.right = bb->x2; + blit->body.destRect.bottom = bb->y2; + + /* rects are relative to dest bb rect */ + for (i = 0; i < num_hits; i++) { + rect->left -= bb->x1; + rect->top -= bb->y1; + rect->right -= bb->x1; + rect->bottom -= bb->y1; + rect++; + } + + return 0; +} + +/** + * vmw_sou_plane_update_surface - Update display unit for surface backed fb. + * @dev_priv: Device private. + * @plane: Plane state. + * @old_state: Old plane state. + * @vfb: Framebuffer which is blitted to display unit + * @out_fence: If non-NULL, will return a ref-counted pointer to vmw_fence_obj. + * The returned fence pointer may be NULL in which case the device + * has already synchronized. + * + * Return: 0 on success or a negative error code on failure. + */ +static int vmw_sou_plane_update_surface(struct vmw_private *dev_priv, + struct drm_plane *plane, + struct drm_plane_state *old_state, + struct vmw_framebuffer *vfb, + struct vmw_fence_obj **out_fence) +{ + struct vmw_du_update_plane_surface srf_update; + + memset(&srf_update, 0, sizeof(struct vmw_du_update_plane_surface)); + srf_update.base.plane = plane; + srf_update.base.old_state = old_state; + srf_update.base.dev_priv = dev_priv; + srf_update.base.du = vmw_crtc_to_du(plane->state->crtc); + srf_update.base.vfb = vfb; + srf_update.base.out_fence = out_fence; + srf_update.base.mutex = &dev_priv->cmdbuf_mutex; + srf_update.base.cpu_blit = false; + srf_update.base.intr = true; + + srf_update.base.calc_fifo_size = vmw_sou_surface_fifo_size; + srf_update.base.post_prepare = vmw_sou_surface_post_prepare; + srf_update.base.pre_clip = vmw_sou_surface_pre_clip; + srf_update.base.clip = vmw_sou_surface_clip_rect; + srf_update.base.post_clip = vmw_sou_surface_post_clip; + + return vmw_du_helper_plane_update(&srf_update.base); +} static void vmw_sou_primary_plane_atomic_update(struct drm_plane *plane, @@ -508,47 +736,28 @@ vmw_sou_primary_plane_atomic_update(struct drm_plane *plane, struct vmw_fence_obj *fence = NULL; int ret; + /* In case of device error, maintain consistent atomic state */ if (crtc && plane->state->fb) { struct vmw_private *dev_priv = vmw_priv(crtc->dev); struct vmw_framebuffer *vfb = vmw_framebuffer_to_vfb(plane->state->fb); - struct drm_vmw_rect vclips; - - vclips.x = crtc->x; - vclips.y = crtc->y; - vclips.w = crtc->mode.hdisplay; - vclips.h = crtc->mode.vdisplay; if (vfb->bo) - ret = vmw_kms_sou_do_bo_dirty(dev_priv, vfb, NULL, - &vclips, 1, 1, true, - &fence, crtc); + ret = vmw_sou_plane_update_bo(dev_priv, plane, + old_state, vfb, &fence); else - ret = vmw_kms_sou_do_surface_dirty(dev_priv, vfb, NULL, - &vclips, NULL, 0, 0, - 1, 1, &fence, crtc); - - /* - * We cannot really fail this function, so if we do, then output - * an error and maintain consistent atomic state. - */ + ret = vmw_sou_plane_update_surface(dev_priv, plane, + old_state, vfb, + &fence); if (ret != 0) DRM_ERROR("Failed to update screen.\n"); } else { - /* - * When disabling a plane, CRTC and FB should always be NULL - * together, otherwise it's an error. - * Here primary plane is being disable so should really blank - * the screen object display unit, if not already done. - */ + /* Do nothing when fb and crtc is NULL (blank crtc) */ return; } + /* For error case vblank event is send from vmw_du_crtc_atomic_flush */ event = crtc->state->event; - /* - * In case of failure and other cases, vblank event will be sent in - * vmw_du_crtc_atomic_flush. - */ if (event && fence) { struct drm_file *file_priv = event->base.file_priv; @@ -639,7 +848,6 @@ static int vmw_sou_init(struct vmw_private *dev_priv, unsigned unit) primary = &sou->base.primary; cursor = &sou->base.cursor; - sou->base.active_implicit = false; sou->base.pref_active = (unit == 0); sou->base.pref_width = dev_priv->initial_width; sou->base.pref_height = dev_priv->initial_height; @@ -665,6 +873,7 @@ static int vmw_sou_init(struct vmw_private *dev_priv, unsigned unit) } drm_plane_helper_add(primary, &vmw_sou_primary_plane_helper_funcs); + drm_plane_enable_fb_damage_clips(primary); /* Initialize cursor plane */ vmw_du_plane_reset(cursor); @@ -692,8 +901,6 @@ static int vmw_sou_init(struct vmw_private *dev_priv, unsigned unit) drm_connector_helper_add(connector, &vmw_sou_connector_helper_funcs); connector->status = vmw_du_connector_detect(connector, true); - vmw_connector_state_to_vcs(connector->state)->is_implicit = false; - ret = drm_encoder_init(dev, encoder, &vmw_screen_object_encoder_funcs, DRM_MODE_ENCODER_VIRTUAL, NULL); @@ -732,12 +939,6 @@ static int vmw_sou_init(struct vmw_private *dev_priv, unsigned unit) dev->mode_config.suggested_x_property, 0); drm_object_attach_property(&connector->base, dev->mode_config.suggested_y_property, 0); - if (dev_priv->implicit_placement_property) - drm_object_attach_property - (&connector->base, - dev_priv->implicit_placement_property, - sou->base.is_implicit); - return 0; err_free_unregister: @@ -763,15 +964,11 @@ int vmw_kms_sou_init_display(struct vmw_private *dev_priv) } ret = -ENOMEM; - dev_priv->num_implicit = 0; - dev_priv->implicit_fb = NULL; ret = drm_vblank_init(dev, VMWGFX_NUM_DISPLAY_UNITS); if (unlikely(ret != 0)) return ret; - vmw_kms_create_implicit_placement_property(dev_priv, false); - for (i = 0; i < VMWGFX_NUM_DISPLAY_UNITS; ++i) vmw_sou_init(dev_priv, i); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c index c3e435f444c1..096c2941a8e4 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c @@ -30,7 +30,7 @@ #include <drm/drm_plane_helper.h> #include <drm/drm_atomic.h> #include <drm/drm_atomic_helper.h> - +#include <drm/drm_damage_helper.h> #define vmw_crtc_to_stdu(x) \ container_of(x, struct vmw_screen_target_display_unit, base.crtc) @@ -92,6 +92,10 @@ struct vmw_stdu_surface_copy { SVGA3dCmdSurfaceCopy body; }; +struct vmw_stdu_update_gb_image { + SVGA3dCmdHeader header; + SVGA3dCmdUpdateGBImage body; +}; /** * struct vmw_screen_target_display_unit @@ -396,13 +400,8 @@ static void vmw_stdu_crtc_mode_set_nofb(struct drm_crtc *crtc) if (!crtc->state->enable) return; - if (stdu->base.is_implicit) { - x = crtc->x; - y = crtc->y; - } else { - x = vmw_conn_state->gui_x; - y = vmw_conn_state->gui_y; - } + x = vmw_conn_state->gui_x; + y = vmw_conn_state->gui_y; vmw_svga_enable(dev_priv); ret = vmw_stdu_define_st(dev_priv, stdu, &crtc->mode, x, y); @@ -417,27 +416,9 @@ static void vmw_stdu_crtc_helper_prepare(struct drm_crtc *crtc) { } - static void vmw_stdu_crtc_atomic_enable(struct drm_crtc *crtc, struct drm_crtc_state *old_state) { - struct drm_plane_state *plane_state = crtc->primary->state; - struct vmw_private *dev_priv; - struct vmw_screen_target_display_unit *stdu; - struct vmw_framebuffer *vfb; - struct drm_framebuffer *fb; - - - stdu = vmw_crtc_to_stdu(crtc); - dev_priv = vmw_priv(crtc->dev); - fb = plane_state->fb; - - vfb = (fb) ? vmw_framebuffer_to_vfb(fb) : NULL; - - if (vfb) - vmw_kms_add_active(dev_priv, &stdu->base, vfb); - else - vmw_kms_del_active(dev_priv, &stdu->base); } static void vmw_stdu_crtc_atomic_disable(struct drm_crtc *crtc, @@ -472,49 +453,6 @@ static void vmw_stdu_crtc_atomic_disable(struct drm_crtc *crtc, } /** - * vmw_stdu_crtc_page_flip - Binds a buffer to a screen target - * - * @crtc: CRTC to attach FB to - * @fb: FB to attach - * @event: Event to be posted. This event should've been alloced - * using k[mz]alloc, and should've been completely initialized. - * @page_flip_flags: Input flags. - * - * If the STDU uses the same display and content buffers, i.e. a true flip, - * this function will replace the existing display buffer with the new content - * buffer. - * - * If the STDU uses different display and content buffers, i.e. a blit, then - * only the content buffer will be updated. - * - * RETURNS: - * 0 on success, error code on failure - */ -static int vmw_stdu_crtc_page_flip(struct drm_crtc *crtc, - struct drm_framebuffer *new_fb, - struct drm_pending_vblank_event *event, - uint32_t flags, - struct drm_modeset_acquire_ctx *ctx) - -{ - struct vmw_private *dev_priv = vmw_priv(crtc->dev); - struct vmw_screen_target_display_unit *stdu = vmw_crtc_to_stdu(crtc); - int ret; - - if (!stdu->defined || !vmw_kms_crtc_flippable(dev_priv, crtc)) - return -EINVAL; - - ret = drm_atomic_helper_page_flip(crtc, new_fb, event, flags, ctx); - if (ret) { - DRM_ERROR("Page flip error %d.\n", ret); - return ret; - } - - return 0; -} - - -/** * vmw_stdu_bo_clip - Callback to encode a suface DMA command cliprect * * @dirty: The closure structure. @@ -986,8 +924,8 @@ static const struct drm_crtc_funcs vmw_stdu_crtc_funcs = { .reset = vmw_du_crtc_reset, .atomic_duplicate_state = vmw_du_crtc_duplicate_state, .atomic_destroy_state = vmw_du_crtc_destroy_state, - .set_config = vmw_kms_set_config, - .page_flip = vmw_stdu_crtc_page_flip, + .set_config = drm_atomic_helper_set_config, + .page_flip = drm_atomic_helper_page_flip, }; @@ -1042,13 +980,10 @@ static const struct drm_connector_funcs vmw_stdu_connector_funcs = { .dpms = vmw_du_connector_dpms, .detect = vmw_du_connector_detect, .fill_modes = vmw_du_connector_fill_modes, - .set_property = vmw_du_connector_set_property, .destroy = vmw_stdu_connector_destroy, .reset = vmw_du_connector_reset, .atomic_duplicate_state = vmw_du_connector_duplicate_state, .atomic_destroy_state = vmw_du_connector_destroy_state, - .atomic_set_property = vmw_du_connector_atomic_set_property, - .atomic_get_property = vmw_du_connector_atomic_get_property, }; @@ -1256,11 +1191,402 @@ out_srf_unref: return ret; } +static uint32_t vmw_stdu_bo_fifo_size(struct vmw_du_update_plane *update, + uint32_t num_hits) +{ + return sizeof(struct vmw_stdu_dma) + sizeof(SVGA3dCopyBox) * num_hits + + sizeof(SVGA3dCmdSurfaceDMASuffix) + + sizeof(struct vmw_stdu_update); +} + +static uint32_t vmw_stdu_bo_fifo_size_cpu(struct vmw_du_update_plane *update, + uint32_t num_hits) +{ + return sizeof(struct vmw_stdu_update_gb_image) + + sizeof(struct vmw_stdu_update); +} + +static uint32_t vmw_stdu_bo_populate_dma(struct vmw_du_update_plane *update, + void *cmd, uint32_t num_hits) +{ + struct vmw_screen_target_display_unit *stdu; + struct vmw_framebuffer_bo *vfbbo; + struct vmw_stdu_dma *cmd_dma = cmd; + + stdu = container_of(update->du, typeof(*stdu), base); + vfbbo = container_of(update->vfb, typeof(*vfbbo), base); + + cmd_dma->header.id = SVGA_3D_CMD_SURFACE_DMA; + cmd_dma->header.size = sizeof(cmd_dma->body) + + sizeof(struct SVGA3dCopyBox) * num_hits + + sizeof(SVGA3dCmdSurfaceDMASuffix); + vmw_bo_get_guest_ptr(&vfbbo->buffer->base, &cmd_dma->body.guest.ptr); + cmd_dma->body.guest.pitch = update->vfb->base.pitches[0]; + cmd_dma->body.host.sid = stdu->display_srf->res.id; + cmd_dma->body.host.face = 0; + cmd_dma->body.host.mipmap = 0; + cmd_dma->body.transfer = SVGA3D_WRITE_HOST_VRAM; + + return sizeof(*cmd_dma); +} + +static uint32_t vmw_stdu_bo_populate_clip(struct vmw_du_update_plane *update, + void *cmd, struct drm_rect *clip, + uint32_t fb_x, uint32_t fb_y) +{ + struct SVGA3dCopyBox *box = cmd; + + box->srcx = fb_x; + box->srcy = fb_y; + box->srcz = 0; + box->x = clip->x1; + box->y = clip->y1; + box->z = 0; + box->w = drm_rect_width(clip); + box->h = drm_rect_height(clip); + box->d = 1; + + return sizeof(*box); +} + +static uint32_t vmw_stdu_bo_populate_update(struct vmw_du_update_plane *update, + void *cmd, struct drm_rect *bb) +{ + struct vmw_screen_target_display_unit *stdu; + struct vmw_framebuffer_bo *vfbbo; + SVGA3dCmdSurfaceDMASuffix *suffix = cmd; + + stdu = container_of(update->du, typeof(*stdu), base); + vfbbo = container_of(update->vfb, typeof(*vfbbo), base); + + suffix->suffixSize = sizeof(*suffix); + suffix->maximumOffset = vfbbo->buffer->base.num_pages * PAGE_SIZE; + + vmw_stdu_populate_update(&suffix[1], stdu->base.unit, bb->x1, bb->x2, + bb->y1, bb->y2); + + return sizeof(*suffix) + sizeof(struct vmw_stdu_update); +} + +static uint32_t vmw_stdu_bo_pre_clip_cpu(struct vmw_du_update_plane *update, + void *cmd, uint32_t num_hits) +{ + struct vmw_du_update_plane_buffer *bo_update = + container_of(update, typeof(*bo_update), base); + + bo_update->fb_left = INT_MAX; + bo_update->fb_top = INT_MAX; + + return 0; +} + +static uint32_t vmw_stdu_bo_clip_cpu(struct vmw_du_update_plane *update, + void *cmd, struct drm_rect *clip, + uint32_t fb_x, uint32_t fb_y) +{ + struct vmw_du_update_plane_buffer *bo_update = + container_of(update, typeof(*bo_update), base); + + bo_update->fb_left = min_t(int, bo_update->fb_left, fb_x); + bo_update->fb_top = min_t(int, bo_update->fb_top, fb_y); + + return 0; +} + +static uint32_t +vmw_stdu_bo_populate_update_cpu(struct vmw_du_update_plane *update, void *cmd, + struct drm_rect *bb) +{ + struct vmw_du_update_plane_buffer *bo_update; + struct vmw_screen_target_display_unit *stdu; + struct vmw_framebuffer_bo *vfbbo; + struct vmw_diff_cpy diff = VMW_CPU_BLIT_DIFF_INITIALIZER(0); + struct vmw_stdu_update_gb_image *cmd_img = cmd; + struct vmw_stdu_update *cmd_update; + struct ttm_buffer_object *src_bo, *dst_bo; + u32 src_offset, dst_offset; + s32 src_pitch, dst_pitch; + s32 width, height; + + bo_update = container_of(update, typeof(*bo_update), base); + stdu = container_of(update->du, typeof(*stdu), base); + vfbbo = container_of(update->vfb, typeof(*vfbbo), base); + + width = bb->x2 - bb->x1; + height = bb->y2 - bb->y1; + + diff.cpp = stdu->cpp; + + dst_bo = &stdu->display_srf->res.backup->base; + dst_pitch = stdu->display_srf->base_size.width * stdu->cpp; + dst_offset = bb->y1 * dst_pitch + bb->x1 * stdu->cpp; + + src_bo = &vfbbo->buffer->base; + src_pitch = update->vfb->base.pitches[0]; + src_offset = bo_update->fb_top * src_pitch + bo_update->fb_left * + stdu->cpp; + + (void) vmw_bo_cpu_blit(dst_bo, dst_offset, dst_pitch, src_bo, + src_offset, src_pitch, width * stdu->cpp, height, + &diff); + + if (drm_rect_visible(&diff.rect)) { + SVGA3dBox *box = &cmd_img->body.box; + + cmd_img->header.id = SVGA_3D_CMD_UPDATE_GB_IMAGE; + cmd_img->header.size = sizeof(cmd_img->body); + cmd_img->body.image.sid = stdu->display_srf->res.id; + cmd_img->body.image.face = 0; + cmd_img->body.image.mipmap = 0; + + box->x = diff.rect.x1; + box->y = diff.rect.y1; + box->z = 0; + box->w = drm_rect_width(&diff.rect); + box->h = drm_rect_height(&diff.rect); + box->d = 1; + + cmd_update = (struct vmw_stdu_update *)&cmd_img[1]; + vmw_stdu_populate_update(cmd_update, stdu->base.unit, + diff.rect.x1, diff.rect.x2, + diff.rect.y1, diff.rect.y2); + + return sizeof(*cmd_img) + sizeof(*cmd_update); + } + + return 0; +} + +/** + * vmw_stdu_plane_update_bo - Update display unit for bo backed fb. + * @dev_priv: device private. + * @plane: plane state. + * @old_state: old plane state. + * @vfb: framebuffer which is blitted to display unit. + * @out_fence: If non-NULL, will return a ref-counted pointer to vmw_fence_obj. + * The returned fence pointer may be NULL in which case the device + * has already synchronized. + * + * Return: 0 on success or a negative error code on failure. + */ +static int vmw_stdu_plane_update_bo(struct vmw_private *dev_priv, + struct drm_plane *plane, + struct drm_plane_state *old_state, + struct vmw_framebuffer *vfb, + struct vmw_fence_obj **out_fence) +{ + struct vmw_du_update_plane_buffer bo_update; + + memset(&bo_update, 0, sizeof(struct vmw_du_update_plane_buffer)); + bo_update.base.plane = plane; + bo_update.base.old_state = old_state; + bo_update.base.dev_priv = dev_priv; + bo_update.base.du = vmw_crtc_to_du(plane->state->crtc); + bo_update.base.vfb = vfb; + bo_update.base.out_fence = out_fence; + bo_update.base.mutex = NULL; + bo_update.base.cpu_blit = !(dev_priv->capabilities & SVGA_CAP_3D); + bo_update.base.intr = false; + + /* + * VM without 3D support don't have surface DMA command and framebuffer + * should be moved out of VRAM. + */ + if (bo_update.base.cpu_blit) { + bo_update.base.calc_fifo_size = vmw_stdu_bo_fifo_size_cpu; + bo_update.base.pre_clip = vmw_stdu_bo_pre_clip_cpu; + bo_update.base.clip = vmw_stdu_bo_clip_cpu; + bo_update.base.post_clip = vmw_stdu_bo_populate_update_cpu; + } else { + bo_update.base.calc_fifo_size = vmw_stdu_bo_fifo_size; + bo_update.base.pre_clip = vmw_stdu_bo_populate_dma; + bo_update.base.clip = vmw_stdu_bo_populate_clip; + bo_update.base.post_clip = vmw_stdu_bo_populate_update; + } + + return vmw_du_helper_plane_update(&bo_update.base); +} + +static uint32_t +vmw_stdu_surface_fifo_size_same_display(struct vmw_du_update_plane *update, + uint32_t num_hits) +{ + struct vmw_framebuffer_surface *vfbs; + uint32_t size = 0; + + vfbs = container_of(update->vfb, typeof(*vfbs), base); + + if (vfbs->is_bo_proxy) + size += sizeof(struct vmw_stdu_update_gb_image) * num_hits; + + size += sizeof(struct vmw_stdu_update); + + return size; +} + +static uint32_t vmw_stdu_surface_fifo_size(struct vmw_du_update_plane *update, + uint32_t num_hits) +{ + struct vmw_framebuffer_surface *vfbs; + uint32_t size = 0; + + vfbs = container_of(update->vfb, typeof(*vfbs), base); + + if (vfbs->is_bo_proxy) + size += sizeof(struct vmw_stdu_update_gb_image) * num_hits; + + size += sizeof(struct vmw_stdu_surface_copy) + sizeof(SVGA3dCopyBox) * + num_hits + sizeof(struct vmw_stdu_update); + + return size; +} + +static uint32_t +vmw_stdu_surface_update_proxy(struct vmw_du_update_plane *update, void *cmd) +{ + struct vmw_framebuffer_surface *vfbs; + struct drm_plane_state *state = update->plane->state; + struct drm_plane_state *old_state = update->old_state; + struct vmw_stdu_update_gb_image *cmd_update = cmd; + struct drm_atomic_helper_damage_iter iter; + struct drm_rect clip; + uint32_t copy_size = 0; + + vfbs = container_of(update->vfb, typeof(*vfbs), base); + + /* + * proxy surface is special where a buffer object type fb is wrapped + * in a surface and need an update gb image command to sync with device. + */ + drm_atomic_helper_damage_iter_init(&iter, old_state, state); + drm_atomic_for_each_plane_damage(&iter, &clip) { + SVGA3dBox *box = &cmd_update->body.box; + + cmd_update->header.id = SVGA_3D_CMD_UPDATE_GB_IMAGE; + cmd_update->header.size = sizeof(cmd_update->body); + cmd_update->body.image.sid = vfbs->surface->res.id; + cmd_update->body.image.face = 0; + cmd_update->body.image.mipmap = 0; + + box->x = clip.x1; + box->y = clip.y1; + box->z = 0; + box->w = drm_rect_width(&clip); + box->h = drm_rect_height(&clip); + box->d = 1; + + copy_size += sizeof(*cmd_update); + cmd_update++; + } + + return copy_size; +} + +static uint32_t +vmw_stdu_surface_populate_copy(struct vmw_du_update_plane *update, void *cmd, + uint32_t num_hits) +{ + struct vmw_screen_target_display_unit *stdu; + struct vmw_framebuffer_surface *vfbs; + struct vmw_stdu_surface_copy *cmd_copy = cmd; + + stdu = container_of(update->du, typeof(*stdu), base); + vfbs = container_of(update->vfb, typeof(*vfbs), base); + + cmd_copy->header.id = SVGA_3D_CMD_SURFACE_COPY; + cmd_copy->header.size = sizeof(cmd_copy->body) + sizeof(SVGA3dCopyBox) * + num_hits; + cmd_copy->body.src.sid = vfbs->surface->res.id; + cmd_copy->body.dest.sid = stdu->display_srf->res.id; + + return sizeof(*cmd_copy); +} + +static uint32_t +vmw_stdu_surface_populate_clip(struct vmw_du_update_plane *update, void *cmd, + struct drm_rect *clip, uint32_t fb_x, + uint32_t fb_y) +{ + struct SVGA3dCopyBox *box = cmd; + + box->srcx = fb_x; + box->srcy = fb_y; + box->srcz = 0; + box->x = clip->x1; + box->y = clip->y1; + box->z = 0; + box->w = drm_rect_width(clip); + box->h = drm_rect_height(clip); + box->d = 1; + + return sizeof(*box); +} + +static uint32_t +vmw_stdu_surface_populate_update(struct vmw_du_update_plane *update, void *cmd, + struct drm_rect *bb) +{ + vmw_stdu_populate_update(cmd, update->du->unit, bb->x1, bb->x2, bb->y1, + bb->y2); + return sizeof(struct vmw_stdu_update); +} /** - * vmw_stdu_primary_plane_atomic_update - formally switches STDU to new plane + * vmw_stdu_plane_update_surface - Update display unit for surface backed fb + * @dev_priv: Device private + * @plane: Plane state + * @old_state: Old plane state + * @vfb: Framebuffer which is blitted to display unit + * @out_fence: If non-NULL, will return a ref-counted pointer to vmw_fence_obj. + * The returned fence pointer may be NULL in which case the device + * has already synchronized. * + * Return: 0 on success or a negative error code on failure. + */ +static int vmw_stdu_plane_update_surface(struct vmw_private *dev_priv, + struct drm_plane *plane, + struct drm_plane_state *old_state, + struct vmw_framebuffer *vfb, + struct vmw_fence_obj **out_fence) +{ + struct vmw_du_update_plane srf_update; + struct vmw_screen_target_display_unit *stdu; + struct vmw_framebuffer_surface *vfbs; + + stdu = vmw_crtc_to_stdu(plane->state->crtc); + vfbs = container_of(vfb, typeof(*vfbs), base); + + memset(&srf_update, 0, sizeof(struct vmw_du_update_plane)); + srf_update.plane = plane; + srf_update.old_state = old_state; + srf_update.dev_priv = dev_priv; + srf_update.du = vmw_crtc_to_du(plane->state->crtc); + srf_update.vfb = vfb; + srf_update.out_fence = out_fence; + srf_update.mutex = &dev_priv->cmdbuf_mutex; + srf_update.cpu_blit = false; + srf_update.intr = true; + + if (vfbs->is_bo_proxy) + srf_update.post_prepare = vmw_stdu_surface_update_proxy; + + if (vfbs->surface->res.id != stdu->display_srf->res.id) { + srf_update.calc_fifo_size = vmw_stdu_surface_fifo_size; + srf_update.pre_clip = vmw_stdu_surface_populate_copy; + srf_update.clip = vmw_stdu_surface_populate_clip; + } else { + srf_update.calc_fifo_size = + vmw_stdu_surface_fifo_size_same_display; + } + + srf_update.post_clip = vmw_stdu_surface_populate_update; + + return vmw_du_helper_plane_update(&srf_update); +} + +/** + * vmw_stdu_primary_plane_atomic_update - formally switches STDU to new plane * @plane: display plane * @old_state: Only used to get crtc info * @@ -1277,17 +1603,14 @@ vmw_stdu_primary_plane_atomic_update(struct drm_plane *plane, struct drm_crtc *crtc = plane->state->crtc; struct vmw_screen_target_display_unit *stdu; struct drm_pending_vblank_event *event; + struct vmw_fence_obj *fence = NULL; struct vmw_private *dev_priv; int ret; - /* - * We cannot really fail this function, so if we do, then output an - * error and maintain consistent atomic state. - */ + /* If case of device error, maintain consistent atomic state */ if (crtc && plane->state->fb) { struct vmw_framebuffer *vfb = vmw_framebuffer_to_vfb(plane->state->fb); - struct drm_vmw_rect vclips; stdu = vmw_crtc_to_stdu(crtc); dev_priv = vmw_priv(crtc->dev); @@ -1295,23 +1618,17 @@ vmw_stdu_primary_plane_atomic_update(struct drm_plane *plane, stdu->content_fb_type = vps->content_fb_type; stdu->cpp = vps->cpp; - vclips.x = crtc->x; - vclips.y = crtc->y; - vclips.w = crtc->mode.hdisplay; - vclips.h = crtc->mode.vdisplay; - ret = vmw_stdu_bind_st(dev_priv, stdu, &stdu->display_srf->res); if (ret) DRM_ERROR("Failed to bind surface to STDU.\n"); if (vfb->bo) - ret = vmw_kms_stdu_dma(dev_priv, NULL, vfb, NULL, NULL, - &vclips, 1, 1, true, false, - crtc); + ret = vmw_stdu_plane_update_bo(dev_priv, plane, + old_state, vfb, &fence); else - ret = vmw_kms_stdu_surface_dirty(dev_priv, vfb, NULL, - &vclips, NULL, 0, 0, - 1, 1, NULL, crtc); + ret = vmw_stdu_plane_update_surface(dev_priv, plane, + old_state, vfb, + &fence); if (ret) DRM_ERROR("Failed to update STDU.\n"); } else { @@ -1319,12 +1636,7 @@ vmw_stdu_primary_plane_atomic_update(struct drm_plane *plane, stdu = vmw_crtc_to_stdu(crtc); dev_priv = vmw_priv(crtc->dev); - /* - * When disabling a plane, CRTC and FB should always be NULL - * together, otherwise it's an error. - * Here primary plane is being disable so blank the screen - * target display unit, if not already done. - */ + /* Blank STDU when fb and crtc are NULL */ if (!stdu->defined) return; @@ -1339,36 +1651,25 @@ vmw_stdu_primary_plane_atomic_update(struct drm_plane *plane, return; } + /* In case of error, vblank event is send in vmw_du_crtc_atomic_flush */ event = crtc->state->event; - /* - * In case of failure and other cases, vblank event will be sent in - * vmw_du_crtc_atomic_flush. - */ - if (event && (ret == 0)) { - struct vmw_fence_obj *fence = NULL; + if (event && fence) { struct drm_file *file_priv = event->base.file_priv; - vmw_execbuf_fence_commands(NULL, dev_priv, &fence, NULL); - - /* - * If fence is NULL, then already sync. - */ - if (fence) { - ret = vmw_event_fence_action_queue( - file_priv, fence, &event->base, - &event->event.vbl.tv_sec, - &event->event.vbl.tv_usec, - true); - if (ret) - DRM_ERROR("Failed to queue event on fence.\n"); - else - crtc->state->event = NULL; - - vmw_fence_obj_unreference(&fence); - } - } else { - (void) vmw_fifo_flush(dev_priv, false); + ret = vmw_event_fence_action_queue(file_priv, + fence, + &event->base, + &event->event.vbl.tv_sec, + &event->event.vbl.tv_usec, + true); + if (ret) + DRM_ERROR("Failed to queue event on fence.\n"); + else + crtc->state->event = NULL; } + + if (fence) + vmw_fence_obj_unreference(&fence); } @@ -1456,11 +1757,6 @@ static int vmw_stdu_init(struct vmw_private *dev_priv, unsigned unit) stdu->base.pref_active = (unit == 0); stdu->base.pref_width = dev_priv->initial_width; stdu->base.pref_height = dev_priv->initial_height; - - /* - * Remove this after enabling atomic because property values can - * only exist in a state object - */ stdu->base.is_implicit = false; /* Initialize primary plane */ @@ -1477,6 +1773,7 @@ static int vmw_stdu_init(struct vmw_private *dev_priv, unsigned unit) } drm_plane_helper_add(primary, &vmw_stdu_primary_plane_helper_funcs); + drm_plane_enable_fb_damage_clips(primary); /* Initialize cursor plane */ vmw_du_plane_reset(cursor); @@ -1505,7 +1802,6 @@ static int vmw_stdu_init(struct vmw_private *dev_priv, unsigned unit) drm_connector_helper_add(connector, &vmw_stdu_connector_helper_funcs); connector->status = vmw_du_connector_detect(connector, false); - vmw_connector_state_to_vcs(connector->state)->is_implicit = false; ret = drm_encoder_init(dev, encoder, &vmw_stdu_encoder_funcs, DRM_MODE_ENCODER_VIRTUAL, NULL); @@ -1543,11 +1839,6 @@ static int vmw_stdu_init(struct vmw_private *dev_priv, unsigned unit) dev->mode_config.suggested_x_property, 0); drm_object_attach_property(&connector->base, dev->mode_config.suggested_y_property, 0); - if (dev_priv->implicit_placement_property) - drm_object_attach_property - (&connector->base, - dev_priv->implicit_placement_property, - stdu->base.is_implicit); return 0; err_free_unregister: @@ -1616,8 +1907,6 @@ int vmw_kms_stdu_init_display(struct vmw_private *dev_priv) dev_priv->active_display_unit = vmw_du_screen_target; - vmw_kms_create_implicit_placement_property(dev_priv, false); - for (i = 0; i < VMWGFX_NUM_DISPLAY_UNITS; ++i) { ret = vmw_stdu_init(dev_priv, i); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c index 154eb09aa91e..e6d75e377dd8 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c @@ -42,3 +42,39 @@ int vmw_mmap(struct file *filp, struct vm_area_struct *vma) dev_priv = vmw_priv(file_priv->minor->dev); return ttm_bo_mmap(filp, vma, &dev_priv->bdev); } + +/* struct vmw_validation_mem callback */ +static int vmw_vmt_reserve(struct vmw_validation_mem *m, size_t size) +{ + static struct ttm_operation_ctx ctx = {.interruptible = false, + .no_wait_gpu = false}; + struct vmw_private *dev_priv = container_of(m, struct vmw_private, vvm); + + return ttm_mem_global_alloc(vmw_mem_glob(dev_priv), size, &ctx); +} + +/* struct vmw_validation_mem callback */ +static void vmw_vmt_unreserve(struct vmw_validation_mem *m, size_t size) +{ + struct vmw_private *dev_priv = container_of(m, struct vmw_private, vvm); + + return ttm_mem_global_free(vmw_mem_glob(dev_priv), size); +} + +/** + * vmw_validation_mem_init_ttm - Interface the validation memory tracker + * to ttm. + * @dev_priv: Pointer to struct vmw_private. The reason we choose a vmw private + * rather than a struct vmw_validation_mem is to make sure assumption in the + * callbacks that struct vmw_private derives from struct vmw_validation_mem + * holds true. + * @gran: The recommended allocation granularity + */ +void vmw_validation_mem_init_ttm(struct vmw_private *dev_priv, size_t gran) +{ + struct vmw_validation_mem *vvm = &dev_priv->vvm; + + vvm->reserve_mem = vmw_vmt_reserve; + vvm->unreserve_mem = vmw_vmt_unreserve; + vvm->gran = gran; +} diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c index 184025fa938e..b3f547fc5d3d 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.c @@ -104,11 +104,25 @@ void *vmw_validation_mem_alloc(struct vmw_validation_context *ctx, return NULL; if (ctx->mem_size_left < size) { - struct page *page = alloc_page(GFP_KERNEL | __GFP_ZERO); + struct page *page; + if (ctx->vm && ctx->vm_size_left < PAGE_SIZE) { + int ret = ctx->vm->reserve_mem(ctx->vm, ctx->vm->gran); + + if (ret) + return NULL; + + ctx->vm_size_left += ctx->vm->gran; + ctx->total_mem += ctx->vm->gran; + } + + page = alloc_page(GFP_KERNEL | __GFP_ZERO); if (!page) return NULL; + if (ctx->vm) + ctx->vm_size_left -= PAGE_SIZE; + list_add_tail(&page->lru, &ctx->page_list); ctx->page_address = page_address(page); ctx->mem_size_left = PAGE_SIZE; @@ -138,6 +152,11 @@ static void vmw_validation_mem_free(struct vmw_validation_context *ctx) } ctx->mem_size_left = 0; + if (ctx->vm && ctx->total_mem) { + ctx->vm->unreserve_mem(ctx->vm, ctx->total_mem); + ctx->total_mem = 0; + ctx->vm_size_left = 0; + } } /** @@ -266,7 +285,7 @@ int vmw_validation_add_bo(struct vmw_validation_context *ctx, val_buf->bo = ttm_bo_get_unless_zero(&vbo->base); if (!val_buf->bo) return -ESRCH; - val_buf->shared = false; + val_buf->num_shared = 0; list_add_tail(&val_buf->head, &ctx->bo_list); bo_node->as_mob = as_mob; bo_node->cpu_blit = cpu_blit; diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.h b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.h index b57e3292c386..3b396fea40d7 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_validation.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_validation.h @@ -34,6 +34,21 @@ #include <drm/ttm/ttm_execbuf_util.h> /** + * struct vmw_validation_mem - Custom interface to provide memory reservations + * for the validation code. + * @reserve_mem: Callback to reserve memory + * @unreserve_mem: Callback to unreserve memory + * @gran: Reservation granularity. Contains a hint how much memory should + * be reserved in each call to @reserve_mem(). A slow implementation may want + * reservation to be done in large batches. + */ +struct vmw_validation_mem { + int (*reserve_mem)(struct vmw_validation_mem *m, size_t size); + void (*unreserve_mem)(struct vmw_validation_mem *m, size_t size); + size_t gran; +}; + +/** * struct vmw_validation_context - Per command submission validation context * @ht: Hash table used to find resource- or buffer object duplicates * @resource_list: List head for resource validation metadata @@ -47,6 +62,10 @@ * buffer objects * @mem_size_left: Free memory left in the last page in @page_list * @page_address: Kernel virtual address of the last page in @page_list + * @vm: A pointer to the memory reservation interface or NULL if no + * memory reservation is needed. + * @vm_size_left: Amount of reserved memory that so far has not been allocated. + * @total_mem: Amount of reserved memory. */ struct vmw_validation_context { struct drm_open_hash *ht; @@ -59,6 +78,9 @@ struct vmw_validation_context { unsigned int merge_dups; unsigned int mem_size_left; u8 *page_address; + struct vmw_validation_mem *vm; + size_t vm_size_left; + size_t total_mem; }; struct vmw_buffer_object; @@ -102,6 +124,21 @@ vmw_validation_has_bos(struct vmw_validation_context *ctx) } /** + * vmw_validation_set_val_mem - Register a validation mem object for + * validation memory reservation + * @ctx: The validation context + * @vm: Pointer to a struct vmw_validation_mem + * + * Must be set before the first attempt to allocate validation memory. + */ +static inline void +vmw_validation_set_val_mem(struct vmw_validation_context *ctx, + struct vmw_validation_mem *vm) +{ + ctx->vm = vm; +} + +/** * vmw_validation_set_ht - Register a hash table for duplicate finding * @ctx: The validation context * @ht: Pointer to a hash table to use for duplicate finding diff --git a/drivers/gpu/drm/xen/Kconfig b/drivers/gpu/drm/xen/Kconfig index 4cca160782ab..f969d486855d 100644 --- a/drivers/gpu/drm/xen/Kconfig +++ b/drivers/gpu/drm/xen/Kconfig @@ -12,6 +12,7 @@ config DRM_XEN_FRONTEND select DRM_KMS_HELPER select VIDEOMODE_HELPERS select XEN_XENBUS_FRONTEND + select XEN_FRONT_PGDIR_SHBUF help Choose this option if you want to enable a para-virtualized frontend DRM/KMS driver for Xen guest OSes. diff --git a/drivers/gpu/drm/xen/Makefile b/drivers/gpu/drm/xen/Makefile index 712afff5ffc3..825905f67faa 100644 --- a/drivers/gpu/drm/xen/Makefile +++ b/drivers/gpu/drm/xen/Makefile @@ -4,7 +4,6 @@ drm_xen_front-objs := xen_drm_front.o \ xen_drm_front_kms.o \ xen_drm_front_conn.o \ xen_drm_front_evtchnl.o \ - xen_drm_front_shbuf.o \ xen_drm_front_cfg.o \ xen_drm_front_gem.o diff --git a/drivers/gpu/drm/xen/xen_drm_front.c b/drivers/gpu/drm/xen/xen_drm_front.c index 6b6d5ab82ec3..4d3d36fc3a5d 100644 --- a/drivers/gpu/drm/xen/xen_drm_front.c +++ b/drivers/gpu/drm/xen/xen_drm_front.c @@ -19,6 +19,7 @@ #include <xen/xen.h> #include <xen/xenbus.h> +#include <xen/xen-front-pgdir-shbuf.h> #include <xen/interface/io/displif.h> #include "xen_drm_front.h" @@ -26,28 +27,20 @@ #include "xen_drm_front_evtchnl.h" #include "xen_drm_front_gem.h" #include "xen_drm_front_kms.h" -#include "xen_drm_front_shbuf.h" struct xen_drm_front_dbuf { struct list_head list; u64 dbuf_cookie; u64 fb_cookie; - struct xen_drm_front_shbuf *shbuf; + + struct xen_front_pgdir_shbuf shbuf; }; -static int dbuf_add_to_list(struct xen_drm_front_info *front_info, - struct xen_drm_front_shbuf *shbuf, u64 dbuf_cookie) +static void dbuf_add_to_list(struct xen_drm_front_info *front_info, + struct xen_drm_front_dbuf *dbuf, u64 dbuf_cookie) { - struct xen_drm_front_dbuf *dbuf; - - dbuf = kzalloc(sizeof(*dbuf), GFP_KERNEL); - if (!dbuf) - return -ENOMEM; - dbuf->dbuf_cookie = dbuf_cookie; - dbuf->shbuf = shbuf; list_add(&dbuf->list, &front_info->dbuf_list); - return 0; } static struct xen_drm_front_dbuf *dbuf_get(struct list_head *dbuf_list, @@ -62,15 +55,6 @@ static struct xen_drm_front_dbuf *dbuf_get(struct list_head *dbuf_list, return NULL; } -static void dbuf_flush_fb(struct list_head *dbuf_list, u64 fb_cookie) -{ - struct xen_drm_front_dbuf *buf, *q; - - list_for_each_entry_safe(buf, q, dbuf_list, list) - if (buf->fb_cookie == fb_cookie) - xen_drm_front_shbuf_flush(buf->shbuf); -} - static void dbuf_free(struct list_head *dbuf_list, u64 dbuf_cookie) { struct xen_drm_front_dbuf *buf, *q; @@ -78,8 +62,8 @@ static void dbuf_free(struct list_head *dbuf_list, u64 dbuf_cookie) list_for_each_entry_safe(buf, q, dbuf_list, list) if (buf->dbuf_cookie == dbuf_cookie) { list_del(&buf->list); - xen_drm_front_shbuf_unmap(buf->shbuf); - xen_drm_front_shbuf_free(buf->shbuf); + xen_front_pgdir_shbuf_unmap(&buf->shbuf); + xen_front_pgdir_shbuf_free(&buf->shbuf); kfree(buf); break; } @@ -91,8 +75,8 @@ static void dbuf_free_all(struct list_head *dbuf_list) list_for_each_entry_safe(buf, q, dbuf_list, list) { list_del(&buf->list); - xen_drm_front_shbuf_unmap(buf->shbuf); - xen_drm_front_shbuf_free(buf->shbuf); + xen_front_pgdir_shbuf_unmap(&buf->shbuf); + xen_front_pgdir_shbuf_free(&buf->shbuf); kfree(buf); } } @@ -171,9 +155,9 @@ int xen_drm_front_dbuf_create(struct xen_drm_front_info *front_info, u32 bpp, u64 size, struct page **pages) { struct xen_drm_front_evtchnl *evtchnl; - struct xen_drm_front_shbuf *shbuf; + struct xen_drm_front_dbuf *dbuf; struct xendispl_req *req; - struct xen_drm_front_shbuf_cfg buf_cfg; + struct xen_front_pgdir_shbuf_cfg buf_cfg; unsigned long flags; int ret; @@ -181,28 +165,29 @@ int xen_drm_front_dbuf_create(struct xen_drm_front_info *front_info, if (unlikely(!evtchnl)) return -EIO; + dbuf = kzalloc(sizeof(*dbuf), GFP_KERNEL); + if (!dbuf) + return -ENOMEM; + + dbuf_add_to_list(front_info, dbuf, dbuf_cookie); + memset(&buf_cfg, 0, sizeof(buf_cfg)); buf_cfg.xb_dev = front_info->xb_dev; + buf_cfg.num_pages = DIV_ROUND_UP(size, PAGE_SIZE); buf_cfg.pages = pages; - buf_cfg.size = size; + buf_cfg.pgdir = &dbuf->shbuf; buf_cfg.be_alloc = front_info->cfg.be_alloc; - shbuf = xen_drm_front_shbuf_alloc(&buf_cfg); - if (IS_ERR(shbuf)) - return PTR_ERR(shbuf); - - ret = dbuf_add_to_list(front_info, shbuf, dbuf_cookie); - if (ret < 0) { - xen_drm_front_shbuf_free(shbuf); - return ret; - } + ret = xen_front_pgdir_shbuf_alloc(&buf_cfg); + if (ret < 0) + goto fail_shbuf_alloc; mutex_lock(&evtchnl->u.req.req_io_lock); spin_lock_irqsave(&front_info->io_lock, flags); req = be_prepare_req(evtchnl, XENDISPL_OP_DBUF_CREATE); req->op.dbuf_create.gref_directory = - xen_drm_front_shbuf_get_dir_start(shbuf); + xen_front_pgdir_shbuf_get_dir_start(&dbuf->shbuf); req->op.dbuf_create.buffer_sz = size; req->op.dbuf_create.dbuf_cookie = dbuf_cookie; req->op.dbuf_create.width = width; @@ -221,7 +206,7 @@ int xen_drm_front_dbuf_create(struct xen_drm_front_info *front_info, if (ret < 0) goto fail; - ret = xen_drm_front_shbuf_map(shbuf); + ret = xen_front_pgdir_shbuf_map(&dbuf->shbuf); if (ret < 0) goto fail; @@ -230,6 +215,7 @@ int xen_drm_front_dbuf_create(struct xen_drm_front_info *front_info, fail: mutex_unlock(&evtchnl->u.req.req_io_lock); +fail_shbuf_alloc: dbuf_free(&front_info->dbuf_list, dbuf_cookie); return ret; } @@ -358,7 +344,6 @@ int xen_drm_front_page_flip(struct xen_drm_front_info *front_info, if (unlikely(conn_idx >= front_info->num_evt_pairs)) return -EINVAL; - dbuf_flush_fb(&front_info->dbuf_list, fb_cookie); evtchnl = &front_info->evt_pairs[conn_idx].req; mutex_lock(&evtchnl->u.req.req_io_lock); diff --git a/drivers/gpu/drm/xen/xen_drm_front_gem.c b/drivers/gpu/drm/xen/xen_drm_front_gem.c index 47ff019d3aef..28bc501af450 100644 --- a/drivers/gpu/drm/xen/xen_drm_front_gem.c +++ b/drivers/gpu/drm/xen/xen_drm_front_gem.c @@ -22,7 +22,6 @@ #include <xen/balloon.h> #include "xen_drm_front.h" -#include "xen_drm_front_shbuf.h" struct xen_gem_object { struct drm_gem_object base; diff --git a/drivers/gpu/drm/xen/xen_drm_front_shbuf.c b/drivers/gpu/drm/xen/xen_drm_front_shbuf.c deleted file mode 100644 index d333b67cc1a0..000000000000 --- a/drivers/gpu/drm/xen/xen_drm_front_shbuf.c +++ /dev/null @@ -1,414 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 OR MIT - -/* - * Xen para-virtual DRM device - * - * Copyright (C) 2016-2018 EPAM Systems Inc. - * - * Author: Oleksandr Andrushchenko <oleksandr_andrushchenko@epam.com> - */ - -#include <drm/drmP.h> - -#if defined(CONFIG_X86) -#include <drm/drm_cache.h> -#endif -#include <linux/errno.h> -#include <linux/mm.h> - -#include <asm/xen/hypervisor.h> -#include <xen/balloon.h> -#include <xen/xen.h> -#include <xen/xenbus.h> -#include <xen/interface/io/ring.h> -#include <xen/interface/io/displif.h> - -#include "xen_drm_front.h" -#include "xen_drm_front_shbuf.h" - -struct xen_drm_front_shbuf_ops { - /* - * Calculate number of grefs required to handle this buffer, - * e.g. if grefs are required for page directory only or the buffer - * pages as well. - */ - void (*calc_num_grefs)(struct xen_drm_front_shbuf *buf); - /* Fill page directory according to para-virtual display protocol. */ - void (*fill_page_dir)(struct xen_drm_front_shbuf *buf); - /* Claim grant references for the pages of the buffer. */ - int (*grant_refs_for_buffer)(struct xen_drm_front_shbuf *buf, - grant_ref_t *priv_gref_head, int gref_idx); - /* Map grant references of the buffer. */ - int (*map)(struct xen_drm_front_shbuf *buf); - /* Unmap grant references of the buffer. */ - int (*unmap)(struct xen_drm_front_shbuf *buf); -}; - -grant_ref_t xen_drm_front_shbuf_get_dir_start(struct xen_drm_front_shbuf *buf) -{ - if (!buf->grefs) - return GRANT_INVALID_REF; - - return buf->grefs[0]; -} - -int xen_drm_front_shbuf_map(struct xen_drm_front_shbuf *buf) -{ - if (buf->ops->map) - return buf->ops->map(buf); - - /* no need to map own grant references */ - return 0; -} - -int xen_drm_front_shbuf_unmap(struct xen_drm_front_shbuf *buf) -{ - if (buf->ops->unmap) - return buf->ops->unmap(buf); - - /* no need to unmap own grant references */ - return 0; -} - -void xen_drm_front_shbuf_flush(struct xen_drm_front_shbuf *buf) -{ -#if defined(CONFIG_X86) - drm_clflush_pages(buf->pages, buf->num_pages); -#endif -} - -void xen_drm_front_shbuf_free(struct xen_drm_front_shbuf *buf) -{ - if (buf->grefs) { - int i; - - for (i = 0; i < buf->num_grefs; i++) - if (buf->grefs[i] != GRANT_INVALID_REF) - gnttab_end_foreign_access(buf->grefs[i], - 0, 0UL); - } - kfree(buf->grefs); - kfree(buf->directory); - kfree(buf); -} - -/* - * number of grefs a page can hold with respect to the - * struct xendispl_page_directory header - */ -#define XEN_DRM_NUM_GREFS_PER_PAGE ((PAGE_SIZE - \ - offsetof(struct xendispl_page_directory, gref)) / \ - sizeof(grant_ref_t)) - -static int get_num_pages_dir(struct xen_drm_front_shbuf *buf) -{ - /* number of pages the page directory consumes itself */ - return DIV_ROUND_UP(buf->num_pages, XEN_DRM_NUM_GREFS_PER_PAGE); -} - -static void backend_calc_num_grefs(struct xen_drm_front_shbuf *buf) -{ - /* only for pages the page directory consumes itself */ - buf->num_grefs = get_num_pages_dir(buf); -} - -static void guest_calc_num_grefs(struct xen_drm_front_shbuf *buf) -{ - /* - * number of pages the page directory consumes itself - * plus grefs for the buffer pages - */ - buf->num_grefs = get_num_pages_dir(buf) + buf->num_pages; -} - -#define xen_page_to_vaddr(page) \ - ((uintptr_t)pfn_to_kaddr(page_to_xen_pfn(page))) - -static int backend_unmap(struct xen_drm_front_shbuf *buf) -{ - struct gnttab_unmap_grant_ref *unmap_ops; - int i, ret; - - if (!buf->pages || !buf->backend_map_handles || !buf->grefs) - return 0; - - unmap_ops = kcalloc(buf->num_pages, sizeof(*unmap_ops), - GFP_KERNEL); - if (!unmap_ops) { - DRM_ERROR("Failed to get memory while unmapping\n"); - return -ENOMEM; - } - - for (i = 0; i < buf->num_pages; i++) { - phys_addr_t addr; - - addr = xen_page_to_vaddr(buf->pages[i]); - gnttab_set_unmap_op(&unmap_ops[i], addr, GNTMAP_host_map, - buf->backend_map_handles[i]); - } - - ret = gnttab_unmap_refs(unmap_ops, NULL, buf->pages, - buf->num_pages); - - for (i = 0; i < buf->num_pages; i++) { - if (unlikely(unmap_ops[i].status != GNTST_okay)) - DRM_ERROR("Failed to unmap page %d: %d\n", - i, unmap_ops[i].status); - } - - if (ret) - DRM_ERROR("Failed to unmap grant references, ret %d", ret); - - kfree(unmap_ops); - kfree(buf->backend_map_handles); - buf->backend_map_handles = NULL; - return ret; -} - -static int backend_map(struct xen_drm_front_shbuf *buf) -{ - struct gnttab_map_grant_ref *map_ops = NULL; - unsigned char *ptr; - int ret, cur_gref, cur_dir_page, cur_page, grefs_left; - - map_ops = kcalloc(buf->num_pages, sizeof(*map_ops), GFP_KERNEL); - if (!map_ops) - return -ENOMEM; - - buf->backend_map_handles = kcalloc(buf->num_pages, - sizeof(*buf->backend_map_handles), - GFP_KERNEL); - if (!buf->backend_map_handles) { - kfree(map_ops); - return -ENOMEM; - } - - /* - * read page directory to get grefs from the backend: for external - * buffer we only allocate buf->grefs for the page directory, - * so buf->num_grefs has number of pages in the page directory itself - */ - ptr = buf->directory; - grefs_left = buf->num_pages; - cur_page = 0; - for (cur_dir_page = 0; cur_dir_page < buf->num_grefs; cur_dir_page++) { - struct xendispl_page_directory *page_dir = - (struct xendispl_page_directory *)ptr; - int to_copy = XEN_DRM_NUM_GREFS_PER_PAGE; - - if (to_copy > grefs_left) - to_copy = grefs_left; - - for (cur_gref = 0; cur_gref < to_copy; cur_gref++) { - phys_addr_t addr; - - addr = xen_page_to_vaddr(buf->pages[cur_page]); - gnttab_set_map_op(&map_ops[cur_page], addr, - GNTMAP_host_map, - page_dir->gref[cur_gref], - buf->xb_dev->otherend_id); - cur_page++; - } - - grefs_left -= to_copy; - ptr += PAGE_SIZE; - } - ret = gnttab_map_refs(map_ops, NULL, buf->pages, buf->num_pages); - - /* save handles even if error, so we can unmap */ - for (cur_page = 0; cur_page < buf->num_pages; cur_page++) { - buf->backend_map_handles[cur_page] = map_ops[cur_page].handle; - if (unlikely(map_ops[cur_page].status != GNTST_okay)) - DRM_ERROR("Failed to map page %d: %d\n", - cur_page, map_ops[cur_page].status); - } - - if (ret) { - DRM_ERROR("Failed to map grant references, ret %d", ret); - backend_unmap(buf); - } - - kfree(map_ops); - return ret; -} - -static void backend_fill_page_dir(struct xen_drm_front_shbuf *buf) -{ - struct xendispl_page_directory *page_dir; - unsigned char *ptr; - int i, num_pages_dir; - - ptr = buf->directory; - num_pages_dir = get_num_pages_dir(buf); - - /* fill only grefs for the page directory itself */ - for (i = 0; i < num_pages_dir - 1; i++) { - page_dir = (struct xendispl_page_directory *)ptr; - - page_dir->gref_dir_next_page = buf->grefs[i + 1]; - ptr += PAGE_SIZE; - } - /* last page must say there is no more pages */ - page_dir = (struct xendispl_page_directory *)ptr; - page_dir->gref_dir_next_page = GRANT_INVALID_REF; -} - -static void guest_fill_page_dir(struct xen_drm_front_shbuf *buf) -{ - unsigned char *ptr; - int cur_gref, grefs_left, to_copy, i, num_pages_dir; - - ptr = buf->directory; - num_pages_dir = get_num_pages_dir(buf); - - /* - * while copying, skip grefs at start, they are for pages - * granted for the page directory itself - */ - cur_gref = num_pages_dir; - grefs_left = buf->num_pages; - for (i = 0; i < num_pages_dir; i++) { - struct xendispl_page_directory *page_dir = - (struct xendispl_page_directory *)ptr; - - if (grefs_left <= XEN_DRM_NUM_GREFS_PER_PAGE) { - to_copy = grefs_left; - page_dir->gref_dir_next_page = GRANT_INVALID_REF; - } else { - to_copy = XEN_DRM_NUM_GREFS_PER_PAGE; - page_dir->gref_dir_next_page = buf->grefs[i + 1]; - } - memcpy(&page_dir->gref, &buf->grefs[cur_gref], - to_copy * sizeof(grant_ref_t)); - ptr += PAGE_SIZE; - grefs_left -= to_copy; - cur_gref += to_copy; - } -} - -static int guest_grant_refs_for_buffer(struct xen_drm_front_shbuf *buf, - grant_ref_t *priv_gref_head, - int gref_idx) -{ - int i, cur_ref, otherend_id; - - otherend_id = buf->xb_dev->otherend_id; - for (i = 0; i < buf->num_pages; i++) { - cur_ref = gnttab_claim_grant_reference(priv_gref_head); - if (cur_ref < 0) - return cur_ref; - - gnttab_grant_foreign_access_ref(cur_ref, otherend_id, - xen_page_to_gfn(buf->pages[i]), - 0); - buf->grefs[gref_idx++] = cur_ref; - } - return 0; -} - -static int grant_references(struct xen_drm_front_shbuf *buf) -{ - grant_ref_t priv_gref_head; - int ret, i, j, cur_ref; - int otherend_id, num_pages_dir; - - ret = gnttab_alloc_grant_references(buf->num_grefs, &priv_gref_head); - if (ret < 0) { - DRM_ERROR("Cannot allocate grant references\n"); - return ret; - } - - otherend_id = buf->xb_dev->otherend_id; - j = 0; - num_pages_dir = get_num_pages_dir(buf); - for (i = 0; i < num_pages_dir; i++) { - unsigned long frame; - - cur_ref = gnttab_claim_grant_reference(&priv_gref_head); - if (cur_ref < 0) - return cur_ref; - - frame = xen_page_to_gfn(virt_to_page(buf->directory + - PAGE_SIZE * i)); - gnttab_grant_foreign_access_ref(cur_ref, otherend_id, frame, 0); - buf->grefs[j++] = cur_ref; - } - - if (buf->ops->grant_refs_for_buffer) { - ret = buf->ops->grant_refs_for_buffer(buf, &priv_gref_head, j); - if (ret) - return ret; - } - - gnttab_free_grant_references(priv_gref_head); - return 0; -} - -static int alloc_storage(struct xen_drm_front_shbuf *buf) -{ - buf->grefs = kcalloc(buf->num_grefs, sizeof(*buf->grefs), GFP_KERNEL); - if (!buf->grefs) - return -ENOMEM; - - buf->directory = kcalloc(get_num_pages_dir(buf), PAGE_SIZE, GFP_KERNEL); - if (!buf->directory) - return -ENOMEM; - - return 0; -} - -/* - * For be allocated buffers we don't need grant_refs_for_buffer as those - * grant references are allocated at backend side - */ -static const struct xen_drm_front_shbuf_ops backend_ops = { - .calc_num_grefs = backend_calc_num_grefs, - .fill_page_dir = backend_fill_page_dir, - .map = backend_map, - .unmap = backend_unmap -}; - -/* For locally granted references we do not need to map/unmap the references */ -static const struct xen_drm_front_shbuf_ops local_ops = { - .calc_num_grefs = guest_calc_num_grefs, - .fill_page_dir = guest_fill_page_dir, - .grant_refs_for_buffer = guest_grant_refs_for_buffer, -}; - -struct xen_drm_front_shbuf * -xen_drm_front_shbuf_alloc(struct xen_drm_front_shbuf_cfg *cfg) -{ - struct xen_drm_front_shbuf *buf; - int ret; - - buf = kzalloc(sizeof(*buf), GFP_KERNEL); - if (!buf) - return ERR_PTR(-ENOMEM); - - if (cfg->be_alloc) - buf->ops = &backend_ops; - else - buf->ops = &local_ops; - - buf->xb_dev = cfg->xb_dev; - buf->num_pages = DIV_ROUND_UP(cfg->size, PAGE_SIZE); - buf->pages = cfg->pages; - - buf->ops->calc_num_grefs(buf); - - ret = alloc_storage(buf); - if (ret) - goto fail; - - ret = grant_references(buf); - if (ret) - goto fail; - - buf->ops->fill_page_dir(buf); - - return buf; - -fail: - xen_drm_front_shbuf_free(buf); - return ERR_PTR(ret); -} diff --git a/drivers/gpu/drm/xen/xen_drm_front_shbuf.h b/drivers/gpu/drm/xen/xen_drm_front_shbuf.h deleted file mode 100644 index 7545c692539e..000000000000 --- a/drivers/gpu/drm/xen/xen_drm_front_shbuf.h +++ /dev/null @@ -1,64 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 OR MIT */ - -/* - * Xen para-virtual DRM device - * - * Copyright (C) 2016-2018 EPAM Systems Inc. - * - * Author: Oleksandr Andrushchenko <oleksandr_andrushchenko@epam.com> - */ - -#ifndef __XEN_DRM_FRONT_SHBUF_H_ -#define __XEN_DRM_FRONT_SHBUF_H_ - -#include <linux/kernel.h> -#include <linux/scatterlist.h> - -#include <xen/grant_table.h> - -struct xen_drm_front_shbuf { - /* - * number of references granted for the backend use: - * - for allocated/imported dma-buf's this holds number of grant - * references for the page directory and pages of the buffer - * - for the buffer provided by the backend this holds number of - * grant references for the page directory as grant references for - * the buffer will be provided by the backend - */ - int num_grefs; - grant_ref_t *grefs; - unsigned char *directory; - - int num_pages; - struct page **pages; - - struct xenbus_device *xb_dev; - - /* these are the ops used internally depending on be_alloc mode */ - const struct xen_drm_front_shbuf_ops *ops; - - /* Xen map handles for the buffer allocated by the backend */ - grant_handle_t *backend_map_handles; -}; - -struct xen_drm_front_shbuf_cfg { - struct xenbus_device *xb_dev; - size_t size; - struct page **pages; - bool be_alloc; -}; - -struct xen_drm_front_shbuf * -xen_drm_front_shbuf_alloc(struct xen_drm_front_shbuf_cfg *cfg); - -grant_ref_t xen_drm_front_shbuf_get_dir_start(struct xen_drm_front_shbuf *buf); - -int xen_drm_front_shbuf_map(struct xen_drm_front_shbuf *buf); - -int xen_drm_front_shbuf_unmap(struct xen_drm_front_shbuf *buf); - -void xen_drm_front_shbuf_flush(struct xen_drm_front_shbuf *buf); - -void xen_drm_front_shbuf_free(struct xen_drm_front_shbuf *buf); - -#endif /* __XEN_DRM_FRONT_SHBUF_H_ */ diff --git a/drivers/gpu/host1x/Makefile b/drivers/gpu/host1x/Makefile index b92016ce09b7..096017b8789d 100644 --- a/drivers/gpu/host1x/Makefile +++ b/drivers/gpu/host1x/Makefile @@ -13,6 +13,7 @@ host1x-y = \ hw/host1x02.o \ hw/host1x04.o \ hw/host1x05.o \ - hw/host1x06.o + hw/host1x06.o \ + hw/host1x07.o obj-$(CONFIG_TEGRA_HOST1X) += host1x.o diff --git a/drivers/gpu/host1x/dev.c b/drivers/gpu/host1x/dev.c index de6bc4e7fa23..419d8929a98f 100644 --- a/drivers/gpu/host1x/dev.c +++ b/drivers/gpu/host1x/dev.c @@ -44,6 +44,7 @@ #include "hw/host1x04.h" #include "hw/host1x05.h" #include "hw/host1x06.h" +#include "hw/host1x07.h" void host1x_hypervisor_writel(struct host1x *host1x, u32 v, u32 r) { @@ -130,7 +131,19 @@ static const struct host1x_info host1x06_info = { .has_hypervisor = true, }; +static const struct host1x_info host1x07_info = { + .nb_channels = 63, + .nb_pts = 704, + .nb_mlocks = 32, + .nb_bases = 0, + .init = host1x07_init, + .sync_offset = 0x0, + .dma_mask = DMA_BIT_MASK(40), + .has_hypervisor = true, +}; + static const struct of_device_id host1x_of_match[] = { + { .compatible = "nvidia,tegra194-host1x", .data = &host1x07_info, }, { .compatible = "nvidia,tegra186-host1x", .data = &host1x06_info, }, { .compatible = "nvidia,tegra210-host1x", .data = &host1x05_info, }, { .compatible = "nvidia,tegra124-host1x", .data = &host1x04_info, }, diff --git a/drivers/gpu/host1x/hw/channel_hw.c b/drivers/gpu/host1x/hw/channel_hw.c index d188f9068b91..95ea81172a83 100644 --- a/drivers/gpu/host1x/hw/channel_hw.c +++ b/drivers/gpu/host1x/hw/channel_hw.c @@ -26,7 +26,6 @@ #include "../intr.h" #include "../job.h" -#define HOST1X_CHANNEL_SIZE 16384 #define TRACE_MAX_LENGTH 128U static void trace_write_gather(struct host1x_cdma *cdma, struct host1x_bo *bo, @@ -203,7 +202,11 @@ static void enable_gather_filter(struct host1x *host, static int host1x_channel_init(struct host1x_channel *ch, struct host1x *dev, unsigned int index) { - ch->regs = dev->regs + index * HOST1X_CHANNEL_SIZE; +#if HOST1X_HW < 6 + ch->regs = dev->regs + index * 0x4000; +#else + ch->regs = dev->regs + index * 0x100; +#endif enable_gather_filter(dev, ch); return 0; } diff --git a/drivers/gpu/host1x/hw/debug_hw_1x06.c b/drivers/gpu/host1x/hw/debug_hw_1x06.c index b503c740c022..8b749516c051 100644 --- a/drivers/gpu/host1x/hw/debug_hw_1x06.c +++ b/drivers/gpu/host1x/hw/debug_hw_1x06.c @@ -62,9 +62,12 @@ static void host1x_debug_show_channel_fifo(struct host1x *host, struct host1x_channel *ch, struct output *o) { - u32 val, rd_ptr, wr_ptr, start, end; +#if HOST1X_HW <= 6 + u32 rd_ptr, wr_ptr, start, end; u32 payload = INVALID_PAYLOAD; unsigned int data_count = 0; +#endif + u32 val; host1x_debug_output(o, "%u: fifo:\n", ch->id); @@ -78,6 +81,7 @@ static void host1x_debug_show_channel_fifo(struct host1x *host, val = host1x_ch_readl(ch, HOST1X_CHANNEL_CMDFIFO_RDATA); host1x_debug_output(o, "CMDFIFO_RDATA %08x\n", val); +#if HOST1X_HW <= 6 /* Peek pointer values are invalid during SLCG, so disable it */ host1x_hypervisor_writel(host, 0x1, HOST1X_HV_ICG_EN_OVERRIDE); @@ -127,6 +131,7 @@ static void host1x_debug_show_channel_fifo(struct host1x *host, host1x_hypervisor_writel(host, 0x0, HOST1X_HV_CMDFIFO_PEEK_CTRL); host1x_hypervisor_writel(host, 0x0, HOST1X_HV_ICG_EN_OVERRIDE); +#endif } static void host1x_debug_show_mlocks(struct host1x *host, struct output *o) diff --git a/drivers/gpu/host1x/hw/host1x07.c b/drivers/gpu/host1x/hw/host1x07.c new file mode 100644 index 000000000000..04b779a53f08 --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x07.c @@ -0,0 +1,44 @@ +/* + * Host1x init for Tegra194 SoCs + * + * Copyright (c) 2018 NVIDIA Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +/* include hw specification */ +#include "host1x07.h" +#include "host1x07_hardware.h" + +/* include code */ +#define HOST1X_HW 7 + +#include "cdma_hw.c" +#include "channel_hw.c" +#include "debug_hw.c" +#include "intr_hw.c" +#include "syncpt_hw.c" + +#include "../dev.h" + +int host1x07_init(struct host1x *host) +{ + host->channel_op = &host1x_channel_ops; + host->cdma_op = &host1x_cdma_ops; + host->cdma_pb_op = &host1x_pushbuffer_ops; + host->syncpt_op = &host1x_syncpt_ops; + host->intr_op = &host1x_intr_ops; + host->debug_op = &host1x_debug_ops; + + return 0; +} diff --git a/drivers/gpu/host1x/hw/host1x07.h b/drivers/gpu/host1x/hw/host1x07.h new file mode 100644 index 000000000000..57b19f354274 --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x07.h @@ -0,0 +1,26 @@ +/* + * Host1x init for Tegra194 SoCs + * + * Copyright (c) 2018 NVIDIA Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef HOST1X_HOST1X07_H +#define HOST1X_HOST1X07_H + +struct host1x; + +int host1x07_init(struct host1x *host); + +#endif diff --git a/drivers/gpu/host1x/hw/host1x07_hardware.h b/drivers/gpu/host1x/hw/host1x07_hardware.h new file mode 100644 index 000000000000..1353e7ab71dd --- /dev/null +++ b/drivers/gpu/host1x/hw/host1x07_hardware.h @@ -0,0 +1,142 @@ +/* + * Tegra host1x Register Offsets for Tegra194 + * + * Copyright (c) 2018 NVIDIA Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +#ifndef __HOST1X_HOST1X07_HARDWARE_H +#define __HOST1X_HOST1X07_HARDWARE_H + +#include <linux/types.h> +#include <linux/bitops.h> + +#include "hw_host1x07_uclass.h" +#include "hw_host1x07_vm.h" +#include "hw_host1x07_hypervisor.h" + +static inline u32 host1x_class_host_wait_syncpt( + unsigned indx, unsigned threshold) +{ + return host1x_uclass_wait_syncpt_indx_f(indx) + | host1x_uclass_wait_syncpt_thresh_f(threshold); +} + +static inline u32 host1x_class_host_load_syncpt_base( + unsigned indx, unsigned threshold) +{ + return host1x_uclass_load_syncpt_base_base_indx_f(indx) + | host1x_uclass_load_syncpt_base_value_f(threshold); +} + +static inline u32 host1x_class_host_wait_syncpt_base( + unsigned indx, unsigned base_indx, unsigned offset) +{ + return host1x_uclass_wait_syncpt_base_indx_f(indx) + | host1x_uclass_wait_syncpt_base_base_indx_f(base_indx) + | host1x_uclass_wait_syncpt_base_offset_f(offset); +} + +static inline u32 host1x_class_host_incr_syncpt_base( + unsigned base_indx, unsigned offset) +{ + return host1x_uclass_incr_syncpt_base_base_indx_f(base_indx) + | host1x_uclass_incr_syncpt_base_offset_f(offset); +} + +static inline u32 host1x_class_host_incr_syncpt( + unsigned cond, unsigned indx) +{ + return host1x_uclass_incr_syncpt_cond_f(cond) + | host1x_uclass_incr_syncpt_indx_f(indx); +} + +static inline u32 host1x_class_host_indoff_reg_write( + unsigned mod_id, unsigned offset, bool auto_inc) +{ + u32 v = host1x_uclass_indoff_indbe_f(0xf) + | host1x_uclass_indoff_indmodid_f(mod_id) + | host1x_uclass_indoff_indroffset_f(offset); + if (auto_inc) + v |= host1x_uclass_indoff_autoinc_f(1); + return v; +} + +static inline u32 host1x_class_host_indoff_reg_read( + unsigned mod_id, unsigned offset, bool auto_inc) +{ + u32 v = host1x_uclass_indoff_indmodid_f(mod_id) + | host1x_uclass_indoff_indroffset_f(offset) + | host1x_uclass_indoff_rwn_read_v(); + if (auto_inc) + v |= host1x_uclass_indoff_autoinc_f(1); + return v; +} + +/* cdma opcodes */ +static inline u32 host1x_opcode_setclass( + unsigned class_id, unsigned offset, unsigned mask) +{ + return (0 << 28) | (offset << 16) | (class_id << 6) | mask; +} + +static inline u32 host1x_opcode_incr(unsigned offset, unsigned count) +{ + return (1 << 28) | (offset << 16) | count; +} + +static inline u32 host1x_opcode_nonincr(unsigned offset, unsigned count) +{ + return (2 << 28) | (offset << 16) | count; +} + +static inline u32 host1x_opcode_mask(unsigned offset, unsigned mask) +{ + return (3 << 28) | (offset << 16) | mask; +} + +static inline u32 host1x_opcode_imm(unsigned offset, unsigned value) +{ + return (4 << 28) | (offset << 16) | value; +} + +static inline u32 host1x_opcode_imm_incr_syncpt(unsigned cond, unsigned indx) +{ + return host1x_opcode_imm(host1x_uclass_incr_syncpt_r(), + host1x_class_host_incr_syncpt(cond, indx)); +} + +static inline u32 host1x_opcode_restart(unsigned address) +{ + return (5 << 28) | (address >> 4); +} + +static inline u32 host1x_opcode_gather(unsigned count) +{ + return (6 << 28) | count; +} + +static inline u32 host1x_opcode_gather_nonincr(unsigned offset, unsigned count) +{ + return (6 << 28) | (offset << 16) | BIT(15) | count; +} + +static inline u32 host1x_opcode_gather_incr(unsigned offset, unsigned count) +{ + return (6 << 28) | (offset << 16) | BIT(15) | BIT(14) | count; +} + +#define HOST1X_OPCODE_NOP host1x_opcode_nonincr(0, 0) + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x06_uclass.h b/drivers/gpu/host1x/hw/hw_host1x06_uclass.h index 4457486c72b0..e599e15bf999 100644 --- a/drivers/gpu/host1x/hw/hw_host1x06_uclass.h +++ b/drivers/gpu/host1x/hw/hw_host1x06_uclass.h @@ -59,7 +59,7 @@ static inline u32 host1x_uclass_incr_syncpt_r(void) host1x_uclass_incr_syncpt_r() static inline u32 host1x_uclass_incr_syncpt_cond_f(u32 v) { - return (v & 0xff) << 8; + return (v & 0xff) << 10; } #define HOST1X_UCLASS_INCR_SYNCPT_COND_F(v) \ host1x_uclass_incr_syncpt_cond_f(v) diff --git a/drivers/gpu/host1x/hw/hw_host1x07_hypervisor.h b/drivers/gpu/host1x/hw/hw_host1x07_hypervisor.h new file mode 100644 index 000000000000..2b99d68d3040 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x07_hypervisor.h @@ -0,0 +1,32 @@ +/* + * Copyright (c) 2018 NVIDIA Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + * + */ + +#define HOST1X_HV_SYNCPT_PROT_EN 0x1ac4 +#define HOST1X_HV_SYNCPT_PROT_EN_CH_EN BIT(1) +#define HOST1X_HV_CH_KERNEL_FILTER_GBUFFER(x) (0x2020 + (x * 4)) +#define HOST1X_HV_CMDFIFO_PEEK_CTRL 0x233c +#define HOST1X_HV_CMDFIFO_PEEK_CTRL_ADDR(x) (x) +#define HOST1X_HV_CMDFIFO_PEEK_CTRL_CHANNEL(x) ((x) << 16) +#define HOST1X_HV_CMDFIFO_PEEK_CTRL_ENABLE BIT(31) +#define HOST1X_HV_CMDFIFO_PEEK_READ 0x2340 +#define HOST1X_HV_CMDFIFO_PEEK_PTRS 0x2344 +#define HOST1X_HV_CMDFIFO_PEEK_PTRS_WR_PTR_V(x) (((x) >> 16) & 0xfff) +#define HOST1X_HV_CMDFIFO_PEEK_PTRS_RD_PTR_V(x) ((x) & 0xfff) +#define HOST1X_HV_CMDFIFO_SETUP(x) (0x2588 + (x * 4)) +#define HOST1X_HV_CMDFIFO_SETUP_LIMIT_V(x) (((x) >> 16) & 0xfff) +#define HOST1X_HV_CMDFIFO_SETUP_BASE_V(x) ((x) & 0xfff) +#define HOST1X_HV_ICG_EN_OVERRIDE 0x2aa8 diff --git a/drivers/gpu/host1x/hw/hw_host1x07_uclass.h b/drivers/gpu/host1x/hw/hw_host1x07_uclass.h new file mode 100644 index 000000000000..7e4e3b377f91 --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x07_uclass.h @@ -0,0 +1,181 @@ +/* + * Copyright (c) 2018 NVIDIA Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + * + */ + + /* + * Function naming determines intended use: + * + * <x>_r(void) : Returns the offset for register <x>. + * + * <x>_w(void) : Returns the word offset for word (4 byte) element <x>. + * + * <x>_<y>_s(void) : Returns size of field <y> of register <x> in bits. + * + * <x>_<y>_f(u32 v) : Returns a value based on 'v' which has been shifted + * and masked to place it at field <y> of register <x>. This value + * can be |'d with others to produce a full register value for + * register <x>. + * + * <x>_<y>_m(void) : Returns a mask for field <y> of register <x>. This + * value can be ~'d and then &'d to clear the value of field <y> for + * register <x>. + * + * <x>_<y>_<z>_f(void) : Returns the constant value <z> after being shifted + * to place it at field <y> of register <x>. This value can be |'d + * with others to produce a full register value for <x>. + * + * <x>_<y>_v(u32 r) : Returns the value of field <y> from a full register + * <x> value 'r' after being shifted to place its LSB at bit 0. + * This value is suitable for direct comparison with other unshifted + * values appropriate for use in field <y> of register <x>. + * + * <x>_<y>_<z>_v(void) : Returns the constant value for <z> defined for + * field <y> of register <x>. This value is suitable for direct + * comparison with unshifted values appropriate for use in field <y> + * of register <x>. + */ + +#ifndef HOST1X_HW_HOST1X07_UCLASS_H +#define HOST1X_HW_HOST1X07_UCLASS_H + +static inline u32 host1x_uclass_incr_syncpt_r(void) +{ + return 0x0; +} +#define HOST1X_UCLASS_INCR_SYNCPT \ + host1x_uclass_incr_syncpt_r() +static inline u32 host1x_uclass_incr_syncpt_cond_f(u32 v) +{ + return (v & 0xff) << 10; +} +#define HOST1X_UCLASS_INCR_SYNCPT_COND_F(v) \ + host1x_uclass_incr_syncpt_cond_f(v) +static inline u32 host1x_uclass_incr_syncpt_indx_f(u32 v) +{ + return (v & 0xff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_INDX_F(v) \ + host1x_uclass_incr_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_r(void) +{ + return 0x8; +} +#define HOST1X_UCLASS_WAIT_SYNCPT \ + host1x_uclass_wait_syncpt_r() +static inline u32 host1x_uclass_wait_syncpt_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_INDX_F(v) \ + host1x_uclass_wait_syncpt_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_thresh_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_THRESH_F(v) \ + host1x_uclass_wait_syncpt_thresh_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_r(void) +{ + return 0x9; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE \ + host1x_uclass_wait_syncpt_base_r() +static inline u32 host1x_uclass_wait_syncpt_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 16; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_wait_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_wait_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffff) << 0; +} +#define HOST1X_UCLASS_WAIT_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_wait_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_load_syncpt_base_r(void) +{ + return 0xb; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE \ + host1x_uclass_load_syncpt_base_r() +static inline u32 host1x_uclass_load_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_load_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_load_syncpt_base_value_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_LOAD_SYNCPT_BASE_VALUE_F(v) \ + host1x_uclass_load_syncpt_base_value_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_base_indx_f(u32 v) +{ + return (v & 0xff) << 24; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_BASE_INDX_F(v) \ + host1x_uclass_incr_syncpt_base_base_indx_f(v) +static inline u32 host1x_uclass_incr_syncpt_base_offset_f(u32 v) +{ + return (v & 0xffffff) << 0; +} +#define HOST1X_UCLASS_INCR_SYNCPT_BASE_OFFSET_F(v) \ + host1x_uclass_incr_syncpt_base_offset_f(v) +static inline u32 host1x_uclass_indoff_r(void) +{ + return 0x2d; +} +#define HOST1X_UCLASS_INDOFF \ + host1x_uclass_indoff_r() +static inline u32 host1x_uclass_indoff_indbe_f(u32 v) +{ + return (v & 0xf) << 28; +} +#define HOST1X_UCLASS_INDOFF_INDBE_F(v) \ + host1x_uclass_indoff_indbe_f(v) +static inline u32 host1x_uclass_indoff_autoinc_f(u32 v) +{ + return (v & 0x1) << 27; +} +#define HOST1X_UCLASS_INDOFF_AUTOINC_F(v) \ + host1x_uclass_indoff_autoinc_f(v) +static inline u32 host1x_uclass_indoff_indmodid_f(u32 v) +{ + return (v & 0xff) << 18; +} +#define HOST1X_UCLASS_INDOFF_INDMODID_F(v) \ + host1x_uclass_indoff_indmodid_f(v) +static inline u32 host1x_uclass_indoff_indroffset_f(u32 v) +{ + return (v & 0xffff) << 2; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) +static inline u32 host1x_uclass_indoff_rwn_read_v(void) +{ + return 1; +} +#define HOST1X_UCLASS_INDOFF_INDROFFSET_F(v) \ + host1x_uclass_indoff_indroffset_f(v) + +#endif diff --git a/drivers/gpu/host1x/hw/hw_host1x07_vm.h b/drivers/gpu/host1x/hw/hw_host1x07_vm.h new file mode 100644 index 000000000000..7e4629e77a2a --- /dev/null +++ b/drivers/gpu/host1x/hw/hw_host1x07_vm.h @@ -0,0 +1,46 @@ +/* + * Copyright (c) 2018 NVIDIA Corporation. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + * + */ + +#define HOST1X_CHANNEL_DMASTART 0x0000 +#define HOST1X_CHANNEL_DMASTART_HI 0x0004 +#define HOST1X_CHANNEL_DMAPUT 0x0008 +#define HOST1X_CHANNEL_DMAPUT_HI 0x000c +#define HOST1X_CHANNEL_DMAGET 0x0010 +#define HOST1X_CHANNEL_DMAGET_HI 0x0014 +#define HOST1X_CHANNEL_DMAEND 0x0018 +#define HOST1X_CHANNEL_DMAEND_HI 0x001c +#define HOST1X_CHANNEL_DMACTRL 0x0020 +#define HOST1X_CHANNEL_DMACTRL_DMASTOP BIT(0) +#define HOST1X_CHANNEL_DMACTRL_DMAGETRST BIT(1) +#define HOST1X_CHANNEL_DMACTRL_DMAINITGET BIT(2) +#define HOST1X_CHANNEL_CMDFIFO_STAT 0x0024 +#define HOST1X_CHANNEL_CMDFIFO_STAT_EMPTY BIT(13) +#define HOST1X_CHANNEL_CMDFIFO_RDATA 0x0028 +#define HOST1X_CHANNEL_CMDP_OFFSET 0x0030 +#define HOST1X_CHANNEL_CMDP_CLASS 0x0034 +#define HOST1X_CHANNEL_CHANNELSTAT 0x0038 +#define HOST1X_CHANNEL_CMDPROC_STOP 0x0048 +#define HOST1X_CHANNEL_TEARDOWN 0x004c + +#define HOST1X_SYNC_SYNCPT_CPU_INCR(x) (0x6400 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_THRESH_CPU0_INT_STATUS(x) (0x6464 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_THRESH_INT_ENABLE_CPU0(x) (0x652c + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_THRESH_INT_DISABLE(x) (0x6590 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT(x) (0x8080 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_INT_THRESH(x) (0x8d00 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_CH_APP(x) (0xa604 + 4 * (x)) +#define HOST1X_SYNC_SYNCPT_CH_APP_CH(v) (((v) & 0x3f) << 8) diff --git a/drivers/gpu/host1x/hw/syncpt_hw.c b/drivers/gpu/host1x/hw/syncpt_hw.c index a23bb3352d02..d946660d47f8 100644 --- a/drivers/gpu/host1x/hw/syncpt_hw.c +++ b/drivers/gpu/host1x/hw/syncpt_hw.c @@ -37,10 +37,12 @@ static void syncpt_restore(struct host1x_syncpt *sp) */ static void syncpt_restore_wait_base(struct host1x_syncpt *sp) { +#if HOST1X_HW < 7 struct host1x *host = sp->host; host1x_sync_writel(host, sp->base_val, HOST1X_SYNC_SYNCPT_BASE(sp->id)); +#endif } /* @@ -48,10 +50,12 @@ static void syncpt_restore_wait_base(struct host1x_syncpt *sp) */ static void syncpt_read_wait_base(struct host1x_syncpt *sp) { +#if HOST1X_HW < 7 struct host1x *host = sp->host; sp->base_val = host1x_sync_readl(host, HOST1X_SYNC_SYNCPT_BASE(sp->id)); +#endif } /* diff --git a/drivers/gpu/ipu-v3/ipu-cpmem.c b/drivers/gpu/ipu-v3/ipu-cpmem.c index a9d2501500a1..163fadb8a33a 100644 --- a/drivers/gpu/ipu-v3/ipu-cpmem.c +++ b/drivers/gpu/ipu-v3/ipu-cpmem.c @@ -259,6 +259,8 @@ EXPORT_SYMBOL_GPL(ipu_cpmem_set_high_priority); void ipu_cpmem_set_buffer(struct ipuv3_channel *ch, int bufnum, dma_addr_t buf) { + WARN_ON_ONCE(buf & 0x7); + if (bufnum) ipu_ch_param_write_field(ch, IPU_FIELD_EBA1, buf >> 3); else @@ -268,6 +270,8 @@ EXPORT_SYMBOL_GPL(ipu_cpmem_set_buffer); void ipu_cpmem_set_uv_offset(struct ipuv3_channel *ch, u32 u_off, u32 v_off) { + WARN_ON_ONCE((u_off & 0x7) || (v_off & 0x7)); + ipu_ch_param_write_field(ch, IPU_FIELD_UBO, u_off / 8); ipu_ch_param_write_field(ch, IPU_FIELD_VBO, v_off / 8); } @@ -435,6 +439,8 @@ void ipu_cpmem_set_yuv_planar_full(struct ipuv3_channel *ch, unsigned int uv_stride, unsigned int u_offset, unsigned int v_offset) { + WARN_ON_ONCE((u_offset & 0x7) || (v_offset & 0x7)); + ipu_ch_param_write_field(ch, IPU_FIELD_SLUV, uv_stride - 1); ipu_ch_param_write_field(ch, IPU_FIELD_UBO, u_offset / 8); ipu_ch_param_write_field(ch, IPU_FIELD_VBO, v_offset / 8); @@ -739,48 +745,56 @@ int ipu_cpmem_set_image(struct ipuv3_channel *ch, struct ipu_image *image) switch (pix->pixelformat) { case V4L2_PIX_FMT_YUV420: offset = Y_OFFSET(pix, image->rect.left, image->rect.top); - u_offset = U_OFFSET(pix, image->rect.left, - image->rect.top) - offset; - v_offset = V_OFFSET(pix, image->rect.left, - image->rect.top) - offset; + u_offset = image->u_offset ? + image->u_offset : U_OFFSET(pix, image->rect.left, + image->rect.top) - offset; + v_offset = image->v_offset ? + image->v_offset : V_OFFSET(pix, image->rect.left, + image->rect.top) - offset; ipu_cpmem_set_yuv_planar_full(ch, pix->bytesperline / 2, u_offset, v_offset); break; case V4L2_PIX_FMT_YVU420: offset = Y_OFFSET(pix, image->rect.left, image->rect.top); - u_offset = U_OFFSET(pix, image->rect.left, - image->rect.top) - offset; - v_offset = V_OFFSET(pix, image->rect.left, - image->rect.top) - offset; + u_offset = image->u_offset ? + image->u_offset : V_OFFSET(pix, image->rect.left, + image->rect.top) - offset; + v_offset = image->v_offset ? + image->v_offset : U_OFFSET(pix, image->rect.left, + image->rect.top) - offset; ipu_cpmem_set_yuv_planar_full(ch, pix->bytesperline / 2, - v_offset, u_offset); + u_offset, v_offset); break; case V4L2_PIX_FMT_YUV422P: offset = Y_OFFSET(pix, image->rect.left, image->rect.top); - u_offset = U2_OFFSET(pix, image->rect.left, - image->rect.top) - offset; - v_offset = V2_OFFSET(pix, image->rect.left, - image->rect.top) - offset; + u_offset = image->u_offset ? + image->u_offset : U2_OFFSET(pix, image->rect.left, + image->rect.top) - offset; + v_offset = image->v_offset ? + image->v_offset : V2_OFFSET(pix, image->rect.left, + image->rect.top) - offset; ipu_cpmem_set_yuv_planar_full(ch, pix->bytesperline / 2, u_offset, v_offset); break; case V4L2_PIX_FMT_NV12: offset = Y_OFFSET(pix, image->rect.left, image->rect.top); - u_offset = UV_OFFSET(pix, image->rect.left, - image->rect.top) - offset; - v_offset = 0; + u_offset = image->u_offset ? + image->u_offset : UV_OFFSET(pix, image->rect.left, + image->rect.top) - offset; + v_offset = image->v_offset ? image->v_offset : 0; ipu_cpmem_set_yuv_planar_full(ch, pix->bytesperline, u_offset, v_offset); break; case V4L2_PIX_FMT_NV16: offset = Y_OFFSET(pix, image->rect.left, image->rect.top); - u_offset = UV2_OFFSET(pix, image->rect.left, - image->rect.top) - offset; - v_offset = 0; + u_offset = image->u_offset ? + image->u_offset : UV2_OFFSET(pix, image->rect.left, + image->rect.top) - offset; + v_offset = image->v_offset ? image->v_offset : 0; ipu_cpmem_set_yuv_planar_full(ch, pix->bytesperline, u_offset, v_offset); diff --git a/drivers/gpu/ipu-v3/ipu-ic.c b/drivers/gpu/ipu-v3/ipu-ic.c index 67cc820253a9..594c3cbc8291 100644 --- a/drivers/gpu/ipu-v3/ipu-ic.c +++ b/drivers/gpu/ipu-v3/ipu-ic.c @@ -442,36 +442,40 @@ unlock: } EXPORT_SYMBOL_GPL(ipu_ic_task_graphics_init); -int ipu_ic_task_init(struct ipu_ic *ic, - int in_width, int in_height, - int out_width, int out_height, - enum ipu_color_space in_cs, - enum ipu_color_space out_cs) +int ipu_ic_task_init_rsc(struct ipu_ic *ic, + int in_width, int in_height, + int out_width, int out_height, + enum ipu_color_space in_cs, + enum ipu_color_space out_cs, + u32 rsc) { struct ipu_ic_priv *priv = ic->priv; - u32 reg, downsize_coeff, resize_coeff; + u32 downsize_coeff, resize_coeff; unsigned long flags; int ret = 0; - /* Setup vertical resizing */ - ret = calc_resize_coeffs(ic, in_height, out_height, - &resize_coeff, &downsize_coeff); - if (ret) - return ret; + if (!rsc) { + /* Setup vertical resizing */ - reg = (downsize_coeff << 30) | (resize_coeff << 16); + ret = calc_resize_coeffs(ic, in_height, out_height, + &resize_coeff, &downsize_coeff); + if (ret) + return ret; + + rsc = (downsize_coeff << 30) | (resize_coeff << 16); - /* Setup horizontal resizing */ - ret = calc_resize_coeffs(ic, in_width, out_width, - &resize_coeff, &downsize_coeff); - if (ret) - return ret; + /* Setup horizontal resizing */ + ret = calc_resize_coeffs(ic, in_width, out_width, + &resize_coeff, &downsize_coeff); + if (ret) + return ret; - reg |= (downsize_coeff << 14) | resize_coeff; + rsc |= (downsize_coeff << 14) | resize_coeff; + } spin_lock_irqsave(&priv->lock, flags); - ipu_ic_write(ic, reg, ic->reg->rsc); + ipu_ic_write(ic, rsc, ic->reg->rsc); /* Setup color space conversion */ ic->in_cs = in_cs; @@ -487,6 +491,16 @@ unlock: spin_unlock_irqrestore(&priv->lock, flags); return ret; } + +int ipu_ic_task_init(struct ipu_ic *ic, + int in_width, int in_height, + int out_width, int out_height, + enum ipu_color_space in_cs, + enum ipu_color_space out_cs) +{ + return ipu_ic_task_init_rsc(ic, in_width, in_height, out_width, + out_height, in_cs, out_cs, 0); +} EXPORT_SYMBOL_GPL(ipu_ic_task_init); int ipu_ic_task_idma_init(struct ipu_ic *ic, struct ipuv3_channel *channel, diff --git a/drivers/gpu/ipu-v3/ipu-image-convert.c b/drivers/gpu/ipu-v3/ipu-image-convert.c index f4081962784c..13103ab86050 100644 --- a/drivers/gpu/ipu-v3/ipu-image-convert.c +++ b/drivers/gpu/ipu-v3/ipu-image-convert.c @@ -37,17 +37,36 @@ * when double_buffering boolean is set). * * Note that the input frame must be split up into the same number - * of tiles as the output frame. + * of tiles as the output frame: * - * FIXME: at this point there is no attempt to deal with visible seams - * at the tile boundaries when upscaling. The seams are caused by a reset - * of the bilinear upscale interpolation when starting a new tile. The - * seams are barely visible for small upscale factors, but become - * increasingly visible as the upscale factor gets larger, since more - * interpolated pixels get thrown out at the tile boundaries. A possilble - * fix might be to overlap tiles of different sizes, but this must be done - * while also maintaining the IDMAC dma buffer address alignment and 8x8 IRT - * alignment restrictions of each tile. + * +---------+-----+ + * +-----+---+ | A | B | + * | A | B | | | | + * +-----+---+ --> +---------+-----+ + * | C | D | | C | D | + * +-----+---+ | | | + * +---------+-----+ + * + * Clockwise 90° rotations are handled by first rescaling into a + * reusable temporary tile buffer and then rotating with the 8x8 + * block rotator, writing to the correct destination: + * + * +-----+-----+ + * | | | + * +-----+---+ +---------+ | C | A | + * | A | B | | A,B, | | | | | + * +-----+---+ --> | C,D | | --> | | | + * | C | D | +---------+ +-----+-----+ + * +-----+---+ | D | B | + * | | | + * +-----+-----+ + * + * If the 8x8 block rotator is used, horizontal or vertical flipping + * is done during the rotation step, otherwise flipping is done + * during the scaling step. + * With rotation or flipping, tile order changes between input and + * output image. Tiles are numbered row major from top left to bottom + * right for both input and output image. */ #define MAX_STRIPES_W 4 @@ -84,6 +103,8 @@ struct ipu_image_convert_dma_chan { struct ipu_image_tile { u32 width; u32 height; + u32 left; + u32 top; /* size and strides are in bytes */ u32 size; u32 stride; @@ -135,6 +156,12 @@ struct ipu_image_convert_ctx { struct ipu_image_convert_image in; struct ipu_image_convert_image out; enum ipu_rotate_mode rot_mode; + u32 downsize_coeff_h; + u32 downsize_coeff_v; + u32 image_resize_coeff_h; + u32 image_resize_coeff_v; + u32 resize_coeffs_h[MAX_STRIPES_W]; + u32 resize_coeffs_v[MAX_STRIPES_H]; /* intermediate buffer for rotation */ struct ipu_image_convert_dma_buf rot_intermediate[2]; @@ -300,12 +327,11 @@ static void dump_format(struct ipu_image_convert_ctx *ctx, struct ipu_image_convert_priv *priv = chan->priv; dev_dbg(priv->ipu->dev, - "task %u: ctx %p: %s format: %dx%d (%dx%d tiles of size %dx%d), %c%c%c%c\n", + "task %u: ctx %p: %s format: %dx%d (%dx%d tiles), %c%c%c%c\n", chan->ic_task, ctx, ic_image->type == IMAGE_CONVERT_OUT ? "Output" : "Input", ic_image->base.pix.width, ic_image->base.pix.height, ic_image->num_cols, ic_image->num_rows, - ic_image->tile[0].width, ic_image->tile[0].height, ic_image->fmt->fourcc & 0xff, (ic_image->fmt->fourcc >> 8) & 0xff, (ic_image->fmt->fourcc >> 16) & 0xff, @@ -353,24 +379,459 @@ static int alloc_dma_buf(struct ipu_image_convert_priv *priv, static inline int num_stripes(int dim) { - if (dim <= 1024) - return 1; - else if (dim <= 2048) + return (dim - 1) / 1024 + 1; +} + +/* + * Calculate downsizing coefficients, which are the same for all tiles, + * and bilinear resizing coefficients, which are used to find the best + * seam positions. + */ +static int calc_image_resize_coefficients(struct ipu_image_convert_ctx *ctx, + struct ipu_image *in, + struct ipu_image *out) +{ + u32 downsized_width = in->rect.width; + u32 downsized_height = in->rect.height; + u32 downsize_coeff_v = 0; + u32 downsize_coeff_h = 0; + u32 resized_width = out->rect.width; + u32 resized_height = out->rect.height; + u32 resize_coeff_h; + u32 resize_coeff_v; + + if (ipu_rot_mode_is_irt(ctx->rot_mode)) { + resized_width = out->rect.height; + resized_height = out->rect.width; + } + + /* Do not let invalid input lead to an endless loop below */ + if (WARN_ON(resized_width == 0 || resized_height == 0)) + return -EINVAL; + + while (downsized_width >= resized_width * 2) { + downsized_width >>= 1; + downsize_coeff_h++; + } + + while (downsized_height >= resized_height * 2) { + downsized_height >>= 1; + downsize_coeff_v++; + } + + /* + * Calculate the bilinear resizing coefficients that could be used if + * we were converting with a single tile. The bottom right output pixel + * should sample as close as possible to the bottom right input pixel + * out of the decimator, but not overshoot it: + */ + resize_coeff_h = 8192 * (downsized_width - 1) / (resized_width - 1); + resize_coeff_v = 8192 * (downsized_height - 1) / (resized_height - 1); + + dev_dbg(ctx->chan->priv->ipu->dev, + "%s: hscale: >>%u, *8192/%u vscale: >>%u, *8192/%u, %ux%u tiles\n", + __func__, downsize_coeff_h, resize_coeff_h, downsize_coeff_v, + resize_coeff_v, ctx->in.num_cols, ctx->in.num_rows); + + if (downsize_coeff_h > 2 || downsize_coeff_v > 2 || + resize_coeff_h > 0x3fff || resize_coeff_v > 0x3fff) + return -EINVAL; + + ctx->downsize_coeff_h = downsize_coeff_h; + ctx->downsize_coeff_v = downsize_coeff_v; + ctx->image_resize_coeff_h = resize_coeff_h; + ctx->image_resize_coeff_v = resize_coeff_v; + + return 0; +} + +#define round_closest(x, y) round_down((x) + (y)/2, (y)) + +/* + * Find the best aligned seam position in the inverval [out_start, out_end]. + * Rotation and image offsets are out of scope. + * + * @out_start: start of inverval, must be within 1024 pixels / lines + * of out_end + * @out_end: end of interval, smaller than or equal to out_edge + * @in_edge: input right / bottom edge + * @out_edge: output right / bottom edge + * @in_align: input alignment, either horizontal 8-byte line start address + * alignment, or pixel alignment due to image format + * @out_align: output alignment, either horizontal 8-byte line start address + * alignment, or pixel alignment due to image format or rotator + * block size + * @in_burst: horizontal input burst size in case of horizontal flip + * @out_burst: horizontal output burst size or rotator block size + * @downsize_coeff: downsizing section coefficient + * @resize_coeff: main processing section resizing coefficient + * @_in_seam: aligned input seam position return value + * @_out_seam: aligned output seam position return value + */ +static void find_best_seam(struct ipu_image_convert_ctx *ctx, + unsigned int out_start, + unsigned int out_end, + unsigned int in_edge, + unsigned int out_edge, + unsigned int in_align, + unsigned int out_align, + unsigned int in_burst, + unsigned int out_burst, + unsigned int downsize_coeff, + unsigned int resize_coeff, + u32 *_in_seam, + u32 *_out_seam) +{ + struct device *dev = ctx->chan->priv->ipu->dev; + unsigned int out_pos; + /* Input / output seam position candidates */ + unsigned int out_seam = 0; + unsigned int in_seam = 0; + unsigned int min_diff = UINT_MAX; + + /* + * Output tiles must start at a multiple of 8 bytes horizontally and + * possibly at an even line horizontally depending on the pixel format. + * Only consider output aligned positions for the seam. + */ + out_start = round_up(out_start, out_align); + for (out_pos = out_start; out_pos < out_end; out_pos += out_align) { + unsigned int in_pos; + unsigned int in_pos_aligned; + unsigned int abs_diff; + + /* + * Tiles in the right row / bottom column may not be allowed to + * overshoot horizontally / vertically. out_burst may be the + * actual DMA burst size, or the rotator block size. + */ + if ((out_burst > 1) && (out_edge - out_pos) % out_burst) + continue; + + /* + * Input sample position, corresponding to out_pos, 19.13 fixed + * point. + */ + in_pos = (out_pos * resize_coeff) << downsize_coeff; + /* + * The closest input sample position that we could actually + * start the input tile at, 19.13 fixed point. + */ + in_pos_aligned = round_closest(in_pos, 8192U * in_align); + + if ((in_burst > 1) && + (in_edge - in_pos_aligned / 8192U) % in_burst) + continue; + + if (in_pos < in_pos_aligned) + abs_diff = in_pos_aligned - in_pos; + else + abs_diff = in_pos - in_pos_aligned; + + if (abs_diff < min_diff) { + in_seam = in_pos_aligned; + out_seam = out_pos; + min_diff = abs_diff; + } + } + + *_out_seam = out_seam; + /* Convert 19.13 fixed point to integer seam position */ + *_in_seam = DIV_ROUND_CLOSEST(in_seam, 8192U); + + dev_dbg(dev, "%s: out_seam %u(%u) in [%u, %u], in_seam %u(%u) diff %u.%03u\n", + __func__, out_seam, out_align, out_start, out_end, + *_in_seam, in_align, min_diff / 8192, + DIV_ROUND_CLOSEST(min_diff % 8192 * 1000, 8192)); +} + +/* + * Tile left edges are required to be aligned to multiples of 8 bytes + * by the IDMAC. + */ +static inline u32 tile_left_align(const struct ipu_image_pixfmt *fmt) +{ + if (fmt->planar) + return fmt->uv_packed ? 8 : 8 * fmt->uv_width_dec; + else + return fmt->bpp == 32 ? 2 : fmt->bpp == 16 ? 4 : 8; +} + +/* + * Tile top edge alignment is only limited by chroma subsampling. + */ +static inline u32 tile_top_align(const struct ipu_image_pixfmt *fmt) +{ + return fmt->uv_height_dec > 1 ? 2 : 1; +} + +static inline u32 tile_width_align(enum ipu_image_convert_type type, + const struct ipu_image_pixfmt *fmt, + enum ipu_rotate_mode rot_mode) +{ + if (type == IMAGE_CONVERT_IN) { + /* + * The IC burst reads 8 pixels at a time. Reading beyond the + * end of the line is usually acceptable. Those pixels are + * ignored, unless the IC has to write the scaled line in + * reverse. + */ + return (!ipu_rot_mode_is_irt(rot_mode) && + (rot_mode & IPU_ROT_BIT_HFLIP)) ? 8 : 2; + } + + /* + * Align to 16x16 pixel blocks for planar 4:2:0 chroma subsampled + * formats to guarantee 8-byte aligned line start addresses in the + * chroma planes when IRT is used. Align to 8x8 pixel IRT block size + * for all other formats. + */ + return (ipu_rot_mode_is_irt(rot_mode) && + fmt->planar && !fmt->uv_packed) ? + 8 * fmt->uv_width_dec : 8; +} + +static inline u32 tile_height_align(enum ipu_image_convert_type type, + const struct ipu_image_pixfmt *fmt, + enum ipu_rotate_mode rot_mode) +{ + if (type == IMAGE_CONVERT_IN || !ipu_rot_mode_is_irt(rot_mode)) return 2; + + /* + * Align to 16x16 pixel blocks for planar 4:2:0 chroma subsampled + * formats to guarantee 8-byte aligned line start addresses in the + * chroma planes when IRT is used. Align to 8x8 pixel IRT block size + * for all other formats. + */ + return (fmt->planar && !fmt->uv_packed) ? 8 * fmt->uv_width_dec : 8; +} + +/* + * Fill in left position and width and for all tiles in an input column, and + * for all corresponding output tiles. If the 90° rotator is used, the output + * tiles are in a row, and output tile top position and height are set. + */ +static void fill_tile_column(struct ipu_image_convert_ctx *ctx, + unsigned int col, + struct ipu_image_convert_image *in, + unsigned int in_left, unsigned int in_width, + struct ipu_image_convert_image *out, + unsigned int out_left, unsigned int out_width) +{ + unsigned int row, tile_idx; + struct ipu_image_tile *in_tile, *out_tile; + + for (row = 0; row < in->num_rows; row++) { + tile_idx = in->num_cols * row + col; + in_tile = &in->tile[tile_idx]; + out_tile = &out->tile[ctx->out_tile_map[tile_idx]]; + + in_tile->left = in_left; + in_tile->width = in_width; + + if (ipu_rot_mode_is_irt(ctx->rot_mode)) { + out_tile->top = out_left; + out_tile->height = out_width; + } else { + out_tile->left = out_left; + out_tile->width = out_width; + } + } +} + +/* + * Fill in top position and height and for all tiles in an input row, and + * for all corresponding output tiles. If the 90° rotator is used, the output + * tiles are in a column, and output tile left position and width are set. + */ +static void fill_tile_row(struct ipu_image_convert_ctx *ctx, unsigned int row, + struct ipu_image_convert_image *in, + unsigned int in_top, unsigned int in_height, + struct ipu_image_convert_image *out, + unsigned int out_top, unsigned int out_height) +{ + unsigned int col, tile_idx; + struct ipu_image_tile *in_tile, *out_tile; + + for (col = 0; col < in->num_cols; col++) { + tile_idx = in->num_cols * row + col; + in_tile = &in->tile[tile_idx]; + out_tile = &out->tile[ctx->out_tile_map[tile_idx]]; + + in_tile->top = in_top; + in_tile->height = in_height; + + if (ipu_rot_mode_is_irt(ctx->rot_mode)) { + out_tile->left = out_top; + out_tile->width = out_height; + } else { + out_tile->top = out_top; + out_tile->height = out_height; + } + } +} + +/* + * Find the best horizontal and vertical seam positions to split into tiles. + * Minimize the fractional part of the input sampling position for the + * top / left pixels of each tile. + */ +static void find_seams(struct ipu_image_convert_ctx *ctx, + struct ipu_image_convert_image *in, + struct ipu_image_convert_image *out) +{ + struct device *dev = ctx->chan->priv->ipu->dev; + unsigned int resized_width = out->base.rect.width; + unsigned int resized_height = out->base.rect.height; + unsigned int col; + unsigned int row; + unsigned int in_left_align = tile_left_align(in->fmt); + unsigned int in_top_align = tile_top_align(in->fmt); + unsigned int out_left_align = tile_left_align(out->fmt); + unsigned int out_top_align = tile_top_align(out->fmt); + unsigned int out_width_align = tile_width_align(out->type, out->fmt, + ctx->rot_mode); + unsigned int out_height_align = tile_height_align(out->type, out->fmt, + ctx->rot_mode); + unsigned int in_right = in->base.rect.width; + unsigned int in_bottom = in->base.rect.height; + unsigned int out_right = out->base.rect.width; + unsigned int out_bottom = out->base.rect.height; + unsigned int flipped_out_left; + unsigned int flipped_out_top; + + if (ipu_rot_mode_is_irt(ctx->rot_mode)) { + /* Switch width/height and align top left to IRT block size */ + resized_width = out->base.rect.height; + resized_height = out->base.rect.width; + out_left_align = out_height_align; + out_top_align = out_width_align; + out_width_align = out_left_align; + out_height_align = out_top_align; + out_right = out->base.rect.height; + out_bottom = out->base.rect.width; + } + + for (col = in->num_cols - 1; col > 0; col--) { + bool allow_in_overshoot = ipu_rot_mode_is_irt(ctx->rot_mode) || + !(ctx->rot_mode & IPU_ROT_BIT_HFLIP); + bool allow_out_overshoot = (col < in->num_cols - 1) && + !(ctx->rot_mode & IPU_ROT_BIT_HFLIP); + unsigned int out_start; + unsigned int out_end; + unsigned int in_left; + unsigned int out_left; + + /* + * Align input width to burst length if the scaling step flips + * horizontally. + */ + + /* Start within 1024 pixels of the right edge */ + out_start = max_t(int, 0, out_right - 1024); + /* End before having to add more columns to the left */ + out_end = min_t(unsigned int, out_right, col * 1024); + + find_best_seam(ctx, out_start, out_end, + in_right, out_right, + in_left_align, out_left_align, + allow_in_overshoot ? 1 : 8 /* burst length */, + allow_out_overshoot ? 1 : out_width_align, + ctx->downsize_coeff_h, ctx->image_resize_coeff_h, + &in_left, &out_left); + + if (ctx->rot_mode & IPU_ROT_BIT_HFLIP) + flipped_out_left = resized_width - out_right; + else + flipped_out_left = out_left; + + fill_tile_column(ctx, col, in, in_left, in_right - in_left, + out, flipped_out_left, out_right - out_left); + + dev_dbg(dev, "%s: col %u: %u, %u -> %u, %u\n", __func__, col, + in_left, in_right - in_left, + flipped_out_left, out_right - out_left); + + in_right = in_left; + out_right = out_left; + } + + flipped_out_left = (ctx->rot_mode & IPU_ROT_BIT_HFLIP) ? + resized_width - out_right : 0; + + fill_tile_column(ctx, 0, in, 0, in_right, + out, flipped_out_left, out_right); + + dev_dbg(dev, "%s: col 0: 0, %u -> %u, %u\n", __func__, + in_right, flipped_out_left, out_right); + + for (row = in->num_rows - 1; row > 0; row--) { + bool allow_overshoot = row < in->num_rows - 1; + unsigned int out_start; + unsigned int out_end; + unsigned int in_top; + unsigned int out_top; + + /* Start within 1024 lines of the bottom edge */ + out_start = max_t(int, 0, out_bottom - 1024); + /* End before having to add more rows above */ + out_end = min_t(unsigned int, out_bottom, row * 1024); + + find_best_seam(ctx, out_start, out_end, + in_bottom, out_bottom, + in_top_align, out_top_align, + 1, allow_overshoot ? 1 : out_height_align, + ctx->downsize_coeff_v, ctx->image_resize_coeff_v, + &in_top, &out_top); + + if ((ctx->rot_mode & IPU_ROT_BIT_VFLIP) ^ + ipu_rot_mode_is_irt(ctx->rot_mode)) + flipped_out_top = resized_height - out_bottom; + else + flipped_out_top = out_top; + + fill_tile_row(ctx, row, in, in_top, in_bottom - in_top, + out, flipped_out_top, out_bottom - out_top); + + dev_dbg(dev, "%s: row %u: %u, %u -> %u, %u\n", __func__, row, + in_top, in_bottom - in_top, + flipped_out_top, out_bottom - out_top); + + in_bottom = in_top; + out_bottom = out_top; + } + + if ((ctx->rot_mode & IPU_ROT_BIT_VFLIP) ^ + ipu_rot_mode_is_irt(ctx->rot_mode)) + flipped_out_top = resized_height - out_bottom; else - return 4; + flipped_out_top = 0; + + fill_tile_row(ctx, 0, in, 0, in_bottom, + out, flipped_out_top, out_bottom); + + dev_dbg(dev, "%s: row 0: 0, %u -> %u, %u\n", __func__, + in_bottom, flipped_out_top, out_bottom); } static void calc_tile_dimensions(struct ipu_image_convert_ctx *ctx, struct ipu_image_convert_image *image) { - int i; + struct ipu_image_convert_chan *chan = ctx->chan; + struct ipu_image_convert_priv *priv = chan->priv; + unsigned int i; for (i = 0; i < ctx->num_tiles; i++) { - struct ipu_image_tile *tile = &image->tile[i]; + struct ipu_image_tile *tile; + const unsigned int row = i / image->num_cols; + const unsigned int col = i % image->num_cols; + + if (image->type == IMAGE_CONVERT_OUT) + tile = &image->tile[ctx->out_tile_map[i]]; + else + tile = &image->tile[i]; - tile->height = image->base.pix.height / image->num_rows; - tile->width = image->base.pix.width / image->num_cols; tile->size = ((tile->height * image->fmt->bpp) >> 3) * tile->width; @@ -383,6 +844,13 @@ static void calc_tile_dimensions(struct ipu_image_convert_ctx *ctx, tile->rot_stride = (image->fmt->bpp * tile->height) >> 3; } + + dev_dbg(priv->ipu->dev, + "task %u: ctx %p: %s@[%u,%u]: %ux%u@%u,%u\n", + chan->ic_task, ctx, + image->type == IMAGE_CONVERT_IN ? "Input" : "Output", + row, col, + tile->width, tile->height, tile->left, tile->top); } } @@ -459,14 +927,14 @@ static void calc_out_tile_map(struct ipu_image_convert_ctx *ctx) } } -static void calc_tile_offsets_planar(struct ipu_image_convert_ctx *ctx, - struct ipu_image_convert_image *image) +static int calc_tile_offsets_planar(struct ipu_image_convert_ctx *ctx, + struct ipu_image_convert_image *image) { struct ipu_image_convert_chan *chan = ctx->chan; struct ipu_image_convert_priv *priv = chan->priv; const struct ipu_image_pixfmt *fmt = image->fmt; unsigned int row, col, tile = 0; - u32 H, w, h, y_stride, uv_stride; + u32 H, top, y_stride, uv_stride; u32 uv_row_off, uv_col_off, uv_off, u_off, v_off, tmp; u32 y_row_off, y_col_off, y_off; u32 y_size, uv_size; @@ -483,13 +951,12 @@ static void calc_tile_offsets_planar(struct ipu_image_convert_ctx *ctx, uv_size = y_size / (fmt->uv_width_dec * fmt->uv_height_dec); for (row = 0; row < image->num_rows; row++) { - w = image->tile[tile].width; - h = image->tile[tile].height; - y_row_off = row * h * y_stride; - uv_row_off = (row * h * uv_stride) / fmt->uv_height_dec; + top = image->tile[tile].top; + y_row_off = top * y_stride; + uv_row_off = (top * uv_stride) / fmt->uv_height_dec; for (col = 0; col < image->num_cols; col++) { - y_col_off = col * w; + y_col_off = image->tile[tile].left; uv_col_off = y_col_off / fmt->uv_width_dec; if (fmt->uv_packed) uv_col_off *= 2; @@ -509,24 +976,30 @@ static void calc_tile_offsets_planar(struct ipu_image_convert_ctx *ctx, image->tile[tile].u_off = u_off; image->tile[tile++].v_off = v_off; - dev_dbg(priv->ipu->dev, - "task %u: ctx %p: %s@[%d,%d]: y_off %08x, u_off %08x, v_off %08x\n", - chan->ic_task, ctx, - image->type == IMAGE_CONVERT_IN ? - "Input" : "Output", row, col, - y_off, u_off, v_off); + if ((y_off & 0x7) || (u_off & 0x7) || (v_off & 0x7)) { + dev_err(priv->ipu->dev, + "task %u: ctx %p: %s@[%d,%d]: " + "y_off %08x, u_off %08x, v_off %08x\n", + chan->ic_task, ctx, + image->type == IMAGE_CONVERT_IN ? + "Input" : "Output", row, col, + y_off, u_off, v_off); + return -EINVAL; + } } } + + return 0; } -static void calc_tile_offsets_packed(struct ipu_image_convert_ctx *ctx, - struct ipu_image_convert_image *image) +static int calc_tile_offsets_packed(struct ipu_image_convert_ctx *ctx, + struct ipu_image_convert_image *image) { struct ipu_image_convert_chan *chan = ctx->chan; struct ipu_image_convert_priv *priv = chan->priv; const struct ipu_image_pixfmt *fmt = image->fmt; unsigned int row, col, tile = 0; - u32 w, h, bpp, stride; + u32 bpp, stride, offset; u32 row_off, col_off; /* setup some convenience vars */ @@ -534,34 +1007,183 @@ static void calc_tile_offsets_packed(struct ipu_image_convert_ctx *ctx, bpp = fmt->bpp; for (row = 0; row < image->num_rows; row++) { - w = image->tile[tile].width; - h = image->tile[tile].height; - row_off = row * h * stride; + row_off = image->tile[tile].top * stride; for (col = 0; col < image->num_cols; col++) { - col_off = (col * w * bpp) >> 3; + col_off = (image->tile[tile].left * bpp) >> 3; + + offset = row_off + col_off; - image->tile[tile].offset = row_off + col_off; + image->tile[tile].offset = offset; image->tile[tile].u_off = 0; image->tile[tile++].v_off = 0; - dev_dbg(priv->ipu->dev, - "task %u: ctx %p: %s@[%d,%d]: phys %08x\n", - chan->ic_task, ctx, - image->type == IMAGE_CONVERT_IN ? - "Input" : "Output", row, col, - row_off + col_off); + if (offset & 0x7) { + dev_err(priv->ipu->dev, + "task %u: ctx %p: %s@[%d,%d]: " + "phys %08x\n", + chan->ic_task, ctx, + image->type == IMAGE_CONVERT_IN ? + "Input" : "Output", row, col, + row_off + col_off); + return -EINVAL; + } } } + + return 0; } -static void calc_tile_offsets(struct ipu_image_convert_ctx *ctx, +static int calc_tile_offsets(struct ipu_image_convert_ctx *ctx, struct ipu_image_convert_image *image) { if (image->fmt->planar) - calc_tile_offsets_planar(ctx, image); + return calc_tile_offsets_planar(ctx, image); + + return calc_tile_offsets_packed(ctx, image); +} + +/* + * Calculate the resizing ratio for the IC main processing section given input + * size, fixed downsizing coefficient, and output size. + * Either round to closest for the next tile's first pixel to minimize seams + * and distortion (for all but right column / bottom row), or round down to + * avoid sampling beyond the edges of the input image for this tile's last + * pixel. + * Returns the resizing coefficient, resizing ratio is 8192.0 / resize_coeff. + */ +static u32 calc_resize_coeff(u32 input_size, u32 downsize_coeff, + u32 output_size, bool allow_overshoot) +{ + u32 downsized = input_size >> downsize_coeff; + + if (allow_overshoot) + return DIV_ROUND_CLOSEST(8192 * downsized, output_size); else - calc_tile_offsets_packed(ctx, image); + return 8192 * (downsized - 1) / (output_size - 1); +} + +/* + * Slightly modify resize coefficients per tile to hide the bilinear + * interpolator reset at tile borders, shifting the right / bottom edge + * by up to a half input pixel. This removes noticeable seams between + * tiles at higher upscaling factors. + */ +static void calc_tile_resize_coefficients(struct ipu_image_convert_ctx *ctx) +{ + struct ipu_image_convert_chan *chan = ctx->chan; + struct ipu_image_convert_priv *priv = chan->priv; + struct ipu_image_tile *in_tile, *out_tile; + unsigned int col, row, tile_idx; + unsigned int last_output; + + for (col = 0; col < ctx->in.num_cols; col++) { + bool closest = (col < ctx->in.num_cols - 1) && + !(ctx->rot_mode & IPU_ROT_BIT_HFLIP); + u32 resized_width; + u32 resize_coeff_h; + + tile_idx = col; + in_tile = &ctx->in.tile[tile_idx]; + out_tile = &ctx->out.tile[ctx->out_tile_map[tile_idx]]; + + if (ipu_rot_mode_is_irt(ctx->rot_mode)) + resized_width = out_tile->height; + else + resized_width = out_tile->width; + + resize_coeff_h = calc_resize_coeff(in_tile->width, + ctx->downsize_coeff_h, + resized_width, closest); + + dev_dbg(priv->ipu->dev, "%s: column %u hscale: *8192/%u\n", + __func__, col, resize_coeff_h); + + + for (row = 0; row < ctx->in.num_rows; row++) { + tile_idx = row * ctx->in.num_cols + col; + in_tile = &ctx->in.tile[tile_idx]; + out_tile = &ctx->out.tile[ctx->out_tile_map[tile_idx]]; + + /* + * With the horizontal scaling factor known, round up + * resized width (output width or height) to burst size. + */ + if (ipu_rot_mode_is_irt(ctx->rot_mode)) + out_tile->height = round_up(resized_width, 8); + else + out_tile->width = round_up(resized_width, 8); + + /* + * Calculate input width from the last accessed input + * pixel given resized width and scaling coefficients. + * Round up to burst size. + */ + last_output = round_up(resized_width, 8) - 1; + if (closest) + last_output++; + in_tile->width = round_up( + (DIV_ROUND_UP(last_output * resize_coeff_h, + 8192) + 1) + << ctx->downsize_coeff_h, 8); + } + + ctx->resize_coeffs_h[col] = resize_coeff_h; + } + + for (row = 0; row < ctx->in.num_rows; row++) { + bool closest = (row < ctx->in.num_rows - 1) && + !(ctx->rot_mode & IPU_ROT_BIT_VFLIP); + u32 resized_height; + u32 resize_coeff_v; + + tile_idx = row * ctx->in.num_cols; + in_tile = &ctx->in.tile[tile_idx]; + out_tile = &ctx->out.tile[ctx->out_tile_map[tile_idx]]; + + if (ipu_rot_mode_is_irt(ctx->rot_mode)) + resized_height = out_tile->width; + else + resized_height = out_tile->height; + + resize_coeff_v = calc_resize_coeff(in_tile->height, + ctx->downsize_coeff_v, + resized_height, closest); + + dev_dbg(priv->ipu->dev, "%s: row %u vscale: *8192/%u\n", + __func__, row, resize_coeff_v); + + for (col = 0; col < ctx->in.num_cols; col++) { + tile_idx = row * ctx->in.num_cols + col; + in_tile = &ctx->in.tile[tile_idx]; + out_tile = &ctx->out.tile[ctx->out_tile_map[tile_idx]]; + + /* + * With the vertical scaling factor known, round up + * resized height (output width or height) to IDMAC + * limitations. + */ + if (ipu_rot_mode_is_irt(ctx->rot_mode)) + out_tile->width = round_up(resized_height, 2); + else + out_tile->height = round_up(resized_height, 2); + + /* + * Calculate input width from the last accessed input + * pixel given resized height and scaling coefficients. + * Align to IDMAC restrictions. + */ + last_output = round_up(resized_height, 2) - 1; + if (closest) + last_output++; + in_tile->height = round_up( + (DIV_ROUND_UP(last_output * resize_coeff_v, + 8192) + 1) + << ctx->downsize_coeff_v, 2); + } + + ctx->resize_coeffs_v[row] = resize_coeff_v; + } } /* @@ -611,7 +1233,8 @@ static void init_idmac_channel(struct ipu_image_convert_ctx *ctx, struct ipuv3_channel *channel, struct ipu_image_convert_image *image, enum ipu_rotate_mode rot_mode, - bool rot_swap_width_height) + bool rot_swap_width_height, + unsigned int tile) { struct ipu_image_convert_chan *chan = ctx->chan; unsigned int burst_size; @@ -621,23 +1244,23 @@ static void init_idmac_channel(struct ipu_image_convert_ctx *ctx, unsigned int tile_idx[2]; if (image->type == IMAGE_CONVERT_OUT) { - tile_idx[0] = ctx->out_tile_map[0]; + tile_idx[0] = ctx->out_tile_map[tile]; tile_idx[1] = ctx->out_tile_map[1]; } else { - tile_idx[0] = 0; + tile_idx[0] = tile; tile_idx[1] = 1; } if (rot_swap_width_height) { - width = image->tile[0].height; - height = image->tile[0].width; - stride = image->tile[0].rot_stride; + width = image->tile[tile_idx[0]].height; + height = image->tile[tile_idx[0]].width; + stride = image->tile[tile_idx[0]].rot_stride; addr0 = ctx->rot_intermediate[0].phys; if (ctx->double_buffering) addr1 = ctx->rot_intermediate[1].phys; } else { - width = image->tile[0].width; - height = image->tile[0].height; + width = image->tile[tile_idx[0]].width; + height = image->tile[tile_idx[0]].height; stride = image->stride; addr0 = image->base.phys0 + image->tile[tile_idx[0]].offset; @@ -655,12 +1278,12 @@ static void init_idmac_channel(struct ipu_image_convert_ctx *ctx, tile_image.pix.pixelformat = image->fmt->fourcc; tile_image.phys0 = addr0; tile_image.phys1 = addr1; - ipu_cpmem_set_image(channel, &tile_image); + if (image->fmt->planar && !rot_swap_width_height) { + tile_image.u_offset = image->tile[tile_idx[0]].u_off; + tile_image.v_offset = image->tile[tile_idx[0]].v_off; + } - if (image->fmt->planar && !rot_swap_width_height) - ipu_cpmem_set_uv_offset(channel, - image->tile[tile_idx[0]].u_off, - image->tile[tile_idx[0]].v_off); + ipu_cpmem_set_image(channel, &tile_image); if (rot_mode) ipu_cpmem_set_rotation(channel, rot_mode); @@ -687,7 +1310,7 @@ static void init_idmac_channel(struct ipu_image_convert_ctx *ctx, ipu_idmac_set_double_buffer(channel, ctx->double_buffering); } -static int convert_start(struct ipu_image_convert_run *run) +static int convert_start(struct ipu_image_convert_run *run, unsigned int tile) { struct ipu_image_convert_ctx *ctx = run->ctx; struct ipu_image_convert_chan *chan = ctx->chan; @@ -695,31 +1318,47 @@ static int convert_start(struct ipu_image_convert_run *run) struct ipu_image_convert_image *s_image = &ctx->in; struct ipu_image_convert_image *d_image = &ctx->out; enum ipu_color_space src_cs, dest_cs; + unsigned int dst_tile = ctx->out_tile_map[tile]; unsigned int dest_width, dest_height; + unsigned int col, row; + u32 rsc; int ret; - dev_dbg(priv->ipu->dev, "%s: task %u: starting ctx %p run %p\n", - __func__, chan->ic_task, ctx, run); + dev_dbg(priv->ipu->dev, "%s: task %u: starting ctx %p run %p tile %u -> %u\n", + __func__, chan->ic_task, ctx, run, tile, dst_tile); src_cs = ipu_pixelformat_to_colorspace(s_image->fmt->fourcc); dest_cs = ipu_pixelformat_to_colorspace(d_image->fmt->fourcc); if (ipu_rot_mode_is_irt(ctx->rot_mode)) { /* swap width/height for resizer */ - dest_width = d_image->tile[0].height; - dest_height = d_image->tile[0].width; + dest_width = d_image->tile[dst_tile].height; + dest_height = d_image->tile[dst_tile].width; } else { - dest_width = d_image->tile[0].width; - dest_height = d_image->tile[0].height; + dest_width = d_image->tile[dst_tile].width; + dest_height = d_image->tile[dst_tile].height; } + row = tile / s_image->num_cols; + col = tile % s_image->num_cols; + + rsc = (ctx->downsize_coeff_v << 30) | + (ctx->resize_coeffs_v[row] << 16) | + (ctx->downsize_coeff_h << 14) | + (ctx->resize_coeffs_h[col]); + + dev_dbg(priv->ipu->dev, "%s: %ux%u -> %ux%u (rsc = 0x%x)\n", + __func__, s_image->tile[tile].width, + s_image->tile[tile].height, dest_width, dest_height, rsc); + /* setup the IC resizer and CSC */ - ret = ipu_ic_task_init(chan->ic, - s_image->tile[0].width, - s_image->tile[0].height, + ret = ipu_ic_task_init_rsc(chan->ic, + s_image->tile[tile].width, + s_image->tile[tile].height, dest_width, dest_height, - src_cs, dest_cs); + src_cs, dest_cs, + rsc); if (ret) { dev_err(priv->ipu->dev, "ipu_ic_task_init failed, %d\n", ret); return ret; @@ -727,27 +1366,27 @@ static int convert_start(struct ipu_image_convert_run *run) /* init the source MEM-->IC PP IDMAC channel */ init_idmac_channel(ctx, chan->in_chan, s_image, - IPU_ROTATE_NONE, false); + IPU_ROTATE_NONE, false, tile); if (ipu_rot_mode_is_irt(ctx->rot_mode)) { /* init the IC PP-->MEM IDMAC channel */ init_idmac_channel(ctx, chan->out_chan, d_image, - IPU_ROTATE_NONE, true); + IPU_ROTATE_NONE, true, tile); /* init the MEM-->IC PP ROT IDMAC channel */ init_idmac_channel(ctx, chan->rotation_in_chan, d_image, - ctx->rot_mode, true); + ctx->rot_mode, true, tile); /* init the destination IC PP ROT-->MEM IDMAC channel */ init_idmac_channel(ctx, chan->rotation_out_chan, d_image, - IPU_ROTATE_NONE, false); + IPU_ROTATE_NONE, false, tile); /* now link IC PP-->MEM to MEM-->IC PP ROT */ ipu_idmac_link(chan->out_chan, chan->rotation_in_chan); } else { /* init the destination IC PP-->MEM IDMAC channel */ init_idmac_channel(ctx, chan->out_chan, d_image, - ctx->rot_mode, false); + ctx->rot_mode, false, tile); } /* enable the IC */ @@ -805,7 +1444,7 @@ static int do_run(struct ipu_image_convert_run *run) list_del(&run->list); chan->current_run = run; - return convert_start(run); + return convert_start(run, 0); } /* hold irqlock when calling */ @@ -896,7 +1535,7 @@ static irqreturn_t do_bh(int irq, void *dev_id) dev_dbg(priv->ipu->dev, "%s: task %u: signaling abort for ctx %p\n", __func__, chan->ic_task, ctx); - complete(&ctx->aborted); + complete_all(&ctx->aborted); } } @@ -908,6 +1547,24 @@ static irqreturn_t do_bh(int irq, void *dev_id) return IRQ_HANDLED; } +static bool ic_settings_changed(struct ipu_image_convert_ctx *ctx) +{ + unsigned int cur_tile = ctx->next_tile - 1; + unsigned int next_tile = ctx->next_tile; + + if (ctx->resize_coeffs_h[cur_tile % ctx->in.num_cols] != + ctx->resize_coeffs_h[next_tile % ctx->in.num_cols] || + ctx->resize_coeffs_v[cur_tile / ctx->in.num_cols] != + ctx->resize_coeffs_v[next_tile / ctx->in.num_cols] || + ctx->in.tile[cur_tile].width != ctx->in.tile[next_tile].width || + ctx->in.tile[cur_tile].height != ctx->in.tile[next_tile].height || + ctx->out.tile[cur_tile].width != ctx->out.tile[next_tile].width || + ctx->out.tile[cur_tile].height != ctx->out.tile[next_tile].height) + return true; + + return false; +} + /* hold irqlock when calling */ static irqreturn_t do_irq(struct ipu_image_convert_run *run) { @@ -951,27 +1608,32 @@ static irqreturn_t do_irq(struct ipu_image_convert_run *run) * not done, place the next tile buffers. */ if (!ctx->double_buffering) { - - src_tile = &s_image->tile[ctx->next_tile]; - dst_idx = ctx->out_tile_map[ctx->next_tile]; - dst_tile = &d_image->tile[dst_idx]; - - ipu_cpmem_set_buffer(chan->in_chan, 0, - s_image->base.phys0 + src_tile->offset); - ipu_cpmem_set_buffer(outch, 0, - d_image->base.phys0 + dst_tile->offset); - if (s_image->fmt->planar) - ipu_cpmem_set_uv_offset(chan->in_chan, - src_tile->u_off, - src_tile->v_off); - if (d_image->fmt->planar) - ipu_cpmem_set_uv_offset(outch, - dst_tile->u_off, - dst_tile->v_off); - - ipu_idmac_select_buffer(chan->in_chan, 0); - ipu_idmac_select_buffer(outch, 0); - + if (ic_settings_changed(ctx)) { + convert_stop(run); + convert_start(run, ctx->next_tile); + } else { + src_tile = &s_image->tile[ctx->next_tile]; + dst_idx = ctx->out_tile_map[ctx->next_tile]; + dst_tile = &d_image->tile[dst_idx]; + + ipu_cpmem_set_buffer(chan->in_chan, 0, + s_image->base.phys0 + + src_tile->offset); + ipu_cpmem_set_buffer(outch, 0, + d_image->base.phys0 + + dst_tile->offset); + if (s_image->fmt->planar) + ipu_cpmem_set_uv_offset(chan->in_chan, + src_tile->u_off, + src_tile->v_off); + if (d_image->fmt->planar) + ipu_cpmem_set_uv_offset(outch, + dst_tile->u_off, + dst_tile->v_off); + + ipu_idmac_select_buffer(chan->in_chan, 0); + ipu_idmac_select_buffer(outch, 0); + } } else if (ctx->next_tile < ctx->num_tiles - 1) { src_tile = &s_image->tile[ctx->next_tile + 1]; @@ -1198,9 +1860,6 @@ static int fill_image(struct ipu_image_convert_ctx *ctx, else ic_image->stride = ic_image->base.pix.bytesperline; - calc_tile_dimensions(ctx, ic_image); - calc_tile_offsets(ctx, ic_image); - return 0; } @@ -1221,40 +1880,11 @@ static unsigned int clamp_align(unsigned int x, unsigned int min, return x; } -/* - * We have to adjust the tile width such that the tile physaddrs and - * U and V plane offsets are multiples of 8 bytes as required by - * the IPU DMA Controller. For the planar formats, this corresponds - * to a pixel alignment of 16 (but use a more formal equation since - * the variables are available). For all the packed formats, 8 is - * good enough. - */ -static inline u32 tile_width_align(const struct ipu_image_pixfmt *fmt) -{ - return fmt->planar ? 8 * fmt->uv_width_dec : 8; -} - -/* - * For tile height alignment, we have to ensure that the output tile - * heights are multiples of 8 lines if the IRT is required by the - * given rotation mode (the IRT performs rotations on 8x8 blocks - * at a time). If the IRT is not used, or for input image tiles, - * 2 lines are good enough. - */ -static inline u32 tile_height_align(enum ipu_image_convert_type type, - enum ipu_rotate_mode rot_mode) -{ - return (type == IMAGE_CONVERT_OUT && - ipu_rot_mode_is_irt(rot_mode)) ? 8 : 2; -} - /* Adjusts input/output images to IPU restrictions */ void ipu_image_convert_adjust(struct ipu_image *in, struct ipu_image *out, enum ipu_rotate_mode rot_mode) { const struct ipu_image_pixfmt *infmt, *outfmt; - unsigned int num_in_rows, num_in_cols; - unsigned int num_out_rows, num_out_cols; u32 w_align, h_align; infmt = get_format(in->pix.pixelformat); @@ -1286,36 +1916,31 @@ void ipu_image_convert_adjust(struct ipu_image *in, struct ipu_image *out, in->pix.height / 4); } - /* get tiling rows/cols from output format */ - num_out_rows = num_stripes(out->pix.height); - num_out_cols = num_stripes(out->pix.width); - if (ipu_rot_mode_is_irt(rot_mode)) { - num_in_rows = num_out_cols; - num_in_cols = num_out_rows; - } else { - num_in_rows = num_out_rows; - num_in_cols = num_out_cols; - } - /* align input width/height */ - w_align = ilog2(tile_width_align(infmt) * num_in_cols); - h_align = ilog2(tile_height_align(IMAGE_CONVERT_IN, rot_mode) * - num_in_rows); + w_align = ilog2(tile_width_align(IMAGE_CONVERT_IN, infmt, rot_mode)); + h_align = ilog2(tile_height_align(IMAGE_CONVERT_IN, infmt, rot_mode)); in->pix.width = clamp_align(in->pix.width, MIN_W, MAX_W, w_align); in->pix.height = clamp_align(in->pix.height, MIN_H, MAX_H, h_align); /* align output width/height */ - w_align = ilog2(tile_width_align(outfmt) * num_out_cols); - h_align = ilog2(tile_height_align(IMAGE_CONVERT_OUT, rot_mode) * - num_out_rows); + w_align = ilog2(tile_width_align(IMAGE_CONVERT_OUT, outfmt, rot_mode)); + h_align = ilog2(tile_height_align(IMAGE_CONVERT_OUT, outfmt, rot_mode)); out->pix.width = clamp_align(out->pix.width, MIN_W, MAX_W, w_align); out->pix.height = clamp_align(out->pix.height, MIN_H, MAX_H, h_align); /* set input/output strides and image sizes */ - in->pix.bytesperline = (in->pix.width * infmt->bpp) >> 3; - in->pix.sizeimage = in->pix.height * in->pix.bytesperline; - out->pix.bytesperline = (out->pix.width * outfmt->bpp) >> 3; - out->pix.sizeimage = out->pix.height * out->pix.bytesperline; + in->pix.bytesperline = infmt->planar ? + clamp_align(in->pix.width, 2 << w_align, MAX_W, w_align) : + clamp_align((in->pix.width * infmt->bpp) >> 3, + 2 << w_align, MAX_W, w_align); + in->pix.sizeimage = infmt->planar ? + (in->pix.height * in->pix.bytesperline * infmt->bpp) >> 3 : + in->pix.height * in->pix.bytesperline; + out->pix.bytesperline = outfmt->planar ? out->pix.width : + (out->pix.width * outfmt->bpp) >> 3; + out->pix.sizeimage = outfmt->planar ? + (out->pix.height * out->pix.bytesperline * outfmt->bpp) >> 3 : + out->pix.height * out->pix.bytesperline; } EXPORT_SYMBOL_GPL(ipu_image_convert_adjust); @@ -1360,6 +1985,7 @@ ipu_image_convert_prepare(struct ipu_soc *ipu, enum ipu_ic_task ic_task, struct ipu_image_convert_chan *chan; struct ipu_image_convert_ctx *ctx; unsigned long flags; + unsigned int i; bool get_res; int ret; @@ -1412,8 +2038,26 @@ ipu_image_convert_prepare(struct ipu_soc *ipu, enum ipu_ic_task ic_task, if (ret) goto out_free; + ret = calc_image_resize_coefficients(ctx, in, out); + if (ret) + goto out_free; + calc_out_tile_map(ctx); + find_seams(ctx, s_image, d_image); + + calc_tile_dimensions(ctx, s_image); + ret = calc_tile_offsets(ctx, s_image); + if (ret) + goto out_free; + + calc_tile_dimensions(ctx, d_image); + ret = calc_tile_offsets(ctx, d_image); + if (ret) + goto out_free; + + calc_tile_resize_coefficients(ctx); + dump_format(ctx, s_image); dump_format(ctx, d_image); @@ -1429,21 +2073,51 @@ ipu_image_convert_prepare(struct ipu_soc *ipu, enum ipu_ic_task ic_task, * for every tile, and therefore would have to be updated for * each buffer which is not possible. So double-buffering is * impossible when either the source or destination images are - * a planar format (YUV420, YUV422P, etc.). + * a planar format (YUV420, YUV422P, etc.). Further, differently + * sized tiles or different resizing coefficients per tile + * prevent double-buffering as well. */ ctx->double_buffering = (ctx->num_tiles > 1 && !s_image->fmt->planar && !d_image->fmt->planar); + for (i = 1; i < ctx->num_tiles; i++) { + if (ctx->in.tile[i].width != ctx->in.tile[0].width || + ctx->in.tile[i].height != ctx->in.tile[0].height || + ctx->out.tile[i].width != ctx->out.tile[0].width || + ctx->out.tile[i].height != ctx->out.tile[0].height) { + ctx->double_buffering = false; + break; + } + } + for (i = 1; i < ctx->in.num_cols; i++) { + if (ctx->resize_coeffs_h[i] != ctx->resize_coeffs_h[0]) { + ctx->double_buffering = false; + break; + } + } + for (i = 1; i < ctx->in.num_rows; i++) { + if (ctx->resize_coeffs_v[i] != ctx->resize_coeffs_v[0]) { + ctx->double_buffering = false; + break; + } + } if (ipu_rot_mode_is_irt(ctx->rot_mode)) { + unsigned long intermediate_size = d_image->tile[0].size; + + for (i = 1; i < ctx->num_tiles; i++) { + if (d_image->tile[i].size > intermediate_size) + intermediate_size = d_image->tile[i].size; + } + ret = alloc_dma_buf(priv, &ctx->rot_intermediate[0], - d_image->tile[0].size); + intermediate_size); if (ret) goto out_free; if (ctx->double_buffering) { ret = alloc_dma_buf(priv, &ctx->rot_intermediate[1], - d_image->tile[0].size); + intermediate_size); if (ret) goto out_free_dmabuf0; } @@ -1524,16 +2198,13 @@ unlock: EXPORT_SYMBOL_GPL(ipu_image_convert_queue); /* Abort any active or pending conversions for this context */ -void ipu_image_convert_abort(struct ipu_image_convert_ctx *ctx) +static void __ipu_image_convert_abort(struct ipu_image_convert_ctx *ctx) { struct ipu_image_convert_chan *chan = ctx->chan; struct ipu_image_convert_priv *priv = chan->priv; struct ipu_image_convert_run *run, *active_run, *tmp; unsigned long flags; int run_count, ret; - bool need_abort; - - reinit_completion(&ctx->aborted); spin_lock_irqsave(&chan->irqlock, flags); @@ -1549,22 +2220,28 @@ void ipu_image_convert_abort(struct ipu_image_convert_ctx *ctx) active_run = (chan->current_run && chan->current_run->ctx == ctx) ? chan->current_run : NULL; - need_abort = (run_count || active_run); + if (active_run) + reinit_completion(&ctx->aborted); - ctx->aborting = need_abort; + ctx->aborting = true; spin_unlock_irqrestore(&chan->irqlock, flags); - if (!need_abort) { + if (!run_count && !active_run) { dev_dbg(priv->ipu->dev, "%s: task %u: no abort needed for ctx %p\n", __func__, chan->ic_task, ctx); return; } + if (!active_run) { + empty_done_q(chan); + return; + } + dev_dbg(priv->ipu->dev, - "%s: task %u: wait for completion: %d runs, active run %p\n", - __func__, chan->ic_task, run_count, active_run); + "%s: task %u: wait for completion: %d runs\n", + __func__, chan->ic_task, run_count); ret = wait_for_completion_timeout(&ctx->aborted, msecs_to_jiffies(10000)); @@ -1572,7 +2249,11 @@ void ipu_image_convert_abort(struct ipu_image_convert_ctx *ctx) dev_warn(priv->ipu->dev, "%s: timeout\n", __func__); force_abort(ctx); } +} +void ipu_image_convert_abort(struct ipu_image_convert_ctx *ctx) +{ + __ipu_image_convert_abort(ctx); ctx->aborting = false; } EXPORT_SYMBOL_GPL(ipu_image_convert_abort); @@ -1586,7 +2267,7 @@ void ipu_image_convert_unprepare(struct ipu_image_convert_ctx *ctx) bool put_res; /* make sure no runs are hanging around */ - ipu_image_convert_abort(ctx); + __ipu_image_convert_abort(ctx); dev_dbg(priv->ipu->dev, "%s: task %u: removing ctx %p\n", __func__, chan->ic_task, ctx); |