iommu/vt-d: patches for v5.8
Subsystems (2)
Lu Baolu
maintainerHi Joerg,
Below patches have been piled up for v5.8. They enable below features:
- Nested Shared Virtual Address (SVA) VT-d support
- Replace Intel SVM with IOMMU SVA APIs
- SVA page request draining support
Can you please consider them for iommu/next?
Best regards,
Lu Baolu
Jacob Pan (9):
iommu/vt-d: Move domain helper to header iommu/vt-d: Use a helper function to skip agaw for SL iommu/vt-d: Add nested translation helper function iommu/vt-d: Add bind guest PASID support iommu/vt-d: Support flushing more translation cache types iommu/vt-d: Add svm/sva invalidate function iommu/vt-d: Add custom allocator for IOASID iommu/vt-d: Report SVA feature with generic flag iommu/vt-d: Replace intel SVM APIs with generic SVA APIs
Lu Baolu (8):
iommu/vt-d: Enlightened PASID allocation
iommu/vt-d: Add get_domain_info() helper
iommu/vt-d: Multiple descriptors per qi_submit_sync() iommu/vt-d: debugfs: Add support to show inv queue internals iommu/vt-d: Disable non-recoverable fault processing before unbind iommu/vt-d: Add page request draining support iommu/vt-d: Remove redundant IOTLB flush
iommu/vt-d: Remove duplicated check in intel_svm_bind_mm()
Tom Murphy (1):
iommu/vt-d: Remove IOVA handling code from the non-dma_ops path
drivers/iommu/dmar.c | 99 ++++-- drivers/iommu/intel-iommu-debugfs.c | 62 ++++ drivers/iommu/intel-iommu.c | 449 +++++++++++++++++++++------- drivers/iommu/intel-pasid.c | 309 +++++++++++++++++-- drivers/iommu/intel-pasid.h | 27 +- drivers/iommu/intel-svm.c | 445 ++++++++++++++++++++++----- drivers/iommu/intel_irq_remapping.c | 2 +- include/linux/intel-iommu.h | 81 ++++- include/linux/intel-svm.h | 94 +----- include/uapi/linux/iommu.h | 5 + 10 files changed, 1252 insertions(+), 321 deletions(-)
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 2/18 iommu/vt-d: Use a helper function to skip agaw for SL
From: Jacob Pan <[email protected]>
An Intel iommu domain uses 5-level page table by default. If the iommu that the domain tries to attach supports less page levels, the top level page tables should be skipped. Add a helper to do this so that it could be used in other places.
--- drivers/iommu/intel-pasid.c | 33 +++++++++++++++++++++++---------- 1 file changed, 23 insertions(+), 10 deletions(-)
drivers/iommu/intel-pasid.c +23 -10
@@ -499,6 +499,25 @@ int intel_pasid_setup_first_level(struct intel_iommu *iommu,return 0 ; } +/* +* Skip top levels of page tables for iommu which has less agaw +* than default. Unnecessary for PT mode. +*/ +static inline int iommu_skip_agaw (struct dmar_domain *domain, +struct intel_iommu *iommu, +struct dma_pte **pgd) +{ +int agaw; + +for (agaw = domain->agaw; agaw > iommu->agaw; agaw--) { + *pgd =phys_to_virt (dma_pte_addr (*pgd)); +if (!dma_pte_present (*pgd)) +return -EINVAL ; + } + +return agaw; +} +/* * Set up the scalable mode pasid entry for second only translation type. */
@@ -522,17 +541,11 @@ int intel_pasid_setup_second_level(struct intel_iommu *iommu,return -EINVAL ; } -/* -* Skip top levels of page tables for iommu which has less agaw -* than default. Unnecessary for PT mode. -*/ pgd = domain->pgd; -for (agaw = domain->agaw; agaw > iommu->agaw; agaw--) { - pgd =phys_to_virt (dma_pte_addr (pgd)); -if (!dma_pte_present (pgd)) { -dev_err (dev,"Invalid domain page table\n" ); -return -EINVAL ; - } + agaw =iommu_skip_agaw (domain, iommu, &pgd); +if (agaw <0 ) { +dev_err (dev,"Invalid domain page table\n" ); +return -EINVAL ; } pgd_val =virt_to_phys (pgd);
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 1/18 iommu/vt-d: Move domain helper to header
In reply to Lu BaoluFrom: Jacob Pan <[email protected]>
Move domain helper to header to be used by SVA code.
--- drivers/iommu/intel-iommu.c | 6 ------ include/linux/intel-iommu.h | 6 ++++++ 2 files changed, 6 insertions(+), 6 deletions(-)
drivers/iommu/intel-iommu.c +0 -6
@@ -441,12 +441,6 @@ static void init_translation_status(struct intel_iommu *iommu) iommu->flags |=VTD_FLAG_TRANS_PRE_ENABLED ; } -/* Convert generic 'struct iommu_domain to private struct dmar_domain */ -static struct dmar_domain *to_dmar_domain (struct iommu_domain *dom) -{ - return container_of(dom, struct dmar_domain, domain); -} -static int __init intel_iommu_setup (char *str) {if (!str)
include/linux/intel-iommu.h +6 -0
@@ -595,6 +595,12 @@ static inline void __iommu_flush_cache(clflush_cache_range (addr, size); } +/* Convert generic struct iommu_domain to private struct dmar_domain */ +static inline struct dmar_domain *to_dmar_domain (struct iommu_domain *dom) +{ + return container_of(dom, struct dmar_domain, domain); +} +/* * 0: readable * 1: writable
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 3/18 iommu/vt-d: Add nested translation helper function
In reply to Lu BaoluFrom: Jacob Pan <[email protected]>
Nested translation mode is supported in VT-d 3.0 Spec.CH 3.8. With PASID granular translation type set to 0x11b, translation result from the first level(FL) also subject to a second level(SL) page table translation. This mode is used for SVA virtualization, where FL performs guest virtual to guest physical translation and SL performs guest physical to host physical translation.
This patch adds a helper function for setting up nested translation where second level comes from a domain and first level comes from a guest PGD.
--- drivers/iommu/intel-iommu.c | 25 ------ drivers/iommu/intel-pasid.c | 174 +++++++++++++++++++++++++++++++++++- drivers/iommu/intel-pasid.h | 10 +++ include/linux/intel-iommu.h | 20 +++++ include/uapi/linux/iommu.h | 5 ++ 5 files changed, 206 insertions(+), 28 deletions(-)
drivers/iommu/intel-iommu.c +0 -25
@@ -296,31 +296,6 @@ static inline void context_clear_entry(struct context_entry *context)static struct dmar_domain *si_domain;static int hw_pass_through =1 ; -/* si_domain contains mulitple devices */ -#define DOMAIN_FLAG_STATIC_IDENTITY BIT(0) - -/* -* This is a DMA domain allocated through the iommu domain allocation -* interface. But one or more devices belonging to this domain have -* been chosen to use a private domain. We should avoid to use the -* map/unmap/iova_to_phys APIs on it. -*/ -#define DOMAIN_FLAG_LOSE_CHILDREN BIT(1) - -/* -* When VT-d works in the scalable mode, it allows DMA translation to -* happen through either first level or second level page table. This -* bit marks that the DMA translation for the domain goes through the -* first level page table, otherwise, it goes through the second level. -*/ -#define DOMAIN_FLAG_USE_FIRST_LEVEL BIT(2) - -/* -* Domain represents a virtual machine which demands iommu nested -* translation mode support. -*/ -#define DOMAIN_FLAG_NESTING_MODE BIT(3) -#define for_each_domain_iommu (idx, domain) \ for (idx = 0; idx < g_num_of_iommus; idx++) \ if (domain->iommu_refcnt[idx])
drivers/iommu/intel-pasid.c +171 -3
201 lines not on this page · raw message
drivers/iommu/intel-pasid.h +10 -0
@@ -36,6 +36,7 @@* to vmalloc or even module mappings. */ #define PASID_FLAG_SUPERVISOR_MODE BIT(0) +#define PASID_FLAG_NESTED BIT(1)/* * The PASID_FLAG_FL5LP flag Indicates using 5-level paging for first-
@@ -51,6 +52,11 @@ struct pasid_entry {u64 val[8 ]; }; +#define PASID_ENTRY_PGTT_FL_ONLY (1) +#define PASID_ENTRY_PGTT_SL_ONLY (2) +#define PASID_ENTRY_PGTT_NESTED (3) +#define PASID_ENTRY_PGTT_PT (4) +/* The representative of a PASID table */ struct pasid_table {void *table;/* pasid table pointer */
@@ -99,6 +105,10 @@ int intel_pasid_setup_second_level(struct intel_iommu *iommu,int intel_pasid_setup_pass_through (struct intel_iommu *iommu,struct dmar_domain *domain,struct device *dev,int pasid); +int intel_pasid_setup_nested (struct intel_iommu *iommu, +struct device *dev,pgd_t *pgd,int pasid, +struct iommu_gpasid_bind_data_vtd *pasid_data, +struct dmar_domain *domain,int addr_width);void intel_pasid_tear_down_entry (struct intel_iommu *iommu,struct device *dev,int pasid);
include/linux/intel-iommu.h +20 -0
@@ -42,6 +42,9 @@#define DMA_FL_PTE_PRESENT BIT_ULL(0)#define DMA_FL_PTE_XD BIT_ULL(63) +#define ADDR_WIDTH_5LEVEL (57) +#define ADDR_WIDTH_4LEVEL (48) +#define CONTEXT_TT_MULTI_LEVEL 0#define CONTEXT_TT_DEV_IOTLB 1#define CONTEXT_TT_PASS_THROUGH 2
@@ -480,6 +483,23 @@ struct context_entry {u64 hi; }; +/* si_domain contains mulitple devices */ +#define DOMAIN_FLAG_STATIC_IDENTITY BIT(0) + +/* +* When VT-d works in the scalable mode, it allows DMA translation to +* happen through either first level or second level page table. This +* bit marks that the DMA translation for the domain goes through the +* first level page table, otherwise, it goes through the second level. +*/ +#define DOMAIN_FLAG_USE_FIRST_LEVEL BIT(1) + +/* +* Domain represents a virtual machine which demands iommu nested +* translation mode support. +*/ +#define DOMAIN_FLAG_NESTING_MODE BIT(2) +struct dmar_domain {int nid;/* node id */
include/uapi/linux/iommu.h +5 -0
@@ -285,6 +285,11 @@ struct iommu_gpasid_bind_data_vtd {__u32 emt; }; +#define IOMMU_SVA_VTD_GPASID_MTS_MASK (IOMMU_SVA_VTD_GPASID_CD | \ + IOMMU_SVA_VTD_GPASID_EMTE | \ + IOMMU_SVA_VTD_GPASID_PCD | \ + IOMMU_SVA_VTD_GPASID_PWT) +/** * struct iommu_gpasid_bind_data - Information about device and guest PASID binding * @version: Version of this data structure
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 4/18 iommu/vt-d: Add bind guest PASID support
In reply to Lu BaoluFrom: Jacob Pan <[email protected]>
When supporting guest SVA with emulated IOMMU, the guest PASID table is shadowed in VMM. Updates to guest vIOMMU PASID table will result in PASID cache flush which will be passed down to the host as bind guest PASID calls.
For the SL page tables, it will be harvested from device's default domain (request w/o PASID), or aux domain in case of mediated device.
.-------------. .---------------------------.
| vIOMMU | | Guest process CR3, FL only|
| | '---------------------------' .----------------/
| PASID Entry |--- PASID cache flush -
'-------------' |
| | V
| | CR3 in GPA
'-------------'
Guest
------| Shadow |--------------------------|--------
v v v
Host
.-------------. .----------------------.
| pIOMMU | | Bind FL for GVA-GPA |
| | '----------------------'
.----------------/ |
| PASID Entry | V (Nested xlate)
'----------------\.------------------------------.
| | |SL for GPA-HPA, default domain|
| | '------------------------------'
'-------------'Where:
- FL = First level/stage one page tables
- SL = Second level/stage two page tables
--- drivers/iommu/intel-iommu.c | 4 + drivers/iommu/intel-svm.c | 200 ++++++++++++++++++++++++++++++++++++ include/linux/intel-iommu.h | 6 +- include/linux/intel-svm.h | 12 +++ 4 files changed, 221 insertions(+), 1 deletion(-)
drivers/iommu/intel-iommu.c +4 -0
@@ -5780,6 +5780,10 @@ const struct iommu_ops intel_iommu_ops = { .is_attach_deferred = intel_iommu_is_attach_deferred, .def_domain_type = device_def_domain_type, .pgsize_bitmap =INTEL_IOMMU_PGSIZES , +#ifdef CONFIG_INTEL_IOMMU_SVM + .sva_bind_gpasid = intel_svm_bind_gpasid, + .sva_unbind_gpasid = intel_svm_unbind_gpasid, +#endif };static void quirk_iommu_igfx (struct pci_dev *dev)
drivers/iommu/intel-svm.c +200 -0
206 lines not on this page · raw message
include/linux/intel-iommu.h +5 -1
@@ -698,7 +698,9 @@ struct dmar_domain *find_domain(struct device *dev);extern void intel_svm_check (struct intel_iommu *iommu);extern int intel_svm_enable_prq (struct intel_iommu *iommu);extern int intel_svm_finish_prq (struct intel_iommu *iommu); - +int intel_svm_bind_gpasid (struct iommu_domain *domain,struct device *dev, +struct iommu_gpasid_bind_data *data); +int intel_svm_unbind_gpasid (struct device *dev,int pasid);struct svm_dev_ops ;struct intel_svm_dev {
@@ -715,9 +717,11 @@ struct intel_svm_dev {struct intel_svm {struct mmu_notifier notifier;struct mm_struct *mm; +struct intel_iommu *iommu;int flags;int pasid; +int gpasid;/* In case that guest PASID is different from host PASID */ struct list_head devs;struct list_head list; };
include/linux/intel-svm.h +12 -0
@@ -44,6 +44,18 @@ struct svm_dev_ops {* do such IOTLB flushes automatically. */ #define SVM_FLAG_SUPERVISOR_MODE (1<<1) +/* +* The SVM_FLAG_GUEST_MODE flag is used when a PASID bind is for guest +* processes. Compared to the host bind, the primary differences are: +* 1. mm life cycle management +* 2. fault reporting +*/ +#define SVM_FLAG_GUEST_MODE (1<<2) +/* +* The SVM_FLAG_GUEST_PASID flag is used when a guest has its own PASID space, +* which requires guest and host PASID translation at both directions. +*/ +#define SVM_FLAG_GUEST_PASID (1<<3)#ifdef CONFIG_INTEL_IOMMU_SVM
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 5/18 iommu/vt-d: Support flushing more translation cache types
In reply to Lu BaoluFrom: Jacob Pan <[email protected]>
When Shared Virtual Memory is exposed to a guest via vIOMMU, scalable IOTLB invalidation may be passed down from outside IOMMU subsystems. This patch adds invalidation functions that can be used for additional translation cache types.
--- drivers/iommu/dmar.c | 39 +++++++++++++++++++++++++++++++++++++ drivers/iommu/intel-pasid.c | 3 ++- include/linux/intel-iommu.h | 21 ++++++++++++++++---- 3 files changed, 58 insertions(+), 5 deletions(-)
drivers/iommu/dmar.c +39 -0
@@ -1421,6 +1421,45 @@ void qi_flush_piotlb(struct intel_iommu *iommu, u16 did, u32 pasid, u64 addr,qi_submit_sync (&desc, iommu); } +/* PASID-based device IOTLB Invalidate */ +void qi_flush_dev_iotlb_pasid (struct intel_iommu *iommu,u16 sid,u16 pfsid, +u32 pasid,u16 qdep,u64 addr, +unsigned int size_order,u64 granu) +{ +unsigned long mask =1UL << (VTD_PAGE_SHIFT + size_order -1 ); +struct qi_desc desc = {.qw1 =0 , .qw2 =0 , .qw3 =0 }; + + desc.qw0 =QI_DEV_EIOTLB_PASID (pasid) |QI_DEV_EIOTLB_SID (sid) | +QI_DEV_EIOTLB_QDEP (qdep) |QI_DEIOTLB_TYPE | +QI_DEV_IOTLB_PFSID (pfsid); + desc.qw1 =QI_DEV_EIOTLB_GLOB (granu); + +/* +* If S bit is 0, we only flush a single page. If S bit is set, +* The least significant zero bit indicates the invalidation address +* range. VT-d spec 6.5.2.6. +* e.g. address bit 12[0] indicates 8KB, 13[0] indicates 16KB. +* size order = 0 is PAGE_SIZE 4KB +* Max Invs Pending (MIP) is set to 0 for now until we have DIT in +* ECAP. +*/ + desc.qw1 |= addr & ~mask; +if (size_order) + desc.qw1 |=QI_DEV_EIOTLB_SIZE ; + +qi_submit_sync (&desc, iommu); +} + +void qi_flush_pasid_cache (struct intel_iommu *iommu,u16 did, +u64 granu,int pasid) +{ +struct qi_desc desc = {.qw1 =0 , .qw2 =0 , .qw3 =0 }; + + desc.qw0 =QI_PC_PASID (pasid) |QI_PC_DID (did) | +QI_PC_GRAN (granu) |QI_PC_TYPE ; +qi_submit_sync (&desc, iommu); +} +/* * Disable Queued Invalidation interface. */
drivers/iommu/intel-pasid.c +2 -1
@@ -375,7 +375,8 @@ pasid_cache_invalidation_with_pasid(struct intel_iommu *iommu, {struct qi_desc desc; - desc.qw0 =QI_PC_DID (did) |QI_PC_PASID_SEL |QI_PC_PASID (pasid); + desc.qw0 =QI_PC_DID (did) |QI_PC_GRAN (QI_PC_PASID_SEL ) | +QI_PC_PASID (pasid) |QI_PC_TYPE ; desc.qw1 =0 ; desc.qw2 =0 ; desc.qw3 =0 ;
include/linux/intel-iommu.h +17 -4
@@ -334,7 +334,7 @@ enum {#define QI_IOTLB_GRAN (gran) (((u64)gran) >> (DMA_TLB_FLUSH_GRANU_OFFSET-4))#define QI_IOTLB_ADDR (addr) (((u64)addr) & VTD_PAGE_MASK)#define QI_IOTLB_IH (ih) (((u64)ih) << 6) -#define QI_IOTLB_AM (am) (((u8)am)) +#define QI_IOTLB_AM (am) (((u8)am) & 0x3f)#define QI_CC_FM (fm) (((u64)fm) << 48)#define QI_CC_SID (sid) (((u64)sid) << 32)
@@ -353,16 +353,21 @@ enum {#define QI_PC_DID (did) (((u64)did) << 16)#define QI_PC_GRAN (gran) (((u64)gran) << 4) -#define QI_PC_ALL_PASIDS (QI_PC_TYPE | QI_PC_GRAN(0)) -#define QI_PC_PASID_SEL (QI_PC_TYPE | QI_PC_GRAN(1)) +/* PASID cache invalidation granu */ +#define QI_PC_ALL_PASIDS 0 +#define QI_PC_PASID_SEL 1#define QI_EIOTLB_ADDR (addr) ((u64)(addr) & VTD_PAGE_MASK)#define QI_EIOTLB_IH (ih) (((u64)ih) << 6) -#define QI_EIOTLB_AM (am) (((u64)am)) +#define QI_EIOTLB_AM (am) (((u64)am) & 0x3f)#define QI_EIOTLB_PASID (pasid) (((u64)pasid) << 32)#define QI_EIOTLB_DID (did) (((u64)did) << 16)#define QI_EIOTLB_GRAN (gran) (((u64)gran) << 4) +/* QI Dev-IOTLB inv granu */ +#define QI_DEV_IOTLB_GRAN_ALL 1 +#define QI_DEV_IOTLB_GRAN_PASID_SEL 0 +#define QI_DEV_EIOTLB_ADDR (a) ((u64)(a) & VTD_PAGE_MASK)#define QI_DEV_EIOTLB_SIZE (((u64)1) << 11)#define QI_DEV_EIOTLB_GLOB (g) ((u64)g)
@@ -679,8 +684,16 @@ extern void qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr, unsigned int size_order, u64 type);extern void qi_flush_dev_iotlb (struct intel_iommu *iommu,u16 sid,u16 pfsid,u16 qdep,u64 addr,unsigned mask); +void qi_flush_piotlb (struct intel_iommu *iommu,u16 did,u32 pasid,u64 addr,unsigned long npages,bool ih); + +void qi_flush_dev_iotlb_pasid (struct intel_iommu *iommu,u16 sid,u16 pfsid, +u32 pasid,u16 qdep,u64 addr, +unsigned int size_order,u64 granu); +void qi_flush_pasid_cache (struct intel_iommu *iommu,u16 did,u64 granu, +int pasid); +extern int qi_submit_sync (struct qi_desc *desc,struct intel_iommu *iommu);extern int dmar_ir_support (void );
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 7/18 iommu/vt-d: Enlightened PASID allocation
In reply to Lu BaoluEnabling IOMMU in a guest requires communication with the host driver for certain aspects. Use of PASID ID to enable Shared Virtual Addressing (SVA) requires managing PASID's in the host. VT-d 3.0 spec provides a Virtual Command Register (VCMD) to facilitate this. Writes to this register in the guest are trapped by vIOMMU which proxies the call to the host driver.
This virtual command interface consists of a capability register, a virtual command register, and a virtual response register. Refer to section 10.4.42, 10.4.43, 10.4.44 for more information.
This patch adds the enlightened PASID allocation/free interfaces via the virtual command interface.
--- drivers/iommu/intel-pasid.c | 57 +++++++++++++++++++++++++++++++++++++ drivers/iommu/intel-pasid.h | 13 ++++++++- include/linux/intel-iommu.h | 1 + 3 files changed, 70 insertions(+), 1 deletion(-)
drivers/iommu/intel-pasid.c +57 -0
@@ -27,6 +27,63 @@static DEFINE_SPINLOCK (pasid_lock);u32 intel_pasid_max_id =PASID_MAX ; +int vcmd_alloc_pasid (struct intel_iommu *iommu,unsigned int *pasid) +{ +unsigned long flags; +u8 status_code; +int ret =0 ; +u64 res; + +raw_spin_lock_irqsave (&iommu->register_lock, flags); +dmar_writeq (iommu->reg +DMAR_VCMD_REG ,VCMD_CMD_ALLOC ); +IOMMU_WAIT_OP (iommu,DMAR_VCRSP_REG , dmar_readq, + !(res &VCMD_VRSP_IP ), res); +raw_spin_unlock_irqrestore (&iommu->register_lock, flags); + + status_code =VCMD_VRSP_SC (res); +switch (status_code) { +case VCMD_VRSP_SC_SUCCESS : + *pasid =VCMD_VRSP_RESULT_PASID (res); +break ; +case VCMD_VRSP_SC_NO_PASID_AVAIL : +pr_info ("IOMMU: %s: No PASID available\n" , iommu->name); + ret = -ENOSPC ; +break ; +default : + ret = -ENODEV ; +pr_warn ("IOMMU: %s: Unexpected error code %d\n" , + iommu->name, status_code); + } + +return ret; +} + +void vcmd_free_pasid (struct intel_iommu *iommu,unsigned int pasid) +{ +unsigned long flags; +u8 status_code; +u64 res; + +raw_spin_lock_irqsave (&iommu->register_lock, flags); +dmar_writeq (iommu->reg +DMAR_VCMD_REG , +VCMD_CMD_OPERAND (pasid) |VCMD_CMD_FREE ); +IOMMU_WAIT_OP (iommu,DMAR_VCRSP_REG , dmar_readq, + !(res &VCMD_VRSP_IP ), res); +raw_spin_unlock_irqrestore (&iommu->register_lock, flags); + + status_code =VCMD_VRSP_SC (res); +switch (status_code) { +case VCMD_VRSP_SC_SUCCESS : +break ; +case VCMD_VRSP_SC_INVALID_PASID : +pr_info ("IOMMU: %s: Invalid PASID\n" , iommu->name); +break ; +default : +pr_warn ("IOMMU: %s: Unexpected error code %d\n" , + iommu->name, status_code); + } +} +/* * Per device pasid table management: */
drivers/iommu/intel-pasid.h +12 -1
@@ -23,6 +23,16 @@#define is_pasid_enabled (entry) (((entry)->lo >> 3) & 0x1)#define get_pasid_dir_size (entry) (1 << ((((entry)->lo >> 9) & 0x7) + 7)) +/* Virtual command interface for enlightened pasid management. */ +#define VCMD_CMD_ALLOC 0x1 +#define VCMD_CMD_FREE 0x2 +#define VCMD_VRSP_IP 0x1 +#define VCMD_VRSP_SC (e) (((e) >> 1) & 0x3) +#define VCMD_VRSP_SC_SUCCESS 0 +#define VCMD_VRSP_SC_NO_PASID_AVAIL 1 +#define VCMD_VRSP_SC_INVALID_PASID 1 +#define VCMD_VRSP_RESULT_PASID (e) (((e) >> 8) & 0xfffff) +#define VCMD_CMD_OPERAND (e) ((e) << 8)/* * Domain ID reserved for pasid entries programmed for first-level * only and pass-through transfer modes.
@@ -111,5 +121,6 @@ int intel_pasid_setup_nested(struct intel_iommu *iommu, struct dmar_domain *domain, int addr_width);void intel_pasid_tear_down_entry (struct intel_iommu *iommu,struct device *dev,int pasid); - +int vcmd_alloc_pasid (struct intel_iommu *iommu,unsigned int *pasid); +void vcmd_free_pasid (struct intel_iommu *iommu,unsigned int pasid);#endif /* __INTEL_PASID_H */
include/linux/intel-iommu.h +1 -0
@@ -169,6 +169,7 @@#define ecap_smpwc (e) (((e) >> 48) & 0x1)#define ecap_flts (e) (((e) >> 47) & 0x1)#define ecap_slts (e) (((e) >> 46) & 0x1) +#define ecap_vcs (e) (((e) >> 44) & 0x1)#define ecap_smts (e) (((e) >> 43) & 0x1)#define ecap_dit (e) ((e >> 41) & 0x1)#define ecap_pasid (e) ((e >> 40) & 0x1)
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 6/18 iommu/vt-d: Add svm/sva invalidate function
In reply to Lu BaoluFrom: Jacob Pan <[email protected]>
When Shared Virtual Address (SVA) is enabled for a guest OS via vIOMMU, we need to provide invalidation support at IOMMU API and driver level. This patch adds Intel VT-d specific function to implement iommu passdown invalidate API for shared virtual address.
The use case is for supporting caching structure invalidation of assigned SVM capable devices. Emulated IOMMU exposes queue invalidation capability and passes down all descriptors from the guest to the physical IOMMU.
The assumption is that guest to host device ID mapping should be resolved prior to calling IOMMU driver. Based on the device handle, host IOMMU driver can replace certain fields before submit to the invalidation queue.
--- drivers/iommu/intel-iommu.c | 171 ++++++++++++++++++++++++++++++++++++ 1 file changed, 171 insertions(+)
drivers/iommu/intel-iommu.c +171 -0
@@ -5242,6 +5242,176 @@ static void intel_iommu_aux_detach_device(struct iommu_domain *domain,aux_domain_remove_dev (to_dmar_domain (domain), dev); } +/* +* 2D array for converting and sanitizing IOMMU generic TLB granularity to +* VT-d granularity. Invalidation is typically included in the unmap operation +* as a result of DMA or VFIO unmap. However, for assigned devices guest +* owns the first level page tables. Invalidations of translation caches in the +* guest are trapped and passed down to the host. +* +* vIOMMU in the guest will only expose first level page tables, therefore +* we do not support IOTLB granularity for request without PASID (second level). +* +* For example, to find the VT-d granularity encoding for IOTLB +* type and page selective granularity within PASID: +* X: indexed by iommu cache type +* Y: indexed by enum iommu_inv_granularity +* [IOMMU_CACHE_INV_TYPE_IOTLB][IOMMU_INV_GRANU_ADDR] +*/ + +const static int +inv_type_granu_table[IOMMU_CACHE_INV_TYPE_NR ][IOMMU_INV_GRANU_NR ] = { +/* +* PASID based IOTLB invalidation: PASID selective (per PASID), +* page selective (address granularity) +*/ + {-EINVAL ,QI_GRAN_NONG_PASID ,QI_GRAN_PSI_PASID }, +/* PASID based dev TLBs */ + {-EINVAL , -EINVAL ,QI_DEV_IOTLB_GRAN_PASID_SEL }, +/* PASID cache */ + {-EINVAL , -EINVAL , -EINVAL } +}; + +static inline int to_vtd_granularity (int type,int granu) +{ +return inv_type_granu_table[type][granu]; +} + +static inline u64 to_vtd_size (u64 granu_size,u64 nr_granules) +{ +u64 nr_pages = (granu_size * nr_granules) >>VTD_PAGE_SHIFT ; + +/* VT-d size is encoded as 2^size of 4K pages, 0 for 4k, 9 for 2MB, etc. +* IOMMU cache invalidate API passes granu_size in bytes, and number of +* granu size in contiguous memory. +*/ +return order_base_2 (nr_pages); +} + +#ifdef CONFIG_INTEL_IOMMU_SVM +static int +intel_iommu_sva_invalidate (struct iommu_domain *domain,struct device *dev, +struct iommu_cache_invalidate_info *inv_info) +{ +struct dmar_domain *dmar_domain =to_dmar_domain (domain); +struct device_domain_info *info; +struct intel_iommu *iommu; +unsigned long flags; +int cache_type; +u8 bus, devfn; +u16 did, sid; +int ret =0 ; +u64 size =0 ; + +if (!inv_info || !dmar_domain || + inv_info->version !=IOMMU_CACHE_INVALIDATE_INFO_VERSION_1 ) +return -EINVAL ; + +if (!dev || !dev_is_pci (dev)) +return -ENODEV ; + + iommu =device_to_iommu (dev, &bus, &devfn); +if (!iommu) +return -ENODEV ; + +if (!(dmar_domain->flags &DOMAIN_FLAG_NESTING_MODE )) +return -EINVAL ; + +spin_lock_irqsave (&device_domain_lock, flags); +spin_lock (&iommu->lock); + info = dev->archdata.iommu; +if (!info) { + ret = -EINVAL ; + goto out_unlock; + } + did = dmar_domain->iommu_did[iommu->seq_id]; + sid =PCI_DEVID (bus, devfn); + +/* Size is only valid in address selective invalidation */ +if (inv_info->granularity !=IOMMU_INV_GRANU_PASID ) + size =to_vtd_size (inv_info->addr_info.granule_size, + inv_info->addr_info.nb_granules); + +for_each_set_bit (cache_type, + (unsigned long *)&inv_info->cache, +IOMMU_CACHE_INV_TYPE_NR ) { +int granu =0 ; +u64 pasid =0 ; + + granu =to_vtd_granularity (cache_type, inv_info->granularity); +if (granu == -EINVAL ) { +pr_err_ratelimited ("Invalid cache type and granu combination %d/%d\n" , + cache_type, inv_info->granularity); +break ; + } + +/* +* PASID is stored in different locations based on the +* granularity. +*/ +if (inv_info->granularity ==IOMMU_INV_GRANU_PASID && + (inv_info->pasid_info.flags &IOMMU_INV_PASID_FLAGS_PASID )) + pasid = inv_info->pasid_info.pasid; +else if (inv_info->granularity ==IOMMU_INV_GRANU_ADDR && + (inv_info->addr_info.flags &IOMMU_INV_ADDR_FLAGS_PASID )) + pasid = inv_info->addr_info.pasid; + +switch (BIT (cache_type)) { +case IOMMU_CACHE_INV_TYPE_IOTLB : +if (inv_info->granularity ==IOMMU_INV_GRANU_ADDR && + size && + (inv_info->addr_info.addr & ((BIT (VTD_PAGE_SHIFT + size)) -1 ))) { +pr_err_ratelimited ("Address out of range, 0x%llx, size order %llu\n" , + inv_info->addr_info.addr, size); + ret = -ERANGE ; + goto out_unlock; + } + +/* +* If granu is PASID-selective, address is ignored. +* We use npages = -1 to indicate that. +*/ +qi_flush_piotlb (iommu, did, pasid, +mm_to_dma_pfn (inv_info->addr_info.addr), + (granu ==QI_GRAN_NONG_PASID ) ?-1 :1 << size, + inv_info->addr_info.flags &IOMMU_INV_ADDR_FLAGS_LEAF ); + +/* +* Always flush device IOTLB if ATS is enabled. vIOMMU +* in the guest may assume IOTLB flush is inclusive, +* which is more efficient. +*/ +if (info->ats_enabled) +qi_flush_dev_iotlb_pasid (iommu, sid, + info->pfsid, pasid, + info->ats_qdep, + inv_info->addr_info.addr, + size, granu); +break ; +case IOMMU_CACHE_INV_TYPE_DEV_IOTLB : +if (info->ats_enabled) +qi_flush_dev_iotlb_pasid (iommu, sid, + info->pfsid, pasid, + info->ats_qdep, + inv_info->addr_info.addr, + size, granu); +else +pr_warn_ratelimited ("Passdown device IOTLB flush w/o ATS!\n" ); +break ; +default : +dev_err_ratelimited (dev,"Unsupported IOMMU invalidation type %d\n" , + cache_type); + ret = -EINVAL ; + } + } +out_unlock: +spin_unlock (&iommu->lock); +spin_unlock_irqrestore (&device_domain_lock, flags); + +return ret; +} +#endif +static int intel_iommu_map (struct iommu_domain *domain,unsigned long iova,phys_addr_t hpa,size_t size,int iommu_prot,gfp_t gfp)
@@ -5781,6 +5951,7 @@ const struct iommu_ops intel_iommu_ops = { .def_domain_type = device_def_domain_type, .pgsize_bitmap =INTEL_IOMMU_PGSIZES , #ifdef CONFIG_INTEL_IOMMU_SVM + .cache_invalidate = intel_iommu_sva_invalidate, .sva_bind_gpasid = intel_svm_bind_gpasid, .sva_unbind_gpasid = intel_svm_unbind_gpasid, #endif
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 8/18 iommu/vt-d: Add custom allocator for IOASID
In reply to Lu BaoluFrom: Jacob Pan <[email protected]>
When VT-d driver runs in the guest, PASID allocation must be performed via virtual command interface. This patch registers a custom IOASID allocator which takes precedence over the default XArray based allocator. The resulting IOASID allocation will always come from the host. This ensures that PASID namespace is system- wide.
Virtual command registers are used in the guest only, to prevent vmexit cost, we cache the capability and store it during initialization.
--- drivers/iommu/dmar.c | 1 + drivers/iommu/intel-iommu.c | 85 +++++++++++++++++++++++++++++++++++++ include/linux/intel-iommu.h | 7 +++ 3 files changed, 93 insertions(+)
drivers/iommu/dmar.c +1 -0
@@ -963,6 +963,7 @@ static int map_iommu(struct intel_iommu *iommu, u64 phys_addr)warn_invalid_dmar (phys_addr," returns all ones" ); goto unmap; } + iommu->vccap =dmar_readq (iommu->reg +DMAR_VCCAP_REG );/* the registers might be more than one page */ map_size = max_t(int, ecap_max_iotlb_offset(iommu->ecap),
drivers/iommu/intel-iommu.c +85 -0
103 lines not on this page · raw message
include/linux/intel-iommu.h +7 -0
@@ -19,6 +19,7 @@#include <linux/iommu.h> #include <linux/io-64-nonatomic-lo-hi.h> #include <linux/dmar.h> +#include <linux/ioasid.h> #include <asm/cacheflush.h> #include <asm/iommu.h>
@@ -195,6 +196,9 @@#define ecap_max_handle_mask (e) ((e >> 20) & 0xf)#define ecap_sc_support (e) ((e >> 7) & 0x1)/* Snooping Control */ +/* Virtual command interface capability */ +#define vccap_pasid (v) (((v) & DMA_VCS_PAS))/* PASID allocation */ +/* IOTLB_REG */ #define DMA_TLB_FLUSH_GRANU_OFFSET 60#define DMA_TLB_GLOBAL_FLUSH (((u64)1) << 60)
@@ -288,6 +292,7 @@/* PRS_REG */ #define DMA_PRS_PPR ((u32)1) +#define DMA_VCS_PAS ((u64)1)#define IOMMU_WAIT_OP (iommu, offset, op, cond, sts) \ do { \
@@ -555,6 +560,7 @@ struct intel_iommu {u64 reg_size;/* size of hw register set */ u64 cap;u64 ecap; +u64 vccap;u32 gcmd;/* Holds TE, EAFL. Don't need SRTP, SFL, WBF */ raw_spinlock_t register_lock;/* protect register handling */ int seq_id;/* sequence id of the iommu */
@@ -575,6 +581,7 @@ struct intel_iommu {#ifdef CONFIG_INTEL_IOMMU_SVM struct page_req_dsc *prq;unsigned char prq_name[16 ];/* Name for PRQ interrupt */ +struct ioasid_allocator_ops pasid_allocator;/* Custom allocator for PASIDs */ #endif struct q_inval *qi;/* Queued invalidation info */ u32 *iommu_state;/* Store iommu states between suspend and resume.*/
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 9/18 iommu/vt-d: Add get_domain_info() helper
In reply to Lu BaoluAdd a get_domain_info() helper to retrieve the valid per-device iommu private data.
--- drivers/iommu/intel-iommu.c | 40 +++++++++++++++++++++++++------------ drivers/iommu/intel-pasid.c | 12 +++++------ drivers/iommu/intel-svm.c | 2 +- include/linux/intel-iommu.h | 1 + 4 files changed, 35 insertions(+), 20 deletions(-)
drivers/iommu/intel-iommu.c +27 -13
@@ -365,6 +365,21 @@ EXPORT_SYMBOL_GPL(intel_iommu_gfx_mapped);#define DUMMY_DEVICE_DOMAIN_INFO ((struct device_domain_info *)(-1))#define DEFER_DEVICE_DOMAIN_INFO ((struct device_domain_info *)(-2)) +struct device_domain_info *get_domain_info (struct device *dev) +{ +struct device_domain_info *info; + +if (!dev) +return NULL ; + + info = dev->archdata.iommu; +if (unlikely (info ==DUMMY_DEVICE_DOMAIN_INFO || + info ==DEFER_DEVICE_DOMAIN_INFO )) +return NULL ; + +return info; +} +DEFINE_SPINLOCK (device_domain_lock);static LIST_HEAD (device_domain_list);
@@ -2429,7 +2444,7 @@ struct dmar_domain *find_domain(struct device *dev) dev = &pci_real_dma_dev (to_pci_dev (dev))->dev;/* No lock here, assumes no domain exit in normal case */ - info = dev->archdata.iommu; + info =get_domain_info (dev);if (likely (info))return info->domain;
@@ -5012,9 +5027,8 @@ static void dmar_remove_one_dev_info(struct device *dev)unsigned long flags;spin_lock_irqsave (&device_domain_lock, flags); - info = dev->archdata.iommu; -if (info && info !=DEFER_DEVICE_DOMAIN_INFO - && info !=DUMMY_DEVICE_DOMAIN_INFO ) + info =get_domain_info (dev); +if (info)__dmar_remove_one_dev_info (info);spin_unlock_irqrestore (&device_domain_lock, flags); }
@@ -5104,7 +5118,7 @@ static void intel_iommu_domain_free(struct iommu_domain *domain)static inline bool is_aux_domain (struct device *dev,struct iommu_domain *domain) { -struct device_domain_info *info = dev->archdata.iommu; +struct device_domain_info *info =get_domain_info (dev);return info && info->auxd_enabled && domain->type ==IOMMU_DOMAIN_UNMANAGED ;
@@ -5113,7 +5127,7 @@ is_aux_domain(struct device *dev, struct iommu_domain *domain)static void auxiliary_link_device (struct dmar_domain *domain,struct device *dev) { -struct device_domain_info *info = dev->archdata.iommu; +struct device_domain_info *info =get_domain_info (dev);assert_spin_locked (&device_domain_lock);if (WARN_ON (!info))
@@ -5126,7 +5140,7 @@ static void auxiliary_link_device(struct dmar_domain *domain,static void auxiliary_unlink_device (struct dmar_domain *domain,struct device *dev) { -struct device_domain_info *info = dev->archdata.iommu; +struct device_domain_info *info =get_domain_info (dev);assert_spin_locked (&device_domain_lock);if (WARN_ON (!info))
@@ -5214,7 +5228,7 @@ static void aux_domain_remove_dev(struct dmar_domain *domain,return ;spin_lock_irqsave (&device_domain_lock, flags); - info = dev->archdata.iommu; + info =get_domain_info (dev); iommu = info->iommu;auxiliary_unlink_device (domain, dev);
@@ -5404,7 +5418,7 @@ intel_iommu_sva_invalidate(struct iommu_domain *domain, struct device *dev,spin_lock_irqsave (&device_domain_lock, flags);spin_lock (&iommu->lock); - info = dev->archdata.iommu; + info =get_domain_info (dev);if (!info) { ret = -EINVAL ; goto out_unlock;
@@ -5768,7 +5782,7 @@ int intel_iommu_enable_pasid(struct intel_iommu *iommu, struct device *dev)spin_lock (&iommu->lock); ret = -EINVAL ; - info = dev->archdata.iommu; + info =get_domain_info (dev);if (!info || !info->pasid_supported) goto out;
@@ -5864,7 +5878,7 @@ static int intel_iommu_enable_auxd(struct device *dev)return -ENODEV ;spin_lock_irqsave (&device_domain_lock, flags); - info = dev->archdata.iommu; + info =get_domain_info (dev); info->auxd_enabled =1 ;spin_unlock_irqrestore (&device_domain_lock, flags);
@@ -5877,7 +5891,7 @@ static int intel_iommu_disable_auxd(struct device *dev)unsigned long flags;spin_lock_irqsave (&device_domain_lock, flags); - info = dev->archdata.iommu; + info =get_domain_info (dev);if (!WARN_ON (!info)) info->auxd_enabled =0 ;spin_unlock_irqrestore (&device_domain_lock, flags);
@@ -5954,7 +5968,7 @@ intel_iommu_dev_disable_feat(struct device *dev, enum iommu_dev_features feat)static bool intel_iommu_dev_feat_enabled (struct device *dev,enum iommu_dev_features feat) { -struct device_domain_info *info = dev->archdata.iommu; +struct device_domain_info *info =get_domain_info (dev);if (feat ==IOMMU_DEV_FEAT_AUX )return scalable_mode_support () && info && info->auxd_enabled;
drivers/iommu/intel-pasid.c +6 -6
@@ -151,7 +151,7 @@ int intel_pasid_alloc_table(struct device *dev)int size;might_sleep (); - info = dev->archdata.iommu; + info =get_domain_info (dev);if (WARN_ON (!info || !dev_is_pci (dev) || info->pasid_table))return -EINVAL ;
@@ -198,7 +198,7 @@ void intel_pasid_free_table(struct device *dev)struct pasid_entry *table;int i, max_pde; - info = dev->archdata.iommu; + info =get_domain_info (dev);if (!info || !dev_is_pci (dev) || !info->pasid_table)return ;
@@ -224,7 +224,7 @@ struct pasid_table *intel_pasid_get_table(struct device *dev) {struct device_domain_info *info; - info = dev->archdata.iommu; + info =get_domain_info (dev);if (!info)return NULL ;
@@ -235,7 +235,7 @@ int intel_pasid_get_dev_max_id(struct device *dev) {struct device_domain_info *info; - info = dev->archdata.iommu; + info =get_domain_info (dev);if (!info || !info->pasid_table)return 0 ;
@@ -256,7 +256,7 @@ struct pasid_entry *intel_pasid_get_entry(struct device *dev, int pasid)return NULL ; dir = pasid_table->table; - info = dev->archdata.iommu; + info =get_domain_info (dev); dir_index = pasid >>PASID_PDE_SHIFT ; index = pasid &PASID_PTE_MASK ;
@@ -462,7 +462,7 @@ devtlb_invalidation_with_pasid(struct intel_iommu *iommu,struct device_domain_info *info;u16 sid, qdep, pfsid; - info = dev->archdata.iommu; + info =get_domain_info (dev);if (!info || !info->ats_enabled)return ;
drivers/iommu/intel-svm.c +1 -1
@@ -503,7 +503,7 @@ int intel_svm_bind_mm(struct device *dev, int *pasid, int flags, struct svm_dev_ goto out; } - info = dev->archdata.iommu; + info =get_domain_info (dev);if (!info || !info->pasid_supported) {kfree (sdev); goto out;
include/linux/intel-iommu.h +1 -0
@@ -714,6 +714,7 @@ int for_each_device_domain(int (*fn)(struct device_domain_info *info,void iommu_flush_write_buffer (struct intel_iommu *iommu);int intel_iommu_enable_pasid (struct intel_iommu *iommu,struct device *dev);struct dmar_domain *find_domain (struct device *dev); +struct device_domain_info *get_domain_info (struct device *dev);#ifdef CONFIG_INTEL_IOMMU_SVM extern void intel_svm_check (struct intel_iommu *iommu);
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 10/18 iommu/vt-d: Report SVA feature with generic flag
In reply to Lu BaoluFrom: Jacob Pan <[email protected]>
Query Shared Virtual Address/Memory capability is a generic feature. SVA feature check is the required first step before calling iommu_sva_bind_device().
VT-d checks SVA feature enabling at per IOMMU level during this step, SVA bind device will check and enable PCI ATS, PRS, and PASID capabilities at device level.
This patch reports Intel SVM as SVA feature such that generic code (e.g. Uacce [1]) can use it.
[1] https://lkml.org/lkml/2020/1/15/604
--- drivers/iommu/intel-iommu.c | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+)
drivers/iommu/intel-iommu.c +18 -0
@@ -5944,6 +5944,14 @@ intel_iommu_dev_has_feat(struct device *dev, enum iommu_dev_features feat)return !!siov_find_pci_dvsec (to_pci_dev (dev)); } +if (feat ==IOMMU_DEV_FEAT_SVA ) { +struct device_domain_info *info =get_domain_info (dev); + +return info && (info->iommu->flags &VTD_FLAG_SVM_CAPABLE ) && + info->pasid_supported && info->pri_supported && + info->ats_supported; + } +return false; }
@@ -5953,6 +5961,16 @@ intel_iommu_dev_enable_feat(struct device *dev, enum iommu_dev_features feat)if (feat ==IOMMU_DEV_FEAT_AUX )return intel_iommu_enable_auxd (dev); +if (feat ==IOMMU_DEV_FEAT_SVA ) { +struct device_domain_info *info =get_domain_info (dev); + +if (!info) +return -EINVAL ; + +if (info->iommu->flags &VTD_FLAG_SVM_CAPABLE ) +return 0 ; + } +return -ENODEV ; }
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 11/18 iommu/vt-d: Replace intel SVM APIs with generic SVA APIs
In reply to Lu BaoluFrom: Jacob Pan <[email protected]>
This patch is an initial step to replace Intel SVM code with the following IOMMU SVA ops:intel_svm_bind_mm() => iommu_sva_bind_device() intel_svm_unbind_mm() => iommu_sva_unbind_device() intel_svm_is_pasid_valid() => iommu_sva_get_pasid()
The features below will continue to work but are not included in this patch in that they are handled mostly within the IOMMU subsystem.
- IO page fault
- mmu notifier
Consolidation of the above will come after merging generic IOMMU sva code[1]. There should not be any changes needed for SVA users such as accelerator device drivers during this time.
--- drivers/iommu/intel-iommu.c | 3 + drivers/iommu/intel-svm.c | 124 ++++++++++++++++++++---------------- include/linux/intel-iommu.h | 6 ++ include/linux/intel-svm.h | 86 ------------------------- 4 files changed, 78 insertions(+), 141 deletions(-)
drivers/iommu/intel-iommu.c +3 -0
@@ -6071,6 +6071,9 @@ const struct iommu_ops intel_iommu_ops = { .cache_invalidate = intel_iommu_sva_invalidate, .sva_bind_gpasid = intel_svm_bind_gpasid, .sva_unbind_gpasid = intel_svm_unbind_gpasid, + .sva_bind = intel_svm_bind, + .sva_unbind = intel_svm_unbind, + .sva_get_pasid = intel_svm_get_pasid,#endif };
drivers/iommu/intel-svm.c +69 -55
180 lines not on this page · raw message
include/linux/intel-iommu.h +6 -0
@@ -723,6 +723,10 @@ extern int intel_svm_finish_prq(struct intel_iommu *iommu);int intel_svm_bind_gpasid (struct iommu_domain *domain,struct device *dev,struct iommu_gpasid_bind_data *data);int intel_svm_unbind_gpasid (struct device *dev,int pasid); +struct iommu_sva *intel_svm_bind (struct device *dev,struct mm_struct *mm, +void *drvdata); +void intel_svm_unbind (struct iommu_sva *handle); +int intel_svm_get_pasid (struct iommu_sva *handle);struct svm_dev_ops ;struct intel_svm_dev {
@@ -730,6 +734,8 @@ struct intel_svm_dev {struct rcu_head rcu;struct device *dev;struct svm_dev_ops *ops; +struct iommu_sva sva; +int pasid;int users;u16 did; u16 dev_iotlb:1;
include/linux/intel-svm.h +0 -86
96 lines not on this page · raw message
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 12/18 iommu/vt-d: Multiple descriptors per qi_submit_sync()
In reply to Lu BaoluCurrent qi_submit_sync() only supports single invalidation descriptor per submission and appends wait descriptor after each submission to poll the hardware completion. This extends the qi_submit_sync() helper to support multiple descriptors, and add an option so that the caller could specify the Page-request Drain (PD) bit in the wait descriptor.
--- drivers/iommu/dmar.c | 63 +++++++++++++++++------------ drivers/iommu/intel-pasid.c | 4 +- drivers/iommu/intel-svm.c | 6 +-- drivers/iommu/intel_irq_remapping.c | 2 +- include/linux/intel-iommu.h | 9 ++++- 5 files changed, 52 insertions(+), 32 deletions(-)
drivers/iommu/dmar.c +38 -25
@@ -1157,12 +1157,11 @@ static inline void reclaim_free_desc(struct q_inval *qi) } } -static int qi_check_fault (struct intel_iommu *iommu,int index) +static int qi_check_fault (struct intel_iommu *iommu,int index,int wait_index) {u32 fault;int head, tail;struct q_inval *qi = iommu->qi; -int wait_index = (index +1 ) %QI_LENGTH ;int shift =qi_shift (iommu);if (qi->desc_status[wait_index] ==QI_ABORT )
@@ -1225,17 +1224,21 @@ static int qi_check_fault(struct intel_iommu *iommu, int index) }/* -* Submit the queued invalidation descriptor to the remapping -* hardware unit and wait for its completion. +* Function to submit invalidation descriptors of all types to the queued +* invalidation interface(QI). Multiple descriptors can be submitted at a +* time, a wait descriptor will be appended to each submission to ensure +* hardware has completed the invalidation before return. Wait descriptors +* can be part of the submission but it will not be polled for completion. */ -int qi_submit_sync (struct qi_desc *desc,struct intel_iommu *iommu) +int qi_submit_sync (struct intel_iommu *iommu,struct qi_desc *desc, +unsigned int count,unsigned long options) { -int rc;struct q_inval *qi = iommu->qi; -int offset, shift, length;struct qi_desc wait_desc;int wait_index, index;unsigned long flags; +int offset, shift; +int rc, i;if (!qi)return 0 ;
@@ -1244,32 +1247,41 @@ int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu) rc =0 ;raw_spin_lock_irqsave (&qi->q_lock, flags); -while (qi->free_cnt <3 ) { +/* +* Check if we have enough empty slots in the queue to submit, +* the calculation is based on: +* # of desc + 1 wait desc + 1 space between head and tail +*/ +while (qi->free_cnt < count +2 ) {raw_spin_unlock_irqrestore (&qi->q_lock, flags);cpu_relax ();raw_spin_lock_irqsave (&qi->q_lock, flags); } index = qi->free_head; - wait_index = (index +1 ) %QI_LENGTH ; + wait_index = (index + count) %QI_LENGTH ; shift =qi_shift (iommu); - length =1 << shift; - qi->desc_status[index] = qi->desc_status[wait_index] =QI_IN_USE ; +for (i =0 ; i < count; i++) { + offset = ((index + i) %QI_LENGTH ) << shift; +memcpy (qi->desc + offset, &desc[i],1 << shift); + qi->desc_status[(index + i) %QI_LENGTH ] =QI_IN_USE ; + } + qi->desc_status[wait_index] =QI_IN_USE ; - offset = index << shift; -memcpy (qi->desc + offset, desc, length); wait_desc.qw0 =QI_IWD_STATUS_DATA (QI_DONE ) |QI_IWD_STATUS_WRITE |QI_IWD_TYPE ; +if (options &QI_OPT_WAIT_DRAIN ) + wait_desc.qw0 |=QI_IWD_PRQ_DRAIN ; wait_desc.qw1 =virt_to_phys (&qi->desc_status[wait_index]); wait_desc.qw2 =0 ; wait_desc.qw3 =0 ; offset = wait_index << shift; -memcpy (qi->desc + offset, &wait_desc, length); +memcpy (qi->desc + offset, &wait_desc,1 << shift); - qi->free_head = (qi->free_head +2 ) %QI_LENGTH ; - qi->free_cnt -=2 ; + qi->free_head = (qi->free_head + count +1 ) %QI_LENGTH ; + qi->free_cnt -= count +1 ;/* * update the HW tail register indicating the presence of
@@ -1285,7 +1297,7 @@ int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu)* a deadlock where the interrupt context can wait indefinitely * for free slots in the queue. */ - rc =qi_check_fault (iommu, index); + rc =qi_check_fault (iommu, index, wait_index);if (rc)break ;
@@ -1294,7 +1306,8 @@ int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu)raw_spin_lock (&qi->q_lock); } - qi->desc_status[index] =QI_DONE ; +for (i =0 ; i < count; i++) + qi->desc_status[(index + i) %QI_LENGTH ] =QI_DONE ;reclaim_free_desc (qi);raw_spin_unlock_irqrestore (&qi->q_lock, flags);
@@ -1318,7 +1331,7 @@ void qi_global_iec(struct intel_iommu *iommu) desc.qw3 =0 ;/* should never fail */ -qi_submit_sync (&desc, iommu); +qi_submit_sync (iommu, &desc,1 ,0 ); } void qi_flush_context(struct intel_iommu *iommu, u16 did, u16 sid, u8 fm,
@@ -1332,7 +1345,7 @@ void qi_flush_context(struct intel_iommu *iommu, u16 did, u16 sid, u8 fm, desc.qw2 =0 ; desc.qw3 =0 ; -qi_submit_sync (&desc, iommu); +qi_submit_sync (iommu, &desc,1 ,0 ); } void qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr,
@@ -1356,7 +1369,7 @@ void qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr, desc.qw2 =0 ; desc.qw3 =0 ; -qi_submit_sync (&desc, iommu); +qi_submit_sync (iommu, &desc,1 ,0 ); } void qi_flush_dev_iotlb(struct intel_iommu *iommu, u16 sid, u16 pfsid,
@@ -1378,7 +1391,7 @@ void qi_flush_dev_iotlb(struct intel_iommu *iommu, u16 sid, u16 pfsid, desc.qw2 =0 ; desc.qw3 =0 ; -qi_submit_sync (&desc, iommu); +qi_submit_sync (iommu, &desc,1 ,0 ); }/* PASID-based IOTLB invalidation */
@@ -1419,7 +1432,7 @@ void qi_flush_piotlb(struct intel_iommu *iommu, u16 did, u32 pasid, u64 addr,QI_EIOTLB_AM (mask); } -qi_submit_sync (&desc, iommu); +qi_submit_sync (iommu, &desc,1 ,0 ); }/* PASID-based device IOTLB Invalidate */
@@ -1448,7 +1461,7 @@ void qi_flush_dev_iotlb_pasid(struct intel_iommu *iommu, u16 sid, u16 pfsid,if (size_order) desc.qw1 |=QI_DEV_EIOTLB_SIZE ; -qi_submit_sync (&desc, iommu); +qi_submit_sync (iommu, &desc,1 ,0 ); } void qi_flush_pasid_cache(struct intel_iommu *iommu, u16 did,
@@ -1458,7 +1471,7 @@ void qi_flush_pasid_cache(struct intel_iommu *iommu, u16 did, desc.qw0 =QI_PC_PASID (pasid) |QI_PC_DID (did) |QI_PC_GRAN (granu) |QI_PC_TYPE ; -qi_submit_sync (&desc, iommu); +qi_submit_sync (iommu, &desc,1 ,0 ); }/*
drivers/iommu/intel-pasid.c +2 -2
@@ -438,7 +438,7 @@ pasid_cache_invalidation_with_pasid(struct intel_iommu *iommu, desc.qw2 =0 ; desc.qw3 =0 ; -qi_submit_sync (&desc, iommu); +qi_submit_sync (iommu, &desc,1 ,0 ); } static void
@@ -452,7 +452,7 @@ iotlb_invalidation_with_pasid(struct intel_iommu *iommu, u16 did, u32 pasid) desc.qw2 =0 ; desc.qw3 =0 ; -qi_submit_sync (&desc, iommu); +qi_submit_sync (iommu, &desc,1 ,0 ); } static void
drivers/iommu/intel-svm.c +3 -3
@@ -138,7 +138,7 @@ static void intel_flush_svm_range_dev (struct intel_svm *svm, struct intel_svm_d } desc.qw2 =0 ; desc.qw3 =0 ; -qi_submit_sync (&desc, svm->iommu); +qi_submit_sync (svm->iommu, &desc,1 ,0 );if (sdev->dev_iotlb) { desc.qw0 = QI_DEV_EIOTLB_PASID(svm->pasid) |
@@ -162,7 +162,7 @@ static void intel_flush_svm_range_dev (struct intel_svm *svm, struct intel_svm_d } desc.qw2 =0 ; desc.qw3 =0 ; -qi_submit_sync (&desc, svm->iommu); +qi_submit_sync (svm->iommu, &desc,1 ,0 ); } }
@@ -846,7 +846,7 @@ static irqreturn_t prq_event_thread(int irq, void *d) sizeof(req->priv_data)); resp.qw2 =0 ; resp.qw3 =0 ; -qi_submit_sync (&resp, iommu); +qi_submit_sync (iommu, &resp,1 ,0 ); } head = (head +sizeof (*req)) &PRQ_RING_MASK ; }
drivers/iommu/intel_irq_remapping.c +1 -1
@@ -151,7 +151,7 @@ static int qi_flush_iec(struct intel_iommu *iommu, int index, int mask) desc.qw2 =0 ; desc.qw3 =0 ; -return qi_submit_sync (&desc, iommu); +return qi_submit_sync (iommu, &desc,1 ,0 ); } static int modify_irte(struct irq_2_iommu *irq_iommu,
include/linux/intel-iommu.h +8 -1
@@ -333,6 +333,7 @@ enum {#define QI_IWD_STATUS_DATA (d) (((u64)d) << 32)#define QI_IWD_STATUS_WRITE (((u64)1) << 5) +#define QI_IWD_PRQ_DRAIN (((u64)1) << 7)#define QI_IOTLB_DID (did) (((u64)did) << 16)#define QI_IOTLB_DR (dr) (((u64)dr) << 7)
@@ -702,7 +703,13 @@ void qi_flush_dev_iotlb_pasid(struct intel_iommu *iommu, u16 sid, u16 pfsid,void qi_flush_pasid_cache (struct intel_iommu *iommu,u16 did,u64 granu,int pasid); -extern int qi_submit_sync (struct qi_desc *desc,struct intel_iommu *iommu); +int qi_submit_sync (struct intel_iommu *iommu,struct qi_desc *desc, +unsigned int count,unsigned long options); +/* +* Options used in qi_submit_sync: +* QI_OPT_WAIT_DRAIN - Wait for PRQ drain completion, spec 6.5.2.8. +*/ +#define QI_OPT_WAIT_DRAIN BIT(0)extern int dmar_ir_support (void );
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 13/18 iommu/vt-d: debugfs: Add support to show inv queue internals
In reply to Lu BaoluExport invalidation queue internals of each iommu device through the debugfs.
Example of such dump on a Skylake machine:
$ sudo cat /sys/kernel/debug/iommu/intel/invalidation_queue
Invalidation queue on IOMMU: dmar1
Base: 0x1672c9000 Head: 80 Tail: 80
Index qw0 qw1 status
0 0000000000000004 0000000000000000 0000000000000000
1 0000000200000025 00000001672be804 0000000000000000
2 0000000000000011 0000000000000000 0000000000000000
3 0000000200000025 00000001672be80c 0000000000000000
4 00000000000000d2 0000000000000000 0000000000000000
5 0000000200000025 00000001672be814 0000000000000000
6 0000000000000014 0000000000000000 0000000000000000
7 0000000200000025 00000001672be81c 0000000000000000
8 0000000000000014 0000000000000000 0000000000000000
9 0000000200000025 00000001672be824 0000000000000000--- drivers/iommu/intel-iommu-debugfs.c | 62 +++++++++++++++++++++++++++++ 1 file changed, 62 insertions(+)
drivers/iommu/intel-iommu-debugfs.c +62 -0
@@ -372,6 +372,66 @@ static int domain_translation_struct_show(struct seq_file *m, void *unused) }DEFINE_SHOW_ATTRIBUTE (domain_translation_struct); +static void invalidation_queue_entry_show (struct seq_file *m, +struct intel_iommu *iommu) +{ +int index, shift =qi_shift (iommu); +struct qi_desc *desc; +int offset; + +if (ecap_smts (iommu->ecap)) +seq_puts (m,"Index\t\tqw0\t\t\tqw1\t\t\tqw2\t\t\tqw3\t\t\tstatus\n" ); +else +seq_puts (m,"Index\t\tqw0\t\t\tqw1\t\t\tstatus\n" ); + +for (index =0 ; index <QI_LENGTH ; index++) { + offset = index << shift; + desc = iommu->qi->desc + offset; +if (ecap_smts (iommu->ecap)) +seq_printf (m,"%5d\t%016llx\t%016llx\t%016llx\t%016llx\t%016x\n" , + index, desc->qw0, desc->qw1, + desc->qw2, desc->qw3, + iommu->qi->desc_status[index]); +else +seq_printf (m,"%5d\t%016llx\t%016llx\t%016x\n" , + index, desc->qw0, desc->qw1, + iommu->qi->desc_status[index]); + } +} + +static int invalidation_queue_show (struct seq_file *m,void *unused) +{ +struct dmar_drhd_unit *drhd; +struct intel_iommu *iommu; +unsigned long flags; +struct q_inval *qi; +int shift; + +rcu_read_lock (); +for_each_active_iommu (iommu, drhd) { + qi = iommu->qi; + shift =qi_shift (iommu); + +if (!qi || !ecap_qis (iommu->ecap)) +continue ; + +seq_printf (m,"Invalidation queue on IOMMU: %s\n" , iommu->name); + +raw_spin_lock_irqsave (&qi->q_lock, flags); +seq_printf (m," Base: 0x%llx\tHead: %lld\tTail: %lld\n" , + (u64 )virt_to_phys (qi->desc), +dmar_readq (iommu->reg +DMAR_IQH_REG ) >> shift, +dmar_readq (iommu->reg +DMAR_IQT_REG ) >> shift); +invalidation_queue_entry_show (m, iommu); +raw_spin_unlock_irqrestore (&qi->q_lock, flags); +seq_putc (m,'\n' ); + } +rcu_read_unlock (); + +return 0 ; +} +DEFINE_SHOW_ATTRIBUTE (invalidation_queue); +#ifdef CONFIG_IRQ_REMAP static void ir_tbl_remap_entry_show (struct seq_file *m,struct intel_iommu *iommu)
@@ -490,6 +550,8 @@ void __init intel_iommu_debugfs_init(void)debugfs_create_file ("domain_translation_struct" ,0444 , intel_iommu_debug,NULL , &domain_translation_struct_fops); +debugfs_create_file ("invalidation_queue" ,0444 , intel_iommu_debug, +NULL , &invalidation_queue_fops);#ifdef CONFIG_IRQ_REMAP debugfs_create_file ("ir_translation_struct" ,0444 , intel_iommu_debug,NULL , &ir_translation_struct_fops);
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 14/18 iommu/vt-d: Disable non-recoverable fault processing before unbind
In reply to Lu BaoluWhen a PASID is used for SVA by the device, it's possible that the PASID entry is cleared before the device flushes all ongoing DMA requests. The IOMMU should tolerate and ignore the non-recoverable faults caused by the untranslated requests from this device.
For example, when an exception happens, the process terminates before the device driver stops DMA and call IOMMU driver to unbind PASID. The flow of process exist is as follows:
do_exit() {
exit_mm() {
mm_put();
exit_mmap() {
intel_invalidate_range() //mmu notifier
tlb_finish_mmu()
mmu_notifier_release(mm) {
intel_iommu_release() {
[2] intel_iommu_teardown_pasid();
intel_iommu_flush_tlbs();
}
}
unmap_vmas();
free_pgtables();
};
}
exit_files(tsk) {
close_files() {
dsa_close();
[1] dsa_stop_dma();
intel_svm_unbind_pasid();
}
}
}Care must be taken on VT-d to avoid unrecoverable faults between the time window of [1] and [2]. [Process exist flow was contributed by Jacob Pan.]
Intel VT-d provides such function through the FPD bit of the PASID entry. This sets FPD bit when PASID entry is changing from present to nonpresent in the mm notifier and will clear it when the pasid is unbound.
--- drivers/iommu/intel-iommu.c | 4 ++-- drivers/iommu/intel-pasid.c | 26 +++++++++++++++++++++----- drivers/iommu/intel-pasid.h | 4 +++- drivers/iommu/intel-svm.c | 9 ++++++--- 4 files changed, 32 insertions(+), 11 deletions(-)
drivers/iommu/intel-iommu.c +2 -2
@@ -5005,7 +5005,7 @@ static void __dmar_remove_one_dev_info(struct device_domain_info *info)if (info->dev) {if (dev_is_pci (info->dev) &&sm_supported (iommu))intel_pasid_tear_down_entry (iommu, info->dev, -PASID_RID2PASID ); +PASID_RID2PASID , false);iommu_disable_dev_iotlb (info);domain_context_clear (iommu, info->dev);
@@ -5234,7 +5234,7 @@ static void aux_domain_remove_dev(struct dmar_domain *domain,auxiliary_unlink_device (domain, dev);spin_lock (&iommu->lock); -intel_pasid_tear_down_entry (iommu, dev, domain->default_pasid); +intel_pasid_tear_down_entry (iommu, dev, domain->default_pasid, false);domain_detach_iommu (domain, iommu);spin_unlock (&iommu->lock);
drivers/iommu/intel-pasid.c +21 -5
@@ -292,7 +292,20 @@ static inline void pasid_clear_entry(struct pasid_entry *pe)WRITE_ONCE (pe->val[7 ],0 ); } -static void intel_pasid_clear_entry (struct device *dev,int pasid) +static inline void pasid_clear_entry_with_fpd (struct pasid_entry *pe) +{ +WRITE_ONCE (pe->val[0 ],PASID_PTE_FPD ); +WRITE_ONCE (pe->val[1 ],0 ); +WRITE_ONCE (pe->val[2 ],0 ); +WRITE_ONCE (pe->val[3 ],0 ); +WRITE_ONCE (pe->val[4 ],0 ); +WRITE_ONCE (pe->val[5 ],0 ); +WRITE_ONCE (pe->val[6 ],0 ); +WRITE_ONCE (pe->val[7 ],0 ); +} + +static void +intel_pasid_clear_entry (struct device *dev,int pasid,bool fault_ignore) {struct pasid_entry *pe;
@@ -300,7 +313,10 @@ static void intel_pasid_clear_entry(struct device *dev, int pasid)if (WARN_ON (!pe))return ; -pasid_clear_entry (pe); +if (fault_ignore &&pasid_pte_is_present (pe)) +pasid_clear_entry_with_fpd (pe); +else +pasid_clear_entry (pe); }static inline void pasid_set_bits (u64 *ptr,u64 mask,u64 bits)
@@ -473,8 +489,8 @@ devtlb_invalidation_with_pasid(struct intel_iommu *iommu,qi_flush_dev_iotlb (iommu, sid, pfsid, qdep,0 ,64 -VTD_PAGE_SHIFT ); } -void intel_pasid_tear_down_entry (struct intel_iommu *iommu, -struct device *dev,int pasid) +void intel_pasid_tear_down_entry (struct intel_iommu *iommu,struct device *dev, +int pasid,bool fault_ignore) {struct pasid_entry *pte;u16 did;
@@ -484,7 +500,7 @@ void intel_pasid_tear_down_entry(struct intel_iommu *iommu,return ; did =pasid_get_domain_id (pte); -intel_pasid_clear_entry (dev, pasid); +intel_pasid_clear_entry (dev, pasid, fault_ignore);if (!ecap_coherent (iommu->ecap))clflush_cache_range (pte,sizeof (*pte));
drivers/iommu/intel-pasid.h +3 -1
@@ -15,6 +15,7 @@#define PASID_MAX 0x100000#define PASID_PTE_MASK 0x3F#define PASID_PTE_PRESENT 1 +#define PASID_PTE_FPD 2#define PDE_PFN_MASK PAGE_MASK#define PASID_PDE_SHIFT 6#define MAX_NR_PASID_BITS 20
@@ -120,7 +121,8 @@ int intel_pasid_setup_nested(struct intel_iommu *iommu,struct iommu_gpasid_bind_data_vtd *pasid_data, struct dmar_domain *domain, int addr_width);void intel_pasid_tear_down_entry (struct intel_iommu *iommu, -struct device *dev,int pasid); +struct device *dev,int pasid, +bool fault_ignore);int vcmd_alloc_pasid (struct intel_iommu *iommu,unsigned int *pasid);void vcmd_free_pasid (struct intel_iommu *iommu,unsigned int pasid);#endif /* __INTEL_PASID_H */
drivers/iommu/intel-svm.c +6 -3
@@ -207,7 +207,8 @@ static void intel_mm_release(struct mmu_notifier *mn, struct mm_struct *mm)*/ rcu_read_lock ();list_for_each_entry_rcu (sdev, &svm->devs, list) { -intel_pasid_tear_down_entry (svm->iommu, sdev->dev, svm->pasid); +intel_pasid_tear_down_entry (svm->iommu, sdev->dev, + svm->pasid, true);intel_flush_svm_range_dev (svm, sdev,0 ,-1 ,0 ); }rcu_read_unlock ();
@@ -396,7 +397,8 @@ int intel_svm_unbind_gpasid(struct device *dev, int pasid) sdev->users--;if (!sdev->users) {list_del_rcu (&sdev->list); -intel_pasid_tear_down_entry (iommu, dev, svm->pasid); +intel_pasid_tear_down_entry (iommu, dev, + svm->pasid, false);intel_flush_svm_range_dev (svm, sdev,0 ,-1 ,0 );/* TODO: Drain in flight PRQ for the PASID since it * may get reused soon, we don't want to
@@ -639,7 +641,8 @@ int intel_svm_unbind_mm(struct device *dev, int pasid)* to use. We have a *shared* PASID table, because it's * large and has to be physically contiguous. So it's * hard to be as defensive as we might like. */ -intel_pasid_tear_down_entry (iommu, dev, svm->pasid); +intel_pasid_tear_down_entry (iommu, dev, + svm->pasid, false);intel_flush_svm_range_dev (svm, sdev,0 ,-1 ,0 );kfree_rcu (sdev, rcu);
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 15/18 iommu/vt-d: Add page request draining support
In reply to Lu BaoluWhen a PASID is stopped or terminated, there can be pending PRQs (requests that haven't received responses) in remapping hardware. This adds the interface to drain page requests and call it when a PASID is terminated.
--- drivers/iommu/intel-svm.c | 107 ++++++++++++++++++++++++++++++++++-- include/linux/intel-iommu.h | 4 ++ 2 files changed, 106 insertions(+), 5 deletions(-)
drivers/iommu/intel-svm.c +102 -5
@@ -23,6 +23,7 @@#include "intel-pasid.h" static irqreturn_t prq_event_thread (int irq,void *d); +static void intel_svm_drain_prq (struct device *dev,int pasid);#define PRQ_ORDER 0
@@ -66,6 +67,8 @@ int intel_svm_enable_prq(struct intel_iommu *iommu)dmar_writeq (iommu->reg +DMAR_PQT_REG ,0ULL );dmar_writeq (iommu->reg +DMAR_PQA_REG ,virt_to_phys (iommu->prq) |PRQ_ORDER ); +init_completion (&iommu->prq_complete); +return 0 ; }
@@ -399,12 +402,8 @@ int intel_svm_unbind_gpasid(struct device *dev, int pasid)list_del_rcu (&sdev->list);intel_pasid_tear_down_entry (iommu, dev, svm->pasid, false); +intel_svm_drain_prq (dev, svm->pasid);intel_flush_svm_range_dev (svm, sdev,0 ,-1 ,0 ); -/* TODO: Drain in flight PRQ for the PASID since it -* may get reused soon, we don't want to -* confuse with its previous life. -* intel_svm_drain_prq(dev, pasid); -*/ kfree_rcu (sdev, rcu);if (list_empty (&svm->devs)) {
@@ -643,6 +642,7 @@ int intel_svm_unbind_mm(struct device *dev, int pasid)* hard to be as defensive as we might like. */ intel_pasid_tear_down_entry (iommu, dev, svm->pasid, false); +intel_svm_drain_prq (dev, svm->pasid);intel_flush_svm_range_dev (svm, sdev,0 ,-1 ,0 );kfree_rcu (sdev, rcu);
@@ -721,6 +721,93 @@ static bool is_canonical_address(u64 addr)return (((saddr << shift) >> shift) == saddr); } +/** +* intel_svm_drain_prq - Drain page requests and responses for a pasid +* @dev: target device +* @pasid: pasid for draining +* +* Drain all pending page requests and responses related to @pasid in both +* software and hardware. This is supposed to be called after the device +* driver has stopped DMA, the pasid entry has been cleared, and both IOTLB +* and DevTLB have been invalidated. +* +* It waits until all pending page requests for @pasid in the page fault +* queue are completed by the prq handling thread. Then follow the steps +* described in VT-d spec CH7.10 to drain all page requests and page +* responses pending in the hardware. +*/ +static void intel_svm_drain_prq (struct device *dev,int pasid) +{ +struct device_domain_info *info; +struct dmar_domain *domain; +struct intel_iommu *iommu; +struct qi_desc desc[3 ]; +struct pci_dev *pdev; +int head, tail; +u16 sid, did; +int qdep; + + info =get_domain_info (dev); +if (WARN_ON (!info || !dev_is_pci (dev))) +return ; + +if (!info->pri_enabled) +return ; + + iommu = info->iommu; + domain = info->domain; + pdev =to_pci_dev (dev); + sid =PCI_DEVID (info->bus, info->devfn); + did = domain->iommu_did[iommu->seq_id]; + qdep =pci_ats_queue_depth (pdev); + +/* +* Check and wait until all pending page requests in the queue are +* handled by the prq handling thread. +*/ +prq_retry: +reinit_completion (&iommu->prq_complete); + tail =dmar_readq (iommu->reg +DMAR_PQT_REG ) &PRQ_RING_MASK ; + head =dmar_readq (iommu->reg +DMAR_PQH_REG ) &PRQ_RING_MASK ; +while (head != tail) { +struct page_req_dsc *req; + + req = &iommu->prq[head /sizeof (*req)]; +if (!req->pasid_present || req->pasid != pasid) { + head = (head +sizeof (*req)) &PRQ_RING_MASK ; +continue ; + } + +wait_for_completion (&iommu->prq_complete); + goto prq_retry; + } + +/* +* Perform steps described in VT-d spec CH7.10 to drain page +* requests and responses in hardware. +*/ +memset (desc,0 ,sizeof (desc)); + desc[0 ].qw0 =QI_IWD_STATUS_DATA (QI_DONE ) | +QI_IWD_FENCE | +QI_IWD_TYPE ; + desc[1 ].qw0 =QI_EIOTLB_PASID (pasid) | +QI_EIOTLB_DID (did) | +QI_EIOTLB_GRAN (QI_GRAN_NONG_PASID ) | +QI_EIOTLB_TYPE ; + desc[2 ].qw0 =QI_DEV_EIOTLB_PASID (pasid) | +QI_DEV_EIOTLB_SID (sid) | +QI_DEV_EIOTLB_QDEP (qdep) | +QI_DEIOTLB_TYPE | +QI_DEV_IOTLB_PFSID (info->pfsid); +qi_retry: +reinit_completion (&iommu->prq_complete); +qi_submit_sync (iommu, desc,3 ,QI_OPT_WAIT_DRAIN ); +if (readl (iommu->reg +DMAR_PRS_REG ) &DMA_PRS_PRO ) { +wait_for_completion (&iommu->prq_complete); + goto qi_retry; + } +} +static irqreturn_t prq_event_thread (int irq,void *d) {struct intel_iommu *iommu = d;
@@ -856,6 +943,16 @@ static irqreturn_t prq_event_thread(int irq, void *d)dmar_writeq (iommu->reg +DMAR_PQH_REG , tail); +/* +* Clear the page request overflow bit and wake up all threads that +* are waiting for the completion of this handling. +*/ +if (readl (iommu->reg +DMAR_PRS_REG ) &DMA_PRS_PRO ) +writel (DMA_PRS_PRO , iommu->reg +DMAR_PRS_REG ); + +if (!completion_done (&iommu->prq_complete)) +complete (&iommu->prq_complete); +return IRQ_RETVAL (handled); }
include/linux/intel-iommu.h +4 -0
@@ -292,6 +292,8 @@/* PRS_REG */ #define DMA_PRS_PPR ((u32)1) +#define DMA_PRS_PRO ((u32)2) +#define DMA_VCS_PAS ((u64)1)#define IOMMU_WAIT_OP (iommu, offset, op, cond, sts) \
@@ -333,6 +335,7 @@ enum {#define QI_IWD_STATUS_DATA (d) (((u64)d) << 32)#define QI_IWD_STATUS_WRITE (((u64)1) << 5) +#define QI_IWD_FENCE (((u64)1) << 6)#define QI_IWD_PRQ_DRAIN (((u64)1) << 7)#define QI_IOTLB_DID (did) (((u64)did) << 16)
@@ -582,6 +585,7 @@ struct intel_iommu {#ifdef CONFIG_INTEL_IOMMU_SVM struct page_req_dsc *prq;unsigned char prq_name[16 ];/* Name for PRQ interrupt */ +struct completion prq_complete;struct ioasid_allocator_ops pasid_allocator;/* Custom allocator for PASIDs */ #endif struct q_inval *qi;/* Queued invalidation info */
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 16/18 iommu/vt-d: Remove redundant IOTLB flush
In reply to Lu BaoluIOTLB flush already included in the PASID tear down and the page request drain process. There is no need to flush again.
--- drivers/iommu/intel-svm.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-)
drivers/iommu/intel-svm.c +1 -5
@@ -209,11 +209,9 @@ static void intel_mm_release(struct mmu_notifier *mn, struct mm_struct *mm)* *has* to handle gracefully without affecting other processes. */ rcu_read_lock (); -list_for_each_entry_rcu (sdev, &svm->devs, list) { +list_for_each_entry_rcu (sdev, &svm->devs, list)intel_pasid_tear_down_entry (svm->iommu, sdev->dev, svm->pasid, true); -intel_flush_svm_range_dev (svm, sdev,0 ,-1 ,0 ); - }rcu_read_unlock (); }
@@ -403,7 +401,6 @@ int intel_svm_unbind_gpasid(struct device *dev, int pasid)intel_pasid_tear_down_entry (iommu, dev, svm->pasid, false);intel_svm_drain_prq (dev, svm->pasid); -intel_flush_svm_range_dev (svm, sdev,0 ,-1 ,0 );kfree_rcu (sdev, rcu);if (list_empty (&svm->devs)) {
@@ -643,7 +640,6 @@ int intel_svm_unbind_mm(struct device *dev, int pasid)intel_pasid_tear_down_entry (iommu, dev, svm->pasid, false);intel_svm_drain_prq (dev, svm->pasid); -intel_flush_svm_range_dev (svm, sdev,0 ,-1 ,0 );kfree_rcu (sdev, rcu);if (list_empty (&svm->devs)) {
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 17/18 iommu/vt-d: Remove duplicated check in intel_svm_bind_mm()
In reply to Lu BaoluThe info and info->pasid_support have already been checked in previous intel_iommu_enable_pasid() call. No need to check again.
--- drivers/iommu/intel-svm.c | 5 ----- 1 file changed, 5 deletions(-)
drivers/iommu/intel-svm.c +0 -5
@@ -501,11 +501,6 @@ intel_svm_bind_mm(struct device *dev, int flags, struct svm_dev_ops *ops, } info =get_domain_info (dev); -if (!info || !info->pasid_supported) { -kfree (sdev); - goto out; - } - sdev->did =FLPT_DEFAULT_DID ; sdev->sid =PCI_DEVID (info->bus, info->devfn);if (info->ats_enabled) {
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerPATCH 18/18 iommu/vt-d: Remove IOVA handling code from the non-dma_ops path
In reply to Lu BaoluFrom: Tom Murphy <[email protected]>
There's no need for the non-dma_ops path to keep track of IOVAs. The whole point of the non-dma_ops path is that it allows the IOVAs to be handled separately. The IOVA handling code removed in this patch is pointless.
--- drivers/iommu/intel-iommu.c | 95 +++++++++++++------------------------ 1 file changed, 32 insertions(+), 63 deletions(-)
drivers/iommu/intel-iommu.c +32 -63
@@ -1892,11 +1892,6 @@ static int dmar_init_reserved_ranges(void)return 0 ; } -static void domain_reserve_special_ranges (struct dmar_domain *domain) -{ -copy_reserved_iova (&reserved_iova_list, &domain->iovad); -} -static inline int guestwidth_to_adjustwidth (int gaw) {int agaw;
@@ -1918,7 +1913,8 @@ static void domain_exit(struct dmar_domain *domain)domain_remove_dev_info (domain);/* destroy iovas */ -put_iova_domain (&domain->iovad); +if (domain->domain.type ==IOMMU_DOMAIN_DMA ) +put_iova_domain (&domain->iovad);if (domain->pgd) {struct page *freelist;
@@ -2627,19 +2623,9 @@ static struct dmar_domain *dmar_insert_one_dev_info(struct intel_iommu *iommu, }static int iommu_domain_identity_map (struct dmar_domain *domain, -unsigned long long start, -unsigned long long end) +unsigned long first_vpfn, +unsigned long last_vpfn) { -unsigned long first_vpfn = start >>VTD_PAGE_SHIFT ; -unsigned long last_vpfn = end >>VTD_PAGE_SHIFT ; - -if (!reserve_iova (&domain->iovad,dma_to_mm_pfn (first_vpfn), -dma_to_mm_pfn (last_vpfn))) { -pr_err ("Reserving iova failed\n" ); -return -ENOMEM ; - } - -pr_debug ("Mapping reserved region %llx-%llx\n" , start, end);/* * RMRR range might have overlap with physical memory range, * clear it first
@@ -2677,7 +2663,8 @@ static int __init si_domain_init(int hw)for_each_mem_pfn_range (i, nid, &start_pfn, &end_pfn,NULL ) { ret =iommu_domain_identity_map (si_domain, -PFN_PHYS (start_pfn),PFN_PHYS (end_pfn)); +mm_to_dma_pfn (start_pfn), +mm_to_dma_pfn (end_pfn));if (ret)return ret; }
@@ -4547,58 +4534,37 @@ static int intel_iommu_memory_notifier(struct notifier_block *nb, unsigned long val, void *v) { struct memory_notify *mhp = v; -unsigned long long start, end; -unsigned long start_vpfn, last_vpfn; +unsigned long start_vpfn =mm_to_dma_pfn (mhp->start_pfn); +unsigned long last_vpfn =mm_to_dma_pfn (mhp->start_pfn + + mhp->nr_pages -1 );switch (val) {case MEM_GOING_ONLINE : - start = mhp->start_pfn <<PAGE_SHIFT ; - end = ((mhp->start_pfn + mhp->nr_pages) <<PAGE_SHIFT ) -1 ; -if (iommu_domain_identity_map (si_domain, start, end)) { -pr_warn ("Failed to build identity map for [%llx-%llx]\n" , - start, end); +if (iommu_domain_identity_map (si_domain, + start_vpfn, last_vpfn)) { +pr_warn ("Failed to build identity map for [%lx-%lx]\n" , + start_vpfn, last_vpfn);return NOTIFY_BAD ; }break ;case MEM_OFFLINE :case MEM_CANCEL_ONLINE : - start_vpfn =mm_to_dma_pfn (mhp->start_pfn); - last_vpfn =mm_to_dma_pfn (mhp->start_pfn + mhp->nr_pages -1 ); -while (start_vpfn <= last_vpfn) { -struct iova *iova; + {struct dmar_drhd_unit *drhd;struct intel_iommu *iommu;struct page *freelist; - iova =find_iova (&si_domain->iovad, start_vpfn); -if (iova ==NULL ) { -pr_debug ("Failed get IOVA for PFN %lx\n" , - start_vpfn); -break ; - } - - iova =split_and_remove_iova (&si_domain->iovad, iova, - start_vpfn, last_vpfn); -if (iova ==NULL ) { -pr_warn ("Failed to split IOVA PFN [%lx-%lx]\n" , - start_vpfn, last_vpfn); -return NOTIFY_BAD ; - } - - freelist =domain_unmap (si_domain, iova->pfn_lo, - iova->pfn_hi); + freelist =domain_unmap (si_domain, + start_vpfn, last_vpfn);rcu_read_lock ();for_each_active_iommu (iommu, drhd)iommu_flush_iotlb_psi (iommu, si_domain, - iova->pfn_lo,iova_size (iova), + start_vpfn, mhp->nr_pages, !freelist,0 );rcu_read_unlock ();dma_free_pagelist (freelist); - - start_vpfn = iova->pfn_hi +1 ; -free_iova_mem (iova); }break ; }
@@ -4626,8 +4592,9 @@ static void free_all_cpu_cached_iovas(unsigned int cpu)for (did =0 ; did <cap_ndoms (iommu->cap); did++) { domain =get_iommu_domain (iommu, (u16 )did); -if (!domain) +if (!domain || domain->domain.type !=IOMMU_DOMAIN_DMA )continue ; +free_cpu_cached_iovas (cpu, &domain->iovad); } }
@@ -5037,9 +5004,6 @@ static int md_domain_init(struct dmar_domain *domain, int guest_width) {int adjust_width; -init_iova_domain (&domain->iovad,VTD_PAGE_SIZE ,IOVA_START_PFN ); -domain_reserve_special_ranges (domain); -/* calculate AGAW */ domain->gaw = guest_width; adjust_width =guestwidth_to_adjustwidth (guest_width);
@@ -5058,11 +5022,21 @@ static int md_domain_init(struct dmar_domain *domain, int guest_width)return 0 ; } +static void intel_init_iova_domain (struct dmar_domain *dmar_domain) +{ +init_iova_domain (&dmar_domain->iovad,VTD_PAGE_SIZE ,IOVA_START_PFN ); +copy_reserved_iova (&reserved_iova_list, &dmar_domain->iovad); + +if (!intel_iommu_strict && +init_iova_flush_queue (&dmar_domain->iovad, + iommu_flush_iova, iova_entry_free)) +pr_info ("iova flush queue initialization failed\n" ); +} +static struct iommu_domain *intel_iommu_domain_alloc (unsigned type) {struct dmar_domain *dmar_domain;struct iommu_domain *domain; -int ret;switch (type) {case IOMMU_DOMAIN_DMA :
@@ -5079,13 +5053,8 @@ static struct iommu_domain *intel_iommu_domain_alloc(unsigned type)return NULL ; } -if (!intel_iommu_strict && type ==IOMMU_DOMAIN_DMA ) { - ret =init_iova_flush_queue (&dmar_domain->iovad, - iommu_flush_iova, - iova_entry_free); -if (ret) -pr_info ("iova flush queue initialization failed\n" ); - } +if (type ==IOMMU_DOMAIN_DMA ) +intel_init_iova_domain (dmar_domain);domain_update_iommu_cap (dmar_domain);
Signature · 6 lines
2.17.1 _______________________________________________ iommu mailing list [email protected] https://lists.linuxfoundation.org/mailman/listinfo/iommu
Joerg Roedel
maintainerQuoted 21 lines · Lu Baolu
Jacob Pan (9):
iommu/vt-d: Move domain helper to header iommu/vt-d: Use a helper function to skip agaw for SL iommu/vt-d: Add nested translation helper function iommu/vt-d: Add bind guest PASID support iommu/vt-d: Support flushing more translation cache types iommu/vt-d: Add svm/sva invalidate function iommu/vt-d: Add custom allocator for IOASID iommu/vt-d: Report SVA feature with generic flag iommu/vt-d: Replace intel SVM APIs with generic SVA APIsLu Baolu (8):
iommu/vt-d: Enlightened PASID allocation
iommu/vt-d: Addget_domain_info()helper
iommu/vt-d: Multiple descriptors perqi_submit_sync()iommu/vt-d: debugfs: Add support to show inv queue internals iommu/vt-d: Disable non-recoverable fault processing before unbind iommu/vt-d: Add page request draining support iommu/vt-d: Remove redundant IOTLB flush
iommu/vt-d: Remove duplicated check inintel_svm_bind_mm()
Quoted 2 lines
Tom Murphy (1):
iommu/vt-d: Remove IOVA handling code from the non-dma_ops path
Applied, thanks Baolu.
_______________________________________________
iommu mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/iommu
Joerg Roedel
maintainerHi Baolu,
Quoted 21 lines · Lu Baolu
Jacob Pan (9):
iommu/vt-d: Move domain helper to header iommu/vt-d: Use a helper function to skip agaw for SL iommu/vt-d: Add nested translation helper function iommu/vt-d: Add bind guest PASID support iommu/vt-d: Support flushing more translation cache types iommu/vt-d: Add svm/sva invalidate function iommu/vt-d: Add custom allocator for IOASID iommu/vt-d: Report SVA feature with generic flag iommu/vt-d: Replace intel SVM APIs with generic SVA APIsLu Baolu (8):
iommu/vt-d: Enlightened PASID allocation
iommu/vt-d: Addget_domain_info()helper
iommu/vt-d: Multiple descriptors perqi_submit_sync()iommu/vt-d: debugfs: Add support to show inv queue internals iommu/vt-d: Disable non-recoverable fault processing before unbind iommu/vt-d: Add page request draining support iommu/vt-d: Remove redundant IOTLB flush
iommu/vt-d: Remove duplicated check inintel_svm_bind_mm()
Quoted 2 lines
Tom Murphy (1):
iommu/vt-d: Remove IOVA handling code from the non-dma_ops path
I get these new compile warnings on x86-32 after pulling in these patches:
drivers/iommu/intel-pasid.c:818:6: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast] drivers/iommu/intel-pasid.c:821:9: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast] drivers/iommu/intel-pasid.c:824:23: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast] drivers/iommu/intel-svm.c:343:45: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
Can you please take a look and send me a fix for these?
Regards,
Joerg
_______________________________________________
iommu mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerHi Joerg,
Quoted 30 lines · Joerg Roedel
Hi Baolu,
Quoted 21 lines · Lu Baolu
Jacob Pan (9):
iommu/vt-d: Move domain helper to header iommu/vt-d: Use a helper function to skip agaw for SL iommu/vt-d: Add nested translation helper function iommu/vt-d: Add bind guest PASID support iommu/vt-d: Support flushing more translation cache types iommu/vt-d: Add svm/sva invalidate function iommu/vt-d: Add custom allocator for IOASID iommu/vt-d: Report SVA feature with generic flag iommu/vt-d: Replace intel SVM APIs with generic SVA APIsLu Baolu (8):
iommu/vt-d: Enlightened PASID allocation
iommu/vt-d: Addget_domain_info()helper
iommu/vt-d: Multiple descriptors perqi_submit_sync()iommu/vt-d: debugfs: Add support to show inv queue internals iommu/vt-d: Disable non-recoverable fault processing before unbind iommu/vt-d: Add page request draining support iommu/vt-d: Remove redundant IOTLB flush
iommu/vt-d: Remove duplicated check inintel_svm_bind_mm()Quoted 2 lines
Tom Murphy (1):
iommu/vt-d: Remove IOVA handling code from the non-dma_ops pathI get these new compile warnings on x86-32 after pulling in these patches:
Quoted 6 lines
drivers/iommu/intel-pasid.c:818:6: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]drivers/iommu/intel-pasid.c:821:9: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]drivers/iommu/intel-pasid.c:824:23: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast]drivers/iommu/intel-svm.c:343:45: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]Can you please take a look and send me a fix for these?
Thanks a lot for catching this. I've post a fix. Please check it.
Best regards,
baolu
_______________________________________________
iommu mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/iommu
[Regression] Re: [PATCH 18/18] iommu/vt-d: Remove IOVA handling code from the non-dma_ops path
In reply to Lu BaoluQuoted 7 lines · On Sat, 16 May 2020 14:21:01 +0800 Lu Baolu <[email protected]> wrote:
From: Tom Murphy <[email protected]>
There's no need for the non-dma_ops path to keep track of IOVAs. The whole point of the non-dma_ops path is that it allows the IOVAs to be handled separately. The IOVA handling code removed in this patch is pointless.
Quoted 5 lines
Signed-off-by Tom Murphy <[email protected]>Signed-off-by Lu Baolu <[email protected]>--- drivers/iommu/intel-iommu.c | 95 +++++++++++++------------------------ 1 file changed, 32 insertions(+), 63 deletions(-)
This commit results in a massive increase in memory use from the VT-d code. I have a 16GB system where I reserve 7168 2MB hugespages for VM usage (14GB), leaving the host with 2GB. I can no longer even boot the host in this configuration. Bisecting to this commit, I find that reverting this change shows the following memory usage difference immediately after boot (no hugepages, nosmp, single user, intel_iommu=on iommu=pt):
@e70b081c6f37:
total used free shared buff/cache available
Mem: 16090860 2219372 13673044 1040 198444 13602664 Swap: 2097148 0 2097148
reverting e70b081c6f37:
total used free shared buff/cache available
Mem: 16090852 101648 15789156 1040 200048 15719572 Swap: 2097148 0 2097148
More than 2GB of additional memory used! There's also a notable stall during bootup for this allocation:
[ 9.703360] DMAR: No ATSR found [ 9.709768] DMAR: dmar0: Using Queued invalidation [ 9.719370] DMAR: dmar1: Using Queued invalidation
### 4+ seconds! ###
[ 14.076387] pci 0000:00:00.0: Adding to iommu group 0 [ 14.086515] pci 0000:00:01.0: Adding to iommu group 1 [ 14.096635] pci 0000:00:02.0: Adding to iommu group 2
Thanks,
Alex
Quoted 178 lines
drivers/iommu/intel-iommu.c +30
-58@@ -1892,11 +1892,6 @@ static int dmar_init_reserved_ranges(void)return 0 ; } -static void domain_reserve_special_ranges (struct dmar_domain *domain) -{ -copy_reserved_iova (&reserved_iova_list, &domain->iovad); -} -static inline int guestwidth_to_adjustwidth (int gaw) {int agaw;@@ -1918,7 +1913,8 @@ static void domain_exit(struct dmar_domain *domain)domain_remove_dev_info (domain);/* destroy iovas */ -put_iova_domain (&domain->iovad); +if (domain->domain.type ==IOMMU_DOMAIN_DMA ) +put_iova_domain (&domain->iovad);if (domain->pgd) {struct page *freelist;@@ -2627,19 +2623,9 @@ static struct dmar_domain *dmar_insert_one_dev_info(struct intel_iommu *iommu, }static int iommu_domain_identity_map (struct dmar_domain *domain, -unsigned long long start, -unsigned long long end) +unsigned long first_vpfn, +unsigned long last_vpfn) { -unsigned long first_vpfn = start >>VTD_PAGE_SHIFT ; -unsigned long last_vpfn = end >>VTD_PAGE_SHIFT ; - -if (!reserve_iova (&domain->iovad,dma_to_mm_pfn (first_vpfn), -dma_to_mm_pfn (last_vpfn))) { -pr_err ("Reserving iova failed\n" ); -return -ENOMEM ; - } - -pr_debug ("Mapping reserved region %llx-%llx\n" , start, end);/* * RMRR range might have overlap with physical memory range, * clear it first @@ -2677,7 +2663,8 @@ static int __init si_domain_init(int hw)for_each_mem_pfn_range (i, nid, &start_pfn, &end_pfn,NULL ) { ret =iommu_domain_identity_map (si_domain, -PFN_PHYS (start_pfn),PFN_PHYS (end_pfn)); +mm_to_dma_pfn (start_pfn), +mm_to_dma_pfn (end_pfn));if (ret)return ret; }@@ -4547,58 +4534,37 @@ static int intel_iommu_memory_notifier(struct notifier_block *nb, unsigned long val, void *v) { struct memory_notify *mhp = v; -unsigned long long start, end; -unsigned long start_vpfn, last_vpfn; +unsigned long start_vpfn =mm_to_dma_pfn (mhp->start_pfn); +unsigned long last_vpfn =mm_to_dma_pfn (mhp->start_pfn + + mhp->nr_pages -1 );switch (val) {case MEM_GOING_ONLINE : - start = mhp->start_pfn <<PAGE_SHIFT ; - end = ((mhp->start_pfn + mhp->nr_pages) <<PAGE_SHIFT ) -1 ; -if (iommu_domain_identity_map (si_domain, start, end)) { -pr_warn ("Failed to build identity map for [%llx-%llx]\n" , - start, end); +if (iommu_domain_identity_map (si_domain, + start_vpfn, last_vpfn)) { +pr_warn ("Failed to build identity map for [%lx-%lx]\n" , + start_vpfn, last_vpfn);return NOTIFY_BAD ; }break ;case MEM_OFFLINE :case MEM_CANCEL_ONLINE : - start_vpfn =mm_to_dma_pfn (mhp->start_pfn); - last_vpfn =mm_to_dma_pfn (mhp->start_pfn + mhp->nr_pages -1 ); -while (start_vpfn <= last_vpfn) { -struct iova *iova; + {struct dmar_drhd_unit *drhd;struct intel_iommu *iommu;struct page *freelist; - iova =find_iova (&si_domain->iovad, start_vpfn); -if (iova ==NULL ) { -pr_debug ("Failed get IOVA for PFN %lx\n" , - start_vpfn); -break ; - } - - iova =split_and_remove_iova (&si_domain->iovad, iova, - start_vpfn, last_vpfn); -if (iova ==NULL ) { -pr_warn ("Failed to split IOVA PFN [%lx-%lx]\n" , - start_vpfn, last_vpfn); -return NOTIFY_BAD ; - } - - freelist =domain_unmap (si_domain, iova->pfn_lo, - iova->pfn_hi); + freelist =domain_unmap (si_domain, + start_vpfn, last_vpfn);rcu_read_lock ();for_each_active_iommu (iommu, drhd)iommu_flush_iotlb_psi (iommu, si_domain, - iova->pfn_lo,iova_size (iova), + start_vpfn, mhp->nr_pages, !freelist,0 );rcu_read_unlock ();dma_free_pagelist (freelist); - - start_vpfn = iova->pfn_hi +1 ; -free_iova_mem (iova); }break ; }@@ -4626,8 +4592,9 @@ static void free_all_cpu_cached_iovas(unsigned int cpu)for (did =0 ; did <cap_ndoms (iommu->cap); did++) { domain =get_iommu_domain (iommu, (u16 )did); -if (!domain) +if (!domain || domain->domain.type !=IOMMU_DOMAIN_DMA )continue ; +free_cpu_cached_iovas (cpu, &domain->iovad); } }@@ -5037,9 +5004,6 @@ static int md_domain_init(struct dmar_domain *domain, int guest_width) {int adjust_width; -init_iova_domain (&domain->iovad,VTD_PAGE_SIZE ,IOVA_START_PFN ); -domain_reserve_special_ranges (domain); -/* calculate AGAW */ domain->gaw = guest_width; adjust_width =guestwidth_to_adjustwidth (guest_width);@@ -5058,11 +5022,21 @@ static int md_domain_init(struct dmar_domain *domain, int guest_width)return 0 ; } +static void intel_init_iova_domain (struct dmar_domain *dmar_domain) +{ +init_iova_domain (&dmar_domain->iovad,VTD_PAGE_SIZE ,IOVA_START_PFN ); +copy_reserved_iova (&reserved_iova_list, &dmar_domain->iovad); + +if (!intel_iommu_strict && +init_iova_flush_queue (&dmar_domain->iovad, + iommu_flush_iova, iova_entry_free)) +pr_info ("iova flush queue initialization failed\n" ); +} +static struct iommu_domain *intel_iommu_domain_alloc (unsigned type) {struct dmar_domain *dmar_domain;struct iommu_domain *domain; -int ret;switch (type) {case IOMMU_DOMAIN_DMA :@@ -5079,13 +5053,8 @@ static struct iommu_domain *intel_iommu_domain_alloc(unsigned type)return NULL ; } -if (!intel_iommu_strict && type ==IOMMU_DOMAIN_DMA ) { - ret = init_iova_flush_queue(&dmar_domain->iovad,
drivers/iommu/intel-iommu.c static struct iommu_domain *intel_iommu_domain_alloc(unsigned type) · line 5060
Quoted 10 lines
drivers/iommu/intel-iommu.c +2
-5- iommu_flush_iova, - iova_entry_free); -if (ret) -pr_info ("iova flush queue initialization failed\n" ); - } +if (type ==IOMMU_DOMAIN_DMA ) +intel_init_iova_domain (dmar_domain);domain_update_iommu_cap (dmar_domain);
_______________________________________________
iommu mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerRe: [Regression] Re: [PATCH 18/18] iommu/vt-d: Remove IOVA handling code from the non-dma_ops path
Hi Alex,
Thanks for the report.
Quoted 40 lines · Alex Williamson
Quoted 7 lines · On Sat, 16 May 2020 14:21:01 +0800 Lu Baolu<[email protected]> wrote:
From: Tom Murphy<[email protected]>
There's no need for the non-dma_ops path to keep track of IOVAs. The whole point of the non-dma_ops path is that it allows the IOVAs to be handled separately. The IOVA handling code removed in this patch is pointless.
Quoted 5 lines
Signed-off-by Tom Murphy<[email protected]>Signed-off-by Lu Baolu<[email protected]>--- drivers/iommu/intel-iommu.c | 95 +++++++++++++------------------------ 1 file changed, 32 insertions(+), 63 deletions(-)This commit results in a massive increase in memory use from the VT-d code. I have a 16GB system where I reserve 7168 2MB hugespages for VM usage (14GB), leaving the host with 2GB. I can no longer even boot the host in this configuration. Bisecting to this commit, I find that reverting this change shows the following memory usage difference immediately after boot (no hugepages, nosmp, single user, intel_iommu=on iommu=pt):
@e70b081c6f37:
total used free shared buff/cache availableMem: 16090860 2219372 13673044 1040 198444 13602664 Swap: 2097148 0 2097148reverting
e70b081c6f37:
total used free shared buff/cache availableMem: 16090852 101648 15789156 1040 200048 15719572 Swap: 2097148 0 2097148More than 2GB of additional memory used! There's also a notable stall during bootup for this allocation:
[ 9.703360] DMAR: No ATSR found [ 9.709768] DMAR: dmar0: Using Queued invalidation [ 9.719370] DMAR: dmar1: Using Queued invalidation
Quoted 5 lines
### 4+ seconds! ###
[ 14.076387] pci 0000:00:00.0: Adding to iommu group 0 [ 14.086515] pci 0000:00:01.0: Adding to iommu group 1 [ 14.096635] pci 0000:00:02.0: Adding to iommu group 2
Can you please try below changes?
drivers/iommu/intel/iommu.c +3 -1
@@ -2705,7 +2705,9 @@ static int __init si_domain_init(int hw) end >> agaw_to_width(si_domain->agaw)))continue ; - ret =iommu_domain_identity_map (si_domain, start, end); + ret =iommu_domain_identity_map (si_domain, +mm_to_dma_pfn (start >>PAGE_SHIFT ), +mm_to_dma_pfn (end >>PAGE_SHIFT ));if (ret)return ret; }
Best regards,
baolu
_______________________________________________
iommu mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/iommu
Re: [Regression] Re: [PATCH 18/18] iommu/vt-d: Remove IOVA handling code from the non-dma_ops path
Quoted 68 lines · On Thu, 18 Jun 2020 09:52:56 +0800 Lu Baolu <[email protected]> wrote:
Hi Alex,
Thanks for the report.
Quoted 40 lines · Alex Williamson
Quoted 7 lines · On Sat, 16 May 2020 14:21:01 +0800 Lu Baolu<[email protected]> wrote:
From: Tom Murphy<[email protected]>
There's no need for the non-dma_ops path to keep track of IOVAs. The whole point of the non-dma_ops path is that it allows the IOVAs to be handled separately. The IOVA handling code removed in this patch is pointless.
Quoted 5 lines
Signed-off-by Tom Murphy<[email protected]>Signed-off-by Lu Baolu<[email protected]>--- drivers/iommu/intel-iommu.c | 95 +++++++++++++------------------------ 1 file changed, 32 insertions(+), 63 deletions(-)This commit results in a massive increase in memory use from the VT-d code. I have a 16GB system where I reserve 7168 2MB hugespages for VM usage (14GB), leaving the host with 2GB. I can no longer even boot the host in this configuration. Bisecting to this commit, I find that reverting this change shows the following memory usage difference immediately after boot (no hugepages, nosmp, single user, intel_iommu=on iommu=pt):
@e70b081c6f37:
total used free shared buff/cache availableMem: 16090860 2219372 13673044 1040 198444 13602664 Swap: 2097148 0 2097148reverting
e70b081c6f37:
total used free shared buff/cache availableMem: 16090852 101648 15789156 1040 200048 15719572 Swap: 2097148 0 2097148More than 2GB of additional memory used! There's also a notable stall during bootup for this allocation:
[ 9.703360] DMAR: No ATSR found [ 9.709768] DMAR: dmar0: Using Queued invalidation [ 9.719370] DMAR: dmar1: Using Queued invalidationQuoted 5 lines
### 4+ seconds! ###
[ 14.076387] pci 0000:00:00.0: Adding to iommu group 0 [ 14.086515] pci 0000:00:01.0: Adding to iommu group 1 [ 14.096635] pci 0000:00:02.0: Adding to iommu group 2Can you please try below changes?
drivers/iommu/intel/iommu.c +3
-1@@ -2705,7 +2705,9 @@ static int __init si_domain_init(int hw) end >> agaw_to_width(si_domain->agaw)))continue ; - ret =iommu_domain_identity_map (si_domain, start, end); + ret =iommu_domain_identity_map (si_domain, +mm_to_dma_pfn (start >>PAGE_SHIFT ), +mm_to_dma_pfn (end >>PAGE_SHIFT ));if (ret)return ret;
Quoted 1 line
}
Yes, this resolves it. Used memory after boot returns to similar numbers as without this commit. Thanks,
Alex
_______________________________________________
iommu mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/iommu
Lu Baolu
maintainerRe: [Regression] Re: [PATCH 18/18] iommu/vt-d: Remove IOVA handling code from the non-dma_ops path
Hi Alex,
Quoted 64 lines · Alex Williamson
Quoted 61 lines · On Thu, 18 Jun 2020 09:52:56 +0800 Lu Baolu <[email protected]> wrote:
Hi Alex,
Thanks for the report.
Quoted 40 lines · Alex Williamson
Quoted 7 lines · On Sat, 16 May 2020 14:21:01 +0800 Lu Baolu<[email protected]> wrote:
From: Tom Murphy<[email protected]>
There's no need for the non-dma_ops path to keep track of IOVAs. The whole point of the non-dma_ops path is that it allows the IOVAs to be handled separately. The IOVA handling code removed in this patch is pointless.
Quoted 5 lines
Signed-off-by Tom Murphy<[email protected]>Signed-off-by Lu Baolu<[email protected]>--- drivers/iommu/intel-iommu.c | 95 +++++++++++++------------------------ 1 file changed, 32 insertions(+), 63 deletions(-)This commit results in a massive increase in memory use from the VT-d code. I have a 16GB system where I reserve 7168 2MB hugespages for VM usage (14GB), leaving the host with 2GB. I can no longer even boot the host in this configuration. Bisecting to this commit, I find that reverting this change shows the following memory usage difference immediately after boot (no hugepages, nosmp, single user, intel_iommu=on iommu=pt):
@e70b081c6f37:
total used free shared buff/cache availableMem: 16090860 2219372 13673044 1040 198444 13602664 Swap: 2097148 0 2097148reverting
e70b081c6f37:
total used free shared buff/cache availableMem: 16090852 101648 15789156 1040 200048 15719572 Swap: 2097148 0 2097148More than 2GB of additional memory used! There's also a notable stall during bootup for this allocation:
[ 9.703360] DMAR: No ATSR found [ 9.709768] DMAR: dmar0: Using Queued invalidation [ 9.719370] DMAR: dmar1: Using Queued invalidationQuoted 5 lines
### 4+ seconds! ###
[ 14.076387] pci 0000:00:00.0: Adding to iommu group 0 [ 14.086515] pci 0000:00:01.0: Adding to iommu group 1 [ 14.096635] pci 0000:00:02.0: Adding to iommu group 2Can you please try below changes?
drivers/iommu/intel/iommu.c +0
-0@@ -2705,7 +2705,9 @@ static int __init si_domain_init(int hw) end >> agaw_to_width(si_domain->agaw)))continue ;
Quoted 11 lines
Quoted 8 lines
drivers/iommu/intel/iommu.c +3
-1- ret =iommu_domain_identity_map (si_domain, start, end); + ret =iommu_domain_identity_map (si_domain, +mm_to_dma_pfn (start >>PAGE_SHIFT ), +mm_to_dma_pfn (end >>PAGE_SHIFT ));if (ret)return ret;}Yes, this resolves it. Used memory after boot returns to similar numbers as without this commit. Thanks,
Thank you! I've posted the fix patch.
Best regards,
baolu
_______________________________________________
iommu mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/iommu