* [PATCH v1 0/5] drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression
@ 2025-09-18 14:25 Nitin Gote
2025-09-18 14:25 ` [PATCH v1 1/5] drm/xe: add VM_BIND DECOMPRESS uapi flag Nitin Gote
` (7 more replies)
0 siblings, 8 replies; 14+ messages in thread
From: Nitin Gote @ 2025-09-18 14:25 UTC (permalink / raw)
To: intel-xe; +Cc: himal.prasad.ghimiray, matthew.brost, thomas.hellstrom,
Nitin Gote
This series implements VM_BIND decompression support in the XE driver
and adds on‑demand decompression at VMA pagefault time.
Tested with https://patchwork.freedesktop.org/series/154711/
which validates basic VM_BIND DECOMPRESS behavior for both
regular and FAULT_MODE VMs.
Nitin Gote (5):
drm/xe: add VM_BIND DECOMPRESS uapi flag
drm/xe: add xe_migrate_resolve wrapper and is_vram_resolve support
drm/xe: add drm_exec helper to atomically lock VM (+ optional BO) with
BOOKKEEP retry
drm/xe: implement VM_BIND decompression in vm_bind_ioctl
drm/xe: perform on-demand decompression on VMA pagefaults
drivers/gpu/drm/xe/xe_gt_pagefault.c | 41 +++++-
drivers/gpu/drm/xe/xe_migrate.c | 90 +++++++++----
drivers/gpu/drm/xe/xe_migrate.h | 7 +
drivers/gpu/drm/xe/xe_vm.c | 187 ++++++++++++++++++++++++++-
include/uapi/drm/xe_drm.h | 1 +
5 files changed, 292 insertions(+), 34 deletions(-)
--
2.25.1
^ permalink raw reply [flat|nested] 14+ messages in thread
* [PATCH v1 1/5] drm/xe: add VM_BIND DECOMPRESS uapi flag
2025-09-18 14:25 [PATCH v1 0/5] drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression Nitin Gote
@ 2025-09-18 14:25 ` Nitin Gote
2025-09-23 17:35 ` Matthew Brost
2025-09-18 14:25 ` [PATCH v1 2/5] drm/xe: add xe_migrate_resolve wrapper and is_vram_resolve support Nitin Gote
` (6 subsequent siblings)
7 siblings, 1 reply; 14+ messages in thread
From: Nitin Gote @ 2025-09-18 14:25 UTC (permalink / raw)
To: intel-xe; +Cc: himal.prasad.ghimiray, matthew.brost, thomas.hellstrom,
Nitin Gote
Add DRM_XE_VM_BIND_FLAG_DECOMPRESS to the userspace API so callers
can request GPU-side decompression on VM_BIND.
This flag is used by subsequent driver changes to trigger scheduling of
GPU work that resolves compressed VRAM pages into an uncompressed PAT
VM mapping.
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
---
include/uapi/drm/xe_drm.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h
index 40ff19f52a8d..6b652a6968a0 100644
--- a/include/uapi/drm/xe_drm.h
+++ b/include/uapi/drm/xe_drm.h
@@ -1119,6 +1119,7 @@ struct drm_xe_vm_bind_op {
#define DRM_XE_VM_BIND_FLAG_DUMPABLE (1 << 3)
#define DRM_XE_VM_BIND_FLAG_CHECK_PXP (1 << 4)
#define DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR (1 << 5)
+#define DRM_XE_VM_BIND_FLAG_DECOMPRESS (1 << 6)
/** @flags: Bind flags */
__u32 flags;
--
2.25.1
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH v1 2/5] drm/xe: add xe_migrate_resolve wrapper and is_vram_resolve support
2025-09-18 14:25 [PATCH v1 0/5] drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression Nitin Gote
2025-09-18 14:25 ` [PATCH v1 1/5] drm/xe: add VM_BIND DECOMPRESS uapi flag Nitin Gote
@ 2025-09-18 14:25 ` Nitin Gote
2025-09-18 14:25 ` [PATCH v1 3/5] drm/xe: add drm_exec helper to atomically lock VM (+ optional BO) with BOOKKEEP retry Nitin Gote
` (5 subsequent siblings)
7 siblings, 0 replies; 14+ messages in thread
From: Nitin Gote @ 2025-09-18 14:25 UTC (permalink / raw)
To: intel-xe; +Cc: himal.prasad.ghimiray, matthew.brost, thomas.hellstrom,
Nitin Gote
Introduce an internal __xe_migrate_copy(..., is_vram_resolve) path and
expose a small wrapper xe_migrate_resolve() that calls it with
is_vram_resolve=true.
For resolve/decompression operations we must ensure the copy code uses
the compression PAT index when appropriate; this change centralizes that
behavior and allows callers to schedule a resolve (decompress) operation
via the migrate API.
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
---
drivers/gpu/drm/xe/xe_migrate.c | 90 +++++++++++++++++++++++----------
drivers/gpu/drm/xe/xe_migrate.h | 7 +++
2 files changed, 70 insertions(+), 27 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_migrate.c b/drivers/gpu/drm/xe/xe_migrate.c
index 1d667fa36cf3..6558a774dbee 100644
--- a/drivers/gpu/drm/xe/xe_migrate.c
+++ b/drivers/gpu/drm/xe/xe_migrate.c
@@ -764,31 +764,13 @@ static u32 xe_migrate_ccs_copy(struct xe_migrate *m,
return flush_flags;
}
-/**
- * xe_migrate_copy() - Copy content of TTM resources.
- * @m: The migration context.
- * @src_bo: The buffer object @src is currently bound to.
- * @dst_bo: If copying between resources created for the same bo, set this to
- * the same value as @src_bo. If copying between buffer objects, set it to
- * the buffer object @dst is currently bound to.
- * @src: The source TTM resource.
- * @dst: The dst TTM resource.
- * @copy_only_ccs: If true copy only CCS metadata
- *
- * Copies the contents of @src to @dst: On flat CCS devices,
- * the CCS metadata is copied as well if needed, or if not present,
- * the CCS metadata of @dst is cleared for security reasons.
- *
- * Return: Pointer to a dma_fence representing the last copy batch, or
- * an error pointer on failure. If there is a failure, any copy operation
- * started by the function call has been synced.
- */
-struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
- struct xe_bo *src_bo,
- struct xe_bo *dst_bo,
- struct ttm_resource *src,
- struct ttm_resource *dst,
- bool copy_only_ccs)
+static struct dma_fence *__xe_migrate_copy(struct xe_migrate *m,
+ struct xe_bo *src_bo,
+ struct xe_bo *dst_bo,
+ struct ttm_resource *src,
+ struct ttm_resource *dst,
+ bool copy_only_ccs,
+ bool is_vram_resolve)
{
struct xe_gt *gt = m->tile->primary_gt;
struct xe_device *xe = gt_to_xe(gt);
@@ -809,8 +791,19 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
bool copy_ccs = xe_device_has_flat_ccs(xe) &&
xe_bo_needs_ccs_pages(src_bo) && xe_bo_needs_ccs_pages(dst_bo);
bool copy_system_ccs = copy_ccs && (!src_is_vram || !dst_is_vram);
- bool use_comp_pat = type_device && xe_device_has_flat_ccs(xe) &&
- GRAPHICS_VER(xe) >= 20 && src_is_vram && !dst_is_vram;
+ bool use_comp_pat;
+
+ /*
+ * For decompression operation, always use the compression PAT index.
+ * Otherwise, only use the compression PAT index for device memory
+ * when copying from VRAM to system memory.
+ */
+ if (is_vram_resolve) {
+ use_comp_pat = true;
+ } else {
+ use_comp_pat = type_device && xe_device_has_flat_ccs(xe) &&
+ GRAPHICS_VER(xe) >= 20 && src_is_vram && !dst_is_vram;
+ }
/* Copying CCS between two different BOs is not supported yet. */
if (XE_WARN_ON(copy_ccs && src_bo != dst_bo))
@@ -969,6 +962,49 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
return fence;
}
+/**
+ * xe_migrate_copy() - Copy content of TTM resources.
+ * @m: The migration context.
+ * @src_bo: The buffer object @src is currently bound to.
+ * @dst_bo: If copying between resources created for the same bo, set this to
+ * the same value as @src_bo. If copying between buffer objects, set it to
+ * the buffer object @dst is currently bound to.
+ * @src: The source TTM resource.
+ * @dst: The dst TTM resource.
+ * @copy_only_ccs: If true copy only CCS metadata
+ *
+ * Copies the contents of @src to @dst: On flat CCS devices,
+ * the CCS metadata is copied as well if needed, or if not present,
+ * the CCS metadata of @dst is cleared for security reasons.
+ *
+ * Return: Pointer to a dma_fence representing the last copy batch, or
+ * an error pointer on failure. If there is a failure, any copy operation
+ * started by the function call has been synced.
+ */
+struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
+ struct xe_bo *src_bo,
+ struct xe_bo *dst_bo,
+ struct ttm_resource *src,
+ struct ttm_resource *dst,
+ bool copy_only_ccs)
+{
+ return __xe_migrate_copy(m, src_bo, dst_bo, src, dst, copy_only_ccs, false);
+}
+
+/**
+ * xe_migrate_resolve() - Resolve and decompress a buffer object if required.
+ * This wrapper forwards to __xe_migrate_copy() with is_vram_resolve=true.
+ */
+struct dma_fence *xe_migrate_resolve(struct xe_migrate *m,
+ struct xe_bo *src_bo,
+ struct xe_bo *dst_bo,
+ struct ttm_resource *src,
+ struct ttm_resource *dst,
+ bool copy_only_ccs)
+{
+ return __xe_migrate_copy(m, src_bo, dst_bo, src, dst, copy_only_ccs, true);
+}
+
/**
* xe_migrate_lrc() - Get the LRC from migrate context.
* @migrate: Migrate context.
diff --git a/drivers/gpu/drm/xe/xe_migrate.h b/drivers/gpu/drm/xe/xe_migrate.h
index 4fad324b6253..ade68ab32a71 100644
--- a/drivers/gpu/drm/xe/xe_migrate.h
+++ b/drivers/gpu/drm/xe/xe_migrate.h
@@ -125,6 +125,13 @@ struct dma_fence *xe_migrate_copy(struct xe_migrate *m,
struct ttm_resource *dst,
bool copy_only_ccs);
+struct dma_fence *xe_migrate_resolve(struct xe_migrate *m,
+ struct xe_bo *src_bo,
+ struct xe_bo *dst_bo,
+ struct ttm_resource *src,
+ struct ttm_resource *dst,
+ bool copy_only_ccs);
+
int xe_migrate_ccs_rw_copy(struct xe_tile *tile, struct xe_exec_queue *q,
struct xe_bo *src_bo,
enum xe_sriov_vf_ccs_rw_ctxs read_write);
--
2.25.1
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH v1 3/5] drm/xe: add drm_exec helper to atomically lock VM (+ optional BO) with BOOKKEEP retry
2025-09-18 14:25 [PATCH v1 0/5] drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression Nitin Gote
2025-09-18 14:25 ` [PATCH v1 1/5] drm/xe: add VM_BIND DECOMPRESS uapi flag Nitin Gote
2025-09-18 14:25 ` [PATCH v1 2/5] drm/xe: add xe_migrate_resolve wrapper and is_vram_resolve support Nitin Gote
@ 2025-09-18 14:25 ` Nitin Gote
2025-09-18 14:25 ` [PATCH v1 4/5] drm/xe: implement VM_BIND decompression in vm_bind_ioctl Nitin Gote
` (4 subsequent siblings)
7 siblings, 0 replies; 14+ messages in thread
From: Nitin Gote @ 2025-09-18 14:25 UTC (permalink / raw)
To: intel-xe; +Cc: himal.prasad.ghimiray, matthew.brost, thomas.hellstrom,
Nitin Gote
Add xe_vm_exec_lock_vm_and_bo(), a drm_exec-based helper that atomically
acquires the VM reservation and an optional BO reservation. When requested
(wait_bookkeep) the helper verifies DMA_RESV_USAGE_BOOKKEEP is clear and,
if not, releases the exec locks, waits outside the reservation and retries.
This centralises the drm_exec + BOOKKEEP wait/retry logic used by vm_bind
paths, avoiding duplicated code and preventing sleeps while holding resv
locks. Caller must call drm_exec_fini(exec) on success and must not
re-acquire vm->lock while exec holds the reservations.
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
---
drivers/gpu/drm/xe/xe_vm.c | 53 ++++++++++++++++++++++++++++++++++++++
1 file changed, 53 insertions(+)
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index 0cacab20ff85..fae88c4c981e 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -3496,6 +3496,59 @@ static int xe_vm_bind_ioctl_validate_bo(struct xe_device *xe, struct xe_bo *bo,
return 0;
}
+/* Helper:
+ * Atomically acquire VM (+ optional BO) reservations using drm_exec.
+ * If wait_bookkeep is true, ensure VM BOOKKEEP is clear before returning:
+ * - if BOOKKEEP fences exist, release exec locks, wait outside and retry.
+ *
+ * On success returns 0 with *exec initialized and the reservations held.
+ * Caller MUST call drm_exec_fini(exec) to release reservations.
+ */
+static int xe_vm_exec_lock_vm_and_bo(struct xe_vm *vm, struct xe_bo *bo,
+ struct drm_exec *exec, bool intr,
+ bool wait_bookkeep)
+{
+ int err = 0, wait;
+
+retry:
+ drm_exec_init(exec, intr ? DRM_EXEC_INTERRUPTIBLE_WAIT : 0, 0);
+ drm_exec_until_all_locked(exec) {
+ err = drm_exec_lock_obj(exec, xe_vm_obj(vm));
+ drm_exec_retry_on_contention(exec);
+ if (err)
+ break;
+
+ if (bo) {
+ err = drm_exec_lock_obj(exec, &bo->ttm.base);
+ drm_exec_retry_on_contention(exec);
+ if (err)
+ break;
+ }
+ }
+
+ if (err) {
+ /* Ensure exec cleaned up if partially initialized */
+ drm_exec_fini(exec);
+ return err;
+ }
+
+ if (wait_bookkeep) {
+ /* If BOOKKEEP fences exist, release exec and wait outside, then retry */
+ if (!dma_resv_test_signaled(xe_vm_resv(vm), DMA_RESV_USAGE_BOOKKEEP)) {
+ drm_exec_fini(exec);
+ wait = dma_resv_wait_timeout(xe_vm_resv(vm),
+ DMA_RESV_USAGE_BOOKKEEP,
+ true, MAX_SCHEDULE_TIMEOUT);
+ if (wait <= 0)
+ return -ETIME;
+ goto retry;
+ }
+ }
+
+ /* Success: exec holds the reservations */
+ return 0;
+}
+
int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
{
struct xe_device *xe = to_xe_device(dev);
--
2.25.1
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH v1 4/5] drm/xe: implement VM_BIND decompression in vm_bind_ioctl
2025-09-18 14:25 [PATCH v1 0/5] drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression Nitin Gote
` (2 preceding siblings ...)
2025-09-18 14:25 ` [PATCH v1 3/5] drm/xe: add drm_exec helper to atomically lock VM (+ optional BO) with BOOKKEEP retry Nitin Gote
@ 2025-09-18 14:25 ` Nitin Gote
2025-09-18 16:13 ` Matthew Auld
2025-09-23 17:26 ` Matthew Brost
2025-09-18 14:25 ` [PATCH v1 5/5] drm/xe: perform on-demand decompression on VMA pagefaults Nitin Gote
` (3 subsequent siblings)
7 siblings, 2 replies; 14+ messages in thread
From: Nitin Gote @ 2025-09-18 14:25 UTC (permalink / raw)
To: intel-xe; +Cc: himal.prasad.ghimiray, matthew.brost, thomas.hellstrom,
Nitin Gote
Implement handling of VM_BIND(..., DECOMPRESS) in xe_vm_bind_ioctl.
- Validate decompression preconditions (VRAM buffer, flat CCS support,
XE2+ hardware, and an uncompressed PAT index).
- Mark the BO with DRM_XE_VM_BIND_FLAG_DECOMPRESS.
- Invalidate any overlapping VMA when required.
- Use helper xe_vm_exec_lock_vm_and_bo(..., wait_bookkeep=true)
to atomically acquire the VM reservation and optional BO reservation and
handle BOOKKEEP wait/retry.
* Drop vm->lock before calling the helper.
* Do VMA invalidation or schedule the migrate resolve while reservations
are held.
* Call drm_exec_fini() to release reservations, then wait on the
migrate/decompression fence without holding reservations and
re-acquire vm->lock to continue.
- Defer decompression when the VM is in fault mode.
This schedules an in-place GPU resolve (xe_migrate_resolve) for
decompression and waits for completion before proceeding with the bind.
The change centralises drm_exec + BOOKKEEP retry logic and avoids
sleeping while holding resv locks.
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
---
drivers/gpu/drm/xe/xe_vm.c | 134 +++++++++++++++++++++++++++++++++++--
1 file changed, 130 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
index fae88c4c981e..cdeb7995eab1 100644
--- a/drivers/gpu/drm/xe/xe_vm.c
+++ b/drivers/gpu/drm/xe/xe_vm.c
@@ -3266,7 +3266,8 @@ ALLOW_ERROR_INJECTION(vm_bind_ioctl_ops_execute, ERRNO);
DRM_XE_VM_BIND_FLAG_NULL | \
DRM_XE_VM_BIND_FLAG_DUMPABLE | \
DRM_XE_VM_BIND_FLAG_CHECK_PXP | \
- DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR)
+ DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR | \
+ DRM_XE_VM_BIND_FLAG_DECOMPRESS)
#ifdef TEST_VM_OPS_ERROR
#define SUPPORTED_FLAGS (SUPPORTED_FLAGS_STUB | FORCE_OP_ERROR)
@@ -3324,6 +3325,7 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm,
bool is_null = flags & DRM_XE_VM_BIND_FLAG_NULL;
bool is_cpu_addr_mirror = flags &
DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR;
+ bool is_decompress = flags & DRM_XE_VM_BIND_FLAG_DECOMPRESS;
u16 pat_index = (*bind_ops)[i].pat_index;
u16 coh_mode;
@@ -3361,7 +3363,7 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm,
(is_null || is_cpu_addr_mirror)) ||
XE_IOCTL_DBG(xe, !obj &&
op == DRM_XE_VM_BIND_OP_MAP &&
- !is_null && !is_cpu_addr_mirror) ||
+ is_decompress && !is_null && !is_cpu_addr_mirror) ||
XE_IOCTL_DBG(xe, !obj &&
op == DRM_XE_VM_BIND_OP_UNMAP_ALL) ||
XE_IOCTL_DBG(xe, addr &&
@@ -3378,8 +3380,8 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm,
op == DRM_XE_VM_BIND_OP_PREFETCH) ||
XE_IOCTL_DBG(xe, prefetch_region &&
op != DRM_XE_VM_BIND_OP_PREFETCH) ||
- XE_IOCTL_DBG(xe, (prefetch_region != DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC &&
- !(BIT(prefetch_region) & xe->info.mem_region_mask))) ||
+ XE_IOCTL_DBG(xe, (prefetch_region != DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC &&
+ !(BIT(prefetch_region) & xe->info.mem_region_mask))) ||
XE_IOCTL_DBG(xe, obj &&
op == DRM_XE_VM_BIND_OP_UNMAP)) {
err = -EINVAL;
@@ -3698,6 +3700,130 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
u64 obj_offset = bind_ops[i].obj_offset;
u32 prefetch_region = bind_ops[i].prefetch_mem_region_instance;
u16 pat_index = bind_ops[i].pat_index;
+ bool do_decompress = flags & DRM_XE_VM_BIND_FLAG_DECOMPRESS;
+
+ /* Handle decompression process */
+ if (do_decompress && bos[i]) {
+ struct xe_vma *existing_vma;
+ struct dma_fence *decomp_fence;
+ struct xe_tile *tile = xe_device_get_root_tile(xe);
+
+ bos[i]->flags |= DRM_XE_VM_BIND_FLAG_DECOMPRESS;
+
+ /* Validate VRAM buffer suitable for decompression or not */
+ if (!mem_type_is_vram(bos[i]->ttm.resource->mem_type)) {
+ drm_err(&vm->xe->drm,
+ "Decompression requires VRAM buffer\n");
+ err = -EINVAL;
+ goto unwind_ops;
+ }
+
+ /* Check hardware support for decompression */
+ if (!xe_device_has_flat_ccs(vm->xe)) {
+ drm_err(&vm->xe->drm,
+ "Decompression requires flat CCS support\n");
+ err = -EOPNOTSUPP;
+ goto unwind_ops;
+ }
+
+ if (GRAPHICS_VER(vm->xe) < 20) {
+ drm_err(&vm->xe->drm,
+ "Decompression requires XE2+ hardware\n");
+ err = -EOPNOTSUPP;
+ goto unwind_ops;
+ }
+
+ /* Validate that user provided an uncompressed PAT index */
+ if (pat_index == vm->xe->pat.idx[XE_CACHE_NONE_COMPRESSION]) {
+ drm_err(&vm->xe->drm,
+ "Decompression requires uncompressed PAT index\n");
+ err = -EINVAL;
+ goto unwind_ops;
+ }
+
+ /* Invalidate VMA so GPU mappings are destroyed */
+ existing_vma = xe_vm_find_overlapping_vma(vm, addr, range);
+ if (existing_vma) {
+ struct xe_bo *existing_bo = xe_vma_bo(existing_vma);
+
+ drm_dbg(&vm->xe->drm,
+ "Found overlapping VMA - automatic invalidation will occur\n");
+
+ /* Drop vm->lock and atomically take VM+BO resvs via drm_exec */
+ up_write(&vm->lock);
+ struct drm_exec exec;
+
+ err = xe_vm_exec_lock_vm_and_bo(vm, existing_bo, &exec, true, true);
+ if (err) {
+ down_write(&vm->lock);
+ goto put_obj;
+ }
+
+ /* Invalidate the VMA while reservations are held. */
+ err = xe_vm_invalidate_vma(existing_vma);
+
+ /* release reservations (drm_exec_fini releases VM+BO resvs) */
+ drm_exec_fini(&exec);
+
+ /* Re-acquire vm->lock */
+ down_write(&vm->lock);
+
+ if (err)
+ goto put_obj;
+ }
+
+ /* Handle decompression differently for fault mode */
+ if (xe_vm_in_fault_mode(vm)) {
+ drm_dbg(&vm->xe->drm,
+ "Deferring decompression for fault mode\n");
+ /* Skip immediate decompression */
+ goto unwind_ops;
+ }
+
+ /* Drop vm->lock and atomically acquire VM+BO reservations via drm_exec.
+ * Hold the reservations while we wait for BOOKKEEP and schedule the
+ * in-place resolve. Release the reservations (drm_exec_fini) before
+ * waiting on the migrate fence so we don't hold resv locks while
+ * sleeping.
+ */
+ up_write(&vm->lock);
+ struct drm_exec exec;
+
+ err = xe_vm_exec_lock_vm_and_bo(vm, bos[i], &exec, true, true);
+ if (err) {
+ drm_err(&vm->xe->drm,
+ "Failed to acquire reservations for decompression: %d\n",
+ err);
+ down_write(&vm->lock);
+ goto unwind_ops;
+ }
+
+ /* With exec holding reservations, schedule the in-place decompression */
+ decomp_fence = xe_migrate_resolve(tile->migrate,
+ bos[i], bos[i],
+ bos[i]->ttm.resource,
+ bos[i]->ttm.resource,
+ false);
+
+ /* release the reservations so scheduler can observe BO fences */
+ drm_exec_fini(&exec);
+
+ if (IS_ERR(decomp_fence)) {
+ err = PTR_ERR(decomp_fence);
+ drm_err(&vm->xe->drm,
+ "In-place decompression failed: %d\n", err);
+ /* vm->lock must be held for unwind_ops cleanup */
+ down_write(&vm->lock);
+ goto unwind_ops;
+ }
+
+ /* Wait for in-place decompression to complete without holding vm->lock */
+ dma_fence_wait(decomp_fence, false);
+ dma_fence_put(decomp_fence);
+
+ /* Now re-acquire vm->lock and continue */
+ down_write(&vm->lock);
+ }
ops[i] = vm_bind_ioctl_ops_create(vm, &vops, bos[i], obj_offset,
addr, range, op, flags,
--
2.25.1
^ permalink raw reply related [flat|nested] 14+ messages in thread
* [PATCH v1 5/5] drm/xe: perform on-demand decompression on VMA pagefaults
2025-09-18 14:25 [PATCH v1 0/5] drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression Nitin Gote
` (3 preceding siblings ...)
2025-09-18 14:25 ` [PATCH v1 4/5] drm/xe: implement VM_BIND decompression in vm_bind_ioctl Nitin Gote
@ 2025-09-18 14:25 ` Nitin Gote
2025-09-23 17:29 ` Matthew Brost
2025-09-18 15:12 ` ✓ CI.KUnit: success for drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression Patchwork
` (2 subsequent siblings)
7 siblings, 1 reply; 14+ messages in thread
From: Nitin Gote @ 2025-09-18 14:25 UTC (permalink / raw)
To: intel-xe; +Cc: himal.prasad.ghimiray, matthew.brost, thomas.hellstrom,
Nitin Gote
When a VMA pagefault occurs and the bound BO has the DECOMPRESS flag,
schedule a resolve (xe_migrate_resolve) to perform per-page decompression
before binding the VMA to the faulting GT.
Behavior:
- schedule xe_migrate_resolve() while holding appropriate reservations,
- wait on the returned fence, then clear the BO's DECOMPRESS flag
on success,
- proceed with the normal rebind path.
This ensures on-demand fault-time decompression works consistently with
the VM_BIND path.
Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
---
drivers/gpu/drm/xe/xe_gt_pagefault.c | 41 ++++++++++++++++++++++++++--
1 file changed, 38 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_gt_pagefault.c b/drivers/gpu/drm/xe/xe_gt_pagefault.c
index a054d6010ae0..4a3f2682ed85 100644
--- a/drivers/gpu/drm/xe/xe_gt_pagefault.c
+++ b/drivers/gpu/drm/xe/xe_gt_pagefault.c
@@ -95,11 +95,14 @@ static int handle_vma_pagefault(struct xe_gt *gt, struct xe_vma *vma,
bool atomic)
{
struct xe_vm *vm = xe_vma_vm(vma);
+ struct xe_bo *bo = xe_vma_bo(vma);
struct xe_tile *tile = gt_to_tile(gt);
struct xe_validation_ctx ctx;
struct drm_exec exec;
struct dma_fence *fence;
int err, needs_vram;
+ bool needs_decompression = false;
+ struct dma_fence *decomp_fence = NULL;
lockdep_assert_held_write(&vm->lock);
@@ -112,8 +115,14 @@ static int handle_vma_pagefault(struct xe_gt *gt, struct xe_vma *vma,
trace_xe_vma_pagefault(vma);
+ /* Check if decompression is needed */
+ if (bo && (bo->flags & DRM_XE_VM_BIND_FLAG_DECOMPRESS)) {
+ needs_decompression = true;
+ drm_dbg(&vm->xe->drm, "Decompressing VMA during page fault handling\n");
+ }
+
/* Check if VMA is valid, opportunistic check only */
- if (vma_is_valid(tile, vma) && !atomic)
+ if (vma_is_valid(tile, vma) && !atomic && !needs_decompression)
return 0;
retry_userptr:
@@ -135,6 +144,21 @@ static int handle_vma_pagefault(struct xe_gt *gt, struct xe_vma *vma,
if (err)
goto unlock_dma_resv;
+ /* Perform decompression inside proper locking context */
+ if (needs_decompression) {
+ decomp_fence = xe_migrate_resolve(tile->migrate, xe_vma_bo(vma),
+ xe_vma_bo(vma),
+ xe_vma_bo(vma)->ttm.resource,
+ xe_vma_bo(vma)->ttm.resource,
+ false);
+ if (IS_ERR(decomp_fence)) {
+ drm_err(&vm->xe->drm,
+ "Decompression failed during page fault handling\n");
+ err = PTR_ERR(decomp_fence);
+ goto unlock_dma_resv;
+ }
+ }
+
/* Bind VMA only to the GT that has faulted */
trace_xe_vma_pf_bind(vma);
xe_vm_set_validation_exec(vm, &exec);
@@ -147,8 +171,19 @@ static int handle_vma_pagefault(struct xe_gt *gt, struct xe_vma *vma,
}
}
- dma_fence_wait(fence, false);
- dma_fence_put(fence);
+ /* Wait for decompression first, then rebind */
+ if (decomp_fence) {
+ dma_fence_wait(decomp_fence, false);
+ dma_fence_put(decomp_fence);
+
+ /* Clear the decompression flag after successful decompression */
+ bo->flags &= ~DRM_XE_VM_BIND_FLAG_DECOMPRESS;
+ }
+
+ if (fence) {
+ dma_fence_wait(fence, false);
+ dma_fence_put(fence);
+ }
unlock_dma_resv:
xe_validation_ctx_fini(&ctx);
--
2.25.1
^ permalink raw reply related [flat|nested] 14+ messages in thread
* ✓ CI.KUnit: success for drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression
2025-09-18 14:25 [PATCH v1 0/5] drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression Nitin Gote
` (4 preceding siblings ...)
2025-09-18 14:25 ` [PATCH v1 5/5] drm/xe: perform on-demand decompression on VMA pagefaults Nitin Gote
@ 2025-09-18 15:12 ` Patchwork
2025-09-18 15:57 ` ✓ Xe.CI.BAT: " Patchwork
2025-09-19 0:37 ` ✓ Xe.CI.Full: " Patchwork
7 siblings, 0 replies; 14+ messages in thread
From: Patchwork @ 2025-09-18 15:12 UTC (permalink / raw)
To: Nitin Gote; +Cc: intel-xe
== Series Details ==
Series: drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression
URL : https://patchwork.freedesktop.org/series/154714/
State : success
== Summary ==
+ trap cleanup EXIT
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/xe/.kunitconfig
[15:11:31] Configuring KUnit Kernel ...
Generating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[15:11:36] 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
[15:12:05] Starting KUnit Kernel (1/1)...
[15:12:05] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[15:12:05] ================== guc_buf (11 subtests) ===================
[15:12:05] [PASSED] test_smallest
[15:12:05] [PASSED] test_largest
[15:12:05] [PASSED] test_granular
[15:12:05] [PASSED] test_unique
[15:12:05] [PASSED] test_overlap
[15:12:05] [PASSED] test_reusable
[15:12:05] [PASSED] test_too_big
[15:12:05] [PASSED] test_flush
[15:12:05] [PASSED] test_lookup
[15:12:05] [PASSED] test_data
[15:12:05] [PASSED] test_class
[15:12:05] ===================== [PASSED] guc_buf =====================
[15:12:05] =================== guc_dbm (7 subtests) ===================
[15:12:05] [PASSED] test_empty
[15:12:05] [PASSED] test_default
[15:12:05] ======================== test_size ========================
[15:12:05] [PASSED] 4
[15:12:05] [PASSED] 8
[15:12:05] [PASSED] 32
[15:12:05] [PASSED] 256
[15:12:05] ==================== [PASSED] test_size ====================
[15:12:05] ======================= test_reuse ========================
[15:12:05] [PASSED] 4
[15:12:05] [PASSED] 8
[15:12:05] [PASSED] 32
[15:12:05] [PASSED] 256
[15:12:05] =================== [PASSED] test_reuse ====================
[15:12:05] =================== test_range_overlap ====================
[15:12:05] [PASSED] 4
[15:12:05] [PASSED] 8
[15:12:05] [PASSED] 32
[15:12:05] [PASSED] 256
[15:12:05] =============== [PASSED] test_range_overlap ================
[15:12:05] =================== test_range_compact ====================
[15:12:05] [PASSED] 4
[15:12:05] [PASSED] 8
[15:12:05] [PASSED] 32
[15:12:05] [PASSED] 256
[15:12:05] =============== [PASSED] test_range_compact ================
[15:12:05] ==================== test_range_spare =====================
[15:12:05] [PASSED] 4
[15:12:05] [PASSED] 8
[15:12:05] [PASSED] 32
[15:12:05] [PASSED] 256
[15:12:05] ================ [PASSED] test_range_spare =================
[15:12:05] ===================== [PASSED] guc_dbm =====================
[15:12:05] =================== guc_idm (6 subtests) ===================
[15:12:05] [PASSED] bad_init
[15:12:05] [PASSED] no_init
[15:12:05] [PASSED] init_fini
[15:12:05] [PASSED] check_used
[15:12:05] [PASSED] check_quota
[15:12:05] [PASSED] check_all
[15:12:05] ===================== [PASSED] guc_idm =====================
[15:12:05] ================== no_relay (3 subtests) ===================
[15:12:05] [PASSED] xe_drops_guc2pf_if_not_ready
[15:12:05] [PASSED] xe_drops_guc2vf_if_not_ready
[15:12:05] [PASSED] xe_rejects_send_if_not_ready
[15:12:05] ==================== [PASSED] no_relay =====================
[15:12:05] ================== pf_relay (14 subtests) ==================
[15:12:05] [PASSED] pf_rejects_guc2pf_too_short
[15:12:05] [PASSED] pf_rejects_guc2pf_too_long
[15:12:05] [PASSED] pf_rejects_guc2pf_no_payload
[15:12:05] [PASSED] pf_fails_no_payload
[15:12:05] [PASSED] pf_fails_bad_origin
[15:12:05] [PASSED] pf_fails_bad_type
[15:12:05] [PASSED] pf_txn_reports_error
[15:12:05] [PASSED] pf_txn_sends_pf2guc
[15:12:05] [PASSED] pf_sends_pf2guc
[15:12:05] [SKIPPED] pf_loopback_nop
[15:12:05] [SKIPPED] pf_loopback_echo
[15:12:05] [SKIPPED] pf_loopback_fail
[15:12:05] [SKIPPED] pf_loopback_busy
[15:12:05] [SKIPPED] pf_loopback_retry
[15:12:05] ==================== [PASSED] pf_relay =====================
[15:12:05] ================== vf_relay (3 subtests) ===================
[15:12:05] [PASSED] vf_rejects_guc2vf_too_short
[15:12:05] [PASSED] vf_rejects_guc2vf_too_long
[15:12:05] [PASSED] vf_rejects_guc2vf_no_payload
[15:12:05] ==================== [PASSED] vf_relay =====================
[15:12:05] ===================== lmtt (1 subtest) =====================
[15:12:05] ======================== test_ops =========================
[15:12:05] [PASSED] 2-level
[15:12:05] [PASSED] multi-level
[15:12:05] ==================== [PASSED] test_ops =====================
[15:12:05] ====================== [PASSED] lmtt =======================
[15:12:05] ================= pf_service (11 subtests) =================
[15:12:05] [PASSED] pf_negotiate_any
[15:12:05] [PASSED] pf_negotiate_base_match
[15:12:05] [PASSED] pf_negotiate_base_newer
[15:12:05] [PASSED] pf_negotiate_base_next
[15:12:05] [SKIPPED] pf_negotiate_base_older
[15:12:05] [PASSED] pf_negotiate_base_prev
[15:12:05] [PASSED] pf_negotiate_latest_match
[15:12:05] [PASSED] pf_negotiate_latest_newer
[15:12:05] [PASSED] pf_negotiate_latest_next
[15:12:05] [SKIPPED] pf_negotiate_latest_older
[15:12:05] [SKIPPED] pf_negotiate_latest_prev
[15:12:05] =================== [PASSED] pf_service ====================
[15:12:05] ================= xe_guc_g2g (2 subtests) ==================
[15:12:05] ============== xe_live_guc_g2g_kunit_default ==============
[15:12:05] ========= [SKIPPED] xe_live_guc_g2g_kunit_default ==========
[15:12:05] ============== xe_live_guc_g2g_kunit_allmem ===============
[15:12:05] ========== [SKIPPED] xe_live_guc_g2g_kunit_allmem ==========
[15:12:05] =================== [SKIPPED] xe_guc_g2g ===================
[15:12:05] =================== xe_mocs (2 subtests) ===================
[15:12:05] ================ xe_live_mocs_kernel_kunit ================
[15:12:05] =========== [SKIPPED] xe_live_mocs_kernel_kunit ============
[15:12:05] ================ xe_live_mocs_reset_kunit =================
[15:12:05] ============ [SKIPPED] xe_live_mocs_reset_kunit ============
[15:12:05] ==================== [SKIPPED] xe_mocs =====================
[15:12:05] ================= xe_migrate (2 subtests) ==================
[15:12:05] ================= xe_migrate_sanity_kunit =================
[15:12:05] ============ [SKIPPED] xe_migrate_sanity_kunit =============
[15:12:05] ================== xe_validate_ccs_kunit ==================
[15:12:05] ============= [SKIPPED] xe_validate_ccs_kunit ==============
[15:12:05] =================== [SKIPPED] xe_migrate ===================
[15:12:05] ================== xe_dma_buf (1 subtest) ==================
[15:12:05] ==================== xe_dma_buf_kunit =====================
[15:12:05] ================ [SKIPPED] xe_dma_buf_kunit ================
[15:12:05] =================== [SKIPPED] xe_dma_buf ===================
[15:12:05] ================= xe_bo_shrink (1 subtest) =================
[15:12:05] =================== xe_bo_shrink_kunit ====================
[15:12:05] =============== [SKIPPED] xe_bo_shrink_kunit ===============
[15:12:05] ================== [SKIPPED] xe_bo_shrink ==================
[15:12:05] ==================== xe_bo (2 subtests) ====================
[15:12:05] ================== xe_ccs_migrate_kunit ===================
[15:12:05] ============== [SKIPPED] xe_ccs_migrate_kunit ==============
[15:12:05] ==================== xe_bo_evict_kunit ====================
[15:12:05] =============== [SKIPPED] xe_bo_evict_kunit ================
[15:12:05] ===================== [SKIPPED] xe_bo ======================
[15:12:05] ==================== args (11 subtests) ====================
[15:12:05] [PASSED] count_args_test
[15:12:05] [PASSED] call_args_example
[15:12:05] [PASSED] call_args_test
[15:12:05] [PASSED] drop_first_arg_example
[15:12:05] [PASSED] drop_first_arg_test
[15:12:05] [PASSED] first_arg_example
[15:12:05] [PASSED] first_arg_test
[15:12:05] [PASSED] last_arg_example
[15:12:05] [PASSED] last_arg_test
[15:12:05] [PASSED] pick_arg_example
[15:12:05] [PASSED] sep_comma_example
[15:12:05] ====================== [PASSED] args =======================
[15:12:05] =================== xe_pci (3 subtests) ====================
[15:12:05] ==================== check_graphics_ip ====================
[15:12:05] [PASSED] 12.00 Xe_LP
[15:12:05] [PASSED] 12.10 Xe_LP+
[15:12:05] [PASSED] 12.55 Xe_HPG
[15:12:05] [PASSED] 12.60 Xe_HPC
[15:12:05] [PASSED] 12.70 Xe_LPG
[15:12:05] [PASSED] 12.71 Xe_LPG
[15:12:05] [PASSED] 12.74 Xe_LPG+
[15:12:05] [PASSED] 20.01 Xe2_HPG
[15:12:05] [PASSED] 20.02 Xe2_HPG
[15:12:05] [PASSED] 20.04 Xe2_LPG
[15:12:05] [PASSED] 30.00 Xe3_LPG
[15:12:05] [PASSED] 30.01 Xe3_LPG
[15:12:05] [PASSED] 30.03 Xe3_LPG
[15:12:05] ================ [PASSED] check_graphics_ip ================
[15:12:05] ===================== check_media_ip ======================
[15:12:05] [PASSED] 12.00 Xe_M
[15:12:05] [PASSED] 12.55 Xe_HPM
[15:12:05] [PASSED] 13.00 Xe_LPM+
[15:12:05] [PASSED] 13.01 Xe2_HPM
[15:12:05] [PASSED] 20.00 Xe2_LPM
[15:12:05] [PASSED] 30.00 Xe3_LPM
[15:12:05] [PASSED] 30.02 Xe3_LPM
[15:12:05] ================= [PASSED] check_media_ip ==================
[15:12:05] ================= check_platform_gt_count =================
[15:12:05] [PASSED] 0x9A60 (TIGERLAKE)
[15:12:05] [PASSED] 0x9A68 (TIGERLAKE)
[15:12:05] [PASSED] 0x9A70 (TIGERLAKE)
[15:12:05] [PASSED] 0x9A40 (TIGERLAKE)
[15:12:05] [PASSED] 0x9A49 (TIGERLAKE)
[15:12:05] [PASSED] 0x9A59 (TIGERLAKE)
[15:12:05] [PASSED] 0x9A78 (TIGERLAKE)
[15:12:05] [PASSED] 0x9AC0 (TIGERLAKE)
[15:12:05] [PASSED] 0x9AC9 (TIGERLAKE)
[15:12:05] [PASSED] 0x9AD9 (TIGERLAKE)
[15:12:05] [PASSED] 0x9AF8 (TIGERLAKE)
[15:12:05] [PASSED] 0x4C80 (ROCKETLAKE)
[15:12:05] [PASSED] 0x4C8A (ROCKETLAKE)
[15:12:05] [PASSED] 0x4C8B (ROCKETLAKE)
[15:12:05] [PASSED] 0x4C8C (ROCKETLAKE)
[15:12:05] [PASSED] 0x4C90 (ROCKETLAKE)
[15:12:05] [PASSED] 0x4C9A (ROCKETLAKE)
[15:12:05] [PASSED] 0x4680 (ALDERLAKE_S)
[15:12:05] [PASSED] 0x4682 (ALDERLAKE_S)
[15:12:05] [PASSED] 0x4688 (ALDERLAKE_S)
[15:12:05] [PASSED] 0x468A (ALDERLAKE_S)
[15:12:05] [PASSED] 0x468B (ALDERLAKE_S)
[15:12:05] [PASSED] 0x4690 (ALDERLAKE_S)
[15:12:05] [PASSED] 0x4692 (ALDERLAKE_S)
[15:12:05] [PASSED] 0x4693 (ALDERLAKE_S)
[15:12:05] [PASSED] 0x46A0 (ALDERLAKE_P)
[15:12:05] [PASSED] 0x46A1 (ALDERLAKE_P)
[15:12:05] [PASSED] 0x46A2 (ALDERLAKE_P)
[15:12:05] [PASSED] 0x46A3 (ALDERLAKE_P)
[15:12:05] [PASSED] 0x46A6 (ALDERLAKE_P)
[15:12:05] [PASSED] 0x46A8 (ALDERLAKE_P)
[15:12:05] [PASSED] 0x46AA (ALDERLAKE_P)
[15:12:05] [PASSED] 0x462A (ALDERLAKE_P)
[15:12:05] [PASSED] 0x4626 (ALDERLAKE_P)
[15:12:05] [PASSED] 0x4628 (ALDERLAKE_P)
[15:12:05] [PASSED] 0x46B0 (ALDERLAKE_P)
[15:12:05] [PASSED] 0x46B1 (ALDERLAKE_P)
[15:12:05] [PASSED] 0x46B2 (ALDERLAKE_P)
[15:12:05] [PASSED] 0x46B3 (ALDERLAKE_P)
[15:12:05] [PASSED] 0x46C0 (ALDERLAKE_P)
[15:12:05] [PASSED] 0x46C1 (ALDERLAKE_P)
[15:12:05] [PASSED] 0x46C2 (ALDERLAKE_P)
[15:12:05] [PASSED] 0x46C3 (ALDERLAKE_P)
[15:12:05] [PASSED] 0x46D0 (ALDERLAKE_N)
[15:12:05] [PASSED] 0x46D1 (ALDERLAKE_N)
[15:12:05] [PASSED] 0x46D2 (ALDERLAKE_N)
[15:12:05] [PASSED] 0x46D3 (ALDERLAKE_N)
[15:12:05] [PASSED] 0x46D4 (ALDERLAKE_N)
[15:12:05] [PASSED] 0xA721 (ALDERLAKE_P)
[15:12:05] [PASSED] 0xA7A1 (ALDERLAKE_P)
[15:12:05] [PASSED] 0xA7A9 (ALDERLAKE_P)
[15:12:05] [PASSED] 0xA7AC (ALDERLAKE_P)
[15:12:05] [PASSED] 0xA7AD (ALDERLAKE_P)
[15:12:05] [PASSED] 0xA720 (ALDERLAKE_P)
[15:12:05] [PASSED] 0xA7A0 (ALDERLAKE_P)
[15:12:05] [PASSED] 0xA7A8 (ALDERLAKE_P)
[15:12:05] [PASSED] 0xA7AA (ALDERLAKE_P)
[15:12:05] [PASSED] 0xA7AB (ALDERLAKE_P)
[15:12:05] [PASSED] 0xA780 (ALDERLAKE_S)
[15:12:05] [PASSED] 0xA781 (ALDERLAKE_S)
[15:12:05] [PASSED] 0xA782 (ALDERLAKE_S)
[15:12:05] [PASSED] 0xA783 (ALDERLAKE_S)
[15:12:05] [PASSED] 0xA788 (ALDERLAKE_S)
[15:12:05] [PASSED] 0xA789 (ALDERLAKE_S)
[15:12:05] [PASSED] 0xA78A (ALDERLAKE_S)
[15:12:05] [PASSED] 0xA78B (ALDERLAKE_S)
[15:12:05] [PASSED] 0x4905 (DG1)
[15:12:05] [PASSED] 0x4906 (DG1)
[15:12:05] [PASSED] 0x4907 (DG1)
[15:12:05] [PASSED] 0x4908 (DG1)
[15:12:05] [PASSED] 0x4909 (DG1)
[15:12:05] [PASSED] 0x56C0 (DG2)
[15:12:05] [PASSED] 0x56C2 (DG2)
[15:12:05] [PASSED] 0x56C1 (DG2)
[15:12:05] [PASSED] 0x7D51 (METEORLAKE)
[15:12:05] [PASSED] 0x7DD1 (METEORLAKE)
[15:12:05] [PASSED] 0x7D41 (METEORLAKE)
[15:12:05] [PASSED] 0x7D67 (METEORLAKE)
[15:12:05] [PASSED] 0xB640 (METEORLAKE)
[15:12:05] [PASSED] 0x56A0 (DG2)
[15:12:05] [PASSED] 0x56A1 (DG2)
[15:12:05] [PASSED] 0x56A2 (DG2)
[15:12:05] [PASSED] 0x56BE (DG2)
[15:12:05] [PASSED] 0x56BF (DG2)
[15:12:05] [PASSED] 0x5690 (DG2)
[15:12:05] [PASSED] 0x5691 (DG2)
[15:12:05] [PASSED] 0x5692 (DG2)
[15:12:05] [PASSED] 0x56A5 (DG2)
[15:12:05] [PASSED] 0x56A6 (DG2)
[15:12:05] [PASSED] 0x56B0 (DG2)
[15:12:05] [PASSED] 0x56B1 (DG2)
[15:12:05] [PASSED] 0x56BA (DG2)
[15:12:05] [PASSED] 0x56BB (DG2)
[15:12:05] [PASSED] 0x56BC (DG2)
[15:12:05] [PASSED] 0x56BD (DG2)
[15:12:05] [PASSED] 0x5693 (DG2)
[15:12:05] [PASSED] 0x5694 (DG2)
[15:12:05] [PASSED] 0x5695 (DG2)
[15:12:05] [PASSED] 0x56A3 (DG2)
[15:12:05] [PASSED] 0x56A4 (DG2)
[15:12:05] [PASSED] 0x56B2 (DG2)
[15:12:05] [PASSED] 0x56B3 (DG2)
[15:12:05] [PASSED] 0x5696 (DG2)
[15:12:05] [PASSED] 0x5697 (DG2)
[15:12:05] [PASSED] 0xB69 (PVC)
[15:12:05] [PASSED] 0xB6E (PVC)
[15:12:05] [PASSED] 0xBD4 (PVC)
[15:12:05] [PASSED] 0xBD5 (PVC)
[15:12:05] [PASSED] 0xBD6 (PVC)
[15:12:05] [PASSED] 0xBD7 (PVC)
[15:12:05] [PASSED] 0xBD8 (PVC)
[15:12:05] [PASSED] 0xBD9 (PVC)
[15:12:05] [PASSED] 0xBDA (PVC)
[15:12:05] [PASSED] 0xBDB (PVC)
[15:12:05] [PASSED] 0xBE0 (PVC)
[15:12:05] [PASSED] 0xBE1 (PVC)
[15:12:05] [PASSED] 0xBE5 (PVC)
[15:12:05] [PASSED] 0x7D40 (METEORLAKE)
[15:12:05] [PASSED] 0x7D45 (METEORLAKE)
[15:12:05] [PASSED] 0x7D55 (METEORLAKE)
[15:12:05] [PASSED] 0x7D60 (METEORLAKE)
[15:12:05] [PASSED] 0x7DD5 (METEORLAKE)
[15:12:05] [PASSED] 0x6420 (LUNARLAKE)
[15:12:05] [PASSED] 0x64A0 (LUNARLAKE)
[15:12:05] [PASSED] 0x64B0 (LUNARLAKE)
[15:12:05] [PASSED] 0xE202 (BATTLEMAGE)
[15:12:05] [PASSED] 0xE209 (BATTLEMAGE)
[15:12:05] [PASSED] 0xE20B (BATTLEMAGE)
[15:12:05] [PASSED] 0xE20C (BATTLEMAGE)
[15:12:05] [PASSED] 0xE20D (BATTLEMAGE)
[15:12:05] [PASSED] 0xE210 (BATTLEMAGE)
[15:12:05] [PASSED] 0xE211 (BATTLEMAGE)
[15:12:05] [PASSED] 0xE212 (BATTLEMAGE)
[15:12:05] [PASSED] 0xE216 (BATTLEMAGE)
[15:12:05] [PASSED] 0xE220 (BATTLEMAGE)
[15:12:05] [PASSED] 0xE221 (BATTLEMAGE)
[15:12:05] [PASSED] 0xE222 (BATTLEMAGE)
[15:12:05] [PASSED] 0xE223 (BATTLEMAGE)
[15:12:05] [PASSED] 0xB080 (PANTHERLAKE)
[15:12:05] [PASSED] 0xB081 (PANTHERLAKE)
[15:12:05] [PASSED] 0xB082 (PANTHERLAKE)
[15:12:05] [PASSED] 0xB083 (PANTHERLAKE)
[15:12:05] [PASSED] 0xB084 (PANTHERLAKE)
[15:12:05] [PASSED] 0xB085 (PANTHERLAKE)
[15:12:05] [PASSED] 0xB086 (PANTHERLAKE)
[15:12:05] [PASSED] 0xB087 (PANTHERLAKE)
[15:12:05] [PASSED] 0xB08F (PANTHERLAKE)
[15:12:05] [PASSED] 0xB090 (PANTHERLAKE)
[15:12:05] [PASSED] 0xB0A0 (PANTHERLAKE)
[15:12:05] [PASSED] 0xB0B0 (PANTHERLAKE)
[15:12:05] [PASSED] 0xFD80 (PANTHERLAKE)
[15:12:05] [PASSED] 0xFD81 (PANTHERLAKE)
[15:12:05] ============= [PASSED] check_platform_gt_count =============
[15:12:05] ===================== [PASSED] xe_pci ======================
[15:12:05] =================== xe_rtp (2 subtests) ====================
[15:12:05] =============== xe_rtp_process_to_sr_tests ================
[15:12:05] [PASSED] coalesce-same-reg
[15:12:05] [PASSED] no-match-no-add
[15:12:05] [PASSED] match-or
[15:12:05] [PASSED] match-or-xfail
[15:12:05] [PASSED] no-match-no-add-multiple-rules
[15:12:05] [PASSED] two-regs-two-entries
[15:12:05] [PASSED] clr-one-set-other
[15:12:05] [PASSED] set-field
[15:12:05] [PASSED] conflict-duplicate
[15:12:05] [PASSED] conflict-not-disjoint
[15:12:05] [PASSED] conflict-reg-type
[15:12:05] =========== [PASSED] xe_rtp_process_to_sr_tests ============
[15:12:05] ================== xe_rtp_process_tests ===================
[15:12:05] [PASSED] active1
[15:12:05] [PASSED] active2
[15:12:05] [PASSED] active-inactive
[15:12:05] [PASSED] inactive-active
[15:12:05] [PASSED] inactive-1st_or_active-inactive
[15:12:05] [PASSED] inactive-2nd_or_active-inactive
[15:12:05] [PASSED] inactive-last_or_active-inactive
[15:12:05] [PASSED] inactive-no_or_active-inactive
[15:12:05] ============== [PASSED] xe_rtp_process_tests ===============
[15:12:05] ===================== [PASSED] xe_rtp ======================
[15:12:05] ==================== xe_wa (1 subtest) =====================
[15:12:05] ======================== xe_wa_gt =========================
[15:12:05] [PASSED] TIGERLAKE B0
[15:12:05] [PASSED] DG1 A0
[15:12:05] [PASSED] DG1 B0
[15:12:05] [PASSED] ALDERLAKE_S A0
[15:12:05] [PASSED] ALDERLAKE_S B0
stty: 'standard input': Inappropriate ioctl for device
[15:12:05] [PASSED] ALDERLAKE_S C0
[15:12:05] [PASSED] ALDERLAKE_S D0
[15:12:05] [PASSED] ALDERLAKE_P A0
[15:12:05] [PASSED] ALDERLAKE_P B0
[15:12:05] [PASSED] ALDERLAKE_P C0
[15:12:05] [PASSED] ALDERLAKE_S RPLS D0
[15:12:05] [PASSED] ALDERLAKE_P RPLU E0
[15:12:05] [PASSED] DG2 G10 C0
[15:12:05] [PASSED] DG2 G11 B1
[15:12:05] [PASSED] DG2 G12 A1
[15:12:05] [PASSED] METEORLAKE 12.70(Xe_LPG) A0 13.00(Xe_LPM+) A0
[15:12:05] [PASSED] METEORLAKE 12.71(Xe_LPG) A0 13.00(Xe_LPM+) A0
[15:12:05] [PASSED] METEORLAKE 12.74(Xe_LPG+) A0 13.00(Xe_LPM+) A0
[15:12:05] [PASSED] LUNARLAKE 20.04(Xe2_LPG) A0 20.00(Xe2_LPM) A0
[15:12:05] [PASSED] LUNARLAKE 20.04(Xe2_LPG) B0 20.00(Xe2_LPM) A0
[15:12:05] [PASSED] BATTLEMAGE 20.01(Xe2_HPG) A0 13.01(Xe2_HPM) A1
[15:12:05] [PASSED] PANTHERLAKE 30.00(Xe3_LPG) A0 30.00(Xe3_LPM) A0
[15:12:05] ==================== [PASSED] xe_wa_gt =====================
[15:12:05] ====================== [PASSED] xe_wa ======================
[15:12:05] ============================================================
[15:12:05] Testing complete. Ran 306 tests: passed: 288, skipped: 18
[15:12:05] Elapsed time: 33.605s total, 4.213s configuring, 29.026s building, 0.323s running
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/tests/.kunitconfig
[15:12:05] Configuring KUnit Kernel ...
Regenerating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[15:12:07] 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
[15:12:31] Starting KUnit Kernel (1/1)...
[15:12:31] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[15:12:31] ============ drm_test_pick_cmdline (2 subtests) ============
[15:12:31] [PASSED] drm_test_pick_cmdline_res_1920_1080_60
[15:12:31] =============== drm_test_pick_cmdline_named ===============
[15:12:31] [PASSED] NTSC
[15:12:31] [PASSED] NTSC-J
[15:12:31] [PASSED] PAL
[15:12:31] [PASSED] PAL-M
[15:12:31] =========== [PASSED] drm_test_pick_cmdline_named ===========
[15:12:31] ============== [PASSED] drm_test_pick_cmdline ==============
[15:12:31] == drm_test_atomic_get_connector_for_encoder (1 subtest) ===
[15:12:31] [PASSED] drm_test_drm_atomic_get_connector_for_encoder
[15:12:31] ==== [PASSED] drm_test_atomic_get_connector_for_encoder ====
[15:12:31] =========== drm_validate_clone_mode (2 subtests) ===========
[15:12:31] ============== drm_test_check_in_clone_mode ===============
[15:12:31] [PASSED] in_clone_mode
[15:12:31] [PASSED] not_in_clone_mode
[15:12:31] ========== [PASSED] drm_test_check_in_clone_mode ===========
[15:12:31] =============== drm_test_check_valid_clones ===============
[15:12:31] [PASSED] not_in_clone_mode
[15:12:31] [PASSED] valid_clone
[15:12:31] [PASSED] invalid_clone
[15:12:31] =========== [PASSED] drm_test_check_valid_clones ===========
[15:12:31] ============= [PASSED] drm_validate_clone_mode =============
[15:12:31] ============= drm_validate_modeset (1 subtest) =============
[15:12:31] [PASSED] drm_test_check_connector_changed_modeset
[15:12:31] ============== [PASSED] drm_validate_modeset ===============
[15:12:31] ====== drm_test_bridge_get_current_state (2 subtests) ======
[15:12:31] [PASSED] drm_test_drm_bridge_get_current_state_atomic
[15:12:31] [PASSED] drm_test_drm_bridge_get_current_state_legacy
[15:12:31] ======== [PASSED] drm_test_bridge_get_current_state ========
[15:12:31] ====== drm_test_bridge_helper_reset_crtc (3 subtests) ======
[15:12:31] [PASSED] drm_test_drm_bridge_helper_reset_crtc_atomic
[15:12:31] [PASSED] drm_test_drm_bridge_helper_reset_crtc_atomic_disabled
[15:12:31] [PASSED] drm_test_drm_bridge_helper_reset_crtc_legacy
[15:12:31] ======== [PASSED] drm_test_bridge_helper_reset_crtc ========
[15:12:31] ============== drm_bridge_alloc (2 subtests) ===============
[15:12:31] [PASSED] drm_test_drm_bridge_alloc_basic
[15:12:31] [PASSED] drm_test_drm_bridge_alloc_get_put
[15:12:31] ================ [PASSED] drm_bridge_alloc =================
[15:12:31] ================== drm_buddy (7 subtests) ==================
[15:12:31] [PASSED] drm_test_buddy_alloc_limit
[15:12:31] [PASSED] drm_test_buddy_alloc_optimistic
[15:12:31] [PASSED] drm_test_buddy_alloc_pessimistic
[15:12:31] [PASSED] drm_test_buddy_alloc_pathological
[15:12:31] [PASSED] drm_test_buddy_alloc_contiguous
[15:12:31] [PASSED] drm_test_buddy_alloc_clear
[15:12:31] [PASSED] drm_test_buddy_alloc_range_bias
[15:12:31] ==================== [PASSED] drm_buddy ====================
[15:12:31] ============= drm_cmdline_parser (40 subtests) =============
[15:12:31] [PASSED] drm_test_cmdline_force_d_only
[15:12:31] [PASSED] drm_test_cmdline_force_D_only_dvi
[15:12:31] [PASSED] drm_test_cmdline_force_D_only_hdmi
[15:12:31] [PASSED] drm_test_cmdline_force_D_only_not_digital
[15:12:31] [PASSED] drm_test_cmdline_force_e_only
[15:12:31] [PASSED] drm_test_cmdline_res
[15:12:31] [PASSED] drm_test_cmdline_res_vesa
[15:12:31] [PASSED] drm_test_cmdline_res_vesa_rblank
[15:12:31] [PASSED] drm_test_cmdline_res_rblank
[15:12:31] [PASSED] drm_test_cmdline_res_bpp
[15:12:31] [PASSED] drm_test_cmdline_res_refresh
[15:12:31] [PASSED] drm_test_cmdline_res_bpp_refresh
[15:12:31] [PASSED] drm_test_cmdline_res_bpp_refresh_interlaced
[15:12:31] [PASSED] drm_test_cmdline_res_bpp_refresh_margins
[15:12:31] [PASSED] drm_test_cmdline_res_bpp_refresh_force_off
[15:12:31] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on
[15:12:31] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on_analog
[15:12:31] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on_digital
[15:12:31] [PASSED] drm_test_cmdline_res_bpp_refresh_interlaced_margins_force_on
[15:12:31] [PASSED] drm_test_cmdline_res_margins_force_on
[15:12:31] [PASSED] drm_test_cmdline_res_vesa_margins
[15:12:31] [PASSED] drm_test_cmdline_name
[15:12:31] [PASSED] drm_test_cmdline_name_bpp
[15:12:31] [PASSED] drm_test_cmdline_name_option
[15:12:31] [PASSED] drm_test_cmdline_name_bpp_option
[15:12:31] [PASSED] drm_test_cmdline_rotate_0
[15:12:31] [PASSED] drm_test_cmdline_rotate_90
[15:12:31] [PASSED] drm_test_cmdline_rotate_180
[15:12:31] [PASSED] drm_test_cmdline_rotate_270
[15:12:31] [PASSED] drm_test_cmdline_hmirror
[15:12:31] [PASSED] drm_test_cmdline_vmirror
[15:12:31] [PASSED] drm_test_cmdline_margin_options
[15:12:31] [PASSED] drm_test_cmdline_multiple_options
[15:12:31] [PASSED] drm_test_cmdline_bpp_extra_and_option
[15:12:31] [PASSED] drm_test_cmdline_extra_and_option
[15:12:31] [PASSED] drm_test_cmdline_freestanding_options
[15:12:31] [PASSED] drm_test_cmdline_freestanding_force_e_and_options
[15:12:31] [PASSED] drm_test_cmdline_panel_orientation
[15:12:31] ================ drm_test_cmdline_invalid =================
[15:12:31] [PASSED] margin_only
[15:12:31] [PASSED] interlace_only
[15:12:31] [PASSED] res_missing_x
[15:12:31] [PASSED] res_missing_y
[15:12:31] [PASSED] res_bad_y
[15:12:31] [PASSED] res_missing_y_bpp
[15:12:31] [PASSED] res_bad_bpp
[15:12:31] [PASSED] res_bad_refresh
[15:12:31] [PASSED] res_bpp_refresh_force_on_off
[15:12:31] [PASSED] res_invalid_mode
[15:12:31] [PASSED] res_bpp_wrong_place_mode
[15:12:31] [PASSED] name_bpp_refresh
[15:12:31] [PASSED] name_refresh
[15:12:31] [PASSED] name_refresh_wrong_mode
[15:12:31] [PASSED] name_refresh_invalid_mode
[15:12:31] [PASSED] rotate_multiple
[15:12:31] [PASSED] rotate_invalid_val
[15:12:31] [PASSED] rotate_truncated
[15:12:31] [PASSED] invalid_option
[15:12:31] [PASSED] invalid_tv_option
[15:12:31] [PASSED] truncated_tv_option
[15:12:31] ============ [PASSED] drm_test_cmdline_invalid =============
[15:12:31] =============== drm_test_cmdline_tv_options ===============
[15:12:31] [PASSED] NTSC
[15:12:31] [PASSED] NTSC_443
[15:12:31] [PASSED] NTSC_J
[15:12:31] [PASSED] PAL
[15:12:31] [PASSED] PAL_M
[15:12:31] [PASSED] PAL_N
[15:12:31] [PASSED] SECAM
[15:12:31] [PASSED] MONO_525
[15:12:31] [PASSED] MONO_625
[15:12:31] =========== [PASSED] drm_test_cmdline_tv_options ===========
[15:12:31] =============== [PASSED] drm_cmdline_parser ================
[15:12:31] ========== drmm_connector_hdmi_init (20 subtests) ==========
[15:12:31] [PASSED] drm_test_connector_hdmi_init_valid
[15:12:31] [PASSED] drm_test_connector_hdmi_init_bpc_8
[15:12:31] [PASSED] drm_test_connector_hdmi_init_bpc_10
[15:12:31] [PASSED] drm_test_connector_hdmi_init_bpc_12
[15:12:31] [PASSED] drm_test_connector_hdmi_init_bpc_invalid
[15:12:31] [PASSED] drm_test_connector_hdmi_init_bpc_null
[15:12:31] [PASSED] drm_test_connector_hdmi_init_formats_empty
[15:12:31] [PASSED] drm_test_connector_hdmi_init_formats_no_rgb
[15:12:31] === drm_test_connector_hdmi_init_formats_yuv420_allowed ===
[15:12:31] [PASSED] supported_formats=0x9 yuv420_allowed=1
[15:12:31] [PASSED] supported_formats=0x9 yuv420_allowed=0
[15:12:31] [PASSED] supported_formats=0x3 yuv420_allowed=1
[15:12:31] [PASSED] supported_formats=0x3 yuv420_allowed=0
[15:12:31] === [PASSED] drm_test_connector_hdmi_init_formats_yuv420_allowed ===
[15:12:31] [PASSED] drm_test_connector_hdmi_init_null_ddc
[15:12:31] [PASSED] drm_test_connector_hdmi_init_null_product
[15:12:31] [PASSED] drm_test_connector_hdmi_init_null_vendor
[15:12:31] [PASSED] drm_test_connector_hdmi_init_product_length_exact
[15:12:31] [PASSED] drm_test_connector_hdmi_init_product_length_too_long
[15:12:31] [PASSED] drm_test_connector_hdmi_init_product_valid
[15:12:31] [PASSED] drm_test_connector_hdmi_init_vendor_length_exact
[15:12:31] [PASSED] drm_test_connector_hdmi_init_vendor_length_too_long
[15:12:31] [PASSED] drm_test_connector_hdmi_init_vendor_valid
[15:12:31] ========= drm_test_connector_hdmi_init_type_valid =========
[15:12:31] [PASSED] HDMI-A
[15:12:31] [PASSED] HDMI-B
[15:12:31] ===== [PASSED] drm_test_connector_hdmi_init_type_valid =====
[15:12:31] ======== drm_test_connector_hdmi_init_type_invalid ========
[15:12:31] [PASSED] Unknown
[15:12:31] [PASSED] VGA
[15:12:31] [PASSED] DVI-I
[15:12:31] [PASSED] DVI-D
[15:12:31] [PASSED] DVI-A
[15:12:31] [PASSED] Composite
[15:12:31] [PASSED] SVIDEO
[15:12:31] [PASSED] LVDS
[15:12:31] [PASSED] Component
[15:12:31] [PASSED] DIN
[15:12:31] [PASSED] DP
[15:12:31] [PASSED] TV
[15:12:31] [PASSED] eDP
[15:12:31] [PASSED] Virtual
[15:12:31] [PASSED] DSI
[15:12:31] [PASSED] DPI
[15:12:31] [PASSED] Writeback
[15:12:31] [PASSED] SPI
[15:12:31] [PASSED] USB
[15:12:31] ==== [PASSED] drm_test_connector_hdmi_init_type_invalid ====
[15:12:31] ============ [PASSED] drmm_connector_hdmi_init =============
[15:12:31] ============= drmm_connector_init (3 subtests) =============
[15:12:31] [PASSED] drm_test_drmm_connector_init
[15:12:31] [PASSED] drm_test_drmm_connector_init_null_ddc
[15:12:31] ========= drm_test_drmm_connector_init_type_valid =========
[15:12:31] [PASSED] Unknown
[15:12:31] [PASSED] VGA
[15:12:31] [PASSED] DVI-I
[15:12:31] [PASSED] DVI-D
[15:12:31] [PASSED] DVI-A
[15:12:31] [PASSED] Composite
[15:12:31] [PASSED] SVIDEO
[15:12:31] [PASSED] LVDS
[15:12:31] [PASSED] Component
[15:12:31] [PASSED] DIN
[15:12:31] [PASSED] DP
[15:12:31] [PASSED] HDMI-A
[15:12:31] [PASSED] HDMI-B
[15:12:31] [PASSED] TV
[15:12:31] [PASSED] eDP
[15:12:31] [PASSED] Virtual
[15:12:31] [PASSED] DSI
[15:12:31] [PASSED] DPI
[15:12:31] [PASSED] Writeback
[15:12:31] [PASSED] SPI
[15:12:31] [PASSED] USB
[15:12:31] ===== [PASSED] drm_test_drmm_connector_init_type_valid =====
[15:12:31] =============== [PASSED] drmm_connector_init ===============
[15:12:31] ========= drm_connector_dynamic_init (6 subtests) ==========
[15:12:31] [PASSED] drm_test_drm_connector_dynamic_init
[15:12:31] [PASSED] drm_test_drm_connector_dynamic_init_null_ddc
[15:12:31] [PASSED] drm_test_drm_connector_dynamic_init_not_added
[15:12:31] [PASSED] drm_test_drm_connector_dynamic_init_properties
[15:12:31] ===== drm_test_drm_connector_dynamic_init_type_valid ======
[15:12:31] [PASSED] Unknown
[15:12:31] [PASSED] VGA
[15:12:31] [PASSED] DVI-I
[15:12:31] [PASSED] DVI-D
[15:12:31] [PASSED] DVI-A
[15:12:31] [PASSED] Composite
[15:12:31] [PASSED] SVIDEO
[15:12:31] [PASSED] LVDS
[15:12:31] [PASSED] Component
[15:12:31] [PASSED] DIN
[15:12:31] [PASSED] DP
[15:12:31] [PASSED] HDMI-A
[15:12:31] [PASSED] HDMI-B
[15:12:31] [PASSED] TV
[15:12:31] [PASSED] eDP
[15:12:31] [PASSED] Virtual
[15:12:31] [PASSED] DSI
[15:12:31] [PASSED] DPI
[15:12:31] [PASSED] Writeback
[15:12:31] [PASSED] SPI
[15:12:31] [PASSED] USB
[15:12:31] = [PASSED] drm_test_drm_connector_dynamic_init_type_valid ==
[15:12:31] ======== drm_test_drm_connector_dynamic_init_name =========
[15:12:31] [PASSED] Unknown
[15:12:31] [PASSED] VGA
[15:12:31] [PASSED] DVI-I
[15:12:31] [PASSED] DVI-D
[15:12:31] [PASSED] DVI-A
[15:12:31] [PASSED] Composite
[15:12:31] [PASSED] SVIDEO
[15:12:31] [PASSED] LVDS
[15:12:31] [PASSED] Component
[15:12:31] [PASSED] DIN
[15:12:31] [PASSED] DP
[15:12:31] [PASSED] HDMI-A
[15:12:31] [PASSED] HDMI-B
[15:12:31] [PASSED] TV
[15:12:31] [PASSED] eDP
[15:12:31] [PASSED] Virtual
[15:12:31] [PASSED] DSI
[15:12:31] [PASSED] DPI
[15:12:31] [PASSED] Writeback
[15:12:31] [PASSED] SPI
[15:12:31] [PASSED] USB
[15:12:31] ==== [PASSED] drm_test_drm_connector_dynamic_init_name =====
[15:12:31] =========== [PASSED] drm_connector_dynamic_init ============
[15:12:31] ==== drm_connector_dynamic_register_early (4 subtests) =====
[15:12:31] [PASSED] drm_test_drm_connector_dynamic_register_early_on_list
[15:12:31] [PASSED] drm_test_drm_connector_dynamic_register_early_defer
[15:12:31] [PASSED] drm_test_drm_connector_dynamic_register_early_no_init
[15:12:31] [PASSED] drm_test_drm_connector_dynamic_register_early_no_mode_object
[15:12:31] ====== [PASSED] drm_connector_dynamic_register_early =======
[15:12:31] ======= drm_connector_dynamic_register (7 subtests) ========
[15:12:31] [PASSED] drm_test_drm_connector_dynamic_register_on_list
[15:12:31] [PASSED] drm_test_drm_connector_dynamic_register_no_defer
[15:12:31] [PASSED] drm_test_drm_connector_dynamic_register_no_init
[15:12:31] [PASSED] drm_test_drm_connector_dynamic_register_mode_object
[15:12:31] [PASSED] drm_test_drm_connector_dynamic_register_sysfs
[15:12:31] [PASSED] drm_test_drm_connector_dynamic_register_sysfs_name
[15:12:31] [PASSED] drm_test_drm_connector_dynamic_register_debugfs
[15:12:31] ========= [PASSED] drm_connector_dynamic_register ==========
[15:12:31] = drm_connector_attach_broadcast_rgb_property (2 subtests) =
[15:12:31] [PASSED] drm_test_drm_connector_attach_broadcast_rgb_property
[15:12:31] [PASSED] drm_test_drm_connector_attach_broadcast_rgb_property_hdmi_connector
[15:12:31] === [PASSED] drm_connector_attach_broadcast_rgb_property ===
[15:12:31] ========== drm_get_tv_mode_from_name (2 subtests) ==========
[15:12:31] ========== drm_test_get_tv_mode_from_name_valid ===========
[15:12:31] [PASSED] NTSC
[15:12:31] [PASSED] NTSC-443
[15:12:31] [PASSED] NTSC-J
[15:12:31] [PASSED] PAL
[15:12:31] [PASSED] PAL-M
[15:12:31] [PASSED] PAL-N
[15:12:31] [PASSED] SECAM
[15:12:31] [PASSED] Mono
[15:12:31] ====== [PASSED] drm_test_get_tv_mode_from_name_valid =======
[15:12:31] [PASSED] drm_test_get_tv_mode_from_name_truncated
[15:12:31] ============ [PASSED] drm_get_tv_mode_from_name ============
[15:12:31] = drm_test_connector_hdmi_compute_mode_clock (12 subtests) =
[15:12:31] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb
[15:12:31] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_10bpc
[15:12:31] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_10bpc_vic_1
[15:12:31] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_12bpc
[15:12:31] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_12bpc_vic_1
[15:12:31] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_double
[15:12:31] = drm_test_connector_hdmi_compute_mode_clock_yuv420_valid =
[15:12:31] [PASSED] VIC 96
[15:12:31] [PASSED] VIC 97
[15:12:31] [PASSED] VIC 101
[15:12:31] [PASSED] VIC 102
[15:12:31] [PASSED] VIC 106
[15:12:31] [PASSED] VIC 107
[15:12:31] === [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_valid ===
[15:12:31] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_10_bpc
[15:12:31] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_12_bpc
[15:12:31] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_8_bpc
[15:12:31] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_10_bpc
[15:12:31] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_12_bpc
[15:12:31] === [PASSED] drm_test_connector_hdmi_compute_mode_clock ====
[15:12:31] == drm_hdmi_connector_get_broadcast_rgb_name (2 subtests) ==
[15:12:31] === drm_test_drm_hdmi_connector_get_broadcast_rgb_name ====
[15:12:31] [PASSED] Automatic
[15:12:31] [PASSED] Full
[15:12:31] [PASSED] Limited 16:235
[15:12:31] === [PASSED] drm_test_drm_hdmi_connector_get_broadcast_rgb_name ===
[15:12:31] [PASSED] drm_test_drm_hdmi_connector_get_broadcast_rgb_name_invalid
[15:12:31] ==== [PASSED] drm_hdmi_connector_get_broadcast_rgb_name ====
[15:12:31] == drm_hdmi_connector_get_output_format_name (2 subtests) ==
[15:12:31] === drm_test_drm_hdmi_connector_get_output_format_name ====
[15:12:31] [PASSED] RGB
[15:12:31] [PASSED] YUV 4:2:0
[15:12:31] [PASSED] YUV 4:2:2
[15:12:31] [PASSED] YUV 4:4:4
[15:12:31] === [PASSED] drm_test_drm_hdmi_connector_get_output_format_name ===
[15:12:31] [PASSED] drm_test_drm_hdmi_connector_get_output_format_name_invalid
[15:12:31] ==== [PASSED] drm_hdmi_connector_get_output_format_name ====
[15:12:31] ============= drm_damage_helper (21 subtests) ==============
[15:12:31] [PASSED] drm_test_damage_iter_no_damage
[15:12:31] [PASSED] drm_test_damage_iter_no_damage_fractional_src
[15:12:31] [PASSED] drm_test_damage_iter_no_damage_src_moved
[15:12:31] [PASSED] drm_test_damage_iter_no_damage_fractional_src_moved
[15:12:31] [PASSED] drm_test_damage_iter_no_damage_not_visible
[15:12:31] [PASSED] drm_test_damage_iter_no_damage_no_crtc
[15:12:31] [PASSED] drm_test_damage_iter_no_damage_no_fb
[15:12:31] [PASSED] drm_test_damage_iter_simple_damage
[15:12:31] [PASSED] drm_test_damage_iter_single_damage
[15:12:31] [PASSED] drm_test_damage_iter_single_damage_intersect_src
[15:12:31] [PASSED] drm_test_damage_iter_single_damage_outside_src
[15:12:31] [PASSED] drm_test_damage_iter_single_damage_fractional_src
[15:12:31] [PASSED] drm_test_damage_iter_single_damage_intersect_fractional_src
[15:12:31] [PASSED] drm_test_damage_iter_single_damage_outside_fractional_src
[15:12:31] [PASSED] drm_test_damage_iter_single_damage_src_moved
[15:12:31] [PASSED] drm_test_damage_iter_single_damage_fractional_src_moved
[15:12:31] [PASSED] drm_test_damage_iter_damage
[15:12:31] [PASSED] drm_test_damage_iter_damage_one_intersect
[15:12:31] [PASSED] drm_test_damage_iter_damage_one_outside
[15:12:31] [PASSED] drm_test_damage_iter_damage_src_moved
[15:12:31] [PASSED] drm_test_damage_iter_damage_not_visible
[15:12:31] ================ [PASSED] drm_damage_helper ================
[15:12:31] ============== drm_dp_mst_helper (3 subtests) ==============
[15:12:31] ============== drm_test_dp_mst_calc_pbn_mode ==============
[15:12:31] [PASSED] Clock 154000 BPP 30 DSC disabled
[15:12:31] [PASSED] Clock 234000 BPP 30 DSC disabled
[15:12:31] [PASSED] Clock 297000 BPP 24 DSC disabled
[15:12:31] [PASSED] Clock 332880 BPP 24 DSC enabled
[15:12:31] [PASSED] Clock 324540 BPP 24 DSC enabled
[15:12:31] ========== [PASSED] drm_test_dp_mst_calc_pbn_mode ==========
[15:12:31] ============== drm_test_dp_mst_calc_pbn_div ===============
[15:12:31] [PASSED] Link rate 2000000 lane count 4
[15:12:31] [PASSED] Link rate 2000000 lane count 2
[15:12:31] [PASSED] Link rate 2000000 lane count 1
[15:12:31] [PASSED] Link rate 1350000 lane count 4
[15:12:31] [PASSED] Link rate 1350000 lane count 2
[15:12:31] [PASSED] Link rate 1350000 lane count 1
[15:12:31] [PASSED] Link rate 1000000 lane count 4
[15:12:31] [PASSED] Link rate 1000000 lane count 2
[15:12:31] [PASSED] Link rate 1000000 lane count 1
[15:12:31] [PASSED] Link rate 810000 lane count 4
[15:12:31] [PASSED] Link rate 810000 lane count 2
[15:12:31] [PASSED] Link rate 810000 lane count 1
[15:12:31] [PASSED] Link rate 540000 lane count 4
[15:12:31] [PASSED] Link rate 540000 lane count 2
[15:12:31] [PASSED] Link rate 540000 lane count 1
[15:12:31] [PASSED] Link rate 270000 lane count 4
[15:12:31] [PASSED] Link rate 270000 lane count 2
[15:12:31] [PASSED] Link rate 270000 lane count 1
[15:12:31] [PASSED] Link rate 162000 lane count 4
[15:12:31] [PASSED] Link rate 162000 lane count 2
[15:12:31] [PASSED] Link rate 162000 lane count 1
[15:12:31] ========== [PASSED] drm_test_dp_mst_calc_pbn_div ===========
[15:12:31] ========= drm_test_dp_mst_sideband_msg_req_decode =========
[15:12:31] [PASSED] DP_ENUM_PATH_RESOURCES with port number
[15:12:31] [PASSED] DP_POWER_UP_PHY with port number
[15:12:31] [PASSED] DP_POWER_DOWN_PHY with port number
[15:12:31] [PASSED] DP_ALLOCATE_PAYLOAD with SDP stream sinks
[15:12:31] [PASSED] DP_ALLOCATE_PAYLOAD with port number
[15:12:31] [PASSED] DP_ALLOCATE_PAYLOAD with VCPI
[15:12:31] [PASSED] DP_ALLOCATE_PAYLOAD with PBN
[15:12:31] [PASSED] DP_QUERY_PAYLOAD with port number
[15:12:31] [PASSED] DP_QUERY_PAYLOAD with VCPI
[15:12:31] [PASSED] DP_REMOTE_DPCD_READ with port number
[15:12:31] [PASSED] DP_REMOTE_DPCD_READ with DPCD address
[15:12:31] [PASSED] DP_REMOTE_DPCD_READ with max number of bytes
[15:12:31] [PASSED] DP_REMOTE_DPCD_WRITE with port number
[15:12:31] [PASSED] DP_REMOTE_DPCD_WRITE with DPCD address
[15:12:31] [PASSED] DP_REMOTE_DPCD_WRITE with data array
[15:12:31] [PASSED] DP_REMOTE_I2C_READ with port number
[15:12:31] [PASSED] DP_REMOTE_I2C_READ with I2C device ID
[15:12:31] [PASSED] DP_REMOTE_I2C_READ with transactions array
[15:12:31] [PASSED] DP_REMOTE_I2C_WRITE with port number
[15:12:31] [PASSED] DP_REMOTE_I2C_WRITE with I2C device ID
[15:12:31] [PASSED] DP_REMOTE_I2C_WRITE with data array
[15:12:31] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream ID
[15:12:31] [PASSED] DP_QUERY_STREAM_ENC_STATUS with client ID
[15:12:31] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream event
[15:12:31] [PASSED] DP_QUERY_STREAM_ENC_STATUS with valid stream event
[15:12:31] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream behavior
[15:12:31] [PASSED] DP_QUERY_STREAM_ENC_STATUS with a valid stream behavior
[15:12:31] ===== [PASSED] drm_test_dp_mst_sideband_msg_req_decode =====
[15:12:31] ================ [PASSED] drm_dp_mst_helper ================
[15:12:31] ================== drm_exec (7 subtests) ===================
[15:12:31] [PASSED] sanitycheck
[15:12:31] [PASSED] test_lock
[15:12:31] [PASSED] test_lock_unlock
[15:12:31] [PASSED] test_duplicates
[15:12:31] [PASSED] test_prepare
[15:12:31] [PASSED] test_prepare_array
[15:12:31] [PASSED] test_multiple_loops
[15:12:31] ==================== [PASSED] drm_exec =====================
[15:12:31] =========== drm_format_helper_test (17 subtests) ===========
[15:12:31] ============== drm_test_fb_xrgb8888_to_gray8 ==============
[15:12:31] [PASSED] single_pixel_source_buffer
[15:12:31] [PASSED] single_pixel_clip_rectangle
[15:12:31] [PASSED] well_known_colors
[15:12:31] [PASSED] destination_pitch
[15:12:31] ========== [PASSED] drm_test_fb_xrgb8888_to_gray8 ==========
[15:12:31] ============= drm_test_fb_xrgb8888_to_rgb332 ==============
[15:12:31] [PASSED] single_pixel_source_buffer
[15:12:31] [PASSED] single_pixel_clip_rectangle
[15:12:31] [PASSED] well_known_colors
[15:12:31] [PASSED] destination_pitch
[15:12:31] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb332 ==========
[15:12:31] ============= drm_test_fb_xrgb8888_to_rgb565 ==============
[15:12:31] [PASSED] single_pixel_source_buffer
[15:12:31] [PASSED] single_pixel_clip_rectangle
[15:12:31] [PASSED] well_known_colors
[15:12:31] [PASSED] destination_pitch
[15:12:31] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb565 ==========
[15:12:31] ============ drm_test_fb_xrgb8888_to_xrgb1555 =============
[15:12:31] [PASSED] single_pixel_source_buffer
[15:12:31] [PASSED] single_pixel_clip_rectangle
[15:12:31] [PASSED] well_known_colors
[15:12:31] [PASSED] destination_pitch
[15:12:31] ======== [PASSED] drm_test_fb_xrgb8888_to_xrgb1555 =========
[15:12:31] ============ drm_test_fb_xrgb8888_to_argb1555 =============
[15:12:31] [PASSED] single_pixel_source_buffer
[15:12:31] [PASSED] single_pixel_clip_rectangle
[15:12:31] [PASSED] well_known_colors
[15:12:31] [PASSED] destination_pitch
[15:12:31] ======== [PASSED] drm_test_fb_xrgb8888_to_argb1555 =========
[15:12:31] ============ drm_test_fb_xrgb8888_to_rgba5551 =============
[15:12:31] [PASSED] single_pixel_source_buffer
[15:12:31] [PASSED] single_pixel_clip_rectangle
[15:12:31] [PASSED] well_known_colors
[15:12:31] [PASSED] destination_pitch
[15:12:31] ======== [PASSED] drm_test_fb_xrgb8888_to_rgba5551 =========
[15:12:31] ============= drm_test_fb_xrgb8888_to_rgb888 ==============
[15:12:31] [PASSED] single_pixel_source_buffer
[15:12:31] [PASSED] single_pixel_clip_rectangle
[15:12:31] [PASSED] well_known_colors
[15:12:31] [PASSED] destination_pitch
[15:12:31] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb888 ==========
[15:12:31] ============= drm_test_fb_xrgb8888_to_bgr888 ==============
[15:12:31] [PASSED] single_pixel_source_buffer
[15:12:31] [PASSED] single_pixel_clip_rectangle
[15:12:31] [PASSED] well_known_colors
[15:12:31] [PASSED] destination_pitch
[15:12:31] ========= [PASSED] drm_test_fb_xrgb8888_to_bgr888 ==========
[15:12:31] ============ drm_test_fb_xrgb8888_to_argb8888 =============
[15:12:31] [PASSED] single_pixel_source_buffer
[15:12:31] [PASSED] single_pixel_clip_rectangle
[15:12:31] [PASSED] well_known_colors
[15:12:31] [PASSED] destination_pitch
[15:12:31] ======== [PASSED] drm_test_fb_xrgb8888_to_argb8888 =========
[15:12:31] =========== drm_test_fb_xrgb8888_to_xrgb2101010 ===========
[15:12:31] [PASSED] single_pixel_source_buffer
[15:12:31] [PASSED] single_pixel_clip_rectangle
[15:12:31] [PASSED] well_known_colors
[15:12:31] [PASSED] destination_pitch
[15:12:31] ======= [PASSED] drm_test_fb_xrgb8888_to_xrgb2101010 =======
[15:12:31] =========== drm_test_fb_xrgb8888_to_argb2101010 ===========
[15:12:31] [PASSED] single_pixel_source_buffer
[15:12:31] [PASSED] single_pixel_clip_rectangle
[15:12:31] [PASSED] well_known_colors
[15:12:31] [PASSED] destination_pitch
[15:12:31] ======= [PASSED] drm_test_fb_xrgb8888_to_argb2101010 =======
[15:12:31] ============== drm_test_fb_xrgb8888_to_mono ===============
[15:12:31] [PASSED] single_pixel_source_buffer
[15:12:31] [PASSED] single_pixel_clip_rectangle
[15:12:31] [PASSED] well_known_colors
[15:12:31] [PASSED] destination_pitch
[15:12:31] ========== [PASSED] drm_test_fb_xrgb8888_to_mono ===========
[15:12:31] ==================== drm_test_fb_swab =====================
[15:12:31] [PASSED] single_pixel_source_buffer
[15:12:31] [PASSED] single_pixel_clip_rectangle
[15:12:31] [PASSED] well_known_colors
[15:12:31] [PASSED] destination_pitch
[15:12:31] ================ [PASSED] drm_test_fb_swab =================
[15:12:31] ============ drm_test_fb_xrgb8888_to_xbgr8888 =============
[15:12:31] [PASSED] single_pixel_source_buffer
[15:12:31] [PASSED] single_pixel_clip_rectangle
[15:12:31] [PASSED] well_known_colors
[15:12:31] [PASSED] destination_pitch
[15:12:31] ======== [PASSED] drm_test_fb_xrgb8888_to_xbgr8888 =========
[15:12:31] ============ drm_test_fb_xrgb8888_to_abgr8888 =============
[15:12:31] [PASSED] single_pixel_source_buffer
[15:12:31] [PASSED] single_pixel_clip_rectangle
[15:12:31] [PASSED] well_known_colors
[15:12:31] [PASSED] destination_pitch
[15:12:31] ======== [PASSED] drm_test_fb_xrgb8888_to_abgr8888 =========
[15:12:31] ================= drm_test_fb_clip_offset =================
[15:12:31] [PASSED] pass through
[15:12:31] [PASSED] horizontal offset
[15:12:31] [PASSED] vertical offset
[15:12:31] [PASSED] horizontal and vertical offset
[15:12:31] [PASSED] horizontal offset (custom pitch)
[15:12:31] [PASSED] vertical offset (custom pitch)
[15:12:31] [PASSED] horizontal and vertical offset (custom pitch)
[15:12:31] ============= [PASSED] drm_test_fb_clip_offset =============
[15:12:31] =================== drm_test_fb_memcpy ====================
[15:12:31] [PASSED] single_pixel_source_buffer: XR24 little-endian (0x34325258)
[15:12:31] [PASSED] single_pixel_source_buffer: XRA8 little-endian (0x38415258)
[15:12:31] [PASSED] single_pixel_source_buffer: YU24 little-endian (0x34325559)
[15:12:31] [PASSED] single_pixel_clip_rectangle: XB24 little-endian (0x34324258)
[15:12:31] [PASSED] single_pixel_clip_rectangle: XRA8 little-endian (0x38415258)
[15:12:31] [PASSED] single_pixel_clip_rectangle: YU24 little-endian (0x34325559)
[15:12:31] [PASSED] well_known_colors: XB24 little-endian (0x34324258)
[15:12:31] [PASSED] well_known_colors: XRA8 little-endian (0x38415258)
[15:12:31] [PASSED] well_known_colors: YU24 little-endian (0x34325559)
[15:12:31] [PASSED] destination_pitch: XB24 little-endian (0x34324258)
[15:12:31] [PASSED] destination_pitch: XRA8 little-endian (0x38415258)
[15:12:31] [PASSED] destination_pitch: YU24 little-endian (0x34325559)
[15:12:31] =============== [PASSED] drm_test_fb_memcpy ================
[15:12:31] ============= [PASSED] drm_format_helper_test ==============
[15:12:31] ================= drm_format (18 subtests) =================
[15:12:31] [PASSED] drm_test_format_block_width_invalid
[15:12:31] [PASSED] drm_test_format_block_width_one_plane
[15:12:31] [PASSED] drm_test_format_block_width_two_plane
[15:12:31] [PASSED] drm_test_format_block_width_three_plane
[15:12:31] [PASSED] drm_test_format_block_width_tiled
[15:12:31] [PASSED] drm_test_format_block_height_invalid
[15:12:31] [PASSED] drm_test_format_block_height_one_plane
[15:12:31] [PASSED] drm_test_format_block_height_two_plane
[15:12:31] [PASSED] drm_test_format_block_height_three_plane
[15:12:31] [PASSED] drm_test_format_block_height_tiled
[15:12:31] [PASSED] drm_test_format_min_pitch_invalid
[15:12:31] [PASSED] drm_test_format_min_pitch_one_plane_8bpp
[15:12:31] [PASSED] drm_test_format_min_pitch_one_plane_16bpp
[15:12:31] [PASSED] drm_test_format_min_pitch_one_plane_24bpp
[15:12:31] [PASSED] drm_test_format_min_pitch_one_plane_32bpp
[15:12:31] [PASSED] drm_test_format_min_pitch_two_plane
[15:12:31] [PASSED] drm_test_format_min_pitch_three_plane_8bpp
[15:12:31] [PASSED] drm_test_format_min_pitch_tiled
[15:12:31] =================== [PASSED] drm_format ====================
[15:12:31] ============== drm_framebuffer (10 subtests) ===============
[15:12:31] ========== drm_test_framebuffer_check_src_coords ==========
[15:12:31] [PASSED] Success: source fits into fb
[15:12:31] [PASSED] Fail: overflowing fb with x-axis coordinate
[15:12:31] [PASSED] Fail: overflowing fb with y-axis coordinate
[15:12:31] [PASSED] Fail: overflowing fb with source width
[15:12:31] [PASSED] Fail: overflowing fb with source height
[15:12:31] ====== [PASSED] drm_test_framebuffer_check_src_coords ======
[15:12:31] [PASSED] drm_test_framebuffer_cleanup
[15:12:31] =============== drm_test_framebuffer_create ===============
[15:12:31] [PASSED] ABGR8888 normal sizes
[15:12:31] [PASSED] ABGR8888 max sizes
[15:12:31] [PASSED] ABGR8888 pitch greater than min required
[15:12:31] [PASSED] ABGR8888 pitch less than min required
[15:12:31] [PASSED] ABGR8888 Invalid width
[15:12:31] [PASSED] ABGR8888 Invalid buffer handle
[15:12:31] [PASSED] No pixel format
[15:12:31] [PASSED] ABGR8888 Width 0
[15:12:31] [PASSED] ABGR8888 Height 0
[15:12:31] [PASSED] ABGR8888 Out of bound height * pitch combination
[15:12:31] [PASSED] ABGR8888 Large buffer offset
[15:12:31] [PASSED] ABGR8888 Buffer offset for inexistent plane
[15:12:31] [PASSED] ABGR8888 Invalid flag
[15:12:31] [PASSED] ABGR8888 Set DRM_MODE_FB_MODIFIERS without modifiers
[15:12:31] [PASSED] ABGR8888 Valid buffer modifier
[15:12:31] [PASSED] ABGR8888 Invalid buffer modifier(DRM_FORMAT_MOD_SAMSUNG_64_32_TILE)
[15:12:31] [PASSED] ABGR8888 Extra pitches without DRM_MODE_FB_MODIFIERS
[15:12:31] [PASSED] ABGR8888 Extra pitches with DRM_MODE_FB_MODIFIERS
[15:12:31] [PASSED] NV12 Normal sizes
[15:12:31] [PASSED] NV12 Max sizes
[15:12:31] [PASSED] NV12 Invalid pitch
[15:12:31] [PASSED] NV12 Invalid modifier/missing DRM_MODE_FB_MODIFIERS flag
[15:12:31] [PASSED] NV12 different modifier per-plane
[15:12:31] [PASSED] NV12 with DRM_FORMAT_MOD_SAMSUNG_64_32_TILE
[15:12:31] [PASSED] NV12 Valid modifiers without DRM_MODE_FB_MODIFIERS
[15:12:31] [PASSED] NV12 Modifier for inexistent plane
[15:12:31] [PASSED] NV12 Handle for inexistent plane
[15:12:31] [PASSED] NV12 Handle for inexistent plane without DRM_MODE_FB_MODIFIERS
[15:12:31] [PASSED] YVU420 DRM_MODE_FB_MODIFIERS set without modifier
[15:12:31] [PASSED] YVU420 Normal sizes
[15:12:31] [PASSED] YVU420 Max sizes
[15:12:31] [PASSED] YVU420 Invalid pitch
[15:12:31] [PASSED] YVU420 Different pitches
[15:12:31] [PASSED] YVU420 Different buffer offsets/pitches
[15:12:31] [PASSED] YVU420 Modifier set just for plane 0, without DRM_MODE_FB_MODIFIERS
[15:12:31] [PASSED] YVU420 Modifier set just for planes 0, 1, without DRM_MODE_FB_MODIFIERS
[15:12:31] [PASSED] YVU420 Modifier set just for plane 0, 1, with DRM_MODE_FB_MODIFIERS
[15:12:31] [PASSED] YVU420 Valid modifier
[15:12:31] [PASSED] YVU420 Different modifiers per plane
[15:12:31] [PASSED] YVU420 Modifier for inexistent plane
[15:12:31] [PASSED] YUV420_10BIT Invalid modifier(DRM_FORMAT_MOD_LINEAR)
[15:12:31] [PASSED] X0L2 Normal sizes
[15:12:31] [PASSED] X0L2 Max sizes
[15:12:31] [PASSED] X0L2 Invalid pitch
[15:12:31] [PASSED] X0L2 Pitch greater than minimum required
[15:12:31] [PASSED] X0L2 Handle for inexistent plane
[15:12:31] [PASSED] X0L2 Offset for inexistent plane, without DRM_MODE_FB_MODIFIERS set
[15:12:31] [PASSED] X0L2 Modifier without DRM_MODE_FB_MODIFIERS set
[15:12:31] [PASSED] X0L2 Valid modifier
[15:12:31] [PASSED] X0L2 Modifier for inexistent plane
[15:12:31] =========== [PASSED] drm_test_framebuffer_create ===========
[15:12:31] [PASSED] drm_test_framebuffer_free
[15:12:31] [PASSED] drm_test_framebuffer_init
[15:12:31] [PASSED] drm_test_framebuffer_init_bad_format
[15:12:31] [PASSED] drm_test_framebuffer_init_dev_mismatch
[15:12:31] [PASSED] drm_test_framebuffer_lookup
[15:12:31] [PASSED] drm_test_framebuffer_lookup_inexistent
[15:12:31] [PASSED] drm_test_framebuffer_modifiers_not_supported
[15:12:31] ================= [PASSED] drm_framebuffer =================
[15:12:31] ================ drm_gem_shmem (8 subtests) ================
[15:12:31] [PASSED] drm_gem_shmem_test_obj_create
[15:12:31] [PASSED] drm_gem_shmem_test_obj_create_private
[15:12:31] [PASSED] drm_gem_shmem_test_pin_pages
[15:12:31] [PASSED] drm_gem_shmem_test_vmap
[15:12:31] [PASSED] drm_gem_shmem_test_get_pages_sgt
[15:12:31] [PASSED] drm_gem_shmem_test_get_sg_table
[15:12:31] [PASSED] drm_gem_shmem_test_madvise
[15:12:31] [PASSED] drm_gem_shmem_test_purge
[15:12:31] ================== [PASSED] drm_gem_shmem ==================
[15:12:31] === drm_atomic_helper_connector_hdmi_check (27 subtests) ===
[15:12:31] [PASSED] drm_test_check_broadcast_rgb_auto_cea_mode
[15:12:31] [PASSED] drm_test_check_broadcast_rgb_auto_cea_mode_vic_1
[15:12:31] [PASSED] drm_test_check_broadcast_rgb_full_cea_mode
[15:12:31] [PASSED] drm_test_check_broadcast_rgb_full_cea_mode_vic_1
[15:12:31] [PASSED] drm_test_check_broadcast_rgb_limited_cea_mode
[15:12:31] [PASSED] drm_test_check_broadcast_rgb_limited_cea_mode_vic_1
[15:12:31] ====== drm_test_check_broadcast_rgb_cea_mode_yuv420 =======
[15:12:31] [PASSED] Automatic
[15:12:31] [PASSED] Full
[15:12:31] [PASSED] Limited 16:235
[15:12:31] == [PASSED] drm_test_check_broadcast_rgb_cea_mode_yuv420 ===
[15:12:31] [PASSED] drm_test_check_broadcast_rgb_crtc_mode_changed
[15:12:31] [PASSED] drm_test_check_broadcast_rgb_crtc_mode_not_changed
[15:12:31] [PASSED] drm_test_check_disable_connector
[15:12:31] [PASSED] drm_test_check_hdmi_funcs_reject_rate
[15:12:31] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_rgb
[15:12:31] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_yuv420
[15:12:31] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_ignore_yuv422
[15:12:31] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_ignore_yuv420
[15:12:31] [PASSED] drm_test_check_driver_unsupported_fallback_yuv420
[15:12:31] [PASSED] drm_test_check_output_bpc_crtc_mode_changed
[15:12:31] [PASSED] drm_test_check_output_bpc_crtc_mode_not_changed
[15:12:31] [PASSED] drm_test_check_output_bpc_dvi
[15:12:31] [PASSED] drm_test_check_output_bpc_format_vic_1
[15:12:31] [PASSED] drm_test_check_output_bpc_format_display_8bpc_only
[15:12:31] [PASSED] drm_test_check_output_bpc_format_display_rgb_only
[15:12:31] [PASSED] drm_test_check_output_bpc_format_driver_8bpc_only
[15:12:31] [PASSED] drm_test_check_output_bpc_format_driver_rgb_only
[15:12:31] [PASSED] drm_test_check_tmds_char_rate_rgb_8bpc
[15:12:31] [PASSED] drm_test_check_tmds_char_rate_rgb_10bpc
[15:12:31] [PASSED] drm_test_check_tmds_char_rate_rgb_12bpc
[15:12:31] ===== [PASSED] drm_atomic_helper_connector_hdmi_check ======
[15:12:31] === drm_atomic_helper_connector_hdmi_reset (6 subtests) ====
[15:12:31] [PASSED] drm_test_check_broadcast_rgb_value
[15:12:31] [PASSED] drm_test_check_bpc_8_value
[15:12:31] [PASSED] drm_test_check_bpc_10_value
[15:12:31] [PASSED] drm_test_check_bpc_12_value
[15:12:31] [PASSED] drm_test_check_format_value
[15:12:31] [PASSED] drm_test_check_tmds_char_value
[15:12:31] ===== [PASSED] drm_atomic_helper_connector_hdmi_reset ======
[15:12:31] = drm_atomic_helper_connector_hdmi_mode_valid (4 subtests) =
[15:12:31] [PASSED] drm_test_check_mode_valid
[15:12:31] [PASSED] drm_test_check_mode_valid_reject
[15:12:31] [PASSED] drm_test_check_mode_valid_reject_rate
[15:12:31] [PASSED] drm_test_check_mode_valid_reject_max_clock
[15:12:31] === [PASSED] drm_atomic_helper_connector_hdmi_mode_valid ===
[15:12:31] ================= drm_managed (2 subtests) =================
[15:12:31] [PASSED] drm_test_managed_release_action
[15:12:31] [PASSED] drm_test_managed_run_action
[15:12:31] =================== [PASSED] drm_managed ===================
[15:12:31] =================== drm_mm (6 subtests) ====================
[15:12:31] [PASSED] drm_test_mm_init
[15:12:31] [PASSED] drm_test_mm_debug
[15:12:31] [PASSED] drm_test_mm_align32
[15:12:31] [PASSED] drm_test_mm_align64
[15:12:31] [PASSED] drm_test_mm_lowest
[15:12:31] [PASSED] drm_test_mm_highest
[15:12:31] ===================== [PASSED] drm_mm ======================
[15:12:31] ============= drm_modes_analog_tv (5 subtests) =============
[15:12:31] [PASSED] drm_test_modes_analog_tv_mono_576i
[15:12:31] [PASSED] drm_test_modes_analog_tv_ntsc_480i
[15:12:31] [PASSED] drm_test_modes_analog_tv_ntsc_480i_inlined
[15:12:31] [PASSED] drm_test_modes_analog_tv_pal_576i
[15:12:31] [PASSED] drm_test_modes_analog_tv_pal_576i_inlined
[15:12:31] =============== [PASSED] drm_modes_analog_tv ===============
[15:12:31] ============== drm_plane_helper (2 subtests) ===============
[15:12:31] =============== drm_test_check_plane_state ================
[15:12:31] [PASSED] clipping_simple
[15:12:31] [PASSED] clipping_rotate_reflect
[15:12:31] [PASSED] positioning_simple
[15:12:31] [PASSED] upscaling
[15:12:31] [PASSED] downscaling
[15:12:31] [PASSED] rounding1
[15:12:31] [PASSED] rounding2
[15:12:31] [PASSED] rounding3
[15:12:31] [PASSED] rounding4
[15:12:31] =========== [PASSED] drm_test_check_plane_state ============
[15:12:31] =========== drm_test_check_invalid_plane_state ============
[15:12:31] [PASSED] positioning_invalid
[15:12:31] [PASSED] upscaling_invalid
[15:12:31] [PASSED] downscaling_invalid
[15:12:31] ======= [PASSED] drm_test_check_invalid_plane_state ========
[15:12:31] ================ [PASSED] drm_plane_helper =================
[15:12:31] ====== drm_connector_helper_tv_get_modes (1 subtest) =======
[15:12:31] ====== drm_test_connector_helper_tv_get_modes_check =======
[15:12:31] [PASSED] None
[15:12:31] [PASSED] PAL
[15:12:31] [PASSED] NTSC
[15:12:31] [PASSED] Both, NTSC Default
[15:12:31] [PASSED] Both, PAL Default
[15:12:31] [PASSED] Both, NTSC Default, with PAL on command-line
[15:12:31] [PASSED] Both, PAL Default, with NTSC on command-line
[15:12:31] == [PASSED] drm_test_connector_helper_tv_get_modes_check ===
[15:12:31] ======== [PASSED] drm_connector_helper_tv_get_modes ========
[15:12:31] ================== drm_rect (9 subtests) ===================
[15:12:31] [PASSED] drm_test_rect_clip_scaled_div_by_zero
[15:12:31] [PASSED] drm_test_rect_clip_scaled_not_clipped
[15:12:31] [PASSED] drm_test_rect_clip_scaled_clipped
[15:12:31] [PASSED] drm_test_rect_clip_scaled_signed_vs_unsigned
[15:12:31] ================= drm_test_rect_intersect =================
[15:12:31] [PASSED] top-left x bottom-right: 2x2+1+1 x 2x2+0+0
[15:12:31] [PASSED] top-right x bottom-left: 2x2+0+0 x 2x2+1-1
[15:12:31] [PASSED] bottom-left x top-right: 2x2+1-1 x 2x2+0+0
[15:12:31] [PASSED] bottom-right x top-left: 2x2+0+0 x 2x2+1+1
[15:12:31] [PASSED] right x left: 2x1+0+0 x 3x1+1+0
[15:12:31] [PASSED] left x right: 3x1+1+0 x 2x1+0+0
[15:12:31] [PASSED] up x bottom: 1x2+0+0 x 1x3+0-1
[15:12:31] [PASSED] bottom x up: 1x3+0-1 x 1x2+0+0
[15:12:31] [PASSED] touching corner: 1x1+0+0 x 2x2+1+1
[15:12:31] [PASSED] touching side: 1x1+0+0 x 1x1+1+0
[15:12:31] [PASSED] equal rects: 2x2+0+0 x 2x2+0+0
[15:12:31] [PASSED] inside another: 2x2+0+0 x 1x1+1+1
[15:12:31] [PASSED] far away: 1x1+0+0 x 1x1+3+6
[15:12:31] [PASSED] points intersecting: 0x0+5+10 x 0x0+5+10
[15:12:31] [PASSED] points not intersecting: 0x0+0+0 x 0x0+5+10
[15:12:31] ============= [PASSED] drm_test_rect_intersect =============
[15:12:31] ================ drm_test_rect_calc_hscale ================
[15:12:31] [PASSED] normal use
[15:12:31] [PASSED] out of max range
[15:12:31] [PASSED] out of min range
[15:12:31] [PASSED] zero dst
[15:12:31] [PASSED] negative src
[15:12:31] [PASSED] negative dst
[15:12:31] ============ [PASSED] drm_test_rect_calc_hscale ============
[15:12:31] ================ drm_test_rect_calc_vscale ================
[15:12:31] [PASSED] normal use
[15:12:31] [PASSED] out of max range
[15:12:31] [PASSED] out of min range
[15:12:31] [PASSED] zero dst
[15:12:31] [PASSED] negative src
stty: 'standard input': Inappropriate ioctl for device
[15:12:31] [PASSED] negative dst
[15:12:31] ============ [PASSED] drm_test_rect_calc_vscale ============
[15:12:31] ================== drm_test_rect_rotate ===================
[15:12:31] [PASSED] reflect-x
[15:12:31] [PASSED] reflect-y
[15:12:31] [PASSED] rotate-0
[15:12:31] [PASSED] rotate-90
[15:12:31] [PASSED] rotate-180
[15:12:31] [PASSED] rotate-270
[15:12:31] ============== [PASSED] drm_test_rect_rotate ===============
[15:12:31] ================ drm_test_rect_rotate_inv =================
[15:12:31] [PASSED] reflect-x
[15:12:31] [PASSED] reflect-y
[15:12:31] [PASSED] rotate-0
[15:12:31] [PASSED] rotate-90
[15:12:31] [PASSED] rotate-180
[15:12:31] [PASSED] rotate-270
[15:12:31] ============ [PASSED] drm_test_rect_rotate_inv =============
[15:12:31] ==================== [PASSED] drm_rect =====================
[15:12:31] ============ drm_sysfb_modeset_test (1 subtest) ============
[15:12:31] ============ drm_test_sysfb_build_fourcc_list =============
[15:12:31] [PASSED] no native formats
[15:12:31] [PASSED] XRGB8888 as native format
[15:12:31] [PASSED] remove duplicates
[15:12:31] [PASSED] convert alpha formats
[15:12:31] [PASSED] random formats
[15:12:31] ======== [PASSED] drm_test_sysfb_build_fourcc_list =========
[15:12:31] ============= [PASSED] drm_sysfb_modeset_test ==============
[15:12:31] ============================================================
[15:12:31] Testing complete. Ran 621 tests: passed: 621
[15:12:31] Elapsed time: 25.594s total, 1.760s configuring, 23.668s building, 0.140s running
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/ttm/tests/.kunitconfig
[15:12:31] Configuring KUnit Kernel ...
Regenerating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[15:12:33] 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
[15:12:42] Starting KUnit Kernel (1/1)...
[15:12:42] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[15:12:42] ================= ttm_device (5 subtests) ==================
[15:12:42] [PASSED] ttm_device_init_basic
[15:12:42] [PASSED] ttm_device_init_multiple
[15:12:42] [PASSED] ttm_device_fini_basic
[15:12:42] [PASSED] ttm_device_init_no_vma_man
[15:12:42] ================== ttm_device_init_pools ==================
[15:12:42] [PASSED] No DMA allocations, no DMA32 required
[15:12:42] [PASSED] DMA allocations, DMA32 required
[15:12:42] [PASSED] No DMA allocations, DMA32 required
[15:12:42] [PASSED] DMA allocations, no DMA32 required
[15:12:42] ============== [PASSED] ttm_device_init_pools ==============
[15:12:42] =================== [PASSED] ttm_device ====================
[15:12:42] ================== ttm_pool (8 subtests) ===================
[15:12:42] ================== ttm_pool_alloc_basic ===================
[15:12:42] [PASSED] One page
[15:12:42] [PASSED] More than one page
[15:12:42] [PASSED] Above the allocation limit
[15:12:42] [PASSED] One page, with coherent DMA mappings enabled
[15:12:42] [PASSED] Above the allocation limit, with coherent DMA mappings enabled
[15:12:42] ============== [PASSED] ttm_pool_alloc_basic ===============
[15:12:42] ============== ttm_pool_alloc_basic_dma_addr ==============
[15:12:42] [PASSED] One page
[15:12:42] [PASSED] More than one page
[15:12:42] [PASSED] Above the allocation limit
[15:12:42] [PASSED] One page, with coherent DMA mappings enabled
[15:12:42] [PASSED] Above the allocation limit, with coherent DMA mappings enabled
[15:12:42] ========== [PASSED] ttm_pool_alloc_basic_dma_addr ==========
[15:12:42] [PASSED] ttm_pool_alloc_order_caching_match
[15:12:42] [PASSED] ttm_pool_alloc_caching_mismatch
[15:12:42] [PASSED] ttm_pool_alloc_order_mismatch
[15:12:42] [PASSED] ttm_pool_free_dma_alloc
[15:12:42] [PASSED] ttm_pool_free_no_dma_alloc
[15:12:42] [PASSED] ttm_pool_fini_basic
[15:12:42] ==================== [PASSED] ttm_pool =====================
[15:12:42] ================ ttm_resource (8 subtests) =================
[15:12:42] ================= ttm_resource_init_basic =================
[15:12:42] [PASSED] Init resource in TTM_PL_SYSTEM
[15:12:42] [PASSED] Init resource in TTM_PL_VRAM
[15:12:42] [PASSED] Init resource in a private placement
[15:12:42] [PASSED] Init resource in TTM_PL_SYSTEM, set placement flags
[15:12:42] ============= [PASSED] ttm_resource_init_basic =============
[15:12:42] [PASSED] ttm_resource_init_pinned
[15:12:42] [PASSED] ttm_resource_fini_basic
[15:12:42] [PASSED] ttm_resource_manager_init_basic
[15:12:42] [PASSED] ttm_resource_manager_usage_basic
[15:12:42] [PASSED] ttm_resource_manager_set_used_basic
[15:12:42] [PASSED] ttm_sys_man_alloc_basic
[15:12:42] [PASSED] ttm_sys_man_free_basic
[15:12:42] ================== [PASSED] ttm_resource ===================
[15:12:42] =================== ttm_tt (15 subtests) ===================
[15:12:42] ==================== ttm_tt_init_basic ====================
[15:12:42] [PASSED] Page-aligned size
[15:12:42] [PASSED] Extra pages requested
[15:12:42] ================ [PASSED] ttm_tt_init_basic ================
[15:12:42] [PASSED] ttm_tt_init_misaligned
[15:12:42] [PASSED] ttm_tt_fini_basic
[15:12:42] [PASSED] ttm_tt_fini_sg
[15:12:42] [PASSED] ttm_tt_fini_shmem
[15:12:42] [PASSED] ttm_tt_create_basic
[15:12:42] [PASSED] ttm_tt_create_invalid_bo_type
[15:12:42] [PASSED] ttm_tt_create_ttm_exists
[15:12:42] [PASSED] ttm_tt_create_failed
[15:12:42] [PASSED] ttm_tt_destroy_basic
[15:12:42] [PASSED] ttm_tt_populate_null_ttm
[15:12:42] [PASSED] ttm_tt_populate_populated_ttm
[15:12:42] [PASSED] ttm_tt_unpopulate_basic
[15:12:42] [PASSED] ttm_tt_unpopulate_empty_ttm
[15:12:42] [PASSED] ttm_tt_swapin_basic
[15:12:42] ===================== [PASSED] ttm_tt ======================
[15:12:42] =================== ttm_bo (14 subtests) ===================
[15:12:42] =========== ttm_bo_reserve_optimistic_no_ticket ===========
[15:12:42] [PASSED] Cannot be interrupted and sleeps
[15:12:42] [PASSED] Cannot be interrupted, locks straight away
[15:12:42] [PASSED] Can be interrupted, sleeps
[15:12:42] ======= [PASSED] ttm_bo_reserve_optimistic_no_ticket =======
[15:12:42] [PASSED] ttm_bo_reserve_locked_no_sleep
[15:12:42] [PASSED] ttm_bo_reserve_no_wait_ticket
[15:12:42] [PASSED] ttm_bo_reserve_double_resv
[15:12:42] [PASSED] ttm_bo_reserve_interrupted
[15:12:42] [PASSED] ttm_bo_reserve_deadlock
[15:12:42] [PASSED] ttm_bo_unreserve_basic
[15:12:42] [PASSED] ttm_bo_unreserve_pinned
[15:12:42] [PASSED] ttm_bo_unreserve_bulk
[15:12:42] [PASSED] ttm_bo_fini_basic
[15:12:42] [PASSED] ttm_bo_fini_shared_resv
[15:12:42] [PASSED] ttm_bo_pin_basic
[15:12:42] [PASSED] ttm_bo_pin_unpin_resource
[15:12:42] [PASSED] ttm_bo_multiple_pin_one_unpin
[15:12:42] ===================== [PASSED] ttm_bo ======================
[15:12:42] ============== ttm_bo_validate (21 subtests) ===============
[15:12:42] ============== ttm_bo_init_reserved_sys_man ===============
[15:12:42] [PASSED] Buffer object for userspace
[15:12:42] [PASSED] Kernel buffer object
[15:12:42] [PASSED] Shared buffer object
[15:12:42] ========== [PASSED] ttm_bo_init_reserved_sys_man ===========
[15:12:42] ============== ttm_bo_init_reserved_mock_man ==============
[15:12:42] [PASSED] Buffer object for userspace
[15:12:42] [PASSED] Kernel buffer object
[15:12:42] [PASSED] Shared buffer object
[15:12:42] ========== [PASSED] ttm_bo_init_reserved_mock_man ==========
[15:12:42] [PASSED] ttm_bo_init_reserved_resv
[15:12:42] ================== ttm_bo_validate_basic ==================
[15:12:42] [PASSED] Buffer object for userspace
[15:12:42] [PASSED] Kernel buffer object
[15:12:42] [PASSED] Shared buffer object
[15:12:42] ============== [PASSED] ttm_bo_validate_basic ==============
[15:12:42] [PASSED] ttm_bo_validate_invalid_placement
[15:12:42] ============= ttm_bo_validate_same_placement ==============
[15:12:42] [PASSED] System manager
[15:12:42] [PASSED] VRAM manager
[15:12:42] ========= [PASSED] ttm_bo_validate_same_placement ==========
[15:12:42] [PASSED] ttm_bo_validate_failed_alloc
[15:12:42] [PASSED] ttm_bo_validate_pinned
[15:12:42] [PASSED] ttm_bo_validate_busy_placement
[15:12:42] ================ ttm_bo_validate_multihop =================
[15:12:42] [PASSED] Buffer object for userspace
[15:12:42] [PASSED] Kernel buffer object
[15:12:42] [PASSED] Shared buffer object
[15:12:42] ============ [PASSED] ttm_bo_validate_multihop =============
[15:12:42] ========== ttm_bo_validate_no_placement_signaled ==========
[15:12:42] [PASSED] Buffer object in system domain, no page vector
[15:12:42] [PASSED] Buffer object in system domain with an existing page vector
[15:12:42] ====== [PASSED] ttm_bo_validate_no_placement_signaled ======
[15:12:42] ======== ttm_bo_validate_no_placement_not_signaled ========
[15:12:42] [PASSED] Buffer object for userspace
[15:12:42] [PASSED] Kernel buffer object
[15:12:42] [PASSED] Shared buffer object
[15:12:42] ==== [PASSED] ttm_bo_validate_no_placement_not_signaled ====
[15:12:42] [PASSED] ttm_bo_validate_move_fence_signaled
[15:12:42] ========= ttm_bo_validate_move_fence_not_signaled =========
[15:12:42] [PASSED] Waits for GPU
[15:12:42] [PASSED] Tries to lock straight away
[15:12:42] ===== [PASSED] ttm_bo_validate_move_fence_not_signaled =====
[15:12:42] [PASSED] ttm_bo_validate_happy_evict
[15:12:42] [PASSED] ttm_bo_validate_all_pinned_evict
[15:12:42] [PASSED] ttm_bo_validate_allowed_only_evict
[15:12:42] [PASSED] ttm_bo_validate_deleted_evict
[15:12:42] [PASSED] ttm_bo_validate_busy_domain_evict
[15:12:42] [PASSED] ttm_bo_validate_evict_gutting
[15:12:42] [PASSED] ttm_bo_validate_recrusive_evict
stty: 'standard input': Inappropriate ioctl for device
[15:12:42] ================= [PASSED] ttm_bo_validate =================
[15:12:42] ============================================================
[15:12:42] Testing complete. Ran 101 tests: passed: 101
[15:12:42] Elapsed time: 11.038s total, 1.764s configuring, 9.058s building, 0.188s running
+ cleanup
++ stat -c %u:%g /kernel
+ chown -R 1003:1003 /kernel
^ permalink raw reply [flat|nested] 14+ messages in thread
* ✓ Xe.CI.BAT: success for drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression
2025-09-18 14:25 [PATCH v1 0/5] drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression Nitin Gote
` (5 preceding siblings ...)
2025-09-18 15:12 ` ✓ CI.KUnit: success for drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression Patchwork
@ 2025-09-18 15:57 ` Patchwork
2025-09-19 0:37 ` ✓ Xe.CI.Full: " Patchwork
7 siblings, 0 replies; 14+ messages in thread
From: Patchwork @ 2025-09-18 15:57 UTC (permalink / raw)
To: Nitin Gote; +Cc: intel-xe
[-- Attachment #1: Type: text/plain, Size: 891 bytes --]
== Series Details ==
Series: drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression
URL : https://patchwork.freedesktop.org/series/154714/
State : success
== Summary ==
CI Bug Log - changes from xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc_BAT -> xe-pw-154714v1_BAT
====================================================
Summary
-------
**SUCCESS**
No regressions found.
Participating hosts (11 -> 11)
------------------------------
No changes in participating hosts
Changes
-------
No changes found
Build changes
-------------
* Linux: xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc -> xe-pw-154714v1
IGT_8543: 8543
xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc: b63885ef3a8242ea4b307ec9427cb4255c82abfc
xe-pw-154714v1: 154714v1
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/index.html
[-- Attachment #2: Type: text/html, Size: 1439 bytes --]
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v1 4/5] drm/xe: implement VM_BIND decompression in vm_bind_ioctl
2025-09-18 14:25 ` [PATCH v1 4/5] drm/xe: implement VM_BIND decompression in vm_bind_ioctl Nitin Gote
@ 2025-09-18 16:13 ` Matthew Auld
2025-09-22 7:57 ` Gote, Nitin R
2025-09-23 17:26 ` Matthew Brost
1 sibling, 1 reply; 14+ messages in thread
From: Matthew Auld @ 2025-09-18 16:13 UTC (permalink / raw)
To: Nitin Gote, intel-xe
Cc: himal.prasad.ghimiray, matthew.brost, thomas.hellstrom
On 18/09/2025 15:25, Nitin Gote wrote:
> Implement handling of VM_BIND(..., DECOMPRESS) in xe_vm_bind_ioctl.
>
> - Validate decompression preconditions (VRAM buffer, flat CCS support,
> XE2+ hardware, and an uncompressed PAT index).
> - Mark the BO with DRM_XE_VM_BIND_FLAG_DECOMPRESS.
> - Invalidate any overlapping VMA when required.
> - Use helper xe_vm_exec_lock_vm_and_bo(..., wait_bookkeep=true)
> to atomically acquire the VM reservation and optional BO reservation and
> handle BOOKKEEP wait/retry.
> * Drop vm->lock before calling the helper.
> * Do VMA invalidation or schedule the migrate resolve while reservations
> are held.
> * Call drm_exec_fini() to release reservations, then wait on the
> migrate/decompression fence without holding reservations and
> re-acquire vm->lock to continue.
> - Defer decompression when the VM is in fault mode.
>
> This schedules an in-place GPU resolve (xe_migrate_resolve) for
> decompression and waits for completion before proceeding with the bind.
> The change centralises drm_exec + BOOKKEEP retry logic and avoids
> sleeping while holding resv locks.
>
> Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
> Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
> ---
> drivers/gpu/drm/xe/xe_vm.c | 134 +++++++++++++++++++++++++++++++++++--
> 1 file changed, 130 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
> index fae88c4c981e..cdeb7995eab1 100644
> --- a/drivers/gpu/drm/xe/xe_vm.c
> +++ b/drivers/gpu/drm/xe/xe_vm.c
> @@ -3266,7 +3266,8 @@ ALLOW_ERROR_INJECTION(vm_bind_ioctl_ops_execute, ERRNO);
> DRM_XE_VM_BIND_FLAG_NULL | \
> DRM_XE_VM_BIND_FLAG_DUMPABLE | \
> DRM_XE_VM_BIND_FLAG_CHECK_PXP | \
> - DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR)
> + DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR | \
> + DRM_XE_VM_BIND_FLAG_DECOMPRESS)
>
> #ifdef TEST_VM_OPS_ERROR
> #define SUPPORTED_FLAGS (SUPPORTED_FLAGS_STUB | FORCE_OP_ERROR)
> @@ -3324,6 +3325,7 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm,
> bool is_null = flags & DRM_XE_VM_BIND_FLAG_NULL;
> bool is_cpu_addr_mirror = flags &
> DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR;
> + bool is_decompress = flags & DRM_XE_VM_BIND_FLAG_DECOMPRESS;
> u16 pat_index = (*bind_ops)[i].pat_index;
> u16 coh_mode;
>
> @@ -3361,7 +3363,7 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm,
> (is_null || is_cpu_addr_mirror)) ||
> XE_IOCTL_DBG(xe, !obj &&
> op == DRM_XE_VM_BIND_OP_MAP &&
> - !is_null && !is_cpu_addr_mirror) ||
> + is_decompress && !is_null && !is_cpu_addr_mirror) ||
> XE_IOCTL_DBG(xe, !obj &&
> op == DRM_XE_VM_BIND_OP_UNMAP_ALL) ||
> XE_IOCTL_DBG(xe, addr &&
> @@ -3378,8 +3380,8 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm,
> op == DRM_XE_VM_BIND_OP_PREFETCH) ||
> XE_IOCTL_DBG(xe, prefetch_region &&
> op != DRM_XE_VM_BIND_OP_PREFETCH) ||
> - XE_IOCTL_DBG(xe, (prefetch_region != DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC &&
> - !(BIT(prefetch_region) & xe->info.mem_region_mask))) ||
> + XE_IOCTL_DBG(xe, (prefetch_region != DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC &&
> + !(BIT(prefetch_region) & xe->info.mem_region_mask))) ||
> XE_IOCTL_DBG(xe, obj &&
> op == DRM_XE_VM_BIND_OP_UNMAP)) {
> err = -EINVAL;
> @@ -3698,6 +3700,130 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
> u64 obj_offset = bind_ops[i].obj_offset;
> u32 prefetch_region = bind_ops[i].prefetch_mem_region_instance;
> u16 pat_index = bind_ops[i].pat_index;
> + bool do_decompress = flags & DRM_XE_VM_BIND_FLAG_DECOMPRESS;
> +
> + /* Handle decompression process */
> + if (do_decompress && bos[i]) {
> + struct xe_vma *existing_vma;
> + struct dma_fence *decomp_fence;
> + struct xe_tile *tile = xe_device_get_root_tile(xe);
> +
> + bos[i]->flags |= DRM_XE_VM_BIND_FLAG_DECOMPRESS;
> +
> + /* Validate VRAM buffer suitable for decompression or not */
> + if (!mem_type_is_vram(bos[i]->ttm.resource->mem_type)) {
I think this can crash with NPD or other nasties since ttm.resource is
dynamic state and we need to hold dma-resv lock if touching it (only
held below AFAICT). Also memory can get evicted or can be deferred until
first use, so this would return an error for those cases?
> + drm_err(&vm->xe->drm,
> + "Decompression requires VRAM buffer\n");
> + err = -EINVAL;
> + goto unwind_ops;
> + }
> +
> + /* Check hardware support for decompression */
> + if (!xe_device_has_flat_ccs(vm->xe)) {
> + drm_err(&vm->xe->drm,
> + "Decompression requires flat CCS support\n");
> + err = -EOPNOTSUPP;
> + goto unwind_ops;
> + }
> +
> + if (GRAPHICS_VER(vm->xe) < 20) {
> + drm_err(&vm->xe->drm,
> + "Decompression requires XE2+ hardware\n");
> + err = -EOPNOTSUPP;
> + goto unwind_ops;
> + }
> +
> + /* Validate that user provided an uncompressed PAT index */
> + if (pat_index == vm->xe->pat.idx[XE_CACHE_NONE_COMPRESSION]) {
XE_CACHE_NONE_COMPRESSION is one of many possible compression enabled
indexes. Why do we only check this one? Also for all of these uAPI
sanity checks I think it's best practice to use XE_IOCTL_DBG.
> + drm_err(&vm->xe->drm,
> + "Decompression requires uncompressed PAT index\n");
> + err = -EINVAL;
> + goto unwind_ops;
> + }
> +
> + /* Invalidate VMA so GPU mappings are destroyed */
> + existing_vma = xe_vm_find_overlapping_vma(vm, addr, range);
> + if (existing_vma) {
> + struct xe_bo *existing_bo = xe_vma_bo(existing_vma);
> +
> + drm_dbg(&vm->xe->drm,
> + "Found overlapping VMA - automatic invalidation will occur\n");
> +
> + /* Drop vm->lock and atomically take VM+BO resvs via drm_exec */
> + up_write(&vm->lock);
> + struct drm_exec exec;
> +
> + err = xe_vm_exec_lock_vm_and_bo(vm, existing_bo, &exec, true, true);
> + if (err) {
> + down_write(&vm->lock);
> + goto put_obj;
> + }
> +
> + /* Invalidate the VMA while reservations are held. */
> + err = xe_vm_invalidate_vma(existing_vma);
> +
> + /* release reservations (drm_exec_fini releases VM+BO resvs) */
> + drm_exec_fini(&exec);
> +
> + /* Re-acquire vm->lock */
> + down_write(&vm->lock);
> +
> + if (err)
> + goto put_obj;
> + }
> +
> + /* Handle decompression differently for fault mode */
> + if (xe_vm_in_fault_mode(vm)) {
> + drm_dbg(&vm->xe->drm,
> + "Deferring decompression for fault mode\n");
> + /* Skip immediate decompression */
> + goto unwind_ops;
> + }
> +
> + /* Drop vm->lock and atomically acquire VM+BO reservations via drm_exec.
> + * Hold the reservations while we wait for BOOKKEEP and schedule the
> + * in-place resolve. Release the reservations (drm_exec_fini) before
> + * waiting on the migrate fence so we don't hold resv locks while
> + * sleeping.
> + */
> + up_write(&vm->lock);
> + struct drm_exec exec;
> +
> + err = xe_vm_exec_lock_vm_and_bo(vm, bos[i], &exec, true, true);
> + if (err) {
> + drm_err(&vm->xe->drm,
> + "Failed to acquire reservations for decompression: %d\n",
> + err);
> + down_write(&vm->lock);
> + goto unwind_ops;
> + }
> +
> + /* With exec holding reservations, schedule the in-place decompression */
> + decomp_fence = xe_migrate_resolve(tile->migrate,
> + bos[i], bos[i],
> + bos[i]->ttm.resource,
> + bos[i]->ttm.resource,
> + false);
> +
> + /* release the reservations so scheduler can observe BO fences */
> + drm_exec_fini(&exec);
> +
> + if (IS_ERR(decomp_fence)) {
> + err = PTR_ERR(decomp_fence);
> + drm_err(&vm->xe->drm,
> + "In-place decompression failed: %d\n", err);
> + /* vm->lock must be held for unwind_ops cleanup */
> + down_write(&vm->lock);
> + goto unwind_ops;
> + }
> +
> + /* Wait for in-place decompression to complete without holding vm->lock */
> + dma_fence_wait(decomp_fence, false);
> + dma_fence_put(decomp_fence);
> +
> + /* Now re-acquire vm->lock and continue */
> + down_write(&vm->lock);
> + }
>
> ops[i] = vm_bind_ioctl_ops_create(vm, &vops, bos[i], obj_offset,
> addr, range, op, flags,
^ permalink raw reply [flat|nested] 14+ messages in thread
* ✓ Xe.CI.Full: success for drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression
2025-09-18 14:25 [PATCH v1 0/5] drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression Nitin Gote
` (6 preceding siblings ...)
2025-09-18 15:57 ` ✓ Xe.CI.BAT: " Patchwork
@ 2025-09-19 0:37 ` Patchwork
7 siblings, 0 replies; 14+ messages in thread
From: Patchwork @ 2025-09-19 0:37 UTC (permalink / raw)
To: Nitin Gote; +Cc: intel-xe
[-- Attachment #1: Type: text/plain, Size: 27759 bytes --]
== Series Details ==
Series: drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression
URL : https://patchwork.freedesktop.org/series/154714/
State : success
== Summary ==
CI Bug Log - changes from xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc_FULL -> xe-pw-154714v1_FULL
====================================================
Summary
-------
**SUCCESS**
No regressions found.
Participating hosts (4 -> 4)
------------------------------
No changes in participating hosts
Known issues
------------
Here are the changes found in xe-pw-154714v1_FULL that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@kms_ccs@ccs-on-another-bo-y-tiled-gen12-rc-ccs-cc@pipe-d-dp-4:
- shard-dg2-set2: NOTRUN -> [SKIP][1] ([Intel XE#455] / [Intel XE#787]) +17 other tests skip
[1]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-dg2-463/igt@kms_ccs@ccs-on-another-bo-y-tiled-gen12-rc-ccs-cc@pipe-d-dp-4.html
* igt@kms_ccs@crc-primary-basic-y-tiled-ccs@pipe-d-hdmi-a-2:
- shard-dg2-set2: NOTRUN -> [SKIP][2] ([Intel XE#787]) +125 other tests skip
[2]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-dg2-432/igt@kms_ccs@crc-primary-basic-y-tiled-ccs@pipe-d-hdmi-a-2.html
* igt@kms_ccs@random-ccs-data-4-tiled-dg2-mc-ccs:
- shard-dg2-set2: [PASS][3] -> [INCOMPLETE][4] ([Intel XE#1727] / [Intel XE#2705] / [Intel XE#3113] / [Intel XE#4212] / [Intel XE#4345] / [Intel XE#4522])
[3]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-dg2-434/igt@kms_ccs@random-ccs-data-4-tiled-dg2-mc-ccs.html
[4]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-dg2-433/igt@kms_ccs@random-ccs-data-4-tiled-dg2-mc-ccs.html
* igt@kms_ccs@random-ccs-data-4-tiled-dg2-mc-ccs@pipe-a-dp-4:
- shard-dg2-set2: [PASS][5] -> [INCOMPLETE][6] ([Intel XE#1727] / [Intel XE#2705] / [Intel XE#3113] / [Intel XE#4212] / [Intel XE#4522])
[5]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-dg2-434/igt@kms_ccs@random-ccs-data-4-tiled-dg2-mc-ccs@pipe-a-dp-4.html
[6]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-dg2-433/igt@kms_ccs@random-ccs-data-4-tiled-dg2-mc-ccs@pipe-a-dp-4.html
* igt@kms_ccs@random-ccs-data-4-tiled-dg2-rc-ccs-cc:
- shard-dg2-set2: [PASS][7] -> [INCOMPLETE][8] ([Intel XE#1727] / [Intel XE#3113] / [Intel XE#4212] / [Intel XE#4345] / [Intel XE#4522])
[7]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-dg2-432/igt@kms_ccs@random-ccs-data-4-tiled-dg2-rc-ccs-cc.html
[8]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-dg2-433/igt@kms_ccs@random-ccs-data-4-tiled-dg2-rc-ccs-cc.html
* igt@kms_ccs@random-ccs-data-4-tiled-dg2-rc-ccs-cc@pipe-b-dp-4:
- shard-dg2-set2: NOTRUN -> [INCOMPLETE][9] ([Intel XE#1727] / [Intel XE#3113] / [Intel XE#4212] / [Intel XE#4522])
[9]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-dg2-433/igt@kms_ccs@random-ccs-data-4-tiled-dg2-rc-ccs-cc@pipe-b-dp-4.html
* igt@kms_ccs@random-ccs-data-4-tiled-dg2-rc-ccs@pipe-a-dp-4:
- shard-dg2-set2: NOTRUN -> [INCOMPLETE][10] ([i915#14968])
[10]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-dg2-434/igt@kms_ccs@random-ccs-data-4-tiled-dg2-rc-ccs@pipe-a-dp-4.html
* igt@kms_cdclk@mode-transition@pipe-a-dp-2:
- shard-dg2-set2: NOTRUN -> [SKIP][11] ([Intel XE#4417]) +3 other tests skip
[11]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-dg2-432/igt@kms_cdclk@mode-transition@pipe-a-dp-2.html
* igt@kms_chamelium_color@ctm-blue-to-red:
- shard-adlp: NOTRUN -> [SKIP][12] ([Intel XE#306])
[12]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-adlp-3/igt@kms_chamelium_color@ctm-blue-to-red.html
* igt@kms_content_protection@legacy@pipe-a-dp-4:
- shard-dg2-set2: NOTRUN -> [FAIL][13] ([Intel XE#1178]) +1 other test fail
[13]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-dg2-463/igt@kms_content_protection@legacy@pipe-a-dp-4.html
* igt@kms_cursor_legacy@cursora-vs-flipb-legacy:
- shard-bmg: [PASS][14] -> [SKIP][15] ([Intel XE#2291]) +4 other tests skip
[14]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-3/igt@kms_cursor_legacy@cursora-vs-flipb-legacy.html
[15]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-6/igt@kms_cursor_legacy@cursora-vs-flipb-legacy.html
* igt@kms_cursor_legacy@flip-vs-cursor-atomic:
- shard-bmg: [PASS][16] -> [FAIL][17] ([Intel XE#4633])
[16]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-4/igt@kms_cursor_legacy@flip-vs-cursor-atomic.html
[17]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-1/igt@kms_cursor_legacy@flip-vs-cursor-atomic.html
* igt@kms_display_modes@extended-mode-basic:
- shard-bmg: [PASS][18] -> [SKIP][19] ([Intel XE#4302])
[18]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-3/igt@kms_display_modes@extended-mode-basic.html
[19]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-6/igt@kms_display_modes@extended-mode-basic.html
* igt@kms_dither@fb-8bpc-vs-panel-6bpc:
- shard-bmg: [PASS][20] -> [SKIP][21] ([Intel XE#1340])
[20]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-8/igt@kms_dither@fb-8bpc-vs-panel-6bpc.html
[21]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-6/igt@kms_dither@fb-8bpc-vs-panel-6bpc.html
* igt@kms_dither@fb-8bpc-vs-panel-6bpc@pipe-a-hdmi-a-2:
- shard-dg2-set2: NOTRUN -> [SKIP][22] ([Intel XE#4494])
[22]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-dg2-432/igt@kms_dither@fb-8bpc-vs-panel-6bpc@pipe-a-hdmi-a-2.html
* igt@kms_flip@2x-plain-flip-fb-recreate-interruptible:
- shard-bmg: [PASS][23] -> [SKIP][24] ([Intel XE#2316]) +5 other tests skip
[23]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-3/igt@kms_flip@2x-plain-flip-fb-recreate-interruptible.html
[24]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-6/igt@kms_flip@2x-plain-flip-fb-recreate-interruptible.html
* igt@kms_flip@basic-flip-vs-dpms:
- shard-adlp: [PASS][25] -> [DMESG-WARN][26] ([Intel XE#4543]) +4 other tests dmesg-warn
[25]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-adlp-2/igt@kms_flip@basic-flip-vs-dpms.html
[26]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-adlp-9/igt@kms_flip@basic-flip-vs-dpms.html
* igt@kms_flip@flip-vs-rmfb-interruptible:
- shard-adlp: [PASS][27] -> [DMESG-WARN][28] ([Intel XE#5208])
[27]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-adlp-2/igt@kms_flip@flip-vs-rmfb-interruptible.html
[28]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-adlp-9/igt@kms_flip@flip-vs-rmfb-interruptible.html
* igt@kms_flip@flip-vs-suspend-interruptible:
- shard-adlp: [PASS][29] -> [DMESG-WARN][30] ([Intel XE#2953] / [Intel XE#4173]) +3 other tests dmesg-warn
[29]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-adlp-1/igt@kms_flip@flip-vs-suspend-interruptible.html
[30]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-adlp-2/igt@kms_flip@flip-vs-suspend-interruptible.html
* igt@kms_flip@plain-flip-fb-recreate@a-edp1:
- shard-lnl: [PASS][31] -> [FAIL][32] ([Intel XE#5337] / [Intel XE#5408]) +1 other test fail
[31]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-lnl-4/igt@kms_flip@plain-flip-fb-recreate@a-edp1.html
[32]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-lnl-8/igt@kms_flip@plain-flip-fb-recreate@a-edp1.html
* igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-shrfb-pgflip-blt:
- shard-adlp: NOTRUN -> [SKIP][33] ([Intel XE#653])
[33]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-adlp-3/igt@kms_frontbuffer_tracking@fbcpsr-1p-primscrn-shrfb-pgflip-blt.html
* igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-spr-indfb-draw-mmap-wc:
- shard-adlp: NOTRUN -> [SKIP][34] ([Intel XE#656])
[34]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-adlp-3/igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-spr-indfb-draw-mmap-wc.html
* igt@kms_frontbuffer_tracking@psr-modesetfrombusy:
- shard-lnl: [PASS][35] -> [DMESG-WARN][36] ([Intel XE#5698])
[35]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-lnl-3/igt@kms_frontbuffer_tracking@psr-modesetfrombusy.html
[36]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-lnl-7/igt@kms_frontbuffer_tracking@psr-modesetfrombusy.html
* igt@kms_hdr@invalid-hdr:
- shard-dg2-set2: [PASS][37] -> [SKIP][38] ([Intel XE#455])
[37]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-dg2-463/igt@kms_hdr@invalid-hdr.html
[38]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-dg2-433/igt@kms_hdr@invalid-hdr.html
* igt@kms_joiner@basic-force-big-joiner:
- shard-bmg: [PASS][39] -> [SKIP][40] ([Intel XE#3012])
[39]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-7/igt@kms_joiner@basic-force-big-joiner.html
[40]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-6/igt@kms_joiner@basic-force-big-joiner.html
* igt@xe_exec_basic@multigpu-no-exec-userptr-invalidate-race:
- shard-dg2-set2: [PASS][41] -> [SKIP][42] ([Intel XE#1392]) +9 other tests skip
[41]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-dg2-435/igt@xe_exec_basic@multigpu-no-exec-userptr-invalidate-race.html
[42]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-dg2-432/igt@xe_exec_basic@multigpu-no-exec-userptr-invalidate-race.html
* igt@xe_exec_fault_mode@twice-userptr-rebind:
- shard-adlp: NOTRUN -> [SKIP][43] ([Intel XE#288] / [Intel XE#5561])
[43]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-adlp-3/igt@xe_exec_fault_mode@twice-userptr-rebind.html
* igt@xe_exec_system_allocator@many-mmap-remap-ro-dontunmap:
- shard-adlp: NOTRUN -> [SKIP][44] ([Intel XE#4915]) +5 other tests skip
[44]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-adlp-3/igt@xe_exec_system_allocator@many-mmap-remap-ro-dontunmap.html
* igt@xe_pat@pat-index-xe2:
- shard-bmg: [PASS][45] -> [FAIL][46] ([Intel XE#5507]) +1 other test fail
[45]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-1/igt@xe_pat@pat-index-xe2.html
[46]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-4/igt@xe_pat@pat-index-xe2.html
* igt@xe_pm_residency@gt-c6-freeze:
- shard-dg2-set2: [PASS][47] -> [INCOMPLETE][48] ([Intel XE#3088]) +1 other test incomplete
[47]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-dg2-434/igt@xe_pm_residency@gt-c6-freeze.html
[48]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-dg2-434/igt@xe_pm_residency@gt-c6-freeze.html
#### Possible fixes ####
* igt@kms_async_flips@alternate-sync-async-flip-atomic:
- shard-bmg: [FAIL][49] ([Intel XE#3718]) -> [PASS][50]
[49]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-3/igt@kms_async_flips@alternate-sync-async-flip-atomic.html
[50]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-5/igt@kms_async_flips@alternate-sync-async-flip-atomic.html
* igt@kms_async_flips@alternate-sync-async-flip-atomic@pipe-c-dp-2:
- shard-bmg: [FAIL][51] -> [PASS][52]
[51]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-3/igt@kms_async_flips@alternate-sync-async-flip-atomic@pipe-c-dp-2.html
[52]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-5/igt@kms_async_flips@alternate-sync-async-flip-atomic@pipe-c-dp-2.html
* igt@kms_atomic_transition@plane-toggle-modeset-transition:
- shard-adlp: [FAIL][53] ([Intel XE#3908]) -> [PASS][54] +1 other test pass
[53]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-adlp-1/igt@kms_atomic_transition@plane-toggle-modeset-transition.html
[54]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-adlp-2/igt@kms_atomic_transition@plane-toggle-modeset-transition.html
* igt@kms_bw@connected-linear-tiling-2-displays-3840x2160p:
- shard-bmg: [SKIP][55] ([Intel XE#2314] / [Intel XE#2894]) -> [PASS][56]
[55]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-6/igt@kms_bw@connected-linear-tiling-2-displays-3840x2160p.html
[56]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-1/igt@kms_bw@connected-linear-tiling-2-displays-3840x2160p.html
* igt@kms_cursor_legacy@basic-flip-after-cursor-varying-size:
- shard-adlp: [DMESG-WARN][57] ([Intel XE#2953] / [Intel XE#4173]) -> [PASS][58] +3 other tests pass
[57]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-adlp-4/igt@kms_cursor_legacy@basic-flip-after-cursor-varying-size.html
[58]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-adlp-8/igt@kms_cursor_legacy@basic-flip-after-cursor-varying-size.html
* igt@kms_cursor_legacy@cursorb-vs-flipb-varying-size:
- shard-bmg: [SKIP][59] ([Intel XE#2291]) -> [PASS][60]
[59]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-6/igt@kms_cursor_legacy@cursorb-vs-flipb-varying-size.html
[60]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-1/igt@kms_cursor_legacy@cursorb-vs-flipb-varying-size.html
* igt@kms_dp_linktrain_fallback@dp-fallback:
- shard-bmg: [SKIP][61] ([Intel XE#4294]) -> [PASS][62]
[61]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-6/igt@kms_dp_linktrain_fallback@dp-fallback.html
[62]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-1/igt@kms_dp_linktrain_fallback@dp-fallback.html
* igt@kms_flip@2x-flip-vs-dpms-on-nop:
- shard-bmg: [SKIP][63] ([Intel XE#2316]) -> [PASS][64] +3 other tests pass
[63]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-6/igt@kms_flip@2x-flip-vs-dpms-on-nop.html
[64]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-1/igt@kms_flip@2x-flip-vs-dpms-on-nop.html
* igt@kms_flip@flip-vs-expired-vblank-interruptible:
- shard-lnl: [FAIL][65] ([Intel XE#301]) -> [PASS][66] +1 other test pass
[65]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-lnl-2/igt@kms_flip@flip-vs-expired-vblank-interruptible.html
[66]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-lnl-1/igt@kms_flip@flip-vs-expired-vblank-interruptible.html
* igt@kms_flip@flip-vs-expired-vblank@c-hdmi-a1:
- shard-adlp: [DMESG-WARN][67] ([Intel XE#4543]) -> [PASS][68]
[67]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-adlp-1/igt@kms_flip@flip-vs-expired-vblank@c-hdmi-a1.html
[68]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-adlp-2/igt@kms_flip@flip-vs-expired-vblank@c-hdmi-a1.html
* igt@kms_vblank@ts-continuation-dpms-suspend:
- shard-adlp: [INCOMPLETE][69] ([Intel XE#4488]) -> [PASS][70] +1 other test pass
[69]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-adlp-8/igt@kms_vblank@ts-continuation-dpms-suspend.html
[70]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-adlp-3/igt@kms_vblank@ts-continuation-dpms-suspend.html
* igt@kms_vblank@ts-continuation-dpms-suspend@pipe-a-hdmi-a-1:
- shard-adlp: [DMESG-FAIL][71] -> [PASS][72]
[71]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-adlp-8/igt@kms_vblank@ts-continuation-dpms-suspend@pipe-a-hdmi-a-1.html
[72]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-adlp-3/igt@kms_vblank@ts-continuation-dpms-suspend@pipe-a-hdmi-a-1.html
* igt@kms_vrr@negative-basic:
- shard-bmg: [SKIP][73] ([Intel XE#1499]) -> [PASS][74]
[73]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-6/igt@kms_vrr@negative-basic.html
[74]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-1/igt@kms_vrr@negative-basic.html
* igt@xe_exec_basic@multigpu-once-rebind:
- shard-dg2-set2: [SKIP][75] ([Intel XE#1392]) -> [PASS][76] +4 other tests pass
[75]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-dg2-432/igt@xe_exec_basic@multigpu-once-rebind.html
[76]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-dg2-433/igt@xe_exec_basic@multigpu-once-rebind.html
* igt@xe_exec_reset@parallel-gt-reset:
- shard-adlp: [DMESG-WARN][77] ([Intel XE#3876]) -> [PASS][78]
[77]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-adlp-8/igt@xe_exec_reset@parallel-gt-reset.html
[78]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-adlp-3/igt@xe_exec_reset@parallel-gt-reset.html
#### Warnings ####
* igt@kms_ccs@random-ccs-data-4-tiled-dg2-rc-ccs@pipe-a-hdmi-a-6:
- shard-dg2-set2: [INCOMPLETE][79] ([Intel XE#1727] / [Intel XE#3113] / [Intel XE#4345]) -> [DMESG-WARN][80] ([Intel XE#1727] / [Intel XE#3113])
[79]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-dg2-434/igt@kms_ccs@random-ccs-data-4-tiled-dg2-rc-ccs@pipe-a-hdmi-a-6.html
[80]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-dg2-434/igt@kms_ccs@random-ccs-data-4-tiled-dg2-rc-ccs@pipe-a-hdmi-a-6.html
* igt@kms_frontbuffer_tracking@drrs-2p-scndscrn-pri-indfb-draw-mmap-wc:
- shard-bmg: [SKIP][81] ([Intel XE#2312]) -> [SKIP][82] ([Intel XE#2311]) +7 other tests skip
[81]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-6/igt@kms_frontbuffer_tracking@drrs-2p-scndscrn-pri-indfb-draw-mmap-wc.html
[82]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-1/igt@kms_frontbuffer_tracking@drrs-2p-scndscrn-pri-indfb-draw-mmap-wc.html
* igt@kms_frontbuffer_tracking@fbc-2p-primscrn-pri-indfb-draw-mmap-wc:
- shard-bmg: [SKIP][83] ([Intel XE#2312]) -> [SKIP][84] ([Intel XE#5390]) +4 other tests skip
[83]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-6/igt@kms_frontbuffer_tracking@fbc-2p-primscrn-pri-indfb-draw-mmap-wc.html
[84]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-1/igt@kms_frontbuffer_tracking@fbc-2p-primscrn-pri-indfb-draw-mmap-wc.html
* igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-indfb-plflip-blt:
- shard-bmg: [SKIP][85] ([Intel XE#5390]) -> [SKIP][86] ([Intel XE#2312]) +3 other tests skip
[85]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-8/igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-indfb-plflip-blt.html
[86]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-6/igt@kms_frontbuffer_tracking@fbc-2p-scndscrn-indfb-plflip-blt.html
* igt@kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-indfb-msflip-blt:
- shard-bmg: [SKIP][87] ([Intel XE#2311]) -> [SKIP][88] ([Intel XE#2312]) +8 other tests skip
[87]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-3/igt@kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-indfb-msflip-blt.html
[88]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-6/igt@kms_frontbuffer_tracking@fbcdrrs-2p-scndscrn-indfb-msflip-blt.html
* igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-cur-indfb-draw-blt:
- shard-bmg: [SKIP][89] ([Intel XE#2313]) -> [SKIP][90] ([Intel XE#2312]) +9 other tests skip
[89]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-3/igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-cur-indfb-draw-blt.html
[90]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-6/igt@kms_frontbuffer_tracking@fbcpsr-2p-primscrn-cur-indfb-draw-blt.html
* igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-shrfb-plflip-blt:
- shard-bmg: [SKIP][91] ([Intel XE#2312]) -> [SKIP][92] ([Intel XE#2313]) +8 other tests skip
[91]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-6/igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-shrfb-plflip-blt.html
[92]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-1/igt@kms_frontbuffer_tracking@fbcpsr-2p-scndscrn-shrfb-plflip-blt.html
* igt@kms_plane_multiple@2x-tiling-y:
- shard-bmg: [SKIP][93] ([Intel XE#4596]) -> [SKIP][94] ([Intel XE#5021])
[93]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-6/igt@kms_plane_multiple@2x-tiling-y.html
[94]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-1/igt@kms_plane_multiple@2x-tiling-y.html
* igt@kms_tiled_display@basic-test-pattern-with-chamelium:
- shard-bmg: [SKIP][95] ([Intel XE#2426]) -> [SKIP][96] ([Intel XE#2509])
[95]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-bmg-3/igt@kms_tiled_display@basic-test-pattern-with-chamelium.html
[96]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-bmg-5/igt@kms_tiled_display@basic-test-pattern-with-chamelium.html
* igt@xe_peer2peer@read:
- shard-dg2-set2: [FAIL][97] ([Intel XE#1173]) -> [SKIP][98] ([Intel XE#1061]) +1 other test skip
[97]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc/shard-dg2-464/igt@xe_peer2peer@read.html
[98]: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/shard-dg2-432/igt@xe_peer2peer@read.html
{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).
[Intel XE#1061]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1061
[Intel XE#1173]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1173
[Intel XE#1178]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1178
[Intel XE#1340]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1340
[Intel XE#1392]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1392
[Intel XE#1499]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1499
[Intel XE#1727]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/1727
[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#2426]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2426
[Intel XE#2509]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2509
[Intel XE#2705]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2705
[Intel XE#288]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/288
[Intel XE#2894]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/2894
[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#3088]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3088
[Intel XE#3113]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3113
[Intel XE#3718]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/3718
[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#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#4302]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4302
[Intel XE#4345]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4345
[Intel XE#4417]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4417
[Intel XE#4488]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4488
[Intel XE#4494]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4494
[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#4633]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4633
[Intel XE#4915]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/4915
[Intel XE#5021]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5021
[Intel XE#5208]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5208
[Intel XE#5337]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5337
[Intel XE#5390]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5390
[Intel XE#5408]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5408
[Intel XE#5507]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5507
[Intel XE#5561]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5561
[Intel XE#5624]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5624
[Intel XE#5698]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/5698
[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#787]: https://gitlab.freedesktop.org/drm/xe/kernel/issues/787
[i915#14968]: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/14968
Build changes
-------------
* Linux: xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc -> xe-pw-154714v1
IGT_8543: 8543
xe-3791-b63885ef3a8242ea4b307ec9427cb4255c82abfc: b63885ef3a8242ea4b307ec9427cb4255c82abfc
xe-pw-154714v1: 154714v1
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-154714v1/index.html
[-- Attachment #2: Type: text/html, Size: 31876 bytes --]
^ permalink raw reply [flat|nested] 14+ messages in thread
* RE: [PATCH v1 4/5] drm/xe: implement VM_BIND decompression in vm_bind_ioctl
2025-09-18 16:13 ` Matthew Auld
@ 2025-09-22 7:57 ` Gote, Nitin R
0 siblings, 0 replies; 14+ messages in thread
From: Gote, Nitin R @ 2025-09-22 7:57 UTC (permalink / raw)
To: Auld, Matthew, intel-xe@lists.freedesktop.org
Cc: Ghimiray, Himal Prasad, Brost, Matthew, Hellstrom, Thomas
Hi Matt,
> -----Original Message-----
> From: Auld, Matthew <matthew.auld@intel.com>
> Sent: Thursday, September 18, 2025 9:43 PM
> To: Gote, Nitin R <nitin.r.gote@intel.com>; intel-xe@lists.freedesktop.org
> Cc: Ghimiray, Himal Prasad <himal.prasad.ghimiray@intel.com>; Brost, Matthew
> <matthew.brost@intel.com>; Hellstrom, Thomas <thomas.hellstrom@intel.com>
> Subject: Re: [PATCH v1 4/5] drm/xe: implement VM_BIND decompression in
> vm_bind_ioctl
>
> On 18/09/2025 15:25, Nitin Gote wrote:
> > Implement handling of VM_BIND(..., DECOMPRESS) in xe_vm_bind_ioctl.
> >
> > - Validate decompression preconditions (VRAM buffer, flat CCS support,
> > XE2+ hardware, and an uncompressed PAT index).
> > - Mark the BO with DRM_XE_VM_BIND_FLAG_DECOMPRESS.
> > - Invalidate any overlapping VMA when required.
> > - Use helper xe_vm_exec_lock_vm_and_bo(..., wait_bookkeep=true)
> > to atomically acquire the VM reservation and optional BO reservation and
> > handle BOOKKEEP wait/retry.
> > * Drop vm->lock before calling the helper.
> > * Do VMA invalidation or schedule the migrate resolve while reservations
> > are held.
> > * Call drm_exec_fini() to release reservations, then wait on the
> > migrate/decompression fence without holding reservations and
> > re-acquire vm->lock to continue.
> > - Defer decompression when the VM is in fault mode.
> >
> > This schedules an in-place GPU resolve (xe_migrate_resolve) for
> > decompression and waits for completion before proceeding with the bind.
> > The change centralises drm_exec + BOOKKEEP retry logic and avoids
> > sleeping while holding resv locks.
> >
> > Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
> > Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_vm.c | 134
> +++++++++++++++++++++++++++++++++++--
> > 1 file changed, 130 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
> > index fae88c4c981e..cdeb7995eab1 100644
> > --- a/drivers/gpu/drm/xe/xe_vm.c
> > +++ b/drivers/gpu/drm/xe/xe_vm.c
> > @@ -3266,7 +3266,8 @@
> ALLOW_ERROR_INJECTION(vm_bind_ioctl_ops_execute, ERRNO);
> > DRM_XE_VM_BIND_FLAG_NULL | \
> > DRM_XE_VM_BIND_FLAG_DUMPABLE | \
> > DRM_XE_VM_BIND_FLAG_CHECK_PXP | \
> > - DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR)
> > + DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR | \
> > + DRM_XE_VM_BIND_FLAG_DECOMPRESS)
> >
> > #ifdef TEST_VM_OPS_ERROR
> > #define SUPPORTED_FLAGS (SUPPORTED_FLAGS_STUB |
> FORCE_OP_ERROR)
> > @@ -3324,6 +3325,7 @@ static int vm_bind_ioctl_check_args(struct xe_device
> *xe, struct xe_vm *vm,
> > bool is_null = flags & DRM_XE_VM_BIND_FLAG_NULL;
> > bool is_cpu_addr_mirror = flags &
> > DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR;
> > + bool is_decompress = flags &
> DRM_XE_VM_BIND_FLAG_DECOMPRESS;
> > u16 pat_index = (*bind_ops)[i].pat_index;
> > u16 coh_mode;
> >
> > @@ -3361,7 +3363,7 @@ static int vm_bind_ioctl_check_args(struct xe_device
> *xe, struct xe_vm *vm,
> > (is_null || is_cpu_addr_mirror)) ||
> > XE_IOCTL_DBG(xe, !obj &&
> > op == DRM_XE_VM_BIND_OP_MAP &&
> > - !is_null && !is_cpu_addr_mirror) ||
> > + is_decompress && !is_null &&
> !is_cpu_addr_mirror) ||
> > XE_IOCTL_DBG(xe, !obj &&
> > op == DRM_XE_VM_BIND_OP_UNMAP_ALL)
> ||
> > XE_IOCTL_DBG(xe, addr &&
> > @@ -3378,8 +3380,8 @@ static int vm_bind_ioctl_check_args(struct xe_device
> *xe, struct xe_vm *vm,
> > op == DRM_XE_VM_BIND_OP_PREFETCH) ||
> > XE_IOCTL_DBG(xe, prefetch_region &&
> > op != DRM_XE_VM_BIND_OP_PREFETCH) ||
> > - XE_IOCTL_DBG(xe, (prefetch_region !=
> DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC &&
> > - !(BIT(prefetch_region) & xe-
> >info.mem_region_mask))) ||
> > + XE_IOCTL_DBG(xe, (prefetch_region !=
> DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC &&
> > + !(BIT(prefetch_region) & xe-
> >info.mem_region_mask))) ||
> > XE_IOCTL_DBG(xe, obj &&
> > op == DRM_XE_VM_BIND_OP_UNMAP)) {
> > err = -EINVAL;
> > @@ -3698,6 +3700,130 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void
> *data, struct drm_file *file)
> > u64 obj_offset = bind_ops[i].obj_offset;
> > u32 prefetch_region =
> bind_ops[i].prefetch_mem_region_instance;
> > u16 pat_index = bind_ops[i].pat_index;
> > + bool do_decompress = flags &
> DRM_XE_VM_BIND_FLAG_DECOMPRESS;
> > +
> > + /* Handle decompression process */
> > + if (do_decompress && bos[i]) {
> > + struct xe_vma *existing_vma;
> > + struct dma_fence *decomp_fence;
> > + struct xe_tile *tile = xe_device_get_root_tile(xe);
> > +
> > + bos[i]->flags |=
> DRM_XE_VM_BIND_FLAG_DECOMPRESS;
> > +
> > + /* Validate VRAM buffer suitable for decompression or
> not */
> > + if (!mem_type_is_vram(bos[i]->ttm.resource-
> >mem_type)) {
>
> I think this can crash with NPD or other nasties since ttm.resource is dynamic
> state and we need to hold dma-resv lock if touching it (only held below AFAICT).
> Also memory can get evicted or can be deferred until first use, so this would
> return an error for those cases?
Thank you for pointing this out. I will move it under drm-resv lock, like after xe_vm_exec_lock_vm_and_bo() call.
>
> > + drm_err(&vm->xe->drm,
> > + "Decompression requires VRAM
> buffer\n");
> > + err = -EINVAL;
> > + goto unwind_ops;
> > + }
> > +
> > + /* Check hardware support for decompression */
> > + if (!xe_device_has_flat_ccs(vm->xe)) {
> > + drm_err(&vm->xe->drm,
> > + "Decompression requires flat CCS
> support\n");
> > + err = -EOPNOTSUPP;
> > + goto unwind_ops;
> > + }
> > +
> > + if (GRAPHICS_VER(vm->xe) < 20) {
> > + drm_err(&vm->xe->drm,
> > + "Decompression requires XE2+
> hardware\n");
> > + err = -EOPNOTSUPP;
> > + goto unwind_ops;
> > + }
> > +
> > + /* Validate that user provided an uncompressed PAT
> index */
> > + if (pat_index == vm->xe-
> >pat.idx[XE_CACHE_NONE_COMPRESSION]) {
>
> XE_CACHE_NONE_COMPRESSION is one of many possible compression enabled
> indexes. Why do we only check this one? Also for all of these uAPI sanity checks I
> think it's best practice to use XE_IOCTL_DBG.
>
Agree, I will check with XE_IOCTL_DBG and update in next version
Thank you,
- Nitin
> > + drm_err(&vm->xe->drm,
> > + "Decompression requires uncompressed
> PAT index\n");
> > + err = -EINVAL;
> > + goto unwind_ops;
> > + }
> > +
> > + /* Invalidate VMA so GPU mappings are destroyed */
> > + existing_vma = xe_vm_find_overlapping_vma(vm, addr,
> range);
> > + if (existing_vma) {
> > + struct xe_bo *existing_bo =
> xe_vma_bo(existing_vma);
> > +
> > + drm_dbg(&vm->xe->drm,
> > + "Found overlapping VMA - automatic
> invalidation will occur\n");
> > +
> > + /* Drop vm->lock and atomically take VM+BO
> resvs via drm_exec */
> > + up_write(&vm->lock);
> > + struct drm_exec exec;
> > +
> > + err = xe_vm_exec_lock_vm_and_bo(vm,
> existing_bo, &exec, true, true);
> > + if (err) {
> > + down_write(&vm->lock);
> > + goto put_obj;
> > + }
> > +
> > + /* Invalidate the VMA while reservations are
> held. */
> > + err = xe_vm_invalidate_vma(existing_vma);
> > +
> > + /* release reservations (drm_exec_fini releases
> VM+BO resvs) */
> > + drm_exec_fini(&exec);
> > +
> > + /* Re-acquire vm->lock */
> > + down_write(&vm->lock);
> > +
> > + if (err)
> > + goto put_obj;
> > + }
> > +
> > + /* Handle decompression differently for fault mode */
> > + if (xe_vm_in_fault_mode(vm)) {
> > + drm_dbg(&vm->xe->drm,
> > + "Deferring decompression for fault
> mode\n");
> > + /* Skip immediate decompression */
> > + goto unwind_ops;
> > + }
> > +
> > + /* Drop vm->lock and atomically acquire VM+BO
> reservations via drm_exec.
> > + * Hold the reservations while we wait for BOOKKEEP
> and schedule the
> > + * in-place resolve. Release the reservations
> (drm_exec_fini) before
> > + * waiting on the migrate fence so we don't hold resv
> locks while
> > + * sleeping.
> > + */
> > + up_write(&vm->lock);
> > + struct drm_exec exec;
> > +
> > + err = xe_vm_exec_lock_vm_and_bo(vm, bos[i], &exec,
> true, true);
> > + if (err) {
> > + drm_err(&vm->xe->drm,
> > + "Failed to acquire reservations for
> decompression: %d\n",
> > + err);
> > + down_write(&vm->lock);
> > + goto unwind_ops;
> > + }
> > +
> > + /* With exec holding reservations, schedule the in-place
> decompression */
> > + decomp_fence = xe_migrate_resolve(tile->migrate,
> > + bos[i], bos[i],
> > + bos[i]->ttm.resource,
> > + bos[i]->ttm.resource,
> > + false);
> > +
> > + /* release the reservations so scheduler can observe BO
> fences */
> > + drm_exec_fini(&exec);
> > +
> > + if (IS_ERR(decomp_fence)) {
> > + err = PTR_ERR(decomp_fence);
> > + drm_err(&vm->xe->drm,
> > + "In-place decompression failed: %d\n",
> err);
> > + /* vm->lock must be held for unwind_ops
> cleanup */
> > + down_write(&vm->lock);
> > + goto unwind_ops;
> > + }
> > +
> > + /* Wait for in-place decompression to complete without
> holding vm->lock */
> > + dma_fence_wait(decomp_fence, false);
> > + dma_fence_put(decomp_fence);
> > +
> > + /* Now re-acquire vm->lock and continue */
> > + down_write(&vm->lock);
> > + }
> >
> > ops[i] = vm_bind_ioctl_ops_create(vm, &vops, bos[i], obj_offset,
> > addr, range, op, flags,
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v1 4/5] drm/xe: implement VM_BIND decompression in vm_bind_ioctl
2025-09-18 14:25 ` [PATCH v1 4/5] drm/xe: implement VM_BIND decompression in vm_bind_ioctl Nitin Gote
2025-09-18 16:13 ` Matthew Auld
@ 2025-09-23 17:26 ` Matthew Brost
1 sibling, 0 replies; 14+ messages in thread
From: Matthew Brost @ 2025-09-23 17:26 UTC (permalink / raw)
To: Nitin Gote; +Cc: intel-xe, himal.prasad.ghimiray, thomas.hellstrom
On Thu, Sep 18, 2025 at 07:55:28PM +0530, Nitin Gote wrote:
> Implement handling of VM_BIND(..., DECOMPRESS) in xe_vm_bind_ioctl.
>
> - Validate decompression preconditions (VRAM buffer, flat CCS support,
> XE2+ hardware, and an uncompressed PAT index).
> - Mark the BO with DRM_XE_VM_BIND_FLAG_DECOMPRESS.
> - Invalidate any overlapping VMA when required.
> - Use helper xe_vm_exec_lock_vm_and_bo(..., wait_bookkeep=true)
> to atomically acquire the VM reservation and optional BO reservation and
> handle BOOKKEEP wait/retry.
> * Drop vm->lock before calling the helper.
> * Do VMA invalidation or schedule the migrate resolve while reservations
> are held.
> * Call drm_exec_fini() to release reservations, then wait on the
> migrate/decompression fence without holding reservations and
> re-acquire vm->lock to continue.
> - Defer decompression when the VM is in fault mode.
>
> This schedules an in-place GPU resolve (xe_migrate_resolve) for
> decompression and waits for completion before proceeding with the bind.
> The change centralises drm_exec + BOOKKEEP retry logic and avoids
> sleeping while holding resv locks.
>
> Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
> Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
> ---
> drivers/gpu/drm/xe/xe_vm.c | 134 +++++++++++++++++++++++++++++++++++--
> 1 file changed, 130 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
> index fae88c4c981e..cdeb7995eab1 100644
> --- a/drivers/gpu/drm/xe/xe_vm.c
> +++ b/drivers/gpu/drm/xe/xe_vm.c
> @@ -3266,7 +3266,8 @@ ALLOW_ERROR_INJECTION(vm_bind_ioctl_ops_execute, ERRNO);
> DRM_XE_VM_BIND_FLAG_NULL | \
> DRM_XE_VM_BIND_FLAG_DUMPABLE | \
> DRM_XE_VM_BIND_FLAG_CHECK_PXP | \
> - DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR)
> + DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR | \
> + DRM_XE_VM_BIND_FLAG_DECOMPRESS)
>
> #ifdef TEST_VM_OPS_ERROR
> #define SUPPORTED_FLAGS (SUPPORTED_FLAGS_STUB | FORCE_OP_ERROR)
> @@ -3324,6 +3325,7 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm,
> bool is_null = flags & DRM_XE_VM_BIND_FLAG_NULL;
> bool is_cpu_addr_mirror = flags &
> DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR;
> + bool is_decompress = flags & DRM_XE_VM_BIND_FLAG_DECOMPRESS;
> u16 pat_index = (*bind_ops)[i].pat_index;
> u16 coh_mode;
>
> @@ -3361,7 +3363,7 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm,
> (is_null || is_cpu_addr_mirror)) ||
> XE_IOCTL_DBG(xe, !obj &&
> op == DRM_XE_VM_BIND_OP_MAP &&
> - !is_null && !is_cpu_addr_mirror) ||
> + is_decompress && !is_null && !is_cpu_addr_mirror) ||
This is not the correct check. What we want is to ensure is_decompress
can only be set on DRM_XE_VM_BIND_OP_MAP ops. So the if statement to
modify is:
Old code:
3567 XE_IOCTL_DBG(xe, op != DRM_XE_VM_BIND_OP_MAP &&
3568 (is_null || is_cpu_addr_mirror)) ||
New code:
3567 XE_IOCTL_DBG(xe, op != DRM_XE_VM_BIND_OP_MAP &&
3568 (is_decompress || is_null || is_cpu_addr_mirror)) ||
> XE_IOCTL_DBG(xe, !obj &&
> op == DRM_XE_VM_BIND_OP_UNMAP_ALL) ||
> XE_IOCTL_DBG(xe, addr &&
> @@ -3378,8 +3380,8 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm,
> op == DRM_XE_VM_BIND_OP_PREFETCH) ||
> XE_IOCTL_DBG(xe, prefetch_region &&
> op != DRM_XE_VM_BIND_OP_PREFETCH) ||
> - XE_IOCTL_DBG(xe, (prefetch_region != DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC &&
> - !(BIT(prefetch_region) & xe->info.mem_region_mask))) ||
> + XE_IOCTL_DBG(xe, (prefetch_region != DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC &&
> + !(BIT(prefetch_region) & xe->info.mem_region_mask))) ||
> XE_IOCTL_DBG(xe, obj &&
> op == DRM_XE_VM_BIND_OP_UNMAP)) {
> err = -EINVAL;
> @@ -3698,6 +3700,130 @@ int xe_vm_bind_ioctl(struct drm_device *dev, void *data, struct drm_file *file)
> u64 obj_offset = bind_ops[i].obj_offset;
> u32 prefetch_region = bind_ops[i].prefetch_mem_region_instance;
> u16 pat_index = bind_ops[i].pat_index;
> + bool do_decompress = flags & DRM_XE_VM_BIND_FLAG_DECOMPRESS;
> +
> + /* Handle decompression process */
> + if (do_decompress && bos[i]) {
This is way to much code and not correct place in bind pipeline to do this.
I think a single function call in vma_lock_and_validate is issue the
decompress what we want.
I'd make this xe_bo.c layer function which calls in xe_migrate.c layer
to issue the decompress and installs returned fence into the BOs or VMs
dma-resv kernel slot. This ensures subsequent bind operations have the
correct dependencies.
> + struct xe_vma *existing_vma;
> + struct dma_fence *decomp_fence;
> + struct xe_tile *tile = xe_device_get_root_tile(xe);
> +
> + bos[i]->flags |= DRM_XE_VM_BIND_FLAG_DECOMPRESS;
> +
> + /* Validate VRAM buffer suitable for decompression or not */
> + if (!mem_type_is_vram(bos[i]->ttm.resource->mem_type)) {
> + drm_err(&vm->xe->drm,
> + "Decompression requires VRAM buffer\n");
> + err = -EINVAL;
> + goto unwind_ops;
> + }
> +
The above validation makes sense here as the BO is required.
> + /* Check hardware support for decompression */
> + if (!xe_device_has_flat_ccs(vm->xe)) {
> + drm_err(&vm->xe->drm,
> + "Decompression requires flat CCS support\n");
> + err = -EOPNOTSUPP;
> + goto unwind_ops;
> + }
> +
> + if (GRAPHICS_VER(vm->xe) < 20) {
> + drm_err(&vm->xe->drm,
> + "Decompression requires XE2+ hardware\n");
> + err = -EOPNOTSUPP;
> + goto unwind_ops;
> + }
> +
> + /* Validate that user provided an uncompressed PAT index */
> + if (pat_index == vm->xe->pat.idx[XE_CACHE_NONE_COMPRESSION]) {
> + drm_err(&vm->xe->drm,
> + "Decompression requires uncompressed PAT index\n");
> + err = -EINVAL;
> + goto unwind_ops;
> + }
The above checks likely can be moved to vm_bind_ioctl_check_args they only
rely on Xe and VM being present which vm_bind_ioctl_check_args has.
> +
> + /* Invalidate VMA so GPU mappings are destroyed */
> + existing_vma = xe_vm_find_overlapping_vma(vm, addr, range);
> + if (existing_vma) {
> + struct xe_bo *existing_bo = xe_vma_bo(existing_vma);
> +
> + drm_dbg(&vm->xe->drm,
> + "Found overlapping VMA - automatic invalidation will occur\n");
> +
> + /* Drop vm->lock and atomically take VM+BO resvs via drm_exec */
> + up_write(&vm->lock);
> + struct drm_exec exec;
> +
> + err = xe_vm_exec_lock_vm_and_bo(vm, existing_bo, &exec, true, true);
> + if (err) {
> + down_write(&vm->lock);
> + goto put_obj;
> + }
> +
> + /* Invalidate the VMA while reservations are held. */
> + err = xe_vm_invalidate_vma(existing_vma);
> +
> + /* release reservations (drm_exec_fini releases VM+BO resvs) */
> + drm_exec_fini(&exec);
> +
> + /* Re-acquire vm->lock */
> + down_write(&vm->lock);
> +
> + if (err)
> + goto put_obj;
> + }
> +
> + /* Handle decompression differently for fault mode */
> + if (xe_vm_in_fault_mode(vm)) {
> + drm_dbg(&vm->xe->drm,
> + "Deferring decompression for fault mode\n");
> + /* Skip immediate decompression */
> + goto unwind_ops;
> + }
> +
> + /* Drop vm->lock and atomically acquire VM+BO reservations via drm_exec.
> + * Hold the reservations while we wait for BOOKKEEP and schedule the
> + * in-place resolve. Release the reservations (drm_exec_fini) before
> + * waiting on the migrate fence so we don't hold resv locks while
> + * sleeping.
> + */
> + up_write(&vm->lock);
> + struct drm_exec exec;
> +
> + err = xe_vm_exec_lock_vm_and_bo(vm, bos[i], &exec, true, true);
> + if (err) {
> + drm_err(&vm->xe->drm,
> + "Failed to acquire reservations for decompression: %d\n",
> + err);
> + down_write(&vm->lock);
> + goto unwind_ops;
> + }
> +
> + /* With exec holding reservations, schedule the in-place decompression */
> + decomp_fence = xe_migrate_resolve(tile->migrate,
> + bos[i], bos[i],
> + bos[i]->ttm.resource,
> + bos[i]->ttm.resource,
> + false);
> +
> + /* release the reservations so scheduler can observe BO fences */
> + drm_exec_fini(&exec);
> +
> + if (IS_ERR(decomp_fence)) {
> + err = PTR_ERR(decomp_fence);
> + drm_err(&vm->xe->drm,
> + "In-place decompression failed: %d\n", err);
> + /* vm->lock must be held for unwind_ops cleanup */
> + down_write(&vm->lock);
> + goto unwind_ops;
> + }
> +
> + /* Wait for in-place decompression to complete without holding vm->lock */
> + dma_fence_wait(decomp_fence, false);
> + dma_fence_put(decomp_fence);
> +
> + /* Now re-acquire vm->lock and continue */
> + down_write(&vm->lock);
Everything between my last comment and above should be in moved to a new
function called in vma_lock_and_validate. Also as noted, no need to
directly wait on the fence, just install it into either VM's or BO's
dma-resv KERNEL slot - that is what xe_bo_validate does too.
Matt
> + }
>
> ops[i] = vm_bind_ioctl_ops_create(vm, &vops, bos[i], obj_offset,
> addr, range, op, flags,
> --
> 2.25.1
>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v1 5/5] drm/xe: perform on-demand decompression on VMA pagefaults
2025-09-18 14:25 ` [PATCH v1 5/5] drm/xe: perform on-demand decompression on VMA pagefaults Nitin Gote
@ 2025-09-23 17:29 ` Matthew Brost
0 siblings, 0 replies; 14+ messages in thread
From: Matthew Brost @ 2025-09-23 17:29 UTC (permalink / raw)
To: Nitin Gote; +Cc: intel-xe, himal.prasad.ghimiray, thomas.hellstrom
On Thu, Sep 18, 2025 at 07:55:29PM +0530, Nitin Gote wrote:
> When a VMA pagefault occurs and the bound BO has the DECOMPRESS flag,
> schedule a resolve (xe_migrate_resolve) to perform per-page decompression
> before binding the VMA to the faulting GT.
>
> Behavior:
> - schedule xe_migrate_resolve() while holding appropriate reservations,
> - wait on the returned fence, then clear the BO's DECOMPRESS flag
> on success,
> - proceed with the normal rebind path.
>
> This ensures on-demand fault-time decompression works consistently with
> the VM_BIND path.
>
Why special case decompress for page faults? I don't really understand
the reasoning. Beyond that, there are quite of few problematic issues
with the way this is implemented but I don't see the point in going over
until I understand why this special cased (i.e., why don't you just do
this in the bind IOCTL?).
Matt
> Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
> Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
> ---
> drivers/gpu/drm/xe/xe_gt_pagefault.c | 41 ++++++++++++++++++++++++++--
> 1 file changed, 38 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_gt_pagefault.c b/drivers/gpu/drm/xe/xe_gt_pagefault.c
> index a054d6010ae0..4a3f2682ed85 100644
> --- a/drivers/gpu/drm/xe/xe_gt_pagefault.c
> +++ b/drivers/gpu/drm/xe/xe_gt_pagefault.c
> @@ -95,11 +95,14 @@ static int handle_vma_pagefault(struct xe_gt *gt, struct xe_vma *vma,
> bool atomic)
> {
> struct xe_vm *vm = xe_vma_vm(vma);
> + struct xe_bo *bo = xe_vma_bo(vma);
> struct xe_tile *tile = gt_to_tile(gt);
> struct xe_validation_ctx ctx;
> struct drm_exec exec;
> struct dma_fence *fence;
> int err, needs_vram;
> + bool needs_decompression = false;
> + struct dma_fence *decomp_fence = NULL;
>
> lockdep_assert_held_write(&vm->lock);
>
> @@ -112,8 +115,14 @@ static int handle_vma_pagefault(struct xe_gt *gt, struct xe_vma *vma,
>
> trace_xe_vma_pagefault(vma);
>
> + /* Check if decompression is needed */
> + if (bo && (bo->flags & DRM_XE_VM_BIND_FLAG_DECOMPRESS)) {
> + needs_decompression = true;
> + drm_dbg(&vm->xe->drm, "Decompressing VMA during page fault handling\n");
> + }
> +
> /* Check if VMA is valid, opportunistic check only */
> - if (vma_is_valid(tile, vma) && !atomic)
> + if (vma_is_valid(tile, vma) && !atomic && !needs_decompression)
> return 0;
>
> retry_userptr:
> @@ -135,6 +144,21 @@ static int handle_vma_pagefault(struct xe_gt *gt, struct xe_vma *vma,
> if (err)
> goto unlock_dma_resv;
>
> + /* Perform decompression inside proper locking context */
> + if (needs_decompression) {
> + decomp_fence = xe_migrate_resolve(tile->migrate, xe_vma_bo(vma),
> + xe_vma_bo(vma),
> + xe_vma_bo(vma)->ttm.resource,
> + xe_vma_bo(vma)->ttm.resource,
> + false);
> + if (IS_ERR(decomp_fence)) {
> + drm_err(&vm->xe->drm,
> + "Decompression failed during page fault handling\n");
> + err = PTR_ERR(decomp_fence);
> + goto unlock_dma_resv;
> + }
> + }
> +
> /* Bind VMA only to the GT that has faulted */
> trace_xe_vma_pf_bind(vma);
> xe_vm_set_validation_exec(vm, &exec);
> @@ -147,8 +171,19 @@ static int handle_vma_pagefault(struct xe_gt *gt, struct xe_vma *vma,
> }
> }
>
> - dma_fence_wait(fence, false);
> - dma_fence_put(fence);
> + /* Wait for decompression first, then rebind */
> + if (decomp_fence) {
> + dma_fence_wait(decomp_fence, false);
> + dma_fence_put(decomp_fence);
> +
> + /* Clear the decompression flag after successful decompression */
> + bo->flags &= ~DRM_XE_VM_BIND_FLAG_DECOMPRESS;
> + }
> +
> + if (fence) {
> + dma_fence_wait(fence, false);
> + dma_fence_put(fence);
> + }
>
> unlock_dma_resv:
> xe_validation_ctx_fini(&ctx);
> --
> 2.25.1
>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH v1 1/5] drm/xe: add VM_BIND DECOMPRESS uapi flag
2025-09-18 14:25 ` [PATCH v1 1/5] drm/xe: add VM_BIND DECOMPRESS uapi flag Nitin Gote
@ 2025-09-23 17:35 ` Matthew Brost
0 siblings, 0 replies; 14+ messages in thread
From: Matthew Brost @ 2025-09-23 17:35 UTC (permalink / raw)
To: Nitin Gote; +Cc: intel-xe, himal.prasad.ghimiray, thomas.hellstrom
On Thu, Sep 18, 2025 at 07:55:25PM +0530, Nitin Gote wrote:
> Add DRM_XE_VM_BIND_FLAG_DECOMPRESS to the userspace API so callers
> can request GPU-side decompression on VM_BIND.
>
> This flag is used by subsequent driver changes to trigger scheduling of
> GPU work that resolves compressed VRAM pages into an uncompressed PAT
> VM mapping.
>
> Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com>
> Signed-off-by: Nitin Gote <nitin.r.gote@intel.com>
> ---
> include/uapi/drm/xe_drm.h | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h
> index 40ff19f52a8d..6b652a6968a0 100644
> --- a/include/uapi/drm/xe_drm.h
> +++ b/include/uapi/drm/xe_drm.h
> @@ -1119,6 +1119,7 @@ struct drm_xe_vm_bind_op {
> #define DRM_XE_VM_BIND_FLAG_DUMPABLE (1 << 3)
> #define DRM_XE_VM_BIND_FLAG_CHECK_PXP (1 << 4)
> #define DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR (1 << 5)
> +#define DRM_XE_VM_BIND_FLAG_DECOMPRESS (1 << 6)
You need kernel doc explaining what this does and the semantics (i.e.,
where this flag is valid).
Matt
> /** @flags: Bind flags */
> __u32 flags;
>
> --
> 2.25.1
>
^ permalink raw reply [flat|nested] 14+ messages in thread
end of thread, other threads:[~2025-09-23 17:35 UTC | newest]
Thread overview: 14+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-09-18 14:25 [PATCH v1 0/5] drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression Nitin Gote
2025-09-18 14:25 ` [PATCH v1 1/5] drm/xe: add VM_BIND DECOMPRESS uapi flag Nitin Gote
2025-09-23 17:35 ` Matthew Brost
2025-09-18 14:25 ` [PATCH v1 2/5] drm/xe: add xe_migrate_resolve wrapper and is_vram_resolve support Nitin Gote
2025-09-18 14:25 ` [PATCH v1 3/5] drm/xe: add drm_exec helper to atomically lock VM (+ optional BO) with BOOKKEEP retry Nitin Gote
2025-09-18 14:25 ` [PATCH v1 4/5] drm/xe: implement VM_BIND decompression in vm_bind_ioctl Nitin Gote
2025-09-18 16:13 ` Matthew Auld
2025-09-22 7:57 ` Gote, Nitin R
2025-09-23 17:26 ` Matthew Brost
2025-09-18 14:25 ` [PATCH v1 5/5] drm/xe: perform on-demand decompression on VMA pagefaults Nitin Gote
2025-09-23 17:29 ` Matthew Brost
2025-09-18 15:12 ` ✓ CI.KUnit: success for drm/xe: add VM_BIND DECOMPRESS support and on‑demand decompression Patchwork
2025-09-18 15:57 ` ✓ Xe.CI.BAT: " Patchwork
2025-09-19 0:37 ` ✓ Xe.CI.Full: " Patchwork
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox