From 9d3a4de4cb8db8e71730e36736272ef041836f68 Mon Sep 17 00:00:00 2001 From: Robin Murphy Date: Thu, 16 Mar 2017 17:00:16 +0000 Subject: iommu: Disambiguate MSI region types The introduction of reserved regions has left a couple of rough edges which we could do with sorting out sooner rather than later. Since we are not yet addressing the potential dynamic aspect of software-managed reservations and presenting them at arbitrary fixed addresses, it is incongruous that we end up displaying hardware vs. software-managed MSI regions to userspace differently, especially since ARM-based systems may actually require one or the other, or even potentially both at once, (which iommu-dma currently has no hope of dealing with at all). Let's resolve the former user-visible inconsistency ASAP before the ABI has been baked into a kernel release, in a way that also lays the groundwork for the latter shortcoming to be addressed by follow-up patches. For clarity, rename the software-managed type to IOMMU_RESV_SW_MSI, use IOMMU_RESV_MSI to describe the hardware type, and document everything a little bit. Since the x86 MSI remapping hardware falls squarely under this meaning of IOMMU_RESV_MSI, apply that type to their regions as well, so that we tell the same story to userspace across all platforms. Secondly, as the various region types require quite different handling, and it really makes little sense to ever try combining them, convert the bitfield-esque #defines to a plain enum in the process before anyone gets the wrong impression. Fixes: d30ddcaa7b02 ("iommu: Add a new type field in iommu_resv_region") Reviewed-by: Eric Auger CC: Alex Williamson CC: David Woodhouse CC: kvm@vger.kernel.org Signed-off-by: Robin Murphy Signed-off-by: Joerg Roedel --- drivers/iommu/iommu.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'drivers/iommu/iommu.c') diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c index 8ea14f41a979..3b67144dead2 100644 --- a/drivers/iommu/iommu.c +++ b/drivers/iommu/iommu.c @@ -72,6 +72,7 @@ static const char * const iommu_group_resv_type_string[] = { [IOMMU_RESV_DIRECT] = "direct", [IOMMU_RESV_RESERVED] = "reserved", [IOMMU_RESV_MSI] = "msi", + [IOMMU_RESV_SW_MSI] = "msi", }; #define IOMMU_GROUP_ATTR(_name, _mode, _show, _store) \ @@ -1743,8 +1744,8 @@ void iommu_put_resv_regions(struct device *dev, struct list_head *list) } struct iommu_resv_region *iommu_alloc_resv_region(phys_addr_t start, - size_t length, - int prot, int type) + size_t length, int prot, + enum iommu_resv_type type) { struct iommu_resv_region *region; -- cgit v1.2.3 From fccb4e3b8ab0957628abec82675691c72f67003e Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Thu, 5 Jan 2017 18:38:26 +0000 Subject: iommu: Allow default domain type to be set on the kernel command line The IOMMU core currently initialises the default domain for each group to IOMMU_DOMAIN_DMA, under the assumption that devices will use IOMMU-backed DMA ops by default. However, in some cases it is desirable for the DMA ops to bypass the IOMMU for performance reasons, reserving use of translation for subsystems such as VFIO that require it for enforcing device isolation. Rather than modify each IOMMU driver to provide different semantics for DMA domains, instead we introduce a command line parameter that can be used to change the type of the default domain. Passthrough can then be specified using "iommu.passthrough=1" on the kernel command line. Signed-off-by: Will Deacon --- Documentation/admin-guide/kernel-parameters.txt | 6 ++++++ drivers/iommu/iommu.c | 28 ++++++++++++++++++++++--- 2 files changed, 31 insertions(+), 3 deletions(-) (limited to 'drivers/iommu/iommu.c') diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index 2ba45caabada..187ac5bc8b40 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -1644,6 +1644,12 @@ nobypass [PPC/POWERNV] Disable IOMMU bypass, using IOMMU for PCI devices. + iommu.passthrough= + [ARM64] Configure DMA to bypass the IOMMU by default. + Format: { "0" | "1" } + 0 - Use IOMMU translation for DMA. + 1 - Bypass the IOMMU for DMA. + unset - Use IOMMU translation for DMA. io7= [HW] IO7 for Marvel based alpha systems See comment before marvel_specify_io7 in diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c index 3b67144dead2..770ba7e7ef4d 100644 --- a/drivers/iommu/iommu.c +++ b/drivers/iommu/iommu.c @@ -36,6 +36,7 @@ static struct kset *iommu_group_kset; static DEFINE_IDA(iommu_group_ida); +static unsigned int iommu_def_domain_type = IOMMU_DOMAIN_DMA; struct iommu_callback_data { const struct iommu_ops *ops; @@ -112,6 +113,18 @@ static int __iommu_attach_group(struct iommu_domain *domain, static void __iommu_detach_group(struct iommu_domain *domain, struct iommu_group *group); +static int __init iommu_set_def_domain_type(char *str) +{ + bool pt; + + if (!str || strtobool(str, &pt)) + return -EINVAL; + + iommu_def_domain_type = pt ? IOMMU_DOMAIN_IDENTITY : IOMMU_DOMAIN_DMA; + return 0; +} +early_param("iommu.passthrough", iommu_set_def_domain_type); + static ssize_t iommu_group_attr_show(struct kobject *kobj, struct attribute *__attr, char *buf) { @@ -1015,10 +1028,19 @@ struct iommu_group *iommu_group_get_for_dev(struct device *dev) * IOMMU driver. */ if (!group->default_domain) { - group->default_domain = __iommu_domain_alloc(dev->bus, - IOMMU_DOMAIN_DMA); + struct iommu_domain *dom; + + dom = __iommu_domain_alloc(dev->bus, iommu_def_domain_type); + if (!dom && iommu_def_domain_type != IOMMU_DOMAIN_DMA) { + dev_warn(dev, + "failed to allocate default IOMMU domain of type %u; falling back to IOMMU_DOMAIN_DMA", + iommu_def_domain_type); + dom = __iommu_domain_alloc(dev->bus, IOMMU_DOMAIN_DMA); + } + + group->default_domain = dom; if (!group->domain) - group->domain = group->default_domain; + group->domain = dom; } ret = iommu_group_add_device(group, dev); -- cgit v1.2.3 From 3ba8775f64484a2b56bf3c88d09a186d819fa348 Mon Sep 17 00:00:00 2001 From: "zhichang.yuan" Date: Tue, 18 Apr 2017 20:51:48 +0800 Subject: iommu: Make iommu_bus_notifier return NOTIFY_DONE rather than error code In iommu_bus_notifier(), when action is BUS_NOTIFY_ADD_DEVICE, it will return 'ops->add_device(dev)' directly. But ops->add_device will return ERR_VAL, such as -ENODEV. These value will make notifier_call_chain() not to traverse the remain nodes in struct notifier_block list. This patch revises iommu_bus_notifier() to return NOTIFY_DONE when some errors happened in ops->add_device(). Signed-off-by: zhichang.yuan Signed-off-by: Joerg Roedel --- drivers/iommu/iommu.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'drivers/iommu/iommu.c') diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c index 8ea14f41a979..9170fd498f46 100644 --- a/drivers/iommu/iommu.c +++ b/drivers/iommu/iommu.c @@ -1082,8 +1082,12 @@ static int iommu_bus_notifier(struct notifier_block *nb, * result in ADD/DEL notifiers to group->notifier */ if (action == BUS_NOTIFY_ADD_DEVICE) { - if (ops->add_device) - return ops->add_device(dev); + if (ops->add_device) { + int ret; + + ret = ops->add_device(dev); + return (ret) ? NOTIFY_DONE : NOTIFY_OK; + } } else if (action == BUS_NOTIFY_REMOVED_DEVICE) { if (ops->remove_device && dev->iommu_group) { ops->remove_device(dev); -- cgit v1.2.3 From 207c6e36f122ebb1164d611c9f34f128313f47d5 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Wed, 26 Apr 2017 15:39:28 +0200 Subject: iommu: Move report_iommu_fault() to iommu.c The function is in no fast-path, there is no need for it to be static inline in a header file. This also removes the need to include iommu trace-points in iommu.h. Signed-off-by: Joerg Roedel --- drivers/iommu/iommu.c | 42 ++++++++++++++++++++++++++++++++++++++++++ include/linux/iommu.h | 41 ++--------------------------------------- 2 files changed, 44 insertions(+), 39 deletions(-) (limited to 'drivers/iommu/iommu.c') diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c index 9170fd498f46..4cb5792cbc21 100644 --- a/drivers/iommu/iommu.c +++ b/drivers/iommu/iommu.c @@ -1655,6 +1655,48 @@ void iommu_domain_window_disable(struct iommu_domain *domain, u32 wnd_nr) } EXPORT_SYMBOL_GPL(iommu_domain_window_disable); +/** + * report_iommu_fault() - report about an IOMMU fault to the IOMMU framework + * @domain: the iommu domain where the fault has happened + * @dev: the device where the fault has happened + * @iova: the faulting address + * @flags: mmu fault flags (e.g. IOMMU_FAULT_READ/IOMMU_FAULT_WRITE/...) + * + * This function should be called by the low-level IOMMU implementations + * whenever IOMMU faults happen, to allow high-level users, that are + * interested in such events, to know about them. + * + * This event may be useful for several possible use cases: + * - mere logging of the event + * - dynamic TLB/PTE loading + * - if restarting of the faulting device is required + * + * Returns 0 on success and an appropriate error code otherwise (if dynamic + * PTE/TLB loading will one day be supported, implementations will be able + * to tell whether it succeeded or not according to this return value). + * + * Specifically, -ENOSYS is returned if a fault handler isn't installed + * (though fault handlers can also return -ENOSYS, in case they want to + * elicit the default behavior of the IOMMU drivers). + */ +int report_iommu_fault(struct iommu_domain *domain, struct device *dev, + unsigned long iova, int flags) +{ + int ret = -ENOSYS; + + /* + * if upper layers showed interest and installed a fault handler, + * invoke it. + */ + if (domain->handler) + ret = domain->handler(domain, dev, iova, flags, + domain->handler_token); + + trace_io_page_fault(dev, iova, flags); + return ret; +} +EXPORT_SYMBOL_GPL(report_iommu_fault); + static int __init iommu_init(void) { iommu_group_kset = kset_create_and_add("iommu_groups", diff --git a/include/linux/iommu.h b/include/linux/iommu.h index 3b4fe4b79d20..abaa0ca848bc 100644 --- a/include/linux/iommu.h +++ b/include/linux/iommu.h @@ -330,46 +330,9 @@ extern int iommu_domain_window_enable(struct iommu_domain *domain, u32 wnd_nr, phys_addr_t offset, u64 size, int prot); extern void iommu_domain_window_disable(struct iommu_domain *domain, u32 wnd_nr); -/** - * report_iommu_fault() - report about an IOMMU fault to the IOMMU framework - * @domain: the iommu domain where the fault has happened - * @dev: the device where the fault has happened - * @iova: the faulting address - * @flags: mmu fault flags (e.g. IOMMU_FAULT_READ/IOMMU_FAULT_WRITE/...) - * - * This function should be called by the low-level IOMMU implementations - * whenever IOMMU faults happen, to allow high-level users, that are - * interested in such events, to know about them. - * - * This event may be useful for several possible use cases: - * - mere logging of the event - * - dynamic TLB/PTE loading - * - if restarting of the faulting device is required - * - * Returns 0 on success and an appropriate error code otherwise (if dynamic - * PTE/TLB loading will one day be supported, implementations will be able - * to tell whether it succeeded or not according to this return value). - * - * Specifically, -ENOSYS is returned if a fault handler isn't installed - * (though fault handlers can also return -ENOSYS, in case they want to - * elicit the default behavior of the IOMMU drivers). - */ -static inline int report_iommu_fault(struct iommu_domain *domain, - struct device *dev, unsigned long iova, int flags) -{ - int ret = -ENOSYS; - /* - * if upper layers showed interest and installed a fault handler, - * invoke it. - */ - if (domain->handler) - ret = domain->handler(domain, dev, iova, flags, - domain->handler_token); - - trace_io_page_fault(dev, iova, flags); - return ret; -} +extern int report_iommu_fault(struct iommu_domain *domain, struct device *dev, + unsigned long iova, int flags); static inline size_t iommu_map_sg(struct iommu_domain *domain, unsigned long iova, struct scatterlist *sg, -- cgit v1.2.3