drm/gpusvm, drm/pagemap, drm/xe: Restructure migration in preparation for multi-device
Subsystems (5)
Thomas Hellström
maintainerv3 cover drm/gpusvm, drm/pagemap, drm/xe: Restructure migration in preparation for multi-device
This patchset modifies the migration part of drm_gpusvm to drm_pagemap and adds a populate_mm() op to drm_pagemap.
The idea is that the device that receives a pagefault determines if it wants to migrate content and to where. It then calls the populate_mm() method of relevant drm_pagemap.
This functionality was mostly already in place, but hard-coded for xe only without going through a pagemap op. Since we might be dealing with separate devices moving forward, it also now becomes the responsibilit of the populate_mm() op to grab any necessary local device runtime pm references and keep them held while its pages are present in an mm (struct mm_struct).
On thing to decide here is whether the populate_mm() callback should sit on a struct drm_pagemap for now while we sort multi-device usability out or whether we should add it (or something equivalent) to struct dev_pagemap.
v2:
- Rebase.
v3:
- Documentation updates (CI, Matt Brost)
- Don't change TTM buffer object type for VRAM allocations (Matt Brost)
Matthew Brost (1):
drm/gpusvm, drm/pagemap: Move migration functionality to drm_pagemap
Thomas Hellström (2):
drm/pagemap: Add a populate_mm op
drm/xe: Implement and use the drm_pagemap populate_mm op
Documentation/gpu/rfc/gpusvm.rst | 12 +- drivers/gpu/drm/Makefile | 6 +- drivers/gpu/drm/drm_gpusvm.c | 760 +----------------------- drivers/gpu/drm/drm_pagemap.c | 832 +++++++++++++++++++++++++++ drivers/gpu/drm/xe/Kconfig | 10 +- drivers/gpu/drm/xe/xe_bo_types.h | 2 +- drivers/gpu/drm/xe/xe_device_types.h | 2 +- drivers/gpu/drm/xe/xe_svm.c | 124 ++-- drivers/gpu/drm/xe/xe_svm.h | 10 +- drivers/gpu/drm/xe/xe_tile.h | 11 + drivers/gpu/drm/xe/xe_vm.c | 2 +- include/drm/drm_gpusvm.h | 96 ---- include/drm/drm_pagemap.h | 135 +++++ 13 files changed, 1090 insertions(+), 912 deletions(-) create mode 100644 drivers/gpu/drm/drm_pagemap.c
Signature · 1 line
2.49.0
Thomas Hellström
maintainerPATCH v3 1/3 drm/gpusvm, drm/pagemap: Move migration functionality to drm_pagemap
From: Matthew Brost <[email protected]>
The migration functionality and track-keeping of per-pagemap VRAM mapped to the CPU mm is not per GPU_vm, but rather per pagemap. This is also reflected by the functions not needing the drm_gpusvm structures. So move to drm_pagemap.
With this, drm_gpusvm shouldn't really access the page zone-device-data since its meaning is internal to drm_pagemap. Currently it's used to reject mapping ranges backed by multiple drm_pagemap allocations. For now, make the zone-device-data a void pointer.
Alter the interface of drm_gpusvm_migrate_to_devmem() to ensure we don't pass a gpusvm pointer.
Rename CONFIG_DRM_XE_DEVMEM_MIRROR to CONFIG_DRM_XE_PAGEMAP.
Matt is listed as author of this commit since he wrote most of the code, and it makes sense to retain his git authorship.
Thomas mostly moved the code around.
v3:
- Kerneldoc fixes (CI)
- Don't update documentation about how the drm_pagemap migration should be interpreted until upcoming patches where the functionality is implemented. (Matt Brost)
--- Documentation/gpu/rfc/gpusvm.rst | 12 +- drivers/gpu/drm/Makefile | 6 +- drivers/gpu/drm/drm_gpusvm.c | 759 +------------------------- drivers/gpu/drm/drm_pagemap.c | 788 +++++++++++++++++++++++++++ drivers/gpu/drm/xe/Kconfig | 10 +- drivers/gpu/drm/xe/xe_bo_types.h | 2 +- drivers/gpu/drm/xe/xe_device_types.h | 2 +- drivers/gpu/drm/xe/xe_svm.c | 47 +- include/drm/drm_gpusvm.h | 96 ---- include/drm/drm_pagemap.h | 101 ++++ 10 files changed, 950 insertions(+), 873 deletions(-) create mode 100644 drivers/gpu/drm/drm_pagemap.c
Documentation/gpu/rfc/gpusvm.rst +9 -3
@@ -73,15 +73,21 @@ Overview of baseline design .. kernel-doc:: drivers/gpu/drm/drm_gpusvm.c :doc: Locking -.. kernel-doc:: drivers/gpu/drm/drm_gpusvm.c - :doc: Migration - .. kernel-doc:: drivers/gpu/drm/drm_gpusvm.c :doc: Partial Unmapping of Ranges .. kernel-doc:: drivers/gpu/drm/drm_gpusvm.c :doc: Examples +Overview of drm_pagemap design +============================== + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Overview + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Migration + Possible future design features ===============================
drivers/gpu/drm/Makefile +5 -1
@@ -104,7 +104,11 @@ obj-$(CONFIG_DRM_PANEL_BACKLIGHT_QUIRKS) += drm_panel_backlight_quirks.o # obj-$(CONFIG_DRM_EXEC) += drm_exec.o obj-$(CONFIG_DRM_GPUVM) += drm_gpuvm.o -obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm.o + +drm_gpusvm_helper-y := \ + drm_gpusvm.o\ + drm_pagemap.o +obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm_helper.o obj-$(CONFIG_DRM_BUDDY) += drm_buddy.o
drivers/gpu/drm/drm_gpusvm.c +12 -747
841 lines not on this page · raw message
drivers/gpu/drm/drm_pagemap.c new file +788 -0
788 lines not on this page · raw message
drivers/gpu/drm/xe/Kconfig +6 -4
@@ -86,14 +86,16 @@ config DRM_XE_GPUSVM If in doubut say "Y". -config DRM_XE_DEVMEM_MIRROR - bool "Enable device memory mirror" +config DRM_XE_PAGEMAP + bool "Enable device memory pool for SVM" depends on DRM_XE_GPUSVM select GET_FREE_REGION default y help - Disable this option only if you want to compile out without device - memory mirror. Will reduce KMD memory footprint when disabled. + Disable this option only if you don't want to expose local device + memory for SVM. Will reduce KMD memory footprint when disabled. + + If in doubut say "Y". config DRM_XE_FORCE_PROBE string "Force probe xe for selected Intel hardware IDs"
drivers/gpu/drm/xe/xe_bo_types.h +1 -1
@@ -86,7 +86,7 @@ struct xe_bo {u16 cpu_caching;/** @devmem_allocation: SVM device memory allocation */ -struct drm_gpusvm_devmem devmem_allocation; +struct drm_pagemap_devmem devmem_allocation;/** @vram_userfault_link: Link into @mem_access.vram_userfault.list */ struct list_head vram_userfault_link;
drivers/gpu/drm/xe/xe_device_types.h +1 -1
@@ -105,7 +105,7 @@ struct xe_vram_region {void __iomem *mapping;/** @ttm: VRAM TTM manager */ struct xe_ttm_vram_mgr ttm; -#if IS_ENABLED (CONFIG_DRM_XE_DEVMEM_MIRROR ) +#if IS_ENABLED (CONFIG_DRM_XE_PAGEMAP )/** @pagemap: Used to remap device memory as ZONE_DEVICE */ struct dev_pagemap pagemap;/**
drivers/gpu/drm/xe/xe_svm.c +27 -20
126 lines not on this page · raw message
include/drm/drm_gpusvm.h +0 -96
117 lines not on this page · raw message
include/drm/drm_pagemap.h +101 -0
111 lines not on this page · raw message
Signature · 1 line
2.49.0
Ghimiray, Himal Prasad
reviewerRe: [PATCH v3 1/3] drm/gpusvm, drm/pagemap: Move migration functionality to drm_pagemap
Quoted 130 lines · Thomas Hellström
From: Matthew Brost <[email protected]>
The migration functionality and track-keeping of per-pagemap VRAM mapped to the CPU mm is not per
GPU_vm, but rather per pagemap. This is also reflected by the functions not needing thedrm_gpusvmstructures. So move todrm_pagemap.With this,
drm_gpusvmshouldn't really access the page zone-device-data since its meaning is internal todrm_pagemap. Currently it's used to reject mapping ranges backed by multipledrm_pagemapallocations. For now, make the zone-device-data a void pointer.Alter the interface of
drm_gpusvm_migrate_to_devmem()to ensure we don't pass a gpusvm pointer.Rename
CONFIG_DRM_XE_DEVMEM_MIRRORtoCONFIG_DRM_XE_PAGEMAP.Matt is listed as author of this commit since he wrote most of the code, and it makes sense to retain his git authorship.
Thomas mostly moved the code around.v3:
- Kerneldoc fixes (CI)
- Don't update documentation about how thedrm_pagemapmigration should be interpreted until upcoming patches where the functionality is implemented. (Matt Brost)Co-developed-by Thomas Hellström <[email protected]>Signed-off-by Thomas Hellström <[email protected]>--- Documentation/gpu/rfc/gpusvm.rst | 12 +- drivers/gpu/drm/Makefile | 6 +- drivers/gpu/drm/drm_gpusvm.c | 759 +------------------------- drivers/gpu/drm/drm_pagemap.c | 788 +++++++++++++++++++++++++++ drivers/gpu/drm/xe/Kconfig | 10 +- drivers/gpu/drm/xe/xe_bo_types.h | 2 +- drivers/gpu/drm/xe/xe_device_types.h | 2 +- drivers/gpu/drm/xe/xe_svm.c | 47 +- include/drm/drm_gpusvm.h | 96 ---- include/drm/drm_pagemap.h | 101 ++++ 10 files changed, 950 insertions(+), 873 deletions(-)create mode 100644
drivers/gpu/drm/drm_pagemap.cDocumentation/gpu/rfc/gpusvm.rst +9
-3@@ -73,15 +73,21 @@ Overview of baseline design .. kernel-doc:: drivers/gpu/drm/drm_gpusvm.c :doc: Locking -.. kernel-doc:: drivers/gpu/drm/drm_gpusvm.c - :doc: Migration - .. kernel-doc:: drivers/gpu/drm/drm_gpusvm.c :doc: Partial Unmapping of Ranges .. kernel-doc:: drivers/gpu/drm/drm_gpusvm.c :doc: Examples +Overview of drm_pagemap design +============================== + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Overview + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Migration + Possible future design features ===============================drivers/gpu/drm/Makefile +5
-1@@ -104,7 +104,11 @@ obj-$(CONFIG_DRM_PANEL_BACKLIGHT_QUIRKS) += drm_panel_backlight_quirks.o # obj-$(CONFIG_DRM_EXEC) += drm_exec.o obj-$(CONFIG_DRM_GPUVM) += drm_gpuvm.o -obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm.o + +drm_gpusvm_helper-y := \ + drm_gpusvm.o\ + drm_pagemap.o +obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm_helper.o obj-$(CONFIG_DRM_BUDDY) += drm_buddy.odrivers/gpu/drm/drm_gpusvm.c +1
-17@@ -8,10 +8,9 @@#include <linux/dma-mapping.h> #include <linux/hmm.h> +#include <linux/hugetlb_inline.h> #include <linux/memremap.h> -#include <linux/migrate.h> #include <linux/mm_types.h> -#include <linux/pagemap.h> #include <linux/slab.h> #include <drm/drm_device.h> @@ -107,21 +106,6 @@* to add annotations to GPU SVM. */ -/** -* DOC: Migration -* -* The migration support is quite simple, allowing migration between RAM and -* device memory at the range granularity. For example, GPU SVM currently does -* not support mixing RAM and device memory pages within a range. This means -* that upon GPU fault, the entire range can be migrated to device memory, and -* upon CPU fault, the entire range is migrated to RAM. Mixed RAM and device -* memory storage within a range could be added in the future if required. -* -* The reasoning for only supporting range granularity is as follows: it -* simplifies the implementation, and range sizes are driver-defined and should -* be relatively small. -*/ -/** * DOC: Partial Unmapping of Ranges *
drivers/gpu/drm/drm_gpusvm.c line 181
Quoted 9 lines
drivers/gpu/drm/drm_gpusvm.c +2
-4@@ -193,10 +177,9 @@* if (driver_migration_policy(range)) { * mmap_read_lock(mm); * devmem = driver_alloc_devmem(); -* err = drm_gpusvm_migrate_to_devmem(gpusvm, range, -* devmem_allocation, -* &ctx); -* mmap_read_unlock(mm); +* err = drm_pagemap_migrate_to_devmem(devmem, gpusvm->mm, gpuva_start, +* gpuva_end, driver_pgmap_owner());
fix doc passing timeslice as parameter.
Quoted 1,082 lines
drivers/gpu/drm/drm_gpusvm.c +9
-726+* mmap_read_unlock(mm); * if (err) // CPU mappings may have changed * goto retry; * } @@ -288,97 +271,6 @@ npages_in_range(unsigned long start, unsigned long end)return (end - start) >>PAGE_SHIFT ; } -/** -* struct drm_gpusvm_zdd - GPU SVM zone device data -* -* @refcount: Reference count for the zdd -* @devmem_allocation: device memory allocation -* @device_private_page_owner: Device private pages owner -* -* This structure serves as a generic wrapper installed in -* page->zone_device_data. It provides infrastructure for looking up a device -* memory allocation upon CPU page fault and asynchronously releasing device -* memory once the CPU has no page references. Asynchronous release is useful -* because CPU page references can be dropped in IRQ contexts, while releasing -* device memory likely requires sleeping locks. -*/ -struct drm_gpusvm_zdd { -struct kref refcount; -struct drm_gpusvm_devmem *devmem_allocation; -void *device_private_page_owner; -}; - -/** -* drm_gpusvm_zdd_alloc() - Allocate a zdd structure. -* @device_private_page_owner: Device private pages owner -* -* This function allocates and initializes a new zdd structure. It sets up the -* reference count and initializes the destroy work. -* -* Return: Pointer to the allocated zdd on success, ERR_PTR() on failure. -*/ -static struct drm_gpusvm_zdd * -drm_gpusvm_zdd_alloc (void *device_private_page_owner) -{ -struct drm_gpusvm_zdd *zdd; - - zdd =kmalloc (sizeof (*zdd),GFP_KERNEL ); -if (!zdd) -return NULL ; - -kref_init (&zdd->refcount); - zdd->devmem_allocation =NULL ; - zdd->device_private_page_owner = device_private_page_owner; - -return zdd; -} - -/** -* drm_gpusvm_zdd_get() - Get a reference to a zdd structure. -* @zdd: Pointer to the zdd structure. -* -* This function increments the reference count of the provided zdd structure. -* -* Return: Pointer to the zdd structure. -*/ -static struct drm_gpusvm_zdd *drm_gpusvm_zdd_get (struct drm_gpusvm_zdd *zdd) -{ -kref_get (&zdd->refcount); -return zdd; -} - -/** -* drm_gpusvm_zdd_destroy() - Destroy a zdd structure. -* @ref: Pointer to the reference count structure. -* -* This function queues the destroy_work of the zdd for asynchronous destruction. -*/ -static void drm_gpusvm_zdd_destroy (struct kref *ref) -{ -struct drm_gpusvm_zdd *zdd = - container_of(ref, struct drm_gpusvm_zdd, refcount); -struct drm_gpusvm_devmem *devmem = zdd->devmem_allocation; - -if (devmem) { -complete_all (&devmem->detached); -if (devmem->ops->devmem_release) - devmem->ops->devmem_release (devmem); - } -kfree (zdd); -} - -/** -* drm_gpusvm_zdd_put() - Put a zdd reference. -* @zdd: Pointer to the zdd structure. -* -* This function decrements the reference count of the provided zdd structure -* and schedules its destruction if the count drops to zero. -*/ -static void drm_gpusvm_zdd_put (struct drm_gpusvm_zdd *zdd) -{ -kref_put (&zdd->refcount, drm_gpusvm_zdd_destroy); -} -/** * drm_gpusvm_range_find() - Find GPU SVM range from GPU SVM notifier * @notifier: Pointer to the GPU SVM notifier structure. @@ -945,7 +837,7 @@ drm_gpusvm_range_chunk_size(struct drm_gpusvm *gpusvm,* process-many-malloc' fails. In the failure case, each process * mallocs 16k but the CPU VMA is ~128k which results in 64k SVM * ranges. When migrating the SVM ranges, some processes fail in -* drm_gpusvm_migrate_to_devmem with 'migrate.cpages != npages' +* drm_pagemap_migrate_to_devmem with 'migrate.cpages != npages' * and then upon drm_gpusvm_range_get_pages device pages from * other processes are collected + faulted in which creates all * sorts of problems. Unsure exactly how this happening, also @@ -1363,7 +1255,7 @@ int drm_gpusvm_range_get_pages(struct drm_gpusvm *gpusvm, .dev_private_owner = gpusvm->device_private_page_owner, };struct mm_struct *mm = gpusvm->mm; -struct drm_gpusvm_zdd *zdd; +void *zdd;unsigned long timeout = jiffies +msecs_to_jiffies (HMM_RANGE_DEFAULT_TIMEOUT );unsigned long i, j;@@ -1465,7 +1357,7 @@ int drm_gpusvm_range_get_pages(struct drm_gpusvm *gpusvm, } pagemap =page_pgmap (page); - dpagemap = zdd->devmem_allocation->dpagemap; + dpagemap =drm_pagemap_page_to_dpagemap (page);if (drm_WARN_ON (gpusvm->drm, !dpagemap)) {/* * Raced. This is not supposed to happen @@ -1489,7 +1381,7 @@ int drm_gpusvm_range_get_pages(struct drm_gpusvm *gpusvm, }else {dma_addr_t addr; -if (is_zone_device_page (page) || zdd) { +if (is_zone_device_page (page) || pagemap) { err = -EOPNOTSUPP ; goto err_unmap; }@@ -1517,7 +1409,7 @@ int drm_gpusvm_range_get_pages(struct drm_gpusvm *gpusvm, flags.has_dma_mapping = true; } -if (zdd) { +if (pagemap) { flags.has_devmem_pages = true; range->dpagemap = dpagemap; }@@ -1545,6 +1437,7 @@ EXPORT_SYMBOL_GPL(drm_gpusvm_range_get_pages);/** * drm_gpusvm_range_unmap_pages() - Unmap pages associated with a GPU SVM range +* drm_gpusvm_range_evict() - Evict GPU SVM range * @gpusvm: Pointer to the GPU SVM structure * @range: Pointer to the GPU SVM range structure * @ctx: GPU SVM context @@ -1575,562 +1468,11 @@ void drm_gpusvm_range_unmap_pages(struct drm_gpusvm *gpusvm,EXPORT_SYMBOL_GPL (drm_gpusvm_range_unmap_pages);/** -* drm_gpusvm_migration_unlock_put_page() - Put a migration page -* @page: Pointer to the page to put -* -* This function unlocks and puts a page. -*/ -static void drm_gpusvm_migration_unlock_put_page (struct page *page) -{ -unlock_page (page); -put_page (page); -} - -/** -* drm_gpusvm_migration_unlock_put_pages() - Put migration pages -* @npages: Number of pages -* @migrate_pfn: Array of migrate page frame numbers -* -* This function unlocks and puts an array of pages. -*/ -static void drm_gpusvm_migration_unlock_put_pages (unsigned long npages, -unsigned long *migrate_pfn) -{ -unsigned long i; - -for (i =0 ; i < npages; ++i) { -struct page *page; - -if (!migrate_pfn[i]) -continue ; - - page =migrate_pfn_to_page (migrate_pfn[i]); -drm_gpusvm_migration_unlock_put_page (page); - migrate_pfn[i] =0 ; - } -} - -/** -* drm_gpusvm_get_devmem_page() - Get a reference to a device memory page -* @page: Pointer to the page -* @zdd: Pointer to the GPU SVM zone device data -* -* This function associates the given page with the specified GPU SVM zone -* device data and initializes it for zone device usage. -*/ -static void drm_gpusvm_get_devmem_page (struct page *page, -struct drm_gpusvm_zdd *zdd) -{ - page->zone_device_data =drm_gpusvm_zdd_get (zdd); -zone_device_page_init (page); -} - -/** -* drm_gpusvm_migrate_map_pages() - Map migration pages for GPU SVM migration -* @dev: The device for which the pages are being mapped -* @dma_addr: Array to store DMA addresses corresponding to mapped pages -* @migrate_pfn: Array of migrate page frame numbers to map -* @npages: Number of pages to map -* @dir: Direction of data transfer (e.g., DMA_BIDIRECTIONAL) -* -* This function maps pages of memory for migration usage in GPU SVM. It -* iterates over each page frame number provided in @migrate_pfn, maps the -* corresponding page, and stores the DMA address in the provided @dma_addr -* array. -* -* Return: 0 on success, -EFAULT if an error occurs during mapping. -*/ -static int drm_gpusvm_migrate_map_pages (struct device *dev, -dma_addr_t *dma_addr, -unsigned long *migrate_pfn, -unsigned long npages, -enum dma_data_direction dir) -{ -unsigned long i; - -for (i =0 ; i < npages; ++i) { -struct page *page =migrate_pfn_to_page (migrate_pfn[i]); - -if (!page) -continue ; - -if (WARN_ON_ONCE (is_zone_device_page (page))) -return -EFAULT ; - - dma_addr[i] =dma_map_page (dev, page,0 ,PAGE_SIZE , dir); -if (dma_mapping_error (dev, dma_addr[i])) -return -EFAULT ; - } - -return 0 ; -} - -/** -* drm_gpusvm_migrate_unmap_pages() - Unmap pages previously mapped for GPU SVM migration -* @dev: The device for which the pages were mapped -* @dma_addr: Array of DMA addresses corresponding to mapped pages -* @npages: Number of pages to unmap -* @dir: Direction of data transfer (e.g., DMA_BIDIRECTIONAL) -* -* This function unmaps previously mapped pages of memory for GPU Shared Virtual -* Memory (SVM). It iterates over each DMA address provided in @dma_addr, checks -* if it's valid and not already unmapped, and unmaps the corresponding page. -*/ -static void drm_gpusvm_migrate_unmap_pages (struct device *dev, -dma_addr_t *dma_addr, -unsigned long npages, -enum dma_data_direction dir) -{ -unsigned long i; - -for (i =0 ; i < npages; ++i) { -if (!dma_addr[i] ||dma_mapping_error (dev, dma_addr[i])) -continue ; - -dma_unmap_page (dev, dma_addr[i],PAGE_SIZE , dir); - } -} - -/** -* drm_gpusvm_migrate_to_devmem() - Migrate GPU SVM range to device memory +* drm_gpusvm_range_evict() - Evict GPU SVM range * @gpusvm: Pointer to the GPU SVM structure -* @range: Pointer to the GPU SVM range structure -* @devmem_allocation: Pointer to the device memory allocation. The caller -* should hold a reference to the device memory allocation, -* which should be dropped via ops->devmem_release or upon -* the failure of this function. -* @ctx: GPU SVM context -* -* This function migrates the specified GPU SVM range to device memory. It -* performs the necessary setup and invokes the driver-specific operations for -* migration to device memory. Upon successful return, @devmem_allocation can -* safely reference @range until ops->devmem_release is called which only upon -* successful return. Expected to be called while holding the mmap lock in read -* mode. -* -* Return: 0 on success, negative error code on failure. -*/ -int drm_gpusvm_migrate_to_devmem (struct drm_gpusvm *gpusvm, -struct drm_gpusvm_range *range, -struct drm_gpusvm_devmem *devmem_allocation, -const struct drm_gpusvm_ctx *ctx) -{ -const struct drm_gpusvm_devmem_ops *ops = devmem_allocation->ops; -unsigned long start =drm_gpusvm_range_start (range), - end =drm_gpusvm_range_end (range); -struct migrate_vma migrate = { - .start = start, - .end = end, - .pgmap_owner = gpusvm->device_private_page_owner, - .flags =MIGRATE_VMA_SELECT_SYSTEM , - }; -struct mm_struct *mm = gpusvm->mm; -unsigned long i, npages =npages_in_range (start, end); -struct vm_area_struct *vas; -struct drm_gpusvm_zdd *zdd =NULL ; -struct page **pages; -dma_addr_t *dma_addr; -void *buf; -int err; - -mmap_assert_locked (gpusvm->mm); - -if (!range->flags.migrate_devmem) -return -EINVAL ; - -if (!ops->populate_devmem_pfn || !ops->copy_to_devmem || - !ops->copy_to_ram) -return -EOPNOTSUPP ; - - vas =vma_lookup (mm, start); -if (!vas) { - err = -ENOENT ; - goto err_out; - } - -if (end > vas->vm_end || start < vas->vm_start) { - err = -EINVAL ; - goto err_out; - } - -if (!vma_is_anonymous (vas)) { - err = -EBUSY ; - goto err_out; - } - - buf =kvcalloc (npages,2 *sizeof (*migrate.src) +sizeof (*dma_addr) + -sizeof (*pages),GFP_KERNEL ); -if (!buf) { - err = -ENOMEM ; - goto err_out; - } - dma_addr = buf + (2 *sizeof (*migrate.src) * npages); - pages = buf + (2 *sizeof (*migrate.src) +sizeof (*dma_addr)) * npages; - - zdd =drm_gpusvm_zdd_alloc (gpusvm->device_private_page_owner); -if (!zdd) { - err = -ENOMEM ; - goto err_free; - } - - migrate.vma = vas; - migrate.src = buf; - migrate.dst = migrate.src + npages; - - err =migrate_vma_setup (&migrate); -if (err) - goto err_free; - -if (!migrate.cpages) { - err = -EFAULT ; - goto err_free; - } - -if (migrate.cpages != npages) { - err = -EBUSY ; - goto err_finalize; - } - - err = ops->populate_devmem_pfn (devmem_allocation, npages, migrate.dst); -if (err) - goto err_finalize; - - err =drm_gpusvm_migrate_map_pages (devmem_allocation->dev, dma_addr, - migrate.src, npages,DMA_TO_DEVICE ); -if (err) - goto err_finalize; - -for (i =0 ; i < npages; ++i) { -struct page *page =pfn_to_page (migrate.dst[i]); - - pages[i] = page; - migrate.dst[i] =migrate_pfn (migrate.dst[i]); -drm_gpusvm_get_devmem_page (page, zdd); - } - - err = ops->copy_to_devmem (pages, dma_addr, npages); -if (err) - goto err_finalize; - -/* Upon success bind devmem allocation to range and zdd */ - devmem_allocation->timeslice_expiration =get_jiffies_64 () + -msecs_to_jiffies (ctx->timeslice_ms); - zdd->devmem_allocation = devmem_allocation;/* Owns ref */ - -err_finalize: -if (err) -drm_gpusvm_migration_unlock_put_pages (npages, migrate.dst); -migrate_vma_pages (&migrate); -migrate_vma_finalize (&migrate); -drm_gpusvm_migrate_unmap_pages (devmem_allocation->dev, dma_addr, npages, -DMA_TO_DEVICE ); -err_free: -if (zdd) -drm_gpusvm_zdd_put (zdd); -kvfree (buf); -err_out: -return err; -} -EXPORT_SYMBOL_GPL (drm_gpusvm_migrate_to_devmem); - -/** -* drm_gpusvm_migrate_populate_ram_pfn() - Populate RAM PFNs for a VM area -* @vas: Pointer to the VM area structure, can be NULL -* @fault_page: Fault page -* @npages: Number of pages to populate -* @mpages: Number of pages to migrate -* @src_mpfn: Source array of migrate PFNs -* @mpfn: Array of migrate PFNs to populate -* @addr: Start address for PFN allocation -* -* This function populates the RAM migrate page frame numbers (PFNs) for the -* specified VM area structure. It allocates and locks pages in the VM area for -* RAM usage. If vas is non-NULL use alloc_page_vma for allocation, if NULL use -* alloc_page for allocation. -* -* Return: 0 on success, negative error code on failure. -*/ -static int drm_gpusvm_migrate_populate_ram_pfn (struct vm_area_struct *vas, -struct page *fault_page, -unsigned long npages, -unsigned long *mpages, -unsigned long *src_mpfn, -unsigned long *mpfn, -unsigned long addr) -{ -unsigned long i; - -for (i =0 ; i < npages; ++i, addr +=PAGE_SIZE ) { -struct page *page, *src_page; - -if (!(src_mpfn[i] &MIGRATE_PFN_MIGRATE )) -continue ; - - src_page =migrate_pfn_to_page (src_mpfn[i]); -if (!src_page) -continue ; - -if (fault_page) { -if (src_page->zone_device_data != - fault_page->zone_device_data) -continue ; - } - -if (vas) - page =alloc_page_vma (GFP_HIGHUSER , vas, addr); -else - page =alloc_page (GFP_HIGHUSER ); - -if (!page) - goto free_pages; - - mpfn[i] =migrate_pfn (page_to_pfn (page)); - } - -for (i =0 ; i < npages; ++i) { -struct page *page =migrate_pfn_to_page (mpfn[i]); - -if (!page) -continue ; - -WARN_ON_ONCE (!trylock_page (page)); - ++*mpages; - } - -return 0 ; - -free_pages: -for (i =0 ; i < npages; ++i) { -struct page *page =migrate_pfn_to_page (mpfn[i]); - -if (!page) -continue ; - -put_page (page); - mpfn[i] =0 ; - } -return -ENOMEM ; -} - -/** -* drm_gpusvm_evict_to_ram() - Evict GPU SVM range to RAM -* @devmem_allocation: Pointer to the device memory allocation -* -* Similar to __drm_gpusvm_migrate_to_ram but does not require mmap lock and -* migration done via migrate_device_* functions. -* -* Return: 0 on success, negative error code on failure. -*/ -int drm_gpusvm_evict_to_ram (struct drm_gpusvm_devmem *devmem_allocation) -{ -const struct drm_gpusvm_devmem_ops *ops = devmem_allocation->ops; -unsigned long npages, mpages =0 ; -struct page **pages; -unsigned long *src, *dst; -dma_addr_t *dma_addr; -void *buf; -int i, err =0 ; -unsigned int retry_count =2 ; - - npages = devmem_allocation->size >>PAGE_SHIFT ; - -retry: -if (!mmget_not_zero (devmem_allocation->mm)) -return -EFAULT ; - - buf =kvcalloc (npages,2 *sizeof (*src) +sizeof (*dma_addr) + -sizeof (*pages),GFP_KERNEL ); -if (!buf) { - err = -ENOMEM ; - goto err_out; - } - src = buf; - dst = buf + (sizeof (*src) * npages); - dma_addr = buf + (2 *sizeof (*src) * npages); - pages = buf + (2 *sizeof (*src) +sizeof (*dma_addr)) * npages; - - err = ops->populate_devmem_pfn (devmem_allocation, npages, src); -if (err) - goto err_free; - - err =migrate_device_pfns (src, npages); -if (err) - goto err_free; - - err =drm_gpusvm_migrate_populate_ram_pfn (NULL ,NULL , npages, &mpages, - src, dst,0 ); -if (err || !mpages) - goto err_finalize; - - err =drm_gpusvm_migrate_map_pages (devmem_allocation->dev, dma_addr, - dst, npages,DMA_FROM_DEVICE ); -if (err) - goto err_finalize; - -for (i =0 ; i < npages; ++i) - pages[i] =migrate_pfn_to_page (src[i]); - - err = ops->copy_to_ram (pages, dma_addr, npages); -if (err) - goto err_finalize; - -err_finalize: -if (err) -drm_gpusvm_migration_unlock_put_pages (npages, dst); -migrate_device_pages (src, dst, npages); -migrate_device_finalize (src, dst, npages); -drm_gpusvm_migrate_unmap_pages (devmem_allocation->dev, dma_addr, npages, -DMA_FROM_DEVICE ); -err_free: -kvfree (buf); -err_out: -mmput_async (devmem_allocation->mm); - -if (completion_done (&devmem_allocation->detached)) -return 0 ; - -if (retry_count--) { -cond_resched (); - goto retry; - } - -return err ?: -EBUSY ; -} -EXPORT_SYMBOL_GPL (drm_gpusvm_evict_to_ram); - -/** -* __drm_gpusvm_migrate_to_ram() - Migrate GPU SVM range to RAM (internal) -* @vas: Pointer to the VM area structure -* @device_private_page_owner: Device private pages owner -* @page: Pointer to the page for fault handling (can be NULL) -* @fault_addr: Fault address -* @size: Size of migration -* -* This internal function performs the migration of the specified GPU SVM range -* to RAM. It sets up the migration, populates + dma maps RAM PFNs, and -* invokes the driver-specific operations for migration to RAM. -* -* Return: 0 on success, negative error code on failure. -*/ -static int __drm_gpusvm_migrate_to_ram (struct vm_area_struct *vas, -void *device_private_page_owner, -struct page *page, -unsigned long fault_addr, -unsigned long size) -{ -struct migrate_vma migrate = { - .vma = vas, - .pgmap_owner = device_private_page_owner, - .flags =MIGRATE_VMA_SELECT_DEVICE_PRIVATE | -MIGRATE_VMA_SELECT_DEVICE_COHERENT , - .fault_page = page, - }; -struct drm_gpusvm_zdd *zdd; -const struct drm_gpusvm_devmem_ops *ops; -struct device *dev =NULL ; -unsigned long npages, mpages =0 ; -struct page **pages; -dma_addr_t *dma_addr; -unsigned long start, end; -void *buf; -int i, err =0 ; - -if (page) { - zdd = page->zone_device_data; -if (time_before64 (get_jiffies_64 (), - zdd->devmem_allocation->timeslice_expiration)) -return 0 ; - } - - start =ALIGN_DOWN (fault_addr, size); - end =ALIGN (fault_addr +1 , size); - -/* Corner where VMA area struct has been partially unmapped */ -if (start < vas->vm_start) - start = vas->vm_start; -if (end > vas->vm_end) - end = vas->vm_end; - - migrate.start = start; - migrate.end = end; - npages =npages_in_range (start, end); - - buf =kvcalloc (npages,2 *sizeof (*migrate.src) +sizeof (*dma_addr) + -sizeof (*pages),GFP_KERNEL ); -if (!buf) { - err = -ENOMEM ; - goto err_out; - } - dma_addr = buf + (2 *sizeof (*migrate.src) * npages); - pages = buf + (2 *sizeof (*migrate.src) +sizeof (*dma_addr)) * npages; - - migrate.vma = vas; - migrate.src = buf; - migrate.dst = migrate.src + npages; - - err =migrate_vma_setup (&migrate); -if (err) - goto err_free; - -/* Raced with another CPU fault, nothing to do */ -if (!migrate.cpages) - goto err_free; - -if (!page) { -for (i =0 ; i < npages; ++i) { -if (!(migrate.src[i] &MIGRATE_PFN_MIGRATE )) -continue ; - - page =migrate_pfn_to_page (migrate.src[i]); -break ; - } - -if (!page) - goto err_finalize; - } - zdd = page->zone_device_data; - ops = zdd->devmem_allocation->ops; - dev = zdd->devmem_allocation->dev; - - err =drm_gpusvm_migrate_populate_ram_pfn (vas, page, npages, &mpages, - migrate.src, migrate.dst, - start); -if (err) - goto err_finalize; - - err =drm_gpusvm_migrate_map_pages (dev, dma_addr, migrate.dst, npages, -DMA_FROM_DEVICE ); -if (err) - goto err_finalize; - -for (i =0 ; i < npages; ++i) - pages[i] =migrate_pfn_to_page (migrate.src[i]); - - err = ops->copy_to_ram (pages, dma_addr, npages); -if (err) - goto err_finalize; - -err_finalize: -if (err) -drm_gpusvm_migration_unlock_put_pages (npages, migrate.dst); -migrate_vma_pages (&migrate); -migrate_vma_finalize (&migrate); -if (dev) -drm_gpusvm_migrate_unmap_pages (dev, dma_addr, npages, -DMA_FROM_DEVICE ); -err_free: -kvfree (buf); -err_out: - -return err; -} - -/** -* drm_gpusvm_range_evict - Evict GPU SVM range * @range: Pointer to the GPU SVM range to be removed * -* This function evicts the specified GPU SVM range. This function will not -* evict coherent pages. +* This function evicts the specified GPU SVM range. * * Return: 0 on success, a negative error code on failure. */ @@ -2182,60 +1524,6 @@ int drm_gpusvm_range_evict(struct drm_gpusvm *gpusvm, }EXPORT_SYMBOL_GPL (drm_gpusvm_range_evict); -/** -* drm_gpusvm_page_free() - Put GPU SVM zone device data associated with a page -* @page: Pointer to the page -* -* This function is a callback used to put the GPU SVM zone device data -* associated with a page when it is being released. -*/ -static void drm_gpusvm_page_free (struct page *page) -{ -drm_gpusvm_zdd_put (page->zone_device_data); -} - -/** -* drm_gpusvm_migrate_to_ram() - Migrate GPU SVM range to RAM (page fault handler) -* @vmf: Pointer to the fault information structure -* -* This function is a page fault handler used to migrate a GPU SVM range to RAM. -* It retrieves the GPU SVM range information from the faulting page and invokes -* the internal migration function to migrate the range back to RAM. -* -* Return: VM_FAULT_SIGBUS on failure, 0 on success. -*/ -static vm_fault_t drm_gpusvm_migrate_to_ram (struct vm_fault *vmf) -{ -struct drm_gpusvm_zdd *zdd = vmf->page->zone_device_data; -int err; - - err =__drm_gpusvm_migrate_to_ram (vmf->vma, - zdd->device_private_page_owner, - vmf->page, vmf->address, - zdd->devmem_allocation->size); - -return err ?VM_FAULT_SIGBUS :0 ; -} - -/* -* drm_gpusvm_pagemap_ops - Device page map operations for GPU SVM -*/ -static const struct dev_pagemap_ops drm_gpusvm_pagemap_ops = { - .page_free = drm_gpusvm_page_free, - .migrate_to_ram = drm_gpusvm_migrate_to_ram, -}; - -/** -* drm_gpusvm_pagemap_ops_get() - Retrieve GPU SVM device page map operations -* -* Return: Pointer to the GPU SVM device page map operations structure. -*/ -const struct dev_pagemap_ops *drm_gpusvm_pagemap_ops_get (void ) -{ -return &drm_gpusvm_pagemap_ops; -} -EXPORT_SYMBOL_GPL (drm_gpusvm_pagemap_ops_get); -/** * drm_gpusvm_has_mapping() - Check if GPU SVM has mapping for the given address range * @gpusvm: Pointer to the GPU SVM structure. @@ -2280,28 +1568,5 @@ void drm_gpusvm_range_set_unmapped(struct drm_gpusvm_range *range, }EXPORT_SYMBOL_GPL (drm_gpusvm_range_set_unmapped); -/** -* drm_gpusvm_devmem_init() - Initialize a GPU SVM device memory allocation -* -* @dev: Pointer to the device structure which device memory allocation belongs to -* @mm: Pointer to the mm_struct for the address space -* @ops: Pointer to the operations structure for GPU SVM device memory -* @dpagemap: The struct drm_pagemap we're allocating from. -* @size: Size of device memory allocation -*/ -void drm_gpusvm_devmem_init (struct drm_gpusvm_devmem *devmem_allocation, -struct device *dev,struct mm_struct *mm, -const struct drm_gpusvm_devmem_ops *ops, -struct drm_pagemap *dpagemap,size_t size) -{ -init_completion (&devmem_allocation->detached); - devmem_allocation->dev = dev; - devmem_allocation->mm = mm; - devmem_allocation->ops = ops; - devmem_allocation->dpagemap = dpagemap; - devmem_allocation->size = size; -} -EXPORT_SYMBOL_GPL (drm_gpusvm_devmem_init); -MODULE_DESCRIPTION ("DRM GPUSVM" );MODULE_LICENSE ("GPL" );drivers/gpu/drm/drm_pagemap.c new file +266
-0@@ -0,0 +1,788 @@ +// SPDX-License-Identifier: GPL-2.0-only OR MIT +/* +* Copyright © 2024-2025 Intel Corporation +*/ + +#include <linux/dma-mapping.h> +#include <linux/migrate.h> +#include <linux/pagemap.h> +#include <drm/drm_pagemap.h> + +/** +* DOC: Overview +* +* The DRM pagemap layer is intended to augment the dev_pagemap functionality by +* providing a way to populate a struct mm_struct virtual range with device +* private pages and to provide helpers to abstract device memory allocations, +* to migrate memory back and forth between device memory and system RAM and +* to handle access (and in the future migration) between devices implementing +* a fast interconnect that is not necessarily visible to the rest of the +* system. +* +* Typically the DRM pagemap receives requests from one or more DRM GPU SVM +* instances to populate struct mm_struct virtual ranges with memory. +*/ + +/** +* DOC: Migration +* +* The migration support is quite simple, allowing migration between RAM and +* device memory at the range granularity. For example, GPU SVM currently does +* not support mixing RAM and device memory pages within a range. This means +* that upon GPU fault, the entire range can be migrated to device memory, and +* upon CPU fault, the entire range is migrated to RAM. Mixed RAM and device +* memory storage within a range could be added in the future if required. +* +* The reasoning for only supporting range granularity is as follows: it +* simplifies the implementation, and range sizes are driver-defined and should +* be relatively small. +* +* +* Key DRM pagemap components: +* +* - Device Memory Allocations: +* Embedded structure containing enough information for the drm_pagemap to +* migrate to / from device memory. +* +* - Device Memory Operations: +* Define the interface for driver-specific device memory operations +* release memory, populate pfns, and copy to / from device memory. +*/ + +/** +* struct drm_pagemap_zdd - GPU SVM zone device data +* +* @refcount: Reference count for the zdd +* @devmem_allocation: device memory allocation +* @device_private_page_owner: Device private pages owner +* +* This structure serves as a generic wrapper installed in +* page->zone_device_data. It provides infrastructure for looking up a device +* memory allocation upon CPU page fault and asynchronously releasing device +* memory once the CPU has no page references. Asynchronous release is useful +* because CPU page references can be dropped in IRQ contexts, while releasing +* device memory likely requires sleeping locks. +*/ +struct drm_pagemap_zdd { +struct kref refcount; +struct drm_pagemap_devmem *devmem_allocation; +void *device_private_page_owner; +}; + +/** +* drm_pagemap_zdd_alloc() - Allocate a zdd structure. +* @device_private_page_owner: Device private pages owner +* +* This function allocates and initializes a new zdd structure. It sets up the +* reference count and initializes the destroy work. +* +* Return: Pointer to the allocated zdd on success, ERR_PTR() on failure. +*/ +static struct drm_pagemap_zdd * +drm_pagemap_zdd_alloc (void *device_private_page_owner) +{ +struct drm_pagemap_zdd *zdd; + + zdd =kmalloc (sizeof (*zdd),GFP_KERNEL ); +if (!zdd) +return NULL ; + +kref_init (&zdd->refcount); + zdd->devmem_allocation =NULL ; + zdd->device_private_page_owner = device_private_page_owner; + +return zdd; +} + +/** +* drm_pagemap_zdd_get() - Get a reference to a zdd structure. +* @zdd: Pointer to the zdd structure. +* +* This function increments the reference count of the provided zdd structure. +* +* Return: Pointer to the zdd structure. +*/ +static struct drm_pagemap_zdd *drm_pagemap_zdd_get (struct drm_pagemap_zdd *zdd) +{ +kref_get (&zdd->refcount); +return zdd; +} + +/** +* drm_pagemap_zdd_destroy() - Destroy a zdd structure. +* @ref: Pointer to the reference count structure. +* +* This function queues the destroy_work of the zdd for asynchronous destruction. +*/ +static void drm_pagemap_zdd_destroy (struct kref *ref) +{ +struct drm_pagemap_zdd *zdd = + container_of(ref, struct drm_pagemap_zdd, refcount); +struct drm_pagemap_devmem *devmem = zdd->devmem_allocation; + +if (devmem) { +complete_all (&devmem->detached); +if (devmem->ops->devmem_release) + devmem->ops->devmem_release (devmem); + } +kfree (zdd); +} + +/** +* drm_pagemap_zdd_put() - Put a zdd reference. +* @zdd: Pointer to the zdd structure. +* +* This function decrements the reference count of the provided zdd structure +* and schedules its destruction if the count drops to zero. +*/ +static void drm_pagemap_zdd_put (struct drm_pagemap_zdd *zdd) +{ +kref_put (&zdd->refcount, drm_pagemap_zdd_destroy); +} + +/** +* drm_pagemap_migration_unlock_put_page() - Put a migration page +* @page: Pointer to the page to put +* +* This function unlocks and puts a page. +*/ +static void drm_pagemap_migration_unlock_put_page (struct page *page) +{ +unlock_page (page); +put_page (page); +} + +/** +* drm_pagemap_migration_unlock_put_pages() - Put migration pages +* @npages: Number of pages +* @migrate_pfn: Array of migrate page frame numbers +* +* This function unlocks and puts an array of pages. +*/ +static void drm_pagemap_migration_unlock_put_pages (unsigned long npages, +unsigned long *migrate_pfn) +{ +unsigned long i; + +for (i =0 ; i < npages; ++i) { +struct page *page; + +if (!migrate_pfn[i]) +continue ; + + page =migrate_pfn_to_page (migrate_pfn[i]); +drm_pagemap_migration_unlock_put_page (page); + migrate_pfn[i] =0 ; + } +} + +/** +* drm_pagemap_get_devmem_page() - Get a reference to a device memory page +* @page: Pointer to the page +* @zdd: Pointer to the GPU SVM zone device data +* +* This function associates the given page with the specified GPU SVM zone +* device data and initializes it for zone device usage. +*/ +static void drm_pagemap_get_devmem_page (struct page *page, +struct drm_pagemap_zdd *zdd) +{ + page->zone_device_data =drm_pagemap_zdd_get (zdd); +zone_device_page_init (page); +} + +/** +* drm_pagemap_migrate_map_pages() - Map migration pages for GPU SVM migration +* @dev: The device for which the pages are being mapped +* @dma_addr: Array to store DMA addresses corresponding to mapped pages +* @migrate_pfn: Array of migrate page frame numbers to map +* @npages: Number of pages to map +* @dir: Direction of data transfer (e.g., DMA_BIDIRECTIONAL) +* +* This function maps pages of memory for migration usage in GPU SVM. It +* iterates over each page frame number provided in @migrate_pfn, maps the +* corresponding page, and stores the DMA address in the provided @dma_addr +* array. +* +* Returns: 0 on success, -EFAULT if an error occurs during mapping. +*/ +static int drm_pagemap_migrate_map_pages (struct device *dev, +dma_addr_t *dma_addr, +unsigned long *migrate_pfn, +unsigned long npages, +enum dma_data_direction dir) +{ +unsigned long i; + +for (i =0 ; i < npages; ++i) { +struct page *page =migrate_pfn_to_page (migrate_pfn[i]); + +if (!page) +continue ; + +if (WARN_ON_ONCE (is_zone_device_page (page))) +return -EFAULT ; + + dma_addr[i] =dma_map_page (dev, page,0 ,PAGE_SIZE , dir); +if (dma_mapping_error (dev, dma_addr[i])) +return -EFAULT ; + } + +return 0 ; +} + +/** +* drm_pagemap_migrate_unmap_pages() - Unmap pages previously mapped for GPU SVM migration +* @dev: The device for which the pages were mapped +* @dma_addr: Array of DMA addresses corresponding to mapped pages +* @npages: Number of pages to unmap +* @dir: Direction of data transfer (e.g., DMA_BIDIRECTIONAL) +* +* This function unmaps previously mapped pages of memory for GPU Shared Virtual +* Memory (SVM). It iterates over each DMA address provided in @dma_addr, checks +* if it's valid and not already unmapped, and unmaps the corresponding page. +*/ +static void drm_pagemap_migrate_unmap_pages (struct device *dev, +dma_addr_t *dma_addr, +unsigned long npages, +enum dma_data_direction dir) +{ +unsigned long i; + +for (i =0 ; i < npages; ++i) { +if (!dma_addr[i] ||dma_mapping_error (dev, dma_addr[i])) +continue ; + +dma_unmap_page (dev, dma_addr[i],PAGE_SIZE , dir); + } +} + +static unsigned long +npages_in_range (unsigned long start,unsigned long end) +{ +return (end - start) >>PAGE_SHIFT ; +} + +/**
drivers/gpu/drm/drm_pagemap.c line 276
Quoted 10 lines
drivers/gpu/drm/drm_pagemap.c new file +10
-0+* drm_pagemap_migrate_to_devmem() - Migrate a struct mm_struct range to device memory +* @devmem_allocation: The device memory allocation to migrate to. +* The caller should hold a reference to the device memory allocation, +* and the reference is consumed by this function unless it returns with +* an error. +* @mm: Pointer to the struct mm_struct. +* @start: Start of the virtual address range to migrate. +* @end: End of the virtual address range to migrate. +* @timeslice_ms: The time requested for the migrated pages to +* be present in the cpu memory map before migrated back.
Shouldn't this be present in gpu or cpu memory map ? We are using this to ensure pagefault can be handled effectively by ensuring pages remain in vram here for prescribed time too.
drivers/gpu/drm/drm_pagemap.c line 284
Quoted 8 lines
drivers/gpu/drm/drm_pagemap.c new file +8
-0+* @pgmap_owner: Not used currently, since only system memory is considered. +* +* This function migrates the specified virtual address range to device memory. +* It performs the necessary setup and invokes the driver-specific operations for +* migration to device memory. Expected to be called while holding the mmap lock in +* at least read mode. +* +* Return: %0 on success, negative error code on failure.
s/%0/0
Quoted 937 lines
drivers/gpu/drm/drm_pagemap.c new file +504
-0+*/ +int drm_pagemap_migrate_to_devmem (struct drm_pagemap_devmem *devmem_allocation, +struct mm_struct *mm, +unsigned long start,unsigned long end, +unsigned long timeslice_ms, +void *pgmap_owner) +{ +const struct drm_pagemap_devmem_ops *ops = devmem_allocation->ops; +struct migrate_vma migrate = { + .start = start, + .end = end, + .pgmap_owner = pgmap_owner, + .flags =MIGRATE_VMA_SELECT_SYSTEM , + }; +unsigned long i, npages =npages_in_range (start, end); +struct vm_area_struct *vas; +struct drm_pagemap_zdd *zdd =NULL ; +struct page **pages; +dma_addr_t *dma_addr; +void *buf; +int err; + +mmap_assert_locked (mm); + +if (!ops->populate_devmem_pfn || !ops->copy_to_devmem || + !ops->copy_to_ram) +return -EOPNOTSUPP ; + + vas =vma_lookup (mm, start); +if (!vas) { + err = -ENOENT ; + goto err_out; + } + +if (end > vas->vm_end || start < vas->vm_start) { + err = -EINVAL ; + goto err_out; + } + +if (!vma_is_anonymous (vas)) { + err = -EBUSY ; + goto err_out; + } + + buf =kvcalloc (npages,2 *sizeof (*migrate.src) +sizeof (*dma_addr) + +sizeof (*pages),GFP_KERNEL ); +if (!buf) { + err = -ENOMEM ; + goto err_out; + } + dma_addr = buf + (2 *sizeof (*migrate.src) * npages); + pages = buf + (2 *sizeof (*migrate.src) +sizeof (*dma_addr)) * npages; + + zdd =drm_pagemap_zdd_alloc (pgmap_owner); +if (!zdd) { + err = -ENOMEM ; + goto err_free; + } + + migrate.vma = vas; + migrate.src = buf; + migrate.dst = migrate.src + npages; + + err =migrate_vma_setup (&migrate); +if (err) + goto err_free; + +if (!migrate.cpages) { + err = -EFAULT ; + goto err_free; + } + +if (migrate.cpages != npages) { + err = -EBUSY ; + goto err_finalize; + } + + err = ops->populate_devmem_pfn (devmem_allocation, npages, migrate.dst); +if (err) + goto err_finalize; + + err =drm_pagemap_migrate_map_pages (devmem_allocation->dev, dma_addr, + migrate.src, npages,DMA_TO_DEVICE ); +if (err) + goto err_finalize; + +for (i =0 ; i < npages; ++i) { +struct page *page =pfn_to_page (migrate.dst[i]); + + pages[i] = page; + migrate.dst[i] =migrate_pfn (migrate.dst[i]); +drm_pagemap_get_devmem_page (page, zdd); + } + + err = ops->copy_to_devmem (pages, dma_addr, npages); +if (err) + goto err_finalize; + +/* Upon success bind devmem allocation to range and zdd */ + devmem_allocation->timeslice_expiration =get_jiffies_64 () + +msecs_to_jiffies (timeslice_ms); + zdd->devmem_allocation = devmem_allocation;/* Owns ref */ + +err_finalize: +if (err) +drm_pagemap_migration_unlock_put_pages (npages, migrate.dst); +migrate_vma_pages (&migrate); +migrate_vma_finalize (&migrate); +drm_pagemap_migrate_unmap_pages (devmem_allocation->dev, dma_addr, npages, +DMA_TO_DEVICE ); +err_free: +if (zdd) +drm_pagemap_zdd_put (zdd); +kvfree (buf); +err_out: +return err; +} +EXPORT_SYMBOL_GPL (drm_pagemap_migrate_to_devmem); + +/** +* drm_pagemap_migrate_populate_ram_pfn() - Populate RAM PFNs for a VM area +* @vas: Pointer to the VM area structure, can be NULL +* @fault_page: Fault page +* @npages: Number of pages to populate +* @mpages: Number of pages to migrate +* @src_mpfn: Source array of migrate PFNs +* @mpfn: Array of migrate PFNs to populate +* @addr: Start address for PFN allocation +* +* This function populates the RAM migrate page frame numbers (PFNs) for the +* specified VM area structure. It allocates and locks pages in the VM area for +* RAM usage. If vas is non-NULL use alloc_page_vma for allocation, if NULL use +* alloc_page for allocation. +* +* Return: 0 on success, negative error code on failure. +*/ +static int drm_pagemap_migrate_populate_ram_pfn (struct vm_area_struct *vas, +struct page *fault_page, +unsigned long npages, +unsigned long *mpages, +unsigned long *src_mpfn, +unsigned long *mpfn, +unsigned long addr) +{ +unsigned long i; + +for (i =0 ; i < npages; ++i, addr +=PAGE_SIZE ) { +struct page *page, *src_page; + +if (!(src_mpfn[i] &MIGRATE_PFN_MIGRATE )) +continue ; + + src_page =migrate_pfn_to_page (src_mpfn[i]); +if (!src_page) +continue ; + +if (fault_page) { +if (src_page->zone_device_data != + fault_page->zone_device_data) +continue ; + } + +if (vas) + page =alloc_page_vma (GFP_HIGHUSER , vas, addr); +else + page =alloc_page (GFP_HIGHUSER ); + +if (!page) + goto free_pages; + + mpfn[i] =migrate_pfn (page_to_pfn (page)); + } + +for (i =0 ; i < npages; ++i) { +struct page *page =migrate_pfn_to_page (mpfn[i]); + +if (!page) +continue ; + +WARN_ON_ONCE (!trylock_page (page)); + ++*mpages; + } + +return 0 ; + +free_pages: +for (i =0 ; i < npages; ++i) { +struct page *page =migrate_pfn_to_page (mpfn[i]); + +if (!page) +continue ; + +put_page (page); + mpfn[i] =0 ; + } +return -ENOMEM ; +} + +/** +* drm_pagemap_evict_to_ram() - Evict GPU SVM range to RAM +* @devmem_allocation: Pointer to the device memory allocation +* +* Similar to __drm_pagemap_migrate_to_ram but does not require mmap lock and +* migration done via migrate_device_* functions. +* +* Return: 0 on success, negative error code on failure. +*/ +int drm_pagemap_evict_to_ram (struct drm_pagemap_devmem *devmem_allocation) +{ +const struct drm_pagemap_devmem_ops *ops = devmem_allocation->ops; +unsigned long npages, mpages =0 ; +struct page **pages; +unsigned long *src, *dst; +dma_addr_t *dma_addr; +void *buf; +int i, err =0 ; +unsigned int retry_count =2 ; + + npages = devmem_allocation->size >>PAGE_SHIFT ; + +retry: +if (!mmget_not_zero (devmem_allocation->mm)) +return -EFAULT ; + + buf =kvcalloc (npages,2 *sizeof (*src) +sizeof (*dma_addr) + +sizeof (*pages),GFP_KERNEL ); +if (!buf) { + err = -ENOMEM ; + goto err_out; + } + src = buf; + dst = buf + (sizeof (*src) * npages); + dma_addr = buf + (2 *sizeof (*src) * npages); + pages = buf + (2 *sizeof (*src) +sizeof (*dma_addr)) * npages; + + err = ops->populate_devmem_pfn (devmem_allocation, npages, src); +if (err) + goto err_free; + + err =migrate_device_pfns (src, npages); +if (err) + goto err_free; + + err =drm_pagemap_migrate_populate_ram_pfn (NULL ,NULL , npages, &mpages, + src, dst,0 ); +if (err || !mpages) + goto err_finalize; + + err =drm_pagemap_migrate_map_pages (devmem_allocation->dev, dma_addr, + dst, npages,DMA_FROM_DEVICE ); +if (err) + goto err_finalize; + +for (i =0 ; i < npages; ++i) + pages[i] =migrate_pfn_to_page (src[i]); + + err = ops->copy_to_ram (pages, dma_addr, npages); +if (err) + goto err_finalize; + +err_finalize: +if (err) +drm_pagemap_migration_unlock_put_pages (npages, dst); +migrate_device_pages (src, dst, npages); +migrate_device_finalize (src, dst, npages); +drm_pagemap_migrate_unmap_pages (devmem_allocation->dev, dma_addr, npages, +DMA_FROM_DEVICE ); +err_free: +kvfree (buf); +err_out: +mmput_async (devmem_allocation->mm); + +if (completion_done (&devmem_allocation->detached)) +return 0 ; + +if (retry_count--) { +cond_resched (); + goto retry; + } + +return err ?: -EBUSY ; +} +EXPORT_SYMBOL_GPL (drm_pagemap_evict_to_ram); + +/** +* __drm_pagemap_migrate_to_ram() - Migrate GPU SVM range to RAM (internal) +* @vas: Pointer to the VM area structure +* @device_private_page_owner: Device private pages owner +* @page: Pointer to the page for fault handling (can be NULL) +* @fault_addr: Fault address +* @size: Size of migration +* +* This internal function performs the migration of the specified GPU SVM range +* to RAM. It sets up the migration, populates + dma maps RAM PFNs, and +* invokes the driver-specific operations for migration to RAM. +* +* Return: 0 on success, negative error code on failure. +*/ +static int __drm_pagemap_migrate_to_ram (struct vm_area_struct *vas, +void *device_private_page_owner, +struct page *page, +unsigned long fault_addr, +unsigned long size) +{ +struct migrate_vma migrate = { + .vma = vas, + .pgmap_owner = device_private_page_owner, + .flags =MIGRATE_VMA_SELECT_DEVICE_PRIVATE | +MIGRATE_VMA_SELECT_DEVICE_COHERENT , + .fault_page = page, + }; +struct drm_pagemap_zdd *zdd; +const struct drm_pagemap_devmem_ops *ops; +struct device *dev =NULL ; +unsigned long npages, mpages =0 ; +struct page **pages; +dma_addr_t *dma_addr; +unsigned long start, end; +void *buf; +int i, err =0 ; + +if (page) { + zdd = page->zone_device_data; +if (time_before64 (get_jiffies_64 (), + zdd->devmem_allocation->timeslice_expiration)) +return 0 ; + } + + start =ALIGN_DOWN (fault_addr, size); + end =ALIGN (fault_addr +1 , size); + +/* Corner where VMA area struct has been partially unmapped */ +if (start < vas->vm_start) + start = vas->vm_start; +if (end > vas->vm_end) + end = vas->vm_end; + + migrate.start = start; + migrate.end = end; + npages =npages_in_range (start, end); + + buf =kvcalloc (npages,2 *sizeof (*migrate.src) +sizeof (*dma_addr) + +sizeof (*pages),GFP_KERNEL ); +if (!buf) { + err = -ENOMEM ; + goto err_out; + } + dma_addr = buf + (2 *sizeof (*migrate.src) * npages); + pages = buf + (2 *sizeof (*migrate.src) +sizeof (*dma_addr)) * npages; + + migrate.vma = vas; + migrate.src = buf; + migrate.dst = migrate.src + npages; + + err =migrate_vma_setup (&migrate); +if (err) + goto err_free; + +/* Raced with another CPU fault, nothing to do */ +if (!migrate.cpages) + goto err_free; + +if (!page) { +for (i =0 ; i < npages; ++i) { +if (!(migrate.src[i] &MIGRATE_PFN_MIGRATE )) +continue ; + + page =migrate_pfn_to_page (migrate.src[i]); +break ; + } + +if (!page) + goto err_finalize; + } + zdd = page->zone_device_data; + ops = zdd->devmem_allocation->ops; + dev = zdd->devmem_allocation->dev; + + err =drm_pagemap_migrate_populate_ram_pfn (vas, page, npages, &mpages, + migrate.src, migrate.dst, + start); +if (err) + goto err_finalize; + + err =drm_pagemap_migrate_map_pages (dev, dma_addr, migrate.dst, npages, +DMA_FROM_DEVICE ); +if (err) + goto err_finalize; + +for (i =0 ; i < npages; ++i) + pages[i] =migrate_pfn_to_page (migrate.src[i]); + + err = ops->copy_to_ram (pages, dma_addr, npages); +if (err) + goto err_finalize; + +err_finalize: +if (err) +drm_pagemap_migration_unlock_put_pages (npages, migrate.dst); +migrate_vma_pages (&migrate); +migrate_vma_finalize (&migrate); +if (dev) +drm_pagemap_migrate_unmap_pages (dev, dma_addr, npages, +DMA_FROM_DEVICE ); +err_free: +kvfree (buf); +err_out: + +return err; +} + +/** +* drm_pagemap_page_free() - Put GPU SVM zone device data associated with a page +* @page: Pointer to the page +* +* This function is a callback used to put the GPU SVM zone device data +* associated with a page when it is being released. +*/ +static void drm_pagemap_page_free (struct page *page) +{ +drm_pagemap_zdd_put (page->zone_device_data); +} + +/** +* drm_pagemap_migrate_to_ram() - Migrate a virtual range to RAM (page fault handler) +* @vmf: Pointer to the fault information structure +* +* This function is a page fault handler used to migrate a virtual range +* to ram. The device memory allocation in which the device page is found is +* migrated in its entirety. +* +* Returns: +* VM_FAULT_SIGBUS on failure, 0 on success. +*/ +static vm_fault_t drm_pagemap_migrate_to_ram (struct vm_fault *vmf) +{ +struct drm_pagemap_zdd *zdd = vmf->page->zone_device_data; +int err; + + err =__drm_pagemap_migrate_to_ram (vmf->vma, + zdd->device_private_page_owner, + vmf->page, vmf->address, + zdd->devmem_allocation->size); + +return err ?VM_FAULT_SIGBUS :0 ; +} + +static const struct dev_pagemap_ops drm_pagemap_pagemap_ops = { + .page_free = drm_pagemap_page_free, + .migrate_to_ram = drm_pagemap_migrate_to_ram, +}; + +/** +* drm_pagemap_pagemap_ops_get() - Retrieve GPU SVM device page map operations +* +* Returns: +* Pointer to the GPU SVM device page map operations structure. +*/ +const struct dev_pagemap_ops *drm_pagemap_pagemap_ops_get (void ) +{ +return &drm_pagemap_pagemap_ops; +} +EXPORT_SYMBOL_GPL (drm_pagemap_pagemap_ops_get); + +/** +* drm_pagemap_devmem_init() - Initialize a drm_pagemap device memory allocation +* +* @devmem_allocation: The struct drm_pagemap_devmem to initialize. +* @dev: Pointer to the device structure which device memory allocation belongs to +* @mm: Pointer to the mm_struct for the address space +* @ops: Pointer to the operations structure for GPU SVM device memory +* @dpagemap: The struct drm_pagemap we're allocating from. +* @size: Size of device memory allocation +*/ +void drm_pagemap_devmem_init (struct drm_pagemap_devmem *devmem_allocation, +struct device *dev,struct mm_struct *mm, +const struct drm_pagemap_devmem_ops *ops, +struct drm_pagemap *dpagemap,size_t size) +{ +init_completion (&devmem_allocation->detached); + devmem_allocation->dev = dev; + devmem_allocation->mm = mm; + devmem_allocation->ops = ops; + devmem_allocation->dpagemap = dpagemap; + devmem_allocation->size = size; +} +EXPORT_SYMBOL_GPL (drm_pagemap_devmem_init); + +/** +* drm_pagemap_page_to_dpagemap() - Return a pointer the drm_pagemap of a page +* @page: The struct page. +* +* Return: A pointer to the struct drm_pagemap of a device private page that +* was populated from the struct drm_pagemap. If the page was *not* populated +* from a struct drm_pagemap, the result is undefined and the function call +* may result in dereferencing and invalid address. +*/ +struct drm_pagemap *drm_pagemap_page_to_dpagemap (struct page *page) +{ +struct drm_pagemap_zdd *zdd = page->zone_device_data; + +return zdd->devmem_allocation->dpagemap; +} +EXPORT_SYMBOL_GPL (drm_pagemap_page_to_dpagemap);drivers/gpu/drm/xe/Kconfig +6
-4@@ -86,14 +86,16 @@ config DRM_XE_GPUSVM If in doubut say "Y". -config DRM_XE_DEVMEM_MIRROR - bool "Enable device memory mirror" +config DRM_XE_PAGEMAP + bool "Enable device memory pool for SVM" depends on DRM_XE_GPUSVM select GET_FREE_REGION default y help - Disable this option only if you want to compile out without device - memory mirror. Will reduce KMD memory footprint when disabled. + Disable this option only if you don't want to expose local device + memory for SVM. Will reduce KMD memory footprint when disabled. + + If in doubut say "Y". config DRM_XE_FORCE_PROBE string "Force probe xe for selected Intel hardware IDs"drivers/gpu/drm/xe/xe_bo_types.h +1
-1@@ -86,7 +86,7 @@ struct xe_bo {u16 cpu_caching;/** @devmem_allocation: SVM device memory allocation */ -struct drm_gpusvm_devmem devmem_allocation; +struct drm_pagemap_devmem devmem_allocation;/** @vram_userfault_link: Link into @mem_access.vram_userfault.list */ struct list_head vram_userfault_link;drivers/gpu/drm/xe/xe_device_types.h +1
-1@@ -105,7 +105,7 @@ struct xe_vram_region {void __iomem *mapping;/** @ttm: VRAM TTM manager */ struct xe_ttm_vram_mgr ttm; -#if IS_ENABLED (CONFIG_DRM_XE_DEVMEM_MIRROR ) +#if IS_ENABLED (CONFIG_DRM_XE_PAGEMAP )/** @pagemap: Used to remap device memory as ZONE_DEVICE */ struct dev_pagemap pagemap;/** drivers/gpu/drm/xe/xe_svm.c +27
-20@@ -329,7 +329,7 @@ static void xe_svm_garbage_collector_work_func(struct work_struct *w)up_write (&vm->lock); } -#if IS_ENABLED (CONFIG_DRM_XE_DEVMEM_MIRROR ) +#if IS_ENABLED (CONFIG_DRM_XE_PAGEMAP )static struct xe_vram_region *page_to_vr (struct page *page) {@@ -517,12 +517,12 @@ static int xe_svm_copy_to_ram(struct page **pages, dma_addr_t *dma_addr,return xe_svm_copy (pages, dma_addr, npages,XE_SVM_COPY_TO_SRAM ); } -static struct xe_bo *to_xe_bo (struct drm_gpusvm_devmem *devmem_allocation) +static struct xe_bo *to_xe_bo (struct drm_pagemap_devmem *devmem_allocation) { return container_of(devmem_allocation, struct xe_bo, devmem_allocation); } -static void xe_svm_devmem_release (struct drm_gpusvm_devmem *devmem_allocation) +static void xe_svm_devmem_release (struct drm_pagemap_devmem *devmem_allocation) {struct xe_bo *bo =to_xe_bo (devmem_allocation);@@ -539,7 +539,7 @@ static struct drm_buddy *tile_to_buddy(struct xe_tile *tile)return &tile->mem.vram.ttm.mm; } -static int xe_svm_populate_devmem_pfn (struct drm_gpusvm_devmem *devmem_allocation, +static int xe_svm_populate_devmem_pfn (struct drm_pagemap_devmem *devmem_allocation,unsigned long npages,unsigned long *pfn) {struct xe_bo *bo =to_xe_bo (devmem_allocation);@@ -562,7 +562,7 @@ static int xe_svm_populate_devmem_pfn(struct drm_gpusvm_devmem *devmem_allocatioreturn 0 ; } -static const struct drm_gpusvm_devmem_ops gpusvm_devmem_ops = { +static const struct drm_pagemap_devmem_ops dpagemap_devmem_ops = { .devmem_release = xe_svm_devmem_release, .populate_devmem_pfn = xe_svm_populate_devmem_pfn, .copy_to_devmem = xe_svm_copy_to_devmem,@@ -714,7 +714,7 @@ u64 xe_svm_find_vma_start(struct xe_vm *vm, u64 start, u64 end, struct xe_vma *v min(end, xe_vma_end(vma))); } -#if IS_ENABLED (CONFIG_DRM_XE_DEVMEM_MIRROR ) +#if IS_ENABLED (CONFIG_DRM_XE_PAGEMAP )static struct xe_vram_region *tile_to_vr (struct xe_tile *tile) {return &tile->mem.vram;@@ -742,6 +742,9 @@ int xe_svm_alloc_vram(struct xe_vm *vm, struct xe_tile *tile,ktime_t end =0 ;int err; +if (!range->base.flags.migrate_devmem) +return -EINVAL ; +range_debug (range,"ALLOCATE VRAM" );if (!mmget_not_zero (mm))@@ -761,19 +764,23 @@ int xe_svm_alloc_vram(struct xe_vm *vm, struct xe_tile *tile, goto unlock; } -drm_gpusvm_devmem_init (&bo->devmem_allocation, - vm->xe->drm.dev, mm, - &gpusvm_devmem_ops, - &tile->mem.vram.dpagemap, -xe_svm_range_size (range)); +drm_pagemap_devmem_init (&bo->devmem_allocation, + vm->xe->drm.dev, mm, + &dpagemap_devmem_ops, + &tile->mem.vram.dpagemap, +xe_svm_range_size (range)); blocks = &to_xe_ttm_vram_mgr_resource (bo->ttm.resource)->blocks;list_for_each_entry (block, blocks, link) block->private = vr;xe_bo_get (bo); - err =drm_gpusvm_migrate_to_devmem (&vm->svm.gpusvm, &range->base, - &bo->devmem_allocation, ctx); + err =drm_pagemap_migrate_to_devmem (&bo->devmem_allocation, + mm, +xe_svm_range_start (range), +xe_svm_range_end (range), + ctx->timeslice_ms, +xe_svm_devm_owner (vm->xe));if (err)xe_svm_devmem_release (&bo->devmem_allocation);@@ -848,13 +855,13 @@ int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,struct drm_gpusvm_ctx ctx = { .read_only =xe_vma_read_only (vma), .devmem_possible =IS_DGFX (vm->xe) && -IS_ENABLED (CONFIG_DRM_XE_DEVMEM_MIRROR ), +IS_ENABLED (CONFIG_DRM_XE_PAGEMAP ), .check_pages_threshold =IS_DGFX (vm->xe) && -IS_ENABLED (CONFIG_DRM_XE_DEVMEM_MIRROR ) ?SZ_64K :0 , +IS_ENABLED (CONFIG_DRM_XE_PAGEMAP ) ?SZ_64K :0 , .devmem_only = atomic &&IS_DGFX (vm->xe) && -IS_ENABLED (CONFIG_DRM_XE_DEVMEM_MIRROR ), +IS_ENABLED (CONFIG_DRM_XE_PAGEMAP ), .timeslice_ms = atomic &&IS_DGFX (vm->xe) && -IS_ENABLED (CONFIG_DRM_XE_DEVMEM_MIRROR ) ? +IS_ENABLED (CONFIG_DRM_XE_PAGEMAP ) ? vm->xe->atomic_svm_timeslice_ms :0 , };struct xe_svm_range *range;@@ -992,7 +999,7 @@ bool xe_svm_has_mapping(struct xe_vm *vm, u64 start, u64 end)*/ int xe_svm_bo_evict (struct xe_bo *bo) { -return drm_gpusvm_evict_to_ram (&bo->devmem_allocation); +return drm_pagemap_evict_to_ram (&bo->devmem_allocation); }/** @@ -1045,7 +1052,7 @@ int xe_svm_range_get_pages(struct xe_vm *vm, struct xe_svm_range *range,return err; } -#if IS_ENABLED (CONFIG_DRM_XE_DEVMEM_MIRROR ) +#if IS_ENABLED (CONFIG_DRM_XE_PAGEMAP ) static struct drm_pagemap_device_addr xe_drm_pagemap_device_map(struct drm_pagemap *dpagemap,@@ -1102,7 +1109,7 @@ int xe_devm_add(struct xe_tile *tile, struct xe_vram_region *vr) vr->pagemap.range.start = res->start; vr->pagemap.range.end = res->end; vr->pagemap.nr_range =1 ; - vr->pagemap.ops =drm_gpusvm_pagemap_ops_get (); + vr->pagemap.ops =drm_pagemap_pagemap_ops_get (); vr->pagemap.owner =xe_svm_devm_owner (xe); addr =devm_memremap_pages (dev, &vr->pagemap);include/drm/drm_gpusvm.h +0
-96@@ -16,91 +16,9 @@ struct drm_gpusvm;struct drm_gpusvm_notifier ;struct drm_gpusvm_ops ;struct drm_gpusvm_range ; -struct drm_gpusvm_devmem ;struct drm_pagemap ;struct drm_pagemap_device_addr ; -/** -* struct drm_gpusvm_devmem_ops - Operations structure for GPU SVM device memory -* -* This structure defines the operations for GPU Shared Virtual Memory (SVM) -* device memory. These operations are provided by the GPU driver to manage device memory -* allocations and perform operations such as migration between device memory and system -* RAM. -*/ -struct drm_gpusvm_devmem_ops { -/** -* @devmem_release: Release device memory allocation (optional) -* @devmem_allocation: device memory allocation -* -* Release device memory allocation and drop a reference to device -* memory allocation. -*/ -void (*devmem_release)(struct drm_gpusvm_devmem *devmem_allocation); - -/** -* @populate_devmem_pfn: Populate device memory PFN (required for migration) -* @devmem_allocation: device memory allocation -* @npages: Number of pages to populate -* @pfn: Array of page frame numbers to populate -* -* Populate device memory page frame numbers (PFN). -* -* Return: 0 on success, a negative error code on failure. -*/ -int (*populate_devmem_pfn)(struct drm_gpusvm_devmem *devmem_allocation, -unsigned long npages,unsigned long *pfn); - -/** -* @copy_to_devmem: Copy to device memory (required for migration) -* @pages: Pointer to array of device memory pages (destination) -* @dma_addr: Pointer to array of DMA addresses (source) -* @npages: Number of pages to copy -* -* Copy pages to device memory. -* -* Return: 0 on success, a negative error code on failure. -*/ -int (*copy_to_devmem)(struct page **pages, -dma_addr_t *dma_addr, -unsigned long npages); - -/** -* @copy_to_ram: Copy to system RAM (required for migration) -* @pages: Pointer to array of device memory pages (source) -* @dma_addr: Pointer to array of DMA addresses (destination) -* @npages: Number of pages to copy -* -* Copy pages to system RAM. -* -* Return: 0 on success, a negative error code on failure. -*/ -int (*copy_to_ram)(struct page **pages, -dma_addr_t *dma_addr, -unsigned long npages); -}; - -/** -* struct drm_gpusvm_devmem - Structure representing a GPU SVM device memory allocation -* -* @dev: Pointer to the device structure which device memory allocation belongs to -* @mm: Pointer to the mm_struct for the address space -* @detached: device memory allocations is detached from device pages -* @ops: Pointer to the operations structure for GPU SVM device memory -* @dpagemap: The struct drm_pagemap of the pages this allocation belongs to. -* @size: Size of device memory allocation -* @timeslice_expiration: Timeslice expiration in jiffies -*/ -struct drm_gpusvm_devmem { -struct device *dev; -struct mm_struct *mm; -struct completion detached; -const struct drm_gpusvm_devmem_ops *ops; -struct drm_pagemap *dpagemap; -size_t size; -u64 timeslice_expiration; -}; -/** * struct drm_gpusvm_ops - Operations structure for GPU SVM * @@ -361,15 +279,6 @@ void drm_gpusvm_range_unmap_pages(struct drm_gpusvm *gpusvm,struct drm_gpusvm_range *range, const struct drm_gpusvm_ctx *ctx); -int drm_gpusvm_migrate_to_devmem (struct drm_gpusvm *gpusvm, -struct drm_gpusvm_range *range, -struct drm_gpusvm_devmem *devmem_allocation, -const struct drm_gpusvm_ctx *ctx); - -int drm_gpusvm_evict_to_ram (struct drm_gpusvm_devmem *devmem_allocation); - -const struct dev_pagemap_ops *drm_gpusvm_pagemap_ops_get (void ); -bool drm_gpusvm_has_mapping (struct drm_gpusvm *gpusvm,unsigned long start,unsigned long end);@@ -380,11 +289,6 @@ drm_gpusvm_range_find(struct drm_gpusvm_notifier *notifier, unsigned long start,void drm_gpusvm_range_set_unmapped (struct drm_gpusvm_range *range,const struct mmu_notifier_range *mmu_range); -void drm_gpusvm_devmem_init (struct drm_gpusvm_devmem *devmem_allocation, -struct device *dev,struct mm_struct *mm, -const struct drm_gpusvm_devmem_ops *ops, -struct drm_pagemap *dpagemap,size_t size); -#ifdef CONFIG_LOCKDEP /** * drm_gpusvm_driver_set_lock() - Set the lock protecting accesses to GPU SVM include/drm/drm_pagemap.h +101
-0@@ -7,6 +7,7 @@#include <linux/types.h> struct drm_pagemap ; +struct drm_pagemap_zdd ;struct device ;/** @@ -104,4 +105,104 @@ struct drm_pagemap {struct device *dev; }; +struct drm_pagemap_devmem ; + +/** +* struct drm_pagemap_devmem_ops - Operations structure for GPU SVM device memory +* +* This structure defines the operations for GPU Shared Virtual Memory (SVM) +* device memory. These operations are provided by the GPU driver to manage device memory +* allocations and perform operations such as migration between device memory and system +* RAM. +*/ +struct drm_pagemap_devmem_ops { +/** +* @devmem_release: Release device memory allocation (optional) +* @devmem_allocation: device memory allocation +* +* Release device memory allocation and drop a reference to device +* memory allocation. +*/ +void (*devmem_release)(struct drm_pagemap_devmem *devmem_allocation); + +/** +* @populate_devmem_pfn: Populate device memory PFN (required for migration) +* @devmem_allocation: device memory allocation +* @npages: Number of pages to populate +* @pfn: Array of page frame numbers to populate +* +* Populate device memory page frame numbers (PFN). +* +* Return: 0 on success, a negative error code on failure. +*/ +int (*populate_devmem_pfn)(struct drm_pagemap_devmem *devmem_allocation, +unsigned long npages,unsigned long *pfn); + +/** +* @copy_to_devmem: Copy to device memory (required for migration) +* @pages: Pointer to array of device memory pages (destination) +* @dma_addr: Pointer to array of DMA addresses (source) +* @npages: Number of pages to copy +* +* Copy pages to device memory. +* +* Return: 0 on success, a negative error code on failure. +*/ +int (*copy_to_devmem)(struct page **pages, +dma_addr_t *dma_addr, +unsigned long npages); + +/** +* @copy_to_ram: Copy to system RAM (required for migration) +* @pages: Pointer to array of device memory pages (source) +* @dma_addr: Pointer to array of DMA addresses (destination) +* @npages: Number of pages to copy +* +* Copy pages to system RAM. +* +* Return: 0 on success, a negative error code on failure. +*/ +int (*copy_to_ram)(struct page **pages, +dma_addr_t *dma_addr, +unsigned long npages); +}; + +/** +* struct drm_pagemap_devmem - Structure representing a GPU SVM device memory allocation +* +* @dev: Pointer to the device structure which device memory allocation belongs to +* @mm: Pointer to the mm_struct for the address space +* @detached: device memory allocations is detached from device pages +* @ops: Pointer to the operations structure for GPU SVM device memory +* @dpagemap: The struct drm_pagemap of the pages this allocation belongs to. +* @size: Size of device memory allocation +* @timeslice_expiration: Timeslice expiration in jiffies +*/ +struct drm_pagemap_devmem { +struct device *dev; +struct mm_struct *mm; +struct completion detached; +const struct drm_pagemap_devmem_ops *ops; +struct drm_pagemap *dpagemap; +size_t size; +u64 timeslice_expiration; +}; + +int drm_pagemap_migrate_to_devmem (struct drm_pagemap_devmem *devmem_allocation, +struct mm_struct *mm, +unsigned long start,unsigned long end, +unsigned long timeslice_ms, +void *pgmap_owner); + +int drm_pagemap_evict_to_ram (struct drm_pagemap_devmem *devmem_allocation); + +const struct dev_pagemap_ops *drm_pagemap_pagemap_ops_get (void ); + +struct drm_pagemap *drm_pagemap_page_to_dpagemap (struct page *page); + +void drm_pagemap_devmem_init (struct drm_pagemap_devmem *devmem_allocation, +struct device *dev,struct mm_struct *mm, +const struct drm_pagemap_devmem_ops *ops, +struct drm_pagemap *dpagemap,size_t size); +#endif
Thomas Hellström
maintainerRe: [PATCH v3 1/3] drm/gpusvm, drm/pagemap: Move migration functionality to drm_pagemap
Quoted 160 lines · Himal Prasad
Quoted 157 lines · Thomas Hellström
From: Matthew Brost <[email protected]>
The migration functionality and track-keeping of per-pagemap VRAM mapped to the CPU mm is not per
GPU_vm, but rather per pagemap. This is also reflected by the functions not needing thedrm_gpusvmstructures. So move todrm_pagemap.With this,
drm_gpusvmshouldn't really access the page zone-device- data
since its meaning is internal todrm_pagemap. Currently it's used to
reject mapping ranges backed by multipledrm_pagemapallocations. For now, make the zone-device-data a void pointer.Alter the interface of
drm_gpusvm_migrate_to_devmem()to ensure we don't
pass a gpusvm pointer.Rename
CONFIG_DRM_XE_DEVMEM_MIRRORtoCONFIG_DRM_XE_PAGEMAP.Matt is listed as author of this commit since he wrote most of the code,
and it makes sense to retain his git authorship.
Thomas mostly moved the code around.v3:
- Kerneldoc fixes (CI)
- Don't update documentation about how thedrm_pagemapmigration should be interpreted until upcoming patches where the functionality is implemented. (Matt Brost)Co-developed-by: Thomas Hellström
<[email protected]>Signed-off-by Thomas Hellström <[email protected]>---
Documentation/gpu/rfc/gpusvm.rst | 12 +- drivers/gpu/drm/Makefile | 6 +- drivers/gpu/drm/drm_gpusvm.c | 759 +---------------------
----
drivers/gpu/drm/drm_pagemap.c | 788
+++++++++++++++++++++++++++
drivers/gpu/drm/xe/Kconfig | 10 +- drivers/gpu/drm/xe/xe_bo_types.h | 2 +-drivers/gpu/drm/xe/xe_device_types.h| 2 +- drivers/gpu/drm/xe/xe_svm.c | 47 +- include/drm/drm_gpusvm.h | 96 ---- include/drm/drm_pagemap.h | 101 ++++10 files changed, 950 insertions(+), 873 deletions(-)create mode 100644
drivers/gpu/drm/drm_pagemap.cDocumentation/gpu/rfc/gpusvm.rst +0
-0@@ -73,15 +73,21 @@ Overview of baseline design.. kernel-doc::
drivers/gpu/drm/drm_gpusvm.c
:doc: Locking
-.. kernel-doc::drivers/gpu/drm/drm_gpusvm.c
- :doc: Migration
-
.. kernel-doc::drivers/gpu/drm/drm_gpusvm.c
:doc: Partial Unmapping of Ranges
.. kernel-doc::drivers/gpu/drm/drm_gpusvm.c
:doc: ExamplesDocumentation/gpu/rfc/gpusvm.rst +9
-0+Overview of drm_pagemap design +============================== + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Overview + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Migration +Possible future design features
===============================drivers/gpu/drm/Makefile +0
-0@@ -104,7 +104,11 @@ obj-$(CONFIG_DRM_PANEL_BACKLIGHT_QUIRKS) +=drm_panel_backlight_quirks.o
#
obj-$(CONFIG_DRM_EXEC) += drm_exec.o
obj-$(CONFIG_DRM_GPUVM) += drm_gpuvm.odrivers/gpu/drm/Makefile +5
-1-obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm.o + +drm_gpusvm_helper-y := \ + drm_gpusvm.o\ + drm_pagemap.o +obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm_helper.o
obj-$(CONFIG_DRM_BUDDY) += drm_buddy.odrivers/gpu/drm/drm_gpusvm.c +0
-0@@ -8,10 +8,9 @@
#include <linux/dma-mapping.h>
#include <linux/hmm.h>
+#include <linux/hugetlb_inline.h>
#include <linux/memremap.h>
-#include <linux/migrate.h>
#include <linux/mm_types.h>
-#include <linux/pagemap.h>
#include <linux/slab.h>
#include <drm/drm_device.h>drivers/gpu/drm/drm_gpusvm.c +0
-0@@ -107,21 +106,6 @@* to add annotations to GPU SVM. */
-/**
- * DOC: Migration
- *
- * The migration support is quite simple, allowing migration between RAM and
- * device memory at the range granularity. For example, GPU SVM currently does
- * not support mixing RAM and device memory pages within a range. This means
- * that upon GPU fault, the entire range can be migrated to device memory, and
- * upon CPU fault, the entire range is migrated to RAM. Mixed RAM and device
- * memory storage within a range could be added in the future if required.
- *
- * The reasoning for only supporting range granularity is as
follows: it
- * simplifies the implementation, and range sizes are driver- defined and should
- * be relatively small.
- */
-
/**
* DOC: Partial Unmapping of Ranges
*drivers/gpu/drm/drm_gpusvm.c +0
-4@@ -193,10 +177,9 @@ * if (driver_migration_policy(range)) { * mmap_read_lock(mm); * devmem = driver_alloc_devmem(); - * err =drm_gpusvm_migrate_to_devmem (gpusvm, range, - * devmem_allocation, - * &ctx); - *mmap_read_unlock (mm);
Quoted 8 lines
Quoted 4 lines
drivers/gpu/drm/drm_gpusvm.c +2
-0+ * err = drm_pagemap_migrate_to_devmem(devmem, gpusvm->mm, gpuva_start, + *
gpuva_end,driver_pgmap_owner());fix doc passing timeslice as parameter.
Will fix.
Quoted 1 line
8<---------------------------------------------------------------------
Quoted 15 lines
Quoted 6 lines
+/**
+ *drm_pagemap_migrate_to_devmem()- Migrate a structmm_structrange to device memory
+ * @devmem_allocation: The device memory allocation to migrate to.
+ * The caller should hold a reference to the device memory allocation,Quoted 8 lines
+ * and the reference is consumed by this function unless it returns with
+ * an error.
+ * @mm: Pointer to the structmm_struct.
+ * @start: Start of the virtual address range to migrate.
+ * @end: End of the virtual address range to migrate.
+ * @timeslice_ms: The time requested for the migrated pages to
+ * be present in the cpu memory map before migrated back.
Quoted 5 lines
Shouldn't this be present in gpu or cpu memory map ? We are using this
to ensure pagefault can be handled effectively by ensuring pages remain
in vram here for prescribed time too.
So with this split, drm_pagemap is responsible for migrating memory and updating the CPU memory map only, whereas drm_gpusvm is responsible for setting up the GPU memory maps. So if it remains in the CPU memory map, then nothing will force the gpu vms to invalidate, unless the gpu driver decides to invalidate itself.
But looking at this i should probably rephrase "before migrated back" to "before being allowed to be migrated back".
Quoted 14 lines
Quoted 13 lines
+ * @pgmap_owner: Not used currently, since only system memory is considered.
+ *
+ * This function migrates the specified virtual address range to device memory.
+ * It performs the necessary setup and invokes the driver-specific operations for
+ * migration to device memory. Expected to be called while holding the mmap lock in
+ * at least read mode.
+ *
+ * Return: %0 on success, negative error code on failure.
Quoted 1 line
s/%0/0
kerneldoc prefers "%" before constants, no?
Thanks,
Thomas
Ghimiray, Himal Prasad
reviewerRe: [PATCH v3 1/3] drm/gpusvm, drm/pagemap: Move migration functionality to drm_pagemap
Quoted 196 lines · Thomas Hellström
Quoted 160 lines · Himal Prasad
Quoted 157 lines · Thomas Hellström
From: Matthew Brost <[email protected]>
The migration functionality and track-keeping of per-pagemap VRAM mapped to the CPU mm is not per
GPU_vm, but rather per pagemap. This is also reflected by the functions not needing thedrm_gpusvmstructures. So move todrm_pagemap.With this,
drm_gpusvmshouldn't really access the page zone-device- data
since its meaning is internal todrm_pagemap. Currently it's used to
reject mapping ranges backed by multipledrm_pagemapallocations. For now, make the zone-device-data a void pointer.Alter the interface of
drm_gpusvm_migrate_to_devmem()to ensure we don't
pass a gpusvm pointer.Rename
CONFIG_DRM_XE_DEVMEM_MIRRORtoCONFIG_DRM_XE_PAGEMAP.Matt is listed as author of this commit since he wrote most of the code,
and it makes sense to retain his git authorship.
Thomas mostly moved the code around.v3:
- Kerneldoc fixes (CI)
- Don't update documentation about how thedrm_pagemapmigration should be interpreted until upcoming patches where the functionality is implemented. (Matt Brost)Co-developed-by: Thomas Hellström
<[email protected]>Signed-off-by Thomas Hellström <[email protected]>--- Documentation/gpu/rfc/gpusvm.rst | 12 +- drivers/gpu/drm/Makefile | 6 +- drivers/gpu/drm/drm_gpusvm.c | 759 +--------------------- ---- drivers/gpu/drm/drm_pagemap.c | 788 +++++++++++++++++++++++++++ drivers/gpu/drm/xe/Kconfig | 10 +- drivers/gpu/drm/xe/xe_bo_types.h | 2 +- drivers/gpu/drm/xe/xe_device_types.h | 2 +- drivers/gpu/drm/xe/xe_svm.c | 47 +- include/drm/drm_gpusvm.h | 96 ---- include/drm/drm_pagemap.h | 101 ++++ 10 files changed, 950 insertions(+), 873 deletions(-)create mode 100644
drivers/gpu/drm/drm_pagemap.cDocumentation/gpu/rfc/gpusvm.rst +9
-3@@ -73,15 +73,21 @@ Overview of baseline design .. kernel-doc:: drivers/gpu/drm/drm_gpusvm.c :doc: Locking -.. kernel-doc:: drivers/gpu/drm/drm_gpusvm.c - :doc: Migration - .. kernel-doc:: drivers/gpu/drm/drm_gpusvm.c :doc: Partial Unmapping of Ranges .. kernel-doc:: drivers/gpu/drm/drm_gpusvm.c :doc: Examples +Overview of drm_pagemap design +============================== + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Overview + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Migration + Possible future design features ===============================drivers/gpu/drm/Makefile +0
-0@@ -104,7 +104,11 @@ obj-$(CONFIG_DRM_PANEL_BACKLIGHT_QUIRKS) +=drm_panel_backlight_quirks.o
#
obj-$(CONFIG_DRM_EXEC) += drm_exec.o
obj-$(CONFIG_DRM_GPUVM) += drm_gpuvm.odrivers/gpu/drm/Makefile +5
-1-obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm.o + +drm_gpusvm_helper-y := \ + drm_gpusvm.o\ + drm_pagemap.o +obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm_helper.o obj-$(CONFIG_DRM_BUDDY) += drm_buddy.odrivers/gpu/drm/drm_gpusvm.c +1
-6@@ -8,10 +8,9 @@ #include <linux/dma-mapping.h> #include <linux/hmm.h> +#include <linux/hugetlb_inline.h> #include <linux/memremap.h> -#include <linux/migrate.h> #include <linux/mm_types.h> -#include <linux/pagemap.h> #include <linux/slab.h> #include <drm/drm_device.h>@@ -107,21 +106,6 @@* to add annotations to GPU SVM. */ -/** -* DOC: Migration -* -* The migration support is quite simple, allowing migration between RAM and
- * device memory at the range granularity. For example, GPU SVM currently does
- * not support mixing RAM and device memory pages within a range. This means
- * that upon GPU fault, the entire range can be migrated to device memory, and
- * upon CPU fault, the entire range is migrated to RAM. Mixed RAM and device
- * memory storage within a range could be added in the future if required.
- *
- * The reasoning for only supporting range granularity is as
follows: it
- * simplifies the implementation, and range sizes are driver- defined and should
- * be relatively small.
- */drivers/gpu/drm/drm_gpusvm.c +0
-5- /*** DOC: Partial Unmapping of Ranges * @@ -193,10 +177,9 @@ * if (driver_migration_policy(range)) { * mmap_read_lock(mm); * devmem = driver_alloc_devmem(); - * err =drm_gpusvm_migrate_to_devmem (gpusvm, range, - * devmem_allocation, - * &ctx); - *mmap_read_unlock (mm);Quoted 8 lines
Quoted 4 lines
drivers/gpu/drm/drm_gpusvm.c +2
-0+ * err = drm_pagemap_migrate_to_devmem(devmem, gpusvm->mm, gpuva_start, + *
gpuva_end,driver_pgmap_owner());fix doc passing timeslice as parameter.
Will fix.
Quoted 1 line
8<---------------------------------------------------------------------
Quoted 15 lines
Quoted 6 lines
+/**
+ *drm_pagemap_migrate_to_devmem()- Migrate a structmm_structrange to device memory
+ * @devmem_allocation: The device memory allocation to migrate to.
+ * The caller should hold a reference to the device memory allocation,Quoted 8 lines
+ * and the reference is consumed by this function unless it returns with
+ * an error.
+ * @mm: Pointer to the structmm_struct.
+ * @start: Start of the virtual address range to migrate.
+ * @end: End of the virtual address range to migrate.
+ * @timeslice_ms: The time requested for the migrated pages to
+ * be present in the cpu memory map before migrated back.Quoted 5 lines
Shouldn't this be present in gpu or cpu memory map ? We are using this
to ensure pagefault can be handled effectively by ensuring pages remain
in vram here for prescribed time too.
Quoted 5 lines
So with this split,
drm_pagemapis responsible for migrating memory and updating the CPU memory map only, whereasdrm_gpusvmis responsible for setting up the GPU memory maps. So if it remains in the CPU memory map, then nothing will force the gpu vms to invalidate, unless the gpu driver decides to invalidate itself.
Thats true.
Quoted 3 lines
But looking at this i should probably rephrase "before migrated back" to "before being allowed to be migrated back".
The confusion for me is that timeslice_ms does not represent the time pages are required to stay in the CPU memory map, but rather the time they must remain in the GPU memory map. We defer migrate_to_smem until this timeslice has expired.
Quoted 16 lines
Quoted 16 lines
Quoted 15 lines
+ * @pgmap_owner: Not used currently, since only system memory is considered.
+ *
+ * This function migrates the specified virtual address range to device memory.
+ * It performs the necessary setup and invokes the driver-specific operations for
+ * migration to device memory. Expected to be called while holding the mmap lock in
+ * at least read mode.
+ *
+ * Return: %0 on success, negative error code on failure.
Quoted 3 lines
Quoted 1 line
s/%0/0
kerneldoc prefers "%" before constants, no?
Thats correct. My miss.
Quoted 6 lines
Thanks,
Thomas
Thomas Hellström
maintainerRe: [PATCH v3 1/3] drm/gpusvm, drm/pagemap: Move migration functionality to drm_pagemap
Quoted 236 lines · Himal Prasad
Quoted 217 lines · Thomas Hellström
Quoted 178 lines · Himal Prasad
Quoted 175 lines · Thomas Hellström
From: Matthew Brost <[email protected]>
The migration functionality and track-keeping of per-pagemap VRAM
mapped to the CPU mm is not perGPU_vm, but rather per pagemap. This is also reflected by the functions not needing thedrm_gpusvm
structures. So move todrm_pagemap.With this,
drm_gpusvmshouldn't really access the page zone- device-
data
since its meaning is internal todrm_pagemap. Currently it's used
to
reject mapping ranges backed by multipledrm_pagemapallocations.
For now, make the zone-device-data a void pointer.Alter the interface of
drm_gpusvm_migrate_to_devmem()to ensure we
don't
pass a gpusvm pointer.Rename
CONFIG_DRM_XE_DEVMEM_MIRRORtoCONFIG_DRM_XE_PAGEMAP.Matt is listed as author of this commit since he wrote most of the
code,
and it makes sense to retain his git authorship. Thomas mostly moved the code around.v3:
- Kerneldoc fixes (CI)
- Don't update documentation about how thedrm_pagemapmigration should be interpreted until upcoming patches where the functionality is implemented. (Matt Brost)Co-developed-by: Thomas Hellström <[email protected]>
Signed-off-by: Thomas Hellström <[email protected]>
---
Documentation/gpu/rfc/gpusvm.rst | 12 +- drivers/gpu/drm/Makefile | 6 +- drivers/gpu/drm/drm_gpusvm.c | 759 +----------------
-----
----
drivers/gpu/drm/drm_pagemap.c | 788
+++++++++++++++++++++++++++
drivers/gpu/drm/xe/Kconfig | 10 +- drivers/gpu/drm/xe/xe_bo_types.h | 2 +-drivers/gpu/drm/xe/xe_device_types.h| 2 +- drivers/gpu/drm/xe/xe_svm.c | 47 +- include/drm/drm_gpusvm.h | 96 ---- include/drm/drm_pagemap.h | 101 ++++10 files changed, 950 insertions(+), 873 deletions(-)create mode 100644
drivers/gpu/drm/drm_pagemap.cDocumentation/gpu/rfc/gpusvm.rst +0
-0@@ -73,15 +73,21 @@ Overview of baseline design.. kernel-doc::
drivers/gpu/drm/drm_gpusvm.c
:doc: Locking
-.. kernel-doc::drivers/gpu/drm/drm_gpusvm.c
- :doc: Migration
-
.. kernel-doc::drivers/gpu/drm/drm_gpusvm.c
:doc: Partial Unmapping of Ranges
.. kernel-doc::drivers/gpu/drm/drm_gpusvm.c
:doc: ExamplesDocumentation/gpu/rfc/gpusvm.rst +9
-0+Overview of drm_pagemap design +============================== + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Overview + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Migration +Possible future design features
===============================drivers/gpu/drm/Makefile +1
-0@@ -104,7 +104,11 @@ obj-$(CONFIG_DRM_PANEL_BACKLIGHT_QUIRKS) +=drm_panel_backlight_quirks.o
#
obj-$(CONFIG_DRM_EXEC) += drm_exec.o
obj-$(CONFIG_DRM_GPUVM) += drm_gpuvm.odrivers/gpu/drm/Makefile +5
-1-obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm.o + +drm_gpusvm_helper-y := \ + drm_gpusvm.o\ + drm_pagemap.o +obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm_helper.o
obj-$(CONFIG_DRM_BUDDY) += drm_buddy.odrivers/gpu/drm/drm_gpusvm.c +0
-0@@ -8,10 +8,9 @@
#include <linux/dma-mapping.h>
#include <linux/hmm.h>
+#include <linux/hugetlb_inline.h>
#include <linux/memremap.h>
-#include <linux/migrate.h>
#include <linux/mm_types.h>
-#include <linux/pagemap.h>
#include <linux/slab.h>
#include <drm/drm_device.h>drivers/gpu/drm/drm_gpusvm.c +0
-0@@ -107,21 +106,6 @@* to add annotations to GPU SVM. */
-/**
- * DOC: Migration
- *
- * The migration support is quite simple, allowing migration between RAM and
- * device memory at the range granularity. For example, GPU SVM
currently does
- * not support mixing RAM and device memory pages within a range.
This means
- * that upon GPU fault, the entire range can be migrated to device
memory, and
- * upon CPU fault, the entire range is migrated to RAM. Mixed RAM
and device
- * memory storage within a range could be added in the future if
required.
- *
- * The reasoning for only supporting range granularity is as
follows: it
- * simplifies the implementation, and range sizes are driver- defined and should
- * be relatively small.
- */
-
/**
* DOC: Partial Unmapping of Ranges
*drivers/gpu/drm/drm_gpusvm.c +0
-4@@ -193,10 +177,9 @@ * if (driver_migration_policy(range)) { * mmap_read_lock(mm); * devmem = driver_alloc_devmem(); - * err =drm_gpusvm_migrate_to_devmem (gpusvm, range, - * devmem_allocation, - * &ctx); - *mmap_read_unlock (mm);Quoted 8 lines
Quoted 4 lines
+ * err =
drm_pagemap_migrate_to_devmem(devmem,gpusvm->mm,gpuva_start,
+ *gpuva_end,driver_pgmap_owner());fix doc passing timeslice as parameter.
Will fix.
Quoted 1 line
8<-----------------------------------------------------------------
----Quoted 17 lines
Quoted 8 lines
+/**
+ *drm_pagemap_migrate_to_devmem()- Migrate a structmm_struct
range to device memory
+ * @devmem_allocation: The device memory allocation to migrate to.
+ * The caller should hold a reference to the device memory allocation,Quoted 8 lines
+ * and the reference is consumed by this function unless it returns with
+ * an error.
+ * @mm: Pointer to the structmm_struct.
+ * @start: Start of the virtual address range to migrate.
+ * @end: End of the virtual address range to migrate.
+ * @timeslice_ms: The time requested for the migrated pages to
+ * be present in the cpu memory map before migrated back.Quoted 5 lines
Shouldn't this be present in gpu or cpu memory map ? We are using this
to ensure pagefault can be handled effectively by ensuring pages remain
in vram here for prescribed time too.Quoted 8 lines
So with this split,
drm_pagemapis responsible for migrating memory and
updating the CPU memory map only, whereasdrm_gpusvmis responsible for
setting up the GPU memory maps. So if it remains in the CPU memory map,
then nothing will force the gpu vms to invalidate, unless the gpu driver decides to invalidate itself.Thats true.
Quoted 4 lines
But looking at this i should probably rephrase "before migrated back"
to "before being allowed to be migrated back".
Quoted 5 lines
The confusion for me is that
timeslice_msdoes not represent the time pages are required to stay in the CPU memory map, but rather the time they must remain in the GPU memory map. We defermigrate_to_smemuntil
this timeslice has expired.
Yeah, although drm_pagemap is not aware of any gpu memory map so it would be incorrect to bring that up in the api docs. Could add some discussion, though, that "this could be used to..." and give the typical gpu use-case?
Thanks,
Thomas
Matthew Brost
maintainerRe: [PATCH v3 1/3] drm/gpusvm, drm/pagemap: Move migration functionality to drm_pagemap
Quoted 198 lines · Thomas Hellström
Quoted 197 lines · Himal Prasad
Quoted 194 lines · Thomas Hellström
Quoted 178 lines · Himal Prasad
Quoted 175 lines · Thomas Hellström
From: Matthew Brost <[email protected]>
The migration functionality and track-keeping of per-pagemap VRAM
mapped to the CPU mm is not perGPU_vm, but rather per pagemap. This is also reflected by the functions not needing thedrm_gpusvm
structures. So move todrm_pagemap.With this,
drm_gpusvmshouldn't really access the page zone- device-
data
since its meaning is internal todrm_pagemap. Currently it's used
to
reject mapping ranges backed by multipledrm_pagemapallocations.
For now, make the zone-device-data a void pointer.Alter the interface of
drm_gpusvm_migrate_to_devmem()to ensure we
don't
pass a gpusvm pointer.Rename
CONFIG_DRM_XE_DEVMEM_MIRRORtoCONFIG_DRM_XE_PAGEMAP.Matt is listed as author of this commit since he wrote most of the
code,
and it makes sense to retain his git authorship. Thomas mostly moved the code around.v3:
- Kerneldoc fixes (CI)
- Don't update documentation about how thedrm_pagemapmigration should be interpreted until upcoming patches where the functionality is implemented. (Matt Brost)Co-developed-by: Thomas Hellström <[email protected]>
Signed-off-by: Thomas Hellström <[email protected]>
---
Documentation/gpu/rfc/gpusvm.rst | 12 +- drivers/gpu/drm/Makefile | 6 +- drivers/gpu/drm/drm_gpusvm.c | 759 +----------------
-----
----
drivers/gpu/drm/drm_pagemap.c | 788
+++++++++++++++++++++++++++
drivers/gpu/drm/xe/Kconfig | 10 +- drivers/gpu/drm/xe/xe_bo_types.h | 2 +-drivers/gpu/drm/xe/xe_device_types.h| 2 +- drivers/gpu/drm/xe/xe_svm.c | 47 +- include/drm/drm_gpusvm.h | 96 ---- include/drm/drm_pagemap.h | 101 ++++10 files changed, 950 insertions(+), 873 deletions(-)create mode 100644
drivers/gpu/drm/drm_pagemap.cDocumentation/gpu/rfc/gpusvm.rst +0
-0@@ -73,15 +73,21 @@ Overview of baseline design.. kernel-doc::
drivers/gpu/drm/drm_gpusvm.c
:doc: Locking
-.. kernel-doc::drivers/gpu/drm/drm_gpusvm.c
- :doc: Migration
-
.. kernel-doc::drivers/gpu/drm/drm_gpusvm.c
:doc: Partial Unmapping of Ranges
.. kernel-doc::drivers/gpu/drm/drm_gpusvm.c
:doc: ExamplesDocumentation/gpu/rfc/gpusvm.rst +9
-0+Overview of drm_pagemap design +============================== + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Overview + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Migration +Possible future design features
===============================drivers/gpu/drm/Makefile +1
-0@@ -104,7 +104,11 @@ obj-$(CONFIG_DRM_PANEL_BACKLIGHT_QUIRKS) +=drm_panel_backlight_quirks.o
#
obj-$(CONFIG_DRM_EXEC) += drm_exec.o
obj-$(CONFIG_DRM_GPUVM) += drm_gpuvm.odrivers/gpu/drm/Makefile +5
-1-obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm.o + +drm_gpusvm_helper-y := \ + drm_gpusvm.o\ + drm_pagemap.o +obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm_helper.o
obj-$(CONFIG_DRM_BUDDY) += drm_buddy.odrivers/gpu/drm/drm_gpusvm.c +0
-0@@ -8,10 +8,9 @@
#include <linux/dma-mapping.h>
#include <linux/hmm.h>
+#include <linux/hugetlb_inline.h>
#include <linux/memremap.h>
-#include <linux/migrate.h>
#include <linux/mm_types.h>
-#include <linux/pagemap.h>
#include <linux/slab.h>
#include <drm/drm_device.h>drivers/gpu/drm/drm_gpusvm.c +0
-0@@ -107,21 +106,6 @@* to add annotations to GPU SVM. */
-/**
- * DOC: Migration
- *
- * The migration support is quite simple, allowing migration between RAM and
- * device memory at the range granularity. For example, GPU SVM
currently does
- * not support mixing RAM and device memory pages within a range.
This means
- * that upon GPU fault, the entire range can be migrated to device
memory, and
- * upon CPU fault, the entire range is migrated to RAM. Mixed RAM
and device
- * memory storage within a range could be added in the future if
required.
- *
- * The reasoning for only supporting range granularity is as
follows: it
- * simplifies the implementation, and range sizes are driver- defined and should
- * be relatively small.
- */
-
/**
* DOC: Partial Unmapping of Ranges
*drivers/gpu/drm/drm_gpusvm.c +0
-4@@ -193,10 +177,9 @@ * if (driver_migration_policy(range)) { * mmap_read_lock(mm); * devmem = driver_alloc_devmem(); - * err =drm_gpusvm_migrate_to_devmem (gpusvm, range, - * devmem_allocation, - * &ctx); - *mmap_read_unlock (mm);Quoted 8 lines
Quoted 4 lines
+ * err =
drm_pagemap_migrate_to_devmem(devmem,gpusvm->mm,gpuva_start,
+ *gpuva_end,driver_pgmap_owner());fix doc passing timeslice as parameter.
Will fix.
Quoted 1 line
8<-----------------------------------------------------------------
----
Quoted 16 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
+/**
+ *drm_pagemap_migrate_to_devmem()- Migrate a structmm_struct
range to device memory
+ * @devmem_allocation: The device memory allocation to migrate to.
+ * The caller should hold a reference to the device memory allocation,Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
+ * and the reference is consumed by this function unless it returns with
+ * an error.
+ * @mm: Pointer to the structmm_struct.
+ * @start: Start of the virtual address range to migrate.
+ * @end: End of the virtual address range to migrate.
+ * @timeslice_ms: The time requested for the migrated pages to
+ * be present in the cpu memory map before migrated back.
As Himal suggests and see below...
s/cpu/gpu
Quoted 23 lines
Quoted 23 lines
Quoted 7 lines
Quoted 6 lines
Shouldn't this be present in gpu or cpu memory map ? We are using this
to ensure pagefault can be handled effectively by ensuring pages remain
in vram here for prescribed time too.Quoted 8 lines
So with this split,
drm_pagemapis responsible for migrating memory and
updating the CPU memory map only, whereasdrm_gpusvmis responsible for
setting up the GPU memory maps. So if it remains in the CPU memory map,
then nothing will force the gpu vms to invalidate, unless the gpu driver decides to invalidate itself.Thats true.
Quoted 4 lines
But looking at this i should probably rephrase "before migrated back"
to "before being allowed to be migrated back".
Quoted 5 lines
Quoted 5 lines
The confusion for me is that
timeslice_msdoes not represent the time pages are required to stay in the CPU memory map, but rather the time they must remain in the GPU memory map. We defermigrate_to_smemuntil
this timeslice has expired.
Ideally, we'd want the timeslice to ensure CPU residency as well. The issue arises because the timeslice is attached to the physical memory (drm_gpusvm_devmem), which is not readily available on the GPU fault handling side. UMD testing showed that ensuring CPU residency is not required for atomics to work, so this part of the implementation was left out. We can revisit and find a solution here if it turns out to be required.
For now, I'd fix the kernel doc above (s/cpu/gpu) with my suggestion and perhaps put in a comment why we don't wait on the GPU fault side.
Matt
Quoted 9 lines
Yeah, although
drm_pagemapis not aware of any gpu memory map so it would be incorrect to bring that up in the api docs. Could add some discussion, though, that "this could be used to..." and give the typical gpu use-case?Thanks,
Thomas
Thomas Hellström
maintainerRe: [PATCH v3 1/3] drm/gpusvm, drm/pagemap: Move migration functionality to drm_pagemap
Quoted 288 lines · Matthew Brost
Quoted 228 lines · Thomas Hellström
Quoted 227 lines · Himal Prasad
Quoted 224 lines · Thomas Hellström
Quoted 195 lines · Himal Prasad
Quoted 192 lines · Thomas Hellström
From: Matthew Brost <[email protected]>
The migration functionality and track-keeping of per- pagemap
VRAM
mapped to the CPU mm is not perGPU_vm, but rather per pagemap.
This is also reflected by the functions not needing thedrm_gpusvm
structures. So move todrm_pagemap.With this,
drm_gpusvmshouldn't really access the page zone-
device-
data
since its meaning is internal todrm_pagemap. Currently it's
used
to
reject mapping ranges backed by multipledrm_pagemapallocations.
For now, make the zone-device-data a void pointer.Alter the interface of
drm_gpusvm_migrate_to_devmem()to ensure
we
don't
pass a gpusvm pointer.Rename
CONFIG_DRM_XE_DEVMEM_MIRRORtoCONFIG_DRM_XE_PAGEMAP.Matt is listed as author of this commit since he wrote most of
the
code,
and it makes sense to retain his git authorship. Thomas mostly moved the code around.v3:
- Kerneldoc fixes (CI)
- Don't update documentation about how thedrm_pagemapmigration should be interpreted until upcoming patches where the functionality is implemented. (Matt Brost)Co-developed-by: Thomas Hellström <[email protected]>
Signed-off-by: Thomas Hellström <[email protected]>
---
Documentation/gpu/rfc/gpusvm.rst | 12 +- drivers/gpu/drm/Makefile | 6 +- drivers/gpu/drm/drm_gpusvm.c | 759 +------------
----
-----
----
drivers/gpu/drm/drm_pagemap.c | 788
+++++++++++++++++++++++++++
drivers/gpu/drm/xe/Kconfig | 10 +- drivers/gpu/drm/xe/xe_bo_types.h | 2 +-drivers/gpu/drm/xe/xe_device_types.h| 2 +- drivers/gpu/drm/xe/xe_svm.c | 47 +- include/drm/drm_gpusvm.h | 96 ---- include/drm/drm_pagemap.h | 101 ++++10 files changed, 950 insertions(+), 873 deletions(-)create mode 100644
drivers/gpu/drm/drm_pagemap.cDocumentation/gpu/rfc/gpusvm.rst +0
-0@@ -73,15 +73,21 @@ Overview of baseline design.. kernel-doc::
drivers/gpu/drm/drm_gpusvm.c
:doc: Locking
-.. kernel-doc::drivers/gpu/drm/drm_gpusvm.c
- :doc: Migration
-
.. kernel-doc::drivers/gpu/drm/drm_gpusvm.c
:doc: Partial Unmapping of Ranges
.. kernel-doc::drivers/gpu/drm/drm_gpusvm.c
:doc: ExamplesDocumentation/gpu/rfc/gpusvm.rst +9
-0+Overview of drm_pagemap design +============================== + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Overview + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Migration +Possible future design features
===============================drivers/gpu/drm/Makefile +1
-0@@ -104,7 +104,11 @@ obj- $(CONFIG_DRM_PANEL_BACKLIGHT_QUIRKS) +=drm_panel_backlight_quirks.o
#
obj-$(CONFIG_DRM_EXEC) += drm_exec.o
obj-$(CONFIG_DRM_GPUVM) += drm_gpuvm.odrivers/gpu/drm/Makefile +5
-1-obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm.o + +drm_gpusvm_helper-y := \ + drm_gpusvm.o\ + drm_pagemap.o +obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm_helper.o
obj-$(CONFIG_DRM_BUDDY) += drm_buddy.odrivers/gpu/drm/drm_gpusvm.c +0
-0@@ -8,10 +8,9 @@
#include <linux/dma-mapping.h>
#include <linux/hmm.h>
+#include <linux/hugetlb_inline.h>
#include <linux/memremap.h>
-#include <linux/migrate.h>
#include <linux/mm_types.h>
-#include <linux/pagemap.h>
#include <linux/slab.h>
#include <drm/drm_device.h>drivers/gpu/drm/drm_gpusvm.c +0
-0@@ -107,21 +106,6 @@* to add annotations to GPU SVM. */
-/**
- * DOC: Migration
- *
- * The migration support is quite simple, allowing migration
between RAM and
- * device memory at the range granularity. For example, GPU
SVM
currently does
- * not support mixing RAM and device memory pages within a range.
This means
- * that upon GPU fault, the entire range can be migrated to
device
memory, and
- * upon CPU fault, the entire range is migrated to RAM. Mixed
RAM
and device
- * memory storage within a range could be added in the future
if
required.
- *
- * The reasoning for only supporting range granularity is as
follows: it
- * simplifies the implementation, and range sizes are driver-
defined and should
- * be relatively small.
- */
-
/**
* DOC: Partial Unmapping of Ranges
*drivers/gpu/drm/drm_gpusvm.c +1
-4@@ -193,10 +177,9 @@ * if (driver_migration_policy(range)) { * mmap_read_lock(mm); * devmem = driver_alloc_devmem(); - * err =drm_gpusvm_migrate_to_devmem (gpusvm, range, - * devmem_allocation, - * &ctx); - *mmap_read_unlock (mm); + * err =Quoted 8 lines
Quoted 4 lines
drm_pagemap_migrate_to_devmem(devmem,
gpusvm->mm,gpuva_start,
+ *gpuva_end,driver_pgmap_owner());fix doc passing timeslice as parameter.
Will fix.
Quoted 1 line
8<-------------------------------------------------------------
----
----Quoted 11 lines
Quoted 11 lines
+/**
+ *drm_pagemap_migrate_to_devmem()- Migrate a structmm_struct
range to device memory
+ * @devmem_allocation: The device memory allocation to migrate
to.
+ * The caller should hold a reference to the device memory allocation,
+ * and the reference is consumed by this function unless itQuoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
returns with
+ * an error.
+ * @mm: Pointer to the structmm_struct.
+ * @start: Start of the virtual address range to migrate.
+ * @end: End of the virtual address range to migrate.
+ * @timeslice_ms: The time requested for the migrated pages to
+ * be present in the cpu memory map before migrated back.As Himal suggests and see below...
s/cpu/gpu
Quoted 29 lines
Quoted 29 lines
Quoted 9 lines
Quoted 8 lines
Shouldn't this be present in gpu or cpu memory map ? We are using
this
to ensure pagefault can be handled effectively by ensuring pages
remain
in vram here for prescribed time too.Quoted 12 lines
So with this split,
drm_pagemapis responsible for migrating memory
and
updating the CPU memory map only, whereasdrm_gpusvmis responsible
for
setting up the GPU memory maps. So if it remains in the CPU memory
map,
then nothing will force the gpu vms to invalidate, unless the gpu
driver decides to invalidate itself.Thats true.
Quoted 4 lines
But looking at this i should probably rephrase "before migrated back"
to "before being allowed to be migrated back".Quoted 7 lines
Quoted 7 lines
The confusion for me is that
timeslice_msdoes not represent the time
pages are required to stay in the CPU memory map, but rather the time
they must remain in the GPU memory map. We defermigrate_to_smemuntil
this timeslice has expired.Ideally, we'd want the timeslice to ensure CPU residency as well. The issue arises because the timeslice is attached to the physical memory (
drm_gpusvm_devmem), which is not readily available on the GPU fault handling side. UMD testing showed that ensuring CPU residency is not required for atomics to work, so this part of the implementation was left out. We can revisit and find a solution here if it turns out to be
required.
Quoted 6 lines
For now, I'd fix the kernel doc above (s/cpu/gpu) with my suggestion and
perhaps put in a comment why we don't wait on the GPU fault side.Matt
Hi, Matt, Please see my second response to Himal.
It might be that there is a confusion between "pages present in the cpu memory map" and "memory present in system memory"?
What I mean with "present in the cpu memory map" is that the pagemap pages (device private memory) is present in the struct mm_struct for at least xx ms. The struct mm is the authoritative information of which pages currently hold the data.
So "migrated pages are present in the cpu memory map" is therefore more or less equivalent to "data is present in device private memory".
So maybe I should rephrase this as
"The time requested for the pagemap pages to be present in @mm"? And add a discussion like in Himal's reply on the typical use-case?
the pagemap itself has no control of how and if the gpu vms choose to expose this. It can only guarantee that the authoritative location of data is the device private memory for timeslice_ms.
/Thomas
Quoted 10 lines
Quoted 10 lines
Yeah, although
drm_pagemapis not aware of any gpu memory map so it would be incorrect to bring that up in the api docs. Could add some discussion, though, that "this could be used to..." and give the typical gpu use-case?Thanks,
Thomas
Matthew Brost
maintainerRe: [PATCH v3 1/3] drm/gpusvm, drm/pagemap: Move migration functionality to drm_pagemap
Quoted 301 lines · Thomas Hellström
Quoted 288 lines · Matthew Brost
Quoted 228 lines · Thomas Hellström
Quoted 227 lines · Himal Prasad
Quoted 224 lines · Thomas Hellström
Quoted 195 lines · Himal Prasad
Quoted 192 lines · Thomas Hellström
From: Matthew Brost <[email protected]>
The migration functionality and track-keeping of per- pagemap
VRAM
mapped to the CPU mm is not perGPU_vm, but rather per pagemap.
This is also reflected by the functions not needing thedrm_gpusvm
structures. So move todrm_pagemap.With this,
drm_gpusvmshouldn't really access the page zone-
device-
data
since its meaning is internal todrm_pagemap. Currently it's
used
to
reject mapping ranges backed by multipledrm_pagemapallocations.
For now, make the zone-device-data a void pointer.Alter the interface of
drm_gpusvm_migrate_to_devmem()to ensure
we
don't
pass a gpusvm pointer.Rename
CONFIG_DRM_XE_DEVMEM_MIRRORtoCONFIG_DRM_XE_PAGEMAP.Matt is listed as author of this commit since he wrote most of
the
code,
and it makes sense to retain his git authorship. Thomas mostly moved the code around.v3:
- Kerneldoc fixes (CI)
- Don't update documentation about how thedrm_pagemapmigration should be interpreted until upcoming patches where the functionality is implemented. (Matt Brost)Co-developed-by: Thomas Hellström <[email protected]>
Signed-off-by: Thomas Hellström <[email protected]>
---
Documentation/gpu/rfc/gpusvm.rst | 12 +- drivers/gpu/drm/Makefile | 6 +- drivers/gpu/drm/drm_gpusvm.c | 759 +------------
----
-----
----
drivers/gpu/drm/drm_pagemap.c | 788
+++++++++++++++++++++++++++
drivers/gpu/drm/xe/Kconfig | 10 +- drivers/gpu/drm/xe/xe_bo_types.h | 2 +-drivers/gpu/drm/xe/xe_device_types.h| 2 +- drivers/gpu/drm/xe/xe_svm.c | 47 +- include/drm/drm_gpusvm.h | 96 ---- include/drm/drm_pagemap.h | 101 ++++10 files changed, 950 insertions(+), 873 deletions(-)create mode 100644
drivers/gpu/drm/drm_pagemap.cDocumentation/gpu/rfc/gpusvm.rst +0
-0@@ -73,15 +73,21 @@ Overview of baseline design.. kernel-doc::
drivers/gpu/drm/drm_gpusvm.c
:doc: Locking
-.. kernel-doc::drivers/gpu/drm/drm_gpusvm.c
- :doc: Migration
-
.. kernel-doc::drivers/gpu/drm/drm_gpusvm.c
:doc: Partial Unmapping of Ranges
.. kernel-doc::drivers/gpu/drm/drm_gpusvm.c
:doc: ExamplesDocumentation/gpu/rfc/gpusvm.rst +9
-0+Overview of drm_pagemap design +============================== + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Overview + +.. kernel-doc:: drivers/gpu/drm/drm_pagemap.c + :doc: Migration +Possible future design features
===============================drivers/gpu/drm/Makefile +1
-0@@ -104,7 +104,11 @@ obj- $(CONFIG_DRM_PANEL_BACKLIGHT_QUIRKS) +=drm_panel_backlight_quirks.o
#
obj-$(CONFIG_DRM_EXEC) += drm_exec.o
obj-$(CONFIG_DRM_GPUVM) += drm_gpuvm.odrivers/gpu/drm/Makefile +5
-1-obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm.o + +drm_gpusvm_helper-y := \ + drm_gpusvm.o\ + drm_pagemap.o +obj-$(CONFIG_DRM_GPUSVM) += drm_gpusvm_helper.o
obj-$(CONFIG_DRM_BUDDY) += drm_buddy.odrivers/gpu/drm/drm_gpusvm.c +0
-0@@ -8,10 +8,9 @@
#include <linux/dma-mapping.h>
#include <linux/hmm.h>
+#include <linux/hugetlb_inline.h>
#include <linux/memremap.h>
-#include <linux/migrate.h>
#include <linux/mm_types.h>
-#include <linux/pagemap.h>
#include <linux/slab.h>
#include <drm/drm_device.h>drivers/gpu/drm/drm_gpusvm.c +0
-0@@ -107,21 +106,6 @@* to add annotations to GPU SVM. */
-/**
- * DOC: Migration
- *
- * The migration support is quite simple, allowing migration
between RAM and
- * device memory at the range granularity. For example, GPU
SVM
currently does
- * not support mixing RAM and device memory pages within a range.
This means
- * that upon GPU fault, the entire range can be migrated to
device
memory, and
- * upon CPU fault, the entire range is migrated to RAM. Mixed
RAM
and device
- * memory storage within a range could be added in the future
if
required.
- *
- * The reasoning for only supporting range granularity is as
follows: it
- * simplifies the implementation, and range sizes are driver-
defined and should
- * be relatively small.
- */
-
/**
* DOC: Partial Unmapping of Ranges
*drivers/gpu/drm/drm_gpusvm.c +1
-4@@ -193,10 +177,9 @@ * if (driver_migration_policy(range)) { * mmap_read_lock(mm); * devmem = driver_alloc_devmem(); - * err =drm_gpusvm_migrate_to_devmem (gpusvm, range, - * devmem_allocation, - * &ctx); - *mmap_read_unlock (mm); + * err =Quoted 8 lines
Quoted 4 lines
drm_pagemap_migrate_to_devmem(devmem,
gpusvm->mm,gpuva_start,
+ *gpuva_end,driver_pgmap_owner());fix doc passing timeslice as parameter.
Will fix.
Quoted 1 line
8<-------------------------------------------------------------
----
----Quoted 11 lines
Quoted 11 lines
+/**
+ *drm_pagemap_migrate_to_devmem()- Migrate a structmm_struct
range to device memory
+ * @devmem_allocation: The device memory allocation to migrate
to.
+ * The caller should hold a reference to the device memory allocation,
+ * and the reference is consumed by this function unless itQuoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
returns with
+ * an error.
+ * @mm: Pointer to the structmm_struct.
+ * @start: Start of the virtual address range to migrate.
+ * @end: End of the virtual address range to migrate.
+ * @timeslice_ms: The time requested for the migrated pages to
+ * be present in the cpu memory map before migrated back.As Himal suggests and see below...
s/cpu/gpu
Quoted 29 lines
Quoted 29 lines
Quoted 9 lines
Quoted 8 lines
Shouldn't this be present in gpu or cpu memory map ? We are using
this
to ensure pagefault can be handled effectively by ensuring pages
remain
in vram here for prescribed time too.Quoted 12 lines
So with this split,
drm_pagemapis responsible for migrating memory
and
updating the CPU memory map only, whereasdrm_gpusvmis responsible
for
setting up the GPU memory maps. So if it remains in the CPU memory
map,
then nothing will force the gpu vms to invalidate, unless the gpu
driver decides to invalidate itself.Thats true.
Quoted 4 lines
But looking at this i should probably rephrase "before migrated back"
to "before being allowed to be migrated back".Quoted 7 lines
Quoted 7 lines
The confusion for me is that
timeslice_msdoes not represent the time
pages are required to stay in the CPU memory map, but rather the time
they must remain in the GPU memory map. We defermigrate_to_smemuntil
this timeslice has expired.Ideally, we'd want the timeslice to ensure CPU residency as well. The issue arises because the timeslice is attached to the physical memory (
drm_gpusvm_devmem), which is not readily available on the GPU fault handling side. UMD testing showed that ensuring CPU residency is not required for atomics to work, so this part of the implementation was left out. We can revisit and find a solution here if it turns out to be
required.Quoted 6 lines
For now, I'd fix the kernel doc above (s/cpu/gpu) with my suggestion and
perhaps put in a comment why we don't wait on the GPU fault side.Matt
Hi, Matt, Please see my second response to Himal.
It might be that there is a confusion between "pages present in the cpu memory map" and "memory present in system memory"?
Quoted 8 lines
What I mean with "present in the cpu memory map" is that the pagemap pages (device private memory) is present in the struct
mm_structfor at least xx ms. The struct mm is the authoritative information of which pages currently hold the data.So "migrated pages are present in the cpu memory map" is therefore more or less equivalent to "data is present in device private memory".
Ah, ok.
Quoted 2 lines
So maybe I should rephrase this as
"The time requested for the pagemap pages to be present in @mm"? And
Yea, that seems better.
Matt
Quoted 21 lines
add a discussion like in Himal's reply on the typical use-case?
the pagemap itself has no control of how and if the gpu vms choose to expose this. It can only guarantee that the authoritative location of data is the device private memory for
timeslice_ms./Thomas
Quoted 10 lines
Quoted 10 lines
Yeah, although
drm_pagemapis not aware of any gpu memory map so it would be incorrect to bring that up in the api docs. Could add some discussion, though, that "this could be used to..." and give the typical gpu use-case?Thanks,
Thomas
Thomas Hellström
maintainerPATCH v3 2/3 drm/pagemap: Add a populate_mm op
In reply to Thomas HellströmAdd an operation to populate a part of a drm_mm with device private memory. Clarify how migration using it is intended to work.
v3:
- Kerneldoc fixes and updates (Matt Brost).
--- drivers/gpu/drm/drm_gpusvm.c | 7 ++-- drivers/gpu/drm/drm_pagemap.c | 67 ++++++++++++++++++++++++++++------- include/drm/drm_pagemap.h | 34 ++++++++++++++++++ 3 files changed, 91 insertions(+), 17 deletions(-)
drivers/gpu/drm/drm_gpusvm.c +2 -5
@@ -175,11 +175,8 @@* } * * if (driver_migration_policy(range)) { -* mmap_read_lock(mm); -* devmem = driver_alloc_devmem(); -* err = drm_pagemap_migrate_to_devmem(devmem, gpusvm->mm, gpuva_start, -* gpuva_end, driver_pgmap_owner()); -* mmap_read_unlock(mm); +* err = drm_pagemap_populate_mm(driver_choose_drm_pagemap(), +* gpuva_start, gpuva_end, gpusvm->mm); * if (err) // CPU mappings may have changed * goto retry; * }
drivers/gpu/drm/drm_pagemap.c +55 -12
87 lines not on this page · raw message
include/drm/drm_pagemap.h +34 -0
@@ -92,6 +92,35 @@ struct drm_pagemap_ops {struct device *dev, struct drm_pagemap_device_addr addr); +/** +* @populate_mm: Populate part of the mm with @dpagemap memory, +* migrating existing data. +* @dpagemap: The struct drm_pagemap managing the memory. +* @start: The virtual start address in @mm +* @end: The virtual end address in @mm +* @mm: Pointer to a live mm. The caller must have an mmget() +* reference. +* +* The caller will have the mm lock at least in read mode. +* Note that there is no guarantee that the memory is resident +* after the function returns, it's best effort only. +* When the mm is not using the memory anymore, +* it will be released. The struct drm_pagemap might have a +* mechanism in place to reclaim the memory and the data will +* then be migrated. Typically to system memory. +* The implementation should hold sufficient runtime power- +* references while pages are used in an address space and +* should ideally guard against hardware device unbind in +* a way such that device pages are migrated back to system +* followed by device page removal. The implementation should +* return -ENODEV after device removal. +* +* Return: 0 if successful. Negative error code on error. +*/ +int (*populate_mm)(struct drm_pagemap *dpagemap, +unsigned long start,unsigned long end, +struct mm_struct *mm, +unsigned long timeslice_ms); };/**
@@ -205,4 +234,9 @@ void drm_pagemap_devmem_init(struct drm_pagemap_devmem *devmem_allocation,const struct drm_pagemap_devmem_ops *ops, struct drm_pagemap *dpagemap, size_t size); +int drm_pagemap_populate_mm (struct drm_pagemap *dpagemap, +unsigned long start,unsigned long end, +struct mm_struct *mm, +unsigned long timeslice_ms); +#endif
Signature · 1 line
2.49.0
Matthew Brost
maintainerRe: [PATCH v3 2/3] drm/pagemap: Add a populate_mm op
Quoted 9 lines · Thomas Hellström
Add an operation to populate a part of a
drm_mmwith device private memory. Clarify how migration using it is intended to work.v3:
- Kerneldoc fixes and updates (Matt Brost).Signed-off-by Thomas Hellström <[email protected]>Reviewed-by Matthew Brost <[email protected]> #v1
Review is still valid on this rev.
Matt
Quoted 171 lines
--- drivers/gpu/drm/drm_gpusvm.c | 7 ++-- drivers/gpu/drm/drm_pagemap.c | 67 ++++++++++++++++++++++++++++------- include/drm/drm_pagemap.h | 34 ++++++++++++++++++ 3 files changed, 91 insertions(+), 17 deletions(-)drivers/gpu/drm/drm_gpusvm.c +2
-5@@ -175,11 +175,8 @@* } * * if (driver_migration_policy(range)) { -* mmap_read_lock(mm); -* devmem = driver_alloc_devmem(); -* err = drm_pagemap_migrate_to_devmem(devmem, gpusvm->mm, gpuva_start, -* gpuva_end, driver_pgmap_owner()); -* mmap_read_unlock(mm); +* err = drm_pagemap_populate_mm(driver_choose_drm_pagemap(), +* gpuva_start, gpuva_end, gpusvm->mm); * if (err) // CPU mappings may have changed * goto retry; * } drivers/gpu/drm/drm_pagemap.c +55
-12@@ -6,6 +6,7 @@#include <linux/dma-mapping.h> #include <linux/migrate.h> #include <linux/pagemap.h> +#include <drm/drm_drv.h> #include <drm/drm_pagemap.h> /** @@ -20,23 +21,30 @@* system. * * Typically the DRM pagemap receives requests from one or more DRM GPU SVM -* instances to populate struct mm_struct virtual ranges with memory. +* instances to populate struct mm_struct virtual ranges with memory, and the +* migration is best effort only and may thus fail. The implementation should +* also handle device unbinding by blocking (return an -ENODEV) error for new +* population requests and after that migrate all device pages to system ram. */ /** * DOC: Migration * -* The migration support is quite simple, allowing migration between RAM and -* device memory at the range granularity. For example, GPU SVM currently does -* not support mixing RAM and device memory pages within a range. This means -* that upon GPU fault, the entire range can be migrated to device memory, and -* upon CPU fault, the entire range is migrated to RAM. Mixed RAM and device -* memory storage within a range could be added in the future if required. -* -* The reasoning for only supporting range granularity is as follows: it -* simplifies the implementation, and range sizes are driver-defined and should -* be relatively small. -* +* Migration granularity typically follows the GPU SVM range requests, but +* if there are clashes, due to races or due to the fact that multiple GPU +* SVM instances have different views of the ranges used, and because of that +* parts of a requested range is already present in the requested device memory, +* the implementation has a variety of options. It can fail and it can choose +* to populate only the part of the range that isn't already in device memory, +* and it can evict the range to system before trying to migrate. Ideally an +* implementation would just try to migrate the missing part of the range and +* allocate just enough memory to do so. +* +* When migrating to system memory as a response to a cpu fault or a device +* memory eviction request, currently a full device memory allocation is +* migrated back to system. Moving forward this might need improvement for +* situations where a single page needs bouncing between system memory and +* device memory due to, for example, atomic operations. * * Key DRM pagemap components: * @@ -786,3 +794,38 @@ struct drm_pagemap *drm_pagemap_page_to_dpagemap(struct page *page)return zdd->devmem_allocation->dpagemap; }EXPORT_SYMBOL_GPL (drm_pagemap_page_to_dpagemap); + +/** +* drm_pagemap_populate_mm() - Populate a virtual range with device memory pages +* @dpagemap: Pointer to the drm_pagemap managing the device memory +* @start: Start of the virtual range to populate. +* @end: End of the virtual range to populate. +* @mm: Pointer to the virtual address space. +* @timeslice_ms: The time requested for the migrated pages to +* be present in the cpu memory map before migrated back. +* +* Attempt to populate a virtual range with device memory pages, +* clearing them or migrating data from the existing pages if necessary. +* The function is best effort only, and implementations may vary +* in how hard they try to satisfy the request. +* +* Return: %0 on success, negative error code on error. If the hardware +* device was removed / unbound the function will return %-ENODEV. +*/ +int drm_pagemap_populate_mm (struct drm_pagemap *dpagemap, +unsigned long start,unsigned long end, +struct mm_struct *mm, +unsigned long timeslice_ms) +{ +int err; + +if (!mmget_not_zero (mm)) +return -EFAULT ; +mmap_read_lock (mm); + err = dpagemap->ops->populate_mm (dpagemap, start, end, mm, + timeslice_ms); +mmap_read_unlock (mm); +mmput (mm); + +return err; +}include/drm/drm_pagemap.h +34
-0@@ -92,6 +92,35 @@ struct drm_pagemap_ops {struct device *dev, struct drm_pagemap_device_addr addr); +/** +* @populate_mm: Populate part of the mm with @dpagemap memory, +* migrating existing data. +* @dpagemap: The struct drm_pagemap managing the memory. +* @start: The virtual start address in @mm +* @end: The virtual end address in @mm +* @mm: Pointer to a live mm. The caller must have an mmget() +* reference. +* +* The caller will have the mm lock at least in read mode. +* Note that there is no guarantee that the memory is resident +* after the function returns, it's best effort only. +* When the mm is not using the memory anymore, +* it will be released. The struct drm_pagemap might have a +* mechanism in place to reclaim the memory and the data will +* then be migrated. Typically to system memory. +* The implementation should hold sufficient runtime power- +* references while pages are used in an address space and +* should ideally guard against hardware device unbind in +* a way such that device pages are migrated back to system +* followed by device page removal. The implementation should +* return -ENODEV after device removal. +* +* Return: 0 if successful. Negative error code on error. +*/ +int (*populate_mm)(struct drm_pagemap *dpagemap, +unsigned long start,unsigned long end, +struct mm_struct *mm, +unsigned long timeslice_ms); };/** @@ -205,4 +234,9 @@ void drm_pagemap_devmem_init(struct drm_pagemap_devmem *devmem_allocation,const struct drm_pagemap_devmem_ops *ops, struct drm_pagemap *dpagemap, size_t size); +int drm_pagemap_populate_mm (struct drm_pagemap *dpagemap, +unsigned long start,unsigned long end, +struct mm_struct *mm, +unsigned long timeslice_ms); +#endif Signature · 1 line
2.49.0
Thomas Hellström
maintainerPATCH v3 3/3 drm/xe: Implement and use the drm_pagemap populate_mm op
In reply to Thomas HellströmAdd runtime PM since we might call populate_mm on a foreign device.
v3:
- Fix a kerneldoc failure (Matt Brost)
- Revert the bo type change from device to kernel (Matt Brost)
--- drivers/gpu/drm/drm_pagemap.c | 1 + drivers/gpu/drm/xe/xe_svm.c | 101 ++++++++++++++++++++-------------- drivers/gpu/drm/xe/xe_svm.h | 10 ++-- drivers/gpu/drm/xe/xe_tile.h | 11 ++++ drivers/gpu/drm/xe/xe_vm.c | 2 +- 5 files changed, 76 insertions(+), 49 deletions(-)
drivers/gpu/drm/drm_pagemap.c +1 -0
@@ -829,3 +829,4 @@ int drm_pagemap_populate_mm(struct drm_pagemap *dpagemap,return err; } +EXPORT_SYMBOL (drm_pagemap_populate_mm);
drivers/gpu/drm/xe/xe_svm.c +59 -42
182 lines not on this page · raw message
drivers/gpu/drm/xe/xe_svm.h +4 -6
@@ -70,8 +70,7 @@ int xe_svm_bo_evict(struct xe_bo *bo);void xe_svm_range_debug (struct xe_svm_range *range,const char *operation); -int xe_svm_alloc_vram (struct xe_vm *vm,struct xe_tile *tile, -struct xe_svm_range *range, +int xe_svm_alloc_vram (struct xe_tile *tile,struct xe_svm_range *range,const struct drm_gpusvm_ctx *ctx); struct xe_svm_range *xe_svm_range_find_or_insert(struct xe_vm *vm, u64 addr,
@@ -237,10 +236,9 @@ void xe_svm_range_debug(struct xe_svm_range *range, const char *operation) { } -static inline -int xe_svm_alloc_vram (struct xe_vm *vm,struct xe_tile *tile, -struct xe_svm_range *range, -const struct drm_gpusvm_ctx *ctx) +static inline int +xe_svm_alloc_vram (struct xe_tile *tile,struct xe_svm_range *range, +const struct drm_gpusvm_ctx *ctx) {return -EOPNOTSUPP ; }
drivers/gpu/drm/xe/xe_tile.h +11 -0
@@ -16,4 +16,15 @@ int xe_tile_init(struct xe_tile *tile);void xe_tile_migrate_wait (struct xe_tile *tile); +#if IS_ENABLED (CONFIG_DRM_XE_PAGEMAP ) +static inline struct drm_pagemap *xe_tile_local_pagemap (struct xe_tile *tile) +{ +return &tile->mem.vram.dpagemap; +} +#else +static inline struct drm_pagemap *xe_tile_local_pagemap (struct xe_tile *tile) +{ +return NULL ; +} +#endif #endif
drivers/gpu/drm/xe/xe_vm.c +1 -1
@@ -2913,7 +2913,7 @@ static int prefetch_ranges(struct xe_vm *vm, struct xe_vma_op *op)if (xe_svm_range_needs_migrate_to_vram (svm_range, vma, region)) { tile = &vm->xe->tiles[region_to_mem_type[region] -XE_PL_VRAM0 ]; - err =xe_svm_alloc_vram (vm, tile, svm_range, &ctx); + err =xe_svm_alloc_vram (tile, svm_range, &ctx);if (err) {drm_dbg (&vm->xe->drm,"VRAM allocation failed, retry from userspace, asid=%u, gpusvm=%p, errno=%pe\n" , vm->usm.asid, &vm->svm.gpusvm,ERR_PTR (err));
Signature · 1 line
2.49.0
Re: [PATCH v3 3/3] drm/xe: Implement and use the drm_pagemap populate_mm op
Quoted 180 lines · Thomas Hellström
Add runtime PM since we might call
populate_mmon a foreign device.v3:
- Fix a kerneldoc failure (Matt Brost)
- Revert the bo type change from device to kernel (Matt Brost)Signed-off-by Thomas Hellström <[email protected]>--- drivers/gpu/drm/drm_pagemap.c | 1 + drivers/gpu/drm/xe/xe_svm.c | 101 ++++++++++++++++++++-------------- drivers/gpu/drm/xe/xe_svm.h | 10 ++-- drivers/gpu/drm/xe/xe_tile.h | 11 ++++ drivers/gpu/drm/xe/xe_vm.c | 2 +- 5 files changed, 76 insertions(+), 49 deletions(-)drivers/gpu/drm/drm_pagemap.c +1
-0@@ -829,3 +829,4 @@ int drm_pagemap_populate_mm(struct drm_pagemap *dpagemap,return err; } +EXPORT_SYMBOL (drm_pagemap_populate_mm);drivers/gpu/drm/xe/xe_svm.c +40
-41@@ -3,13 +3,17 @@* Copyright © 2024 Intel Corporation */ +#include <drm/drm_drv.h> +#include "xe_bo.h" #include "xe_gt_stats.h" #include "xe_gt_tlb_invalidation.h" #include "xe_migrate.h" #include "xe_module.h" +#include "xe_pm.h" #include "xe_pt.h" #include "xe_svm.h" +#include "xe_tile.h" #include "xe_ttm_vram_mgr.h" #include "xe_vm.h" #include "xe_vm_types.h" @@ -525,8 +529,10 @@ static struct xe_bo *to_xe_bo(struct drm_pagemap_devmem *devmem_allocation)static void xe_svm_devmem_release (struct drm_pagemap_devmem *devmem_allocation) {struct xe_bo *bo =to_xe_bo (devmem_allocation); +struct xe_device *xe =xe_bo_device (bo);xe_bo_put_async (bo); +xe_pm_runtime_put (xe); }static u64 block_offset_to_pfn (struct xe_vram_region *vr,u64 offset)@@ -720,76 +726,63 @@ static struct xe_vram_region *tile_to_vr(struct xe_tile *tile)return &tile->mem.vram; } -/** -* xe_svm_alloc_vram()- Allocate device memory pages for range, -* migrating existing data. -* @vm: The VM. -* @tile: tile to allocate vram from -* @range: SVM range -* @ctx: DRM GPU SVM context -* -* Return: 0 on success, error code on failure. -*/ -int xe_svm_alloc_vram (struct xe_vm *vm,struct xe_tile *tile, -struct xe_svm_range *range, -const struct drm_gpusvm_ctx *ctx) +static int xe_drm_pagemap_populate_mm (struct drm_pagemap *dpagemap, +unsigned long start,unsigned long end, +struct mm_struct *mm, +unsigned long timeslice_ms) { -struct mm_struct *mm = vm->svm.gpusvm.mm; +struct xe_tile *tile =container_of (dpagemap,typeof (*tile), mem.vram.dpagemap); +struct xe_device *xe =tile_to_xe (tile); +struct device *dev = xe->drm.dev;struct xe_vram_region *vr =tile_to_vr (tile);struct drm_buddy_block *block;struct list_head *blocks;struct xe_bo *bo; -ktime_t end =0 ; -int err; +ktime_t time_end =0 ; +int err, idx; -if (!range->base.flags.migrate_devmem) -return -EINVAL ; - -range_debug (range,"ALLOCATE VRAM" ); +if (!drm_dev_enter (&xe->drm, &idx)) +return -ENODEV ; -if (!mmget_not_zero (mm)) -return -EFAULT ; -mmap_read_lock (mm); +xe_pm_runtime_get (xe); -retry: - bo =xe_bo_create_locked (tile_to_xe (tile),NULL ,NULL , -xe_svm_range_size (range), + retry: + bo =xe_bo_create_locked (tile_to_xe (tile),NULL ,NULL , end - start, ttm_bo_type_device,XE_BO_FLAG_VRAM_IF_DGFX (tile) |XE_BO_FLAG_CPU_ADDR_MIRROR );if (IS_ERR (bo)) { err =PTR_ERR (bo); -if (xe_vm_validate_should_retry (NULL , err, &end)) +if (xe_vm_validate_should_retry (NULL , err, &time_end)) goto retry; - goto unlock; + goto out_pm_put; } -drm_pagemap_devmem_init (&bo->devmem_allocation, - vm->xe->drm.dev, mm, +drm_pagemap_devmem_init (&bo->devmem_allocation, dev, mm, &dpagemap_devmem_ops, &tile->mem.vram.dpagemap, -xe_svm_range_size (range)); + end - start); blocks = &to_xe_ttm_vram_mgr_resource (bo->ttm.resource)->blocks;list_for_each_entry (block, blocks, link) block->private = vr;xe_bo_get (bo); - err =drm_pagemap_migrate_to_devmem (&bo->devmem_allocation, - mm, -xe_svm_range_start (range), -xe_svm_range_end (range), - ctx->timeslice_ms, -xe_svm_devm_owner (vm->xe)); + +/* Ensure the device has a pm ref while there are device pages active. */ +xe_pm_runtime_get_noresume (xe); + err =drm_pagemap_migrate_to_devmem (&bo->devmem_allocation, mm, + start, end, timeslice_ms, +xe_svm_devm_owner (xe));if (err)xe_svm_devmem_release (&bo->devmem_allocation);xe_bo_unlock (bo);xe_bo_put (bo); -unlock: -mmap_read_unlock (mm); -mmput (mm); +out_pm_put: +xe_pm_runtime_put (xe); +drm_dev_exit (idx);return err; }@@ -898,7 +891,7 @@ int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,if (--migrate_try_count >=0 &&xe_svm_range_needs_migrate_to_vram (range, vma,IS_DGFX (vm->xe))) { - err =xe_svm_alloc_vram (vm, tile, range, &ctx); + err =xe_svm_alloc_vram (tile, range, &ctx); ctx.timeslice_ms <<=1 ;/* Double timeslice if we have to retry */ if (err) {if (migrate_try_count || !ctx.devmem_only) {@@ -1054,6 +1047,29 @@ int xe_svm_range_get_pages(struct xe_vm *vm, struct xe_svm_range *range,#if IS_ENABLED (CONFIG_DRM_XE_PAGEMAP ) +/** +* xe_svm_alloc_vram()- Allocate device memory pages for range, +* migrating existing data. +* @tile: tile to allocate vram from +* @range: SVM range +* @ctx: DRM GPU SVM context
drivers/gpu/drm/xe/xe_svm.c int xe_svm_range_get_pages(struct xe_vm *vm, struct xe_svm_range *range, · line 1065
Quoted 10 lines
drivers/gpu/drm/xe/xe_svm.c +10
-0+* +* Return: 0 on success, error code on failure. +*/ +int xe_svm_alloc_vram (struct xe_tile *tile,struct xe_svm_range *range, +const struct drm_gpusvm_ctx *ctx) +{ +struct drm_pagemap *dpagemap; + +range_debug (range,"ALLOCATE VRAM" ); +
I thought I had suggested a check here on range->base.flags.migrate_devmem, that seems to have missed.
Otherwise patch LGTM.
With a check (or an assert) on range->base.flags.migrate_devmem here:
Matt
Quoted 91 lines
drivers/gpu/drm/xe/xe_svm.c +9
-1+ dpagemap =xe_tile_local_pagemap (tile); +return drm_pagemap_populate_mm (dpagemap,xe_svm_range_start (range), +xe_svm_range_end (range), + range->base.gpusvm->mm, + ctx->timeslice_ms); +} + static struct drm_pagemap_device_addr xe_drm_pagemap_device_map(struct drm_pagemap *dpagemap, struct device *dev,@@ -1078,6 +1094,7 @@ xe_drm_pagemap_device_map(struct drm_pagemap *dpagemap,static const struct drm_pagemap_ops xe_drm_pagemap_ops = { .device_map = xe_drm_pagemap_device_map, + .populate_mm = xe_drm_pagemap_populate_mm, };/** @@ -1130,7 +1147,7 @@ int xe_devm_add(struct xe_tile *tile, struct xe_vram_region *vr) return 0; }#else -int xe_svm_alloc_vram (struct xe_vm *vm,struct xe_tile *tile, +int xe_svm_alloc_vram (struct xe_tile *tile,struct xe_svm_range *range,const struct drm_gpusvm_ctx *ctx) {drivers/gpu/drm/xe/xe_svm.h +4
-6@@ -70,8 +70,7 @@ int xe_svm_bo_evict(struct xe_bo *bo);void xe_svm_range_debug (struct xe_svm_range *range,const char *operation); -int xe_svm_alloc_vram (struct xe_vm *vm,struct xe_tile *tile, -struct xe_svm_range *range, +int xe_svm_alloc_vram (struct xe_tile *tile,struct xe_svm_range *range,const struct drm_gpusvm_ctx *ctx); struct xe_svm_range *xe_svm_range_find_or_insert(struct xe_vm *vm, u64 addr,@@ -237,10 +236,9 @@ void xe_svm_range_debug(struct xe_svm_range *range, const char *operation) { } -static inline -int xe_svm_alloc_vram (struct xe_vm *vm,struct xe_tile *tile, -struct xe_svm_range *range, -const struct drm_gpusvm_ctx *ctx) +static inline int +xe_svm_alloc_vram (struct xe_tile *tile,struct xe_svm_range *range, +const struct drm_gpusvm_ctx *ctx) {return -EOPNOTSUPP ; }drivers/gpu/drm/xe/xe_tile.h +11
-0@@ -16,4 +16,15 @@ int xe_tile_init(struct xe_tile *tile);void xe_tile_migrate_wait (struct xe_tile *tile); +#if IS_ENABLED (CONFIG_DRM_XE_PAGEMAP ) +static inline struct drm_pagemap *xe_tile_local_pagemap (struct xe_tile *tile) +{ +return &tile->mem.vram.dpagemap; +} +#else +static inline struct drm_pagemap *xe_tile_local_pagemap (struct xe_tile *tile) +{ +return NULL ; +} +#endif #endif drivers/gpu/drm/xe/xe_vm.c +1
-1@@ -2913,7 +2913,7 @@ static int prefetch_ranges(struct xe_vm *vm, struct xe_vma_op *op)if (xe_svm_range_needs_migrate_to_vram (svm_range, vma, region)) { tile = &vm->xe->tiles[region_to_mem_type[region] -XE_PL_VRAM0 ]; - err =xe_svm_alloc_vram (vm, tile, svm_range, &ctx); + err =xe_svm_alloc_vram (tile, svm_range, &ctx);if (err) {drm_dbg (&vm->xe->drm,"VRAM allocation failed, retry from userspace, asid=%u, gpusvm=%p, errno=%pe\n" , vm->usm.asid, &vm->svm.gpusvm,ERR_PTR (err));Signature · 1 line
2.49.0
Thomas Hellström
maintainerRe: [PATCH v3 3/3] drm/xe: Implement and use the drm_pagemap populate_mm op
Quoted 214 lines · Matthew Brost
Quoted 213 lines · Thomas Hellström
Add runtime PM since we might call
populate_mmon a foreign device.v3:
- Fix a kerneldoc failure (Matt Brost)
- Revert the bo type change from device to kernel (Matt Brost)Signed-off-by Thomas Hellström <[email protected]>---
drivers/gpu/drm/drm_pagemap.c | 1 + drivers/gpu/drm/xe/xe_svm.c | 101 ++++++++++++++++++++----------
----
drivers/gpu/drm/xe/xe_svm.h | 10 ++-- drivers/gpu/drm/xe/xe_tile.h | 11 ++++ drivers/gpu/drm/xe/xe_vm.c | 2 +-5 files changed, 76 insertions(+), 49 deletions(-)drivers/gpu/drm/drm_pagemap.c +0
-0@@ -829,3 +829,4 @@ int drm_pagemap_populate_mm(struct drm_pagemap *dpagemap,
return err;
}drivers/gpu/drm/drm_pagemap.c +1
-0+EXPORT_SYMBOL (drm_pagemap_populate_mm);drivers/gpu/drm/xe/xe_svm.c +0
-0@@ -3,13 +3,17 @@* Copyright © 2024 Intel Corporation */ drivers/gpu/drm/xe/xe_svm.c +2
-0+#include <drm/drm_drv.h> +#include "
xe_bo.h"
#include "xe_gt_stats.h"
#include "xe_gt_tlb_invalidation.h"
#include "xe_migrate.h"
#include "xe_module.h"
+#include "xe_pm.h"
#include "xe_pt.h"
#include "xe_svm.h"
+#include "xe_tile.h"
#include "xe_ttm_vram_mgr.h"
#include "xe_vm.h"
#include "xe_vm_types.h"drivers/gpu/drm/xe/xe_svm.c +0
-0@@ -525,8 +529,10 @@ static struct xe_bo *to_xe_bo(struct drm_pagemap_devmem *devmem_allocation)static void
xe_svm_devmem_release(struct drm_pagemap_devmem *devmem_allocation)
{
structxe_bo*bo =to_xe_bo(devmem_allocation);drivers/gpu/drm/xe/xe_svm.c +1
-0+struct xe_device *xe =xe_bo_device (bo);
xe_bo_put_async(bo);drivers/gpu/drm/xe/xe_svm.c +1
-0+xe_pm_runtime_put (xe);}
static u64block_offset_to_pfn(struct xe_vram_region *vr, u64 offset)drivers/gpu/drm/xe/xe_svm.c +0
-0@@ -720,76 +726,63 @@ static struct xe_vram_region *tile_to_vr(struct xe_tile *tile) return &tile->mem.vram; }drivers/gpu/drm/xe/xe_svm.c +1
-13-/** -* xe_svm_alloc_vram()- Allocate device memory pages for range, -* migrating existing data. -* @vm: The VM. -* @tile: tile to allocate vram from -* @range: SVM range -* @ctx: DRM GPU SVM context -* -* Return: 0 on success, error code on failure. -*/ -int xe_svm_alloc_vram (struct xe_vm *vm,struct xe_tile *tile, - struct xe_svm_range *range, - const struct drm_gpusvm_ctx *ctx) +static int xe_drm_pagemap_populate_mm(struct drm_pagemap*dpagemap,
drivers/gpu/drm/xe/xe_svm.c +1
-0+ unsigned long start,unsigned long end,
drivers/gpu/drm/xe/xe_svm.c +2
-0+ struct mm_struct *mm, + unsigned long timeslice_ms){
drivers/gpu/drm/xe/xe_svm.c +1
-1-struct mm_struct *mm = vm->svm.gpusvm.mm; + struct xe_tile *tile = container_of(dpagemap,
typeof(*tile), mem.vram.dpagemap);drivers/gpu/drm/xe/xe_svm.c +2
-0+struct xe_device *xe =tile_to_xe (tile); +struct device *dev = xe->drm.dev;struct
xe_vram_region*vr =tile_to_vr(tile); structdrm_buddy_block*block;
structlist_head*blocks;
structxe_bo*bo;drivers/gpu/drm/xe/xe_svm.c +2
-2-ktime_t end =0 ; -int err; +ktime_t time_end =0 ; +int err, idx;drivers/gpu/drm/xe/xe_svm.c +2
-4-if (!range->base.flags.migrate_devmem) -return -EINVAL ; - -range_debug (range,"ALLOCATE VRAM" ); +if (!drm_dev_enter (&xe->drm, &idx)) +return -ENODEV ;drivers/gpu/drm/xe/xe_svm.c +1
-3-if (!mmget_not_zero (mm)) -return -EFAULT ; -mmap_read_lock (mm); +xe_pm_runtime_get (xe);drivers/gpu/drm/xe/xe_svm.c +2
-4-retry: - bo = xe_bo_create_locked(tile_to_xe(tile), NULL, NULL, - xe_svm_range_size(range), + retry: + bo = xe_bo_create_locked(tile_to_xe(tile), NULL, NULL, end - start,
ttm_bo_type_device,XE_BO_FLAG_VRAM_IF_DGFX(tile)|XE_BO_FLAG_CPU_ADDR_MIRROR); if (IS_ERR(bo)) {
err =PTR_ERR(bo);drivers/gpu/drm/xe/xe_svm.c +1
-1-if (xe_vm_validate_should_retry (NULL , err, &end)) + if (xe_vm_validate_should_retry(NULL, err,
&time_end))
goto retry;drivers/gpu/drm/xe/xe_svm.c +1
-1- goto unlock; + goto out_pm_put;}
drivers/gpu/drm/xe/xe_svm.c +1
-2- drm_pagemap_devmem_init(&bo->devmem_allocation, - vm->xe->drm.dev, mm, + drm_pagemap_devmem_init(&bo->devmem_allocation, dev, mm,
&dpagemap_devmem_ops,&tile->mem.vram.dpagemap,drivers/gpu/drm/xe/xe_svm.c +1
-1- xe_svm_range_size(range)); + end - start);
blocks = &to_xe_ttm_vram_mgr_resource(bo->ttm.resource)-Quoted 1 line
blocks;
list_for_each_entry(block, blocks, link)
block->private= vr;
xe_bo_get(bo);drivers/gpu/drm/xe/xe_svm.c +0
-1- err = drm_pagemap_migrate_to_devmem(&bo-Quoted 1 line
devmem_allocation,drivers/gpu/drm/xe/xe_svm.c +0
-2- mm, -
xe_svm_range_start(range),drivers/gpu/drm/xe/xe_svm.c +0
-1-
xe_svm_range_end(range),drivers/gpu/drm/xe/xe_svm.c +0
-2- ctx->timeslice_ms, - xe_svm_devm_owner(vm-Quoted 1 line
xe));
drivers/gpu/drm/xe/xe_svm.c +2
-0+ +/* Ensure the device has a pm ref while there are device pages active. */
drivers/gpu/drm/xe/xe_svm.c +2
-0+xe_pm_runtime_get_noresume (xe); + err = drm_pagemap_migrate_to_devmem(&bo-Quoted 1 line
devmem_allocation, mm,drivers/gpu/drm/xe/xe_svm.c +1
-0+ start, end,
timeslice_ms,drivers/gpu/drm/xe/xe_svm.c +1
-0+
xe_svm_devm_owner(xe));
if (err)xe_svm_devmem_release(&bo->devmem_allocation);
xe_bo_unlock(bo);
xe_bo_put(bo);drivers/gpu/drm/xe/xe_svm.c +3
-3-unlock: -mmap_read_unlock (mm); -mmput (mm); +out_pm_put: +xe_pm_runtime_put (xe); +drm_dev_exit (idx);
return err;
}drivers/gpu/drm/xe/xe_svm.c +0
-0@@ -898,7 +891,7 @@ int xe_svm_handle_pagefault(struct xe_vm *vm, struct xe_vma *vma,
if (--migrate_try_count >= 0 &&xe_svm_range_needs_migrate_to_vram(range, vma, IS_DGFX(vm->xe))) {drivers/gpu/drm/xe/xe_svm.c +1
-1- err =xe_svm_alloc_vram (vm, tile, range, &ctx); + err =xe_svm_alloc_vram (tile, range, &ctx);ctx.timeslice_ms <<= 1; /* Double
timeslice if we have to retry */
if (err) {
if (migrate_try_count|| !ctx.devmem_only) {drivers/gpu/drm/xe/xe_svm.c +0
-0@@ -1054,6 +1047,29 @@ int xe_svm_range_get_pages(struct xe_vm *vm, struct xe_svm_range *range,
#ifIS_ENABLED(CONFIG_DRM_XE_PAGEMAP)drivers/gpu/drm/xe/xe_svm.c +10
-0+/** +* xe_svm_alloc_vram()- Allocate device memory pages for range, +* migrating existing data. +* @tile: tile to allocate vram from +* @range: SVM range +* @ctx: DRM GPU SVM context +* +* Return: 0 on success, error code on failure. +*/ +int xe_svm_alloc_vram(struct xe_tile *tile, struct xe_svm_range*range,
drivers/gpu/drm/xe/xe_svm.c +1
-0+ const struct drm_gpusvm_ctx *ctx)
Quoted 8 lines
drivers/gpu/drm/xe/xe_svm.c int xe_svm_range_get_pages(struct xe_vm *vm, struct xe_svm_range *range, · line 1065
Quoted 5 lines
drivers/gpu/drm/xe/xe_svm.c +5
-0+{ +struct drm_pagemap *dpagemap; + +range_debug (range,"ALLOCATE VRAM" ); +I thought I had suggested a check here on
range->base.flags.migrate_devmem, that seems to have missed.
You're right. I missed to fix that. will update.
Thanks for reviewing.
/Thomas
Quoted 120 lines
Otherwise patch LGTM.
With a check (or an assert) on
range->base.flags.migrate_devmemhere:Reviewed-by Matthew Brost <[email protected]>Matt
Quoted 112 lines
drivers/gpu/drm/xe/xe_svm.c +2
-0+ dpagemap =xe_tile_local_pagemap (tile); + return drm_pagemap_populate_mm(dpagemap,
xe_svm_range_start(range),drivers/gpu/drm/xe/xe_svm.c +5
-0+ xe_svm_range_end(range), + range->base.gpusvm->mm, + ctx->timeslice_ms); +} +static struct
drm_pagemap_device_addrxe_drm_pagemap_device_map(structdrm_pagemap*dpagemap, struct device *dev,drivers/gpu/drm/xe/xe_svm.c +0
-0@@ -1078,6 +1094,7 @@ xe_drm_pagemap_device_map(struct drm_pagemap *dpagemap,
static const structdrm_pagemap_opsxe_drm_pagemap_ops= { .device_map =xe_drm_pagemap_device_map,drivers/gpu/drm/xe/xe_svm.c +1
-0+ .populate_mm = xe_drm_pagemap_populate_mm,};
/**drivers/gpu/drm/xe/xe_svm.c +1
-1@@ -1130,7 +1147,7 @@ int xe_devm_add(struct xe_tile *tile, struct xe_vram_region *vr) return 0; } #else -int xe_svm_alloc_vram (struct xe_vm *vm,struct xe_tile *tile, +int xe_svm_alloc_vram (struct xe_tile *tile, struct xe_svm_range *range, const struct drm_gpusvm_ctx *ctx){
drivers/gpu/drm/xe/xe_svm.h +0
-0@@ -70,8 +70,7 @@ int xe_svm_bo_evict(struct xe_bo *bo);
voidxe_svm_range_debug(struct xe_svm_range *range, const char *operation);drivers/gpu/drm/xe/xe_svm.h +1
-2-int xe_svm_alloc_vram(struct xe_vm *vm, struct xe_tile *tile, - struct xe_svm_range *range, +int xe_svm_alloc_vram(struct xe_tile *tile, struct xe_svm_range*range,
const structdrm_gpusvm_ctx*ctx);
structxe_svm_range*xe_svm_range_find_or_insert(structxe_vm*vm, u64 addr,drivers/gpu/drm/xe/xe_svm.h +0
-0@@ -237,10 +236,9 @@ void xe_svm_range_debug(struct xe_svm_range *range, const char *operation) { }drivers/gpu/drm/xe/xe_svm.h +2
-4-static inline -int xe_svm_alloc_vram (struct xe_vm *vm,struct xe_tile *tile, - struct xe_svm_range *range, - const struct drm_gpusvm_ctx *ctx) +static inline int +xe_svm_alloc_vram(struct xe_tile *tile, struct xe_svm_range*range,
drivers/gpu/drm/xe/xe_svm.h +1
-0+ const struct drm_gpusvm_ctx *ctx){
return-EOPNOTSUPP;
}drivers/gpu/drm/xe/xe_tile.h +0
-0@@ -16,4 +16,15 @@ int xe_tile_init(struct xe_tile *tile);
voidxe_tile_migrate_wait(struct xe_tile *tile);drivers/gpu/drm/xe/xe_tile.h +2
-0+#if IS_ENABLED (CONFIG_DRM_XE_PAGEMAP ) +static inline struct drm_pagemap *xe_tile_local_pagemap(struct
xe_tile*tile)drivers/gpu/drm/xe/xe_tile.h +5
-0+{ +return &tile->mem.vram.dpagemap; +} +#else +static inline struct drm_pagemap *xe_tile_local_pagemap(struct
xe_tile*tile)drivers/gpu/drm/xe/xe_tile.h +4
-0+{ +return NULL ; +} +#endif #endif
drivers/gpu/drm/xe/xe_vm.c +0
-0@@ -2913,7 +2913,7 @@ static int prefetch_ranges(struct xe_vm *vm, struct xe_vma_op *op)
if (xe_svm_range_needs_migrate_to_vram(svm_range, vma, region)) {
tile = &vm->xe-Quoted 1 line
tiles[region_to_mem_type[region] -
XE_PL_VRAM0];drivers/gpu/drm/xe/xe_vm.c +0
-1- err = xe_svm_alloc_vram(vm, tile,
svm_range, &ctx);drivers/gpu/drm/xe/xe_vm.c +1
-0+ err = xe_svm_alloc_vram(tile, svm_range,&ctx);
if (err) { drm_dbg(&vm->xe->drm, "VRAM allocation failed, retry from userspace, asid=%u, gpusvm=%p, errno=%pe\n",
vm->usm.asid, &vm-Quoted 1 line
svm.gpusvm,
ERR_PTR(err));Signature · 1 line
2.49.0
Patchwork
CI✗ CI.checkpatch: warning for drm/gpusvm, drm/pagemap, drm/xe: Restructure migration in preparation for multi-device (rev3)
In reply to Thomas Hellström== Series Details ==
== Series Details ==
Series: drm/gpusvm, drm/pagemap, drm/xe: Restructure migration in preparation for multi-device (rev3)
URL : https://patchwork.freedesktop.org/series/147902/
State : warning
== Summary ==
+ KERNEL=/kernel
+ git clone https://gitlab.freedesktop.org/drm/maintainer-tools mt Cloning into 'mt'...
warning: redirecting to https://gitlab.freedesktop.org/drm/maintainer-tools.git/
+ git -C mt rev-list -n1 origin/master 202708c00696422fd217223bb679a353a5936e23
+ cd /kernel
+ git config --global --add safe.directory /kernel
+ git log -n1
commit 3dd17d3f125ecdf4c1940f3691ecc96f146eea95
Author: Thomas Hellström <[email protected]> Date: Fri Jun 13 16:02:19 2025 +0200
drm/xe: Implement and use the drm_pagemap populate_mm op
Add runtime PM since we might call populate_mm on a foreign device.
v3:
- Fix a kerneldoc failure (Matt Brost)
- Revert the bo type change from device to kernel (Matt Brost)
+ /mt/dim checkpatch 4d016d6e602638e0ebc3895331224e057508c07a drm-intel 4b5f765f273b drm/gpusvm, drm/pagemap: Move migration functionality to drm_pagemap -:944: WARNING:FILE_PATH_CHANGES: added, moved or deleted file(s), does MAINTAINERS need updating? #944:
new file mode 100644
-:1747: WARNING:CONFIG_DESCRIPTION: please write a help paragraph that fully describes the config symbol with at least 4 lines #1747: FILE: drivers/gpu/drm/xe/Kconfig:89:
+config DRM_XE_PAGEMAP
+ bool "Enable device memory pool for SVM"
depends on DRM_XE_GPUSVM
select GET_FREE_REGION
default y
help
+ Disable this option only if you don't want to expose local device + Disable this option only if you don't want to expose local device + Disable this option only if you don't want to expose local device + memory for SVM. Will reduce KMD memory footprint when disabled.
+
+ If in doubut say "Y".
config DRM_XE_FORCE_PROBE
string "Force probe xe for selected Intel hardware IDs"
-:2169: ERROR:NO_AUTHOR_SIGN_OFF: Missing Signed-off-by: line by nominal patch author 'Matthew Brost <[email protected]>'
total: 1 errors, 2 warnings, 0 checks, 2055 lines checked 006871746ffd drm/pagemap: Add a populate_mm op 3dd17d3f125e drm/xe: Implement and use the drm_pagemap populate_mm op
Patchwork
CI✓ CI.KUnit: success for drm/gpusvm, drm/pagemap, drm/xe: Restructure migration in preparation for multi-device (rev3)
In reply to Thomas Hellström== Series Details ==
== Series Details ==
Series: drm/gpusvm, drm/pagemap, drm/xe: Restructure migration in preparation for multi-device (rev3)
URL : https://patchwork.freedesktop.org/series/147902/
State : success
== Summary ==
+ trap cleanup EXIT
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/xe/.kunitconfig [15:56:45] Configuring KUnit Kernel ...
Generating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig [15:56:49] Building KUnit Kernel ... Populating config with: $ make ARCH=um O=.kunit olddefconfig Building with: $ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48 [15:57:16] Starting KUnit Kernel (1/1)... [15:57:16] ============================================================ Running tests with: $ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt [15:57:17] ================== guc_buf (11 subtests) =================== [15:57:17] [PASSED] test_smallest [15:57:17] [PASSED] test_largest [15:57:17] [PASSED] test_granular [15:57:17] [PASSED] test_unique [15:57:17] [PASSED] test_overlap [15:57:17] [PASSED] test_reusable [15:57:17] [PASSED] test_too_big [15:57:17] [PASSED] test_flush [15:57:17] [PASSED] test_lookup [15:57:17] [PASSED] test_data [15:57:17] [PASSED] test_class [15:57:17] ===================== [PASSED] guc_buf ===================== [15:57:17] =================== guc_dbm (7 subtests) =================== [15:57:17] [PASSED] test_empty [15:57:17] [PASSED] test_default [15:57:17] ======================== test_size ======================== [15:57:17] [PASSED] 4 [15:57:17] [PASSED] 8 [15:57:17] [PASSED] 32 [15:57:17] [PASSED] 256 [15:57:17] ==================== [PASSED] test_size ==================== [15:57:17] ======================= test_reuse ======================== [15:57:17] [PASSED] 4 [15:57:17] [PASSED] 8 [15:57:17] [PASSED] 32 [15:57:17] [PASSED] 256 [15:57:17] =================== [PASSED] test_reuse ==================== [15:57:17] =================== test_range_overlap ==================== [15:57:17] [PASSED] 4 [15:57:17] [PASSED] 8 [15:57:17] [PASSED] 32 [15:57:17] [PASSED] 256 [15:57:17] =============== [PASSED] test_range_overlap ================ [15:57:17] =================== test_range_compact ==================== [15:57:17] [PASSED] 4 [15:57:17] [PASSED] 8 [15:57:17] [PASSED] 32 [15:57:17] [PASSED] 256 [15:57:17] =============== [PASSED] test_range_compact ================ [15:57:17] ==================== test_range_spare ===================== [15:57:17] [PASSED] 4 [15:57:17] [PASSED] 8 [15:57:17] [PASSED] 32 [15:57:17] [PASSED] 256 [15:57:17] ================ [PASSED] test_range_spare ================= [15:57:17] ===================== [PASSED] guc_dbm ===================== [15:57:17] =================== guc_idm (6 subtests) =================== [15:57:17] [PASSED] bad_init [15:57:17] [PASSED] no_init [15:57:17] [PASSED] init_fini [15:57:17] [PASSED] check_used [15:57:17] [PASSED] check_quota [15:57:17] [PASSED] check_all [15:57:17] ===================== [PASSED] guc_idm ===================== [15:57:17] ================== no_relay (3 subtests) =================== [15:57:17] [PASSED] xe_drops_guc2pf_if_not_ready [15:57:17] [PASSED] xe_drops_guc2vf_if_not_ready [15:57:17] [PASSED] xe_rejects_send_if_not_ready [15:57:17] ==================== [PASSED] no_relay ===================== [15:57:17] ================== pf_relay (14 subtests) ================== [15:57:17] [PASSED] pf_rejects_guc2pf_too_short [15:57:17] [PASSED] pf_rejects_guc2pf_too_long [15:57:17] [PASSED] pf_rejects_guc2pf_no_payload [15:57:17] [PASSED] pf_fails_no_payload [15:57:17] [PASSED] pf_fails_bad_origin [15:57:17] [PASSED] pf_fails_bad_type [15:57:17] [PASSED] pf_txn_reports_error [15:57:17] [PASSED] pf_txn_sends_pf2guc [15:57:17] [PASSED] pf_sends_pf2guc [15:57:17] [SKIPPED] pf_loopback_nop [15:57:17] [SKIPPED] pf_loopback_echo [15:57:17] [SKIPPED] pf_loopback_fail [15:57:17] [SKIPPED] pf_loopback_busy [15:57:17] [SKIPPED] pf_loopback_retry [15:57:17] ==================== [PASSED] pf_relay ===================== [15:57:17] ================== vf_relay (3 subtests) =================== [15:57:17] [PASSED] vf_rejects_guc2vf_too_short [15:57:17] [PASSED] vf_rejects_guc2vf_too_long [15:57:17] [PASSED] vf_rejects_guc2vf_no_payload [15:57:17] ==================== [PASSED] vf_relay ===================== [15:57:17] ================= pf_service (11 subtests) ================= [15:57:17] [PASSED] pf_negotiate_any [15:57:17] [PASSED] pf_negotiate_base_match [15:57:17] [PASSED] pf_negotiate_base_newer [15:57:17] [PASSED] pf_negotiate_base_next [15:57:17] [SKIPPED] pf_negotiate_base_older [15:57:17] [PASSED] pf_negotiate_base_prev [15:57:17] [PASSED] pf_negotiate_latest_match [15:57:17] [PASSED] pf_negotiate_latest_newer [15:57:17] [PASSED] pf_negotiate_latest_next [15:57:17] [SKIPPED] pf_negotiate_latest_older [15:57:17] [SKIPPED] pf_negotiate_latest_prev [15:57:17] =================== [PASSED] pf_service ==================== [15:57:17] ===================== lmtt (1 subtest) ===================== [15:57:17] ======================== test_ops ========================= [15:57:17] [PASSED] 2-level [15:57:17] [PASSED] multi-level [15:57:17] ==================== [PASSED] test_ops ===================== [15:57:17] ====================== [PASSED] lmtt ======================= [15:57:17] =================== xe_mocs (2 subtests) =================== [15:57:17] ================ xe_live_mocs_kernel_kunit ================ [15:57:17] =========== [SKIPPED] xe_live_mocs_kernel_kunit ============ [15:57:17] ================ xe_live_mocs_reset_kunit ================= [15:57:17] ============ [SKIPPED] xe_live_mocs_reset_kunit ============ [15:57:17] ==================== [SKIPPED] xe_mocs ===================== [15:57:17] ================= xe_migrate (2 subtests) ================== [15:57:17] ================= xe_migrate_sanity_kunit ================= [15:57:17] ============ [SKIPPED] xe_migrate_sanity_kunit ============= [15:57:17] ================== xe_validate_ccs_kunit ================== [15:57:17] ============= [SKIPPED] xe_validate_ccs_kunit ============== [15:57:17] =================== [SKIPPED] xe_migrate =================== [15:57:17] ================== xe_dma_buf (1 subtest) ================== [15:57:17] ==================== xe_dma_buf_kunit ===================== [15:57:17] ================ [SKIPPED] xe_dma_buf_kunit ================ [15:57:17] =================== [SKIPPED] xe_dma_buf =================== [15:57:17] ================= xe_bo_shrink (1 subtest) ================= [15:57:17] =================== xe_bo_shrink_kunit ==================== [15:57:17] =============== [SKIPPED] xe_bo_shrink_kunit =============== [15:57:17] ================== [SKIPPED] xe_bo_shrink ================== [15:57:17] ==================== xe_bo (2 subtests) ==================== [15:57:17] ================== xe_ccs_migrate_kunit =================== [15:57:17] ============== [SKIPPED] xe_ccs_migrate_kunit ============== [15:57:17] ==================== xe_bo_evict_kunit ==================== [15:57:17] =============== [SKIPPED] xe_bo_evict_kunit ================ [15:57:17] ===================== [SKIPPED] xe_bo ====================== [15:57:17] ==================== args (11 subtests) ==================== [15:57:17] [PASSED] count_args_test [15:57:17] [PASSED] call_args_example [15:57:17] [PASSED] call_args_test [15:57:17] [PASSED] drop_first_arg_example [15:57:17] [PASSED] drop_first_arg_test [15:57:17] [PASSED] first_arg_example [15:57:17] [PASSED] first_arg_test [15:57:17] [PASSED] last_arg_example [15:57:17] [PASSED] last_arg_test [15:57:17] [PASSED] pick_arg_example [15:57:17] [PASSED] sep_comma_example [15:57:17] ====================== [PASSED] args ======================= [15:57:17] =================== xe_pci (2 subtests) ==================== [15:57:17] [PASSED] xe_gmdid_graphics_ip [15:57:17] [PASSED] xe_gmdid_media_ip [15:57:17] ===================== [PASSED] xe_pci ====================== [15:57:17] =================== xe_rtp (2 subtests) ==================== [15:57:17] =============== xe_rtp_process_to_sr_tests ================ [15:57:17] [PASSED] coalesce-same-reg [15:57:17] [PASSED] no-match-no-add [15:57:17] [PASSED] match-or [15:57:17] [PASSED] match-or-xfail [15:57:17] [PASSED] no-match-no-add-multiple-rules [15:57:17] [PASSED] two-regs-two-entries [15:57:17] [PASSED] clr-one-set-other [15:57:17] [PASSED] set-field [15:57:17] [PASSED] conflict-duplicate [15:57:17] [PASSED] conflict-not-disjoint stty: 'standard input': Inappropriate ioctl for device [15:57:17] [PASSED] conflict-reg-type [15:57:17] =========== [PASSED] xe_rtp_process_to_sr_tests ============ [15:57:17] ================== xe_rtp_process_tests =================== [15:57:17] [PASSED] active1 [15:57:17] [PASSED] active2 [15:57:17] [PASSED] active-inactive [15:57:17] [PASSED] inactive-active [15:57:17] [PASSED] inactive-1st_or_active-inactive [15:57:17] [PASSED] inactive-2nd_or_active-inactive [15:57:17] [PASSED] inactive-last_or_active-inactive [15:57:17] [PASSED] inactive-no_or_active-inactive [15:57:17] ============== [PASSED] xe_rtp_process_tests =============== [15:57:17] ===================== [PASSED] xe_rtp ====================== [15:57:17] ==================== xe_wa (1 subtest) ===================== [15:57:17] ======================== xe_wa_gt ========================= [15:57:17] [PASSED] TIGERLAKE (B0) [15:57:17] [PASSED] DG1 (A0) [15:57:17] [PASSED] DG1 (B0) [15:57:17] [PASSED] ALDERLAKE_S (A0) [15:57:17] [PASSED] ALDERLAKE_S (B0) [15:57:17] [PASSED] ALDERLAKE_S (C0) [15:57:17] [PASSED] ALDERLAKE_S (D0) [15:57:17] [PASSED] ALDERLAKE_P (A0) [15:57:17] [PASSED] ALDERLAKE_P (B0) [15:57:17] [PASSED] ALDERLAKE_P (C0) [15:57:17] [PASSED] ALDERLAKE_S_RPLS (D0) [15:57:17] [PASSED] ALDERLAKE_P_RPLU (E0) [15:57:17] [PASSED] DG2_G10 (C0) [15:57:17] [PASSED] DG2_G11 (B1) [15:57:17] [PASSED] DG2_G12 (A1) [15:57:17] [PASSED] METEORLAKE (g:A0, m:A0) [15:57:17] [PASSED] METEORLAKE (g:A0, m:A0) [15:57:17] [PASSED] METEORLAKE (g:A0, m:A0) [15:57:17] [PASSED] LUNARLAKE (g:A0, m:A0) [15:57:17] [PASSED] LUNARLAKE (g:B0, m:A0) [15:57:17] [PASSED] BATTLEMAGE (g:A0, m:A1) [15:57:17] ==================== [PASSED] xe_wa_gt ===================== [15:57:17] ====================== [PASSED] xe_wa ====================== [15:57:17] ============================================================ [15:57:17] Testing complete. Ran 133 tests: passed: 117, skipped: 16 [15:57:17] Elapsed time: 31.549s total, 4.146s configuring, 27.086s building, 0.299s running
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/tests/.kunitconfig [15:57:17] Configuring KUnit Kernel ...
Regenerating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig [15:57:19] Building KUnit Kernel ... Populating config with: $ make ARCH=um O=.kunit olddefconfig Building with: $ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48 [15:57:41] Starting KUnit Kernel (1/1)... [15:57:41] ============================================================ Running tests with: $ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt [15:57:41] == drm_test_atomic_get_connector_for_encoder (1 subtest) === [15:57:41] [PASSED] drm_test_drm_atomic_get_connector_for_encoder [15:57:41] ==== [PASSED] drm_test_atomic_get_connector_for_encoder ==== [15:57:41] =========== drm_validate_clone_mode (2 subtests) =========== [15:57:41] ============== drm_test_check_in_clone_mode =============== [15:57:41] [PASSED] in_clone_mode [15:57:41] [PASSED] not_in_clone_mode [15:57:41] ========== [PASSED] drm_test_check_in_clone_mode =========== [15:57:41] =============== drm_test_check_valid_clones =============== [15:57:41] [PASSED] not_in_clone_mode [15:57:41] [PASSED] valid_clone [15:57:41] [PASSED] invalid_clone [15:57:41] =========== [PASSED] drm_test_check_valid_clones =========== [15:57:41] ============= [PASSED] drm_validate_clone_mode ============= [15:57:41] ============= drm_validate_modeset (1 subtest) ============= [15:57:41] [PASSED] drm_test_check_connector_changed_modeset [15:57:41] ============== [PASSED] drm_validate_modeset =============== [15:57:41] ====== drm_test_bridge_get_current_state (2 subtests) ====== [15:57:41] [PASSED] drm_test_drm_bridge_get_current_state_atomic [15:57:41] [PASSED] drm_test_drm_bridge_get_current_state_legacy [15:57:41] ======== [PASSED] drm_test_bridge_get_current_state ======== [15:57:41] ====== drm_test_bridge_helper_reset_crtc (3 subtests) ====== [15:57:41] [PASSED] drm_test_drm_bridge_helper_reset_crtc_atomic [15:57:41] [PASSED] drm_test_drm_bridge_helper_reset_crtc_atomic_disabled [15:57:41] [PASSED] drm_test_drm_bridge_helper_reset_crtc_legacy [15:57:41] ======== [PASSED] drm_test_bridge_helper_reset_crtc ======== [15:57:41] ============== drm_bridge_alloc (2 subtests) =============== [15:57:41] [PASSED] drm_test_drm_bridge_alloc_basic [15:57:41] [PASSED] drm_test_drm_bridge_alloc_get_put [15:57:41] ================ [PASSED] drm_bridge_alloc ================= [15:57:41] ================== drm_buddy (7 subtests) ================== [15:57:41] [PASSED] drm_test_buddy_alloc_limit [15:57:41] [PASSED] drm_test_buddy_alloc_optimistic [15:57:41] [PASSED] drm_test_buddy_alloc_pessimistic [15:57:41] [PASSED] drm_test_buddy_alloc_pathological [15:57:41] [PASSED] drm_test_buddy_alloc_contiguous [15:57:41] [PASSED] drm_test_buddy_alloc_clear [15:57:41] [PASSED] drm_test_buddy_alloc_range_bias [15:57:41] ==================== [PASSED] drm_buddy ==================== [15:57:41] ============= drm_cmdline_parser (40 subtests) ============= [15:57:41] [PASSED] drm_test_cmdline_force_d_only [15:57:41] [PASSED] drm_test_cmdline_force_D_only_dvi [15:57:41] [PASSED] drm_test_cmdline_force_D_only_hdmi [15:57:41] [PASSED] drm_test_cmdline_force_D_only_not_digital [15:57:41] [PASSED] drm_test_cmdline_force_e_only [15:57:41] [PASSED] drm_test_cmdline_res [15:57:41] [PASSED] drm_test_cmdline_res_vesa [15:57:41] [PASSED] drm_test_cmdline_res_vesa_rblank [15:57:41] [PASSED] drm_test_cmdline_res_rblank [15:57:41] [PASSED] drm_test_cmdline_res_bpp [15:57:41] [PASSED] drm_test_cmdline_res_refresh [15:57:41] [PASSED] drm_test_cmdline_res_bpp_refresh [15:57:41] [PASSED] drm_test_cmdline_res_bpp_refresh_interlaced [15:57:41] [PASSED] drm_test_cmdline_res_bpp_refresh_margins [15:57:41] [PASSED] drm_test_cmdline_res_bpp_refresh_force_off [15:57:41] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on [15:57:41] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on_analog [15:57:41] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on_digital [15:57:41] [PASSED] drm_test_cmdline_res_bpp_refresh_interlaced_margins_force_on [15:57:41] [PASSED] drm_test_cmdline_res_margins_force_on [15:57:41] [PASSED] drm_test_cmdline_res_vesa_margins [15:57:41] [PASSED] drm_test_cmdline_name [15:57:41] [PASSED] drm_test_cmdline_name_bpp [15:57:41] [PASSED] drm_test_cmdline_name_option [15:57:41] [PASSED] drm_test_cmdline_name_bpp_option [15:57:41] [PASSED] drm_test_cmdline_rotate_0 [15:57:41] [PASSED] drm_test_cmdline_rotate_90 [15:57:41] [PASSED] drm_test_cmdline_rotate_180 [15:57:41] [PASSED] drm_test_cmdline_rotate_270 [15:57:41] [PASSED] drm_test_cmdline_hmirror [15:57:41] [PASSED] drm_test_cmdline_vmirror [15:57:41] [PASSED] drm_test_cmdline_margin_options [15:57:41] [PASSED] drm_test_cmdline_multiple_options [15:57:41] [PASSED] drm_test_cmdline_bpp_extra_and_option [15:57:41] [PASSED] drm_test_cmdline_extra_and_option [15:57:41] [PASSED] drm_test_cmdline_freestanding_options [15:57:41] [PASSED] drm_test_cmdline_freestanding_force_e_and_options [15:57:41] [PASSED] drm_test_cmdline_panel_orientation [15:57:41] ================ drm_test_cmdline_invalid ================= [15:57:41] [PASSED] margin_only [15:57:41] [PASSED] interlace_only [15:57:41] [PASSED] res_missing_x [15:57:41] [PASSED] res_missing_y [15:57:41] [PASSED] res_bad_y [15:57:41] [PASSED] res_missing_y_bpp [15:57:41] [PASSED] res_bad_bpp [15:57:41] [PASSED] res_bad_refresh [15:57:41] [PASSED] res_bpp_refresh_force_on_off [15:57:41] [PASSED] res_invalid_mode [15:57:41] [PASSED] res_bpp_wrong_place_mode [15:57:41] [PASSED] name_bpp_refresh [15:57:41] [PASSED] name_refresh [15:57:41] [PASSED] name_refresh_wrong_mode [15:57:41] [PASSED] name_refresh_invalid_mode [15:57:41] [PASSED] rotate_multiple [15:57:41] [PASSED] rotate_invalid_val [15:57:41] [PASSED] rotate_truncated [15:57:41] [PASSED] invalid_option [15:57:41] [PASSED] invalid_tv_option [15:57:41] [PASSED] truncated_tv_option [15:57:41] ============ [PASSED] drm_test_cmdline_invalid ============= [15:57:41] =============== drm_test_cmdline_tv_options =============== [15:57:41] [PASSED] NTSC [15:57:41] [PASSED] NTSC_443 [15:57:41] [PASSED] NTSC_J [15:57:41] [PASSED] PAL [15:57:41] [PASSED] PAL_M [15:57:41] [PASSED] PAL_N [15:57:41] [PASSED] SECAM [15:57:41] [PASSED] MONO_525 [15:57:41] [PASSED] MONO_625 [15:57:41] =========== [PASSED] drm_test_cmdline_tv_options =========== [15:57:41] =============== [PASSED] drm_cmdline_parser ================ [15:57:41] ========== drmm_connector_hdmi_init (20 subtests) ========== [15:57:41] [PASSED] drm_test_connector_hdmi_init_valid [15:57:41] [PASSED] drm_test_connector_hdmi_init_bpc_8 [15:57:41] [PASSED] drm_test_connector_hdmi_init_bpc_10 [15:57:41] [PASSED] drm_test_connector_hdmi_init_bpc_12 [15:57:41] [PASSED] drm_test_connector_hdmi_init_bpc_invalid [15:57:41] [PASSED] drm_test_connector_hdmi_init_bpc_null [15:57:41] [PASSED] drm_test_connector_hdmi_init_formats_empty [15:57:41] [PASSED] drm_test_connector_hdmi_init_formats_no_rgb [15:57:41] === drm_test_connector_hdmi_init_formats_yuv420_allowed === [15:57:41] [PASSED] supported_formats=0x9 yuv420_allowed=1 [15:57:41] [PASSED] supported_formats=0x9 yuv420_allowed=0 [15:57:41] [PASSED] supported_formats=0x3 yuv420_allowed=1 [15:57:41] [PASSED] supported_formats=0x3 yuv420_allowed=0 [15:57:41] === [PASSED] drm_test_connector_hdmi_init_formats_yuv420_allowed === [15:57:41] [PASSED] drm_test_connector_hdmi_init_null_ddc [15:57:41] [PASSED] drm_test_connector_hdmi_init_null_product [15:57:41] [PASSED] drm_test_connector_hdmi_init_null_vendor [15:57:41] [PASSED] drm_test_connector_hdmi_init_product_length_exact [15:57:41] [PASSED] drm_test_connector_hdmi_init_product_length_too_long [15:57:41] [PASSED] drm_test_connector_hdmi_init_product_valid [15:57:41] [PASSED] drm_test_connector_hdmi_init_vendor_length_exact [15:57:41] [PASSED] drm_test_connector_hdmi_init_vendor_length_too_long [15:57:41] [PASSED] drm_test_connector_hdmi_init_vendor_valid [15:57:41] ========= drm_test_connector_hdmi_init_type_valid ========= [15:57:41] [PASSED] HDMI-A [15:57:41] [PASSED] HDMI-B [15:57:41] ===== [PASSED] drm_test_connector_hdmi_init_type_valid ===== [15:57:41] ======== drm_test_connector_hdmi_init_type_invalid ======== [15:57:41] [PASSED] Unknown [15:57:41] [PASSED] VGA [15:57:41] [PASSED] DVI-I [15:57:41] [PASSED] DVI-D [15:57:41] [PASSED] DVI-A [15:57:41] [PASSED] Composite [15:57:41] [PASSED] SVIDEO [15:57:41] [PASSED] LVDS [15:57:41] [PASSED] Component [15:57:41] [PASSED] DIN [15:57:41] [PASSED] DP [15:57:41] [PASSED] TV [15:57:41] [PASSED] eDP [15:57:41] [PASSED] Virtual [15:57:41] [PASSED] DSI [15:57:41] [PASSED] DPI [15:57:41] [PASSED] Writeback [15:57:41] [PASSED] SPI [15:57:41] [PASSED] USB [15:57:41] ==== [PASSED] drm_test_connector_hdmi_init_type_invalid ==== [15:57:41] ============ [PASSED] drmm_connector_hdmi_init ============= [15:57:41] ============= drmm_connector_init (3 subtests) ============= [15:57:41] [PASSED] drm_test_drmm_connector_init [15:57:41] [PASSED] drm_test_drmm_connector_init_null_ddc [15:57:41] ========= drm_test_drmm_connector_init_type_valid ========= [15:57:41] [PASSED] Unknown [15:57:41] [PASSED] VGA [15:57:41] [PASSED] DVI-I [15:57:41] [PASSED] DVI-D [15:57:41] [PASSED] DVI-A [15:57:41] [PASSED] Composite [15:57:41] [PASSED] SVIDEO [15:57:41] [PASSED] LVDS [15:57:41] [PASSED] Component [15:57:41] [PASSED] DIN [15:57:41] [PASSED] DP [15:57:41] [PASSED] HDMI-A [15:57:41] [PASSED] HDMI-B [15:57:41] [PASSED] TV [15:57:41] [PASSED] eDP [15:57:41] [PASSED] Virtual [15:57:41] [PASSED] DSI [15:57:41] [PASSED] DPI [15:57:41] [PASSED] Writeback [15:57:41] [PASSED] SPI [15:57:41] [PASSED] USB [15:57:41] ===== [PASSED] drm_test_drmm_connector_init_type_valid ===== [15:57:41] =============== [PASSED] drmm_connector_init =============== [15:57:41] ========= drm_connector_dynamic_init (6 subtests) ========== [15:57:41] [PASSED] drm_test_drm_connector_dynamic_init [15:57:41] [PASSED] drm_test_drm_connector_dynamic_init_null_ddc [15:57:41] [PASSED] drm_test_drm_connector_dynamic_init_not_added [15:57:41] [PASSED] drm_test_drm_connector_dynamic_init_properties [15:57:41] ===== drm_test_drm_connector_dynamic_init_type_valid ====== [15:57:41] [PASSED] Unknown [15:57:41] [PASSED] VGA [15:57:41] [PASSED] DVI-I [15:57:41] [PASSED] DVI-D [15:57:41] [PASSED] DVI-A [15:57:41] [PASSED] Composite [15:57:41] [PASSED] SVIDEO [15:57:41] [PASSED] LVDS [15:57:41] [PASSED] Component [15:57:41] [PASSED] DIN [15:57:41] [PASSED] DP [15:57:41] [PASSED] HDMI-A [15:57:41] [PASSED] HDMI-B [15:57:41] [PASSED] TV [15:57:41] [PASSED] eDP [15:57:41] [PASSED] Virtual [15:57:41] [PASSED] DSI [15:57:41] [PASSED] DPI [15:57:41] [PASSED] Writeback [15:57:41] [PASSED] SPI [15:57:41] [PASSED] USB [15:57:41] = [PASSED] drm_test_drm_connector_dynamic_init_type_valid == [15:57:41] ======== drm_test_drm_connector_dynamic_init_name ========= [15:57:41] [PASSED] Unknown [15:57:41] [PASSED] VGA [15:57:41] [PASSED] DVI-I [15:57:41] [PASSED] DVI-D [15:57:41] [PASSED] DVI-A [15:57:41] [PASSED] Composite [15:57:41] [PASSED] SVIDEO [15:57:41] [PASSED] LVDS [15:57:41] [PASSED] Component [15:57:41] [PASSED] DIN [15:57:41] [PASSED] DP [15:57:41] [PASSED] HDMI-A [15:57:41] [PASSED] HDMI-B [15:57:41] [PASSED] TV [15:57:41] [PASSED] eDP [15:57:41] [PASSED] Virtual [15:57:41] [PASSED] DSI [15:57:41] [PASSED] DPI [15:57:41] [PASSED] Writeback [15:57:41] [PASSED] SPI [15:57:41] [PASSED] USB [15:57:41] ==== [PASSED] drm_test_drm_connector_dynamic_init_name ===== [15:57:41] =========== [PASSED] drm_connector_dynamic_init ============ [15:57:41] ==== drm_connector_dynamic_register_early (4 subtests) ===== [15:57:41] [PASSED] drm_test_drm_connector_dynamic_register_early_on_list [15:57:41] [PASSED] drm_test_drm_connector_dynamic_register_early_defer [15:57:41] [PASSED] drm_test_drm_connector_dynamic_register_early_no_init [15:57:41] [PASSED] drm_test_drm_connector_dynamic_register_early_no_mode_object [15:57:41] ====== [PASSED] drm_connector_dynamic_register_early ======= [15:57:41] ======= drm_connector_dynamic_register (7 subtests) ======== [15:57:41] [PASSED] drm_test_drm_connector_dynamic_register_on_list [15:57:41] [PASSED] drm_test_drm_connector_dynamic_register_no_defer [15:57:41] [PASSED] drm_test_drm_connector_dynamic_register_no_init [15:57:41] [PASSED] drm_test_drm_connector_dynamic_register_mode_object [15:57:41] [PASSED] drm_test_drm_connector_dynamic_register_sysfs [15:57:41] [PASSED] drm_test_drm_connector_dynamic_register_sysfs_name [15:57:41] [PASSED] drm_test_drm_connector_dynamic_register_debugfs [15:57:41] ========= [PASSED] drm_connector_dynamic_register ========== [15:57:41] = drm_connector_attach_broadcast_rgb_property (2 subtests) = [15:57:41] [PASSED] drm_test_drm_connector_attach_broadcast_rgb_property [15:57:41] [PASSED] drm_test_drm_connector_attach_broadcast_rgb_property_hdmi_connector [15:57:41] === [PASSED] drm_connector_attach_broadcast_rgb_property === [15:57:41] ========== drm_get_tv_mode_from_name (2 subtests) ========== [15:57:41] ========== drm_test_get_tv_mode_from_name_valid =========== [15:57:41] [PASSED] NTSC [15:57:41] [PASSED] NTSC-443 [15:57:41] [PASSED] NTSC-J [15:57:41] [PASSED] PAL [15:57:41] [PASSED] PAL-M [15:57:41] [PASSED] PAL-N [15:57:41] [PASSED] SECAM [15:57:41] [PASSED] Mono [15:57:41] ====== [PASSED] drm_test_get_tv_mode_from_name_valid ======= [15:57:41] [PASSED] drm_test_get_tv_mode_from_name_truncated [15:57:41] ============ [PASSED] drm_get_tv_mode_from_name ============ [15:57:41] = drm_test_connector_hdmi_compute_mode_clock (12 subtests) = [15:57:41] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb [15:57:41] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_10bpc [15:57:41] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_10bpc_vic_1 [15:57:41] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_12bpc [15:57:41] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_12bpc_vic_1 [15:57:41] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_double [15:57:41] = drm_test_connector_hdmi_compute_mode_clock_yuv420_valid = [15:57:41] [PASSED] VIC 96 [15:57:41] [PASSED] VIC 97 [15:57:41] [PASSED] VIC 101 [15:57:41] [PASSED] VIC 102 [15:57:41] [PASSED] VIC 106 [15:57:41] [PASSED] VIC 107 [15:57:41] === [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_valid === [15:57:41] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_10_bpc [15:57:41] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_12_bpc [15:57:41] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_8_bpc [15:57:41] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_10_bpc [15:57:41] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_12_bpc [15:57:41] === [PASSED] drm_test_connector_hdmi_compute_mode_clock ==== [15:57:41] == drm_hdmi_connector_get_broadcast_rgb_name (2 subtests) == [15:57:41] === drm_test_drm_hdmi_connector_get_broadcast_rgb_name ==== [15:57:41] [PASSED] Automatic [15:57:41] [PASSED] Full [15:57:41] [PASSED] Limited 16:235 [15:57:41] === [PASSED] drm_test_drm_hdmi_connector_get_broadcast_rgb_name === [15:57:41] [PASSED] drm_test_drm_hdmi_connector_get_broadcast_rgb_name_invalid [15:57:41] ==== [PASSED] drm_hdmi_connector_get_broadcast_rgb_name ==== [15:57:41] == drm_hdmi_connector_get_output_format_name (2 subtests) == [15:57:41] === drm_test_drm_hdmi_connector_get_output_format_name ==== [15:57:41] [PASSED] RGB [15:57:41] [PASSED] YUV 4:2:0 [15:57:41] [PASSED] YUV 4:2:2 [15:57:41] [PASSED] YUV 4:4:4 [15:57:41] === [PASSED] drm_test_drm_hdmi_connector_get_output_format_name === [15:57:41] [PASSED] drm_test_drm_hdmi_connector_get_output_format_name_invalid [15:57:41] ==== [PASSED] drm_hdmi_connector_get_output_format_name ==== [15:57:41] ============= drm_damage_helper (21 subtests) ============== [15:57:41] [PASSED] drm_test_damage_iter_no_damage [15:57:41] [PASSED] drm_test_damage_iter_no_damage_fractional_src [15:57:41] [PASSED] drm_test_damage_iter_no_damage_src_moved [15:57:41] [PASSED] drm_test_damage_iter_no_damage_fractional_src_moved [15:57:41] [PASSED] drm_test_damage_iter_no_damage_not_visible [15:57:41] [PASSED] drm_test_damage_iter_no_damage_no_crtc [15:57:41] [PASSED] drm_test_damage_iter_no_damage_no_fb [15:57:41] [PASSED] drm_test_damage_iter_simple_damage [15:57:41] [PASSED] drm_test_damage_iter_single_damage [15:57:41] [PASSED] drm_test_damage_iter_single_damage_intersect_src [15:57:41] [PASSED] drm_test_damage_iter_single_damage_outside_src [15:57:41] [PASSED] drm_test_damage_iter_single_damage_fractional_src [15:57:41] [PASSED] drm_test_damage_iter_single_damage_intersect_fractional_src [15:57:41] [PASSED] drm_test_damage_iter_single_damage_outside_fractional_src [15:57:41] [PASSED] drm_test_damage_iter_single_damage_src_moved [15:57:41] [PASSED] drm_test_damage_iter_single_damage_fractional_src_moved [15:57:41] [PASSED] drm_test_damage_iter_damage [15:57:41] [PASSED] drm_test_damage_iter_damage_one_intersect [15:57:41] [PASSED] drm_test_damage_iter_damage_one_outside [15:57:41] [PASSED] drm_test_damage_iter_damage_src_moved [15:57:41] [PASSED] drm_test_damage_iter_damage_not_visible [15:57:41] ================ [PASSED] drm_damage_helper ================ [15:57:41] ============== drm_dp_mst_helper (3 subtests) ============== [15:57:41] ============== drm_test_dp_mst_calc_pbn_mode ============== [15:57:41] [PASSED] Clock 154000 BPP 30 DSC disabled [15:57:41] [PASSED] Clock 234000 BPP 30 DSC disabled [15:57:41] [PASSED] Clock 297000 BPP 24 DSC disabled [15:57:41] [PASSED] Clock 332880 BPP 24 DSC enabled [15:57:41] [PASSED] Clock 324540 BPP 24 DSC enabled [15:57:41] ========== [PASSED] drm_test_dp_mst_calc_pbn_mode ========== [15:57:41] ============== drm_test_dp_mst_calc_pbn_div =============== [15:57:41] [PASSED] Link rate 2000000 lane count 4 [15:57:41] [PASSED] Link rate 2000000 lane count 2 [15:57:41] [PASSED] Link rate 2000000 lane count 1 [15:57:41] [PASSED] Link rate 1350000 lane count 4 [15:57:41] [PASSED] Link rate 1350000 lane count 2 [15:57:41] [PASSED] Link rate 1350000 lane count 1 [15:57:41] [PASSED] Link rate 1000000 lane count 4 [15:57:41] [PASSED] Link rate 1000000 lane count 2 [15:57:41] [PASSED] Link rate 1000000 lane count 1 [15:57:41] [PASSED] Link rate 810000 lane count 4 [15:57:41] [PASSED] Link rate 810000 lane count 2 [15:57:41] [PASSED] Link rate 810000 lane count 1 [15:57:41] [PASSED] Link rate 540000 lane count 4 [15:57:41] [PASSED] Link rate 540000 lane count 2 [15:57:41] [PASSED] Link rate 540000 lane count 1 [15:57:41] [PASSED] Link rate 270000 lane count 4 [15:57:41] [PASSED] Link rate 270000 lane count 2 [15:57:41] [PASSED] Link rate 270000 lane count 1 [15:57:41] [PASSED] Link rate 162000 lane count 4 [15:57:41] [PASSED] Link rate 162000 lane count 2 [15:57:41] [PASSED] Link rate 162000 lane count 1 [15:57:41] ========== [PASSED] drm_test_dp_mst_calc_pbn_div =========== [15:57:41] ========= drm_test_dp_mst_sideband_msg_req_decode ========= [15:57:41] [PASSED] DP_ENUM_PATH_RESOURCES with port number [15:57:41] [PASSED] DP_POWER_UP_PHY with port number [15:57:41] [PASSED] DP_POWER_DOWN_PHY with port number [15:57:41] [PASSED] DP_ALLOCATE_PAYLOAD with SDP stream sinks [15:57:41] [PASSED] DP_ALLOCATE_PAYLOAD with port number [15:57:41] [PASSED] DP_ALLOCATE_PAYLOAD with VCPI [15:57:41] [PASSED] DP_ALLOCATE_PAYLOAD with PBN [15:57:41] [PASSED] DP_QUERY_PAYLOAD with port number [15:57:41] [PASSED] DP_QUERY_PAYLOAD with VCPI [15:57:41] [PASSED] DP_REMOTE_DPCD_READ with port number [15:57:41] [PASSED] DP_REMOTE_DPCD_READ with DPCD address [15:57:41] [PASSED] DP_REMOTE_DPCD_READ with max number of bytes [15:57:41] [PASSED] DP_REMOTE_DPCD_WRITE with port number [15:57:41] [PASSED] DP_REMOTE_DPCD_WRITE with DPCD address [15:57:41] [PASSED] DP_REMOTE_DPCD_WRITE with data array [15:57:41] [PASSED] DP_REMOTE_I2C_READ with port number [15:57:41] [PASSED] DP_REMOTE_I2C_READ with I2C device ID [15:57:41] [PASSED] DP_REMOTE_I2C_READ with transactions array [15:57:41] [PASSED] DP_REMOTE_I2C_WRITE with port number [15:57:41] [PASSED] DP_REMOTE_I2C_WRITE with I2C device ID [15:57:41] [PASSED] DP_REMOTE_I2C_WRITE with data array [15:57:41] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream ID [15:57:41] [PASSED] DP_QUERY_STREAM_ENC_STATUS with client ID [15:57:41] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream event [15:57:41] [PASSED] DP_QUERY_STREAM_ENC_STATUS with valid stream event [15:57:41] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream behavior [15:57:41] [PASSED] DP_QUERY_STREAM_ENC_STATUS with a valid stream behavior [15:57:41] ===== [PASSED] drm_test_dp_mst_sideband_msg_req_decode ===== [15:57:41] ================ [PASSED] drm_dp_mst_helper ================ [15:57:41] ================== drm_exec (7 subtests) =================== [15:57:41] [PASSED] sanitycheck [15:57:41] [PASSED] test_lock [15:57:41] [PASSED] test_lock_unlock [15:57:41] [PASSED] test_duplicates [15:57:41] [PASSED] test_prepare [15:57:41] [PASSED] test_prepare_array [15:57:41] [PASSED] test_multiple_loops [15:57:41] ==================== [PASSED] drm_exec ===================== [15:57:41] =========== drm_format_helper_test (18 subtests) =========== [15:57:41] ============== drm_test_fb_xrgb8888_to_gray8 ============== [15:57:41] [PASSED] single_pixel_source_buffer [15:57:41] [PASSED] single_pixel_clip_rectangle [15:57:41] [PASSED] well_known_colors [15:57:41] [PASSED] destination_pitch [15:57:41] ========== [PASSED] drm_test_fb_xrgb8888_to_gray8 ========== [15:57:41] ============= drm_test_fb_xrgb8888_to_rgb332 ============== [15:57:41] [PASSED] single_pixel_source_buffer [15:57:41] [PASSED] single_pixel_clip_rectangle [15:57:41] [PASSED] well_known_colors [15:57:41] [PASSED] destination_pitch [15:57:41] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb332 ========== [15:57:41] ============= drm_test_fb_xrgb8888_to_rgb565 ============== [15:57:41] [PASSED] single_pixel_source_buffer [15:57:41] [PASSED] single_pixel_clip_rectangle [15:57:41] [PASSED] well_known_colors [15:57:41] [PASSED] destination_pitch [15:57:41] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb565 ========== [15:57:41] ============ drm_test_fb_xrgb8888_to_xrgb1555 ============= [15:57:41] [PASSED] single_pixel_source_buffer [15:57:41] [PASSED] single_pixel_clip_rectangle [15:57:41] [PASSED] well_known_colors [15:57:41] [PASSED] destination_pitch [15:57:41] ======== [PASSED] drm_test_fb_xrgb8888_to_xrgb1555 ========= [15:57:41] ============ drm_test_fb_xrgb8888_to_argb1555 ============= [15:57:41] [PASSED] single_pixel_source_buffer [15:57:41] [PASSED] single_pixel_clip_rectangle [15:57:41] [PASSED] well_known_colors [15:57:41] [PASSED] destination_pitch [15:57:41] ======== [PASSED] drm_test_fb_xrgb8888_to_argb1555 ========= [15:57:41] ============ drm_test_fb_xrgb8888_to_rgba5551 ============= [15:57:41] [PASSED] single_pixel_source_buffer [15:57:41] [PASSED] single_pixel_clip_rectangle [15:57:41] [PASSED] well_known_colors [15:57:41] [PASSED] destination_pitch [15:57:41] ======== [PASSED] drm_test_fb_xrgb8888_to_rgba5551 ========= [15:57:41] ============= drm_test_fb_xrgb8888_to_rgb888 ============== [15:57:41] [PASSED] single_pixel_source_buffer [15:57:41] [PASSED] single_pixel_clip_rectangle [15:57:41] [PASSED] well_known_colors [15:57:41] [PASSED] destination_pitch [15:57:41] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb888 ========== [15:57:41] ============= drm_test_fb_xrgb8888_to_bgr888 ============== [15:57:41] [PASSED] single_pixel_source_buffer [15:57:41] [PASSED] single_pixel_clip_rectangle [15:57:41] [PASSED] well_known_colors [15:57:41] [PASSED] destination_pitch [15:57:41] ========= [PASSED] drm_test_fb_xrgb8888_to_bgr888 ========== [15:57:41] ============ drm_test_fb_xrgb8888_to_argb8888 ============= [15:57:41] [PASSED] single_pixel_source_buffer [15:57:41] [PASSED] single_pixel_clip_rectangle [15:57:41] [PASSED] well_known_colors [15:57:41] [PASSED] destination_pitch [15:57:41] ======== [PASSED] drm_test_fb_xrgb8888_to_argb8888 ========= [15:57:41] =========== drm_test_fb_xrgb8888_to_xrgb2101010 =========== [15:57:41] [PASSED] single_pixel_source_buffer [15:57:41] [PASSED] single_pixel_clip_rectangle [15:57:41] [PASSED] well_known_colors [15:57:41] [PASSED] destination_pitch [15:57:41] ======= [PASSED] drm_test_fb_xrgb8888_to_xrgb2101010 ======= [15:57:41] =========== drm_test_fb_xrgb8888_to_argb2101010 =========== [15:57:41] [PASSED] single_pixel_source_buffer [15:57:41] [PASSED] single_pixel_clip_rectangle [15:57:41] [PASSED] well_known_colors [15:57:41] [PASSED] destination_pitch [15:57:41] ======= [PASSED] drm_test_fb_xrgb8888_to_argb2101010 ======= [15:57:41] ============== drm_test_fb_xrgb8888_to_mono =============== [15:57:41] [PASSED] single_pixel_source_buffer [15:57:41] [PASSED] single_pixel_clip_rectangle [15:57:41] [PASSED] well_known_colors [15:57:41] [PASSED] destination_pitch [15:57:41] ========== [PASSED] drm_test_fb_xrgb8888_to_mono =========== [15:57:41] ==================== drm_test_fb_swab ===================== [15:57:41] [PASSED] single_pixel_source_buffer [15:57:41] [PASSED] single_pixel_clip_rectangle [15:57:41] [PASSED] well_known_colors [15:57:41] [PASSED] destination_pitch [15:57:41] ================ [PASSED] drm_test_fb_swab ================= [15:57:41] ============ drm_test_fb_xrgb8888_to_xbgr8888 ============= [15:57:41] [PASSED] single_pixel_source_buffer [15:57:41] [PASSED] single_pixel_clip_rectangle [15:57:41] [PASSED] well_known_colors [15:57:41] [PASSED] destination_pitch [15:57:41] ======== [PASSED] drm_test_fb_xrgb8888_to_xbgr8888 ========= [15:57:41] ============ drm_test_fb_xrgb8888_to_abgr8888 ============= [15:57:41] [PASSED] single_pixel_source_buffer [15:57:41] [PASSED] single_pixel_clip_rectangle [15:57:41] [PASSED] well_known_colors [15:57:41] [PASSED] destination_pitch [15:57:41] ======== [PASSED] drm_test_fb_xrgb8888_to_abgr8888 ========= [15:57:41] ================= drm_test_fb_clip_offset ================= [15:57:41] [PASSED] pass through [15:57:41] [PASSED] horizontal offset [15:57:41] [PASSED] vertical offset [15:57:41] [PASSED] horizontal and vertical offset [15:57:41] [PASSED] horizontal offset (custom pitch) [15:57:41] [PASSED] vertical offset (custom pitch) [15:57:41] [PASSED] horizontal and vertical offset (custom pitch) [15:57:41] ============= [PASSED] drm_test_fb_clip_offset ============= [15:57:41] ============== drm_test_fb_build_fourcc_list ============== [15:57:41] [PASSED] no native formats [15:57:41] [PASSED] XRGB8888 as native format [15:57:41] [PASSED] remove duplicates [15:57:41] [PASSED] convert alpha formats [15:57:41] [PASSED] random formats [15:57:41] ========== [PASSED] drm_test_fb_build_fourcc_list ========== [15:57:41] =================== drm_test_fb_memcpy ==================== [15:57:41] [PASSED] single_pixel_source_buffer: XR24 little-endian (0x34325258) [15:57:41] [PASSED] single_pixel_source_buffer: XRA8 little-endian (0x38415258) [15:57:41] [PASSED] single_pixel_source_buffer: YU24 little-endian (0x34325559) [15:57:41] [PASSED] single_pixel_clip_rectangle: XB24 little-endian (0x34324258) [15:57:41] [PASSED] single_pixel_clip_rectangle: XRA8 little-endian (0x38415258) [15:57:41] [PASSED] single_pixel_clip_rectangle: YU24 little-endian (0x34325559) [15:57:41] [PASSED] well_known_colors: XB24 little-endian (0x34324258) [15:57:41] [PASSED] well_known_colors: XRA8 little-endian (0x38415258) [15:57:41] [PASSED] well_known_colors: YU24 little-endian (0x34325559) [15:57:41] [PASSED] destination_pitch: XB24 little-endian (0x34324258) [15:57:41] [PASSED] destination_pitch: XRA8 little-endian (0x38415258) [15:57:41] [PASSED] destination_pitch: YU24 little-endian (0x34325559) [15:57:41] =============== [PASSED] drm_test_fb_memcpy ================ [15:57:41] ============= [PASSED] drm_format_helper_test ============== [15:57:41] ================= drm_format (18 subtests) ================= [15:57:41] [PASSED] drm_test_format_block_width_invalid [15:57:41] [PASSED] drm_test_format_block_width_one_plane [15:57:41] [PASSED] drm_test_format_block_width_two_plane [15:57:41] [PASSED] drm_test_format_block_width_three_plane [15:57:41] [PASSED] drm_test_format_block_width_tiled [15:57:41] [PASSED] drm_test_format_block_height_invalid [15:57:41] [PASSED] drm_test_format_block_height_one_plane [15:57:41] [PASSED] drm_test_format_block_height_two_plane [15:57:41] [PASSED] drm_test_format_block_height_three_plane [15:57:41] [PASSED] drm_test_format_block_height_tiled [15:57:41] [PASSED] drm_test_format_min_pitch_invalid [15:57:41] [PASSED] drm_test_format_min_pitch_one_plane_8bpp [15:57:41] [PASSED] drm_test_format_min_pitch_one_plane_16bpp [15:57:41] [PASSED] drm_test_format_min_pitch_one_plane_24bpp [15:57:41] [PASSED] drm_test_format_min_pitch_one_plane_32bpp [15:57:41] [PASSED] drm_test_format_min_pitch_two_plane [15:57:41] [PASSED] drm_test_format_min_pitch_three_plane_8bpp [15:57:41] [PASSED] drm_test_format_min_pitch_tiled [15:57:41] =================== [PASSED] drm_format ==================== [15:57:41] ============== drm_framebuffer (10 subtests) =============== [15:57:41] ========== drm_test_framebuffer_check_src_coords ========== [15:57:41] [PASSED] Success: source fits into fb [15:57:41] [PASSED] Fail: overflowing fb with x-axis coordinate [15:57:41] [PASSED] Fail: overflowing fb with y-axis coordinate [15:57:41] [PASSED] Fail: overflowing fb with source width [15:57:41] [PASSED] Fail: overflowing fb with source height [15:57:41] ====== [PASSED] drm_test_framebuffer_check_src_coords ====== [15:57:41] [PASSED] drm_test_framebuffer_cleanup [15:57:41] =============== drm_test_framebuffer_create =============== [15:57:41] [PASSED] ABGR8888 normal sizes [15:57:41] [PASSED] ABGR8888 max sizes [15:57:41] [PASSED] ABGR8888 pitch greater than min required [15:57:41] [PASSED] ABGR8888 pitch less than min required [15:57:41] [PASSED] ABGR8888 Invalid width [15:57:41] [PASSED] ABGR8888 Invalid buffer handle [15:57:41] [PASSED] No pixel format [15:57:41] [PASSED] ABGR8888 Width 0 [15:57:41] [PASSED] ABGR8888 Height 0 [15:57:41] [PASSED] ABGR8888 Out of bound height * pitch combination [15:57:41] [PASSED] ABGR8888 Large buffer offset [15:57:41] [PASSED] ABGR8888 Buffer offset for inexistent plane [15:57:41] [PASSED] ABGR8888 Invalid flag [15:57:41] [PASSED] ABGR8888 Set DRM_MODE_FB_MODIFIERS without modifiers [15:57:41] [PASSED] ABGR8888 Valid buffer modifier [15:57:41] [PASSED] ABGR8888 Invalid buffer modifier(DRM_FORMAT_MOD_SAMSUNG_64_32_TILE) [15:57:41] [PASSED] ABGR8888 Extra pitches without DRM_MODE_FB_MODIFIERS [15:57:41] [PASSED] ABGR8888 Extra pitches with DRM_MODE_FB_MODIFIERS [15:57:41] [PASSED] NV12 Normal sizes [15:57:41] [PASSED] NV12 Max sizes [15:57:41] [PASSED] NV12 Invalid pitch [15:57:41] [PASSED] NV12 Invalid modifier/missing DRM_MODE_FB_MODIFIERS flag [15:57:41] [PASSED] NV12 different modifier per-plane [15:57:41] [PASSED] NV12 with DRM_FORMAT_MOD_SAMSUNG_64_32_TILE [15:57:41] [PASSED] NV12 Valid modifiers without DRM_MODE_FB_MODIFIERS [15:57:41] [PASSED] NV12 Modifier for inexistent plane [15:57:41] [PASSED] NV12 Handle for inexistent plane [15:57:41] [PASSED] NV12 Handle for inexistent plane without DRM_MODE_FB_MODIFIERS [15:57:41] [PASSED] YVU420 DRM_MODE_FB_MODIFIERS set without modifier [15:57:41] [PASSED] YVU420 Normal sizes [15:57:41] [PASSED] YVU420 Max sizes [15:57:41] [PASSED] YVU420 Invalid pitch [15:57:41] [PASSED] YVU420 Different pitches [15:57:41] [PASSED] YVU420 Different buffer offsets/pitches [15:57:41] [PASSED] YVU420 Modifier set just for plane 0, without DRM_MODE_FB_MODIFIERS [15:57:41] [PASSED] YVU420 Modifier set just for planes 0, 1, without DRM_MODE_FB_MODIFIERS [15:57:41] [PASSED] YVU420 Modifier set just for plane 0, 1, with DRM_MODE_FB_MODIFIERS [15:57:41] [PASSED] YVU420 Valid modifier [15:57:41] [PASSED] YVU420 Different modifiers per plane [15:57:41] [PASSED] YVU420 Modifier for inexistent plane [15:57:41] [PASSED] YUV420_10BIT Invalid modifier(DRM_FORMAT_MOD_LINEAR) [15:57:41] [PASSED] X0L2 Normal sizes [15:57:41] [PASSED] X0L2 Max sizes [15:57:41] [PASSED] X0L2 Invalid pitch [15:57:41] [PASSED] X0L2 Pitch greater than minimum required [15:57:41] [PASSED] X0L2 Handle for inexistent plane [15:57:41] [PASSED] X0L2 Offset for inexistent plane, without DRM_MODE_FB_MODIFIERS set [15:57:41] [PASSED] X0L2 Modifier without DRM_MODE_FB_MODIFIERS set [15:57:41] [PASSED] X0L2 Valid modifier [15:57:41] [PASSED] X0L2 Modifier for inexistent plane [15:57:41] =========== [PASSED] drm_test_framebuffer_create =========== [15:57:41] [PASSED] drm_test_framebuffer_free [15:57:41] [PASSED] drm_test_framebuffer_init [15:57:41] [PASSED] drm_test_framebuffer_init_bad_format [15:57:41] [PASSED] drm_test_framebuffer_init_dev_mismatch [15:57:41] [PASSED] drm_test_framebuffer_lookup [15:57:41] [PASSED] drm_test_framebuffer_lookup_inexistent [15:57:41] [PASSED] drm_test_framebuffer_modifiers_not_supported [15:57:41] ================= [PASSED] drm_framebuffer ================= [15:57:41] ================ drm_gem_shmem (8 subtests) ================ [15:57:41] [PASSED] drm_gem_shmem_test_obj_create [15:57:41] [PASSED] drm_gem_shmem_test_obj_create_private [15:57:41] [PASSED] drm_gem_shmem_test_pin_pages [15:57:41] [PASSED] drm_gem_shmem_test_vmap [15:57:41] [PASSED] drm_gem_shmem_test_get_pages_sgt [15:57:41] [PASSED] drm_gem_shmem_test_get_sg_table [15:57:41] [PASSED] drm_gem_shmem_test_madvise [15:57:41] [PASSED] drm_gem_shmem_test_purge [15:57:41] ================== [PASSED] drm_gem_shmem ================== [15:57:41] === drm_atomic_helper_connector_hdmi_check (27 subtests) === [15:57:41] [PASSED] drm_test_check_broadcast_rgb_auto_cea_mode [15:57:41] [PASSED] drm_test_check_broadcast_rgb_auto_cea_mode_vic_1 [15:57:41] [PASSED] drm_test_check_broadcast_rgb_full_cea_mode [15:57:41] [PASSED] drm_test_check_broadcast_rgb_full_cea_mode_vic_1 [15:57:41] [PASSED] drm_test_check_broadcast_rgb_limited_cea_mode [15:57:41] [PASSED] drm_test_check_broadcast_rgb_limited_cea_mode_vic_1 [15:57:41] ====== drm_test_check_broadcast_rgb_cea_mode_yuv420 ======= [15:57:41] [PASSED] Automatic [15:57:41] [PASSED] Full [15:57:41] [PASSED] Limited 16:235 [15:57:41] == [PASSED] drm_test_check_broadcast_rgb_cea_mode_yuv420 === [15:57:41] [PASSED] drm_test_check_broadcast_rgb_crtc_mode_changed [15:57:41] [PASSED] drm_test_check_broadcast_rgb_crtc_mode_not_changed [15:57:41] [PASSED] drm_test_check_disable_connector [15:57:41] [PASSED] drm_test_check_hdmi_funcs_reject_rate [15:57:41] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_rgb [15:57:41] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_yuv420 [15:57:41] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_ignore_yuv422 [15:57:41] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_ignore_yuv420 [15:57:41] [PASSED] drm_test_check_driver_unsupported_fallback_yuv420 [15:57:41] [PASSED] drm_test_check_output_bpc_crtc_mode_changed [15:57:41] [PASSED] drm_test_check_output_bpc_crtc_mode_not_changed [15:57:41] [PASSED] drm_test_check_output_bpc_dvi [15:57:41] [PASSED] drm_test_check_output_bpc_format_vic_1 [15:57:41] [PASSED] drm_test_check_output_bpc_format_display_8bpc_only [15:57:41] [PASSED] drm_test_check_output_bpc_format_display_rgb_only [15:57:41] [PASSED] drm_test_check_output_bpc_format_driver_8bpc_only [15:57:41] [PASSED] drm_test_check_output_bpc_format_driver_rgb_only [15:57:41] [PASSED] drm_test_check_tmds_char_rate_rgb_8bpc [15:57:41] [PASSED] drm_test_check_tmds_char_rate_rgb_10bpc [15:57:41] [PASSED] drm_test_check_tmds_char_rate_rgb_12bpc [15:57:41] ===== [PASSED] drm_atomic_helper_connector_hdmi_check ====== [15:57:41] === drm_atomic_helper_connector_hdmi_reset (6 subtests) ==== [15:57:41] [PASSED] drm_test_check_broadcast_rgb_value [15:57:41] [PASSED] drm_test_check_bpc_8_value [15:57:41] [PASSED] drm_test_check_bpc_10_value [15:57:41] [PASSED] drm_test_check_bpc_12_value [15:57:41] [PASSED] drm_test_check_format_value [15:57:41] [PASSED] drm_test_check_tmds_char_value [15:57:41] ===== [PASSED] drm_atomic_helper_connector_hdmi_reset ====== [15:57:41] = drm_atomic_helper_connector_hdmi_mode_valid (4 subtests) = [15:57:41] [PASSED] drm_test_check_mode_valid [15:57:41] [PASSED] drm_test_check_mode_valid_reject [15:57:41] [PASSED] drm_test_check_mode_valid_reject_rate [15:57:41] [PASSED] drm_test_check_mode_valid_reject_max_clock [15:57:41] === [PASSED] drm_atomic_helper_connector_hdmi_mode_valid === [15:57:41] ================= drm_managed (2 subtests) ================= [15:57:41] [PASSED] drm_test_managed_release_action [15:57:41] [PASSED] drm_test_managed_run_action [15:57:41] =================== [PASSED] drm_managed =================== [15:57:41] =================== drm_mm (6 subtests) ==================== [15:57:41] [PASSED] drm_test_mm_init [15:57:41] [PASSED] drm_test_mm_debug [15:57:41] [PASSED] drm_test_mm_align32 [15:57:41] [PASSED] drm_test_mm_align64 [15:57:41] [PASSED] drm_test_mm_lowest [15:57:41] [PASSED] drm_test_mm_highest [15:57:41] ===================== [PASSED] drm_mm ====================== [15:57:41] ============= drm_modes_analog_tv (5 subtests) ============= [15:57:41] [PASSED] drm_test_modes_analog_tv_mono_576i [15:57:41] [PASSED] drm_test_modes_analog_tv_ntsc_480i [15:57:41] [PASSED] drm_test_modes_analog_tv_ntsc_480i_inlined [15:57:41] [PASSED] drm_test_modes_analog_tv_pal_576i [15:57:41] [PASSED] drm_test_modes_analog_tv_pal_576i_inlined [15:57:41] =============== [PASSED] drm_modes_analog_tv =============== [15:57:41] ============== drm_plane_helper (2 subtests) =============== [15:57:41] =============== drm_test_check_plane_state ================ [15:57:41] [PASSED] clipping_simple [15:57:41] [PASSED] clipping_rotate_reflect [15:57:41] [PASSED] positioning_simple [15:57:41] [PASSED] upscaling [15:57:41] [PASSED] downscaling [15:57:41] [PASSED] rounding1 [15:57:41] [PASSED] rounding2 [15:57:41] [PASSED] rounding3 [15:57:41] [PASSED] rounding4 [15:57:41] =========== [PASSED] drm_test_check_plane_state ============ [15:57:41] =========== drm_test_check_invalid_plane_state ============ [15:57:41] [PASSED] positioning_invalid [15:57:41] [PASSED] upscaling_invalid [15:57:41] [PASSED] downscaling_invalid [15:57:41] ======= [PASSED] drm_test_check_invalid_plane_state ======== [15:57:41] ================ [PASSED] drm_plane_helper ================= [15:57:41] ====== drm_connector_helper_tv_get_modes (1 subtest) ======= [15:57:41] ====== drm_test_connector_helper_tv_get_modes_check ======= [15:57:41] [PASSED] None [15:57:41] [PASSED] PAL [15:57:41] [PASSED] NTSC [15:57:41] [PASSED] Both, NTSC Default [15:57:41] [PASSED] Both, PAL Default [15:57:41] [PASSED] Both, NTSC Default, with PAL on command-line [15:57:41] [PASSED] Both, PAL Default, with NTSC on command-line [15:57:41] == [PASSED] drm_test_connector_helper_tv_get_modes_check === [15:57:41] ======== [PASSED] drm_connector_helper_tv_get_modes ======== [15:57:41] ================== drm_rect (9 subtests) =================== [15:57:41] [PASSED] drm_test_rect_clip_scaled_div_by_zero [15:57:41] [PASSED] drm_test_rect_clip_scaled_not_clipped [15:57:41] [PASSED] drm_test_rect_clip_scaled_clipped [15:57:41] [PASSED] drm_test_rect_clip_scaled_signed_vs_unsigned [15:57:41] ================= drm_test_rect_intersect ================= [15:57:41] [PASSED] top-left x bottom-right: 2x2+1+1 x 2x2+0+0 [15:57:41] [PASSED] top-right x bottom-left: 2x2+0+0 x 2x2+1-1 [15:57:41] [PASSED] bottom-left x top-right: 2x2+1-1 x 2x2+0+0 [15:57:41] [PASSED] bottom-right x top-left: 2x2+0+0 x 2x2+1+1 [15:57:41] [PASSED] right x left: 2x1+0+0 x 3x1+1+0 [15:57:41] [PASSED] left x right: 3x1+1+0 x 2x1+0+0 [15:57:41] [PASSED] up x bottom: 1x2+0+0 x 1x3+0-1 [15:57:41] [PASSED] bottom x up: 1x3+0-1 x 1x2+0+0 [15:57:41] [PASSED] touching corner: 1x1+0+0 x 2x2+1+1 [15:57:41] [PASSED] touching side: 1x1+0+0 x 1x1+1+0 [15:57:41] [PASSED] equal rects: 2x2+0+0 x 2x2+0+0 [15:57:41] [PASSED] inside another: 2x2+0+0 x 1x1+1+1 [15:57:41] [PASSED] far away: 1x1+0+0 x 1x1+3+6 [15:57:41] [PASSED] points intersecting: 0x0+5+10 x 0x0+5+10 [15:57:41] [PASSED] points not intersecting: 0x0+0+0 x 0x0+5+10 [15:57:41] ============= [PASSED] drm_test_rect_intersect ============= [15:57:41] ================ drm_test_rect_calc_hscale ================ [15:57:41] [PASSED] normal use [15:57:41] [PASSED] out of max range [15:57:41] [PASSED] out of min range [15:57:41] [PASSED] zero dst [15:57:41] [PASSED] negative src [15:57:41] [PASSED] negative dst [15:57:41] ============ [PASSED] drm_test_rect_calc_hscale ============ [15:57:41] ================ drm_test_rect_calc_vscale ================ [15:57:41] [PASSED] normal use [15:57:41] [PASSED] out of max range [15:57:41] [PASSED] out of min range [15:57:41] [PASSED] zero dst [15:57:41] [PASSED] negative src [15:57:41] [PASSED] negative dst stty: 'standard input': Inappropriate ioctl for device [15:57:41] ============ [PASSED] drm_test_rect_calc_vscale ============ [15:57:41] ================== drm_test_rect_rotate =================== [15:57:41] [PASSED] reflect-x [15:57:41] [PASSED] reflect-y [15:57:41] [PASSED] rotate-0 [15:57:41] [PASSED] rotate-90 [15:57:41] [PASSED] rotate-180 [15:57:41] [PASSED] rotate-270 [15:57:41] ============== [PASSED] drm_test_rect_rotate =============== [15:57:41] ================ drm_test_rect_rotate_inv ================= [15:57:41] [PASSED] reflect-x [15:57:41] [PASSED] reflect-y [15:57:41] [PASSED] rotate-0 [15:57:41] [PASSED] rotate-90 [15:57:41] [PASSED] rotate-180 [15:57:41] [PASSED] rotate-270 [15:57:41] ============ [PASSED] drm_test_rect_rotate_inv ============= [15:57:41] ==================== [PASSED] drm_rect ===================== [15:57:41] ============================================================ [15:57:41] Testing complete. Ran 616 tests: passed: 616 [15:57:41] Elapsed time: 23.865s total, 1.748s configuring, 21.899s building, 0.195s running
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/ttm/tests/.kunitconfig [15:57:41] Configuring KUnit Kernel ...
Regenerating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig [15:57:43] Building KUnit Kernel ... Populating config with: $ make ARCH=um O=.kunit olddefconfig Building with: $ make all compile_commands.json scripts_gdb ARCH=um O=.kunit --jobs=48 [15:57:51] Starting KUnit Kernel (1/1)... [15:57:51] ============================================================ Running tests with: $ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt [15:57:51] ================= ttm_device (5 subtests) ================== [15:57:51] [PASSED] ttm_device_init_basic [15:57:51] [PASSED] ttm_device_init_multiple [15:57:51] [PASSED] ttm_device_fini_basic [15:57:51] [PASSED] ttm_device_init_no_vma_man [15:57:51] ================== ttm_device_init_pools ================== [15:57:51] [PASSED] No DMA allocations, no DMA32 required [15:57:51] [PASSED] DMA allocations, DMA32 required [15:57:51] [PASSED] No DMA allocations, DMA32 required [15:57:51] [PASSED] DMA allocations, no DMA32 required [15:57:51] ============== [PASSED] ttm_device_init_pools ============== [15:57:51] =================== [PASSED] ttm_device ==================== [15:57:51] ================== ttm_pool (8 subtests) =================== [15:57:51] ================== ttm_pool_alloc_basic =================== [15:57:51] [PASSED] One page [15:57:51] [PASSED] More than one page [15:57:51] [PASSED] Above the allocation limit [15:57:51] [PASSED] One page, with coherent DMA mappings enabled [15:57:51] [PASSED] Above the allocation limit, with coherent DMA mappings enabled [15:57:51] ============== [PASSED] ttm_pool_alloc_basic =============== [15:57:51] ============== ttm_pool_alloc_basic_dma_addr ============== [15:57:51] [PASSED] One page [15:57:51] [PASSED] More than one page [15:57:51] [PASSED] Above the allocation limit [15:57:51] [PASSED] One page, with coherent DMA mappings enabled [15:57:51] [PASSED] Above the allocation limit, with coherent DMA mappings enabled [15:57:51] ========== [PASSED] ttm_pool_alloc_basic_dma_addr ========== [15:57:51] [PASSED] ttm_pool_alloc_order_caching_match [15:57:51] [PASSED] ttm_pool_alloc_caching_mismatch [15:57:51] [PASSED] ttm_pool_alloc_order_mismatch [15:57:51] [PASSED] ttm_pool_free_dma_alloc [15:57:51] [PASSED] ttm_pool_free_no_dma_alloc [15:57:51] [PASSED] ttm_pool_fini_basic [15:57:51] ==================== [PASSED] ttm_pool ===================== [15:57:51] ================ ttm_resource (8 subtests) ================= [15:57:51] ================= ttm_resource_init_basic ================= [15:57:51] [PASSED] Init resource in TTM_PL_SYSTEM [15:57:51] [PASSED] Init resource in TTM_PL_VRAM [15:57:51] [PASSED] Init resource in a private placement [15:57:51] [PASSED] Init resource in TTM_PL_SYSTEM, set placement flags [15:57:51] ============= [PASSED] ttm_resource_init_basic ============= [15:57:51] [PASSED] ttm_resource_init_pinned [15:57:51] [PASSED] ttm_resource_fini_basic [15:57:51] [PASSED] ttm_resource_manager_init_basic [15:57:51] [PASSED] ttm_resource_manager_usage_basic [15:57:51] [PASSED] ttm_resource_manager_set_used_basic [15:57:51] [PASSED] ttm_sys_man_alloc_basic [15:57:51] [PASSED] ttm_sys_man_free_basic [15:57:51] ================== [PASSED] ttm_resource =================== [15:57:51] =================== ttm_tt (15 subtests) =================== [15:57:51] ==================== ttm_tt_init_basic ==================== [15:57:51] [PASSED] Page-aligned size [15:57:51] [PASSED] Extra pages requested [15:57:51] ================ [PASSED] ttm_tt_init_basic ================ [15:57:51] [PASSED] ttm_tt_init_misaligned [15:57:51] [PASSED] ttm_tt_fini_basic [15:57:51] [PASSED] ttm_tt_fini_sg [15:57:51] [PASSED] ttm_tt_fini_shmem [15:57:51] [PASSED] ttm_tt_create_basic [15:57:51] [PASSED] ttm_tt_create_invalid_bo_type [15:57:51] [PASSED] ttm_tt_create_ttm_exists [15:57:51] [PASSED] ttm_tt_create_failed [15:57:51] [PASSED] ttm_tt_destroy_basic [15:57:51] [PASSED] ttm_tt_populate_null_ttm [15:57:51] [PASSED] ttm_tt_populate_populated_ttm [15:57:51] [PASSED] ttm_tt_unpopulate_basic [15:57:51] [PASSED] ttm_tt_unpopulate_empty_ttm [15:57:51] [PASSED] ttm_tt_swapin_basic [15:57:51] ===================== [PASSED] ttm_tt ====================== [15:57:51] =================== ttm_bo (14 subtests) =================== [15:57:51] =========== ttm_bo_reserve_optimistic_no_ticket =========== [15:57:51] [PASSED] Cannot be interrupted and sleeps [15:57:51] [PASSED] Cannot be interrupted, locks straight away [15:57:51] [PASSED] Can be interrupted, sleeps [15:57:51] ======= [PASSED] ttm_bo_reserve_optimistic_no_ticket ======= [15:57:51] [PASSED] ttm_bo_reserve_locked_no_sleep [15:57:51] [PASSED] ttm_bo_reserve_no_wait_ticket [15:57:51] [PASSED] ttm_bo_reserve_double_resv [15:57:51] [PASSED] ttm_bo_reserve_interrupted [15:57:51] [PASSED] ttm_bo_reserve_deadlock [15:57:51] [PASSED] ttm_bo_unreserve_basic [15:57:51] [PASSED] ttm_bo_unreserve_pinned [15:57:51] [PASSED] ttm_bo_unreserve_bulk [15:57:51] [PASSED] ttm_bo_put_basic [15:57:51] [PASSED] ttm_bo_put_shared_resv [15:57:51] [PASSED] ttm_bo_pin_basic [15:57:51] [PASSED] ttm_bo_pin_unpin_resource [15:57:51] [PASSED] ttm_bo_multiple_pin_one_unpin [15:57:51] ===================== [PASSED] ttm_bo ====================== [15:57:51] ============== ttm_bo_validate (22 subtests) =============== [15:57:51] ============== ttm_bo_init_reserved_sys_man =============== [15:57:51] [PASSED] Buffer object for userspace [15:57:51] [PASSED] Kernel buffer object [15:57:51] [PASSED] Shared buffer object [15:57:51] ========== [PASSED] ttm_bo_init_reserved_sys_man =========== [15:57:51] ============== ttm_bo_init_reserved_mock_man ============== [15:57:51] [PASSED] Buffer object for userspace [15:57:51] [PASSED] Kernel buffer object [15:57:51] [PASSED] Shared buffer object [15:57:51] ========== [PASSED] ttm_bo_init_reserved_mock_man ========== [15:57:51] [PASSED] ttm_bo_init_reserved_resv [15:57:51] ================== ttm_bo_validate_basic ================== [15:57:51] [PASSED] Buffer object for userspace [15:57:51] [PASSED] Kernel buffer object [15:57:51] [PASSED] Shared buffer object [15:57:51] ============== [PASSED] ttm_bo_validate_basic ============== [15:57:51] [PASSED] ttm_bo_validate_invalid_placement [15:57:51] ============= ttm_bo_validate_same_placement ============== [15:57:51] [PASSED] System manager [15:57:51] [PASSED] VRAM manager [15:57:51] ========= [PASSED] ttm_bo_validate_same_placement ========== [15:57:51] [PASSED] ttm_bo_validate_failed_alloc [15:57:51] [PASSED] ttm_bo_validate_pinned [15:57:51] [PASSED] ttm_bo_validate_busy_placement [15:57:51] ================ ttm_bo_validate_multihop ================= [15:57:51] [PASSED] Buffer object for userspace [15:57:51] [PASSED] Kernel buffer object [15:57:51] [PASSED] Shared buffer object [15:57:51] ============ [PASSED] ttm_bo_validate_multihop ============= [15:57:51] ========== ttm_bo_validate_no_placement_signaled ========== [15:57:51] [PASSED] Buffer object in system domain, no page vector [15:57:51] [PASSED] Buffer object in system domain with an existing page vector [15:57:51] ====== [PASSED] ttm_bo_validate_no_placement_signaled ====== [15:57:51] ======== ttm_bo_validate_no_placement_not_signaled ======== [15:57:51] [PASSED] Buffer object for userspace [15:57:51] [PASSED] Kernel buffer object [15:57:51] [PASSED] Shared buffer object [15:57:51] ==== [PASSED] ttm_bo_validate_no_placement_not_signaled ==== [15:57:51] [PASSED] ttm_bo_validate_move_fence_signaled [15:57:51] ========= ttm_bo_validate_move_fence_not_signaled ========= [15:57:51] [PASSED] Waits for GPU [15:57:51] [PASSED] Tries to lock straight away [15:57:51] ===== [PASSED] ttm_bo_validate_move_fence_not_signaled ===== [15:57:51] [PASSED] ttm_bo_validate_swapout [15:57:51] [PASSED] ttm_bo_validate_happy_evict [15:57:51] [PASSED] ttm_bo_validate_all_pinned_evict [15:57:51] [PASSED] ttm_bo_validate_allowed_only_evict [15:57:51] [PASSED] ttm_bo_validate_deleted_evict [15:57:51] [PASSED] ttm_bo_validate_busy_domain_evict [15:57:51] [PASSED] ttm_bo_validate_evict_gutting [15:57:51] [PASSED] ttm_bo_validate_recrusive_evict stty: 'standard input': Inappropriate ioctl for device [15:57:51] ================= [PASSED] ttm_bo_validate ================= [15:57:51] ============================================================ [15:57:51] Testing complete. Ran 102 tests: passed: 102 [15:57:51] Elapsed time: 10.385s total, 1.730s configuring, 7.988s building, 0.567s running
+ cleanup
++ stat -c %u:%g /kernel
+ chown -R 1003:1003 /kernel
Patchwork
CI✓ Xe.CI.BAT: success for drm/gpusvm, drm/pagemap, drm/xe: Restructure migration in preparation for multi-device (rev3)
In reply to Thomas Hellström== Series Details ==
== Series Details ==
Series: drm/gpusvm, drm/pagemap, drm/xe: Restructure migration in preparation for multi-device (rev3)
URL : https://patchwork.freedesktop.org/series/147902/
State : success
== Summary ==
CI Bug Log - changes from xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4_BAT -> xe-pw-147902v3_BAT
====================================================
Summary
-------
**SUCCESS**
No regressions found.
Participating hosts (8 -> 8)
------------------------------
No changes in participating hosts
Changes
-------
No changes found
Build changes
-------------
* Linux: xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4 -> xe-pw-147902v3
IGT_8411: d5b5d2bb4f8795a98ea58376a128b74f654b7ec1 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4: 3cd32052aee17b5e94d85ffd8ca83fb1b37189d4
xe-pw-147902v3: 147902v3
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/index.html
Patchwork
CI✗ Xe.CI.Full: failure for drm/gpusvm, drm/pagemap, drm/xe: Restructure migration in preparation for multi-device (rev3)
In reply to Thomas Hellström== Series Details ==
== Series Details ==
Series: drm/gpusvm, drm/pagemap, drm/xe: Restructure migration in preparation for multi-device (rev3)
URL : https://patchwork.freedesktop.org/series/147902/
State : failure
== Summary ==
CI Bug Log - changes from xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4_FULL -> xe-pw-147902v3_FULL
====================================================
Summary
-------
**FAILURE**
Serious unknown changes coming with xe-pw-147902v3_FULL absolutely need to be verified manually.
If you think the reported changes have nothing to do with the changes
introduced in xe-pw-147902v3_FULL, please notify your bug team ([email protected]) to allow them to document this new failure mode, which will reduce false positives in CI.
Participating hosts (4 -> 4)
------------------------------
No changes in participating hosts
Possible new issues
-------------------
Here are the unknown changes that may have been introduced in xe-pw-147902v3_FULL:
### IGT changes ###
#### Possible regressions ####
* igt@kms_big_fb@linear-16bpp-rotate-180:
- shard-bmg: [PASS][1] -> [DMESG-WARN][2] +35 other tests dmesg-warn
[1]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-6/igt@[email protected]
[2]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-3/igt@[email protected]
* igt@kms_cursor_edge_walk@256x256-right-edge:
- shard-lnl: [PASS][3] -> [DMESG-WARN][4] +66 other tests dmesg-warn
[3]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-lnl-8/igt@[email protected]
[4]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-1/igt@[email protected]
* igt@kms_flip@basic-plain-flip@a-dp2:
- shard-bmg: NOTRUN -> [DMESG-WARN][5]
[5]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-3/igt@kms_flip@[email protected]
* igt@kms_pm_rpm@universal-planes-dpms:
- shard-bmg: [PASS][6] -> [SKIP][7]
[6]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-4/igt@[email protected]
[7]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-6/igt@[email protected]
* igt@kms_vblank@query-forked:
- shard-lnl: NOTRUN -> [DMESG-WARN][8] +3 other tests dmesg-warn
[8]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-8/igt@[email protected]
* igt@xe_exec_store@long-shader-bb-check@gt0-drm_xe_engine_class_render0-bb-system-target-system:
- shard-bmg: [PASS][9] -> [DMESG-FAIL][10] +24 other tests dmesg-fail
[9]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-3/igt@xe_exec_store@long-shader-bb-check@gt0-drm_xe_engine_class_render0-bb-system-target-system.html
[10]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-6/igt@xe_exec_store@long-shader-bb-check@gt0-drm_xe_engine_class_render0-bb-system-target-system.html
* igt@xe_exec_system_allocator@many-execqueues-mmap-shared:
- shard-lnl: [PASS][11] -> [DMESG-FAIL][12] +32 other tests dmesg-fail
[11]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-lnl-8/igt@[email protected]
[12]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-1/igt@[email protected]
* igt@xe_exec_system_allocator@process-many-execqueues-mmap-remap:
- shard-lnl: [PASS][13] -> [INCOMPLETE][14] +1058 other tests incomplete
[13]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-lnl-4/igt@[email protected]
[14]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-4/igt@[email protected]
* igt@xe_exec_system_allocator@process-many-mmap-shared-remap:
- shard-bmg: [PASS][15] -> [FAIL][16] +20 other tests fail
[15]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-5/igt@[email protected]
[16]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-5/igt@[email protected]
* igt@xe_exec_system_allocator@process-many-mmap-shared-remap-eocheck:
- shard-lnl: NOTRUN -> [INCOMPLETE][17] +8 other tests incomplete
[17]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-8/igt@xe_exec_system_allocator@process-many-mmap-shared-remap-eocheck.html
* igt@xe_exec_system_allocator@threads-shared-vm-many-malloc-race:
- shard-bmg: [PASS][18] -> [INCOMPLETE][19] +888 other tests incomplete
[18]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-4/igt@[email protected]
[19]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-5/igt@[email protected]
* igt@xe_fault_injection@exec-queue-create-fail-xe_vm_add_compute_exec_queue:
- shard-lnl: [PASS][20] -> [ABORT][21] +1 other test abort
[20]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-lnl-8/igt@xe_fault_injection@exec-queue-create-fail-xe_vm_add_compute_exec_queue.html
[21]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-1/igt@xe_fault_injection@exec-queue-create-fail-xe_vm_add_compute_exec_queue.html
* igt@xe_fault_injection@inject-fault-probe-function-xe_device_probe_early:
- shard-bmg: [PASS][22] -> [ABORT][23] +6 other tests abort
[22]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-2/igt@xe_fault_injection@inject-fault-probe-function-xe_device_probe_early.html
[23]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-5/igt@xe_fault_injection@inject-fault-probe-function-xe_device_probe_early.html
* igt@xe_pmu@engine-activity-idle:
- shard-lnl: [PASS][24] -> [FAIL][25] +87 other tests fail
[24]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-lnl-1/igt@[email protected]
[25]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-3/igt@[email protected]
* igt@xe_query@query-cs-cycles:
- shard-lnl: [PASS][26] -> [TIMEOUT][27]
[26]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-lnl-7/igt@[email protected]
[27]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-2/igt@[email protected]
#### Warnings ####
* igt@kms_content_protection@uevent:
- shard-bmg: [FAIL][28] ([Intel XE#1188]) -> [DMESG-FAIL][29] +1 other test dmesg-fail
[28]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-2/igt@[email protected]
[29]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-5/igt@[email protected]
* igt@kms_dp_linktrain_fallback@dp-fallback:
- shard-bmg: [SKIP][30] ([Intel XE#4294]) -> [DMESG-WARN][31]
[30]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-6/igt@[email protected]
[31]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-2/igt@[email protected]
#### Suppressed ####
The following results come from untrusted machines, tests, or statuses. They do not affect the overall result.
* {igt@xe_pmu@engine-activity-multi-client}:
- shard-lnl: [PASS][32] -> [TIMEOUT][33] +1 other test timeout
[32]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-lnl-7/igt@[email protected]
[33]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-2/igt@[email protected]
New tests
---------
New tests have been introduced between xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4_FULL and xe-pw-147902v3_FULL:
### New IGT tests (1) ###
* igt@xe_exec_balancer:
- Statuses :
- Exec time: [None] s
Known issues
------------
Here are the changes found in xe-pw-147902v3_FULL that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@kms_atomic_transition@plane-all-modeset-transition-fencing:
- shard-bmg: [PASS][34] -> [FAIL][35] ([Intel XE#3908]) +1 other test fail
[34]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-4/igt@[email protected]
[35]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-7/igt@[email protected]
* igt@kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-180-async-flip:
- shard-adlp: NOTRUN -> [SKIP][36] ([Intel XE#1124])
[36]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@kms_big_fb@4-tiled-max-hw-stride-64bpp-rotate-180-async-flip.html
* igt@kms_big_fb@x-tiled-16bpp-rotate-270:
- shard-dg2-set2: NOTRUN -> [SKIP][37] ([Intel XE#316])
[37]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@kms_big_fb@yf-tiled-64bpp-rotate-90:
- shard-dg2-set2: NOTRUN -> [SKIP][38] ([Intel XE#1124]) +2 other tests skip
[38]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@kms_bw@linear-tiling-4-displays-2560x1440p:
- shard-dg2-set2: NOTRUN -> [SKIP][39] ([Intel XE#367])
[39]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@kms_ccs@bad-aux-stride-y-tiled-gen12-mc-ccs@pipe-c-hdmi-a-6:
- shard-dg2-set2: NOTRUN -> [SKIP][40] ([Intel XE#787]) +13 other tests skip
[40]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@kms_ccs@[email protected]
* igt@kms_ccs@bad-rotation-90-4-tiled-lnl-ccs@pipe-c-dp-2:
- shard-bmg: NOTRUN -> [SKIP][41] ([Intel XE#2652] / [Intel XE#787]) +3 other tests skip
[41]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-2/igt@kms_ccs@[email protected]
* igt@kms_ccs@ccs-on-another-bo-y-tiled-gen12-rc-ccs-cc@pipe-d-dp-4:
- shard-dg2-set2: NOTRUN -> [SKIP][42] ([Intel XE#455] / [Intel XE#787]) +3 other tests skip
[42]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@kms_ccs@[email protected]
* igt@kms_ccs@crc-primary-basic-4-tiled-dg2-mc-ccs:
- shard-lnl: NOTRUN -> [SKIP][43] ([Intel XE#2887]) +1 other test skip
[43]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-8/igt@[email protected]
* igt@kms_ccs@crc-primary-rotation-180-4-tiled-dg2-mc-ccs:
- shard-adlp: NOTRUN -> [SKIP][44] ([Intel XE#455] / [Intel XE#787]) +1 other test skip
[44]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
* igt@kms_ccs@crc-primary-rotation-180-4-tiled-dg2-mc-ccs@pipe-b-hdmi-a-1:
- shard-adlp: NOTRUN -> [SKIP][45] ([Intel XE#787]) +2 other tests skip
[45]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@kms_ccs@[email protected]
* igt@kms_ccs@crc-primary-suspend-4-tiled-bmg-ccs:
- shard-bmg: [PASS][46] -> [INCOMPLETE][47] ([Intel XE#3862]) +1 other test incomplete
[46]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-2/igt@[email protected]
[47]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-6/igt@[email protected]
* igt@kms_ccs@crc-sprite-planes-basic-4-tiled-lnl-ccs:
- shard-dg2-set2: NOTRUN -> [SKIP][48] ([Intel XE#2907])
[48]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@kms_ccs@random-ccs-data-4-tiled-dg2-mc-ccs:
- shard-dg2-set2: [PASS][49] -> [INCOMPLETE][50] ([Intel XE#2705] / [Intel XE#4212])
[49]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-dg2-435/igt@[email protected]
[50]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@kms_ccs@random-ccs-data-4-tiled-dg2-mc-ccs@pipe-a-dp-4:
- shard-dg2-set2: [PASS][51] -> [INCOMPLETE][52] ([Intel XE#4212])
[51]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-dg2-435/igt@kms_ccs@[email protected]
[52]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@kms_ccs@[email protected]
* igt@kms_ccs@random-ccs-data-4-tiled-dg2-rc-ccs:
- shard-dg2-set2: [PASS][53] -> [INCOMPLETE][54] ([Intel XE#1727] / [Intel XE#2705] / [Intel XE#3113] / [Intel XE#4212] / [Intel XE#4345] / [Intel XE#4522]) +1 other test incomplete
[53]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-dg2-435/igt@[email protected]
[54]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-433/igt@[email protected]
* igt@kms_chamelium_color@ctm-green-to-red:
- shard-dg2-set2: NOTRUN -> [SKIP][55] ([Intel XE#306])
[55]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@kms_chamelium_edid@hdmi-edid-change-during-hibernate:
- shard-adlp: NOTRUN -> [SKIP][56] ([Intel XE#373])
[56]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
* igt@kms_chamelium_hpd@dp-hpd-enable-disable-mode:
- shard-dg2-set2: NOTRUN -> [SKIP][57] ([Intel XE#373]) +2 other tests skip
[57]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@kms_content_protection@dp-mst-lic-type-1:
- shard-adlp: NOTRUN -> [SKIP][58] ([Intel XE#307])
[58]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
* igt@kms_content_protection@srm@pipe-a-dp-2:
- shard-bmg: NOTRUN -> [FAIL][59] ([Intel XE#1178]) +1 other test fail
[59]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-2/igt@kms_content_protection@[email protected]
* igt@kms_content_protection@uevent:
- shard-lnl: NOTRUN -> [SKIP][60] ([Intel XE#5176])
[60]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-8/igt@[email protected]
* igt@kms_cursor_legacy@cursorb-vs-flipb-atomic-transitions-varying-size:
- shard-bmg: [PASS][61] -> [SKIP][62] ([Intel XE#2291]) +2 other tests skip
[61]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-7/igt@kms_cursor_legacy@cursorb-vs-flipb-atomic-transitions-varying-size.html
[62]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-6/igt@kms_cursor_legacy@cursorb-vs-flipb-atomic-transitions-varying-size.html
* igt@kms_dp_link_training@non-uhbr-sst:
- shard-lnl: NOTRUN -> [SKIP][63] ([Intel XE#4354])
[63]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-8/igt@[email protected]
* igt@kms_dp_linktrain_fallback@dsc-fallback:
- shard-dg2-set2: NOTRUN -> [SKIP][64] ([Intel XE#4331])
[64]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@kms_flip@2x-flip-vs-dpms-on-nop:
- shard-bmg: [PASS][65] -> [SKIP][66] ([Intel XE#2316]) +5 other tests skip
[65]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-4/igt@[email protected]
[66]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-6/igt@[email protected]
* igt@kms_flip@2x-flip-vs-expired-vblank-interruptible@ab-hdmi-a6-dp4:
- shard-dg2-set2: [PASS][67] -> [FAIL][68] ([Intel XE#301] / [Intel XE#3321]) +1 other test fail
[67]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-dg2-436/igt@kms_flip@[email protected]
[68]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@kms_flip@[email protected]
* igt@kms_flip@2x-flip-vs-expired-vblank-interruptible@cd-hdmi-a6-dp4:
- shard-dg2-set2: [PASS][69] -> [FAIL][70] ([Intel XE#301]) +6 other tests fail
[69]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-dg2-436/igt@kms_flip@[email protected]
[70]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@kms_flip@[email protected]
* igt@kms_flip@2x-flip-vs-suspend-interruptible:
- shard-lnl: NOTRUN -> [SKIP][71] ([Intel XE#1421])
[71]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-8/igt@[email protected]
* igt@kms_flip@flip-vs-expired-vblank-interruptible@a-edp1:
- shard-lnl: [PASS][72] -> [FAIL][73] ([Intel XE#301])
[72]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-lnl-4/igt@kms_flip@[email protected]
[73]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-7/igt@kms_flip@[email protected]
* igt@kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling@pipe-a-valid-mode:
- shard-dg2-set2: NOTRUN -> [SKIP][74] ([Intel XE#455]) +1 other test skip
[74]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@kms_flip_scaled_crc@flip-32bpp-ytileccs-to-64bpp-ytile-upscaling@pipe-a-valid-mode.html
* igt@kms_frontbuffer_tracking@drrs-1p-primscrn-cur-indfb-draw-render:
- shard-lnl: NOTRUN -> [SKIP][75] ([Intel XE#651]) +1 other test skip
[75]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-8/igt@kms_frontbuffer_tracking@drrs-1p-primscrn-cur-indfb-draw-render.html
* igt@kms_frontbuffer_tracking@drrs-1p-primscrn-spr-indfb-onoff:
- shard-dg2-set2: NOTRUN -> [SKIP][76] ([Intel XE#651]) +7 other tests skip
[76]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@kms_frontbuffer_tracking@drrs-2p-scndscrn-shrfb-plflip-blt:
- shard-adlp: NOTRUN -> [SKIP][77] ([Intel XE#656]) +2 other tests skip
[77]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
* igt@kms_frontbuffer_tracking@fbc-2p-primscrn-indfb-pgflip-blt:
- shard-dg2-set2: [PASS][78] -> [INCOMPLETE][79] ([Intel XE#4842])
[78]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-dg2-435/igt@[email protected]
[79]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-435/igt@[email protected]
* igt@kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-pri-indfb-draw-mmap-wc:
- shard-adlp: NOTRUN -> [SKIP][80] ([Intel XE#651])
[80]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@kms_frontbuffer_tracking@fbcdrrs-1p-primscrn-pri-indfb-draw-mmap-wc.html
* igt@kms_frontbuffer_tracking@psr-2p-primscrn-pri-indfb-draw-blt:
- shard-dg2-set2: NOTRUN -> [SKIP][81] ([Intel XE#653]) +7 other tests skip
[81]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@kms_frontbuffer_tracking@psr-rgb101010-draw-render:
- shard-adlp: NOTRUN -> [SKIP][82] ([Intel XE#653]) +2 other tests skip
[82]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
* igt@kms_hdr@static-toggle:
- shard-bmg: [PASS][83] -> [SKIP][84] ([Intel XE#1503]) +1 other test skip
[83]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-4/igt@[email protected]
[84]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-6/igt@[email protected]
* igt@kms_joiner@basic-force-ultra-joiner:
- shard-lnl: NOTRUN -> [SKIP][85] ([Intel XE#2934])
[85]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-8/igt@[email protected]
* igt@kms_joiner@invalid-modeset-force-big-joiner:
- shard-adlp: NOTRUN -> [SKIP][86] ([Intel XE#3012])
[86]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
* igt@kms_joiner@switch-modeset-ultra-joiner-big-joiner:
- shard-dg2-set2: NOTRUN -> [SKIP][87] ([Intel XE#2925])
[87]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@kms_lease@lease-uevent:
- shard-adlp: [PASS][88] -> [DMESG-WARN][89] ([Intel XE#2953] / [Intel XE#4173]) +2 other tests dmesg-warn
[88]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-adlp-3/igt@[email protected]
[89]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-3/igt@[email protected]
* igt@kms_plane_multiple@2x-tiling-4:
- shard-bmg: [PASS][90] -> [SKIP][91] ([Intel XE#4596])
[90]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-3/igt@[email protected]
[91]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-6/igt@[email protected]
* igt@kms_pm_backlight@bad-brightness:
- shard-dg2-set2: NOTRUN -> [SKIP][92] ([Intel XE#870])
[92]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@kms_psr2_sf@psr2-cursor-plane-move-continuous-sf:
- shard-dg2-set2: NOTRUN -> [SKIP][93] ([Intel XE#1489])
[93]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@kms_psr@fbc-pr-sprite-render:
- shard-adlp: NOTRUN -> [SKIP][94] ([Intel XE#2850] / [Intel XE#929]) +1 other test skip
[94]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
* igt@kms_psr@pr-primary-blt:
- shard-lnl: NOTRUN -> [SKIP][95] ([Intel XE#1406])
[95]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-8/igt@[email protected]
* igt@kms_psr@psr-dpms:
- shard-dg2-set2: NOTRUN -> [SKIP][96] ([Intel XE#2850] / [Intel XE#929]) +3 other tests skip
[96]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@kms_scaling_modes@scaling-mode-center:
- shard-adlp: NOTRUN -> [SKIP][97] ([Intel XE#455]) +2 other tests skip
[97]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
* igt@kms_vrr@cmrr@pipe-a-edp-1:
- shard-lnl: [PASS][98] -> [FAIL][99] ([Intel XE#4459]) +1 other test fail
[98]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-lnl-7/igt@kms_vrr@[email protected]
[99]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-1/igt@kms_vrr@[email protected]
* igt@sriov_basic@enable-vfs-autoprobe-off:
- shard-dg2-set2: NOTRUN -> [SKIP][100] ([Intel XE#1091] / [Intel XE#2849])
[100]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@xe_copy_basic@mem-set-linear-0x3fff:
- shard-dg2-set2: NOTRUN -> [SKIP][101] ([Intel XE#1126])
[101]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@xe_eu_stall@invalid-gt-id:
- shard-dg2-set2: NOTRUN -> [SKIP][102] ([Intel XE#4497])
[102]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@xe_eudebug@basic-vm-bind-extended-discovery:
- shard-lnl: NOTRUN -> [SKIP][103] ([Intel XE#4837])
[103]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-8/igt@[email protected]
* igt@xe_eudebug@discovery-race-sigint:
- shard-adlp: NOTRUN -> [SKIP][104] ([Intel XE#4837]) +1 other test skip
[104]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
* igt@xe_eudebug_online@single-step-one:
- shard-dg2-set2: NOTRUN -> [SKIP][105] ([Intel XE#4837]) +2 other tests skip
[105]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@xe_evict@evict-beng-small-external:
- shard-adlp: NOTRUN -> [SKIP][106] ([Intel XE#261] / [Intel XE#688])
[106]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
* igt@xe_exec_basic@multigpu-many-execqueues-many-vm-bindexecqueue-userptr-invalidate-race:
- shard-lnl: NOTRUN -> [SKIP][107] ([Intel XE#1392])
[107]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-8/igt@xe_exec_basic@multigpu-many-execqueues-many-vm-bindexecqueue-userptr-invalidate-race.html
* igt@xe_exec_basic@multigpu-once-userptr-invalidate-race:
- shard-adlp: NOTRUN -> [SKIP][108] ([Intel XE#1392])
[108]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
* igt@xe_exec_fault_mode@twice-bindexecqueue-rebind-imm:
- shard-adlp: NOTRUN -> [SKIP][109] ([Intel XE#288])
[109]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
* igt@xe_exec_fault_mode@twice-userptr-prefetch:
- shard-dg2-set2: NOTRUN -> [SKIP][110] ([Intel XE#288]) +5 other tests skip
[110]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@xe_exec_system_allocator@many-execqueues-malloc-busy:
- shard-lnl: [PASS][111] -> [INCOMPLETE][112] ([Intel XE#5111])
[111]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-lnl-6/igt@[email protected]
[112]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-3/igt@[email protected]
* igt@xe_exec_system_allocator@once-new-bo-map-nomemset:
- shard-adlp: NOTRUN -> [SKIP][113] ([Intel XE#4915]) +24 other tests skip
[113]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
* igt@xe_exec_system_allocator@threads-many-stride-mmap-huge-nomemset:
- shard-lnl: NOTRUN -> [SKIP][114] ([Intel XE#4943])
[114]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-8/igt@xe_exec_system_allocator@threads-many-stride-mmap-huge-nomemset.html
* igt@xe_exec_system_allocator@threads-many-stride-mmap-remap-eocheck:
- shard-dg2-set2: NOTRUN -> [SKIP][115] ([Intel XE#4915]) +57 other tests skip
[115]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@xe_exec_system_allocator@threads-many-stride-mmap-remap-eocheck.html
* igt@xe_fault_injection@probe-fail-guc-xe_guc_ct_send_recv:
- shard-adlp: NOTRUN -> [ABORT][116] ([Intel XE#4917])
[116]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@xe_fault_injection@probe-fail-guc-xe_guc_ct_send_recv.html
* igt@xe_media_fill@media-fill:
- shard-dg2-set2: NOTRUN -> [SKIP][117] ([Intel XE#560])
[117]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@xe_oa@oa-exponents:
- shard-adlp: NOTRUN -> [SKIP][118] ([Intel XE#2541] / [Intel XE#3573])
[118]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
* igt@xe_oa@oa-formats:
- shard-dg2-set2: NOTRUN -> [SKIP][119] ([Intel XE#2541] / [Intel XE#3573])
[119]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@xe_pm@d3cold-basic:
- shard-adlp: NOTRUN -> [SKIP][120] ([Intel XE#2284] / [Intel XE#366])
[120]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
* igt@xe_pm@s4-mocs:
- shard-adlp: [PASS][121] -> [ABORT][122] ([Intel XE#1794])
[121]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-adlp-1/igt@[email protected]
[122]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-9/igt@[email protected]
* igt@xe_pmu@gt-frequency:
- shard-dg2-set2: [PASS][123] -> [FAIL][124] ([Intel XE#4819]) +1 other test fail
[123]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-dg2-433/igt@[email protected]
[124]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-435/igt@[email protected]
* igt@xe_pxp@pxp-stale-bo-bind-post-suspend:
- shard-dg2-set2: NOTRUN -> [SKIP][125] ([Intel XE#4733])
[125]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@[email protected]
* igt@xe_pxp@pxp-stale-queue-post-suspend:
- shard-adlp: NOTRUN -> [SKIP][126] ([Intel XE#4733])
[126]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
* igt@xe_query@multigpu-query-invalid-uc-fw-version-mbz:
- shard-adlp: NOTRUN -> [SKIP][127] ([Intel XE#944])
[127]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@[email protected]
#### Possible fixes ####
* igt@kms_async_flips@async-flip-with-page-flip-events-linear@pipe-c-edp-1:
- shard-lnl: [FAIL][128] ([Intel XE#911]) -> [PASS][129] +3 other tests pass
[128]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-lnl-7/igt@kms_async_flips@[email protected]
[129]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-8/igt@kms_async_flips@[email protected]
* igt@kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-180-hflip-async-flip:
- shard-adlp: [DMESG-FAIL][130] ([Intel XE#4543]) -> [PASS][131]
[130]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-adlp-1/igt@kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-180-hflip-async-flip.html
[131]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-4/igt@kms_big_fb@y-tiled-max-hw-stride-64bpp-rotate-180-hflip-async-flip.html
* igt@kms_bw@connected-linear-tiling-2-displays-2160x1440p:
- shard-bmg: [SKIP][132] ([Intel XE#2314] / [Intel XE#2894]) -> [PASS][133]
[132]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-6/igt@[email protected]
[133]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-2/igt@[email protected]
* igt@kms_ccs@crc-primary-suspend-4-tiled-dg2-rc-ccs-cc@pipe-d-dp-4:
- shard-dg2-set2: [INCOMPLETE][134] ([Intel XE#3862]) -> [PASS][135] +1 other test pass
[134]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-dg2-436/igt@kms_ccs@[email protected]
[135]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-464/igt@kms_ccs@[email protected]
* igt@kms_cursor_legacy@cursora-vs-flipb-legacy:
- shard-bmg: [SKIP][136] ([Intel XE#2291]) -> [PASS][137] +3 other tests pass
[136]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-6/igt@[email protected]
[137]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-3/igt@[email protected]
* igt@kms_flip@2x-flip-vs-blocking-wf-vblank@ab-hdmi-a6-dp4:
- shard-dg2-set2: [FAIL][138] ([Intel XE#886]) -> [PASS][139] +1 other test pass
[138]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-dg2-435/igt@kms_flip@[email protected]
[139]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-435/igt@kms_flip@[email protected]
* igt@kms_flip@2x-flip-vs-dpms:
- shard-bmg: [SKIP][140] ([Intel XE#2316]) -> [PASS][141] +2 other tests pass
[140]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-6/igt@[email protected]
[141]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-2/igt@[email protected]
* igt@kms_flip@flip-vs-absolute-wf_vblank:
- shard-lnl: [FAIL][142] ([Intel XE#886]) -> [PASS][143] +3 other tests pass
[142]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-lnl-1/igt@kms_flip@flip-vs-absolute-wf_vblank.html
[143]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-3/igt@kms_flip@flip-vs-absolute-wf_vblank.html
* igt@kms_flip@flip-vs-expired-vblank-interruptible@b-hdmi-a6:
- shard-dg2-set2: [FAIL][144] ([Intel XE#301]) -> [PASS][145]
[144]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-dg2-436/igt@kms_flip@[email protected]
[145]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-436/igt@kms_flip@[email protected]
* igt@kms_flip@wf_vblank-ts-check@a-hdmi-a1:
- shard-adlp: [FAIL][146] ([Intel XE#886]) -> [PASS][147] +1 other test pass
[146]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-adlp-9/igt@kms_flip@[email protected]
[147]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-4/igt@kms_flip@[email protected]
* igt@kms_hdr@static-toggle-dpms:
- shard-bmg: [SKIP][148] ([Intel XE#1503]) -> [PASS][149]
[148]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-6/igt@[email protected]
[149]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-2/igt@[email protected]
* igt@kms_joiner@invalid-modeset-force-big-joiner:
- shard-bmg: [SKIP][150] ([Intel XE#3012]) -> [PASS][151]
[150]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-6/igt@[email protected]
[151]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-4/igt@[email protected]
* igt@kms_plane_scaling@plane-downscale-factor-0-5-with-modifiers@pipe-b:
- shard-adlp: [DMESG-WARN][152] ([Intel XE#2953] / [Intel XE#4173]) -> [PASS][153] +4 other tests pass
[152]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-adlp-1/igt@kms_plane_scaling@[email protected]
[153]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-2/igt@kms_plane_scaling@[email protected]
* igt@kms_setmode@clone-exclusive-crtc:
- shard-bmg: [SKIP][154] ([Intel XE#1435]) -> [PASS][155]
[154]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-6/igt@[email protected]
[155]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-2/igt@[email protected]
* igt@xe_exec_threads@threads-hang-fd-userptr-invalidate:
- shard-dg2-set2: [DMESG-WARN][156] ([Intel XE#3876]) -> [PASS][157]
[156]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-dg2-435/igt@[email protected]
[157]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-dg2-435/igt@[email protected]
* igt@xe_pm@s4-basic-exec:
- shard-lnl: [ABORT][158] ([Intel XE#1794]) -> [PASS][159]
[158]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-lnl-2/igt@[email protected]
[159]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-lnl-8/igt@[email protected]
#### Warnings ####
* igt@kms_content_protection@legacy:
- shard-bmg: [SKIP][160] ([Intel XE#2341]) -> [FAIL][161] ([Intel XE#1178]) +1 other test fail
[160]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-6/igt@[email protected]
[161]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-2/igt@[email protected]
* igt@kms_frontbuffer_tracking@drrs-2p-primscrn-cur-indfb-draw-render:
- shard-bmg: [SKIP][162] ([Intel XE#2311]) -> [SKIP][163] ([Intel XE#2312]) +9 other tests skip
[162]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-4/igt@kms_frontbuffer_tracking@drrs-2p-primscrn-cur-indfb-draw-render.html
[163]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-6/igt@kms_frontbuffer_tracking@drrs-2p-primscrn-cur-indfb-draw-render.html
* igt@kms_frontbuffer_tracking@drrs-2p-primscrn-pri-indfb-draw-mmap-wc:
- shard-bmg: [SKIP][164] ([Intel XE#2312]) -> [SKIP][165] ([Intel XE#2311]) +15 other tests skip
[164]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-6/igt@kms_frontbuffer_tracking@drrs-2p-primscrn-pri-indfb-draw-mmap-wc.html
[165]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-3/igt@kms_frontbuffer_tracking@drrs-2p-primscrn-pri-indfb-draw-mmap-wc.html
* igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-draw-mmap-wc:
- shard-bmg: [SKIP][166] ([Intel XE#4141]) -> [SKIP][167] ([Intel XE#2312]) +7 other tests skip
[166]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-4/igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-draw-mmap-wc.html
[167]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-6/igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-draw-mmap-wc.html
* igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-spr-indfb-onoff:
- shard-bmg: [SKIP][168] ([Intel XE#2312]) -> [SKIP][169] ([Intel XE#4141]) +8 other tests skip
[168]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-6/igt@[email protected]
[169]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-2/igt@[email protected]
* igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-indfb-draw-render:
- shard-bmg: [SKIP][170] ([Intel XE#2313]) -> [SKIP][171] ([Intel XE#2312]) +11 other tests skip
[170]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-7/igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-indfb-draw-render.html
[171]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-6/igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-pri-indfb-draw-render.html
* igt@kms_frontbuffer_tracking@psr-2p-primscrn-spr-indfb-fullscreen:
- shard-bmg: [SKIP][172] ([Intel XE#2312]) -> [SKIP][173] ([Intel XE#2313]) +14 other tests skip
[172]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-6/igt@kms_frontbuffer_tracking@psr-2p-primscrn-spr-indfb-fullscreen.html
[173]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-2/igt@kms_frontbuffer_tracking@psr-2p-primscrn-spr-indfb-fullscreen.html
* igt@kms_plane_multiple@2x-tiling-y:
- shard-bmg: [SKIP][174] ([Intel XE#5021]) -> [SKIP][175] ([Intel XE#4596])
[174]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-bmg-4/igt@[email protected]
[175]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-bmg-6/igt@[email protected]
* igt@xe_pmu@gt-c6-idle:
- shard-adlp: [DMESG-WARN][176] ([Intel XE#5214]) -> [ABORT][177] ([Intel XE#5214]) +4 other tests abort
[176]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-adlp-4/igt@[email protected]
[177]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-3/igt@[email protected]
* igt@xe_sriov_scheduling@equal-throughput:
- shard-adlp: [ABORT][178] ([Intel XE#5214]) -> [INCOMPLETE][179] ([Intel XE#5214]) +1 other test incomplete
[178]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4/shard-adlp-3/igt@[email protected]
[179]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/shard-adlp-9/igt@[email protected]
{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).
[Intel XE#1091]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1091
[Intel XE#1124]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1124
[Intel XE#1126]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1126
[Intel XE#1178]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1178
[Intel XE#1188]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1188
[Intel XE#1392]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1392
[Intel XE#1406]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1406
[Intel XE#1421]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1421
[Intel XE#1435]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1435
[Intel XE#1489]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1489
[Intel XE#1503]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1503
[Intel XE#1727]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1727
[Intel XE#1794]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1794
[Intel XE#2284]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2284
[Intel XE#2291]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2291
[Intel XE#2311]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2311
[Intel XE#2312]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2312
[Intel XE#2313]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2313
[Intel XE#2314]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2314
[Intel XE#2316]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2316
[Intel XE#2341]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2341
[Intel XE#2541]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2541
[Intel XE#261]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/261
[Intel XE#2652]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2652
[Intel XE#2705]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2705
[Intel XE#2849]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2849
[Intel XE#2850]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2850
[Intel XE#288]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/288
[Intel XE#2887]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2887
[Intel XE#2894]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2894
[Intel XE#2907]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2907
[Intel XE#2925]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2925
[Intel XE#2934]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2934
[Intel XE#2953]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2953
[Intel XE#301]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/301
[Intel XE#3012]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3012
[Intel XE#306]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/306
[Intel XE#307]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/307
[Intel XE#3113]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3113
[Intel XE#316]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/316
[Intel XE#3321]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3321
[Intel XE#3573]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3573
[Intel XE#366]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/366
[Intel XE#367]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/367
[Intel XE#373]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/373
[Intel XE#3862]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3862
[Intel XE#3876]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3876
[Intel XE#3908]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3908
[Intel XE#4141]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4141
[Intel XE#4173]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4173
[Intel XE#4212]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4212
[Intel XE#4294]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4294
[Intel XE#4331]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4331
[Intel XE#4345]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4345
[Intel XE#4354]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4354
[Intel XE#4459]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4459
[Intel XE#4497]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4497
[Intel XE#4522]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4522
[Intel XE#4543]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4543
[Intel XE#455]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/455
[Intel XE#4596]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4596
[Intel XE#4733]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4733
[Intel XE#4819]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4819
[Intel XE#4837]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4837
[Intel XE#4842]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4842
[Intel XE#4915]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4915
[Intel XE#4917]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4917
[Intel XE#4943]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4943
[Intel XE#5021]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5021
[Intel XE#5111]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5111
[Intel XE#5176]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5176
[Intel XE#5214]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5214
[Intel XE#560]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/560
[Intel XE#651]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/651
[Intel XE#653]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/653
[Intel XE#656]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/656
[Intel XE#688]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/688
[Intel XE#787]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/787
[Intel XE#870]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/870
[Intel XE#886]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/886
[Intel XE#911]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/911
[Intel XE#929]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/929
[Intel XE#944]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/944
Build changes
-------------
* Linux: xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4 -> xe-pw-147902v3
IGT_8411: d5b5d2bb4f8795a98ea58376a128b74f654b7ec1 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
xe-3251-3cd32052aee17b5e94d85ffd8ca83fb1b37189d4: 3cd32052aee17b5e94d85ffd8ca83fb1b37189d4
xe-pw-147902v3: 147902v3
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-147902v3/index.html