From eb8451e0aee1816395506e97965d0917754072de Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Thu, 27 Feb 2014 14:41:14 +0100 Subject: xtensa: Export __invalidate_icache_range When modular code calls flush_icache_range(): ERROR: "__invalidate_icache_range" [drivers/misc/lkdtm.ko] undefined! make[1]: *** [__modpost] Error 1 Signed-off-by: Geert Uytterhoeven Reviewed-by: Kees Cook Signed-off-by: Max Filippov --- arch/xtensa/kernel/xtensa_ksyms.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch') diff --git a/arch/xtensa/kernel/xtensa_ksyms.c b/arch/xtensa/kernel/xtensa_ksyms.c index 80b33ed51f31..88eab32be29e 100644 --- a/arch/xtensa/kernel/xtensa_ksyms.c +++ b/arch/xtensa/kernel/xtensa_ksyms.c @@ -105,6 +105,7 @@ EXPORT_SYMBOL(csum_partial_copy_generic); * Architecture-specific symbols */ EXPORT_SYMBOL(__xtensa_copy_user); +EXPORT_SYMBOL(__invalidate_icache_range); /* * Kernel hacking ... -- cgit v1.2.3 From a3cfda9d2f9ca067a2757b9d85d98cfbaf1e8c63 Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Fri, 21 Mar 2014 21:05:16 +0400 Subject: xtensa: export __{invalidate,flush}_dcache_range Signed-off-by: Max Filippov --- arch/xtensa/kernel/xtensa_ksyms.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'arch') diff --git a/arch/xtensa/kernel/xtensa_ksyms.c b/arch/xtensa/kernel/xtensa_ksyms.c index 88eab32be29e..4d2872fd9bb5 100644 --- a/arch/xtensa/kernel/xtensa_ksyms.c +++ b/arch/xtensa/kernel/xtensa_ksyms.c @@ -20,6 +20,7 @@ #include #include +#include #include #include #include @@ -128,3 +129,8 @@ EXPORT_SYMBOL(common_exception_return); #ifdef CONFIG_FUNCTION_TRACER EXPORT_SYMBOL(_mcount); #endif + +EXPORT_SYMBOL(__invalidate_dcache_range); +#if XCHAL_DCACHE_IS_WRITEBACK +EXPORT_SYMBOL(__flush_dcache_range); +#endif -- cgit v1.2.3 From 25df8198f4b257cf6db4d4f000c53accfa9c28f8 Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Tue, 18 Feb 2014 15:29:11 +0400 Subject: xtensa: enable sorting extable at build time Signed-off-by: Max Filippov Acked-by: David Daney --- arch/xtensa/Kconfig | 1 + scripts/sortextable.c | 5 +++++ 2 files changed, 6 insertions(+) (limited to 'arch') diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig index c87ae7c6e5f9..df6f86ca6c1a 100644 --- a/arch/xtensa/Kconfig +++ b/arch/xtensa/Kconfig @@ -14,6 +14,7 @@ config XTENSA select GENERIC_PCI_IOMAP select ARCH_WANT_IPC_PARSE_VERSION select ARCH_WANT_OPTIONAL_GPIOLIB + select BUILDTIME_EXTABLE_SORT select CLONE_BACKWARDS select IRQ_DOMAIN select HAVE_OPROFILE diff --git a/scripts/sortextable.c b/scripts/sortextable.c index cc49062acdee..1052d4834a44 100644 --- a/scripts/sortextable.c +++ b/scripts/sortextable.c @@ -35,6 +35,10 @@ #define EM_ARCOMPACT 93 #endif +#ifndef EM_XTENSA +#define EM_XTENSA 94 +#endif + #ifndef EM_AARCH64 #define EM_AARCH64 183 #endif @@ -281,6 +285,7 @@ do_file(char const *const fname) case EM_AARCH64: case EM_MICROBLAZE: case EM_MIPS: + case EM_XTENSA: break; } /* end switch */ -- cgit v1.2.3 From 9ba067f93f1eec0d241f002812806b873dd4f802 Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Sun, 23 Mar 2014 03:17:43 +0400 Subject: xtensa: split bootparam and kernel meminfo Bootparam meminfo is a bootloader ABI, kernel meminfo is for the kernel bookkeeping, keep them separate. Kernel doesn't care of memory region types, so drop the type field and don't pass it to add_sysmem_bank. Move kernel sysmem structures and prototypes to asm/sysmem.h and sysmem variable and add_sysmem_bank to mm/init.c Signed-off-by: Max Filippov --- arch/xtensa/include/asm/bootparam.h | 13 ++--------- arch/xtensa/include/asm/sysmem.h | 33 ++++++++++++++++++++++++++++ arch/xtensa/kernel/setup.c | 43 ++++++++----------------------------- arch/xtensa/mm/init.c | 17 +++++++++++++++ 4 files changed, 61 insertions(+), 45 deletions(-) create mode 100644 arch/xtensa/include/asm/sysmem.h (limited to 'arch') diff --git a/arch/xtensa/include/asm/bootparam.h b/arch/xtensa/include/asm/bootparam.h index 23392c5630ce..892aab399ac8 100644 --- a/arch/xtensa/include/asm/bootparam.h +++ b/arch/xtensa/include/asm/bootparam.h @@ -37,23 +37,14 @@ typedef struct bp_tag { unsigned long data[0]; /* data */ } bp_tag_t; -typedef struct meminfo { +struct bp_meminfo { unsigned long type; unsigned long start; unsigned long end; -} meminfo_t; - -#define SYSMEM_BANKS_MAX 5 +}; #define MEMORY_TYPE_CONVENTIONAL 0x1000 #define MEMORY_TYPE_NONE 0x2000 -typedef struct sysmem_info { - int nr_banks; - meminfo_t bank[SYSMEM_BANKS_MAX]; -} sysmem_info_t; - -extern sysmem_info_t sysmem; - #endif #endif diff --git a/arch/xtensa/include/asm/sysmem.h b/arch/xtensa/include/asm/sysmem.h new file mode 100644 index 000000000000..fe7ad750a158 --- /dev/null +++ b/arch/xtensa/include/asm/sysmem.h @@ -0,0 +1,33 @@ +/* + * sysmem-related prototypes. + * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (C) 2014 Cadence Design Systems Inc. + */ + +#ifndef _XTENSA_SYSMEM_H +#define _XTENSA_SYSMEM_H + +#define SYSMEM_BANKS_MAX 31 + +struct meminfo { + unsigned long start; + unsigned long end; +}; + +struct sysmem_info { + int nr_banks; + struct meminfo bank[SYSMEM_BANKS_MAX]; +}; + +extern struct sysmem_info sysmem; + +int add_sysmem_bank(unsigned long start, unsigned long end); +int mem_reserve(unsigned long, unsigned long, int); +void bootmem_init(void); +void zones_init(void); + +#endif /* _XTENSA_SYSMEM_H */ diff --git a/arch/xtensa/kernel/setup.c b/arch/xtensa/kernel/setup.c index 84fe931bb60e..df2b1d6fc843 100644 --- a/arch/xtensa/kernel/setup.c +++ b/arch/xtensa/kernel/setup.c @@ -50,6 +50,7 @@ #include #include #include +#include #include @@ -88,12 +89,6 @@ static char __initdata command_line[COMMAND_LINE_SIZE]; static char default_command_line[COMMAND_LINE_SIZE] __initdata = CONFIG_CMDLINE; #endif -sysmem_info_t __initdata sysmem; - -extern int mem_reserve(unsigned long, unsigned long, int); -extern void bootmem_init(void); -extern void zones_init(void); - /* * Boot parameter parsing. * @@ -113,31 +108,14 @@ typedef struct tagtable { /* parse current tag */ -static int __init add_sysmem_bank(unsigned long type, unsigned long start, - unsigned long end) -{ - if (sysmem.nr_banks >= SYSMEM_BANKS_MAX) { - printk(KERN_WARNING - "Ignoring memory bank 0x%08lx size %ldKB\n", - start, end - start); - return -EINVAL; - } - sysmem.bank[sysmem.nr_banks].type = type; - sysmem.bank[sysmem.nr_banks].start = PAGE_ALIGN(start); - sysmem.bank[sysmem.nr_banks].end = end & PAGE_MASK; - sysmem.nr_banks++; - - return 0; -} - static int __init parse_tag_mem(const bp_tag_t *tag) { - meminfo_t *mi = (meminfo_t *)(tag->data); + struct bp_meminfo *mi = (struct bp_meminfo *)(tag->data); if (mi->type != MEMORY_TYPE_CONVENTIONAL) return -1; - return add_sysmem_bank(mi->type, mi->start, mi->end); + return add_sysmem_bank(mi->start, mi->end); } __tagtable(BP_TAG_MEMORY, parse_tag_mem); @@ -146,8 +124,8 @@ __tagtable(BP_TAG_MEMORY, parse_tag_mem); static int __init parse_tag_initrd(const bp_tag_t* tag) { - meminfo_t* mi; - mi = (meminfo_t*)(tag->data); + struct bp_meminfo *mi = (struct bp_meminfo *)(tag->data); + initrd_start = (unsigned long)__va(mi->start); initrd_end = (unsigned long)__va(mi->end); @@ -255,7 +233,7 @@ void __init early_init_dt_add_memory_arch(u64 base, u64 size) return; size &= PAGE_MASK; - add_sysmem_bank(MEMORY_TYPE_CONVENTIONAL, base, base + size); + add_sysmem_bank(base, base + size); } void * __init early_init_dt_alloc_memory_arch(u64 size, u64 align) @@ -292,8 +270,6 @@ device_initcall(xtensa_device_probe); void __init init_arch(bp_tag_t *bp_start) { - sysmem.nr_banks = 0; - /* Parse boot parameters */ if (bp_start) @@ -304,10 +280,9 @@ void __init init_arch(bp_tag_t *bp_start) #endif if (sysmem.nr_banks == 0) { - sysmem.nr_banks = 1; - sysmem.bank[0].start = PLATFORM_DEFAULT_MEM_START; - sysmem.bank[0].end = PLATFORM_DEFAULT_MEM_START - + PLATFORM_DEFAULT_MEM_SIZE; + add_sysmem_bank(PLATFORM_DEFAULT_MEM_START, + PLATFORM_DEFAULT_MEM_START + + PLATFORM_DEFAULT_MEM_SIZE); } #ifdef CONFIG_CMDLINE_BOOL diff --git a/arch/xtensa/mm/init.c b/arch/xtensa/mm/init.c index aff108df92d3..4f78264e8bd8 100644 --- a/arch/xtensa/mm/init.c +++ b/arch/xtensa/mm/init.c @@ -27,6 +27,23 @@ #include #include #include +#include + +struct sysmem_info sysmem __initdata; + +int __init add_sysmem_bank(unsigned long start, unsigned long end) +{ + if (sysmem.nr_banks >= SYSMEM_BANKS_MAX) { + pr_warn("Ignoring memory bank 0x%08lx size %ldKB\n", + start, end - start); + return -EINVAL; + } + sysmem.bank[sysmem.nr_banks].start = PAGE_ALIGN(start); + sysmem.bank[sysmem.nr_banks].end = end & PAGE_MASK; + sysmem.nr_banks++; + + return 0; +} /* * mem_reserve(start, end, must_exist) -- cgit v1.2.3 From 9d4b52df4b1242e6ba9a00db5f8d62083a56709f Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Sun, 23 Mar 2014 03:34:44 +0400 Subject: xtensa: keep sysmem banks ordered in add_sysmem_bank Rewrite add_sysmem_bank so that it keeps bank order and merges adjacent/overlapping banks. Signed-off-by: Max Filippov --- arch/xtensa/include/asm/sysmem.h | 5 ++ arch/xtensa/mm/init.c | 103 +++++++++++++++++++++++++++++++++++++-- 2 files changed, 103 insertions(+), 5 deletions(-) (limited to 'arch') diff --git a/arch/xtensa/include/asm/sysmem.h b/arch/xtensa/include/asm/sysmem.h index fe7ad750a158..c015c5c8e3f7 100644 --- a/arch/xtensa/include/asm/sysmem.h +++ b/arch/xtensa/include/asm/sysmem.h @@ -18,6 +18,11 @@ struct meminfo { unsigned long end; }; +/* + * Bank array is sorted by .start. + * Banks don't overlap and there's at least one page gap + * between adjacent bank entries. + */ struct sysmem_info { int nr_banks; struct meminfo bank[SYSMEM_BANKS_MAX]; diff --git a/arch/xtensa/mm/init.c b/arch/xtensa/mm/init.c index 4f78264e8bd8..79c0c3d52ae3 100644 --- a/arch/xtensa/mm/init.c +++ b/arch/xtensa/mm/init.c @@ -8,6 +8,7 @@ * for more details. * * Copyright (C) 2001 - 2005 Tensilica Inc. + * Copyright (C) 2014 Cadence Design Systems Inc. * * Chris Zankel * Joe Taylor @@ -31,17 +32,109 @@ struct sysmem_info sysmem __initdata; +/* + * Find bank with maximal .start such that bank.start <= start + */ +static inline struct meminfo * __init find_bank(unsigned long start) +{ + unsigned i; + struct meminfo *it = NULL; + + for (i = 0; i < sysmem.nr_banks; ++i) + if (sysmem.bank[i].start <= start) + it = sysmem.bank + i; + else + break; + return it; +} + +/* + * Move all memory banks starting at 'from' to a new place at 'to', + * adjust nr_banks accordingly. + * Both 'from' and 'to' must be inside the sysmem.bank. + * + * Returns: 0 (success), -ENOMEM (not enough space in the sysmem.bank). + */ +static int __init move_banks(struct meminfo *to, struct meminfo *from) +{ + unsigned n = sysmem.nr_banks - (from - sysmem.bank); + + if (to > from && to - from + sysmem.nr_banks > SYSMEM_BANKS_MAX) + return -ENOMEM; + if (to != from) + memmove(to, from, n * sizeof(struct meminfo)); + sysmem.nr_banks += to - from; + return 0; +} + +/* + * Add new bank to sysmem. Resulting sysmem is the union of bytes of the + * original sysmem and the new bank. + * + * Returns: 0 (success), < 0 (error) + */ int __init add_sysmem_bank(unsigned long start, unsigned long end) { - if (sysmem.nr_banks >= SYSMEM_BANKS_MAX) { - pr_warn("Ignoring memory bank 0x%08lx size %ldKB\n", + unsigned i; + struct meminfo *it = NULL; + unsigned long sz; + unsigned long bank_sz = 0; + + if (start == end || + (start < end) != (PAGE_ALIGN(start) < (end & PAGE_MASK))) { + pr_warn("Ignoring small memory bank 0x%08lx size: %ld bytes\n", start, end - start); return -EINVAL; } - sysmem.bank[sysmem.nr_banks].start = PAGE_ALIGN(start); - sysmem.bank[sysmem.nr_banks].end = end & PAGE_MASK; - sysmem.nr_banks++; + start = PAGE_ALIGN(start); + end &= PAGE_MASK; + sz = end - start; + + it = find_bank(start); + + if (it) + bank_sz = it->end - it->start; + + if (it && bank_sz >= start - it->start) { + if (end - it->start > bank_sz) + it->end = end; + else + return 0; + } else { + if (!it) + it = sysmem.bank; + else + ++it; + + if (it - sysmem.bank < sysmem.nr_banks && + it->start - start <= sz) { + it->start = start; + if (it->end - it->start < sz) + it->end = end; + else + return 0; + } else { + if (move_banks(it + 1, it) < 0) { + pr_warn("Ignoring memory bank 0x%08lx size %ld bytes\n", + start, end - start); + return -EINVAL; + } + it->start = start; + it->end = end; + return 0; + } + } + sz = it->end - it->start; + for (i = it + 1 - sysmem.bank; i < sysmem.nr_banks; ++i) + if (sysmem.bank[i].start - it->start <= sz) { + if (sz < sysmem.bank[i].end - it->start) + it->end = sysmem.bank[i].end; + } else { + break; + } + + move_banks(it + 1, sysmem.bank + i); return 0; } -- cgit v1.2.3 From 6232791833785ae591b211609f6f7c4faa7c6e55 Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Sun, 23 Mar 2014 03:24:45 +0400 Subject: xtensa: keep sysmem banks ordered in mem_reserve Rewrite mem_reserve so that it keeps bank order. Also make its return code more traditional. Signed-off-by: Max Filippov --- arch/xtensa/kernel/setup.c | 2 +- arch/xtensa/mm/init.c | 82 ++++++++++++++++++++++++++++------------------ 2 files changed, 51 insertions(+), 33 deletions(-) (limited to 'arch') diff --git a/arch/xtensa/kernel/setup.c b/arch/xtensa/kernel/setup.c index df2b1d6fc843..017c06aba9b2 100644 --- a/arch/xtensa/kernel/setup.c +++ b/arch/xtensa/kernel/setup.c @@ -462,7 +462,7 @@ void __init setup_arch(char **cmdline_p) #ifdef CONFIG_BLK_DEV_INITRD if (initrd_start < initrd_end) { initrd_is_mapped = mem_reserve(__pa(initrd_start), - __pa(initrd_end), 0); + __pa(initrd_end), 0) == 0; initrd_below_start_ok = 1; } else { initrd_start = 0; diff --git a/arch/xtensa/mm/init.c b/arch/xtensa/mm/init.c index 79c0c3d52ae3..5d23697957bf 100644 --- a/arch/xtensa/mm/init.c +++ b/arch/xtensa/mm/init.c @@ -142,6 +142,8 @@ int __init add_sysmem_bank(unsigned long start, unsigned long end) * mem_reserve(start, end, must_exist) * * Reserve some memory from the memory pool. + * If must_exist is set and a part of the region being reserved does not exist + * memory map is not altered. * * Parameters: * start Start of region, @@ -149,53 +151,69 @@ int __init add_sysmem_bank(unsigned long start, unsigned long end) * must_exist Must exist in memory pool. * * Returns: - * 0 (memory area couldn't be mapped) - * -1 (success) + * 0 (success) + * < 0 (error) */ int __init mem_reserve(unsigned long start, unsigned long end, int must_exist) { - int i; - - if (start == end) - return 0; + struct meminfo *it; + struct meminfo *rm = NULL; + unsigned long sz; + unsigned long bank_sz = 0; start = start & PAGE_MASK; end = PAGE_ALIGN(end); + sz = end - start; + if (!sz) + return -EINVAL; - for (i = 0; i < sysmem.nr_banks; i++) - if (start < sysmem.bank[i].end - && end >= sysmem.bank[i].start) - break; + it = find_bank(start); + + if (it) + bank_sz = it->end - it->start; - if (i == sysmem.nr_banks) { - if (must_exist) - printk (KERN_WARNING "mem_reserve: [0x%0lx, 0x%0lx) " - "not in any region!\n", start, end); - return 0; + if ((!it || end - it->start > bank_sz) && must_exist) { + pr_warn("mem_reserve: [0x%0lx, 0x%0lx) not in any region!\n", + start, end); + return -EINVAL; } - if (start > sysmem.bank[i].start) { - if (end < sysmem.bank[i].end) { - /* split entry */ - if (sysmem.nr_banks >= SYSMEM_BANKS_MAX) - panic("meminfo overflow\n"); - sysmem.bank[sysmem.nr_banks].start = end; - sysmem.bank[sysmem.nr_banks].end = sysmem.bank[i].end; - sysmem.nr_banks++; + if (it && start - it->start < bank_sz) { + if (start == it->start) { + if (end - it->start < bank_sz) { + it->start = end; + return 0; + } else { + rm = it; + } + } else { + it->end = start; + if (end - it->start < bank_sz) + return add_sysmem_bank(end, + it->start + bank_sz); + ++it; } - sysmem.bank[i].end = start; + } - } else if (end < sysmem.bank[i].end) { - sysmem.bank[i].start = end; + if (!it) + it = sysmem.bank; - } else { - /* remove entry */ - sysmem.nr_banks--; - sysmem.bank[i].start = sysmem.bank[sysmem.nr_banks].start; - sysmem.bank[i].end = sysmem.bank[sysmem.nr_banks].end; + for (; it < sysmem.bank + sysmem.nr_banks; ++it) { + if (it->end - start <= sz) { + if (!rm) + rm = it; + } else { + if (it->start - start < sz) + it->start = end; + break; + } } - return -1; + + if (rm) + move_banks(rm, it); + + return 0; } -- cgit v1.2.3 From 06bd2824f7dcbfb8dcd13519239a53d13298d238 Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Fri, 21 Mar 2014 21:04:40 +0400 Subject: xtensa: handle memmap kernel option This option is useful for reserving memory regions for secondary cores in AMP configurations. Implement the following memmap variants: - memmap=nn[KMG]@ss[KMG]: force usage of a specific region of memory; - memmap=nn[KMG]$ss[KMG]: mark specified memory as reserved; - memmap=nn[KMG]: set end of memory. Signed-off-by: Max Filippov --- arch/xtensa/kernel/setup.c | 1 + arch/xtensa/mm/init.c | 50 ++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 51 insertions(+) (limited to 'arch') diff --git a/arch/xtensa/kernel/setup.c b/arch/xtensa/kernel/setup.c index 017c06aba9b2..9757bb74e532 100644 --- a/arch/xtensa/kernel/setup.c +++ b/arch/xtensa/kernel/setup.c @@ -507,6 +507,7 @@ void __init setup_arch(char **cmdline_p) __pa(&_Level6InterruptVector_text_end), 0); #endif + parse_early_param(); bootmem_init(); unflatten_and_copy_device_tree(); diff --git a/arch/xtensa/mm/init.c b/arch/xtensa/mm/init.c index 5d23697957bf..d70ba9333f44 100644 --- a/arch/xtensa/mm/init.c +++ b/arch/xtensa/mm/init.c @@ -332,3 +332,53 @@ void free_initmem(void) { free_initmem_default(-1); } + +static void __init parse_memmap_one(char *p) +{ + char *oldp; + unsigned long start_at, mem_size; + + if (!p) + return; + + oldp = p; + mem_size = memparse(p, &p); + if (p == oldp) + return; + + switch (*p) { + case '@': + start_at = memparse(p + 1, &p); + add_sysmem_bank(start_at, start_at + mem_size); + break; + + case '$': + start_at = memparse(p + 1, &p); + mem_reserve(start_at, start_at + mem_size, 0); + break; + + case 0: + mem_reserve(mem_size, 0, 0); + break; + + default: + pr_warn("Unrecognized memmap syntax: %s\n", p); + break; + } +} + +static int __init parse_memmap_opt(char *str) +{ + while (str) { + char *k = strchr(str, ','); + + if (k) + *k++ = 0; + + parse_memmap_one(str); + str = k; + } + + return 0; +} +early_param("memmap", parse_memmap_opt); -- cgit v1.2.3 From 8585b316bbed9339412d267c1fd8839dd059d69f Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Sun, 23 Mar 2014 03:26:46 +0400 Subject: xtensa: dump sysmem from the bootmem_init Debug dump of physical memory configuration. Useful for inspection of resulting memory map, esp. in the presence of memmap= kernel option. Signed-off-by: Max Filippov --- arch/xtensa/mm/init.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'arch') diff --git a/arch/xtensa/mm/init.c b/arch/xtensa/mm/init.c index d70ba9333f44..03bd025307e3 100644 --- a/arch/xtensa/mm/init.c +++ b/arch/xtensa/mm/init.c @@ -32,6 +32,17 @@ struct sysmem_info sysmem __initdata; +static void __init sysmem_dump(void) +{ + unsigned i; + + pr_debug("Sysmem:\n"); + for (i = 0; i < sysmem.nr_banks; ++i) + pr_debug(" 0x%08lx - 0x%08lx (%ldK)\n", + sysmem.bank[i].start, sysmem.bank[i].end, + (sysmem.bank[i].end - sysmem.bank[i].start) >> 10); +} + /* * Find bank with maximal .start such that bank.start <= start */ @@ -227,6 +238,7 @@ void __init bootmem_init(void) unsigned long bootmap_start, bootmap_size; int i; + sysmem_dump(); max_low_pfn = max_pfn = 0; min_low_pfn = ~0; -- cgit v1.2.3 From 04c6b3e2b5e5c1dbd99ad7620033eafd05ff4c26 Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Fri, 14 Feb 2014 14:08:48 +0400 Subject: xtensa: optimize local_flush_tlb_kernel_range Don't flush whole TLB if only a small kernel range is requested. Signed-off-by: Max Filippov --- arch/xtensa/include/asm/tlbflush.h | 11 ++++------- arch/xtensa/kernel/smp.c | 15 +++++++++++++++ arch/xtensa/mm/tlb.c | 15 +++++++++++++++ 3 files changed, 34 insertions(+), 7 deletions(-) (limited to 'arch') diff --git a/arch/xtensa/include/asm/tlbflush.h b/arch/xtensa/include/asm/tlbflush.h index fc34274ce41b..06875feb27c2 100644 --- a/arch/xtensa/include/asm/tlbflush.h +++ b/arch/xtensa/include/asm/tlbflush.h @@ -36,6 +36,7 @@ void local_flush_tlb_page(struct vm_area_struct *vma, unsigned long page); void local_flush_tlb_range(struct vm_area_struct *vma, unsigned long start, unsigned long end); +void local_flush_tlb_kernel_range(unsigned long start, unsigned long end); #ifdef CONFIG_SMP @@ -44,12 +45,7 @@ void flush_tlb_mm(struct mm_struct *); void flush_tlb_page(struct vm_area_struct *, unsigned long); void flush_tlb_range(struct vm_area_struct *, unsigned long, unsigned long); - -static inline void flush_tlb_kernel_range(unsigned long start, - unsigned long end) -{ - flush_tlb_all(); -} +void flush_tlb_kernel_range(unsigned long start, unsigned long end); #else /* !CONFIG_SMP */ @@ -58,7 +54,8 @@ static inline void flush_tlb_kernel_range(unsigned long start, #define flush_tlb_page(vma, page) local_flush_tlb_page(vma, page) #define flush_tlb_range(vma, vmaddr, end) local_flush_tlb_range(vma, vmaddr, \ end) -#define flush_tlb_kernel_range(start, end) local_flush_tlb_all() +#define flush_tlb_kernel_range(start, end) local_flush_tlb_kernel_range(start, \ + end) #endif /* CONFIG_SMP */ diff --git a/arch/xtensa/kernel/smp.c b/arch/xtensa/kernel/smp.c index aa8bd8717927..40b5a3771fb0 100644 --- a/arch/xtensa/kernel/smp.c +++ b/arch/xtensa/kernel/smp.c @@ -496,6 +496,21 @@ void flush_tlb_range(struct vm_area_struct *vma, on_each_cpu(ipi_flush_tlb_range, &fd, 1); } +static void ipi_flush_tlb_kernel_range(void *arg) +{ + struct flush_data *fd = arg; + local_flush_tlb_kernel_range(fd->addr1, fd->addr2); +} + +void flush_tlb_kernel_range(unsigned long start, unsigned long end) +{ + struct flush_data fd = { + .addr1 = start, + .addr2 = end, + }; + on_each_cpu(ipi_flush_tlb_kernel_range, &fd, 1); +} + /* Cache flush functions */ static void ipi_flush_cache_all(void *arg) diff --git a/arch/xtensa/mm/tlb.c b/arch/xtensa/mm/tlb.c index ade623826788..5ece856c5725 100644 --- a/arch/xtensa/mm/tlb.c +++ b/arch/xtensa/mm/tlb.c @@ -149,6 +149,21 @@ void local_flush_tlb_page(struct vm_area_struct *vma, unsigned long page) local_irq_restore(flags); } +void local_flush_tlb_kernel_range(unsigned long start, unsigned long end) +{ + if (end > start && start >= TASK_SIZE && end <= PAGE_OFFSET && + end - start < _TLB_ENTRIES << PAGE_SHIFT) { + start &= PAGE_MASK; + while (start < end) { + invalidate_itlb_mapping(start); + invalidate_dtlb_mapping(start); + start += PAGE_SIZE; + } + } else { + local_flush_tlb_all(); + } +} + #ifdef CONFIG_DEBUG_TLB_SANITY static unsigned get_pte_for_vaddr(unsigned vaddr) -- cgit v1.2.3 From 65559100655c6ed6ce2e17ffc8d4f3852bc2858a Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Tue, 4 Feb 2014 02:17:09 +0400 Subject: xtensa: add HIGHMEM support Introduce fixmap area just below the vmalloc region. Use it for atomic mapping of high memory pages. High memory on cores with cache aliasing is not supported and is still to be implemented. Fail build for such configurations for now. Signed-off-by: Max Filippov --- arch/xtensa/Kconfig | 18 ++++++++++ arch/xtensa/include/asm/fixmap.h | 58 +++++++++++++++++++++++++++++++ arch/xtensa/include/asm/highmem.h | 45 +++++++++++++++++++++++- arch/xtensa/include/asm/pgtable.h | 4 +++ arch/xtensa/mm/Makefile | 1 + arch/xtensa/mm/cache.c | 7 +++- arch/xtensa/mm/highmem.c | 72 +++++++++++++++++++++++++++++++++++++++ arch/xtensa/mm/init.c | 45 ++++++++++++++++-------- arch/xtensa/mm/mmu.c | 36 ++++++++++++++++++++ 9 files changed, 270 insertions(+), 16 deletions(-) create mode 100644 arch/xtensa/include/asm/fixmap.h create mode 100644 arch/xtensa/mm/highmem.c (limited to 'arch') diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig index df6f86ca6c1a..17b31982c566 100644 --- a/arch/xtensa/Kconfig +++ b/arch/xtensa/Kconfig @@ -190,6 +190,24 @@ config INITIALIZE_XTENSA_MMU_INSIDE_VMLINUX If in doubt, say Y. +config HIGHMEM + bool "High Memory Support" + help + Linux can use the full amount of RAM in the system by + default. However, the default MMUv2 setup only maps the + lowermost 128 MB of memory linearly to the areas starting + at 0xd0000000 (cached) and 0xd8000000 (uncached). + When there are more than 128 MB memory in the system not + all of it can be "permanently mapped" by the kernel. + The physical memory that's not permanently mapped is called + "high memory". + + If you are compiling a kernel which will never run on a + machine with more than 128 MB total physical RAM, answer + N here. + + If unsure, say Y. + endmenu config XTENSA_CALIBRATE_CCOUNT diff --git a/arch/xtensa/include/asm/fixmap.h b/arch/xtensa/include/asm/fixmap.h new file mode 100644 index 000000000000..9f6c33d0428a --- /dev/null +++ b/arch/xtensa/include/asm/fixmap.h @@ -0,0 +1,58 @@ +/* + * fixmap.h: compile-time virtual memory allocation + * + * This file is subject to the terms and conditions of the GNU General Public + * License. See the file "COPYING" in the main directory of this archive + * for more details. + * + * Copyright (C) 1998 Ingo Molnar + * + * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999 + */ + +#ifndef _ASM_FIXMAP_H +#define _ASM_FIXMAP_H + +#include +#ifdef CONFIG_HIGHMEM +#include +#include +#endif + +/* + * Here we define all the compile-time 'special' virtual + * addresses. The point is to have a constant address at + * compile time, but to set the physical address only + * in the boot process. We allocate these special addresses + * from the end of the consistent memory region backwards. + * Also this lets us do fail-safe vmalloc(), we + * can guarantee that these special addresses and + * vmalloc()-ed addresses never overlap. + * + * these 'compile-time allocated' memory buffers are + * fixed-size 4k pages. (or larger if used with an increment + * higher than 1) use fixmap_set(idx,phys) to associate + * physical memory with fixmap indices. + */ +enum fixed_addresses { +#ifdef CONFIG_HIGHMEM + /* reserved pte's for temporary kernel mappings */ + FIX_KMAP_BEGIN, + FIX_KMAP_END = FIX_KMAP_BEGIN + (KM_TYPE_NR * NR_CPUS) - 1, +#endif + __end_of_fixed_addresses +}; + +#define FIXADDR_TOP (VMALLOC_START - PAGE_SIZE) +#define FIXADDR_SIZE (__end_of_fixed_addresses << PAGE_SHIFT) +#define FIXADDR_START ((FIXADDR_TOP - FIXADDR_SIZE) & PMD_MASK) + +#include + +#define kmap_get_fixmap_pte(vaddr) \ + pte_offset_kernel( \ + pmd_offset(pud_offset(pgd_offset_k(vaddr), (vaddr)), (vaddr)), \ + (vaddr) \ + ) + +#endif diff --git a/arch/xtensa/include/asm/highmem.h b/arch/xtensa/include/asm/highmem.h index 80be15124697..2653ef5d55f1 100644 --- a/arch/xtensa/include/asm/highmem.h +++ b/arch/xtensa/include/asm/highmem.h @@ -6,11 +6,54 @@ * this archive for more details. * * Copyright (C) 2003 - 2005 Tensilica Inc. + * Copyright (C) 2014 Cadence Design Systems Inc. */ #ifndef _XTENSA_HIGHMEM_H #define _XTENSA_HIGHMEM_H -extern void flush_cache_kmaps(void); +#include +#include +#include +#include + +#define PKMAP_BASE (FIXADDR_START - PMD_SIZE) +#define LAST_PKMAP PTRS_PER_PTE +#define LAST_PKMAP_MASK (LAST_PKMAP - 1) +#define PKMAP_NR(virt) (((virt) - PKMAP_BASE) >> PAGE_SHIFT) +#define PKMAP_ADDR(nr) (PKMAP_BASE + ((nr) << PAGE_SHIFT)) + +#define kmap_prot PAGE_KERNEL + +extern pte_t *pkmap_page_table; + +void *kmap_high(struct page *page); +void kunmap_high(struct page *page); + +static inline void *kmap(struct page *page) +{ + BUG_ON(in_interrupt()); + if (!PageHighMem(page)) + return page_address(page); + return kmap_high(page); +} + +static inline void kunmap(struct page *page) +{ + BUG_ON(in_interrupt()); + if (!PageHighMem(page)) + return; + kunmap_high(page); +} + +static inline void flush_cache_kmaps(void) +{ + flush_cache_all(); +} + +void *kmap_atomic(struct page *page); +void __kunmap_atomic(void *kvaddr); + +void kmap_init(void); #endif diff --git a/arch/xtensa/include/asm/pgtable.h b/arch/xtensa/include/asm/pgtable.h index 216446295ada..4b0ca35a93b1 100644 --- a/arch/xtensa/include/asm/pgtable.h +++ b/arch/xtensa/include/asm/pgtable.h @@ -310,6 +310,10 @@ set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pteval) update_pte(ptep, pteval); } +static inline void set_pte(pte_t *ptep, pte_t pteval) +{ + update_pte(ptep, pteval); +} static inline void set_pmd(pmd_t *pmdp, pmd_t pmdval) diff --git a/arch/xtensa/mm/Makefile b/arch/xtensa/mm/Makefile index f0b646d2f843..f54f78e24d7b 100644 --- a/arch/xtensa/mm/Makefile +++ b/arch/xtensa/mm/Makefile @@ -4,3 +4,4 @@ obj-y := init.o cache.o misc.o obj-$(CONFIG_MMU) += fault.o mmu.o tlb.o +obj-$(CONFIG_HIGHMEM) += highmem.o diff --git a/arch/xtensa/mm/cache.c b/arch/xtensa/mm/cache.c index ba4c47f291b1..63cbb867dadd 100644 --- a/arch/xtensa/mm/cache.c +++ b/arch/xtensa/mm/cache.c @@ -59,6 +59,10 @@ * */ +#if (DCACHE_WAY_SIZE > PAGE_SIZE) && defined(CONFIG_HIGHMEM) +#error "HIGHMEM is not supported on cores with aliasing cache." +#endif + #if (DCACHE_WAY_SIZE > PAGE_SIZE) && XCHAL_DCACHE_IS_WRITEBACK /* @@ -179,10 +183,11 @@ update_mmu_cache(struct vm_area_struct * vma, unsigned long addr, pte_t *ptep) #else if (!PageReserved(page) && !test_bit(PG_arch_1, &page->flags) && (vma->vm_flags & VM_EXEC) != 0) { - unsigned long paddr = (unsigned long) page_address(page); + unsigned long paddr = (unsigned long)kmap_atomic(page); __flush_dcache_page(paddr); __invalidate_icache_page(paddr); set_bit(PG_arch_1, &page->flags); + kunmap_atomic((void *)paddr); } #endif } diff --git a/arch/xtensa/mm/highmem.c b/arch/xtensa/mm/highmem.c new file mode 100644 index 000000000000..17a8c0d6fd17 --- /dev/null +++ b/arch/xtensa/mm/highmem.c @@ -0,0 +1,72 @@ +/* + * High memory support for Xtensa architecture + * + * This file is subject to the terms and conditions of the GNU General + * Public License. See the file "COPYING" in the main directory of + * this archive for more details. + * + * Copyright (C) 2014 Cadence Design Systems Inc. + */ + +#include +#include +#include + +static pte_t *kmap_pte; + +void *kmap_atomic(struct page *page) +{ + enum fixed_addresses idx; + unsigned long vaddr; + int type; + + pagefault_disable(); + if (!PageHighMem(page)) + return page_address(page); + + type = kmap_atomic_idx_push(); + idx = type + KM_TYPE_NR * smp_processor_id(); + vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); +#ifdef CONFIG_DEBUG_HIGHMEM + BUG_ON(!pte_none(*(kmap_pte - idx))); +#endif + set_pte(kmap_pte - idx, mk_pte(page, PAGE_KERNEL_EXEC)); + + return (void *)vaddr; +} +EXPORT_SYMBOL(kmap_atomic); + +void __kunmap_atomic(void *kvaddr) +{ + int idx, type; + + if (kvaddr >= (void *)FIXADDR_START && + kvaddr < (void *)FIXADDR_TOP) { + type = kmap_atomic_idx(); + idx = type + KM_TYPE_NR * smp_processor_id(); + + /* + * Force other mappings to Oops if they'll try to access this + * pte without first remap it. Keeping stale mappings around + * is a bad idea also, in case the page changes cacheability + * attributes or becomes a protected page in a hypervisor. + */ + pte_clear(&init_mm, kvaddr, kmap_pte - idx); + local_flush_tlb_kernel_range((unsigned long)kvaddr, + (unsigned long)kvaddr + PAGE_SIZE); + + kmap_atomic_idx_pop(); + } + + pagefault_enable(); +} +EXPORT_SYMBOL(__kunmap_atomic); + +void __init kmap_init(void) +{ + unsigned long kmap_vstart; + + /* cache the first kmap pte */ + kmap_vstart = __fix_to_virt(FIX_KMAP_BEGIN); + kmap_pte = kmap_get_fixmap_pte(kmap_vstart); +} diff --git a/arch/xtensa/mm/init.c b/arch/xtensa/mm/init.c index 03bd025307e3..4224256bb215 100644 --- a/arch/xtensa/mm/init.c +++ b/arch/xtensa/mm/init.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include #include @@ -296,19 +297,13 @@ void __init bootmem_init(void) void __init zones_init(void) { - unsigned long zones_size[MAX_NR_ZONES]; - int i; - /* All pages are DMA-able, so we put them all in the DMA zone. */ - - zones_size[ZONE_DMA] = max_low_pfn - ARCH_PFN_OFFSET; - for (i = 1; i < MAX_NR_ZONES; i++) - zones_size[i] = 0; - + unsigned long zones_size[MAX_NR_ZONES] = { + [ZONE_DMA] = max_low_pfn - ARCH_PFN_OFFSET, #ifdef CONFIG_HIGHMEM - zones_size[ZONE_HIGHMEM] = max_pfn - max_low_pfn; + [ZONE_HIGHMEM] = max_pfn - max_low_pfn, #endif - + }; free_area_init_node(0, zones_size, ARCH_PFN_OFFSET, NULL); } @@ -318,16 +313,38 @@ void __init zones_init(void) void __init mem_init(void) { - max_mapnr = max_low_pfn - ARCH_PFN_OFFSET; - high_memory = (void *) __va(max_low_pfn << PAGE_SHIFT); - #ifdef CONFIG_HIGHMEM -#error HIGHGMEM not implemented in init.c + unsigned long tmp; + + reset_all_zones_managed_pages(); + for (tmp = max_low_pfn; tmp < max_pfn; tmp++) + free_highmem_page(pfn_to_page(tmp)); #endif + max_mapnr = max_pfn - ARCH_PFN_OFFSET; + high_memory = (void *)__va(max_low_pfn << PAGE_SHIFT); + free_all_bootmem(); mem_init_print_info(NULL); + pr_info("virtual kernel memory layout:\n" +#ifdef CONFIG_HIGHMEM + " pkmap : 0x%08lx - 0x%08lx (%5lu kB)\n" + " fixmap : 0x%08lx - 0x%08lx (%5lu kB)\n" +#endif + " vmalloc : 0x%08x - 0x%08x (%5u MB)\n" + " lowmem : 0x%08x - 0x%08lx (%5lu MB)\n", +#ifdef CONFIG_HIGHMEM + PKMAP_BASE, PKMAP_BASE + LAST_PKMAP * PAGE_SIZE, + (LAST_PKMAP*PAGE_SIZE) >> 10, + FIXADDR_START, FIXADDR_TOP, + (FIXADDR_TOP - FIXADDR_START) >> 10, +#endif + VMALLOC_START, VMALLOC_END, + (VMALLOC_END - VMALLOC_START) >> 20, + PAGE_OFFSET, PAGE_OFFSET + + (max_low_pfn - min_low_pfn) * PAGE_SIZE, + ((max_low_pfn - min_low_pfn) * PAGE_SIZE) >> 20); } #ifdef CONFIG_BLK_DEV_INITRD diff --git a/arch/xtensa/mm/mmu.c b/arch/xtensa/mm/mmu.c index 861203e958da..3429b483d9f8 100644 --- a/arch/xtensa/mm/mmu.c +++ b/arch/xtensa/mm/mmu.c @@ -3,6 +3,7 @@ * * Extracted from init.c */ +#include #include #include #include @@ -16,9 +17,44 @@ #include #include +#if defined(CONFIG_HIGHMEM) +static void * __init init_pmd(unsigned long vaddr) +{ + pgd_t *pgd = pgd_offset_k(vaddr); + pmd_t *pmd = pmd_offset(pgd, vaddr); + + if (pmd_none(*pmd)) { + unsigned i; + pte_t *pte = alloc_bootmem_low_pages(PAGE_SIZE); + + for (i = 0; i < 1024; i++) + pte_clear(NULL, 0, pte + i); + + set_pmd(pmd, __pmd(((unsigned long)pte) & PAGE_MASK)); + BUG_ON(pte != pte_offset_kernel(pmd, 0)); + pr_debug("%s: vaddr: 0x%08lx, pmd: 0x%p, pte: 0x%p\n", + __func__, vaddr, pmd, pte); + return pte; + } else { + return pte_offset_kernel(pmd, 0); + } +} + +static void __init fixedrange_init(void) +{ + BUILD_BUG_ON(FIXADDR_SIZE > PMD_SIZE); + init_pmd(__fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK); +} +#endif + void __init paging_init(void) { memset(swapper_pg_dir, 0, PAGE_SIZE); +#ifdef CONFIG_HIGHMEM + fixedrange_init(); + pkmap_page_table = init_pmd(PKMAP_BASE); + kmap_init(); +#endif } /* -- cgit v1.2.3 From 08a7bbf624072bed1cb35e5b4db7d538580f731c Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Sun, 23 Feb 2014 03:48:32 +0400 Subject: xtensa: xtfpga: introduce SoC I/O bus Signed-off-by: Max Filippov --- arch/xtensa/boot/dts/xtfpga-flash-16m.dtsi | 48 ++++++++++++++++-------------- arch/xtensa/boot/dts/xtfpga-flash-4m.dtsi | 32 ++++++++++---------- arch/xtensa/boot/dts/xtfpga.dtsi | 37 +++++++++++++---------- 3 files changed, 64 insertions(+), 53 deletions(-) (limited to 'arch') diff --git a/arch/xtensa/boot/dts/xtfpga-flash-16m.dtsi b/arch/xtensa/boot/dts/xtfpga-flash-16m.dtsi index e5703c7beeb6..1d97203c18e7 100644 --- a/arch/xtensa/boot/dts/xtfpga-flash-16m.dtsi +++ b/arch/xtensa/boot/dts/xtfpga-flash-16m.dtsi @@ -1,26 +1,28 @@ / { - flash: flash@f8000000 { - #address-cells = <1>; - #size-cells = <1>; - compatible = "cfi-flash"; - reg = <0xf8000000 0x01000000>; - bank-width = <2>; - device-width = <2>; - partition@0x0 { - label = "boot loader area"; - reg = <0x00000000 0x00400000>; + soc { + flash: flash@08000000 { + #address-cells = <1>; + #size-cells = <1>; + compatible = "cfi-flash"; + reg = <0x08000000 0x01000000>; + bank-width = <2>; + device-width = <2>; + partition@0x0 { + label = "boot loader area"; + reg = <0x00000000 0x00400000>; + }; + partition@0x400000 { + label = "kernel image"; + reg = <0x00400000 0x00600000>; + }; + partition@0xa00000 { + label = "data"; + reg = <0x00a00000 0x005e0000>; + }; + partition@0xfe0000 { + label = "boot environment"; + reg = <0x00fe0000 0x00020000>; + }; }; - partition@0x400000 { - label = "kernel image"; - reg = <0x00400000 0x00600000>; - }; - partition@0xa00000 { - label = "data"; - reg = <0x00a00000 0x005e0000>; - }; - partition@0xfe0000 { - label = "boot environment"; - reg = <0x00fe0000 0x00020000>; - }; - }; + }; }; diff --git a/arch/xtensa/boot/dts/xtfpga-flash-4m.dtsi b/arch/xtensa/boot/dts/xtfpga-flash-4m.dtsi index 6f9c10d6b689..d1c621ca8be1 100644 --- a/arch/xtensa/boot/dts/xtfpga-flash-4m.dtsi +++ b/arch/xtensa/boot/dts/xtfpga-flash-4m.dtsi @@ -1,18 +1,20 @@ / { - flash: flash@f8000000 { - #address-cells = <1>; - #size-cells = <1>; - compatible = "cfi-flash"; - reg = <0xf8000000 0x00400000>; - bank-width = <2>; - device-width = <2>; - partition@0x0 { - label = "boot loader area"; - reg = <0x00000000 0x003f0000>; + soc { + flash: flash@08000000 { + #address-cells = <1>; + #size-cells = <1>; + compatible = "cfi-flash"; + reg = <0x08000000 0x00400000>; + bank-width = <2>; + device-width = <2>; + partition@0x0 { + label = "boot loader area"; + reg = <0x00000000 0x003f0000>; + }; + partition@0x3f0000 { + label = "boot environment"; + reg = <0x003f0000 0x00010000>; + }; }; - partition@0x3f0000 { - label = "boot environment"; - reg = <0x003f0000 0x00010000>; - }; - }; + }; }; diff --git a/arch/xtensa/boot/dts/xtfpga.dtsi b/arch/xtensa/boot/dts/xtfpga.dtsi index e7370b11348e..dec9178840f6 100644 --- a/arch/xtensa/boot/dts/xtfpga.dtsi +++ b/arch/xtensa/boot/dts/xtfpga.dtsi @@ -42,21 +42,28 @@ }; }; - serial0: serial@fd050020 { - device_type = "serial"; - compatible = "ns16550a"; - no-loopback-test; - reg = <0xfd050020 0x20>; - reg-shift = <2>; - interrupts = <0 1>; /* external irq 0 */ - clocks = <&osc>; - }; + soc { + #address-cells = <1>; + #size-cells = <1>; + compatible = "simple-bus"; + ranges = <0x00000000 0xf0000000 0x10000000>; - enet0: ethoc@fd030000 { - compatible = "opencores,ethoc"; - reg = <0xfd030000 0x4000 0xfd800000 0x4000>; - interrupts = <1 1>; /* external irq 1 */ - local-mac-address = [00 50 c2 13 6f 00]; - clocks = <&osc>; + serial0: serial@0d050020 { + device_type = "serial"; + compatible = "ns16550a"; + no-loopback-test; + reg = <0x0d050020 0x20>; + reg-shift = <2>; + interrupts = <0 1>; /* external irq 0 */ + clocks = <&osc>; + }; + + enet0: ethoc@0d030000 { + compatible = "opencores,ethoc"; + reg = <0x0d030000 0x4000 0x0d800000 0x4000>; + interrupts = <1 1>; /* external irq 1 */ + local-mac-address = [00 50 c2 13 6f 00]; + clocks = <&osc>; + }; }; }; -- cgit v1.2.3 From 9c602629e34bad88a464e08de08118e80beee0d8 Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Sun, 28 Apr 2013 19:15:56 +0400 Subject: xtensa: add support for KC705 Signed-off-by: Max Filippov --- arch/xtensa/boot/dts/kc705.dts | 11 +++++++++++ arch/xtensa/boot/dts/xtfpga-flash-128m.dtsi | 28 ++++++++++++++++++++++++++++ 2 files changed, 39 insertions(+) create mode 100644 arch/xtensa/boot/dts/kc705.dts create mode 100644 arch/xtensa/boot/dts/xtfpga-flash-128m.dtsi (limited to 'arch') diff --git a/arch/xtensa/boot/dts/kc705.dts b/arch/xtensa/boot/dts/kc705.dts new file mode 100644 index 000000000000..742a347be67a --- /dev/null +++ b/arch/xtensa/boot/dts/kc705.dts @@ -0,0 +1,11 @@ +/dts-v1/; +/include/ "xtfpga.dtsi" +/include/ "xtfpga-flash-128m.dtsi" + +/ { + compatible = "cdns,xtensa-kc705"; + memory@0 { + device_type = "memory"; + reg = <0x00000000 0x08000000>; + }; +}; diff --git a/arch/xtensa/boot/dts/xtfpga-flash-128m.dtsi b/arch/xtensa/boot/dts/xtfpga-flash-128m.dtsi new file mode 100644 index 000000000000..d3a88e029873 --- /dev/null +++ b/arch/xtensa/boot/dts/xtfpga-flash-128m.dtsi @@ -0,0 +1,28 @@ +/ { + soc { + flash: flash@00000000 { + #address-cells = <1>; + #size-cells = <1>; + compatible = "cfi-flash"; + reg = <0x00000000 0x08000000>; + bank-width = <2>; + device-width = <2>; + partition@0x0 { + label = "data"; + reg = <0x00000000 0x06000000>; + }; + partition@0x6000000 { + label = "boot loader area"; + reg = <0x06000000 0x00800000>; + }; + partition@0x6800000 { + label = "kernel image"; + reg = <0x06800000 0x017e0000>; + }; + partition@0x7fe0000 { + label = "boot environment"; + reg = <0x07fe0000 0x00020000>; + }; + }; + }; +}; -- cgit v1.2.3 From cfe8255f0afcf5d2d1dbfa093223cafe38773fa7 Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Thu, 10 Apr 2014 02:40:00 +0400 Subject: xtensa: xt2000: drop redundant sysmem initialization sysmem structure initialization in xt2000 platform_init is identical to the one done in init_arch just before the call to platform_init. Signed-off-by: Max Filippov --- arch/xtensa/platforms/xt2000/setup.c | 12 +----------- 1 file changed, 1 insertion(+), 11 deletions(-) (limited to 'arch') diff --git a/arch/xtensa/platforms/xt2000/setup.c b/arch/xtensa/platforms/xt2000/setup.c index f9bc87966290..b90555cb8089 100644 --- a/arch/xtensa/platforms/xt2000/setup.c +++ b/arch/xtensa/platforms/xt2000/setup.c @@ -92,18 +92,8 @@ void __init platform_setup(char** cmdline) /* early initialization */ -extern sysmem_info_t __initdata sysmem; - -void platform_init(bp_tag_t* first) +void __init platform_init(bp_tag_t *first) { - /* Set default memory block if not provided by the bootloader. */ - - if (sysmem.nr_banks == 0) { - sysmem.nr_banks = 1; - sysmem.bank[0].start = PLATFORM_DEFAULT_MEM_START; - sysmem.bank[0].end = PLATFORM_DEFAULT_MEM_START - + PLATFORM_DEFAULT_MEM_SIZE; - } } /* Heartbeat. Let the LED blink. */ -- cgit v1.2.3 From 8f2dd677bec68fb55904799a82674c9e64b23be3 Mon Sep 17 00:00:00 2001 From: Behan Webster Date: Mon, 21 Apr 2014 22:40:27 -0700 Subject: x86: LLVMLinux: Wrap -mno-80387 with cc-option Wrap -mno-80387 gcc options with cc-option so they don't break clang. Signed-off-by: Behan Webster Cc: torvalds@linux-foundation.org Cc: dwmw2@infradead.org Cc: pageexec@freemail.hu Link: http://lkml.kernel.org/r/1398145227-25053-1-git-send-email-behanw@converseincode.com Signed-off-by: Ingo Molnar --- arch/x86/Makefile | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/x86/Makefile b/arch/x86/Makefile index d1b7c377a234..ce6ad7e6a7d7 100644 --- a/arch/x86/Makefile +++ b/arch/x86/Makefile @@ -83,7 +83,9 @@ else KBUILD_CFLAGS += -m64 # Don't autogenerate traditional x87, MMX or SSE instructions - KBUILD_CFLAGS += -mno-mmx -mno-sse -mno-80387 -mno-fp-ret-in-387 + KBUILD_CFLAGS += -mno-mmx -mno-sse + KBUILD_CFLAGS += $(call cc-option,-mno-80387) + KBUILD_CFLAGS += $(call cc-option,-mno-fp-ret-in-387) # Use -mpreferred-stack-boundary=3 if supported. KBUILD_CFLAGS += $(call cc-option,-mpreferred-stack-boundary=3) -- cgit v1.2.3 From ae107d06137ce210ea21d1423443d3638599f297 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Tue, 22 Apr 2014 20:40:25 +0200 Subject: dt: Fix binding typos in clock-names and interrupt-names s/interrupts-names/interrupt-names/g s/clocks-names/clock-names/g Some of the binding files and device tree files get this wrong and the kernel won't be able to pick it up. Fix them up now so that they don't get widely used. Signed-off-by: Geert Uytterhoeven Acked-by : Patrice Chotard Signed-off-by: Grant Likely --- Documentation/devicetree/bindings/net/socfpga-dwmac.txt | 2 +- Documentation/devicetree/bindings/net/stmmac.txt | 2 +- Documentation/devicetree/bindings/pinctrl/pinctrl-st.txt | 4 ++-- .../devicetree/bindings/sound/davinci-mcasp-audio.txt | 2 +- arch/arm/boot/dts/am33xx.dtsi | 4 ++-- arch/arm/boot/dts/am4372.dtsi | 4 ++-- arch/arm/boot/dts/stih415-pinctrl.dtsi | 10 +++++----- arch/arm/boot/dts/stih416-pinctrl.dtsi | 10 +++++----- drivers/of/irq.c | 2 +- 9 files changed, 20 insertions(+), 20 deletions(-) (limited to 'arch') diff --git a/Documentation/devicetree/bindings/net/socfpga-dwmac.txt b/Documentation/devicetree/bindings/net/socfpga-dwmac.txt index 636f0ac4e223..2a60cd3e8d5d 100644 --- a/Documentation/devicetree/bindings/net/socfpga-dwmac.txt +++ b/Documentation/devicetree/bindings/net/socfpga-dwmac.txt @@ -23,5 +23,5 @@ gmac0: ethernet@ff700000 { interrupt-names = "macirq"; mac-address = [00 00 00 00 00 00];/* Filled in by U-Boot */ clocks = <&emac_0_clk>; - clocks-names = "stmmaceth"; + clock-names = "stmmaceth"; }; diff --git a/Documentation/devicetree/bindings/net/stmmac.txt b/Documentation/devicetree/bindings/net/stmmac.txt index 80c1fb8bfbb8..a2acd2b26baf 100644 --- a/Documentation/devicetree/bindings/net/stmmac.txt +++ b/Documentation/devicetree/bindings/net/stmmac.txt @@ -33,7 +33,7 @@ Optional properties: - max-frame-size: See ethernet.txt file in the same directory - clocks: If present, the first clock should be the GMAC main clock, further clocks may be specified in derived bindings. -- clocks-names: One name for each entry in the clocks property, the +- clock-names: One name for each entry in the clocks property, the first one should be "stmmaceth". Examples: diff --git a/Documentation/devicetree/bindings/pinctrl/pinctrl-st.txt b/Documentation/devicetree/bindings/pinctrl/pinctrl-st.txt index 4bd5be0e5e7d..26bcb18f4e60 100644 --- a/Documentation/devicetree/bindings/pinctrl/pinctrl-st.txt +++ b/Documentation/devicetree/bindings/pinctrl/pinctrl-st.txt @@ -83,7 +83,7 @@ Example: reg = <0xfe61f080 0x4>; reg-names = "irqmux"; interrupts = ; - interrupts-names = "irqmux"; + interrupt-names = "irqmux"; ranges = <0 0xfe610000 0x5000>; PIO0: gpio@fe610000 { @@ -165,7 +165,7 @@ sdhci0:sdhci@fe810000{ interrupt-parent = <&PIO3>; #interrupt-cells = <2>; interrupts = <3 IRQ_TYPE_LEVEL_HIGH>; /* Interrupt line via PIO3-3 */ - interrupts-names = "card-detect"; + interrupt-names = "card-detect"; pinctrl-names = "default"; pinctrl-0 = <&pinctrl_mmc>; }; diff --git a/Documentation/devicetree/bindings/sound/davinci-mcasp-audio.txt b/Documentation/devicetree/bindings/sound/davinci-mcasp-audio.txt index 569b26c4a81e..60ca07996458 100644 --- a/Documentation/devicetree/bindings/sound/davinci-mcasp-audio.txt +++ b/Documentation/devicetree/bindings/sound/davinci-mcasp-audio.txt @@ -47,7 +47,7 @@ mcasp0: mcasp0@1d00000 { reg = <0x100000 0x3000>; reg-names "mpu"; interrupts = <82>, <83>; - interrupts-names = "tx", "rx"; + interrupt-names = "tx", "rx"; op-mode = <0>; /* MCASP_IIS_MODE */ tdm-slots = <2>; serial-dir = < diff --git a/arch/arm/boot/dts/am33xx.dtsi b/arch/arm/boot/dts/am33xx.dtsi index 9770e35f2536..3bce3c9dbd24 100644 --- a/arch/arm/boot/dts/am33xx.dtsi +++ b/arch/arm/boot/dts/am33xx.dtsi @@ -802,7 +802,7 @@ <0x46000000 0x400000>; reg-names = "mpu", "dat"; interrupts = <80>, <81>; - interrupts-names = "tx", "rx"; + interrupt-names = "tx", "rx"; status = "disabled"; dmas = <&edma 8>, <&edma 9>; @@ -816,7 +816,7 @@ <0x46400000 0x400000>; reg-names = "mpu", "dat"; interrupts = <82>, <83>; - interrupts-names = "tx", "rx"; + interrupt-names = "tx", "rx"; status = "disabled"; dmas = <&edma 10>, <&edma 11>; diff --git a/arch/arm/boot/dts/am4372.dtsi b/arch/arm/boot/dts/am4372.dtsi index 36d523a26831..d1f8707ff1df 100644 --- a/arch/arm/boot/dts/am4372.dtsi +++ b/arch/arm/boot/dts/am4372.dtsi @@ -691,7 +691,7 @@ <0x46000000 0x400000>; reg-names = "mpu", "dat"; interrupts = <80>, <81>; - interrupts-names = "tx", "rx"; + interrupt-names = "tx", "rx"; status = "disabled"; dmas = <&edma 8>, <&edma 9>; @@ -705,7 +705,7 @@ <0x46400000 0x400000>; reg-names = "mpu", "dat"; interrupts = <82>, <83>; - interrupts-names = "tx", "rx"; + interrupt-names = "tx", "rx"; status = "disabled"; dmas = <&edma 10>, <&edma 11>; diff --git a/arch/arm/boot/dts/stih415-pinctrl.dtsi b/arch/arm/boot/dts/stih415-pinctrl.dtsi index f09fb10a3791..81df870e5ee6 100644 --- a/arch/arm/boot/dts/stih415-pinctrl.dtsi +++ b/arch/arm/boot/dts/stih415-pinctrl.dtsi @@ -49,7 +49,7 @@ reg = <0xfe61f080 0x4>; reg-names = "irqmux"; interrupts = ; - interrupts-names = "irqmux"; + interrupt-names = "irqmux"; ranges = <0 0xfe610000 0x5000>; PIO0: gpio@fe610000 { @@ -187,7 +187,7 @@ reg = <0xfee0f080 0x4>; reg-names = "irqmux"; interrupts = ; - interrupts-names = "irqmux"; + interrupt-names = "irqmux"; ranges = <0 0xfee00000 0x8000>; PIO5: gpio@fee00000 { @@ -282,7 +282,7 @@ reg = <0xfe82f080 0x4>; reg-names = "irqmux"; interrupts = ; - interrupts-names = "irqmux"; + interrupt-names = "irqmux"; ranges = <0 0xfe820000 0x8000>; PIO13: gpio@fe820000 { @@ -423,7 +423,7 @@ reg = <0xfd6bf080 0x4>; reg-names = "irqmux"; interrupts = ; - interrupts-names = "irqmux"; + interrupt-names = "irqmux"; ranges = <0 0xfd6b0000 0x3000>; PIO100: gpio@fd6b0000 { @@ -460,7 +460,7 @@ reg = <0xfd33f080 0x4>; reg-names = "irqmux"; interrupts = ; - interrupts-names = "irqmux"; + interrupt-names = "irqmux"; ranges = <0 0xfd330000 0x5000>; PIO103: gpio@fd330000 { diff --git a/arch/arm/boot/dts/stih416-pinctrl.dtsi b/arch/arm/boot/dts/stih416-pinctrl.dtsi index aeea304086eb..250d5ecc951e 100644 --- a/arch/arm/boot/dts/stih416-pinctrl.dtsi +++ b/arch/arm/boot/dts/stih416-pinctrl.dtsi @@ -53,7 +53,7 @@ reg = <0xfe61f080 0x4>; reg-names = "irqmux"; interrupts = ; - interrupts-names = "irqmux"; + interrupt-names = "irqmux"; ranges = <0 0xfe610000 0x6000>; PIO0: gpio@fe610000 { @@ -201,7 +201,7 @@ reg = <0xfee0f080 0x4>; reg-names = "irqmux"; interrupts = ; - interrupts-names = "irqmux"; + interrupt-names = "irqmux"; ranges = <0 0xfee00000 0x10000>; PIO5: gpio@fee00000 { @@ -333,7 +333,7 @@ reg = <0xfe82f080 0x4>; reg-names = "irqmux"; interrupts = ; - interrupts-names = "irqmux"; + interrupt-names = "irqmux"; ranges = <0 0xfe820000 0x6000>; PIO13: gpio@fe820000 { @@ -461,7 +461,7 @@ reg = <0xfd6bf080 0x4>; reg-names = "irqmux"; interrupts = ; - interrupts-names = "irqmux"; + interrupt-names = "irqmux"; ranges = <0 0xfd6b0000 0x3000>; PIO100: gpio@fd6b0000 { @@ -498,7 +498,7 @@ reg = <0xfd33f080 0x4>; reg-names = "irqmux"; interrupts = ; - interrupts-names = "irqmux"; + interrupt-names = "irqmux"; ranges = <0 0xfd330000 0x5000>; PIO103: gpio@fd330000 { diff --git a/drivers/of/irq.c b/drivers/of/irq.c index 9bcf2cf19357..e2e4c548a42f 100644 --- a/drivers/of/irq.c +++ b/drivers/of/irq.c @@ -364,7 +364,7 @@ int of_irq_to_resource(struct device_node *dev, int index, struct resource *r) memset(r, 0, sizeof(*r)); /* - * Get optional "interrupts-names" property to add a name + * Get optional "interrupt-names" property to add a name * to the resource. */ of_property_read_string_index(dev, "interrupt-names", index, -- cgit v1.2.3 From 9f7ff8931e3c5ddc8535476971ec9501e9555c05 Mon Sep 17 00:00:00 2001 From: Stephane Eranian Date: Wed, 23 Apr 2014 19:04:19 +0200 Subject: perf/x86: Fix RAPL rdmsrl_safe() usage This patch fixes a bug introduced by: 24223657806a ("perf/x86/intel: Use rdmsrl_safe() when initializing RAPL PMU") The rdmsrl_safe() function returns 0 on success. The current code was failing to detect the RAPL PMU on real hardware (missing /sys/devices/power) because the return value of rdmsrl_safe() was misinterpreted. Signed-off-by: Stephane Eranian Acked-by: Borislav Petkov Acked-by: Venkatesh Srinivas Cc: peterz@infradead.org Cc: zheng.z.yan@intel.com Link: http://lkml.kernel.org/r/20140423170418.GA12767@quad Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event_intel_rapl.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/x86/kernel/cpu/perf_event_intel_rapl.c b/arch/x86/kernel/cpu/perf_event_intel_rapl.c index 7c87424d4140..619f7699487a 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_rapl.c +++ b/arch/x86/kernel/cpu/perf_event_intel_rapl.c @@ -543,7 +543,8 @@ static int rapl_cpu_prepare(int cpu) if (phys_id < 0) return -1; - if (!rdmsrl_safe(MSR_RAPL_POWER_UNIT, &msr_rapl_power_unit_bits)) + /* protect rdmsrl() to handle virtualization */ + if (rdmsrl_safe(MSR_RAPL_POWER_UNIT, &msr_rapl_power_unit_bits)) return -1; pmu = kzalloc_node(sizeof(*pmu), GFP_KERNEL, cpu_to_node(cpu)); -- cgit v1.2.3 From 6e0de817594c61f3b392a9245deeb09609ec707d Mon Sep 17 00:00:00 2001 From: Martin Schwidefsky Date: Fri, 25 Apr 2014 10:53:44 +0200 Subject: s390/bpf,jit: initialize A register if 1st insn is BPF_S_LDX_B_MSH The A register needs to be initialized to zero in the prolog if the first instruction of the BPF program is BPF_S_LDX_B_MSH to prevent leaking the content of %r5 to user space. Cc: Signed-off-by: Martin Schwidefsky --- arch/s390/net/bpf_jit_comp.c | 1 - 1 file changed, 1 deletion(-) (limited to 'arch') diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c index 9c36dc398f90..452d3ebd9d0f 100644 --- a/arch/s390/net/bpf_jit_comp.c +++ b/arch/s390/net/bpf_jit_comp.c @@ -276,7 +276,6 @@ static void bpf_jit_noleaks(struct bpf_jit *jit, struct sock_filter *filter) case BPF_S_LD_W_IND: case BPF_S_LD_H_IND: case BPF_S_LD_B_IND: - case BPF_S_LDX_B_MSH: case BPF_S_LD_IMM: case BPF_S_LD_MEM: case BPF_S_MISC_TXA: -- cgit v1.2.3 From 4e4468fac4381b92eb333d94256e7fb8350f3de3 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Fri, 25 Apr 2014 11:46:04 +0100 Subject: ARM: KVM: disable KVM in Kconfig on big-endian systems KVM currently crashes and burns on big-endian hosts, so don't allow it to be selected until we've got that fixed. Cc: Marc Zyngier Signed-off-by: Will Deacon Signed-off-by: Christoffer Dall --- arch/arm/kvm/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/arm/kvm/Kconfig b/arch/arm/kvm/Kconfig index 466bd299b1a8..4be5bb150bdd 100644 --- a/arch/arm/kvm/Kconfig +++ b/arch/arm/kvm/Kconfig @@ -23,7 +23,7 @@ config KVM select HAVE_KVM_CPU_RELAX_INTERCEPT select KVM_MMIO select KVM_ARM_HOST - depends on ARM_VIRT_EXT && ARM_LPAE + depends on ARM_VIRT_EXT && ARM_LPAE && !CPU_BIG_ENDIAN ---help--- Support hosting virtualized guest machines. You will also need to select one or more of the processor modules below. -- cgit v1.2.3 From 28ea3c7529388c1e4177336a0e655c6c7272fd1c Mon Sep 17 00:00:00 2001 From: Jeff Mahoney Date: Sun, 27 Apr 2014 18:10:43 -0400 Subject: powerpc: Export flush_icache_range Commit aac416fc38c (lkdtm: flush icache and report actions) calls flush_icache_range from a module. It's exported on most architectures that implement it, but not on powerpc. This patch exports it to fix the module link failure. Signed-off-by: Jeff Mahoney Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/ppc_ksyms.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch') diff --git a/arch/powerpc/kernel/ppc_ksyms.c b/arch/powerpc/kernel/ppc_ksyms.c index 3bd77edd7610..450850a49dce 100644 --- a/arch/powerpc/kernel/ppc_ksyms.c +++ b/arch/powerpc/kernel/ppc_ksyms.c @@ -120,6 +120,7 @@ EXPORT_SYMBOL(giveup_spe); EXPORT_SYMBOL(flush_instruction_cache); #endif EXPORT_SYMBOL(flush_dcache_range); +EXPORT_SYMBOL(flush_icache_range); #ifdef CONFIG_SMP #ifdef CONFIG_PPC32 -- cgit v1.2.3 From 4fb8d027dca0236c811272d342cf185569d91311 Mon Sep 17 00:00:00 2001 From: Li Zhong Date: Mon, 28 Apr 2014 08:29:51 +0800 Subject: powerpc: Fix Oops in rtas_stop_self() commit 41dd03a9 may cause Oops in rtas_stop_self(). The reason is that the rtas_args was moved into stack space. For a box with more that 4GB RAM, the stack could easily be outside 32bit range, but RTAS is 32bit. So the patch moves rtas_args away from stack by adding static before it. Signed-off-by: Li Zhong Signed-off-by: Anton Blanchard Cc: stable@vger.kernel.org # 3.14+ Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/pseries/hotplug-cpu.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/platforms/pseries/hotplug-cpu.c b/arch/powerpc/platforms/pseries/hotplug-cpu.c index 9b8e05078a63..20d62975856f 100644 --- a/arch/powerpc/platforms/pseries/hotplug-cpu.c +++ b/arch/powerpc/platforms/pseries/hotplug-cpu.c @@ -88,13 +88,14 @@ void set_default_offline_state(int cpu) static void rtas_stop_self(void) { - struct rtas_args args = { - .token = cpu_to_be32(rtas_stop_self_token), + static struct rtas_args args = { .nargs = 0, .nret = 1, .rets = &args.args[0], }; + args.token = cpu_to_be32(rtas_stop_self_token); + local_irq_disable(); BUG_ON(rtas_stop_self_token == RTAS_UNKNOWN_SERVICE); -- cgit v1.2.3 From ba9a32b176c31ec4d3d7314bb2e529a85423f7a4 Mon Sep 17 00:00:00 2001 From: Joel Stanley Date: Thu, 24 Apr 2014 16:55:33 +0930 Subject: powerpc/powernv: Fix sysparam sysfs error handling When a sysparam query in OPAL returned a negative value (error code), sysfs would spew out a decent chunk of memory; almost 64K more than expected. This was traced to a sign/unsigned mix up in the OPAL sysparam sysfs code at sys_param_show. The return value of sys_param_show is a ssize_t, calculated using return ret ? ret : attr->param_size; Alan Modra explains: "attr->param_size" is an unsigned int, "ret" an int, so the overall expression has type unsigned int. Result is that ret is cast to unsigned int before being cast to ssize_t. Instead of using the ternary operator, set ret to the param_size if an error is not detected. The same bug exists in the sysfs write callback; this patch fixes it in the same way. A note on debugging this next time: on my system gcc will warn about this if compiled with -Wsign-compare, which is not enabled by -Wall, only -Wextra. Signed-off-by: Joel Stanley Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/powernv/opal-sysparam.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/platforms/powernv/opal-sysparam.c b/arch/powerpc/platforms/powernv/opal-sysparam.c index 6b614726baf2..a51ea189aec3 100644 --- a/arch/powerpc/platforms/powernv/opal-sysparam.c +++ b/arch/powerpc/platforms/powernv/opal-sysparam.c @@ -121,9 +121,10 @@ static ssize_t sys_param_show(struct kobject *kobj, memcpy(buf, param_data_buf, attr->param_size); + ret = attr->param_size; out: mutex_unlock(&opal_sysparam_mutex); - return ret ? ret : attr->param_size; + return ret; } static ssize_t sys_param_store(struct kobject *kobj, @@ -138,7 +139,9 @@ static ssize_t sys_param_store(struct kobject *kobj, ret = opal_set_sys_param(attr->param_id, attr->param_size, param_data_buf); mutex_unlock(&opal_sysparam_mutex); - return ret ? ret : count; + if (!ret) + ret = count; + return ret; } void __init opal_sys_param_init(void) -- cgit v1.2.3 From b8569d230416c490065098684adf13c1751cb830 Mon Sep 17 00:00:00 2001 From: Joel Stanley Date: Thu, 24 Apr 2014 16:55:34 +0930 Subject: powerpc/powernv: Use ssize_t for sysparam return values The OPAL calls are returning int64_t values, which the sysparam code stores in an int, and the sysfs callback returns ssize_t. Make code a easier to read by consistently using ssize_t. Signed-off-by: Joel Stanley Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/powernv/opal-sysparam.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/platforms/powernv/opal-sysparam.c b/arch/powerpc/platforms/powernv/opal-sysparam.c index a51ea189aec3..0675e1a90b40 100644 --- a/arch/powerpc/platforms/powernv/opal-sysparam.c +++ b/arch/powerpc/platforms/powernv/opal-sysparam.c @@ -39,10 +39,11 @@ struct param_attr { struct kobj_attribute kobj_attr; }; -static int opal_get_sys_param(u32 param_id, u32 length, void *buffer) +static ssize_t opal_get_sys_param(u32 param_id, u32 length, void *buffer) { struct opal_msg msg; - int ret, token; + ssize_t ret; + int token; token = opal_async_get_token_interruptible(); if (token < 0) { @@ -59,7 +60,7 @@ static int opal_get_sys_param(u32 param_id, u32 length, void *buffer) ret = opal_async_wait_response(token, &msg); if (ret) { - pr_err("%s: Failed to wait for the async response, %d\n", + pr_err("%s: Failed to wait for the async response, %zd\n", __func__, ret); goto out_token; } @@ -111,7 +112,7 @@ static ssize_t sys_param_show(struct kobject *kobj, { struct param_attr *attr = container_of(kobj_attr, struct param_attr, kobj_attr); - int ret; + ssize_t ret; mutex_lock(&opal_sysparam_mutex); ret = opal_get_sys_param(attr->param_id, attr->param_size, @@ -132,7 +133,7 @@ static ssize_t sys_param_store(struct kobject *kobj, { struct param_attr *attr = container_of(kobj_attr, struct param_attr, kobj_attr); - int ret; + ssize_t ret; mutex_lock(&opal_sysparam_mutex); memcpy(param_data_buf, buf, count); -- cgit v1.2.3 From 85390378f0a6abf3bdb10c3d504d77862a2d0ec9 Mon Sep 17 00:00:00 2001 From: Joel Stanley Date: Thu, 24 Apr 2014 16:55:35 +0930 Subject: powerpc/powernv: Check sysfs size before copying The sysparam code currently uses the userspace supplied number of bytes when memcpy()ing in to a local 64-byte buffer. Limit the maximum number of bytes by the size of the buffer. Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/powernv/opal-sysparam.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'arch') diff --git a/arch/powerpc/platforms/powernv/opal-sysparam.c b/arch/powerpc/platforms/powernv/opal-sysparam.c index 0675e1a90b40..83f5f2a1c563 100644 --- a/arch/powerpc/platforms/powernv/opal-sysparam.c +++ b/arch/powerpc/platforms/powernv/opal-sysparam.c @@ -135,6 +135,10 @@ static ssize_t sys_param_store(struct kobject *kobj, kobj_attr); ssize_t ret; + /* MAX_PARAM_DATA_LEN is sizeof(param_data_buf) */ + if (count > MAX_PARAM_DATA_LEN) + count = MAX_PARAM_DATA_LEN; + mutex_lock(&opal_sysparam_mutex); memcpy(param_data_buf, buf, count); ret = opal_set_sys_param(attr->param_id, attr->param_size, -- cgit v1.2.3 From 16003d235bdb4d1ebcecfe4915c6e6d8c95f4a0c Mon Sep 17 00:00:00 2001 From: Joel Stanley Date: Thu, 24 Apr 2014 16:55:36 +0930 Subject: powerpc/powernv: Fix typos in sysparam code Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/powernv/opal-sysparam.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/platforms/powernv/opal-sysparam.c b/arch/powerpc/platforms/powernv/opal-sysparam.c index 83f5f2a1c563..c59e77a84721 100644 --- a/arch/powerpc/platforms/powernv/opal-sysparam.c +++ b/arch/powerpc/platforms/powernv/opal-sysparam.c @@ -222,13 +222,13 @@ void __init opal_sys_param_init(void) } if (of_property_read_u32_array(sysparam, "param-len", size, count)) { - pr_err("SYSPARAM: Missing propery param-len in the DT\n"); + pr_err("SYSPARAM: Missing property param-len in the DT\n"); goto out_free_perm; } if (of_property_read_u8_array(sysparam, "param-perm", perm, count)) { - pr_err("SYSPARAM: Missing propery param-perm in the DT\n"); + pr_err("SYSPARAM: Missing property param-perm in the DT\n"); goto out_free_perm; } -- cgit v1.2.3 From 63aecfb20a6629f1ca91d7a052f87988ceb6dd53 Mon Sep 17 00:00:00 2001 From: Joel Stanley Date: Thu, 24 Apr 2014 16:55:37 +0930 Subject: powerpc/powernv: Check sysparam size before creation The size of the sysparam sysfs files is determined from the device tree at boot. However the buffer is hard coded to 64 bytes. If we encounter a parameter that is larger than 64, or miss-parse the device tree, the buffer will overflow when reading or writing to the parameter. Check it at discovery time, and if the parameter is too large, do not create a sysfs entry for it. Signed-off-by: Joel Stanley Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/powernv/opal-sysparam.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'arch') diff --git a/arch/powerpc/platforms/powernv/opal-sysparam.c b/arch/powerpc/platforms/powernv/opal-sysparam.c index c59e77a84721..d202f9bc3683 100644 --- a/arch/powerpc/platforms/powernv/opal-sysparam.c +++ b/arch/powerpc/platforms/powernv/opal-sysparam.c @@ -241,6 +241,12 @@ void __init opal_sys_param_init(void) /* For each of the parameters, populate the parameter attributes */ for (i = 0; i < count; i++) { + if (size[i] > MAX_PARAM_DATA_LEN) { + pr_warn("SYSPARAM: Not creating parameter %d as size " + "exceeds buffer length\n", i); + continue; + } + sysfs_attr_init(&attr[i].kobj_attr.attr); attr[i].param_id = id[i]; attr[i].param_size = size[i]; -- cgit v1.2.3 From 298b34d7d578c3b4325248b823f3d83a8ea1541c Mon Sep 17 00:00:00 2001 From: Benjamin Herrenschmidt Date: Thu, 24 Apr 2014 16:14:25 +1000 Subject: powerpc/powernv: Fix kexec races going back to OPAL We have a subtle race when sending CPUs back to OPAL on kexec. We mark them as "in real mode" right before we send them down. Once we've booted the new kernel, it might try to call opal_reinit_cpus() to change endianness, and that requires all CPUs to be spinning inside OPAL. However there is no synchronization here and we've observed cases where the returning CPUs hadn't established their new state inside OPAL before opal_reinit_cpus() is called, causing it to fail. The proper fix is to actually wait for them to go down all the way from the kexec'ing kernel. Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/powernv/setup.c | 48 ++++++++++++++++++++++++++++++++-- 1 file changed, 46 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/platforms/powernv/setup.c b/arch/powerpc/platforms/powernv/setup.c index 61cf8fa9c61b..8723d32632f5 100644 --- a/arch/powerpc/platforms/powernv/setup.c +++ b/arch/powerpc/platforms/powernv/setup.c @@ -162,18 +162,62 @@ static void pnv_shutdown(void) } #ifdef CONFIG_KEXEC +static void pnv_kexec_wait_secondaries_down(void) +{ + int my_cpu, i, notified = -1; + + my_cpu = get_cpu(); + + for_each_online_cpu(i) { + uint8_t status; + int64_t rc; + + if (i == my_cpu) + continue; + + for (;;) { + rc = opal_query_cpu_status(get_hard_smp_processor_id(i), + &status); + if (rc != OPAL_SUCCESS || status != OPAL_THREAD_STARTED) + break; + barrier(); + if (i != notified) { + printk(KERN_INFO "kexec: waiting for cpu %d " + "(physical %d) to enter OPAL\n", + i, paca[i].hw_cpu_id); + notified = i; + } + } + } +} + static void pnv_kexec_cpu_down(int crash_shutdown, int secondary) { xics_kexec_teardown_cpu(secondary); - /* Return secondary CPUs to firmware on OPAL v3 */ - if (firmware_has_feature(FW_FEATURE_OPALv3) && secondary) { + /* On OPAL v3, we return all CPUs to firmware */ + + if (!firmware_has_feature(FW_FEATURE_OPALv3)) + return; + + if (secondary) { + /* Return secondary CPUs to firmware on OPAL v3 */ mb(); get_paca()->kexec_state = KEXEC_STATE_REAL_MODE; mb(); /* Return the CPU to OPAL */ opal_return_cpu(); + } else if (crash_shutdown) { + /* + * On crash, we don't wait for secondaries to go + * down as they might be unreachable or hung, so + * instead we just wait a bit and move on. + */ + mdelay(1); + } else { + /* Primary waits for the secondaries to have reached OPAL */ + pnv_kexec_wait_secondaries_down(); } } #endif /* CONFIG_KEXEC */ -- cgit v1.2.3 From cc146d1db0673396429ce4f906350d684ef4ca2a Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Thu, 24 Apr 2014 07:25:34 +1000 Subject: powerpc/powernv: Fix little endian issues in OPAL flash code With this patch I was able to update firmware on an LE kernel. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/powernv/opal-flash.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/platforms/powernv/opal-flash.c b/arch/powerpc/platforms/powernv/opal-flash.c index 714ef972406b..a968fd1888b3 100644 --- a/arch/powerpc/platforms/powernv/opal-flash.c +++ b/arch/powerpc/platforms/powernv/opal-flash.c @@ -131,11 +131,15 @@ static DEFINE_MUTEX(image_data_mutex); */ static inline void opal_flash_validate(void) { - struct validate_flash_t *args_buf = &validate_flash_data; + long ret; + void *buf = validate_flash_data.buf; + __be32 size, result; + + ret = opal_validate_flash(__pa(buf), &size, &result); - args_buf->status = opal_validate_flash(__pa(args_buf->buf), - &(args_buf->buf_size), - &(args_buf->result)); + validate_flash_data.status = ret; + validate_flash_data.buf_size = be32_to_cpu(size); + validate_flash_data.result = be32_to_cpu(result); } /* -- cgit v1.2.3 From 3f28c5af3964c11e61e9a58df77cae5ebdb8209e Mon Sep 17 00:00:00 2001 From: Wei Yang Date: Wed, 23 Apr 2014 10:26:32 +0800 Subject: powerpc/powernv: Reduce multi-hit of iommu_add_device() During the EEH hotplug event, iommu_add_device() will be invoked three times and two of them will trigger warning or error. The three times to invoke the iommu_add_device() are: pci_device_add ... set_iommu_table_base_and_group <- 1st time, fail device_add ... tce_iommu_bus_notifier <- 2nd time, succees pcibios_add_pci_devices ... pcibios_setup_bus_devices <- 3rd time, re-attach The first time fails, since the dev->kobj->sd is not initialized. The dev->kobj->sd is initialized in device_add(). The third time's warning is triggered by the re-attach of the iommu_group. After applying this patch, the error iommu_tce: 0003:05:00.0 has not been added, ret=-14 and the warning [ 204.123609] ------------[ cut here ]------------ [ 204.123645] WARNING: at arch/powerpc/kernel/iommu.c:1125 [ 204.123680] Modules linked in: xt_CHECKSUM nf_conntrack_netbios_ns nf_conntrack_broadcast ipt_MASQUERADE ip6t_REJECT bnep bluetooth 6lowpan_iphc rfkill xt_conntrack ebtable_nat ebtable_broute bridge stp llc mlx4_ib ib_sa ib_mad ib_core ib_addr ebtable_filter ebtables ip6table_nat nf_conntrack_ipv6 nf_defrag_ipv6 nf_nat_ipv6 ip6table_mangle ip6table_security ip6table_raw ip6table_filter ip6_tables iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack iptable_mangle iptable_security iptable_raw bnx2x tg3 mlx4_core nfsd ptp mdio ses libcrc32c nfs_acl enclosure be2net pps_core shpchp lockd kvm uinput sunrpc binfmt_misc lpfc scsi_transport_fc ipr scsi_tgt [ 204.124356] CPU: 18 PID: 650 Comm: eehd Not tainted 3.14.0-rc5yw+ #102 [ 204.124400] task: c0000027ed485670 ti: c0000027ed50c000 task.ti: c0000027ed50c000 [ 204.124453] NIP: c00000000003cf80 LR: c00000000006c648 CTR: c00000000006c5c0 [ 204.124506] REGS: c0000027ed50f440 TRAP: 0700 Not tainted (3.14.0-rc5yw+) [ 204.124558] MSR: 9000000000029032 CR: 88008084 XER: 20000000 [ 204.124682] CFAR: c00000000006c644 SOFTE: 1 GPR00: c00000000006c648 c0000027ed50f6c0 c000000001398380 c0000027ec260300 GPR04: c0000027ea92c000 c00000000006ad00 c0000000016e41b0 0000000000000110 GPR08: c0000000012cd4c0 0000000000000001 c0000027ec2602ff 0000000000000062 GPR12: 0000000028008084 c00000000fdca200 c0000000000d1d90 c0000027ec281a80 GPR16: 0000000000000000 0000000000000000 0000000000000000 0000000000000000 GPR20: 0000000000000000 0000000000000000 0000000000000000 0000000000000001 GPR24: 000000005342697b 0000000000002906 c000001fe6ac9800 c000001fe6ac9800 GPR28: 0000000000000000 c0000000016e3a80 c0000027ea92c090 c0000027ea92c000 [ 204.125353] NIP [c00000000003cf80] .iommu_add_device+0x30/0x1f0 [ 204.125399] LR [c00000000006c648] .pnv_pci_ioda_dma_dev_setup+0x88/0xb0 [ 204.125443] Call Trace: [ 204.125464] [c0000027ed50f6c0] [c0000027ed50f750] 0xc0000027ed50f750 (unreliable) [ 204.125526] [c0000027ed50f750] [c00000000006c648] .pnv_pci_ioda_dma_dev_setup+0x88/0xb0 [ 204.125588] [c0000027ed50f7d0] [c000000000069cc8] .pnv_pci_dma_dev_setup+0x78/0x340 [ 204.125650] [c0000027ed50f870] [c000000000044408] .pcibios_setup_device+0x88/0x2f0 [ 204.125712] [c0000027ed50f940] [c000000000046040] .pcibios_setup_bus_devices+0x60/0xd0 [ 204.125774] [c0000027ed50f9c0] [c000000000043acc] .pcibios_add_pci_devices+0xdc/0x1c0 [ 204.125837] [c0000027ed50fa50] [c00000000086f970] .eeh_reset_device+0x36c/0x4f0 [ 204.125939] [c0000027ed50fb20] [c00000000003a2d8] .eeh_handle_normal_event+0x448/0x480 [ 204.126068] [c0000027ed50fbc0] [c00000000003a35c] .eeh_handle_event+0x4c/0x340 [ 204.126192] [c0000027ed50fc80] [c00000000003a74c] .eeh_event_handler+0xfc/0x1b0 [ 204.126319] [c0000027ed50fd30] [c0000000000d1ea0] .kthread+0x110/0x130 [ 204.126430] [c0000027ed50fe30] [c00000000000a460] .ret_from_kernel_thread+0x5c/0x7c [ 204.126556] Instruction dump: [ 204.126610] 7c0802a6 fba1ffe8 fbc1fff0 fbe1fff8 f8010010 f821ff71 7c7e1b78 60000000 [ 204.126787] 60000000 e87e0298 3143ffff 7d2a1910 <0b090000> 2fa90000 40de00c8 ebfe0218 [ 204.126966] ---[ end trace 6e7aefd80add2973 ]--- are cleared. This patch removes iommu_add_device() in pnv_pci_ioda_dma_dev_setup(), which revert part of the change in commit d905c5df(PPC: POWERNV: move iommu_add_device earlier). Signed-off-by: Wei Yang Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/powernv/pci-ioda.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c index 3b2b4fb3585b..1d992f137c48 100644 --- a/arch/powerpc/platforms/powernv/pci-ioda.c +++ b/arch/powerpc/platforms/powernv/pci-ioda.c @@ -462,7 +462,7 @@ static void pnv_pci_ioda_dma_dev_setup(struct pnv_phb *phb, struct pci_dev *pdev pe = &phb->ioda.pe_array[pdn->pe_number]; WARN_ON(get_dma_ops(&pdev->dev) != &dma_iommu_ops); - set_iommu_table_base_and_group(&pdev->dev, &pe->tce32_table); + set_iommu_table_base(&pdev->dev, &pe->tce32_table); } static int pnv_pci_ioda_dma_set_mask(struct pnv_phb *phb, -- cgit v1.2.3 From 4966bfa1b3347ee75e6d93859a2e8ce9a662390c Mon Sep 17 00:00:00 2001 From: Wei Yang Date: Wed, 23 Apr 2014 10:26:33 +0800 Subject: powerpc/powernv: Release the refcount for pci_dev On PowerNV platform, we are holding an unnecessary refcount on a pci_dev, which leads to the pci_dev is not destroyed when hotplugging a pci device. This patch release the unnecessary refcount. Signed-off-by: Wei Yang Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/powernv/pci-ioda.c | 1 - 1 file changed, 1 deletion(-) (limited to 'arch') diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c index 1d992f137c48..98824aa99173 100644 --- a/arch/powerpc/platforms/powernv/pci-ioda.c +++ b/arch/powerpc/platforms/powernv/pci-ioda.c @@ -343,7 +343,6 @@ static void pnv_ioda_setup_same_PE(struct pci_bus *bus, struct pnv_ioda_pe *pe) pci_name(dev)); continue; } - pci_dev_get(dev); pdn->pcidev = dev; pdn->pe_number = pe->pe_number; pe->dma_weight += pnv_ioda_dma_weight(dev); -- cgit v1.2.3 From 2bad742388d23c15f2cb355bb992046e30065a11 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Tue, 22 Apr 2014 15:01:22 +1000 Subject: powerpc/powernv: Use uint64_t instead of size_t in OPAL APIs Using size_t in our APIs is asking for trouble, especially when some OPAL calls use size_t pointers. Signed-off-by: Anton Blanchard Reviewed-by: Stewart Smith Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/include/asm/opal.h | 12 ++++++------ arch/powerpc/platforms/powernv/opal-elog.c | 2 +- 2 files changed, 7 insertions(+), 7 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/include/asm/opal.h b/arch/powerpc/include/asm/opal.h index a2efdaa020b0..2a34485f7e0a 100644 --- a/arch/powerpc/include/asm/opal.h +++ b/arch/powerpc/include/asm/opal.h @@ -858,8 +858,8 @@ int64_t opal_lpc_write(uint32_t chip_id, enum OpalLPCAddressType addr_type, int64_t opal_lpc_read(uint32_t chip_id, enum OpalLPCAddressType addr_type, uint32_t addr, __be32 *data, uint32_t sz); -int64_t opal_read_elog(uint64_t buffer, size_t size, uint64_t log_id); -int64_t opal_get_elog_size(uint64_t *log_id, size_t *size, uint64_t *elog_type); +int64_t opal_read_elog(uint64_t buffer, uint64_t size, uint64_t log_id); +int64_t opal_get_elog_size(uint64_t *log_id, uint64_t *size, uint64_t *elog_type); int64_t opal_write_elog(uint64_t buffer, uint64_t size, uint64_t offset); int64_t opal_send_ack_elog(uint64_t log_id); void opal_resend_pending_logs(void); @@ -874,13 +874,13 @@ int64_t opal_dump_read(uint32_t dump_id, uint64_t buffer); int64_t opal_dump_ack(uint32_t dump_id); int64_t opal_dump_resend_notification(void); -int64_t opal_get_msg(uint64_t buffer, size_t size); -int64_t opal_check_completion(uint64_t buffer, size_t size, uint64_t token); +int64_t opal_get_msg(uint64_t buffer, uint64_t size); +int64_t opal_check_completion(uint64_t buffer, uint64_t size, uint64_t token); int64_t opal_sync_host_reboot(void); int64_t opal_get_param(uint64_t token, uint32_t param_id, uint64_t buffer, - size_t length); + uint64_t length); int64_t opal_set_param(uint64_t token, uint32_t param_id, uint64_t buffer, - size_t length); + uint64_t length); int64_t opal_sensor_read(uint32_t sensor_hndl, int token, __be32 *sensor_data); /* Internal functions */ diff --git a/arch/powerpc/platforms/powernv/opal-elog.c b/arch/powerpc/platforms/powernv/opal-elog.c index ef7bc2a97862..7e3821e611a8 100644 --- a/arch/powerpc/platforms/powernv/opal-elog.c +++ b/arch/powerpc/platforms/powernv/opal-elog.c @@ -238,7 +238,7 @@ static struct elog_obj *create_elog_obj(uint64_t id, size_t size, uint64_t type) static void elog_work_fn(struct work_struct *work) { - size_t elog_size; + uint64_t elog_size; uint64_t log_id; uint64_t elog_type; int rc; -- cgit v1.2.3 From e2c8b93e65935498d9bd419cec900221ace70c5b Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Tue, 22 Apr 2014 15:01:23 +1000 Subject: powerpc/powernv: Remove some OPAL function declaration duplication We had some duplication of the internal OPAL functions. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/include/asm/opal.h | 12 ++---------- 1 file changed, 2 insertions(+), 10 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/include/asm/opal.h b/arch/powerpc/include/asm/opal.h index 2a34485f7e0a..cb7d52ed86d2 100644 --- a/arch/powerpc/include/asm/opal.h +++ b/arch/powerpc/include/asm/opal.h @@ -884,7 +884,8 @@ int64_t opal_set_param(uint64_t token, uint32_t param_id, uint64_t buffer, int64_t opal_sensor_read(uint32_t sensor_hndl, int token, __be32 *sensor_data); /* Internal functions */ -extern int early_init_dt_scan_opal(unsigned long node, const char *uname, int depth, void *data); +extern int early_init_dt_scan_opal(unsigned long node, const char *uname, + int depth, void *data); extern int early_init_dt_scan_recoverable_ranges(unsigned long node, const char *uname, int depth, void *data); @@ -893,10 +894,6 @@ extern int opal_put_chars(uint32_t vtermno, const char *buf, int total_len); extern void hvc_opal_init_early(void); -/* Internal functions */ -extern int early_init_dt_scan_opal(unsigned long node, const char *uname, - int depth, void *data); - extern int opal_notifier_register(struct notifier_block *nb); extern int opal_notifier_unregister(struct notifier_block *nb); @@ -906,9 +903,6 @@ extern void opal_notifier_enable(void); extern void opal_notifier_disable(void); extern void opal_notifier_update_evt(uint64_t evt_mask, uint64_t evt_val); -extern int opal_get_chars(uint32_t vtermno, char *buf, int count); -extern int opal_put_chars(uint32_t vtermno, const char *buf, int total_len); - extern int __opal_async_get_token(void); extern int opal_async_get_token_interruptible(void); extern int __opal_async_release_token(int token); @@ -916,8 +910,6 @@ extern int opal_async_release_token(int token); extern int opal_async_wait_response(uint64_t token, struct opal_msg *msg); extern int opal_get_sensor_data(u32 sensor_hndl, u32 *sensor_data); -extern void hvc_opal_init_early(void); - struct rtc_time; extern int opal_set_rtc_time(struct rtc_time *tm); extern void opal_get_rtc_time(struct rtc_time *tm); -- cgit v1.2.3 From 56b4c993124d15f3b198cc757ba49a0022b5d695 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Tue, 22 Apr 2014 15:01:24 +1000 Subject: powerpc/powernv: Fix little endian issues with opal_do_notifier calls The bitmap in opal_poll_events and opal_handle_interrupt is big endian, so we need to byteswap it on little endian builds. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/powernv/opal.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/platforms/powernv/opal.c b/arch/powerpc/platforms/powernv/opal.c index 49d2f00019e5..17cfc70082aa 100644 --- a/arch/powerpc/platforms/powernv/opal.c +++ b/arch/powerpc/platforms/powernv/opal.c @@ -242,14 +242,14 @@ void opal_notifier_update_evt(uint64_t evt_mask, void opal_notifier_enable(void) { int64_t rc; - uint64_t evt = 0; + __be64 evt = 0; atomic_set(&opal_notifier_hold, 0); /* Process pending events */ rc = opal_poll_events(&evt); if (rc == OPAL_SUCCESS && evt) - opal_do_notifier(evt); + opal_do_notifier(be64_to_cpu(evt)); } void opal_notifier_disable(void) @@ -529,7 +529,7 @@ static irqreturn_t opal_interrupt(int irq, void *data) opal_handle_interrupt(virq_to_hw(irq), &events); - opal_do_notifier(events); + opal_do_notifier(be64_to_cpu(events)); return IRQ_HANDLED; } -- cgit v1.2.3 From 14ad0c58d5df6e5911a5413abdc2a9be6a8acb51 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Tue, 22 Apr 2014 15:01:25 +1000 Subject: powerpc/powernv: Fix little endian issues in OPAL error log code Fix little endian issues with the OPAL error log code. Signed-off-by: Anton Blanchard Reviewed-by: Stewart Smith Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/include/asm/opal.h | 2 +- arch/powerpc/platforms/powernv/opal-elog.c | 9 ++++++++- 2 files changed, 9 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/include/asm/opal.h b/arch/powerpc/include/asm/opal.h index cb7d52ed86d2..1a752ac8c0ba 100644 --- a/arch/powerpc/include/asm/opal.h +++ b/arch/powerpc/include/asm/opal.h @@ -859,7 +859,7 @@ int64_t opal_lpc_read(uint32_t chip_id, enum OpalLPCAddressType addr_type, uint32_t addr, __be32 *data, uint32_t sz); int64_t opal_read_elog(uint64_t buffer, uint64_t size, uint64_t log_id); -int64_t opal_get_elog_size(uint64_t *log_id, uint64_t *size, uint64_t *elog_type); +int64_t opal_get_elog_size(__be64 *log_id, __be64 *size, __be64 *elog_type); int64_t opal_write_elog(uint64_t buffer, uint64_t size, uint64_t offset); int64_t opal_send_ack_elog(uint64_t log_id); void opal_resend_pending_logs(void); diff --git a/arch/powerpc/platforms/powernv/opal-elog.c b/arch/powerpc/platforms/powernv/opal-elog.c index 7e3821e611a8..10268c41d830 100644 --- a/arch/powerpc/platforms/powernv/opal-elog.c +++ b/arch/powerpc/platforms/powernv/opal-elog.c @@ -238,18 +238,25 @@ static struct elog_obj *create_elog_obj(uint64_t id, size_t size, uint64_t type) static void elog_work_fn(struct work_struct *work) { + __be64 size; + __be64 id; + __be64 type; uint64_t elog_size; uint64_t log_id; uint64_t elog_type; int rc; char name[2+16+1]; - rc = opal_get_elog_size(&log_id, &elog_size, &elog_type); + rc = opal_get_elog_size(&id, &size, &type); if (rc != OPAL_SUCCESS) { pr_err("ELOG: Opal log read failed\n"); return; } + elog_size = be64_to_cpu(size); + log_id = be64_to_cpu(id); + elog_type = be64_to_cpu(type); + BUG_ON(elog_size > OPAL_MAX_ERRLOG_SIZE); if (elog_size >= OPAL_MAX_ERRLOG_SIZE) -- cgit v1.2.3 From 3441f04b4b62758a798f9fbbf2047dfedf0329a5 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Tue, 22 Apr 2014 15:01:26 +1000 Subject: powerpc/powernv: Create OPAL sglist helper functions and fix endian issues We have two copies of code that creates an OPAL sg list. Consolidate these into a common set of helpers and fix the endian issues. The flash interface embedded a version number in the num_entries field, whereas the dump interface did did not. Since versioning wasn't added to the flash interface and it is impossible to add this in a backwards compatible way, just remove it. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/include/asm/opal.h | 14 ++-- arch/powerpc/platforms/powernv/opal-dump.c | 81 +-------------------- arch/powerpc/platforms/powernv/opal-flash.c | 106 +--------------------------- arch/powerpc/platforms/powernv/opal.c | 63 +++++++++++++++++ 4 files changed, 76 insertions(+), 188 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/include/asm/opal.h b/arch/powerpc/include/asm/opal.h index 1a752ac8c0ba..afb0fedd1214 100644 --- a/arch/powerpc/include/asm/opal.h +++ b/arch/powerpc/include/asm/opal.h @@ -41,14 +41,14 @@ struct opal_takeover_args { * size except the last one in the list to be as well. */ struct opal_sg_entry { - void *data; - long length; + __be64 data; + __be64 length; }; -/* sg list */ +/* SG list */ struct opal_sg_list { - unsigned long num_entries; - struct opal_sg_list *next; + __be64 length; + __be64 next; struct opal_sg_entry entry[]; }; @@ -929,6 +929,10 @@ extern int opal_resync_timebase(void); extern void opal_lpc_init(void); +struct opal_sg_list *opal_vmalloc_to_sg_list(void *vmalloc_addr, + unsigned long vmalloc_size); +void opal_free_sg_list(struct opal_sg_list *sg); + #endif /* __ASSEMBLY__ */ #endif /* __OPAL_H */ diff --git a/arch/powerpc/platforms/powernv/opal-dump.c b/arch/powerpc/platforms/powernv/opal-dump.c index b9827b0d87e4..f0b472427c57 100644 --- a/arch/powerpc/platforms/powernv/opal-dump.c +++ b/arch/powerpc/platforms/powernv/opal-dump.c @@ -209,80 +209,6 @@ static struct kobj_type dump_ktype = { .default_attrs = dump_default_attrs, }; -static void free_dump_sg_list(struct opal_sg_list *list) -{ - struct opal_sg_list *sg1; - while (list) { - sg1 = list->next; - kfree(list); - list = sg1; - } - list = NULL; -} - -static struct opal_sg_list *dump_data_to_sglist(struct dump_obj *dump) -{ - struct opal_sg_list *sg1, *list = NULL; - void *addr; - int64_t size; - - addr = dump->buffer; - size = dump->size; - - sg1 = kzalloc(PAGE_SIZE, GFP_KERNEL); - if (!sg1) - goto nomem; - - list = sg1; - sg1->num_entries = 0; - while (size > 0) { - /* Translate virtual address to physical address */ - sg1->entry[sg1->num_entries].data = - (void *)(vmalloc_to_pfn(addr) << PAGE_SHIFT); - - if (size > PAGE_SIZE) - sg1->entry[sg1->num_entries].length = PAGE_SIZE; - else - sg1->entry[sg1->num_entries].length = size; - - sg1->num_entries++; - if (sg1->num_entries >= SG_ENTRIES_PER_NODE) { - sg1->next = kzalloc(PAGE_SIZE, GFP_KERNEL); - if (!sg1->next) - goto nomem; - - sg1 = sg1->next; - sg1->num_entries = 0; - } - addr += PAGE_SIZE; - size -= PAGE_SIZE; - } - return list; - -nomem: - pr_err("%s : Failed to allocate memory\n", __func__); - free_dump_sg_list(list); - return NULL; -} - -static void sglist_to_phy_addr(struct opal_sg_list *list) -{ - struct opal_sg_list *sg, *next; - - for (sg = list; sg; sg = next) { - next = sg->next; - /* Don't translate NULL pointer for last entry */ - if (sg->next) - sg->next = (struct opal_sg_list *)__pa(sg->next); - else - sg->next = NULL; - - /* Convert num_entries to length */ - sg->num_entries = - sg->num_entries * sizeof(struct opal_sg_entry) + 16; - } -} - static int64_t dump_read_info(uint32_t *id, uint32_t *size, uint32_t *type) { int rc; @@ -314,15 +240,12 @@ static int64_t dump_read_data(struct dump_obj *dump) } /* Generate SG list */ - list = dump_data_to_sglist(dump); + list = opal_vmalloc_to_sg_list(dump->buffer, dump->size); if (!list) { rc = -ENOMEM; goto out; } - /* Translate sg list addr to real address */ - sglist_to_phy_addr(list); - /* First entry address */ addr = __pa(list); @@ -341,7 +264,7 @@ static int64_t dump_read_data(struct dump_obj *dump) __func__, dump->id); /* Free SG list */ - free_dump_sg_list(list); + opal_free_sg_list(list); out: return rc; diff --git a/arch/powerpc/platforms/powernv/opal-flash.c b/arch/powerpc/platforms/powernv/opal-flash.c index a968fd1888b3..dc487ff04704 100644 --- a/arch/powerpc/platforms/powernv/opal-flash.c +++ b/arch/powerpc/platforms/powernv/opal-flash.c @@ -79,9 +79,6 @@ /* XXX: Assume candidate image size is <= 1GB */ #define MAX_IMAGE_SIZE 0x40000000 -/* Flash sg list version */ -#define SG_LIST_VERSION (1UL) - /* Image status */ enum { IMAGE_INVALID, @@ -271,94 +268,12 @@ static ssize_t manage_store(struct kobject *kobj, return count; } -/* - * Free sg list - */ -static void free_sg_list(struct opal_sg_list *list) -{ - struct opal_sg_list *sg1; - while (list) { - sg1 = list->next; - kfree(list); - list = sg1; - } - list = NULL; -} - -/* - * Build candidate image scatter gather list - * - * list format: - * ----------------------------------- - * | VER (8) | Entry length in bytes | - * ----------------------------------- - * | Pointer to next entry | - * ----------------------------------- - * | Address of memory area 1 | - * ----------------------------------- - * | Length of memory area 1 | - * ----------------------------------- - * | ......... | - * ----------------------------------- - * | ......... | - * ----------------------------------- - * | Address of memory area N | - * ----------------------------------- - * | Length of memory area N | - * ----------------------------------- - */ -static struct opal_sg_list *image_data_to_sglist(void) -{ - struct opal_sg_list *sg1, *list = NULL; - void *addr; - int size; - - addr = image_data.data; - size = image_data.size; - - sg1 = kzalloc(PAGE_SIZE, GFP_KERNEL); - if (!sg1) - return NULL; - - list = sg1; - sg1->num_entries = 0; - while (size > 0) { - /* Translate virtual address to physical address */ - sg1->entry[sg1->num_entries].data = - (void *)(vmalloc_to_pfn(addr) << PAGE_SHIFT); - - if (size > PAGE_SIZE) - sg1->entry[sg1->num_entries].length = PAGE_SIZE; - else - sg1->entry[sg1->num_entries].length = size; - - sg1->num_entries++; - if (sg1->num_entries >= SG_ENTRIES_PER_NODE) { - sg1->next = kzalloc(PAGE_SIZE, GFP_KERNEL); - if (!sg1->next) { - pr_err("%s : Failed to allocate memory\n", - __func__); - goto nomem; - } - - sg1 = sg1->next; - sg1->num_entries = 0; - } - addr += PAGE_SIZE; - size -= PAGE_SIZE; - } - return list; -nomem: - free_sg_list(list); - return NULL; -} - /* * OPAL update flash */ static int opal_flash_update(int op) { - struct opal_sg_list *sg, *list, *next; + struct opal_sg_list *list; unsigned long addr; int64_t rc = OPAL_PARAMETER; @@ -368,30 +283,13 @@ static int opal_flash_update(int op) goto flash; } - list = image_data_to_sglist(); + list = opal_vmalloc_to_sg_list(image_data.data, image_data.size); if (!list) goto invalid_img; /* First entry address */ addr = __pa(list); - /* Translate sg list address to absolute */ - for (sg = list; sg; sg = next) { - next = sg->next; - /* Don't translate NULL pointer for last entry */ - if (sg->next) - sg->next = (struct opal_sg_list *)__pa(sg->next); - else - sg->next = NULL; - - /* - * Convert num_entries to version/length format - * to satisfy OPAL. - */ - sg->num_entries = (SG_LIST_VERSION << 56) | - (sg->num_entries * sizeof(struct opal_sg_entry) + 16); - } - pr_alert("FLASH: Image is %u bytes\n", image_data.size); pr_alert("FLASH: Image update requested\n"); pr_alert("FLASH: Image will be updated during system reboot\n"); diff --git a/arch/powerpc/platforms/powernv/opal.c b/arch/powerpc/platforms/powernv/opal.c index 17cfc70082aa..360ad80c754c 100644 --- a/arch/powerpc/platforms/powernv/opal.c +++ b/arch/powerpc/platforms/powernv/opal.c @@ -638,3 +638,66 @@ void opal_shutdown(void) /* Export this so that test modules can use it */ EXPORT_SYMBOL_GPL(opal_invalid_call); + +/* Convert a region of vmalloc memory to an opal sg list */ +struct opal_sg_list *opal_vmalloc_to_sg_list(void *vmalloc_addr, + unsigned long vmalloc_size) +{ + struct opal_sg_list *sg, *first = NULL; + unsigned long i = 0; + + sg = kzalloc(PAGE_SIZE, GFP_KERNEL); + if (!sg) + goto nomem; + + first = sg; + + while (vmalloc_size > 0) { + uint64_t data = vmalloc_to_pfn(vmalloc_addr) << PAGE_SHIFT; + uint64_t length = min(vmalloc_size, PAGE_SIZE); + + sg->entry[i].data = cpu_to_be64(data); + sg->entry[i].length = cpu_to_be64(length); + i++; + + if (i >= SG_ENTRIES_PER_NODE) { + struct opal_sg_list *next; + + next = kzalloc(PAGE_SIZE, GFP_KERNEL); + if (!next) + goto nomem; + + sg->length = cpu_to_be64( + i * sizeof(struct opal_sg_entry) + 16); + i = 0; + sg->next = cpu_to_be64(__pa(next)); + sg = next; + } + + vmalloc_addr += length; + vmalloc_size -= length; + } + + sg->length = cpu_to_be64(i * sizeof(struct opal_sg_entry) + 16); + + return first; + +nomem: + pr_err("%s : Failed to allocate memory\n", __func__); + opal_free_sg_list(first); + return NULL; +} + +void opal_free_sg_list(struct opal_sg_list *sg) +{ + while (sg) { + uint64_t next = be64_to_cpu(sg->next); + + kfree(sg); + + if (next) + sg = __va(next); + else + sg = NULL; + } +} -- cgit v1.2.3 From 2d6b63bbdd7a649ea17fae19f8bab07d4b292a09 Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Tue, 22 Apr 2014 15:01:27 +1000 Subject: powerpc/powernv: Fix little endian issues in OPAL dump code Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/include/asm/opal.h | 4 ++-- arch/powerpc/platforms/powernv/opal-dump.c | 13 +++++++++---- 2 files changed, 11 insertions(+), 6 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/include/asm/opal.h b/arch/powerpc/include/asm/opal.h index afb0fedd1214..66ad7a74116f 100644 --- a/arch/powerpc/include/asm/opal.h +++ b/arch/powerpc/include/asm/opal.h @@ -868,8 +868,8 @@ int64_t opal_validate_flash(uint64_t buffer, uint32_t *size, uint32_t *result); int64_t opal_manage_flash(uint8_t op); int64_t opal_update_flash(uint64_t blk_list); int64_t opal_dump_init(uint8_t dump_type); -int64_t opal_dump_info(uint32_t *dump_id, uint32_t *dump_size); -int64_t opal_dump_info2(uint32_t *dump_id, uint32_t *dump_size, uint32_t *dump_type); +int64_t opal_dump_info(__be32 *dump_id, __be32 *dump_size); +int64_t opal_dump_info2(__be32 *dump_id, __be32 *dump_size, __be32 *dump_type); int64_t opal_dump_read(uint32_t dump_id, uint64_t buffer); int64_t opal_dump_ack(uint32_t dump_id); int64_t opal_dump_resend_notification(void); diff --git a/arch/powerpc/platforms/powernv/opal-dump.c b/arch/powerpc/platforms/powernv/opal-dump.c index f0b472427c57..788a1977b9a5 100644 --- a/arch/powerpc/platforms/powernv/opal-dump.c +++ b/arch/powerpc/platforms/powernv/opal-dump.c @@ -209,15 +209,20 @@ static struct kobj_type dump_ktype = { .default_attrs = dump_default_attrs, }; -static int64_t dump_read_info(uint32_t *id, uint32_t *size, uint32_t *type) +static int64_t dump_read_info(uint32_t *dump_id, uint32_t *dump_size, uint32_t *dump_type) { + __be32 id, size, type; int rc; - *type = 0xffffffff; - rc = opal_dump_info2(id, size, type); + type = cpu_to_be32(0xffffffff); + rc = opal_dump_info2(&id, &size, &type); if (rc == OPAL_PARAMETER) - rc = opal_dump_info(id, size); + rc = opal_dump_info(&id, &size); + + *dump_id = be32_to_cpu(id); + *dump_size = be32_to_cpu(size); + *dump_type = be32_to_cpu(type); if (rc) pr_warn("%s: Failed to get dump info (%d)\n", -- cgit v1.2.3 From 29ef7a3e26283606f181000b11e5fd37d73ab98f Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Mon, 21 Apr 2014 10:37:36 +0530 Subject: powerpc/mm: Fix tlbie to add AVAL fields for 64K pages The if condition check was based on a draft ISA doc. Remove the same. Signed-off-by: Aneesh Kumar K.V Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/mm/hash_native_64.c | 38 ++++++++++++++++---------------------- 1 file changed, 16 insertions(+), 22 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/mm/hash_native_64.c b/arch/powerpc/mm/hash_native_64.c index 3ea26c25590b..cf1d325eae8b 100644 --- a/arch/powerpc/mm/hash_native_64.c +++ b/arch/powerpc/mm/hash_native_64.c @@ -82,17 +82,14 @@ static inline void __tlbie(unsigned long vpn, int psize, int apsize, int ssize) va &= ~((1ul << mmu_psize_defs[apsize].shift) - 1); va |= penc << 12; va |= ssize << 8; - /* Add AVAL part */ - if (psize != apsize) { - /* - * MPSS, 64K base page size and 16MB parge page size - * We don't need all the bits, but rest of the bits - * must be ignored by the processor. - * vpn cover upto 65 bits of va. (0...65) and we need - * 58..64 bits of va. - */ - va |= (vpn & 0xfe); - } + /* + * AVAL bits: + * We don't need all the bits, but rest of the bits + * must be ignored by the processor. + * vpn cover upto 65 bits of va. (0...65) and we need + * 58..64 bits of va. + */ + va |= (vpn & 0xfe); /* AVAL */ va |= 1; /* L */ asm volatile(ASM_FTR_IFCLR("tlbie %0,1", PPC_TLBIE(%1,%0), %2) : : "r" (va), "r"(0), "i" (CPU_FTR_ARCH_206) @@ -133,17 +130,14 @@ static inline void __tlbiel(unsigned long vpn, int psize, int apsize, int ssize) va &= ~((1ul << mmu_psize_defs[apsize].shift) - 1); va |= penc << 12; va |= ssize << 8; - /* Add AVAL part */ - if (psize != apsize) { - /* - * MPSS, 64K base page size and 16MB parge page size - * We don't need all the bits, but rest of the bits - * must be ignored by the processor. - * vpn cover upto 65 bits of va. (0...65) and we need - * 58..64 bits of va. - */ - va |= (vpn & 0xfe); - } + /* + * AVAL bits: + * We don't need all the bits, but rest of the bits + * must be ignored by the processor. + * vpn cover upto 65 bits of va. (0...65) and we need + * 58..64 bits of va. + */ + va |= (vpn & 0xfe); va |= 1; /* L */ asm volatile(".long 0x7c000224 | (%0 << 11) | (1 << 21)" : : "r"(va) : "memory"); -- cgit v1.2.3 From e98bf005d5feafb3a089868420c1f3743d304e14 Mon Sep 17 00:00:00 2001 From: Cody P Schafer Date: Tue, 15 Apr 2014 10:10:50 -0700 Subject: powerpc/perf/hv_24x7: Probe errors changed to pr_debug(), padding fixed fixup for "powerpc/perf: Add support for the hv 24x7 interface" Makes the "not enabled" message less awful (and hides it in most cases). Signed-off-by: Cody P Schafer Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/perf/hv-24x7.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/perf/hv-24x7.c b/arch/powerpc/perf/hv-24x7.c index 297c91051413..f5bca73d92b8 100644 --- a/arch/powerpc/perf/hv-24x7.c +++ b/arch/powerpc/perf/hv-24x7.c @@ -485,13 +485,13 @@ static int hv_24x7_init(void) struct hv_perf_caps caps; if (!firmware_has_feature(FW_FEATURE_LPAR)) { - pr_info("not a virtualized system, not enabling\n"); + pr_debug("not a virtualized system, not enabling\n"); return -ENODEV; } hret = hv_perf_caps_get(&caps); if (hret) { - pr_info("could not obtain capabilities, error 0x%80lx, not enabling\n", + pr_debug("could not obtain capabilities, not enabling, rc=%ld\n", hret); return -ENODEV; } -- cgit v1.2.3 From 0a8cf9e28c094cd934deaedfd3ec07830f2a63cf Mon Sep 17 00:00:00 2001 From: Cody P Schafer Date: Tue, 15 Apr 2014 10:10:51 -0700 Subject: powerpc/perf/hv_gpci: Probe failures use pr_debug(), and padding reduced fixup for "powerpc/perf: Add support for the hv gpci (get performance counter info) interface". Makes the "not enabled" message less awful (and hidden unless debugging). Signed-off-by: Cody P Schafer Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/perf/hv-gpci.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/perf/hv-gpci.c b/arch/powerpc/perf/hv-gpci.c index 278ba7b9c2b5..8fee1dccaaec 100644 --- a/arch/powerpc/perf/hv-gpci.c +++ b/arch/powerpc/perf/hv-gpci.c @@ -273,13 +273,13 @@ static int hv_gpci_init(void) struct hv_perf_caps caps; if (!firmware_has_feature(FW_FEATURE_LPAR)) { - pr_info("not a virtualized system, not enabling\n"); + pr_debug("not a virtualized system, not enabling\n"); return -ENODEV; } hret = hv_perf_caps_get(&caps); if (hret) { - pr_info("could not obtain capabilities, error 0x%80lx, not enabling\n", + pr_debug("could not obtain capabilities, not enabling, rc=%ld\n", hret); return -ENODEV; } -- cgit v1.2.3 From 58a685c2d8f0616b5c6cfa0f7a9f2d081af3c6fd Mon Sep 17 00:00:00 2001 From: Cody P Schafer Date: Tue, 15 Apr 2014 10:10:52 -0700 Subject: powerpc/perf/hv-gpci: Make device attr static Signed-off-by: Cody P Schafer Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/perf/hv-gpci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/powerpc/perf/hv-gpci.c b/arch/powerpc/perf/hv-gpci.c index 8fee1dccaaec..c9d399a2df82 100644 --- a/arch/powerpc/perf/hv-gpci.c +++ b/arch/powerpc/perf/hv-gpci.c @@ -78,7 +78,7 @@ static ssize_t kernel_version_show(struct device *dev, return sprintf(page, "0x%x\n", COUNTER_INFO_VERSION_CURRENT); } -DEVICE_ATTR_RO(kernel_version); +static DEVICE_ATTR_RO(kernel_version); HV_CAPS_ATTR(version, "0x%x\n"); HV_CAPS_ATTR(ga, "%d\n"); HV_CAPS_ATTR(expanded, "%d\n"); -- cgit v1.2.3 From 78d13166b189ff1741327d748a7f8d73e65d70bd Mon Sep 17 00:00:00 2001 From: Cody P Schafer Date: Tue, 15 Apr 2014 10:10:53 -0700 Subject: powerpc/perf/hv-24x7: Use (unsigned long) not (u32) values when calling plpar_hcall_norets() Signed-off-by: Cody P Schafer Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/perf/hv-24x7.c | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/perf/hv-24x7.c b/arch/powerpc/perf/hv-24x7.c index f5bca73d92b8..3e8f60a1d53f 100644 --- a/arch/powerpc/perf/hv-24x7.c +++ b/arch/powerpc/perf/hv-24x7.c @@ -155,16 +155,28 @@ static ssize_t read_offset_data(void *dest, size_t dest_len, return copy_len; } -static unsigned long h_get_24x7_catalog_page(char page[static 4096], - u32 version, u32 index) +static unsigned long h_get_24x7_catalog_page_(unsigned long phys_4096, + unsigned long version, + unsigned long index) { - WARN_ON(!IS_ALIGNED((unsigned long)page, 4096)); + pr_devel("h_get_24x7_catalog_page(0x%lx, %lu, %lu)", + phys_4096, + version, + index); + WARN_ON(!IS_ALIGNED(phys_4096, 4096)); return plpar_hcall_norets(H_GET_24X7_CATALOG_PAGE, - virt_to_phys(page), + phys_4096, version, index); } +static unsigned long h_get_24x7_catalog_page(char page[static 4096], + u32 version, u32 index) +{ + return h_get_24x7_catalog_page_(virt_to_phys(page), + version, index); +} + static ssize_t catalog_read(struct file *filp, struct kobject *kobj, struct bin_attribute *bin_attr, char *buf, loff_t offset, size_t count) -- cgit v1.2.3 From 1ee9fcc1a021e54454a23b107a47d10fb603508c Mon Sep 17 00:00:00 2001 From: Cody P Schafer Date: Tue, 15 Apr 2014 10:10:54 -0700 Subject: powerpc/perf/hv-24x7: Remove [static 4096], sparse chokes on it Signed-off-by: Cody P Schafer Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/perf/hv-24x7.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/powerpc/perf/hv-24x7.c b/arch/powerpc/perf/hv-24x7.c index 3e8f60a1d53f..95a67f81f22a 100644 --- a/arch/powerpc/perf/hv-24x7.c +++ b/arch/powerpc/perf/hv-24x7.c @@ -170,7 +170,7 @@ static unsigned long h_get_24x7_catalog_page_(unsigned long phys_4096, index); } -static unsigned long h_get_24x7_catalog_page(char page[static 4096], +static unsigned long h_get_24x7_catalog_page(char page[], u32 version, u32 index) { return h_get_24x7_catalog_page_(virt_to_phys(page), -- cgit v1.2.3 From bbad3e50e8ed2d48294c141e687d411430cacbcf Mon Sep 17 00:00:00 2001 From: Cody P Schafer Date: Tue, 15 Apr 2014 10:10:55 -0700 Subject: powerpc/perf/hv-24x7: Catalog version number is be64, not be32 The catalog version number was changed from a be32 (with proceeding 32bits of padding) to a be64, update the code to treat it as a be64 Signed-off-by: Cody P Schafer Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/perf/hv-24x7.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/perf/hv-24x7.c b/arch/powerpc/perf/hv-24x7.c index 95a67f81f22a..e0766b82e165 100644 --- a/arch/powerpc/perf/hv-24x7.c +++ b/arch/powerpc/perf/hv-24x7.c @@ -171,7 +171,7 @@ static unsigned long h_get_24x7_catalog_page_(unsigned long phys_4096, } static unsigned long h_get_24x7_catalog_page(char page[], - u32 version, u32 index) + u64 version, u32 index) { return h_get_24x7_catalog_page_(virt_to_phys(page), version, index); @@ -185,7 +185,7 @@ static ssize_t catalog_read(struct file *filp, struct kobject *kobj, ssize_t ret = 0; size_t catalog_len = 0, catalog_page_len = 0, page_count = 0; loff_t page_offset = 0; - uint32_t catalog_version_num = 0; + uint64_t catalog_version_num = 0; void *page = kmem_cache_alloc(hv_page_cache, GFP_USER); struct hv_24x7_catalog_page_0 *page_0 = page; if (!page) @@ -197,7 +197,7 @@ static ssize_t catalog_read(struct file *filp, struct kobject *kobj, goto e_free; } - catalog_version_num = be32_to_cpu(page_0->version); + catalog_version_num = be64_to_cpu(page_0->version); catalog_page_len = be32_to_cpu(page_0->length); catalog_len = catalog_page_len * 4096; @@ -220,8 +220,9 @@ static ssize_t catalog_read(struct file *filp, struct kobject *kobj, page, 4096, page_offset * 4096); e_free: if (hret) - pr_err("h_get_24x7_catalog_page(ver=%d, page=%lld) failed: rc=%ld\n", - catalog_version_num, page_offset, hret); + pr_err("h_get_24x7_catalog_page(ver=%lld, page=%lld) failed:" + " rc=%ld\n", + catalog_version_num, page_offset, hret); kfree(page); pr_devel("catalog_read: offset=%lld(%lld) count=%zu(%zu) catalog_len=%zu(%zu) => %zd\n", @@ -255,7 +256,7 @@ e_free: \ static DEVICE_ATTR_RO(_name) PAGE_0_ATTR(catalog_version, "%lld\n", - (unsigned long long)be32_to_cpu(page_0->version)); + (unsigned long long)be64_to_cpu(page_0->version)); PAGE_0_ATTR(catalog_len, "%lld\n", (unsigned long long)be32_to_cpu(page_0->length) * 4096); static BIN_ATTR_RO(catalog, 0/* real length varies */); -- cgit v1.2.3 From a2dd5da77f2cc9fc9ff11ea6b699556254e94a6c Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Mon, 14 Apr 2014 21:54:05 +1000 Subject: powerpc: Rename duplicate COMMAND_LINE_SIZE define We have two definitions of COMMAND_LINE_SIZE, one for the kernel and one for the boot wrapper. I assume this is so the boot wrapper can be self sufficient and not rely on kernel headers. Having two defines with the same name is confusing, I just updated the wrong one when trying to bump it. Make the boot wrapper define unique by calling it BOOT_COMMAND_LINE_SIZE. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/boot/main.c | 8 ++++---- arch/powerpc/boot/ops.h | 2 +- arch/powerpc/boot/ps3.c | 4 ++-- 3 files changed, 7 insertions(+), 7 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/boot/main.c b/arch/powerpc/boot/main.c index a28f02165e97..d367a0aece2a 100644 --- a/arch/powerpc/boot/main.c +++ b/arch/powerpc/boot/main.c @@ -139,18 +139,18 @@ static struct addr_range prep_initrd(struct addr_range vmlinux, void *chosen, * edit the command line passed to vmlinux (by setting /chosen/bootargs). * The buffer is put in it's own section so that tools may locate it easier. */ -static char cmdline[COMMAND_LINE_SIZE] +static char cmdline[BOOT_COMMAND_LINE_SIZE] __attribute__((__section__("__builtin_cmdline"))); static void prep_cmdline(void *chosen) { if (cmdline[0] == '\0') - getprop(chosen, "bootargs", cmdline, COMMAND_LINE_SIZE-1); + getprop(chosen, "bootargs", cmdline, BOOT_COMMAND_LINE_SIZE-1); printf("\n\rLinux/PowerPC load: %s", cmdline); /* If possible, edit the command line */ if (console_ops.edit_cmdline) - console_ops.edit_cmdline(cmdline, COMMAND_LINE_SIZE); + console_ops.edit_cmdline(cmdline, BOOT_COMMAND_LINE_SIZE); printf("\n\r"); /* Put the command line back into the devtree for the kernel */ @@ -174,7 +174,7 @@ void start(void) * built-in command line wasn't set by an external tool */ if ((loader_info.cmdline_len > 0) && (cmdline[0] == '\0')) memmove(cmdline, loader_info.cmdline, - min(loader_info.cmdline_len, COMMAND_LINE_SIZE-1)); + min(loader_info.cmdline_len, BOOT_COMMAND_LINE_SIZE-1)); if (console_ops.open && (console_ops.open() < 0)) exit(); diff --git a/arch/powerpc/boot/ops.h b/arch/powerpc/boot/ops.h index b3218ce451bb..339e041e8dda 100644 --- a/arch/powerpc/boot/ops.h +++ b/arch/powerpc/boot/ops.h @@ -15,7 +15,7 @@ #include "types.h" #include "string.h" -#define COMMAND_LINE_SIZE 512 +#define BOOT_COMMAND_LINE_SIZE 512 #define MAX_PATH_LEN 256 #define MAX_PROP_LEN 256 /* What should this be? */ diff --git a/arch/powerpc/boot/ps3.c b/arch/powerpc/boot/ps3.c index 9954d98871d0..4ec2d86d3c50 100644 --- a/arch/powerpc/boot/ps3.c +++ b/arch/powerpc/boot/ps3.c @@ -47,13 +47,13 @@ BSS_STACK(4096); * The buffer is put in it's own section so that tools may locate it easier. */ -static char cmdline[COMMAND_LINE_SIZE] +static char cmdline[BOOT_COMMAND_LINE_SIZE] __attribute__((__section__("__builtin_cmdline"))); static void prep_cmdline(void *chosen) { if (cmdline[0] == '\0') - getprop(chosen, "bootargs", cmdline, COMMAND_LINE_SIZE-1); + getprop(chosen, "bootargs", cmdline, BOOT_COMMAND_LINE_SIZE-1); else setprop_str(chosen, "bootargs", cmdline); -- cgit v1.2.3 From a5980d064fe292584da73ddd180f4f140f9ab97d Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Mon, 14 Apr 2014 21:54:52 +1000 Subject: powerpc: Bump COMMAND_LINE_SIZE to 2048 I've had a report that the current limit is too small for an automated network based installer. Bump it. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/include/uapi/asm/setup.h | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/powerpc/include/uapi/asm/setup.h b/arch/powerpc/include/uapi/asm/setup.h index 552df83f1a49..ae3fb68cb28e 100644 --- a/arch/powerpc/include/uapi/asm/setup.h +++ b/arch/powerpc/include/uapi/asm/setup.h @@ -1 +1,6 @@ -#include +#ifndef _UAPI_ASM_POWERPC_SETUP_H +#define _UAPI_ASM_POWERPC_SETUP_H + +#define COMMAND_LINE_SIZE 2048 + +#endif /* _UAPI_ASM_POWERPC_SETUP_H */ -- cgit v1.2.3 From 579a53cafdd641f2d67ba4f4fbd9cc4180b8717d Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Mon, 14 Apr 2014 21:55:25 +1000 Subject: powerpc: Bump BOOT_COMMAND_LINE_SIZE to 2048 Bump the boot wrapper BOOT_COMMAND_LINE_SIZE to match the kernel. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/boot/ops.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/powerpc/boot/ops.h b/arch/powerpc/boot/ops.h index 339e041e8dda..8aad3c55aeda 100644 --- a/arch/powerpc/boot/ops.h +++ b/arch/powerpc/boot/ops.h @@ -15,7 +15,7 @@ #include "types.h" #include "string.h" -#define BOOT_COMMAND_LINE_SIZE 512 +#define BOOT_COMMAND_LINE_SIZE 2048 #define MAX_PATH_LEN 256 #define MAX_PROP_LEN 256 /* What should this be? */ -- cgit v1.2.3 From 0c930692107be4b91c3eeecfd9c9d600039f966f Mon Sep 17 00:00:00 2001 From: Anton Blanchard Date: Mon, 14 Apr 2014 21:23:32 +1000 Subject: powerpc: Fix error return in rtas_flash module init module_init should return 0 or a negative errno. Signed-off-by: Anton Blanchard Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kernel/rtas_flash.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/powerpc/kernel/rtas_flash.c b/arch/powerpc/kernel/rtas_flash.c index 2f3cdb01506d..658e89d2025b 100644 --- a/arch/powerpc/kernel/rtas_flash.c +++ b/arch/powerpc/kernel/rtas_flash.c @@ -705,7 +705,7 @@ static int __init rtas_flash_init(void) if (rtas_token("ibm,update-flash-64-and-reboot") == RTAS_UNKNOWN_SERVICE) { pr_info("rtas_flash: no firmware flash support\n"); - return 1; + return -EINVAL; } rtas_validate_flash_data.buf = kzalloc(VALIDATE_BUF_SIZE, GFP_KERNEL); -- cgit v1.2.3 From 42dbfc8649737cb622b2a7e02045401c4c09561c Mon Sep 17 00:00:00 2001 From: Li Zhong Date: Thu, 10 Apr 2014 16:25:31 +0800 Subject: powerpc/pseries: Protect remove_memory() with device hotplug lock While testing memory hot-remove, I found following dead lock: Process #1141 is drmgr, trying to remove some memory, i.e. memory499. It holds the memory_hotplug_mutex, and blocks when trying to remove file "online" under dir memory499, in kernfs_drain(), at wait_event(root->deactivate_waitq, atomic_read(&kn->active) == KN_DEACTIVATED_BIAS); Process #1120 is trying to online memory499 by echo 1 > memory499/online In .kernfs_fop_write, it uses kernfs_get_active() to increase &kn->active, thus blocking process #1141. While itself is blocked later when trying to acquire memory_hotplug_mutex, which is held by process The backtrace of both processes are shown below: [] 0xc000000001b18600 [] .__switch_to+0x144/0x200 [] .online_pages+0x74/0x7b0 [] .memory_subsys_online+0x9c/0x150 [] .device_online+0xb8/0x120 [] .online_store+0xb4/0xc0 [] .dev_attr_store+0x64/0xa0 [] .sysfs_kf_write+0x7c/0xb0 [] .kernfs_fop_write+0x154/0x1e0 [] .vfs_write+0xe0/0x260 [] .SyS_write+0x64/0x110 [] syscall_exit+0x0/0x7c [] 0xc000000001b18600 [] .__switch_to+0x144/0x200 [] .__kernfs_remove+0x204/0x300 [] .kernfs_remove_by_name_ns+0x68/0xf0 [] .sysfs_remove_file_ns+0x38/0x60 [] .device_remove_attrs+0x54/0xc0 [] .device_del+0x158/0x250 [] .device_unregister+0x34/0xa0 [] .unregister_memory_section+0x164/0x170 [] .__remove_pages+0x108/0x4c0 [] .arch_remove_memory+0x60/0xc0 [] .remove_memory+0x8c/0xe0 [] .pseries_remove_memblock+0xd4/0x160 [] .pseries_memory_notifier+0x27c/0x290 [] .notifier_call_chain+0x8c/0x100 [] .__blocking_notifier_call_chain+0x6c/0xe0 [] .of_property_notify+0x7c/0xc0 [] .of_update_property+0x3c/0x1b0 [] .ofdt_write+0x3dc/0x740 [] .proc_reg_write+0xac/0x110 [] .vfs_write+0xe0/0x260 [] .SyS_write+0x64/0x110 [] syscall_exit+0x0/0x7c This patch uses lock_device_hotplug() to protect remove_memory() called in pseries_remove_memblock(), which is also stated before function remove_memory(): * NOTE: The caller must call lock_device_hotplug() to serialize hotplug * and online/offline operations before this call, as required by * try_offline_node(). */ void __ref remove_memory(int nid, u64 start, u64 size) With this lock held, the other process(#1120 above) trying to online the memory block will retry the system call when calling lock_device_hotplug_sysfs(), and finally find No such device error. Signed-off-by: Li Zhong Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/pseries/hotplug-memory.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'arch') diff --git a/arch/powerpc/platforms/pseries/hotplug-memory.c b/arch/powerpc/platforms/pseries/hotplug-memory.c index 573b488fc48b..7f75c94af822 100644 --- a/arch/powerpc/platforms/pseries/hotplug-memory.c +++ b/arch/powerpc/platforms/pseries/hotplug-memory.c @@ -100,10 +100,10 @@ static int pseries_remove_memblock(unsigned long base, unsigned int memblock_siz start_pfn = base >> PAGE_SHIFT; - if (!pfn_valid(start_pfn)) { - memblock_remove(base, memblock_size); - return 0; - } + lock_device_hotplug(); + + if (!pfn_valid(start_pfn)) + goto out; block_sz = memory_block_size_bytes(); sections_per_block = block_sz / MIN_MEMORY_BLOCK_SIZE; @@ -114,8 +114,10 @@ static int pseries_remove_memblock(unsigned long base, unsigned int memblock_siz base += MIN_MEMORY_BLOCK_SIZE; } +out: /* Update memory regions for memory remove */ memblock_remove(base, memblock_size); + unlock_device_hotplug(); return 0; } -- cgit v1.2.3 From f203891117323dd31c11adcbba60452c933d604e Mon Sep 17 00:00:00 2001 From: Preeti U Murthy Date: Fri, 11 Apr 2014 16:01:48 +0530 Subject: ppc/powernv: Set the runlatch bits correctly for offline cpus Up until now we have been setting the runlatch bits for a busy CPU and clearing it when a CPU enters idle state. The runlatch bit has thus been consistent with the utilization of a CPU as long as the CPU is online. However when a CPU is hotplugged out the runlatch bit is not cleared. It needs to be cleared to indicate an unused CPU. Hence this patch has the runlatch bit cleared for an offline CPU just before entering an idle state and sets it immediately after it exits the idle state. Signed-off-by: Preeti U Murthy Acked-by: Paul Mackerras Reviewed-by: Srivatsa S. Bhat Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/platforms/powernv/smp.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'arch') diff --git a/arch/powerpc/platforms/powernv/smp.c b/arch/powerpc/platforms/powernv/smp.c index 908672bdcea6..bf5fcd452168 100644 --- a/arch/powerpc/platforms/powernv/smp.c +++ b/arch/powerpc/platforms/powernv/smp.c @@ -30,6 +30,7 @@ #include #include #include +#include #include "powernv.h" @@ -156,7 +157,9 @@ static void pnv_smp_cpu_kill_self(void) */ mtspr(SPRN_LPCR, mfspr(SPRN_LPCR) & ~(u64)LPCR_PECE1); while (!generic_check_cpu_restart(cpu)) { + ppc64_runlatch_off(); power7_nap(); + ppc64_runlatch_on(); if (!generic_check_cpu_restart(cpu)) { DBG("CPU%d Unexpected exit while offline !\n", cpu); /* We may be getting an IPI, so we re-enable -- cgit v1.2.3 From fd17dc7b9a8d57150ee0121c4c83edbb7e26eece Mon Sep 17 00:00:00 2001 From: Preeti U Murthy Date: Fri, 11 Apr 2014 16:01:58 +0530 Subject: ppc/kvm: Set the runlatch bit of a CPU just before starting guest The secondary threads in the core are kept offline before launching guests in kvm on powerpc: "371fefd6f2dc4666:KVM: PPC: Allow book3s_hv guests to use SMT processor modes." Hence their runlatch bits are cleared. When the secondary threads are called in to start a guest, their runlatch bits need to be set to indicate that they are busy. The primary thread has its runlatch bit set though, but there is no harm in setting this bit once again. Hence set the runlatch bit for all threads before they start guest. Signed-off-by: Preeti U Murthy Acked-by: Paul Mackerras Reviewed-by: Srivatsa S. Bhat Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kvm/book3s_hv_rmhandlers.S | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'arch') diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S index ffbb871c2bd8..b25407533618 100644 --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S @@ -242,6 +242,12 @@ kvm_novcpu_exit: */ .globl kvm_start_guest kvm_start_guest: + + /* Set runlatch bit the minute you wake up from nap */ + mfspr r1, SPRN_CTRLF + ori r1, r1, 1 + mtspr SPRN_CTRLT, r1 + ld r2,PACATOC(r13) li r0,KVM_HWTHREAD_IN_KVM -- cgit v1.2.3 From 582b910edafd283dfab78f41f437a92a65ee5103 Mon Sep 17 00:00:00 2001 From: Preeti U Murthy Date: Fri, 11 Apr 2014 16:02:08 +0530 Subject: ppc/kvm: Clear the runlatch bit of a vcpu before napping When the guest cedes the vcpu or the vcpu has no guest to run it naps. Clear the runlatch bit of the vcpu before napping to indicate an idle cpu. Signed-off-by: Preeti U Murthy Acked-by: Paul Mackerras Reviewed-by: Srivatsa S. Bhat Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/kvm/book3s_hv_rmhandlers.S | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S index b25407533618..b031f932c0cc 100644 --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S @@ -315,6 +315,11 @@ kvm_no_guest: li r0, KVM_HWTHREAD_IN_NAP stb r0, HSTATE_HWTHREAD_STATE(r13) kvm_do_nap: + /* Clear the runlatch bit before napping */ + mfspr r2, SPRN_CTRLF + clrrdi r2, r2, 1 + mtspr SPRN_CTRLT, r2 + li r3, LPCR_PECE0 mfspr r4, SPRN_LPCR rlwimi r4, r3, 0, LPCR_PECE0 | LPCR_PECE1 @@ -2005,8 +2010,13 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_206) /* * Take a nap until a decrementer or external or doobell interrupt - * occurs, with PECE1, PECE0 and PECEDP set in LPCR + * occurs, with PECE1, PECE0 and PECEDP set in LPCR. Also clear the + * runlatch bit before napping. */ + mfspr r2, SPRN_CTRLF + clrrdi r2, r2, 1 + mtspr SPRN_CTRLT, r2 + li r0,1 stb r0,HSTATE_HWTHREAD_REQ(r13) mfspr r5,SPRN_LPCR -- cgit v1.2.3 From e4565362c7adc31201135c4b6d649fc1bdc3bf20 Mon Sep 17 00:00:00 2001 From: Alistair Popple Date: Tue, 8 Apr 2014 14:20:19 +1000 Subject: powerpc/4xx: Fix section mismatch in ppc4xx_pci.c This patch fixes this section mismatch: WARNING: vmlinux.o(.text+0x1efc4): Section mismatch in reference from the function apm821xx_pciex_init_port_hw() to the function .init.text:ppc4xx_pciex_wait_on_sdr.isra.9() The function apm821xx_pciex_init_port_hw() references the function __init ppc4xx_pciex_wait_on_sdr.isra.9(). This is often because apm821xx_pciex_init_port_hw lacks a __init annotation or the annotation of ppc4xx_pciex_wait_on_sdr.isra.9 is wrong. apm821xx_pciex_init_port_hw is only referenced by a struct in __initdata, so it should be safe to add __init to apm821xx_pciex_init_port_hw. Signed-off-by: Alistair Popple Signed-off-by: Benjamin Herrenschmidt --- arch/powerpc/sysdev/ppc4xx_pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/powerpc/sysdev/ppc4xx_pci.c b/arch/powerpc/sysdev/ppc4xx_pci.c index 64603a10b863..4914fd3f41ec 100644 --- a/arch/powerpc/sysdev/ppc4xx_pci.c +++ b/arch/powerpc/sysdev/ppc4xx_pci.c @@ -1058,7 +1058,7 @@ static int __init apm821xx_pciex_core_init(struct device_node *np) return 1; } -static int apm821xx_pciex_init_port_hw(struct ppc4xx_pciex_port *port) +static int __init apm821xx_pciex_init_port_hw(struct ppc4xx_pciex_port *port) { u32 val; -- cgit v1.2.3 From 39025ba38278f3003ee538409f7c98970620ef49 Mon Sep 17 00:00:00 2001 From: Oren Twaig Date: Mon, 28 Apr 2014 10:21:37 +0300 Subject: x86/vsmp: Fix irq routing Correct IRQ routing in case a vSMP box is detected but the Interrupt Routing Comply (IRC) value is set to "comply", which leads to incorrect IRQ routing. Before the patch: When a vSMP box was detected and IRC was set to "comply", users (and the kernel) couldn't effectively set the destination of the IRQs. This is because the hook inside vsmp_64.c always setup all CPUs as the IRQ destination using cpumask_setall() as the return value for IRQ allocation mask. Later, this "overrided" mask caused the kernel to set the IRQ destination to the lowest online CPU in the mask (CPU0 usually). After the patch: When the IRC is set to "comply", users (and the kernel) can control the destination of the IRQs as we will not be changing the default "apic->vector_allocation_domain". Signed-off-by: Oren Twaig Acked-by: Shai Fultheim Link: http://lkml.kernel.org/r/1398669697-2123-1-git-send-email-oren@scalemp.com [ Minor readability edits. ] Signed-off-by: Ingo Molnar --- arch/x86/kernel/vsmp_64.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c index f6584a90aba3..5edc34b5b951 100644 --- a/arch/x86/kernel/vsmp_64.c +++ b/arch/x86/kernel/vsmp_64.c @@ -26,6 +26,9 @@ #define TOPOLOGY_REGISTER_OFFSET 0x10 +/* Flag below is initialized once during vSMP PCI initialization. */ +static int irq_routing_comply = 1; + #if defined CONFIG_PCI && defined CONFIG_PARAVIRT /* * Interrupt control on vSMPowered systems: @@ -101,6 +104,10 @@ static void __init set_vsmp_pv_ops(void) #ifdef CONFIG_SMP if (cap & ctl & BIT(8)) { ctl &= ~BIT(8); + + /* Interrupt routing set to ignore */ + irq_routing_comply = 0; + #ifdef CONFIG_PROC_FS /* Don't let users change irq affinity via procfs */ no_irq_affinity = 1; @@ -218,7 +225,9 @@ static void vsmp_apic_post_init(void) { /* need to update phys_pkg_id */ apic->phys_pkg_id = apicid_phys_pkg_id; - apic->vector_allocation_domain = fill_vector_allocation_domain; + + if (!irq_routing_comply) + apic->vector_allocation_domain = fill_vector_allocation_domain; } void __init vsmp_init(void) -- cgit v1.2.3 From fe2b201b3be91575857227da7ef21f661bc460b1 Mon Sep 17 00:00:00 2001 From: Bandan Das Date: Mon, 21 Apr 2014 15:20:14 -0400 Subject: KVM: x86: Check for host supported fields in shadow vmcs We track shadow vmcs fields through two static lists, one for read only and another for r/w fields. However, with addition of new vmcs fields, not all fields may be supported on all hosts. If so, copy_vmcs12_to_shadow() trying to vmwrite on unsupported hosts will result in a vmwrite error. For example, commit 36be0b9deb23161 introduced GUEST_BNDCFGS, which is not supported by all processors. Filter out host unsupported fields before letting guests use shadow vmcs Signed-off-by: Bandan Das Signed-off-by: Paolo Bonzini --- arch/x86/kvm/vmx.c | 53 +++++++++++++++++++++++++++++++++++++++++------------ 1 file changed, 41 insertions(+), 12 deletions(-) (limited to 'arch') diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 1f68c5831924..33e8c028842f 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -503,7 +503,7 @@ static inline struct vcpu_vmx *to_vmx(struct kvm_vcpu *vcpu) [number##_HIGH] = VMCS12_OFFSET(name)+4 -static const unsigned long shadow_read_only_fields[] = { +static unsigned long shadow_read_only_fields[] = { /* * We do NOT shadow fields that are modified when L0 * traps and emulates any vmx instruction (e.g. VMPTRLD, @@ -526,10 +526,10 @@ static const unsigned long shadow_read_only_fields[] = { GUEST_LINEAR_ADDRESS, GUEST_PHYSICAL_ADDRESS }; -static const int max_shadow_read_only_fields = +static int max_shadow_read_only_fields = ARRAY_SIZE(shadow_read_only_fields); -static const unsigned long shadow_read_write_fields[] = { +static unsigned long shadow_read_write_fields[] = { GUEST_RIP, GUEST_RSP, GUEST_CR0, @@ -558,7 +558,7 @@ static const unsigned long shadow_read_write_fields[] = { HOST_FS_SELECTOR, HOST_GS_SELECTOR }; -static const int max_shadow_read_write_fields = +static int max_shadow_read_write_fields = ARRAY_SIZE(shadow_read_write_fields); static const unsigned short vmcs_field_to_offset_table[] = { @@ -3009,6 +3009,41 @@ static void free_kvm_area(void) } } +static void init_vmcs_shadow_fields(void) +{ + int i, j; + + /* No checks for read only fields yet */ + + for (i = j = 0; i < max_shadow_read_write_fields; i++) { + switch (shadow_read_write_fields[i]) { + case GUEST_BNDCFGS: + if (!vmx_mpx_supported()) + continue; + break; + default: + break; + } + + if (j < i) + shadow_read_write_fields[j] = + shadow_read_write_fields[i]; + j++; + } + max_shadow_read_write_fields = j; + + /* shadowed fields guest access without vmexit */ + for (i = 0; i < max_shadow_read_write_fields; i++) { + clear_bit(shadow_read_write_fields[i], + vmx_vmwrite_bitmap); + clear_bit(shadow_read_write_fields[i], + vmx_vmread_bitmap); + } + for (i = 0; i < max_shadow_read_only_fields; i++) + clear_bit(shadow_read_only_fields[i], + vmx_vmread_bitmap); +} + static __init int alloc_kvm_area(void) { int cpu; @@ -3039,6 +3074,8 @@ static __init int hardware_setup(void) enable_vpid = 0; if (!cpu_has_vmx_shadow_vmcs()) enable_shadow_vmcs = 0; + if (enable_shadow_vmcs) + init_vmcs_shadow_fields(); if (!cpu_has_vmx_ept() || !cpu_has_vmx_ept_4levels()) { @@ -8803,14 +8840,6 @@ static int __init vmx_init(void) memset(vmx_vmread_bitmap, 0xff, PAGE_SIZE); memset(vmx_vmwrite_bitmap, 0xff, PAGE_SIZE); - /* shadowed read/write fields */ - for (i = 0; i < max_shadow_read_write_fields; i++) { - clear_bit(shadow_read_write_fields[i], vmx_vmwrite_bitmap); - clear_bit(shadow_read_write_fields[i], vmx_vmread_bitmap); - } - /* shadowed read only fields */ - for (i = 0; i < max_shadow_read_only_fields; i++) - clear_bit(shadow_read_only_fields[i], vmx_vmread_bitmap); /* * Allow direct access to the PC debug port (it is often used for I/O -- cgit v1.2.3 From 62a08ae2a5763aabeee98264605236b001503e0c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 24 Apr 2014 09:50:53 +0200 Subject: genirq: x86: Ensure that dynamic irq allocation does not conflict On x86 the allocation of irq descriptors may allocate interrupts which are in the range of the GSI interrupts. That's wrong as those interrupts are hardwired and we don't have the irq domain translation like PPC. So one of these interrupts can be hooked up later to one of the devices which are hard wired to it and the io_apic init code for that particular interrupt line happily reuses that descriptor with a completely different configuration so hell breaks lose. Inside x86 we allocate dynamic interrupts from above nr_gsi_irqs, except for a few usage sites which have not yet blown up in our face for whatever reason. But for drivers which need an irq range, like the GPIO drivers, we have no limit in place and we don't want to expose such a detail to a driver. To cure this introduce a function which an architecture can implement to impose a lower bound on the dynamic interrupt allocations. Implement it for x86 and set the lower bound to nr_gsi_irqs, which is the end of the hardwired interrupt space, so all dynamic allocations happen above. That not only allows the GPIO driver to work sanely, it also protects the bogus callsites of create_irq_nr() in hpet, uv, irq_remapping and htirq code. They need to be cleaned up as well, but that's a separate issue. Reported-by: Jin Yao Signed-off-by: Thomas Gleixner Tested-by: Mika Westerberg Cc: Mathias Nyman Cc: Linus Torvalds Cc: Grant Likely Cc: H. Peter Anvin Cc: Rafael J. Wysocki Cc: Andy Shevchenko Cc: Krogerus Heikki Cc: Linus Walleij Link: http://lkml.kernel.org/r/alpine.DEB.2.02.1404241617360.28206@ionos.tec.linutronix.de Signed-off-by: Thomas Gleixner --- arch/x86/kernel/apic/io_apic.c | 5 +++++ include/linux/irq.h | 2 ++ kernel/irq/irqdesc.c | 7 +++++++ kernel/softirq.c | 5 +++++ 4 files changed, 19 insertions(+) (limited to 'arch') diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 6ad4658de705..d23aa82e7a7b 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -3425,6 +3425,11 @@ int get_nr_irqs_gsi(void) return nr_irqs_gsi; } +unsigned int arch_dynirq_lower_bound(unsigned int from) +{ + return from < nr_irqs_gsi ? nr_irqs_gsi : from; +} + int __init arch_probe_nr_irqs(void) { int nr; diff --git a/include/linux/irq.h b/include/linux/irq.h index 10a0b1ac4ea0..5c57efb863d0 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -603,6 +603,8 @@ static inline u32 irq_get_trigger_type(unsigned int irq) return d ? irqd_get_trigger_type(d) : 0; } +unsigned int arch_dynirq_lower_bound(unsigned int from); + int __irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node, struct module *owner); diff --git a/kernel/irq/irqdesc.c b/kernel/irq/irqdesc.c index a7174617616b..bb07f2928f4b 100644 --- a/kernel/irq/irqdesc.c +++ b/kernel/irq/irqdesc.c @@ -363,6 +363,13 @@ __irq_alloc_descs(int irq, unsigned int from, unsigned int cnt, int node, if (from > irq) return -EINVAL; from = irq; + } else { + /* + * For interrupts which are freely allocated the + * architecture can force a lower bound to the @from + * argument. x86 uses this to exclude the GSI space. + */ + from = arch_dynirq_lower_bound(from); } mutex_lock(&sparse_irq_lock); diff --git a/kernel/softirq.c b/kernel/softirq.c index b50990a5bea0..33e4648ae0e7 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -779,3 +779,8 @@ int __init __weak arch_early_irq_init(void) { return 0; } + +unsigned int __weak arch_dynirq_lower_bound(unsigned int from) +{ + return from; +} -- cgit v1.2.3 From 5d4e08c45a6cf8f1ab3c7fa375007635ac569165 Mon Sep 17 00:00:00 2001 From: Mark Salter Date: Fri, 28 Mar 2014 14:25:19 +0000 Subject: arm: KVM: fix possible misalignment of PGDs and bounce page The kvm/mmu code shared by arm and arm64 uses kalloc() to allocate a bounce page (if hypervisor init code crosses page boundary) and hypervisor PGDs. The problem is that kalloc() does not guarantee the proper alignment. In the case of the bounce page, the page sized buffer allocated may also cross a page boundary negating the purpose and leading to a hang during kvm initialization. Likewise the PGDs allocated may not meet the minimum alignment requirements of the underlying MMU. This patch uses __get_free_page() to guarantee the worst case alignment needs of the bounce page and PGDs on both arm and arm64. Cc: # 3.10+ Signed-off-by: Mark Salter Acked-by: Marc Zyngier Signed-off-by: Christoffer Dall --- arch/arm/kvm/mmu.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) (limited to 'arch') diff --git a/arch/arm/kvm/mmu.c b/arch/arm/kvm/mmu.c index 80bb1e6c2c29..16f804938b8f 100644 --- a/arch/arm/kvm/mmu.c +++ b/arch/arm/kvm/mmu.c @@ -42,6 +42,8 @@ static unsigned long hyp_idmap_start; static unsigned long hyp_idmap_end; static phys_addr_t hyp_idmap_vector; +#define pgd_order get_order(PTRS_PER_PGD * sizeof(pgd_t)) + #define kvm_pmd_huge(_x) (pmd_huge(_x) || pmd_trans_huge(_x)) static void kvm_tlb_flush_vmid_ipa(struct kvm *kvm, phys_addr_t ipa) @@ -293,14 +295,14 @@ void free_boot_hyp_pgd(void) if (boot_hyp_pgd) { unmap_range(NULL, boot_hyp_pgd, hyp_idmap_start, PAGE_SIZE); unmap_range(NULL, boot_hyp_pgd, TRAMPOLINE_VA, PAGE_SIZE); - kfree(boot_hyp_pgd); + free_pages((unsigned long)boot_hyp_pgd, pgd_order); boot_hyp_pgd = NULL; } if (hyp_pgd) unmap_range(NULL, hyp_pgd, TRAMPOLINE_VA, PAGE_SIZE); - kfree(init_bounce_page); + free_page((unsigned long)init_bounce_page); init_bounce_page = NULL; mutex_unlock(&kvm_hyp_pgd_mutex); @@ -330,7 +332,7 @@ void free_hyp_pgds(void) for (addr = VMALLOC_START; is_vmalloc_addr((void*)addr); addr += PGDIR_SIZE) unmap_range(NULL, hyp_pgd, KERN_TO_HYP(addr), PGDIR_SIZE); - kfree(hyp_pgd); + free_pages((unsigned long)hyp_pgd, pgd_order); hyp_pgd = NULL; } @@ -1024,7 +1026,7 @@ int kvm_mmu_init(void) size_t len = __hyp_idmap_text_end - __hyp_idmap_text_start; phys_addr_t phys_base; - init_bounce_page = kmalloc(PAGE_SIZE, GFP_KERNEL); + init_bounce_page = (void *)__get_free_page(GFP_KERNEL); if (!init_bounce_page) { kvm_err("Couldn't allocate HYP init bounce page\n"); err = -ENOMEM; @@ -1050,8 +1052,9 @@ int kvm_mmu_init(void) (unsigned long)phys_base); } - hyp_pgd = kzalloc(PTRS_PER_PGD * sizeof(pgd_t), GFP_KERNEL); - boot_hyp_pgd = kzalloc(PTRS_PER_PGD * sizeof(pgd_t), GFP_KERNEL); + hyp_pgd = (pgd_t *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, pgd_order); + boot_hyp_pgd = (pgd_t *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, pgd_order); + if (!hyp_pgd || !boot_hyp_pgd) { kvm_err("Hyp mode PGD not allocated\n"); err = -ENOMEM; -- cgit v1.2.3 From 8aa9e85adac609588eeec356e5a85059b3b819ba Mon Sep 17 00:00:00 2001 From: Vineet Gupta Date: Wed, 30 Apr 2014 15:26:45 +0530 Subject: ARC: !PREEMPT: Ensure Return to kernel mode is IRQ safe There was a very small race window where resume to kernel mode from a Exception Path (or pure kernel mode which is true for most of ARC exceptions anyways), was not disabling interrupts in restore_regs, clobbering the exception regs Anton found the culprit call flow (after many sleepless nights) | 1. we got a Trap from user land | 2. started to service it. | 3. While doing some stuff on user-land memory (I think it is padzero()), | we got a DataTlbMiss | 4. On return from it we are taking "resume_kernel_mode" path | 5. NEED_RESHED is not set, so we go to "return from exception" path in | restore regs. | 6. there seems to be IRQ happening Signed-off-by: Vineet Gupta Cc: #3.10, 3.12, 3.13, 3.14 Cc: Anton Kolesov Cc: Francois Bedard Signed-off-by: Linus Torvalds --- arch/arc/kernel/entry.S | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'arch') diff --git a/arch/arc/kernel/entry.S b/arch/arc/kernel/entry.S index 819dd5f7eb05..29b82adbf0b4 100644 --- a/arch/arc/kernel/entry.S +++ b/arch/arc/kernel/entry.S @@ -614,11 +614,13 @@ resume_user_mode_begin: resume_kernel_mode: -#ifdef CONFIG_PREEMPT - - ; This is a must for preempt_schedule_irq() + ; Disable Interrupts from this point on + ; CONFIG_PREEMPT: This is a must for preempt_schedule_irq() + ; !CONFIG_PREEMPT: To ensure restore_regs is intr safe IRQ_DISABLE r9 +#ifdef CONFIG_PREEMPT + ; Can't preempt if preemption disabled GET_CURR_THR_INFO_FROM_SP r10 ld r8, [r10, THREAD_INFO_PREEMPT_COUNT] -- cgit v1.2.3 From b7e1bd964812808710b48a5f9a58f7ddfb1d4e59 Mon Sep 17 00:00:00 2001 From: Vineet Gupta Date: Fri, 18 Apr 2014 13:38:35 +0530 Subject: Hexagon: Delete stale barrier.h Commit 93ea02bb8435 ("arch: Clean up asm/barrier.h implementations") wired generic barrier.h for hexagon, but failed to delete the existing file. Cc: Richard Kuo Cc: Peter Zijlstra Signed-off-by: Vineet Gupta Compile-tested-by: Guenter Roeck Signed-off-by: Linus Torvalds --- arch/hexagon/include/asm/barrier.h | 37 ------------------------------------- 1 file changed, 37 deletions(-) delete mode 100644 arch/hexagon/include/asm/barrier.h (limited to 'arch') diff --git a/arch/hexagon/include/asm/barrier.h b/arch/hexagon/include/asm/barrier.h deleted file mode 100644 index 4e863daea25b..000000000000 --- a/arch/hexagon/include/asm/barrier.h +++ /dev/null @@ -1,37 +0,0 @@ -/* - * Memory barrier definitions for the Hexagon architecture - * - * Copyright (c) 2010-2011, 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. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA - * 02110-1301, USA. - */ - -#ifndef _ASM_BARRIER_H -#define _ASM_BARRIER_H - -#define rmb() barrier() -#define read_barrier_depends() barrier() -#define wmb() barrier() -#define mb() barrier() -#define smp_rmb() barrier() -#define smp_read_barrier_depends() barrier() -#define smp_wmb() barrier() -#define smp_mb() barrier() - -/* Set a value and use a memory barrier. Used by the scheduler somewhere. */ -#define set_mb(var, value) \ - do { var = value; mb(); } while (0) - -#endif /* _ASM_BARRIER_H */ -- cgit v1.2.3 From e0d8898d76a785453bfaf6cd08b830a7d5189f78 Mon Sep 17 00:00:00 2001 From: John David Anglin Date: Sun, 27 Apr 2014 16:20:47 -0400 Subject: parisc: remove _STK_LIM_MAX override There are only a couple of architectures that override _STK_LIM_MAX to a non-infinity value. This changes the stack allocation semantics in subtle ways. For example, GNU make changes its stack allocation to the hard maximum defined by _STK_LIM_MAX. As a results, threads executed by processes running under make are allocated a stack size of _STK_LIM_MAX rather than a sensible default value. This causes various thread stress tests to fail when they can't muster more than about 50 threads. The attached change implements the default behavior used by the majority of architectures. Signed-off-by: John David Anglin Reviewed-by: Carlos O'Donell Cc: stable@vger.kernel.org # 3.14 Signed-off-by: Helge Deller --- arch/parisc/include/uapi/asm/resource.h | 1 - 1 file changed, 1 deletion(-) (limited to 'arch') diff --git a/arch/parisc/include/uapi/asm/resource.h b/arch/parisc/include/uapi/asm/resource.h index 8b06343b62ed..090483c47dbb 100644 --- a/arch/parisc/include/uapi/asm/resource.h +++ b/arch/parisc/include/uapi/asm/resource.h @@ -1,7 +1,6 @@ #ifndef _ASM_PARISC_RESOURCE_H #define _ASM_PARISC_RESOURCE_H -#define _STK_LIM_MAX 10 * _STK_LIM #include #endif -- cgit v1.2.3 From 8a415e534dbda905b829320f90c9ce5f7359debf Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Tue, 29 Apr 2014 16:13:22 +0200 Subject: parisc: Use generic uapi/asm/resource.h file Signed-off-by: Helge Deller --- arch/parisc/include/uapi/asm/Kbuild | 3 ++- arch/parisc/include/uapi/asm/resource.h | 6 ------ 2 files changed, 2 insertions(+), 7 deletions(-) delete mode 100644 arch/parisc/include/uapi/asm/resource.h (limited to 'arch') diff --git a/arch/parisc/include/uapi/asm/Kbuild b/arch/parisc/include/uapi/asm/Kbuild index a580642555b6..348356c99514 100644 --- a/arch/parisc/include/uapi/asm/Kbuild +++ b/arch/parisc/include/uapi/asm/Kbuild @@ -1,6 +1,8 @@ # UAPI Header export list include include/uapi/asm-generic/Kbuild.asm +generic-y += resource.h + header-y += bitsperlong.h header-y += byteorder.h header-y += errno.h @@ -13,7 +15,6 @@ header-y += msgbuf.h header-y += pdc.h header-y += posix_types.h header-y += ptrace.h -header-y += resource.h header-y += sembuf.h header-y += setup.h header-y += shmbuf.h diff --git a/arch/parisc/include/uapi/asm/resource.h b/arch/parisc/include/uapi/asm/resource.h deleted file mode 100644 index 090483c47dbb..000000000000 --- a/arch/parisc/include/uapi/asm/resource.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef _ASM_PARISC_RESOURCE_H -#define _ASM_PARISC_RESOURCE_H - -#include - -#endif -- cgit v1.2.3 From aa3449ee9c87d9b7660dd1493248abcc57769e31 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Wed, 30 Apr 2014 19:37:48 -0700 Subject: sparc64: Fix argument sign extension for compat_sys_futex(). Only the second argument, 'op', is signed. Signed-off-by: David S. Miller --- arch/sparc/kernel/sys32.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/sparc/kernel/sys32.S b/arch/sparc/kernel/sys32.S index f7c72b6efc27..d066eb18650c 100644 --- a/arch/sparc/kernel/sys32.S +++ b/arch/sparc/kernel/sys32.S @@ -44,7 +44,7 @@ SIGN1(sys32_timer_settime, compat_sys_timer_settime, %o1) SIGN1(sys32_io_submit, compat_sys_io_submit, %o1) SIGN1(sys32_mq_open, compat_sys_mq_open, %o1) SIGN1(sys32_select, compat_sys_select, %o0) -SIGN3(sys32_futex, compat_sys_futex, %o1, %o2, %o5) +SIGN1(sys32_futex, compat_sys_futex, %o1) SIGN1(sys32_recvfrom, compat_sys_recvfrom, %o0) SIGN1(sys32_recvmsg, compat_sys_recvmsg, %o0) SIGN1(sys32_sendmsg, compat_sys_sendmsg, %o0) -- cgit v1.2.3 From 49b6c01f4c1de3b5e5427ac5aba80f9f6d27837a Mon Sep 17 00:00:00 2001 From: Kirill Tkhai Date: Thu, 17 Apr 2014 00:45:24 +0400 Subject: sparc64: Make itc_sync_lock raw One more place where we must not be able to be preempted or to be interrupted in RT. Always actually disable interrupts during synchronization cycle. Signed-off-by: Kirill Tkhai Signed-off-by: David S. Miller --- arch/sparc/kernel/smp_64.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'arch') diff --git a/arch/sparc/kernel/smp_64.c b/arch/sparc/kernel/smp_64.c index 9781048161ab..745a3633ce14 100644 --- a/arch/sparc/kernel/smp_64.c +++ b/arch/sparc/kernel/smp_64.c @@ -149,7 +149,7 @@ void cpu_panic(void) #define NUM_ROUNDS 64 /* magic value */ #define NUM_ITERS 5 /* likewise */ -static DEFINE_SPINLOCK(itc_sync_lock); +static DEFINE_RAW_SPINLOCK(itc_sync_lock); static unsigned long go[SLAVE + 1]; #define DEBUG_TICK_SYNC 0 @@ -257,7 +257,7 @@ static void smp_synchronize_one_tick(int cpu) go[MASTER] = 0; membar_safe("#StoreLoad"); - spin_lock_irqsave(&itc_sync_lock, flags); + raw_spin_lock_irqsave(&itc_sync_lock, flags); { for (i = 0; i < NUM_ROUNDS*NUM_ITERS; i++) { while (!go[MASTER]) @@ -268,7 +268,7 @@ static void smp_synchronize_one_tick(int cpu) membar_safe("#StoreLoad"); } } - spin_unlock_irqrestore(&itc_sync_lock, flags); + raw_spin_unlock_irqrestore(&itc_sync_lock, flags); } #if defined(CONFIG_SUN_LDOMS) && defined(CONFIG_HOTPLUG_CPU) -- cgit v1.2.3 From 5f35eb0e29ca26da82febe49d7698dbeb8882ea0 Mon Sep 17 00:00:00 2001 From: Dave Young Date: Thu, 1 May 2014 21:15:48 +0800 Subject: x86/efi: earlyprintk=efi,keep fix earlyprintk=efi,keep will cause kernel hangs while freeing initmem like below: VFS: Mounted root (ext4 filesystem) readonly on device 254:2. devtmpfs: mounted Freeing unused kernel memory: 880K (ffffffff817d4000 - ffffffff818b0000) It is caused by efi earlyprintk use __init function which will be freed later. Such as early_efi_write is marked as __init, also it will use early_ioremap which is init function as well. To fix this issue, I added early initcall early_efi_map_fb which maps the whole efi fb for later use. OTOH, adding a wrapper function early_efi_map which calls early_ioremap before ioremap is available. With this patch applied efi boot ok with earlyprintk=efi,keep console=efi Signed-off-by: Dave Young Signed-off-by: Matt Fleming --- arch/x86/platform/efi/early_printk.c | 83 +++++++++++++++++++++++++++--------- 1 file changed, 64 insertions(+), 19 deletions(-) (limited to 'arch') diff --git a/arch/x86/platform/efi/early_printk.c b/arch/x86/platform/efi/early_printk.c index 81b506d5befd..524142117296 100644 --- a/arch/x86/platform/efi/early_printk.c +++ b/arch/x86/platform/efi/early_printk.c @@ -14,48 +14,92 @@ static const struct font_desc *font; static u32 efi_x, efi_y; +static void *efi_fb; +static bool early_efi_keep; -static __init void early_efi_clear_scanline(unsigned int y) +/* + * efi earlyprintk need use early_ioremap to map the framebuffer. + * But early_ioremap is not usable for earlyprintk=efi,keep, ioremap should + * be used instead. ioremap will be available after paging_init() which is + * earlier than initcall callbacks. Thus adding this early initcall function + * early_efi_map_fb to map the whole efi framebuffer. + */ +static __init int early_efi_map_fb(void) { - unsigned long base, *dst; - u16 len; + unsigned long base, size; + + if (!early_efi_keep) + return 0; base = boot_params.screen_info.lfb_base; - len = boot_params.screen_info.lfb_linelength; + size = boot_params.screen_info.lfb_size; + efi_fb = ioremap(base, size); + + return efi_fb ? 0 : -ENOMEM; +} +early_initcall(early_efi_map_fb); + +/* + * early_efi_map maps efi framebuffer region [start, start + len -1] + * In case earlyprintk=efi,keep we have the whole framebuffer mapped already + * so just return the offset efi_fb + start. + */ +static __init_refok void *early_efi_map(unsigned long start, unsigned long len) +{ + unsigned long base; + + base = boot_params.screen_info.lfb_base; + + if (efi_fb) + return (efi_fb + start); + else + return early_ioremap(base + start, len); +} - dst = early_ioremap(base + y*len, len); +static __init_refok void early_efi_unmap(void *addr, unsigned long len) +{ + if (!efi_fb) + early_iounmap(addr, len); +} + +static void early_efi_clear_scanline(unsigned int y) +{ + unsigned long *dst; + u16 len; + + len = boot_params.screen_info.lfb_linelength; + dst = early_efi_map(y*len, len); if (!dst) return; memset(dst, 0, len); - early_iounmap(dst, len); + early_efi_unmap(dst, len); } -static __init void early_efi_scroll_up(void) +static void early_efi_scroll_up(void) { - unsigned long base, *dst, *src; + unsigned long *dst, *src; u16 len; u32 i, height; - base = boot_params.screen_info.lfb_base; len = boot_params.screen_info.lfb_linelength; height = boot_params.screen_info.lfb_height; for (i = 0; i < height - font->height; i++) { - dst = early_ioremap(base + i*len, len); + dst = early_efi_map(i*len, len); if (!dst) return; - src = early_ioremap(base + (i + font->height) * len, len); + src = early_efi_map((i + font->height) * len, len); if (!src) { - early_iounmap(dst, len); + early_efi_unmap(dst, len); return; } memmove(dst, src, len); - early_iounmap(src, len); - early_iounmap(dst, len); + early_efi_unmap(src, len); + early_efi_unmap(dst, len); } } @@ -79,16 +123,14 @@ static void early_efi_write_char(u32 *dst, unsigned char c, unsigned int h) } } -static __init void +static void early_efi_write(struct console *con, const char *str, unsigned int num) { struct screen_info *si; - unsigned long base; unsigned int len; const char *s; void *dst; - base = boot_params.screen_info.lfb_base; si = &boot_params.screen_info; len = si->lfb_linelength; @@ -109,7 +151,7 @@ early_efi_write(struct console *con, const char *str, unsigned int num) for (h = 0; h < font->height; h++) { unsigned int n, x; - dst = early_ioremap(base + (efi_y + h) * len, len); + dst = early_efi_map((efi_y + h) * len, len); if (!dst) return; @@ -123,7 +165,7 @@ early_efi_write(struct console *con, const char *str, unsigned int num) s++; } - early_iounmap(dst, len); + early_efi_unmap(dst, len); } num -= count; @@ -179,6 +221,9 @@ static __init int early_efi_setup(struct console *con, char *options) for (i = 0; i < (yres - efi_y) / font->height; i++) early_efi_scroll_up(); + /* early_console_register will unset CON_BOOT in case ,keep */ + if (!(con->flags & CON_BOOT)) + early_efi_keep = true; return 0; } -- cgit v1.2.3 From da6e4cb67c6dd1f72257c0a4a97c26dc4e80d3a7 Mon Sep 17 00:00:00 2001 From: Dave Anderson Date: Tue, 15 Apr 2014 18:53:24 +0100 Subject: arm64: Fix for the arm64 kern_addr_valid() function Fix for the arm64 kern_addr_valid() function to recognize virtual addresses in the kernel logical memory map. The function fails as written because it does not check whether the addresses in that region are mapped at the pmd level to 2MB or 512MB pages, continues the page table walk to the pte level, and issues a garbage value to pfn_valid(). Tested on 4K-page and 64K-page kernels. Signed-off-by: Dave Anderson Signed-off-by: Catalin Marinas --- arch/arm64/mm/mmu.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'arch') diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c index 6b7e89569a3a..0a472c41a67f 100644 --- a/arch/arm64/mm/mmu.c +++ b/arch/arm64/mm/mmu.c @@ -374,6 +374,9 @@ int kern_addr_valid(unsigned long addr) if (pmd_none(*pmd)) return 0; + if (pmd_sect(*pmd)) + return pfn_valid(pmd_pfn(*pmd)); + pte = pte_offset_kernel(pmd, addr); if (pte_none(*pte)) return 0; -- cgit v1.2.3 From f774b7d10e2155e52b92dfce2f8cb099a6d6d0e6 Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Mon, 28 Apr 2014 19:50:06 +0100 Subject: arm64: fixmap: fix missing sub-page offset for earlyprintk Commit d57c33c5daa4 (add generic fixmap.h) added (among other similar things) set_fixmap_io to deal with early ioremap of devices. More recently, commit bf4b558eba92 (arm64: add early_ioremap support) converted the arm64 earlyprintk to use set_fixmap_io. A side effect of this conversion is that my virtual machines have stopped booting when I pass "earlyprintk=uart8250-8bit,0x3f8" to the guest kernel. Turns out that the new earlyprintk code doesn't care at all about sub-page offsets, and just assumes that the earlyprintk device will be page-aligned. Obviously, that doesn't play well with the above example. Further investigation shows that set_fixmap_io uses __set_fixmap instead of __set_fixmap_offset. A fix is to introduce a set_fixmap_offset_io that uses the latter, and to remove the superflous call to fix_to_virt (which only returns the value that set_fixmap_io has already given us). With this applied, my VMs are back in business. Tested on a Cortex-A57 platform with kvmtool as platform emulation. Cc: Will Deacon Acked-by: Mark Salter Acked-by: Arnd Bergmann Signed-off-by: Marc Zyngier Signed-off-by: Catalin Marinas --- arch/arm64/kernel/early_printk.c | 6 ++---- include/asm-generic/fixmap.h | 3 +++ 2 files changed, 5 insertions(+), 4 deletions(-) (limited to 'arch') diff --git a/arch/arm64/kernel/early_printk.c b/arch/arm64/kernel/early_printk.c index ffbbdde7aba1..2dc36d00addf 100644 --- a/arch/arm64/kernel/early_printk.c +++ b/arch/arm64/kernel/early_printk.c @@ -143,10 +143,8 @@ static int __init setup_early_printk(char *buf) } /* no options parsing yet */ - if (paddr) { - set_fixmap_io(FIX_EARLYCON_MEM_BASE, paddr); - early_base = (void __iomem *)fix_to_virt(FIX_EARLYCON_MEM_BASE); - } + if (paddr) + early_base = (void __iomem *)set_fixmap_offset_io(FIX_EARLYCON_MEM_BASE, paddr); printch = match->printch; early_console = &early_console_dev; diff --git a/include/asm-generic/fixmap.h b/include/asm-generic/fixmap.h index 5a64ca4621f3..f23174fb9ec4 100644 --- a/include/asm-generic/fixmap.h +++ b/include/asm-generic/fixmap.h @@ -93,5 +93,8 @@ static inline unsigned long virt_to_fix(const unsigned long vaddr) #define set_fixmap_io(idx, phys) \ __set_fixmap(idx, phys, FIXMAP_PAGE_IO) +#define set_fixmap_offset_io(idx, phys) \ + __set_fixmap_offset(idx, phys, FIXMAP_PAGE_IO) + #endif /* __ASSEMBLY__ */ #endif /* __ASM_GENERIC_FIXMAP_H */ -- cgit v1.2.3 From c7a4a7658d689f664050c45493d79adf053f226e Mon Sep 17 00:00:00 2001 From: Ritesh Harjani Date: Wed, 23 Apr 2014 06:29:46 +0100 Subject: arm64: Make default dma_ops to be noncoherent Currently arm64 dma_ops is by default made coherent which makes it opposite in default policy from arm. Make default dma_ops to be noncoherent (same as arm), as currently there aren't any dma-capable drivers which assumes coherent ops Signed-off-by: Ritesh Harjani Acked-by: Will Deacon Signed-off-by: Catalin Marinas --- arch/arm64/mm/dma-mapping.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c index 0ba347e59f06..1f65963a9c04 100644 --- a/arch/arm64/mm/dma-mapping.c +++ b/arch/arm64/mm/dma-mapping.c @@ -311,7 +311,7 @@ static int __init swiotlb_late_init(void) { size_t swiotlb_size = min(SZ_64M, MAX_ORDER_NR_PAGES << PAGE_SHIFT); - dma_ops = &coherent_swiotlb_dma_ops; + dma_ops = &noncoherent_swiotlb_dma_ops; return swiotlb_late_init_with_default_size(swiotlb_size); } -- cgit v1.2.3 From 6ecba8eb51b7d23fda66388a5420be7d8688b186 Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Fri, 25 Apr 2014 15:31:45 +0100 Subject: arm64: Use bus notifiers to set per-device coherent DMA ops Recently, the default DMA ops have been changed to non-coherent for alignment with 32-bit ARM platforms (and DT files). This patch adds bus notifiers to be able to set the coherent DMA ops (with no cache maintenance) for devices explicitly marked as coherent via the "dma-coherent" DT property. Signed-off-by: Catalin Marinas --- arch/arm64/kernel/setup.c | 2 +- arch/arm64/mm/dma-mapping.c | 33 ++++++++++++++++++++++++++++++++- 2 files changed, 33 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c index 93e7df8968fe..7ec784653b29 100644 --- a/arch/arm64/kernel/setup.c +++ b/arch/arm64/kernel/setup.c @@ -396,7 +396,7 @@ static int __init arm64_device_init(void) of_platform_populate(NULL, of_default_bus_match_table, NULL, NULL); return 0; } -arch_initcall(arm64_device_init); +arch_initcall_sync(arm64_device_init); static DEFINE_PER_CPU(struct cpu, cpu_data); diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c index 1f65963a9c04..c851eb44dc50 100644 --- a/arch/arm64/mm/dma-mapping.c +++ b/arch/arm64/mm/dma-mapping.c @@ -22,8 +22,11 @@ #include #include #include +#include +#include #include #include +#include #include @@ -305,17 +308,45 @@ struct dma_map_ops coherent_swiotlb_dma_ops = { }; EXPORT_SYMBOL(coherent_swiotlb_dma_ops); +static int dma_bus_notifier(struct notifier_block *nb, + unsigned long event, void *_dev) +{ + struct device *dev = _dev; + + if (event != BUS_NOTIFY_ADD_DEVICE) + return NOTIFY_DONE; + + if (of_property_read_bool(dev->of_node, "dma-coherent")) + set_dma_ops(dev, &coherent_swiotlb_dma_ops); + + return NOTIFY_OK; +} + +static struct notifier_block platform_bus_nb = { + .notifier_call = dma_bus_notifier, +}; + +static struct notifier_block amba_bus_nb = { + .notifier_call = dma_bus_notifier, +}; + extern int swiotlb_late_init_with_default_size(size_t default_size); static int __init swiotlb_late_init(void) { size_t swiotlb_size = min(SZ_64M, MAX_ORDER_NR_PAGES << PAGE_SHIFT); + /* + * These must be registered before of_platform_populate(). + */ + bus_register_notifier(&platform_bus_type, &platform_bus_nb); + bus_register_notifier(&amba_bustype, &amba_bus_nb); + dma_ops = &noncoherent_swiotlb_dma_ops; return swiotlb_late_init_with_default_size(swiotlb_size); } -subsys_initcall(swiotlb_late_init); +arch_initcall(swiotlb_late_init); #define PREALLOC_DMA_DEBUG_ENTRIES 4096 -- cgit v1.2.3 From 7a8d1ec16dfbb7785e82ccc97b0076cc34911701 Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Fri, 25 Apr 2014 16:39:49 +0100 Subject: arm64: Mark the Applied Micro X-Gene SATA controller as DMA coherent Since the default DMA ops for arm64 are non-coherent, mark the X-Gene controller explicitly as dma-coherent to avoid additional cache maintenance. Signed-off-by: Catalin Marinas Cc: Loc Ho --- Documentation/devicetree/bindings/ata/apm-xgene.txt | 3 +++ arch/arm64/boot/dts/apm-storm.dtsi | 3 +++ 2 files changed, 6 insertions(+) (limited to 'arch') diff --git a/Documentation/devicetree/bindings/ata/apm-xgene.txt b/Documentation/devicetree/bindings/ata/apm-xgene.txt index 7bcfbf59810e..a668f0e7d001 100644 --- a/Documentation/devicetree/bindings/ata/apm-xgene.txt +++ b/Documentation/devicetree/bindings/ata/apm-xgene.txt @@ -24,6 +24,7 @@ Required properties: * "sata-phy" for the SATA 6.0Gbps PHY Optional properties: +- dma-coherent : Present if dma operations are coherent - status : Shall be "ok" if enabled or "disabled" if disabled. Default is "ok". @@ -55,6 +56,7 @@ Example: <0x0 0x1f22e000 0x0 0x1000>, <0x0 0x1f227000 0x0 0x1000>; interrupts = <0x0 0x87 0x4>; + dma-coherent; status = "ok"; clocks = <&sataclk 0>; phys = <&phy2 0>; @@ -69,6 +71,7 @@ Example: <0x0 0x1f23e000 0x0 0x1000>, <0x0 0x1f237000 0x0 0x1000>; interrupts = <0x0 0x88 0x4>; + dma-coherent; status = "ok"; clocks = <&sataclk 0>; phys = <&phy3 0>; diff --git a/arch/arm64/boot/dts/apm-storm.dtsi b/arch/arm64/boot/dts/apm-storm.dtsi index 93f4b2dd9248..f8c40a66e65d 100644 --- a/arch/arm64/boot/dts/apm-storm.dtsi +++ b/arch/arm64/boot/dts/apm-storm.dtsi @@ -307,6 +307,7 @@ <0x0 0x1f21e000 0x0 0x1000>, <0x0 0x1f217000 0x0 0x1000>; interrupts = <0x0 0x86 0x4>; + dma-coherent; status = "disabled"; clocks = <&sata01clk 0>; phys = <&phy1 0>; @@ -321,6 +322,7 @@ <0x0 0x1f22e000 0x0 0x1000>, <0x0 0x1f227000 0x0 0x1000>; interrupts = <0x0 0x87 0x4>; + dma-coherent; status = "ok"; clocks = <&sata23clk 0>; phys = <&phy2 0>; @@ -334,6 +336,7 @@ <0x0 0x1f23d000 0x0 0x1000>, <0x0 0x1f23e000 0x0 0x1000>; interrupts = <0x0 0x88 0x4>; + dma-coherent; status = "ok"; clocks = <&sata45clk 0>; phys = <&phy3 0>; -- cgit v1.2.3 From 16ce8a30e6102d7aabd4e4518d255f330290fa3f Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sat, 3 May 2014 22:25:33 -0700 Subject: sparc64: Normalize NMI watchdog logging and behavior. Bring this code in line with the perf based generic NMI watchdog in kernel/watchdog.c (which we should convert over to at some point). In particular, don't do anything super fancy when the watchdog triggers, and specifically don't do a do_exit() which only makes things worse. Either panic(), or WARN(). The latter of which will do all of the actions such as give us a stack backtrace. Signed-off-by: David S. Miller --- arch/sparc/kernel/nmi.c | 21 +++++---------------- 1 file changed, 5 insertions(+), 16 deletions(-) (limited to 'arch') diff --git a/arch/sparc/kernel/nmi.c b/arch/sparc/kernel/nmi.c index 6479256fd5a4..337094556916 100644 --- a/arch/sparc/kernel/nmi.c +++ b/arch/sparc/kernel/nmi.c @@ -68,27 +68,16 @@ EXPORT_SYMBOL(touch_nmi_watchdog); static void die_nmi(const char *str, struct pt_regs *regs, int do_panic) { + int this_cpu = smp_processor_id(); + if (notify_die(DIE_NMIWATCHDOG, str, regs, 0, pt_regs_trap_type(regs), SIGINT) == NOTIFY_STOP) return; - console_verbose(); - bust_spinlocks(1); - - printk(KERN_EMERG "%s", str); - printk(" on CPU%d, ip %08lx, registers:\n", - smp_processor_id(), regs->tpc); - show_regs(regs); - dump_stack(); - - bust_spinlocks(0); - if (do_panic || panic_on_oops) - panic("Non maskable interrupt"); - - nmi_exit(); - local_irq_enable(); - do_exit(SIGBUS); + panic("Watchdog detected hard LOCKUP on cpu %d", this_cpu); + else + WARN(1, "Watchdog detected hard LOCKUP on cpu %d", this_cpu); } notrace __kprobes void perfctr_irq(int irq, struct pt_regs *regs) -- cgit v1.2.3 From 5b1e94fa439a3227beefad58c28c17f68287a8e9 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 20 Apr 2014 21:55:01 -0400 Subject: sparc64: Fix executable bit testing in set_pmd_at() paths. This code was mistakenly using the exec bit from the PMD in all cases, even when the PMD isn't a huge PMD. If it's not a huge PMD, test the exec bit in the individual ptes down in tlb_batch_pmd_scan(). Signed-off-by: David S. Miller --- arch/sparc/mm/tlb.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) (limited to 'arch') diff --git a/arch/sparc/mm/tlb.c b/arch/sparc/mm/tlb.c index b12cb5e72812..c07667fcd66e 100644 --- a/arch/sparc/mm/tlb.c +++ b/arch/sparc/mm/tlb.c @@ -134,7 +134,7 @@ no_cache_flush: #ifdef CONFIG_TRANSPARENT_HUGEPAGE static void tlb_batch_pmd_scan(struct mm_struct *mm, unsigned long vaddr, - pmd_t pmd, bool exec) + pmd_t pmd) { unsigned long end; pte_t *pte; @@ -142,8 +142,11 @@ static void tlb_batch_pmd_scan(struct mm_struct *mm, unsigned long vaddr, pte = pte_offset_map(&pmd, vaddr); end = vaddr + HPAGE_SIZE; while (vaddr < end) { - if (pte_val(*pte) & _PAGE_VALID) + if (pte_val(*pte) & _PAGE_VALID) { + bool exec = pte_exec(*pte); + tlb_batch_add_one(mm, vaddr, exec); + } pte++; vaddr += PAGE_SIZE; } @@ -177,15 +180,15 @@ void set_pmd_at(struct mm_struct *mm, unsigned long addr, } if (!pmd_none(orig)) { - pte_t orig_pte = __pte(pmd_val(orig)); - bool exec = pte_exec(orig_pte); - addr &= HPAGE_MASK; if (pmd_trans_huge(orig)) { + pte_t orig_pte = __pte(pmd_val(orig)); + bool exec = pte_exec(orig_pte); + tlb_batch_add_one(mm, addr, exec); tlb_batch_add_one(mm, addr + REAL_HPAGE_SIZE, exec); } else { - tlb_batch_pmd_scan(mm, addr, orig, exec); + tlb_batch_pmd_scan(mm, addr, orig); } } } -- cgit v1.2.3 From 51e5ef1bb7ab0e5fa7de4e802da5ab22fe35f0bf Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Thu, 24 Apr 2014 13:58:02 -0700 Subject: sparc64: Fix huge PMD invalidation. On sparc64 "present" and "valid" are seperate PTE bits, this allows us to naturally distinguish between the user explicitly asking for PROT_NONE with mprotect() and other situations. However we weren't handling this properly in the huge PMD paths. First of all, the page table walker in the TSB miss path only checks for _PAGE_PMD_HUGE. So the generic pmdp_invalidate() would clear _PAGE_PRESENT but the TLB miss paths would still load it into the TLB as a valid huge PMD. Fix this by clearing the valid bit in pmdp_invalidate(), and also checking the valid bit in USER_PGTABLE_CHECK_PMD_HUGE using "brgez" since _PAGE_VALID is bit 63 in both the sun4u and sun4v pte layouts. Signed-off-by: David S. Miller --- arch/sparc/include/asm/pgtable_64.h | 18 ++++-------------- arch/sparc/include/asm/tsb.h | 3 ++- arch/sparc/mm/tlb.c | 11 +++++++++++ 3 files changed, 17 insertions(+), 15 deletions(-) (limited to 'arch') diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h index 0f9e94537eee..e3db4b8e689b 100644 --- a/arch/sparc/include/asm/pgtable_64.h +++ b/arch/sparc/include/asm/pgtable_64.h @@ -719,20 +719,6 @@ static inline pmd_t pmd_mkwrite(pmd_t pmd) return __pmd(pte_val(pte)); } -static inline pmd_t pmd_mknotpresent(pmd_t pmd) -{ - unsigned long mask; - - if (tlb_type == hypervisor) - mask = _PAGE_PRESENT_4V; - else - mask = _PAGE_PRESENT_4U; - - pmd_val(pmd) &= ~mask; - - return pmd; -} - static inline pmd_t pmd_mksplitting(pmd_t pmd) { pte_t pte = __pte(pmd_val(pmd)); @@ -893,6 +879,10 @@ extern void update_mmu_cache(struct vm_area_struct *, unsigned long, pte_t *); extern void update_mmu_cache_pmd(struct vm_area_struct *vma, unsigned long addr, pmd_t *pmd); +#define __HAVE_ARCH_PMDP_INVALIDATE +extern void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address, + pmd_t *pmdp); + #define __HAVE_ARCH_PGTABLE_DEPOSIT extern void pgtable_trans_huge_deposit(struct mm_struct *mm, pmd_t *pmdp, pgtable_t pgtable); diff --git a/arch/sparc/include/asm/tsb.h b/arch/sparc/include/asm/tsb.h index 2230f80d9fe3..90916f955cac 100644 --- a/arch/sparc/include/asm/tsb.h +++ b/arch/sparc/include/asm/tsb.h @@ -171,7 +171,8 @@ extern struct tsb_phys_patch_entry __tsb_phys_patch, __tsb_phys_patch_end; andcc REG1, REG2, %g0; \ be,pt %xcc, 700f; \ sethi %hi(4 * 1024 * 1024), REG2; \ - andn REG1, REG2, REG1; \ + brgez,pn REG1, FAIL_LABEL; \ + andn REG1, REG2, REG1; \ and VADDR, REG2, REG2; \ brlz,pt REG1, PTE_LABEL; \ or REG1, REG2, REG1; \ diff --git a/arch/sparc/mm/tlb.c b/arch/sparc/mm/tlb.c index c07667fcd66e..b89aba217e3b 100644 --- a/arch/sparc/mm/tlb.c +++ b/arch/sparc/mm/tlb.c @@ -193,6 +193,17 @@ void set_pmd_at(struct mm_struct *mm, unsigned long addr, } } +void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address, + pmd_t *pmdp) +{ + pmd_t entry = *pmdp; + + pmd_val(entry) &= ~_PAGE_VALID; + + set_pmd_at(vma->vm_mm, address, pmdp, entry); + flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE); +} + void pgtable_trans_huge_deposit(struct mm_struct *mm, pmd_t *pmdp, pgtable_t pgtable) { -- cgit v1.2.3 From 04df419de34104d8818b8c5cffaa062fa36d20ea Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Fri, 25 Apr 2014 10:21:12 -0700 Subject: sparc64: Fix bugs in get_user_pages_fast() wrt. THP. The large PMD path needs to check _PAGE_VALID not _PAGE_PRESENT, to decide if it needs to bail and return 0. pmd_large() should therefore just check _PAGE_PMD_HUGE. Calls to gup_huge_pmd() are guarded with a check of pmd_large(), so we just need to add a valid bit check. Signed-off-by: David S. Miller --- arch/sparc/include/asm/pgtable_64.h | 2 +- arch/sparc/mm/gup.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h index e3db4b8e689b..1f281030769c 100644 --- a/arch/sparc/include/asm/pgtable_64.h +++ b/arch/sparc/include/asm/pgtable_64.h @@ -633,7 +633,7 @@ static inline unsigned long pmd_large(pmd_t pmd) { pte_t pte = __pte(pmd_val(pmd)); - return (pte_val(pte) & _PAGE_PMD_HUGE) && pte_present(pte); + return pte_val(pte) & _PAGE_PMD_HUGE; } #ifdef CONFIG_TRANSPARENT_HUGEPAGE diff --git a/arch/sparc/mm/gup.c b/arch/sparc/mm/gup.c index c4d3da68b800..1aed0432c64b 100644 --- a/arch/sparc/mm/gup.c +++ b/arch/sparc/mm/gup.c @@ -73,7 +73,7 @@ static int gup_huge_pmd(pmd_t *pmdp, pmd_t pmd, unsigned long addr, struct page *head, *page, *tail; int refs; - if (!pmd_large(pmd)) + if (!(pmd_val(pmd) & _PAGE_VALID)) return 0; if (write && !pmd_write(pmd)) -- cgit v1.2.3 From c2e4e676adb40ea764af79d3e08be954e14a0f4c Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sun, 27 Apr 2014 21:01:56 -0700 Subject: sparc64: Fix hex values in comment above pte_modify(). When _PAGE_SPECIAL and _PAGE_PMD_HUGE were added to the mask, the comment was not updated. Signed-off-by: David S. Miller --- arch/sparc/include/asm/pgtable_64.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h index 1f281030769c..50b2d78225ac 100644 --- a/arch/sparc/include/asm/pgtable_64.h +++ b/arch/sparc/include/asm/pgtable_64.h @@ -258,8 +258,8 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t prot) { unsigned long mask, tmp; - /* SUN4U: 0x600307ffffffecb8 (negated == 0x9ffcf80000001347) - * SUN4V: 0x30ffffffffffee17 (negated == 0xcf000000000011e8) + /* SUN4U: 0x630107ffffffecb8 (negated == 0x9cfef80000001347) + * SUN4V: 0x33ffffffffffee17 (negated == 0xcc000000000011e8) * * Even if we use negation tricks the result is still a 6 * instruction sequence, so don't try to play fancy and just -- cgit v1.2.3 From eaf85da82669b057f20c4e438dc2566b51a83af6 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 28 Apr 2014 19:11:27 -0700 Subject: sparc64: Don't use _PAGE_PRESENT in pte_modify() mask. Signed-off-by: David S. Miller --- arch/sparc/include/asm/pgtable_64.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'arch') diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h index 50b2d78225ac..ac5f7087a50d 100644 --- a/arch/sparc/include/asm/pgtable_64.h +++ b/arch/sparc/include/asm/pgtable_64.h @@ -258,8 +258,8 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t prot) { unsigned long mask, tmp; - /* SUN4U: 0x630107ffffffecb8 (negated == 0x9cfef80000001347) - * SUN4V: 0x33ffffffffffee17 (negated == 0xcc000000000011e8) + /* SUN4U: 0x630107ffffffec38 (negated == 0x9cfef800000013c7) + * SUN4V: 0x33ffffffffffee07 (negated == 0xcc000000000011f8) * * Even if we use negation tricks the result is still a 6 * instruction sequence, so don't try to play fancy and just @@ -289,10 +289,10 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t prot) " .previous\n" : "=r" (mask), "=r" (tmp) : "i" (_PAGE_PADDR_4U | _PAGE_MODIFIED_4U | _PAGE_ACCESSED_4U | - _PAGE_CP_4U | _PAGE_CV_4U | _PAGE_E_4U | _PAGE_PRESENT_4U | + _PAGE_CP_4U | _PAGE_CV_4U | _PAGE_E_4U | _PAGE_SPECIAL | _PAGE_PMD_HUGE | _PAGE_SZALL_4U), "i" (_PAGE_PADDR_4V | _PAGE_MODIFIED_4V | _PAGE_ACCESSED_4V | - _PAGE_CP_4V | _PAGE_CV_4V | _PAGE_E_4V | _PAGE_PRESENT_4V | + _PAGE_CP_4V | _PAGE_CV_4V | _PAGE_E_4V | _PAGE_SPECIAL | _PAGE_PMD_HUGE | _PAGE_SZALL_4V)); return __pte((pte_val(pte) & mask) | (pgprot_val(prot) & ~mask)); -- cgit v1.2.3 From d037d16372bbe4d580342bebbb8826821ad9edf0 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 28 Apr 2014 23:50:08 -0700 Subject: sparc64: Handle 32-bit tasks properly in compute_effective_address(). If we have a 32-bit task we must chop off the top 32-bits of the 64-bit value just as the cpu would. Signed-off-by: David S. Miller --- arch/sparc/kernel/unaligned_64.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) (limited to 'arch') diff --git a/arch/sparc/kernel/unaligned_64.c b/arch/sparc/kernel/unaligned_64.c index 3c1a7cb31579..35ab8b60d256 100644 --- a/arch/sparc/kernel/unaligned_64.c +++ b/arch/sparc/kernel/unaligned_64.c @@ -166,17 +166,23 @@ static unsigned long *fetch_reg_addr(unsigned int reg, struct pt_regs *regs) unsigned long compute_effective_address(struct pt_regs *regs, unsigned int insn, unsigned int rd) { + int from_kernel = (regs->tstate & TSTATE_PRIV) != 0; unsigned int rs1 = (insn >> 14) & 0x1f; unsigned int rs2 = insn & 0x1f; - int from_kernel = (regs->tstate & TSTATE_PRIV) != 0; + unsigned long addr; if (insn & 0x2000) { maybe_flush_windows(rs1, 0, rd, from_kernel); - return (fetch_reg(rs1, regs) + sign_extend_imm13(insn)); + addr = (fetch_reg(rs1, regs) + sign_extend_imm13(insn)); } else { maybe_flush_windows(rs1, rs2, rd, from_kernel); - return (fetch_reg(rs1, regs) + fetch_reg(rs2, regs)); + addr = (fetch_reg(rs1, regs) + fetch_reg(rs2, regs)); } + + if (!from_kernel && test_thread_flag(TIF_32BIT)) + addr &= 0xffffffff; + + return addr; } /* This is just to make gcc think die_if_kernel does return... */ -- cgit v1.2.3 From 70ffc6ebaead783ac8dafb1e87df0039bb043596 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Mon, 28 Apr 2014 23:52:11 -0700 Subject: sparc64: Fix top-level fault handling bugs. Make get_user_insn() able to cope with huge PMDs. Next, make do_fault_siginfo() more robust when get_user_insn() can't actually fetch the instruction. In particular, use the MMU announced fault address when that happens, instead of calling compute_effective_address() and computing garbage. Signed-off-by: David S. Miller --- arch/sparc/mm/fault_64.c | 82 ++++++++++++++++++++++++++++++------------------ 1 file changed, 52 insertions(+), 30 deletions(-) (limited to 'arch') diff --git a/arch/sparc/mm/fault_64.c b/arch/sparc/mm/fault_64.c index 69bb818fdd79..a8ff0d1a3b69 100644 --- a/arch/sparc/mm/fault_64.c +++ b/arch/sparc/mm/fault_64.c @@ -96,38 +96,51 @@ static unsigned int get_user_insn(unsigned long tpc) pte_t *ptep, pte; unsigned long pa; u32 insn = 0; - unsigned long pstate; - if (pgd_none(*pgdp)) - goto outret; + if (pgd_none(*pgdp) || unlikely(pgd_bad(*pgdp))) + goto out; pudp = pud_offset(pgdp, tpc); - if (pud_none(*pudp)) - goto outret; - pmdp = pmd_offset(pudp, tpc); - if (pmd_none(*pmdp)) - goto outret; + if (pud_none(*pudp) || unlikely(pud_bad(*pudp))) + goto out; /* This disables preemption for us as well. */ - __asm__ __volatile__("rdpr %%pstate, %0" : "=r" (pstate)); - __asm__ __volatile__("wrpr %0, %1, %%pstate" - : : "r" (pstate), "i" (PSTATE_IE)); - ptep = pte_offset_map(pmdp, tpc); - pte = *ptep; - if (!pte_present(pte)) - goto out; + local_irq_disable(); - pa = (pte_pfn(pte) << PAGE_SHIFT); - pa += (tpc & ~PAGE_MASK); + pmdp = pmd_offset(pudp, tpc); + if (pmd_none(*pmdp) || unlikely(pmd_bad(*pmdp))) + goto out_irq_enable; + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + if (pmd_trans_huge(*pmdp)) { + if (pmd_trans_splitting(*pmdp)) + goto out_irq_enable; - /* Use phys bypass so we don't pollute dtlb/dcache. */ - __asm__ __volatile__("lduwa [%1] %2, %0" - : "=r" (insn) - : "r" (pa), "i" (ASI_PHYS_USE_EC)); + pa = pmd_pfn(*pmdp) << PAGE_SHIFT; + pa += tpc & ~HPAGE_MASK; + /* Use phys bypass so we don't pollute dtlb/dcache. */ + __asm__ __volatile__("lduwa [%1] %2, %0" + : "=r" (insn) + : "r" (pa), "i" (ASI_PHYS_USE_EC)); + } else +#endif + { + ptep = pte_offset_map(pmdp, tpc); + pte = *ptep; + if (pte_present(pte)) { + pa = (pte_pfn(pte) << PAGE_SHIFT); + pa += (tpc & ~PAGE_MASK); + + /* Use phys bypass so we don't pollute dtlb/dcache. */ + __asm__ __volatile__("lduwa [%1] %2, %0" + : "=r" (insn) + : "r" (pa), "i" (ASI_PHYS_USE_EC)); + } + pte_unmap(ptep); + } +out_irq_enable: + local_irq_enable(); out: - pte_unmap(ptep); - __asm__ __volatile__("wrpr %0, 0x0, %%pstate" : : "r" (pstate)); -outret: return insn; } @@ -153,7 +166,8 @@ show_signal_msg(struct pt_regs *regs, int sig, int code, } static void do_fault_siginfo(int code, int sig, struct pt_regs *regs, - unsigned int insn, int fault_code) + unsigned long fault_addr, unsigned int insn, + int fault_code) { unsigned long addr; siginfo_t info; @@ -161,10 +175,18 @@ static void do_fault_siginfo(int code, int sig, struct pt_regs *regs, info.si_code = code; info.si_signo = sig; info.si_errno = 0; - if (fault_code & FAULT_CODE_ITLB) + if (fault_code & FAULT_CODE_ITLB) { addr = regs->tpc; - else - addr = compute_effective_address(regs, insn, 0); + } else { + /* If we were able to probe the faulting instruction, use it + * to compute a precise fault address. Otherwise use the fault + * time provided address which may only have page granularity. + */ + if (insn) + addr = compute_effective_address(regs, insn, 0); + else + addr = fault_addr; + } info.si_addr = (void __user *) addr; info.si_trapno = 0; @@ -239,7 +261,7 @@ static void __kprobes do_kernel_fault(struct pt_regs *regs, int si_code, /* The si_code was set to make clear whether * this was a SEGV_MAPERR or SEGV_ACCERR fault. */ - do_fault_siginfo(si_code, SIGSEGV, regs, insn, fault_code); + do_fault_siginfo(si_code, SIGSEGV, regs, address, insn, fault_code); return; } @@ -525,7 +547,7 @@ do_sigbus: * Send a sigbus, regardless of whether we were in kernel * or user mode. */ - do_fault_siginfo(BUS_ADRERR, SIGBUS, regs, insn, fault_code); + do_fault_siginfo(BUS_ADRERR, SIGBUS, regs, address, insn, fault_code); /* Kernel mode? Handle exceptions or die */ if (regs->tstate & TSTATE_PRIV) -- cgit v1.2.3 From ee73887e92a69ae0a5cda21c68ea75a27804c944 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 29 Apr 2014 12:58:03 -0700 Subject: sparc64: Fix range check in kern_addr_valid(). In commit b2d438348024b75a1ee8b66b85d77f569a5dfed8 ("sparc64: Make PAGE_OFFSET variable."), the MAX_PHYS_ADDRESS_BITS value was increased (to 47). This constant reference to '41UL' was missed. Signed-off-by: David S. Miller --- arch/sparc/include/asm/pgtable_64.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h index ac5f7087a50d..ff97960b6242 100644 --- a/arch/sparc/include/asm/pgtable_64.h +++ b/arch/sparc/include/asm/pgtable_64.h @@ -916,7 +916,7 @@ static inline bool kern_addr_valid(unsigned long addr) { unsigned long paddr = __pa(addr); - if ((paddr >> 41UL) != 0UL) + if ((paddr >> MAX_PHYS_ADDRESS_BITS) != 0UL) return false; return test_bit(paddr >> 22, sparc64_valid_addr_bitmap); } -- cgit v1.2.3 From 0eef331a3d0ee970dcbebd1bd5fcb57ca33ece01 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Sat, 3 May 2014 22:52:50 -0700 Subject: sparc64: Use 'ILOG2_4MB' instead of constant '22'. Signed-off-by: David S. Miller --- arch/sparc/include/asm/pgtable_64.h | 2 +- arch/sparc/kernel/head_64.S | 4 ++-- arch/sparc/kernel/ktlb.S | 2 +- arch/sparc/mm/init_64.c | 12 ++++++------ 4 files changed, 10 insertions(+), 10 deletions(-) (limited to 'arch') diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h index ff97960b6242..0ad0a1285bd0 100644 --- a/arch/sparc/include/asm/pgtable_64.h +++ b/arch/sparc/include/asm/pgtable_64.h @@ -918,7 +918,7 @@ static inline bool kern_addr_valid(unsigned long addr) if ((paddr >> MAX_PHYS_ADDRESS_BITS) != 0UL) return false; - return test_bit(paddr >> 22, sparc64_valid_addr_bitmap); + return test_bit(paddr >> ILOG2_4MB, sparc64_valid_addr_bitmap); } extern int page_in_phys_avail(unsigned long paddr); diff --git a/arch/sparc/kernel/head_64.S b/arch/sparc/kernel/head_64.S index 26b706a1867d..452f04fe8da6 100644 --- a/arch/sparc/kernel/head_64.S +++ b/arch/sparc/kernel/head_64.S @@ -282,8 +282,8 @@ sun4v_chip_type: stx %l2, [%l4 + 0x0] ldx [%sp + 2047 + 128 + 0x50], %l3 ! physaddr low /* 4MB align */ - srlx %l3, 22, %l3 - sllx %l3, 22, %l3 + srlx %l3, ILOG2_4MB, %l3 + sllx %l3, ILOG2_4MB, %l3 stx %l3, [%l4 + 0x8] /* Leave service as-is, "call-method" */ diff --git a/arch/sparc/kernel/ktlb.S b/arch/sparc/kernel/ktlb.S index 542e96ac4d39..605d49204580 100644 --- a/arch/sparc/kernel/ktlb.S +++ b/arch/sparc/kernel/ktlb.S @@ -277,7 +277,7 @@ kvmap_dtlb_load: #ifdef CONFIG_SPARSEMEM_VMEMMAP kvmap_vmemmap: sub %g4, %g5, %g5 - srlx %g5, 22, %g5 + srlx %g5, ILOG2_4MB, %g5 sethi %hi(vmemmap_table), %g1 sllx %g5, 3, %g5 or %g1, %lo(vmemmap_table), %g1 diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c index eafbc65c9c47..ed3c969a5f4c 100644 --- a/arch/sparc/mm/init_64.c +++ b/arch/sparc/mm/init_64.c @@ -588,7 +588,7 @@ static void __init remap_kernel(void) int i, tlb_ent = sparc64_highest_locked_tlbent(); tte_vaddr = (unsigned long) KERNBASE; - phys_page = (prom_boot_mapping_phys_low >> 22UL) << 22UL; + phys_page = (prom_boot_mapping_phys_low >> ILOG2_4MB) << ILOG2_4MB; tte_data = kern_large_tte(phys_page); kern_locked_tte_data = tte_data; @@ -1881,7 +1881,7 @@ void __init paging_init(void) BUILD_BUG_ON(NR_CPUS > 4096); - kern_base = (prom_boot_mapping_phys_low >> 22UL) << 22UL; + kern_base = (prom_boot_mapping_phys_low >> ILOG2_4MB) << ILOG2_4MB; kern_size = (unsigned long)&_end - (unsigned long)KERNBASE; /* Invalidate both kernel TSBs. */ @@ -1937,7 +1937,7 @@ void __init paging_init(void) shift = kern_base + PAGE_OFFSET - ((unsigned long)KERNBASE); real_end = (unsigned long)_end; - num_kernel_image_mappings = DIV_ROUND_UP(real_end - KERNBASE, 1 << 22); + num_kernel_image_mappings = DIV_ROUND_UP(real_end - KERNBASE, 1 << ILOG2_4MB); printk("Kernel: Using %d locked TLB entries for main kernel image.\n", num_kernel_image_mappings); @@ -2094,7 +2094,7 @@ static void __init setup_valid_addr_bitmap_from_pavail(unsigned long *bitmap) if (new_start <= old_start && new_end >= (old_start + PAGE_SIZE)) { - set_bit(old_start >> 22, bitmap); + set_bit(old_start >> ILOG2_4MB, bitmap); goto do_next_page; } } @@ -2143,7 +2143,7 @@ void __init mem_init(void) addr = PAGE_OFFSET + kern_base; last = PAGE_ALIGN(kern_size) + addr; while (addr < last) { - set_bit(__pa(addr) >> 22, sparc64_valid_addr_bitmap); + set_bit(__pa(addr) >> ILOG2_4MB, sparc64_valid_addr_bitmap); addr += PAGE_SIZE; } @@ -2267,7 +2267,7 @@ int __meminit vmemmap_populate(unsigned long vstart, unsigned long vend, void *block; if (!(*vmem_pp & _PAGE_VALID)) { - block = vmemmap_alloc_block(1UL << 22, node); + block = vmemmap_alloc_block(1UL << ILOG2_4MB, node); if (!block) return -ENOMEM; -- cgit v1.2.3 From 26cf432551d749e7d581db33529507a711c6eaab Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 29 Apr 2014 13:03:27 -0700 Subject: sparc64: Add basic validations to {pud,pmd}_bad(). Instead of returning false we should at least check the most basic things, otherwise page table corruptions will be very difficult to debug. PMD and PTE tables are of size PAGE_SIZE, so none of the sub-PAGE_SIZE bits should be set. We also complement this with a check that the physical address the pud/pmd points to is valid memory. PowerPC was used as a guide while implementating this. Signed-off-by: David S. Miller --- arch/sparc/include/asm/pgtable_64.h | 46 +++++++++++++++++++++++++------------ 1 file changed, 31 insertions(+), 15 deletions(-) (limited to 'arch') diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h index 0ad0a1285bd0..baccf35621c0 100644 --- a/arch/sparc/include/asm/pgtable_64.h +++ b/arch/sparc/include/asm/pgtable_64.h @@ -71,6 +71,23 @@ #include +extern unsigned long sparc64_valid_addr_bitmap[]; + +/* Needs to be defined here and not in linux/mm.h, as it is arch dependent */ +static inline bool __kern_addr_valid(unsigned long paddr) +{ + if ((paddr >> MAX_PHYS_ADDRESS_BITS) != 0UL) + return false; + return test_bit(paddr >> ILOG2_4MB, sparc64_valid_addr_bitmap); +} + +static inline bool kern_addr_valid(unsigned long addr) +{ + unsigned long paddr = __pa(addr); + + return __kern_addr_valid(paddr); +} + /* Entries per page directory level. */ #define PTRS_PER_PTE (1UL << (PAGE_SHIFT-3)) #define PTRS_PER_PMD (1UL << PMD_BITS) @@ -743,6 +760,20 @@ static inline int pmd_present(pmd_t pmd) #define pmd_none(pmd) (!pmd_val(pmd)) +/* pmd_bad() is only called on non-trans-huge PMDs. Our encoding is + * very simple, it's just the physical address. PTE tables are of + * size PAGE_SIZE so make sure the sub-PAGE_SIZE bits are clear and + * the top bits outside of the range of any physical address size we + * support are clear as well. We also validate the physical itself. + */ +#define pmd_bad(pmd) ((pmd_val(pmd) & ~PAGE_MASK) || \ + !__kern_addr_valid(pmd_val(pmd))) + +#define pud_none(pud) (!pud_val(pud)) + +#define pud_bad(pud) ((pud_val(pud) & ~PAGE_MASK) || \ + !__kern_addr_valid(pud_val(pud))) + #ifdef CONFIG_TRANSPARENT_HUGEPAGE extern void set_pmd_at(struct mm_struct *mm, unsigned long addr, pmd_t *pmdp, pmd_t pmd); @@ -776,10 +807,7 @@ static inline unsigned long __pmd_page(pmd_t pmd) #define pud_page_vaddr(pud) \ ((unsigned long) __va(pud_val(pud))) #define pud_page(pud) virt_to_page((void *)pud_page_vaddr(pud)) -#define pmd_bad(pmd) (0) #define pmd_clear(pmdp) (pmd_val(*(pmdp)) = 0UL) -#define pud_none(pud) (!pud_val(pud)) -#define pud_bad(pud) (0) #define pud_present(pud) (pud_val(pud) != 0U) #define pud_clear(pudp) (pud_val(*(pudp)) = 0UL) @@ -909,18 +937,6 @@ extern unsigned long pte_file(pte_t); extern pte_t pgoff_to_pte(unsigned long); #define PTE_FILE_MAX_BITS (64UL - PAGE_SHIFT - 1UL) -extern unsigned long sparc64_valid_addr_bitmap[]; - -/* Needs to be defined here and not in linux/mm.h, as it is arch dependent */ -static inline bool kern_addr_valid(unsigned long addr) -{ - unsigned long paddr = __pa(addr); - - if ((paddr >> MAX_PHYS_ADDRESS_BITS) != 0UL) - return false; - return test_bit(paddr >> ILOG2_4MB, sparc64_valid_addr_bitmap); -} - extern int page_in_phys_avail(unsigned long paddr); /* -- cgit v1.2.3 From fe866433f843b080246ce729b5e6b27b5f5d9a58 Mon Sep 17 00:00:00 2001 From: "David S. Miller" Date: Tue, 29 Apr 2014 13:28:23 -0700 Subject: sparc64: Give more detailed information in {pgd,pmd}_ERROR() and kill pte_ERROR(). pte_ERROR() is not used anywhere, delete it. For pgd_ERROR() and pmd_ERROR(), output something similar to x86, giving the address of the pgd/pmd as well as it's value. Also provide the caller, since these macros are invoked from pgd_clear_bad() and pmd_clear_bad() which provides little context as to what high level operation was occuring when the BAD state was detected. Signed-off-by: David S. Miller --- arch/sparc/include/asm/pgtable_64.h | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'arch') diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h index baccf35621c0..fde5abaac0cc 100644 --- a/arch/sparc/include/asm/pgtable_64.h +++ b/arch/sparc/include/asm/pgtable_64.h @@ -96,9 +96,12 @@ static inline bool kern_addr_valid(unsigned long addr) /* Kernel has a separate 44bit address space. */ #define FIRST_USER_ADDRESS 0 -#define pte_ERROR(e) __builtin_trap() -#define pmd_ERROR(e) __builtin_trap() -#define pgd_ERROR(e) __builtin_trap() +#define pmd_ERROR(e) \ + pr_err("%s:%d: bad pmd %p(%016lx) seen at (%pS)\n", \ + __FILE__, __LINE__, &(e), pmd_val(e), __builtin_return_address(0)) +#define pgd_ERROR(e) \ + pr_err("%s:%d: bad pgd %p(%016lx) seen at (%pS)\n", \ + __FILE__, __LINE__, &(e), pgd_val(e), __builtin_return_address(0)) #endif /* !(__ASSEMBLY__) */ -- cgit v1.2.3 From 55b441be5cd600bf645a01b14900880a09941d4c Mon Sep 17 00:00:00 2001 From: Max Filippov Date: Thu, 1 May 2014 04:46:36 +0400 Subject: xtensa: ISS: don't depend on CONFIG_TTY Build console support only when CONFIG_TTY is selected. This restores ISS as the default platform for allnoconfig builds. Signed-off-by: Max Filippov Signed-off-by: Chris Zankel --- arch/xtensa/Kconfig | 1 - arch/xtensa/platforms/iss/Makefile | 3 ++- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/xtensa/Kconfig b/arch/xtensa/Kconfig index 17b31982c566..2ca8819979b4 100644 --- a/arch/xtensa/Kconfig +++ b/arch/xtensa/Kconfig @@ -243,7 +243,6 @@ choice config XTENSA_PLATFORM_ISS bool "ISS" - depends on TTY select XTENSA_CALIBRATE_CCOUNT select SERIAL_CONSOLE help diff --git a/arch/xtensa/platforms/iss/Makefile b/arch/xtensa/platforms/iss/Makefile index d2369b799c50..b3e89291cfba 100644 --- a/arch/xtensa/platforms/iss/Makefile +++ b/arch/xtensa/platforms/iss/Makefile @@ -4,6 +4,7 @@ # "prom monitor" library routines under Linux. # -obj-y = console.o setup.o +obj-y = setup.o +obj-$(CONFIG_TTY) += console.o obj-$(CONFIG_NET) += network.o obj-$(CONFIG_BLK_DEV_SIMDISK) += simdisk.o -- cgit v1.2.3 From ac008fe0a3236729751ccde655c215b436dfdaeb Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Mon, 5 May 2014 15:23:35 -0700 Subject: x86, build: Don't get confused by local symbols arch/x86/crypto/sha1_avx2_x86_64_asm.S introduced _end as a local symbol, which broke the build under certain circumstances. Although the wisdom of _end as a local symbol can definitely be questioned, the build should not break for that reason. Thus, filter the output of nm to only get global symbols of appropriate type. Reported-by: Andy Lutomirski Cc: Chandramouli Narayanan Cc: Herbert Xu Signed-off-by: H. Peter Anvin Link: http://lkml.kernel.org/n/tip-uxm3j3w3odglcwhafwq5tjqu@git.kernel.org --- arch/x86/boot/Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/x86/boot/Makefile b/arch/x86/boot/Makefile index abb9eba61b50..dbe8dd2fe247 100644 --- a/arch/x86/boot/Makefile +++ b/arch/x86/boot/Makefile @@ -71,7 +71,7 @@ $(obj)/vmlinux.bin: $(obj)/compressed/vmlinux FORCE SETUP_OBJS = $(addprefix $(obj)/,$(setup-y)) -sed-voffset := -e 's/^\([0-9a-fA-F]*\) . \(_text\|_end\)$$/\#define VO_\2 0x\1/p' +sed-voffset := -e 's/^\([0-9a-fA-F]*\) [ABCDGRSTVW] \(_text\|_end\)$$/\#define VO_\2 0x\1/p' quiet_cmd_voffset = VOFFSET $@ cmd_voffset = $(NM) $< | sed -n $(sed-voffset) > $@ @@ -80,7 +80,7 @@ targets += voffset.h $(obj)/voffset.h: vmlinux FORCE $(call if_changed,voffset) -sed-zoffset := -e 's/^\([0-9a-fA-F]*\) . \(startup_32\|startup_64\|efi32_stub_entry\|efi64_stub_entry\|efi_pe_entry\|input_data\|_end\|z_.*\)$$/\#define ZO_\2 0x\1/p' +sed-zoffset := -e 's/^\([0-9a-fA-F]*\) [ABCDGRSTVW] \(startup_32\|startup_64\|efi32_stub_entry\|efi64_stub_entry\|efi_pe_entry\|input_data\|_end\|z_.*\)$$/\#define ZO_\2 0x\1/p' quiet_cmd_zoffset = ZOFFSET $@ cmd_zoffset = $(NM) $< | sed -n $(sed-zoffset) > $@ -- cgit v1.2.3 From 2605fc216fa492f9e7c488bdc7f687cd6dcc703b Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Fri, 2 May 2014 00:44:37 +0200 Subject: asmlinkage, x86: Add explicit __visible to arch/x86/* As requested by Linus add explicit __visible to the asmlinkage users. This marks all functions visible to assembler. Tree sweep for arch/x86/* Signed-off-by: Andi Kleen Link: http://lkml.kernel.org/r/1398984278-29319-3-git-send-email-andi@firstfloor.org Signed-off-by: H. Peter Anvin --- arch/x86/boot/compressed/misc.c | 2 +- arch/x86/kernel/acpi/sleep.c | 2 +- arch/x86/kernel/apic/io_apic.c | 2 +- arch/x86/kernel/cpu/mcheck/therm_throt.c | 4 ++-- arch/x86/kernel/cpu/mcheck/threshold.c | 4 ++-- arch/x86/kernel/head32.c | 2 +- arch/x86/kernel/head64.c | 2 +- arch/x86/kernel/process_64.c | 2 +- arch/x86/kernel/smp.c | 2 +- arch/x86/kernel/traps.c | 6 +++--- arch/x86/kernel/vsmp_64.c | 6 +++--- arch/x86/kvm/x86.c | 2 +- arch/x86/lguest/boot.c | 4 ++-- arch/x86/math-emu/errors.c | 16 ++++++++-------- arch/x86/platform/olpc/olpc-xo1-pm.c | 2 +- arch/x86/power/hibernate_64.c | 2 +- arch/x86/xen/enlighten.c | 2 +- arch/x86/xen/irq.c | 6 +++--- 18 files changed, 34 insertions(+), 34 deletions(-) (limited to 'arch') diff --git a/arch/x86/boot/compressed/misc.c b/arch/x86/boot/compressed/misc.c index 17684615374b..57ab74df7eea 100644 --- a/arch/x86/boot/compressed/misc.c +++ b/arch/x86/boot/compressed/misc.c @@ -354,7 +354,7 @@ static void parse_elf(void *output) free(phdrs); } -asmlinkage void *decompress_kernel(void *rmode, memptr heap, +asmlinkage __visible void *decompress_kernel(void *rmode, memptr heap, unsigned char *input_data, unsigned long input_len, unsigned char *output, diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index 3a2ae4c88948..31368207837c 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -31,7 +31,7 @@ static char temp_stack[4096]; * * Wrapper around acpi_enter_sleep_state() to be called by assmebly. */ -acpi_status asmlinkage x86_acpi_enter_sleep_state(u8 state) +acpi_status asmlinkage __visible x86_acpi_enter_sleep_state(u8 state) { return acpi_enter_sleep_state(state); } diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index d23aa82e7a7b..992060e09897 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -2189,7 +2189,7 @@ void send_cleanup_vector(struct irq_cfg *cfg) cfg->move_in_progress = 0; } -asmlinkage void smp_irq_move_cleanup_interrupt(void) +asmlinkage __visible void smp_irq_move_cleanup_interrupt(void) { unsigned vector, me; diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c index d921b7ee6595..36a1bb6d1ee0 100644 --- a/arch/x86/kernel/cpu/mcheck/therm_throt.c +++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c @@ -429,14 +429,14 @@ static inline void __smp_thermal_interrupt(void) smp_thermal_vector(); } -asmlinkage void smp_thermal_interrupt(struct pt_regs *regs) +asmlinkage __visible void smp_thermal_interrupt(struct pt_regs *regs) { entering_irq(); __smp_thermal_interrupt(); exiting_ack_irq(); } -asmlinkage void smp_trace_thermal_interrupt(struct pt_regs *regs) +asmlinkage __visible void smp_trace_thermal_interrupt(struct pt_regs *regs) { entering_irq(); trace_thermal_apic_entry(THERMAL_APIC_VECTOR); diff --git a/arch/x86/kernel/cpu/mcheck/threshold.c b/arch/x86/kernel/cpu/mcheck/threshold.c index fe6b1c86645b..7245980186ee 100644 --- a/arch/x86/kernel/cpu/mcheck/threshold.c +++ b/arch/x86/kernel/cpu/mcheck/threshold.c @@ -24,14 +24,14 @@ static inline void __smp_threshold_interrupt(void) mce_threshold_vector(); } -asmlinkage void smp_threshold_interrupt(void) +asmlinkage __visible void smp_threshold_interrupt(void) { entering_irq(); __smp_threshold_interrupt(); exiting_ack_irq(); } -asmlinkage void smp_trace_threshold_interrupt(void) +asmlinkage __visible void smp_trace_threshold_interrupt(void) { entering_irq(); trace_threshold_apic_entry(THRESHOLD_APIC_VECTOR); diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c index c61a14a4a310..d6c1b9836995 100644 --- a/arch/x86/kernel/head32.c +++ b/arch/x86/kernel/head32.c @@ -29,7 +29,7 @@ static void __init i386_default_early_setup(void) reserve_ebda_region(); } -asmlinkage void __init i386_start_kernel(void) +asmlinkage __visible void __init i386_start_kernel(void) { sanitize_boot_params(&boot_params); diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c index 85126ccbdf6b..068054f4bf20 100644 --- a/arch/x86/kernel/head64.c +++ b/arch/x86/kernel/head64.c @@ -137,7 +137,7 @@ static void __init copy_bootdata(char *real_mode_data) } } -asmlinkage void __init x86_64_start_kernel(char * real_mode_data) +asmlinkage __visible void __init x86_64_start_kernel(char * real_mode_data) { int i; diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 9c0280f93d05..898d077617a9 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -52,7 +52,7 @@ asmlinkage extern void ret_from_fork(void); -asmlinkage DEFINE_PER_CPU(unsigned long, old_rsp); +__visible DEFINE_PER_CPU(unsigned long, old_rsp); /* Prints also some state that isn't saved in the pt_regs */ void __show_regs(struct pt_regs *regs, int all) diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c index 7c3a5a61f2e4..be8e1bde07aa 100644 --- a/arch/x86/kernel/smp.c +++ b/arch/x86/kernel/smp.c @@ -168,7 +168,7 @@ static int smp_stop_nmi_callback(unsigned int val, struct pt_regs *regs) * this function calls the 'stop' function on all other CPUs in the system. */ -asmlinkage void smp_reboot_interrupt(void) +asmlinkage __visible void smp_reboot_interrupt(void) { ack_APIC_irq(); irq_enter(); diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 57409f6b8c62..f73b5d435bdc 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -357,7 +357,7 @@ exit: * for scheduling or signal handling. The actual stack switch is done in * entry.S */ -asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs) +asmlinkage __visible __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs) { struct pt_regs *regs = eregs; /* Did already sync */ @@ -601,11 +601,11 @@ do_spurious_interrupt_bug(struct pt_regs *regs, long error_code) #endif } -asmlinkage void __attribute__((weak)) smp_thermal_interrupt(void) +asmlinkage __visible void __attribute__((weak)) smp_thermal_interrupt(void) { } -asmlinkage void __attribute__((weak)) smp_threshold_interrupt(void) +asmlinkage __visible void __attribute__((weak)) smp_threshold_interrupt(void) { } diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c index 5edc34b5b951..b99b9ad8540c 100644 --- a/arch/x86/kernel/vsmp_64.c +++ b/arch/x86/kernel/vsmp_64.c @@ -36,7 +36,7 @@ static int irq_routing_comply = 1; * and vice versa. */ -asmlinkage unsigned long vsmp_save_fl(void) +asmlinkage __visible unsigned long vsmp_save_fl(void) { unsigned long flags = native_save_fl(); @@ -56,7 +56,7 @@ __visible void vsmp_restore_fl(unsigned long flags) } PV_CALLEE_SAVE_REGS_THUNK(vsmp_restore_fl); -asmlinkage void vsmp_irq_disable(void) +asmlinkage __visible void vsmp_irq_disable(void) { unsigned long flags = native_save_fl(); @@ -64,7 +64,7 @@ asmlinkage void vsmp_irq_disable(void) } PV_CALLEE_SAVE_REGS_THUNK(vsmp_irq_disable); -asmlinkage void vsmp_irq_enable(void) +asmlinkage __visible void vsmp_irq_enable(void) { unsigned long flags = native_save_fl(); diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 8b8fc0b792ba..b6c0bacca9bd 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -280,7 +280,7 @@ int kvm_set_apic_base(struct kvm_vcpu *vcpu, struct msr_data *msr_info) } EXPORT_SYMBOL_GPL(kvm_set_apic_base); -asmlinkage void kvm_spurious_fault(void) +asmlinkage __visible void kvm_spurious_fault(void) { /* Fault while not rebooting. We want the trace. */ BUG(); diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c index ad1fb5f53925..aae94132bc24 100644 --- a/arch/x86/lguest/boot.c +++ b/arch/x86/lguest/boot.c @@ -233,13 +233,13 @@ static void lguest_end_context_switch(struct task_struct *next) * flags word contains all kind of stuff, but in practice Linux only cares * about the interrupt flag. Our "save_flags()" just returns that. */ -asmlinkage unsigned long lguest_save_fl(void) +asmlinkage __visible unsigned long lguest_save_fl(void) { return lguest_data.irq_enabled; } /* Interrupts go off... */ -asmlinkage void lguest_irq_disable(void) +asmlinkage __visible void lguest_irq_disable(void) { lguest_data.irq_enabled = 0; } diff --git a/arch/x86/math-emu/errors.c b/arch/x86/math-emu/errors.c index a5449089cd9f..9e6545f269e5 100644 --- a/arch/x86/math-emu/errors.c +++ b/arch/x86/math-emu/errors.c @@ -302,7 +302,7 @@ static struct { 0x242 in div_Xsig.S */ -asmlinkage void FPU_exception(int n) +asmlinkage __visible void FPU_exception(int n) { int i, int_type; @@ -492,7 +492,7 @@ int real_2op_NaN(FPU_REG const *b, u_char tagb, /* Invalid arith operation on Valid registers */ /* Returns < 0 if the exception is unmasked */ -asmlinkage int arith_invalid(int deststnr) +asmlinkage __visible int arith_invalid(int deststnr) { EXCEPTION(EX_Invalid); @@ -507,7 +507,7 @@ asmlinkage int arith_invalid(int deststnr) } /* Divide a finite number by zero */ -asmlinkage int FPU_divide_by_zero(int deststnr, u_char sign) +asmlinkage __visible int FPU_divide_by_zero(int deststnr, u_char sign) { FPU_REG *dest = &st(deststnr); int tag = TAG_Valid; @@ -539,7 +539,7 @@ int set_precision_flag(int flags) } /* This may be called often, so keep it lean */ -asmlinkage void set_precision_flag_up(void) +asmlinkage __visible void set_precision_flag_up(void) { if (control_word & CW_Precision) partial_status |= (SW_Precision | SW_C1); /* The masked response */ @@ -548,7 +548,7 @@ asmlinkage void set_precision_flag_up(void) } /* This may be called often, so keep it lean */ -asmlinkage void set_precision_flag_down(void) +asmlinkage __visible void set_precision_flag_down(void) { if (control_word & CW_Precision) { /* The masked response */ partial_status &= ~SW_C1; @@ -557,7 +557,7 @@ asmlinkage void set_precision_flag_down(void) EXCEPTION(EX_Precision); } -asmlinkage int denormal_operand(void) +asmlinkage __visible int denormal_operand(void) { if (control_word & CW_Denormal) { /* The masked response */ partial_status |= SW_Denorm_Op; @@ -568,7 +568,7 @@ asmlinkage int denormal_operand(void) } } -asmlinkage int arith_overflow(FPU_REG *dest) +asmlinkage __visible int arith_overflow(FPU_REG *dest) { int tag = TAG_Valid; @@ -596,7 +596,7 @@ asmlinkage int arith_overflow(FPU_REG *dest) } -asmlinkage int arith_underflow(FPU_REG *dest) +asmlinkage __visible int arith_underflow(FPU_REG *dest) { int tag = TAG_Valid; diff --git a/arch/x86/platform/olpc/olpc-xo1-pm.c b/arch/x86/platform/olpc/olpc-xo1-pm.c index ff0174dda810..a9acde72d4ed 100644 --- a/arch/x86/platform/olpc/olpc-xo1-pm.c +++ b/arch/x86/platform/olpc/olpc-xo1-pm.c @@ -75,7 +75,7 @@ static int xo1_power_state_enter(suspend_state_t pm_state) return 0; } -asmlinkage int xo1_do_sleep(u8 sleep_state) +asmlinkage __visible int xo1_do_sleep(u8 sleep_state) { void *pgd_addr = __va(read_cr3()); diff --git a/arch/x86/power/hibernate_64.c b/arch/x86/power/hibernate_64.c index 304fca20d96e..35e2bb6c0f37 100644 --- a/arch/x86/power/hibernate_64.c +++ b/arch/x86/power/hibernate_64.c @@ -23,7 +23,7 @@ extern __visible const void __nosave_begin, __nosave_end; /* Defined in hibernate_asm_64.S */ -extern asmlinkage int restore_image(void); +extern asmlinkage __visible int restore_image(void); /* * Address to jump to in the last phase of restore in order to get to the image diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 201d09a7c46b..c34bfc4bbe7f 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -1515,7 +1515,7 @@ static void __init xen_pvh_early_guest_init(void) } /* First C function to be called on Xen boot */ -asmlinkage void __init xen_start_kernel(void) +asmlinkage __visible void __init xen_start_kernel(void) { struct physdev_set_iopl set_iopl; int rc; diff --git a/arch/x86/xen/irq.c b/arch/x86/xen/irq.c index 08f763de26fe..a1207cb6472a 100644 --- a/arch/x86/xen/irq.c +++ b/arch/x86/xen/irq.c @@ -23,7 +23,7 @@ void xen_force_evtchn_callback(void) (void)HYPERVISOR_xen_version(0, NULL); } -asmlinkage unsigned long xen_save_fl(void) +asmlinkage __visible unsigned long xen_save_fl(void) { struct vcpu_info *vcpu; unsigned long flags; @@ -63,7 +63,7 @@ __visible void xen_restore_fl(unsigned long flags) } PV_CALLEE_SAVE_REGS_THUNK(xen_restore_fl); -asmlinkage void xen_irq_disable(void) +asmlinkage __visible void xen_irq_disable(void) { /* There's a one instruction preempt window here. We need to make sure we're don't switch CPUs between getting the vcpu @@ -74,7 +74,7 @@ asmlinkage void xen_irq_disable(void) } PV_CALLEE_SAVE_REGS_THUNK(xen_irq_disable); -asmlinkage void xen_irq_enable(void) +asmlinkage __visible void xen_irq_enable(void) { struct vcpu_info *vcpu; -- cgit v1.2.3 From aadca6fa4068ad1f92c492bc8507b7ed350825a2 Mon Sep 17 00:00:00 2001 From: Christian Gmeiner Date: Wed, 7 May 2014 09:01:54 +0200 Subject: x86/reboot: Add reboot quirk for Certec BPC600 Certec BPC600 needs reboot=pci to actually reboot. Signed-off-by: Christian Gmeiner Cc: Matthew Garrett Cc: Li Aubrey Cc: Andrew Morton Cc: Dave Jones Cc: Fenghua Yu Cc: Linus Torvalds Link: http://lkml.kernel.org/r/1399446114-2147-1-git-send-email-christian.gmeiner@gmail.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/reboot.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'arch') diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 3399d3a99730..52b1157c53eb 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -191,6 +191,16 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = { }, }, + /* Certec */ + { /* Handle problems with rebooting on Certec BPC600 */ + .callback = set_pci_reboot, + .ident = "Certec BPC600", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Certec"), + DMI_MATCH(DMI_PRODUCT_NAME, "BPC600"), + }, + }, + /* Dell */ { /* Handle problems with rebooting on Dell DXP061 */ .callback = set_bios_reboot, -- cgit v1.2.3 From 14262d67fe348018af368a07430fbc06eadeabb1 Mon Sep 17 00:00:00 2001 From: George Spelvin Date: Wed, 7 May 2014 17:05:52 -0400 Subject: x86-64, build: Fix stack protector Makefile breakage with 32-bit userland If you are using a 64-bit kernel with 32-bit userland, then scripts/gcc-x86_64-has-stack-protector.sh invokes 32-bit gcc with -mcmodel=kernel, which produces: :1:0: error: code model 'kernel' not supported in the 32 bit mode and trips the "broken compiler" test at arch/x86/Makefile:120. There are several places a fix is possible, but the following seems cleanest. (But it's minimal; it would also be possible to factor out a bunch of stuff from the two branches of the if.) Signed-off-by: George Spelvin Link: http://lkml.kernel.org/r/20140507210552.7581.qmail@ns.horizon.com Cc: # v3.14 Signed-off-by: H. Peter Anvin --- arch/x86/Makefile | 1 + 1 file changed, 1 insertion(+) (limited to 'arch') diff --git a/arch/x86/Makefile b/arch/x86/Makefile index ce6ad7e6a7d7..33f71b01fd22 100644 --- a/arch/x86/Makefile +++ b/arch/x86/Makefile @@ -79,6 +79,7 @@ else UTS_MACHINE := x86_64 CHECKFLAGS += -D__x86_64__ -m64 + biarch := -m64 KBUILD_AFLAGS += -m64 KBUILD_CFLAGS += -m64 -- cgit v1.2.3 From f10f383d8414bfe3357e24432ed8a26eeb58ffb8 Mon Sep 17 00:00:00 2001 From: Feng Tang Date: Thu, 24 Apr 2014 16:18:17 +0800 Subject: x86/hpet: Make boot_hpet_disable extern HPET on some platform has accuracy problem. Making "boot_hpet_disable" extern so that we can runtime disable the HPET timer by using quirk to check the platform. Signed-off-by: Feng Tang Cc: Clemens Ladisch Cc: John Stultz Cc: Linus Torvalds Link: http://lkml.kernel.org/r/1398327498-13163-1-git-send-email-feng.tang@intel.com Signed-off-by: Ingo Molnar --- arch/x86/include/asm/hpet.h | 1 + arch/x86/kernel/hpet.c | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/x86/include/asm/hpet.h b/arch/x86/include/asm/hpet.h index b18df579c0e9..36f7125945e3 100644 --- a/arch/x86/include/asm/hpet.h +++ b/arch/x86/include/asm/hpet.h @@ -63,6 +63,7 @@ /* hpet memory map physical address */ extern unsigned long hpet_address; extern unsigned long force_hpet_address; +extern int boot_hpet_disable; extern u8 hpet_blockid; extern int hpet_force_user; extern u8 hpet_msi_disable; diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 8d80ae011603..4177bfbc80b0 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -88,7 +88,7 @@ static inline void hpet_clear_mapping(void) /* * HPET command line enable / disable */ -static int boot_hpet_disable; +int boot_hpet_disable; int hpet_force_user; static int hpet_verbose; -- cgit v1.2.3 From 62187910b0fc7a75cfec9c30fda58ce2f39d689b Mon Sep 17 00:00:00 2001 From: Feng Tang Date: Thu, 24 Apr 2014 16:18:18 +0800 Subject: x86/intel: Add quirk to disable HPET for the Baytrail platform HPET on current Baytrail platform has accuracy problem to be used as reliable clocksource/clockevent, so add a early quirk to disable it. Signed-off-by: Feng Tang Cc: Clemens Ladisch Cc: John Stultz Cc: Linus Torvalds Link: http://lkml.kernel.org/r/1398327498-13163-2-git-send-email-feng.tang@intel.com Signed-off-by: Ingo Molnar --- arch/x86/kernel/early-quirks.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) (limited to 'arch') diff --git a/arch/x86/kernel/early-quirks.c b/arch/x86/kernel/early-quirks.c index 6e2537c32190..6cda0baeac9d 100644 --- a/arch/x86/kernel/early-quirks.c +++ b/arch/x86/kernel/early-quirks.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -530,6 +531,15 @@ static void __init intel_graphics_stolen(int num, int slot, int func) } } +static void __init force_disable_hpet(int num, int slot, int func) +{ +#ifdef CONFIG_HPET_TIMER + boot_hpet_disable = 1; + pr_info("x86/hpet: Will disable the HPET for this platform because it's not reliable\n"); +#endif +} + + #define QFLAG_APPLY_ONCE 0x1 #define QFLAG_APPLIED 0x2 #define QFLAG_DONE (QFLAG_APPLY_ONCE|QFLAG_APPLIED) @@ -567,6 +577,12 @@ static struct chipset early_qrk[] __initdata = { PCI_BASE_CLASS_BRIDGE, 0, intel_remapping_check }, { PCI_VENDOR_ID_INTEL, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA, PCI_ANY_ID, QFLAG_APPLY_ONCE, intel_graphics_stolen }, + /* + * HPET on current version of Baytrail platform has accuracy + * problems, disable it for now: + */ + { PCI_VENDOR_ID_INTEL, 0x0f00, + PCI_CLASS_BRIDGE_HOST, PCI_ANY_ID, 0, force_disable_hpet}, {} }; -- cgit v1.2.3 From 722a0d22d028bd74061cf582de1764884e73674f Mon Sep 17 00:00:00 2001 From: Andres Freund Date: Fri, 9 May 2014 03:29:16 +0200 Subject: x86: Fix typo preventing msr_set/clear_bit from having an effect Due to a typo the msr accessor function introduced in 22085a66c2fab6cf9b9393c056a3600a6b4735de didn't have any lasting effects because they accidentally wrote the old value back. After c0a639ad0bc6b178b46996bd1f821a04643e2bde this at the very least this causes cpuid limits not to be lifted on some cpus leading to missing capabilities for those. Signed-off-by: Andres Freund Link: http://lkml.kernel.org/r/1399598957-7011-2-git-send-email-andres@anarazel.de Cc: Borislav Petkov Signed-off-by: H. Peter Anvin --- arch/x86/lib/msr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/x86/lib/msr.c b/arch/x86/lib/msr.c index db9db446b71a..43623739c7cf 100644 --- a/arch/x86/lib/msr.c +++ b/arch/x86/lib/msr.c @@ -76,7 +76,7 @@ static inline int __flip_bit(u32 msr, u8 bit, bool set) if (m1.q == m.q) return 0; - err = msr_write(msr, &m); + err = msr_write(msr, &m1); if (err) return err; -- cgit v1.2.3 From c45f77364ba060395b7eff1bf45e6c537f913380 Mon Sep 17 00:00:00 2001 From: Andres Freund Date: Fri, 9 May 2014 03:29:17 +0200 Subject: x86: Fix typo in MSR_IA32_MISC_ENABLE_LIMIT_CPUID macro The spuriously added semicolon didn't have any effect because the macro isn't currently in use. c0a639ad0bc6b178b46996bd1f821a04643e2bde Signed-off-by: Andres Freund Link: http://lkml.kernel.org/r/1399598957-7011-3-git-send-email-andres@anarazel.de Cc: Borislav Petkov Signed-off-by: H. Peter Anvin --- arch/x86/include/uapi/asm/msr-index.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/x86/include/uapi/asm/msr-index.h b/arch/x86/include/uapi/asm/msr-index.h index c827ace3121b..fcf2b3ae1bf0 100644 --- a/arch/x86/include/uapi/asm/msr-index.h +++ b/arch/x86/include/uapi/asm/msr-index.h @@ -384,7 +384,7 @@ #define MSR_IA32_MISC_ENABLE_MWAIT_BIT 18 #define MSR_IA32_MISC_ENABLE_MWAIT (1ULL << MSR_IA32_MISC_ENABLE_MWAIT_BIT) #define MSR_IA32_MISC_ENABLE_LIMIT_CPUID_BIT 22 -#define MSR_IA32_MISC_ENABLE_LIMIT_CPUID (1ULL << MSR_IA32_MISC_ENABLE_LIMIT_CPUID_BIT); +#define MSR_IA32_MISC_ENABLE_LIMIT_CPUID (1ULL << MSR_IA32_MISC_ENABLE_LIMIT_CPUID_BIT) #define MSR_IA32_MISC_ENABLE_XTPR_DISABLE_BIT 23 #define MSR_IA32_MISC_ENABLE_XTPR_DISABLE (1ULL << MSR_IA32_MISC_ENABLE_XTPR_DISABLE_BIT) #define MSR_IA32_MISC_ENABLE_XD_DISABLE_BIT 34 -- cgit v1.2.3 From 28b92e09e25bdc0ae864b22eacf195a74f861389 Mon Sep 17 00:00:00 2001 From: Boris Ostrovsky Date: Fri, 9 May 2014 11:11:27 -0400 Subject: x86, vdso, time: Cast tv_nsec to u64 for proper shifting in update_vsyscall() With tk->wall_to_monotonic.tv_nsec being a 32-bit value on 32-bit systems, (tk->wall_to_monotonic.tv_nsec << tk->shift) in update_vsyscall() may lose upper bits or, worse, add them since compiler will do this: (u64)(tk->wall_to_monotonic.tv_nsec << tk->shift) instead of ((u64)tk->wall_to_monotonic.tv_nsec << tk->shift) So if, for example, tv_nsec is 0x800000 and shift is 8 we will end up with 0xffffffff80000000 instead of 0x80000000. And then we are stuck in the subsequent 'while' loop. We need an explicit cast. Signed-off-by: Boris Ostrovsky Link: http://lkml.kernel.org/r/1399648287-15178-1-git-send-email-boris.ostrovsky@oracle.com Acked-by: Konrad Rzeszutek Wilk Cc: # v3.14 Signed-off-by: H. Peter Anvin --- arch/x86/kernel/vsyscall_gtod.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch') diff --git a/arch/x86/kernel/vsyscall_gtod.c b/arch/x86/kernel/vsyscall_gtod.c index f9c6e56e14b5..9531fbb123ba 100644 --- a/arch/x86/kernel/vsyscall_gtod.c +++ b/arch/x86/kernel/vsyscall_gtod.c @@ -43,7 +43,7 @@ void update_vsyscall(struct timekeeper *tk) vdata->monotonic_time_sec = tk->xtime_sec + tk->wall_to_monotonic.tv_sec; vdata->monotonic_time_snsec = tk->xtime_nsec - + (tk->wall_to_monotonic.tv_nsec + + ((u64)tk->wall_to_monotonic.tv_nsec << tk->shift); while (vdata->monotonic_time_snsec >= (((u64)NSEC_PER_SEC) << tk->shift)) { -- cgit v1.2.3