* [PATCH 0/2] Fix fs_reclaim deadlock caused by CCS save/restore
@ 2026-01-29 12:51 Satyanarayana K V P
2026-01-29 12:51 ` [PATCH 1/2] drm/xe/vf: Fix fs_reclaim warning with CCS save/restore BB allocation Satyanarayana K V P
` (3 more replies)
0 siblings, 4 replies; 10+ messages in thread
From: Satyanarayana K V P @ 2026-01-29 12:51 UTC (permalink / raw)
To: intel-xe; +Cc: Satyanarayana K V P
CCS save/restore batch buffers are currently attached during BO allocation
and detached during BO teardown. The shrinker also triggers xe_bo_move(),
which is used for both allocation and deletion paths.
When BO allocation and shrinking occur concurrently, a circular locking
dependency involving fs_reclaim and swap_guard can occur, leading to a
deadlock such as:
======================================================
WARNING: possible circular locking dependency detected
------------------------------------------------------
CPU0 CPU1
---- ----
lock(fs_reclaim);
lock(&sa_manager->swap_guard);
lock(fs_reclaim);
lock(&sa_manager->swap_guard);
*** DEADLOCK ***
=====================================================
To avoid this, allocate CCS save/restore BB BOs using GFP_ATOMIC,
preventing reclaim from being invoked in this context.
Satyanarayana K V P (2):
drm/xe/vf: Fix fs_reclaim warning with CCS save/restore BB allocation
drm/xe/sa: Add lockdep annotations for SA manager swap_guard
drivers/gpu/drm/xe/xe_bb.c | 4 ++--
drivers/gpu/drm/xe/xe_sa.c | 6 ++++++
2 files changed, 8 insertions(+), 2 deletions(-)
--
2.43.0
^ permalink raw reply [flat|nested] 10+ messages in thread
* [PATCH 1/2] drm/xe/vf: Fix fs_reclaim warning with CCS save/restore BB allocation
2026-01-29 12:51 [PATCH 0/2] Fix fs_reclaim deadlock caused by CCS save/restore Satyanarayana K V P
@ 2026-01-29 12:51 ` Satyanarayana K V P
2026-01-29 15:39 ` Thomas Hellström
2026-01-29 12:51 ` [PATCH 2/2] drm/xe/sa: Add lockdep annotations for SA manager swap_guard Satyanarayana K V P
` (2 subsequent siblings)
3 siblings, 1 reply; 10+ messages in thread
From: Satyanarayana K V P @ 2026-01-29 12:51 UTC (permalink / raw)
To: intel-xe; +Cc: Satyanarayana K V P, Matthew Brost, Michal Wajdeczko,
Matthew Auld
CCS save/restore batch buffers are attached during BO allocation and
detached during BO teardown. The shrinker triggers xe_bo_move(), which is
used for both allocation and deletion paths.
When BO allocation and shrinking occur concurrently, a circular locking
dependency involving fs_reclaim and swap_guard can occur, leading to a
deadlock such as:
======================================================
WARNING: possible circular locking dependency detected
------------------------------------------------------
CPU0 CPU1
---- ----
lock(fs_reclaim);
lock(&sa_manager->swap_guard);
lock(fs_reclaim);
lock(&sa_manager->swap_guard);
*** DEADLOCK ***
=====================================================
To avoid this, allocate CCS save/restore BB BOs using GFP_ATOMIC,
preventing reclaim from being invoked in this context.
Fixes: 864690cf4dd62 ("drm/xe/vf: Attach and detach CCS copy commands with BO")
Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Suggested-by: Matthew Brost <matthew.brost@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
---
drivers/gpu/drm/xe/xe_bb.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/xe/xe_bb.c b/drivers/gpu/drm/xe/xe_bb.c
index 8b678297aaa2..355365625df9 100644
--- a/drivers/gpu/drm/xe/xe_bb.c
+++ b/drivers/gpu/drm/xe/xe_bb.c
@@ -62,7 +62,7 @@ struct xe_bb *xe_bb_new(struct xe_gt *gt, u32 dwords, bool usm)
struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt, u32 dwords,
enum xe_sriov_vf_ccs_rw_ctxs ctx_id)
{
- struct xe_bb *bb = kmalloc(sizeof(*bb), GFP_KERNEL);
+ struct xe_bb *bb = kmalloc(sizeof(*bb), GFP_ATOMIC);
struct xe_device *xe = gt_to_xe(gt);
struct xe_sa_manager *bb_pool;
int err;
@@ -78,7 +78,7 @@ struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt, u32 dwords,
*/
bb_pool = xe->sriov.vf.ccs.contexts[ctx_id].mem.ccs_bb_pool;
- bb->bo = xe_sa_bo_new(bb_pool, 4 * (dwords + 1));
+ bb->bo = __xe_sa_bo_new(bb_pool, 4 * (dwords + 1), GFP_ATOMIC);
if (IS_ERR(bb->bo)) {
err = PTR_ERR(bb->bo);
--
2.43.0
^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH 2/2] drm/xe/sa: Add lockdep annotations for SA manager swap_guard
2026-01-29 12:51 [PATCH 0/2] Fix fs_reclaim deadlock caused by CCS save/restore Satyanarayana K V P
2026-01-29 12:51 ` [PATCH 1/2] drm/xe/vf: Fix fs_reclaim warning with CCS save/restore BB allocation Satyanarayana K V P
@ 2026-01-29 12:51 ` Satyanarayana K V P
2026-01-29 13:32 ` ✓ CI.KUnit: success for Fix fs_reclaim deadlock caused by CCS save/restore Patchwork
2026-01-29 14:13 ` ✓ Xe.CI.BAT: " Patchwork
3 siblings, 0 replies; 10+ messages in thread
From: Satyanarayana K V P @ 2026-01-29 12:51 UTC (permalink / raw)
To: intel-xe; +Cc: Satyanarayana K V P, Matthew Brost, Michal Wajdeczko,
Matthew Auld
Annotate the SA manager init path to model taking swap_guard while under
reclaim context. This helps lockdep catch potential circular dependencies
between fs_reclaim and swap_guard in debug builds.
Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
Suggested-by: Matthew Brost <matthew.brost@intel.com>
Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
---
drivers/gpu/drm/xe/xe_sa.c | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/drivers/gpu/drm/xe/xe_sa.c b/drivers/gpu/drm/xe/xe_sa.c
index b738102575d4..5efbb5a09f77 100644
--- a/drivers/gpu/drm/xe/xe_sa.c
+++ b/drivers/gpu/drm/xe/xe_sa.c
@@ -89,6 +89,12 @@ struct xe_sa_manager *__xe_sa_bo_manager_init(struct xe_tile *tile, u32 size,
if (ret)
return ERR_PTR(ret);
+ if (IS_ENABLED(CONFIG_PROVE_LOCKING)) {
+ fs_reclaim_acquire(GFP_KERNEL);
+ might_lock(&sa_manager->swap_guard);
+ fs_reclaim_release(GFP_KERNEL);
+ }
+
shadow = xe_managed_bo_create_pin_map(xe, tile, size,
XE_BO_FLAG_VRAM_IF_DGFX(tile) |
XE_BO_FLAG_GGTT |
--
2.43.0
^ permalink raw reply related [flat|nested] 10+ messages in thread
* ✓ CI.KUnit: success for Fix fs_reclaim deadlock caused by CCS save/restore
2026-01-29 12:51 [PATCH 0/2] Fix fs_reclaim deadlock caused by CCS save/restore Satyanarayana K V P
2026-01-29 12:51 ` [PATCH 1/2] drm/xe/vf: Fix fs_reclaim warning with CCS save/restore BB allocation Satyanarayana K V P
2026-01-29 12:51 ` [PATCH 2/2] drm/xe/sa: Add lockdep annotations for SA manager swap_guard Satyanarayana K V P
@ 2026-01-29 13:32 ` Patchwork
2026-01-29 14:13 ` ✓ Xe.CI.BAT: " Patchwork
3 siblings, 0 replies; 10+ messages in thread
From: Patchwork @ 2026-01-29 13:32 UTC (permalink / raw)
To: Satyanarayana K V P; +Cc: intel-xe
== Series Details ==
Series: Fix fs_reclaim deadlock caused by CCS save/restore
URL : https://patchwork.freedesktop.org/series/160841/
State : success
== Summary ==
+ trap cleanup EXIT
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/xe/.kunitconfig
[13:31:02] Configuring KUnit Kernel ...
Generating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[13:31:06] 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
[13:31:38] Starting KUnit Kernel (1/1)...
[13:31:38] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[13:31:38] ================== guc_buf (11 subtests) ===================
[13:31:38] [PASSED] test_smallest
[13:31:38] [PASSED] test_largest
[13:31:38] [PASSED] test_granular
[13:31:38] [PASSED] test_unique
[13:31:38] [PASSED] test_overlap
[13:31:38] [PASSED] test_reusable
[13:31:38] [PASSED] test_too_big
[13:31:38] [PASSED] test_flush
[13:31:38] [PASSED] test_lookup
[13:31:38] [PASSED] test_data
[13:31:38] [PASSED] test_class
[13:31:38] ===================== [PASSED] guc_buf =====================
[13:31:38] =================== guc_dbm (7 subtests) ===================
[13:31:38] [PASSED] test_empty
[13:31:38] [PASSED] test_default
[13:31:38] ======================== test_size ========================
[13:31:38] [PASSED] 4
[13:31:38] [PASSED] 8
[13:31:38] [PASSED] 32
[13:31:38] [PASSED] 256
[13:31:38] ==================== [PASSED] test_size ====================
[13:31:38] ======================= test_reuse ========================
[13:31:38] [PASSED] 4
[13:31:38] [PASSED] 8
[13:31:38] [PASSED] 32
[13:31:38] [PASSED] 256
[13:31:38] =================== [PASSED] test_reuse ====================
[13:31:38] =================== test_range_overlap ====================
[13:31:38] [PASSED] 4
[13:31:38] [PASSED] 8
[13:31:38] [PASSED] 32
[13:31:38] [PASSED] 256
[13:31:38] =============== [PASSED] test_range_overlap ================
[13:31:38] =================== test_range_compact ====================
[13:31:38] [PASSED] 4
[13:31:38] [PASSED] 8
[13:31:38] [PASSED] 32
[13:31:38] [PASSED] 256
[13:31:38] =============== [PASSED] test_range_compact ================
[13:31:38] ==================== test_range_spare =====================
[13:31:38] [PASSED] 4
[13:31:38] [PASSED] 8
[13:31:38] [PASSED] 32
[13:31:38] [PASSED] 256
[13:31:38] ================ [PASSED] test_range_spare =================
[13:31:38] ===================== [PASSED] guc_dbm =====================
[13:31:38] =================== guc_idm (6 subtests) ===================
[13:31:38] [PASSED] bad_init
[13:31:38] [PASSED] no_init
[13:31:38] [PASSED] init_fini
[13:31:38] [PASSED] check_used
[13:31:38] [PASSED] check_quota
[13:31:38] [PASSED] check_all
[13:31:38] ===================== [PASSED] guc_idm =====================
[13:31:38] ================== no_relay (3 subtests) ===================
[13:31:38] [PASSED] xe_drops_guc2pf_if_not_ready
[13:31:38] [PASSED] xe_drops_guc2vf_if_not_ready
[13:31:38] [PASSED] xe_rejects_send_if_not_ready
[13:31:38] ==================== [PASSED] no_relay =====================
[13:31:38] ================== pf_relay (14 subtests) ==================
[13:31:38] [PASSED] pf_rejects_guc2pf_too_short
[13:31:38] [PASSED] pf_rejects_guc2pf_too_long
[13:31:38] [PASSED] pf_rejects_guc2pf_no_payload
[13:31:38] [PASSED] pf_fails_no_payload
[13:31:38] [PASSED] pf_fails_bad_origin
[13:31:38] [PASSED] pf_fails_bad_type
[13:31:38] [PASSED] pf_txn_reports_error
[13:31:38] [PASSED] pf_txn_sends_pf2guc
[13:31:38] [PASSED] pf_sends_pf2guc
[13:31:38] [SKIPPED] pf_loopback_nop
[13:31:38] [SKIPPED] pf_loopback_echo
[13:31:38] [SKIPPED] pf_loopback_fail
[13:31:38] [SKIPPED] pf_loopback_busy
[13:31:38] [SKIPPED] pf_loopback_retry
[13:31:38] ==================== [PASSED] pf_relay =====================
[13:31:38] ================== vf_relay (3 subtests) ===================
[13:31:38] [PASSED] vf_rejects_guc2vf_too_short
[13:31:38] [PASSED] vf_rejects_guc2vf_too_long
[13:31:38] [PASSED] vf_rejects_guc2vf_no_payload
[13:31:38] ==================== [PASSED] vf_relay =====================
[13:31:38] ================ pf_gt_config (6 subtests) =================
[13:31:38] [PASSED] fair_contexts_1vf
[13:31:38] [PASSED] fair_doorbells_1vf
[13:31:38] [PASSED] fair_ggtt_1vf
[13:31:38] ====================== fair_contexts ======================
[13:31:38] [PASSED] 1 VF
[13:31:38] [PASSED] 2 VFs
[13:31:38] [PASSED] 3 VFs
[13:31:38] [PASSED] 4 VFs
[13:31:38] [PASSED] 5 VFs
[13:31:38] [PASSED] 6 VFs
[13:31:38] [PASSED] 7 VFs
[13:31:38] [PASSED] 8 VFs
[13:31:38] [PASSED] 9 VFs
[13:31:38] [PASSED] 10 VFs
[13:31:38] [PASSED] 11 VFs
[13:31:38] [PASSED] 12 VFs
[13:31:38] [PASSED] 13 VFs
[13:31:38] [PASSED] 14 VFs
[13:31:38] [PASSED] 15 VFs
[13:31:38] [PASSED] 16 VFs
[13:31:38] [PASSED] 17 VFs
[13:31:38] [PASSED] 18 VFs
[13:31:38] [PASSED] 19 VFs
[13:31:38] [PASSED] 20 VFs
[13:31:38] [PASSED] 21 VFs
[13:31:38] [PASSED] 22 VFs
[13:31:38] [PASSED] 23 VFs
[13:31:38] [PASSED] 24 VFs
[13:31:38] [PASSED] 25 VFs
[13:31:38] [PASSED] 26 VFs
[13:31:38] [PASSED] 27 VFs
[13:31:38] [PASSED] 28 VFs
[13:31:38] [PASSED] 29 VFs
[13:31:38] [PASSED] 30 VFs
[13:31:38] [PASSED] 31 VFs
[13:31:38] [PASSED] 32 VFs
[13:31:38] [PASSED] 33 VFs
[13:31:38] [PASSED] 34 VFs
[13:31:38] [PASSED] 35 VFs
[13:31:38] [PASSED] 36 VFs
[13:31:38] [PASSED] 37 VFs
[13:31:38] [PASSED] 38 VFs
[13:31:38] [PASSED] 39 VFs
[13:31:38] [PASSED] 40 VFs
[13:31:38] [PASSED] 41 VFs
[13:31:38] [PASSED] 42 VFs
[13:31:38] [PASSED] 43 VFs
[13:31:38] [PASSED] 44 VFs
[13:31:38] [PASSED] 45 VFs
[13:31:38] [PASSED] 46 VFs
[13:31:38] [PASSED] 47 VFs
[13:31:38] [PASSED] 48 VFs
[13:31:38] [PASSED] 49 VFs
[13:31:38] [PASSED] 50 VFs
[13:31:38] [PASSED] 51 VFs
[13:31:38] [PASSED] 52 VFs
[13:31:38] [PASSED] 53 VFs
[13:31:38] [PASSED] 54 VFs
[13:31:38] [PASSED] 55 VFs
[13:31:38] [PASSED] 56 VFs
[13:31:38] [PASSED] 57 VFs
[13:31:38] [PASSED] 58 VFs
[13:31:38] [PASSED] 59 VFs
[13:31:38] [PASSED] 60 VFs
[13:31:38] [PASSED] 61 VFs
[13:31:38] [PASSED] 62 VFs
[13:31:38] [PASSED] 63 VFs
[13:31:38] ================== [PASSED] fair_contexts ==================
[13:31:38] ===================== fair_doorbells ======================
[13:31:38] [PASSED] 1 VF
[13:31:38] [PASSED] 2 VFs
[13:31:38] [PASSED] 3 VFs
[13:31:38] [PASSED] 4 VFs
[13:31:38] [PASSED] 5 VFs
[13:31:38] [PASSED] 6 VFs
[13:31:38] [PASSED] 7 VFs
[13:31:38] [PASSED] 8 VFs
[13:31:38] [PASSED] 9 VFs
[13:31:38] [PASSED] 10 VFs
[13:31:38] [PASSED] 11 VFs
[13:31:38] [PASSED] 12 VFs
[13:31:38] [PASSED] 13 VFs
[13:31:38] [PASSED] 14 VFs
[13:31:38] [PASSED] 15 VFs
[13:31:38] [PASSED] 16 VFs
[13:31:38] [PASSED] 17 VFs
[13:31:38] [PASSED] 18 VFs
[13:31:38] [PASSED] 19 VFs
[13:31:38] [PASSED] 20 VFs
[13:31:38] [PASSED] 21 VFs
[13:31:38] [PASSED] 22 VFs
[13:31:38] [PASSED] 23 VFs
[13:31:38] [PASSED] 24 VFs
[13:31:38] [PASSED] 25 VFs
[13:31:38] [PASSED] 26 VFs
[13:31:38] [PASSED] 27 VFs
[13:31:38] [PASSED] 28 VFs
[13:31:38] [PASSED] 29 VFs
[13:31:38] [PASSED] 30 VFs
[13:31:38] [PASSED] 31 VFs
[13:31:38] [PASSED] 32 VFs
[13:31:38] [PASSED] 33 VFs
[13:31:38] [PASSED] 34 VFs
[13:31:38] [PASSED] 35 VFs
[13:31:38] [PASSED] 36 VFs
[13:31:38] [PASSED] 37 VFs
[13:31:38] [PASSED] 38 VFs
[13:31:38] [PASSED] 39 VFs
[13:31:38] [PASSED] 40 VFs
[13:31:38] [PASSED] 41 VFs
[13:31:38] [PASSED] 42 VFs
[13:31:38] [PASSED] 43 VFs
[13:31:38] [PASSED] 44 VFs
[13:31:38] [PASSED] 45 VFs
[13:31:38] [PASSED] 46 VFs
[13:31:38] [PASSED] 47 VFs
[13:31:38] [PASSED] 48 VFs
[13:31:38] [PASSED] 49 VFs
[13:31:38] [PASSED] 50 VFs
[13:31:38] [PASSED] 51 VFs
[13:31:38] [PASSED] 52 VFs
[13:31:38] [PASSED] 53 VFs
[13:31:38] [PASSED] 54 VFs
[13:31:38] [PASSED] 55 VFs
[13:31:38] [PASSED] 56 VFs
[13:31:38] [PASSED] 57 VFs
[13:31:38] [PASSED] 58 VFs
[13:31:38] [PASSED] 59 VFs
[13:31:38] [PASSED] 60 VFs
[13:31:38] [PASSED] 61 VFs
[13:31:38] [PASSED] 62 VFs
[13:31:38] [PASSED] 63 VFs
[13:31:38] ================= [PASSED] fair_doorbells ==================
[13:31:38] ======================== fair_ggtt ========================
[13:31:38] [PASSED] 1 VF
[13:31:38] [PASSED] 2 VFs
[13:31:38] [PASSED] 3 VFs
[13:31:38] [PASSED] 4 VFs
[13:31:38] [PASSED] 5 VFs
[13:31:38] [PASSED] 6 VFs
[13:31:38] [PASSED] 7 VFs
[13:31:38] [PASSED] 8 VFs
[13:31:38] [PASSED] 9 VFs
[13:31:38] [PASSED] 10 VFs
[13:31:38] [PASSED] 11 VFs
[13:31:38] [PASSED] 12 VFs
[13:31:38] [PASSED] 13 VFs
[13:31:38] [PASSED] 14 VFs
[13:31:38] [PASSED] 15 VFs
[13:31:38] [PASSED] 16 VFs
[13:31:38] [PASSED] 17 VFs
[13:31:38] [PASSED] 18 VFs
[13:31:38] [PASSED] 19 VFs
[13:31:38] [PASSED] 20 VFs
[13:31:38] [PASSED] 21 VFs
[13:31:38] [PASSED] 22 VFs
[13:31:38] [PASSED] 23 VFs
[13:31:38] [PASSED] 24 VFs
[13:31:38] [PASSED] 25 VFs
[13:31:38] [PASSED] 26 VFs
[13:31:38] [PASSED] 27 VFs
[13:31:38] [PASSED] 28 VFs
[13:31:38] [PASSED] 29 VFs
[13:31:38] [PASSED] 30 VFs
[13:31:38] [PASSED] 31 VFs
[13:31:38] [PASSED] 32 VFs
[13:31:38] [PASSED] 33 VFs
[13:31:38] [PASSED] 34 VFs
[13:31:38] [PASSED] 35 VFs
[13:31:38] [PASSED] 36 VFs
[13:31:38] [PASSED] 37 VFs
[13:31:38] [PASSED] 38 VFs
[13:31:38] [PASSED] 39 VFs
[13:31:38] [PASSED] 40 VFs
[13:31:38] [PASSED] 41 VFs
[13:31:38] [PASSED] 42 VFs
[13:31:38] [PASSED] 43 VFs
[13:31:38] [PASSED] 44 VFs
[13:31:38] [PASSED] 45 VFs
[13:31:38] [PASSED] 46 VFs
[13:31:38] [PASSED] 47 VFs
[13:31:38] [PASSED] 48 VFs
[13:31:38] [PASSED] 49 VFs
[13:31:38] [PASSED] 50 VFs
[13:31:38] [PASSED] 51 VFs
[13:31:38] [PASSED] 52 VFs
[13:31:38] [PASSED] 53 VFs
[13:31:38] [PASSED] 54 VFs
[13:31:38] [PASSED] 55 VFs
[13:31:38] [PASSED] 56 VFs
[13:31:38] [PASSED] 57 VFs
[13:31:38] [PASSED] 58 VFs
[13:31:38] [PASSED] 59 VFs
[13:31:38] [PASSED] 60 VFs
[13:31:38] [PASSED] 61 VFs
[13:31:38] [PASSED] 62 VFs
[13:31:38] [PASSED] 63 VFs
[13:31:38] ==================== [PASSED] fair_ggtt ====================
[13:31:38] ================== [PASSED] pf_gt_config ===================
[13:31:38] ===================== lmtt (1 subtest) =====================
[13:31:38] ======================== test_ops =========================
[13:31:38] [PASSED] 2-level
[13:31:38] [PASSED] multi-level
[13:31:38] ==================== [PASSED] test_ops =====================
[13:31:38] ====================== [PASSED] lmtt =======================
[13:31:38] ================= pf_service (11 subtests) =================
[13:31:38] [PASSED] pf_negotiate_any
[13:31:38] [PASSED] pf_negotiate_base_match
[13:31:38] [PASSED] pf_negotiate_base_newer
[13:31:38] [PASSED] pf_negotiate_base_next
[13:31:38] [SKIPPED] pf_negotiate_base_older
[13:31:38] [PASSED] pf_negotiate_base_prev
[13:31:38] [PASSED] pf_negotiate_latest_match
[13:31:38] [PASSED] pf_negotiate_latest_newer
[13:31:38] [PASSED] pf_negotiate_latest_next
[13:31:38] [SKIPPED] pf_negotiate_latest_older
[13:31:38] [SKIPPED] pf_negotiate_latest_prev
[13:31:38] =================== [PASSED] pf_service ====================
[13:31:38] ================= xe_guc_g2g (2 subtests) ==================
[13:31:38] ============== xe_live_guc_g2g_kunit_default ==============
[13:31:38] ========= [SKIPPED] xe_live_guc_g2g_kunit_default ==========
[13:31:38] ============== xe_live_guc_g2g_kunit_allmem ===============
[13:31:38] ========== [SKIPPED] xe_live_guc_g2g_kunit_allmem ==========
[13:31:38] =================== [SKIPPED] xe_guc_g2g ===================
[13:31:38] =================== xe_mocs (2 subtests) ===================
[13:31:38] ================ xe_live_mocs_kernel_kunit ================
[13:31:38] =========== [SKIPPED] xe_live_mocs_kernel_kunit ============
[13:31:38] ================ xe_live_mocs_reset_kunit =================
[13:31:38] ============ [SKIPPED] xe_live_mocs_reset_kunit ============
[13:31:38] ==================== [SKIPPED] xe_mocs =====================
[13:31:38] ================= xe_migrate (2 subtests) ==================
[13:31:38] ================= xe_migrate_sanity_kunit =================
[13:31:38] ============ [SKIPPED] xe_migrate_sanity_kunit =============
[13:31:38] ================== xe_validate_ccs_kunit ==================
[13:31:38] ============= [SKIPPED] xe_validate_ccs_kunit ==============
[13:31:38] =================== [SKIPPED] xe_migrate ===================
[13:31:38] ================== xe_dma_buf (1 subtest) ==================
[13:31:38] ==================== xe_dma_buf_kunit =====================
[13:31:38] ================ [SKIPPED] xe_dma_buf_kunit ================
[13:31:38] =================== [SKIPPED] xe_dma_buf ===================
[13:31:38] ================= xe_bo_shrink (1 subtest) =================
[13:31:38] =================== xe_bo_shrink_kunit ====================
[13:31:38] =============== [SKIPPED] xe_bo_shrink_kunit ===============
[13:31:38] ================== [SKIPPED] xe_bo_shrink ==================
[13:31:38] ==================== xe_bo (2 subtests) ====================
[13:31:38] ================== xe_ccs_migrate_kunit ===================
[13:31:38] ============== [SKIPPED] xe_ccs_migrate_kunit ==============
[13:31:38] ==================== xe_bo_evict_kunit ====================
[13:31:38] =============== [SKIPPED] xe_bo_evict_kunit ================
[13:31:38] ===================== [SKIPPED] xe_bo ======================
[13:31:38] ==================== args (13 subtests) ====================
[13:31:38] [PASSED] count_args_test
[13:31:38] [PASSED] call_args_example
[13:31:38] [PASSED] call_args_test
[13:31:38] [PASSED] drop_first_arg_example
[13:31:38] [PASSED] drop_first_arg_test
[13:31:38] [PASSED] first_arg_example
[13:31:38] [PASSED] first_arg_test
[13:31:38] [PASSED] last_arg_example
[13:31:38] [PASSED] last_arg_test
[13:31:38] [PASSED] pick_arg_example
[13:31:38] [PASSED] if_args_example
[13:31:38] [PASSED] if_args_test
[13:31:38] [PASSED] sep_comma_example
[13:31:38] ====================== [PASSED] args =======================
[13:31:38] =================== xe_pci (3 subtests) ====================
[13:31:38] ==================== check_graphics_ip ====================
[13:31:38] [PASSED] 12.00 Xe_LP
[13:31:38] [PASSED] 12.10 Xe_LP+
[13:31:38] [PASSED] 12.55 Xe_HPG
[13:31:38] [PASSED] 12.60 Xe_HPC
[13:31:38] [PASSED] 12.70 Xe_LPG
[13:31:38] [PASSED] 12.71 Xe_LPG
[13:31:38] [PASSED] 12.74 Xe_LPG+
[13:31:38] [PASSED] 20.01 Xe2_HPG
[13:31:38] [PASSED] 20.02 Xe2_HPG
[13:31:38] [PASSED] 20.04 Xe2_LPG
[13:31:38] [PASSED] 30.00 Xe3_LPG
[13:31:38] [PASSED] 30.01 Xe3_LPG
[13:31:38] [PASSED] 30.03 Xe3_LPG
[13:31:38] [PASSED] 30.04 Xe3_LPG
[13:31:38] [PASSED] 30.05 Xe3_LPG
[13:31:38] [PASSED] 35.11 Xe3p_XPC
[13:31:38] ================ [PASSED] check_graphics_ip ================
[13:31:38] ===================== check_media_ip ======================
[13:31:38] [PASSED] 12.00 Xe_M
[13:31:38] [PASSED] 12.55 Xe_HPM
[13:31:38] [PASSED] 13.00 Xe_LPM+
[13:31:38] [PASSED] 13.01 Xe2_HPM
[13:31:38] [PASSED] 20.00 Xe2_LPM
[13:31:38] [PASSED] 30.00 Xe3_LPM
[13:31:38] [PASSED] 30.02 Xe3_LPM
[13:31:38] [PASSED] 35.00 Xe3p_LPM
[13:31:38] [PASSED] 35.03 Xe3p_HPM
[13:31:38] ================= [PASSED] check_media_ip ==================
[13:31:38] =================== check_platform_desc ===================
[13:31:38] [PASSED] 0x9A60 (TIGERLAKE)
[13:31:38] [PASSED] 0x9A68 (TIGERLAKE)
[13:31:38] [PASSED] 0x9A70 (TIGERLAKE)
[13:31:38] [PASSED] 0x9A40 (TIGERLAKE)
[13:31:38] [PASSED] 0x9A49 (TIGERLAKE)
[13:31:38] [PASSED] 0x9A59 (TIGERLAKE)
[13:31:38] [PASSED] 0x9A78 (TIGERLAKE)
[13:31:38] [PASSED] 0x9AC0 (TIGERLAKE)
[13:31:38] [PASSED] 0x9AC9 (TIGERLAKE)
[13:31:38] [PASSED] 0x9AD9 (TIGERLAKE)
[13:31:38] [PASSED] 0x9AF8 (TIGERLAKE)
[13:31:38] [PASSED] 0x4C80 (ROCKETLAKE)
[13:31:38] [PASSED] 0x4C8A (ROCKETLAKE)
[13:31:38] [PASSED] 0x4C8B (ROCKETLAKE)
[13:31:38] [PASSED] 0x4C8C (ROCKETLAKE)
[13:31:38] [PASSED] 0x4C90 (ROCKETLAKE)
[13:31:38] [PASSED] 0x4C9A (ROCKETLAKE)
[13:31:38] [PASSED] 0x4680 (ALDERLAKE_S)
[13:31:38] [PASSED] 0x4682 (ALDERLAKE_S)
[13:31:38] [PASSED] 0x4688 (ALDERLAKE_S)
[13:31:38] [PASSED] 0x468A (ALDERLAKE_S)
[13:31:38] [PASSED] 0x468B (ALDERLAKE_S)
[13:31:38] [PASSED] 0x4690 (ALDERLAKE_S)
[13:31:38] [PASSED] 0x4692 (ALDERLAKE_S)
[13:31:38] [PASSED] 0x4693 (ALDERLAKE_S)
[13:31:38] [PASSED] 0x46A0 (ALDERLAKE_P)
[13:31:38] [PASSED] 0x46A1 (ALDERLAKE_P)
[13:31:38] [PASSED] 0x46A2 (ALDERLAKE_P)
[13:31:38] [PASSED] 0x46A3 (ALDERLAKE_P)
[13:31:38] [PASSED] 0x46A6 (ALDERLAKE_P)
[13:31:38] [PASSED] 0x46A8 (ALDERLAKE_P)
[13:31:38] [PASSED] 0x46AA (ALDERLAKE_P)
[13:31:38] [PASSED] 0x462A (ALDERLAKE_P)
[13:31:38] [PASSED] 0x4626 (ALDERLAKE_P)
[13:31:38] [PASSED] 0x4628 (ALDERLAKE_P)
stty: 'standard input': Inappropriate ioctl for device
[13:31:38] [PASSED] 0x46B0 (ALDERLAKE_P)
[13:31:38] [PASSED] 0x46B1 (ALDERLAKE_P)
[13:31:38] [PASSED] 0x46B2 (ALDERLAKE_P)
[13:31:38] [PASSED] 0x46B3 (ALDERLAKE_P)
[13:31:38] [PASSED] 0x46C0 (ALDERLAKE_P)
[13:31:38] [PASSED] 0x46C1 (ALDERLAKE_P)
[13:31:38] [PASSED] 0x46C2 (ALDERLAKE_P)
[13:31:38] [PASSED] 0x46C3 (ALDERLAKE_P)
[13:31:38] [PASSED] 0x46D0 (ALDERLAKE_N)
[13:31:38] [PASSED] 0x46D1 (ALDERLAKE_N)
[13:31:38] [PASSED] 0x46D2 (ALDERLAKE_N)
[13:31:38] [PASSED] 0x46D3 (ALDERLAKE_N)
[13:31:38] [PASSED] 0x46D4 (ALDERLAKE_N)
[13:31:38] [PASSED] 0xA721 (ALDERLAKE_P)
[13:31:38] [PASSED] 0xA7A1 (ALDERLAKE_P)
[13:31:38] [PASSED] 0xA7A9 (ALDERLAKE_P)
[13:31:38] [PASSED] 0xA7AC (ALDERLAKE_P)
[13:31:38] [PASSED] 0xA7AD (ALDERLAKE_P)
[13:31:38] [PASSED] 0xA720 (ALDERLAKE_P)
[13:31:38] [PASSED] 0xA7A0 (ALDERLAKE_P)
[13:31:38] [PASSED] 0xA7A8 (ALDERLAKE_P)
[13:31:38] [PASSED] 0xA7AA (ALDERLAKE_P)
[13:31:38] [PASSED] 0xA7AB (ALDERLAKE_P)
[13:31:38] [PASSED] 0xA780 (ALDERLAKE_S)
[13:31:38] [PASSED] 0xA781 (ALDERLAKE_S)
[13:31:38] [PASSED] 0xA782 (ALDERLAKE_S)
[13:31:38] [PASSED] 0xA783 (ALDERLAKE_S)
[13:31:38] [PASSED] 0xA788 (ALDERLAKE_S)
[13:31:38] [PASSED] 0xA789 (ALDERLAKE_S)
[13:31:38] [PASSED] 0xA78A (ALDERLAKE_S)
[13:31:38] [PASSED] 0xA78B (ALDERLAKE_S)
[13:31:38] [PASSED] 0x4905 (DG1)
[13:31:38] [PASSED] 0x4906 (DG1)
[13:31:38] [PASSED] 0x4907 (DG1)
[13:31:38] [PASSED] 0x4908 (DG1)
[13:31:38] [PASSED] 0x4909 (DG1)
[13:31:38] [PASSED] 0x56C0 (DG2)
[13:31:38] [PASSED] 0x56C2 (DG2)
[13:31:38] [PASSED] 0x56C1 (DG2)
[13:31:38] [PASSED] 0x7D51 (METEORLAKE)
[13:31:38] [PASSED] 0x7DD1 (METEORLAKE)
[13:31:38] [PASSED] 0x7D41 (METEORLAKE)
[13:31:38] [PASSED] 0x7D67 (METEORLAKE)
[13:31:38] [PASSED] 0xB640 (METEORLAKE)
[13:31:38] [PASSED] 0x56A0 (DG2)
[13:31:38] [PASSED] 0x56A1 (DG2)
[13:31:38] [PASSED] 0x56A2 (DG2)
[13:31:38] [PASSED] 0x56BE (DG2)
[13:31:38] [PASSED] 0x56BF (DG2)
[13:31:38] [PASSED] 0x5690 (DG2)
[13:31:38] [PASSED] 0x5691 (DG2)
[13:31:38] [PASSED] 0x5692 (DG2)
[13:31:38] [PASSED] 0x56A5 (DG2)
[13:31:38] [PASSED] 0x56A6 (DG2)
[13:31:38] [PASSED] 0x56B0 (DG2)
[13:31:38] [PASSED] 0x56B1 (DG2)
[13:31:38] [PASSED] 0x56BA (DG2)
[13:31:38] [PASSED] 0x56BB (DG2)
[13:31:38] [PASSED] 0x56BC (DG2)
[13:31:38] [PASSED] 0x56BD (DG2)
[13:31:38] [PASSED] 0x5693 (DG2)
[13:31:38] [PASSED] 0x5694 (DG2)
[13:31:38] [PASSED] 0x5695 (DG2)
[13:31:38] [PASSED] 0x56A3 (DG2)
[13:31:38] [PASSED] 0x56A4 (DG2)
[13:31:38] [PASSED] 0x56B2 (DG2)
[13:31:38] [PASSED] 0x56B3 (DG2)
[13:31:38] [PASSED] 0x5696 (DG2)
[13:31:38] [PASSED] 0x5697 (DG2)
[13:31:38] [PASSED] 0xB69 (PVC)
[13:31:38] [PASSED] 0xB6E (PVC)
[13:31:38] [PASSED] 0xBD4 (PVC)
[13:31:38] [PASSED] 0xBD5 (PVC)
[13:31:38] [PASSED] 0xBD6 (PVC)
[13:31:38] [PASSED] 0xBD7 (PVC)
[13:31:38] [PASSED] 0xBD8 (PVC)
[13:31:38] [PASSED] 0xBD9 (PVC)
[13:31:38] [PASSED] 0xBDA (PVC)
[13:31:38] [PASSED] 0xBDB (PVC)
[13:31:38] [PASSED] 0xBE0 (PVC)
[13:31:38] [PASSED] 0xBE1 (PVC)
[13:31:38] [PASSED] 0xBE5 (PVC)
[13:31:38] [PASSED] 0x7D40 (METEORLAKE)
[13:31:38] [PASSED] 0x7D45 (METEORLAKE)
[13:31:38] [PASSED] 0x7D55 (METEORLAKE)
[13:31:38] [PASSED] 0x7D60 (METEORLAKE)
[13:31:38] [PASSED] 0x7DD5 (METEORLAKE)
[13:31:38] [PASSED] 0x6420 (LUNARLAKE)
[13:31:38] [PASSED] 0x64A0 (LUNARLAKE)
[13:31:38] [PASSED] 0x64B0 (LUNARLAKE)
[13:31:38] [PASSED] 0xE202 (BATTLEMAGE)
[13:31:38] [PASSED] 0xE209 (BATTLEMAGE)
[13:31:38] [PASSED] 0xE20B (BATTLEMAGE)
[13:31:38] [PASSED] 0xE20C (BATTLEMAGE)
[13:31:38] [PASSED] 0xE20D (BATTLEMAGE)
[13:31:38] [PASSED] 0xE210 (BATTLEMAGE)
[13:31:38] [PASSED] 0xE211 (BATTLEMAGE)
[13:31:38] [PASSED] 0xE212 (BATTLEMAGE)
[13:31:38] [PASSED] 0xE216 (BATTLEMAGE)
[13:31:38] [PASSED] 0xE220 (BATTLEMAGE)
[13:31:38] [PASSED] 0xE221 (BATTLEMAGE)
[13:31:38] [PASSED] 0xE222 (BATTLEMAGE)
[13:31:38] [PASSED] 0xE223 (BATTLEMAGE)
[13:31:38] [PASSED] 0xB080 (PANTHERLAKE)
[13:31:38] [PASSED] 0xB081 (PANTHERLAKE)
[13:31:38] [PASSED] 0xB082 (PANTHERLAKE)
[13:31:38] [PASSED] 0xB083 (PANTHERLAKE)
[13:31:38] [PASSED] 0xB084 (PANTHERLAKE)
[13:31:38] [PASSED] 0xB085 (PANTHERLAKE)
[13:31:38] [PASSED] 0xB086 (PANTHERLAKE)
[13:31:38] [PASSED] 0xB087 (PANTHERLAKE)
[13:31:38] [PASSED] 0xB08F (PANTHERLAKE)
[13:31:38] [PASSED] 0xB090 (PANTHERLAKE)
[13:31:38] [PASSED] 0xB0A0 (PANTHERLAKE)
[13:31:38] [PASSED] 0xB0B0 (PANTHERLAKE)
[13:31:38] [PASSED] 0xFD80 (PANTHERLAKE)
[13:31:38] [PASSED] 0xFD81 (PANTHERLAKE)
[13:31:38] [PASSED] 0xD740 (NOVALAKE_S)
[13:31:38] [PASSED] 0xD741 (NOVALAKE_S)
[13:31:38] [PASSED] 0xD742 (NOVALAKE_S)
[13:31:38] [PASSED] 0xD743 (NOVALAKE_S)
[13:31:38] [PASSED] 0xD744 (NOVALAKE_S)
[13:31:38] [PASSED] 0xD745 (NOVALAKE_S)
[13:31:38] [PASSED] 0x674C (CRESCENTISLAND)
[13:31:38] =============== [PASSED] check_platform_desc ===============
[13:31:38] ===================== [PASSED] xe_pci ======================
[13:31:38] =================== xe_rtp (2 subtests) ====================
[13:31:38] =============== xe_rtp_process_to_sr_tests ================
[13:31:38] [PASSED] coalesce-same-reg
[13:31:38] [PASSED] no-match-no-add
[13:31:38] [PASSED] match-or
[13:31:38] [PASSED] match-or-xfail
[13:31:38] [PASSED] no-match-no-add-multiple-rules
[13:31:38] [PASSED] two-regs-two-entries
[13:31:38] [PASSED] clr-one-set-other
[13:31:38] [PASSED] set-field
[13:31:38] [PASSED] conflict-duplicate
[13:31:38] [PASSED] conflict-not-disjoint
[13:31:38] [PASSED] conflict-reg-type
[13:31:38] =========== [PASSED] xe_rtp_process_to_sr_tests ============
[13:31:38] ================== xe_rtp_process_tests ===================
[13:31:38] [PASSED] active1
[13:31:38] [PASSED] active2
[13:31:38] [PASSED] active-inactive
[13:31:38] [PASSED] inactive-active
[13:31:38] [PASSED] inactive-1st_or_active-inactive
[13:31:38] [PASSED] inactive-2nd_or_active-inactive
[13:31:38] [PASSED] inactive-last_or_active-inactive
[13:31:38] [PASSED] inactive-no_or_active-inactive
[13:31:38] ============== [PASSED] xe_rtp_process_tests ===============
[13:31:38] ===================== [PASSED] xe_rtp ======================
[13:31:38] ==================== xe_wa (1 subtest) =====================
[13:31:38] ======================== xe_wa_gt =========================
[13:31:38] [PASSED] TIGERLAKE B0
[13:31:38] [PASSED] DG1 A0
[13:31:38] [PASSED] DG1 B0
[13:31:38] [PASSED] ALDERLAKE_S A0
[13:31:38] [PASSED] ALDERLAKE_S B0
[13:31:38] [PASSED] ALDERLAKE_S C0
[13:31:38] [PASSED] ALDERLAKE_S D0
[13:31:38] [PASSED] ALDERLAKE_P A0
[13:31:38] [PASSED] ALDERLAKE_P B0
[13:31:38] [PASSED] ALDERLAKE_P C0
[13:31:38] [PASSED] ALDERLAKE_S RPLS D0
[13:31:38] [PASSED] ALDERLAKE_P RPLU E0
[13:31:38] [PASSED] DG2 G10 C0
[13:31:38] [PASSED] DG2 G11 B1
[13:31:38] [PASSED] DG2 G12 A1
[13:31:38] [PASSED] METEORLAKE 12.70(Xe_LPG) A0 13.00(Xe_LPM+) A0
[13:31:38] [PASSED] METEORLAKE 12.71(Xe_LPG) A0 13.00(Xe_LPM+) A0
[13:31:38] [PASSED] METEORLAKE 12.74(Xe_LPG+) A0 13.00(Xe_LPM+) A0
[13:31:38] [PASSED] LUNARLAKE 20.04(Xe2_LPG) A0 20.00(Xe2_LPM) A0
[13:31:38] [PASSED] LUNARLAKE 20.04(Xe2_LPG) B0 20.00(Xe2_LPM) A0
[13:31:38] [PASSED] BATTLEMAGE 20.01(Xe2_HPG) A0 13.01(Xe2_HPM) A1
[13:31:38] [PASSED] PANTHERLAKE 30.00(Xe3_LPG) A0 30.00(Xe3_LPM) A0
[13:31:38] ==================== [PASSED] xe_wa_gt =====================
[13:31:38] ====================== [PASSED] xe_wa ======================
[13:31:38] ============================================================
[13:31:38] Testing complete. Ran 512 tests: passed: 494, skipped: 18
[13:31:38] Elapsed time: 36.278s total, 4.207s configuring, 31.554s building, 0.470s running
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/tests/.kunitconfig
[13:31:38] Configuring KUnit Kernel ...
Regenerating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[13:31:40] 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
[13:32:06] Starting KUnit Kernel (1/1)...
[13:32:06] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[13:32:06] ============ drm_test_pick_cmdline (2 subtests) ============
[13:32:06] [PASSED] drm_test_pick_cmdline_res_1920_1080_60
[13:32:06] =============== drm_test_pick_cmdline_named ===============
[13:32:06] [PASSED] NTSC
[13:32:06] [PASSED] NTSC-J
[13:32:06] [PASSED] PAL
[13:32:06] [PASSED] PAL-M
[13:32:06] =========== [PASSED] drm_test_pick_cmdline_named ===========
[13:32:06] ============== [PASSED] drm_test_pick_cmdline ==============
[13:32:06] == drm_test_atomic_get_connector_for_encoder (1 subtest) ===
[13:32:06] [PASSED] drm_test_drm_atomic_get_connector_for_encoder
[13:32:06] ==== [PASSED] drm_test_atomic_get_connector_for_encoder ====
[13:32:06] =========== drm_validate_clone_mode (2 subtests) ===========
[13:32:06] ============== drm_test_check_in_clone_mode ===============
[13:32:06] [PASSED] in_clone_mode
[13:32:06] [PASSED] not_in_clone_mode
[13:32:06] ========== [PASSED] drm_test_check_in_clone_mode ===========
[13:32:06] =============== drm_test_check_valid_clones ===============
[13:32:06] [PASSED] not_in_clone_mode
[13:32:06] [PASSED] valid_clone
[13:32:06] [PASSED] invalid_clone
[13:32:06] =========== [PASSED] drm_test_check_valid_clones ===========
[13:32:06] ============= [PASSED] drm_validate_clone_mode =============
[13:32:06] ============= drm_validate_modeset (1 subtest) =============
[13:32:06] [PASSED] drm_test_check_connector_changed_modeset
[13:32:06] ============== [PASSED] drm_validate_modeset ===============
[13:32:06] ====== drm_test_bridge_get_current_state (2 subtests) ======
[13:32:06] [PASSED] drm_test_drm_bridge_get_current_state_atomic
[13:32:06] [PASSED] drm_test_drm_bridge_get_current_state_legacy
[13:32:06] ======== [PASSED] drm_test_bridge_get_current_state ========
[13:32:06] ====== drm_test_bridge_helper_reset_crtc (3 subtests) ======
[13:32:06] [PASSED] drm_test_drm_bridge_helper_reset_crtc_atomic
[13:32:06] [PASSED] drm_test_drm_bridge_helper_reset_crtc_atomic_disabled
[13:32:06] [PASSED] drm_test_drm_bridge_helper_reset_crtc_legacy
[13:32:06] ======== [PASSED] drm_test_bridge_helper_reset_crtc ========
[13:32:06] ============== drm_bridge_alloc (2 subtests) ===============
[13:32:06] [PASSED] drm_test_drm_bridge_alloc_basic
[13:32:06] [PASSED] drm_test_drm_bridge_alloc_get_put
[13:32:06] ================ [PASSED] drm_bridge_alloc =================
[13:32:06] ================== drm_buddy (9 subtests) ==================
[13:32:06] [PASSED] drm_test_buddy_alloc_limit
[13:32:06] [PASSED] drm_test_buddy_alloc_optimistic
[13:32:06] [PASSED] drm_test_buddy_alloc_pessimistic
[13:32:06] [PASSED] drm_test_buddy_alloc_pathological
[13:32:06] [PASSED] drm_test_buddy_alloc_contiguous
[13:32:06] [PASSED] drm_test_buddy_alloc_clear
[13:32:06] [PASSED] drm_test_buddy_alloc_range_bias
[13:32:06] [PASSED] drm_test_buddy_fragmentation_performance
[13:32:06] [PASSED] drm_test_buddy_alloc_exceeds_max_order
[13:32:06] ==================== [PASSED] drm_buddy ====================
[13:32:06] ============= drm_cmdline_parser (40 subtests) =============
[13:32:06] [PASSED] drm_test_cmdline_force_d_only
[13:32:06] [PASSED] drm_test_cmdline_force_D_only_dvi
[13:32:06] [PASSED] drm_test_cmdline_force_D_only_hdmi
[13:32:06] [PASSED] drm_test_cmdline_force_D_only_not_digital
[13:32:06] [PASSED] drm_test_cmdline_force_e_only
[13:32:06] [PASSED] drm_test_cmdline_res
[13:32:06] [PASSED] drm_test_cmdline_res_vesa
[13:32:06] [PASSED] drm_test_cmdline_res_vesa_rblank
[13:32:06] [PASSED] drm_test_cmdline_res_rblank
[13:32:06] [PASSED] drm_test_cmdline_res_bpp
[13:32:06] [PASSED] drm_test_cmdline_res_refresh
[13:32:06] [PASSED] drm_test_cmdline_res_bpp_refresh
[13:32:06] [PASSED] drm_test_cmdline_res_bpp_refresh_interlaced
[13:32:06] [PASSED] drm_test_cmdline_res_bpp_refresh_margins
[13:32:06] [PASSED] drm_test_cmdline_res_bpp_refresh_force_off
[13:32:06] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on
[13:32:06] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on_analog
[13:32:06] [PASSED] drm_test_cmdline_res_bpp_refresh_force_on_digital
[13:32:06] [PASSED] drm_test_cmdline_res_bpp_refresh_interlaced_margins_force_on
[13:32:06] [PASSED] drm_test_cmdline_res_margins_force_on
[13:32:06] [PASSED] drm_test_cmdline_res_vesa_margins
[13:32:06] [PASSED] drm_test_cmdline_name
[13:32:06] [PASSED] drm_test_cmdline_name_bpp
[13:32:06] [PASSED] drm_test_cmdline_name_option
[13:32:06] [PASSED] drm_test_cmdline_name_bpp_option
[13:32:06] [PASSED] drm_test_cmdline_rotate_0
[13:32:06] [PASSED] drm_test_cmdline_rotate_90
[13:32:06] [PASSED] drm_test_cmdline_rotate_180
[13:32:06] [PASSED] drm_test_cmdline_rotate_270
[13:32:06] [PASSED] drm_test_cmdline_hmirror
[13:32:06] [PASSED] drm_test_cmdline_vmirror
[13:32:06] [PASSED] drm_test_cmdline_margin_options
[13:32:06] [PASSED] drm_test_cmdline_multiple_options
[13:32:06] [PASSED] drm_test_cmdline_bpp_extra_and_option
[13:32:06] [PASSED] drm_test_cmdline_extra_and_option
[13:32:06] [PASSED] drm_test_cmdline_freestanding_options
[13:32:06] [PASSED] drm_test_cmdline_freestanding_force_e_and_options
[13:32:06] [PASSED] drm_test_cmdline_panel_orientation
[13:32:06] ================ drm_test_cmdline_invalid =================
[13:32:06] [PASSED] margin_only
[13:32:06] [PASSED] interlace_only
[13:32:06] [PASSED] res_missing_x
[13:32:06] [PASSED] res_missing_y
[13:32:06] [PASSED] res_bad_y
[13:32:06] [PASSED] res_missing_y_bpp
[13:32:06] [PASSED] res_bad_bpp
[13:32:06] [PASSED] res_bad_refresh
[13:32:06] [PASSED] res_bpp_refresh_force_on_off
[13:32:06] [PASSED] res_invalid_mode
[13:32:06] [PASSED] res_bpp_wrong_place_mode
[13:32:06] [PASSED] name_bpp_refresh
[13:32:06] [PASSED] name_refresh
[13:32:06] [PASSED] name_refresh_wrong_mode
[13:32:06] [PASSED] name_refresh_invalid_mode
[13:32:06] [PASSED] rotate_multiple
[13:32:06] [PASSED] rotate_invalid_val
[13:32:06] [PASSED] rotate_truncated
[13:32:06] [PASSED] invalid_option
[13:32:06] [PASSED] invalid_tv_option
[13:32:06] [PASSED] truncated_tv_option
[13:32:06] ============ [PASSED] drm_test_cmdline_invalid =============
[13:32:06] =============== drm_test_cmdline_tv_options ===============
[13:32:06] [PASSED] NTSC
[13:32:06] [PASSED] NTSC_443
[13:32:06] [PASSED] NTSC_J
[13:32:06] [PASSED] PAL
[13:32:06] [PASSED] PAL_M
[13:32:06] [PASSED] PAL_N
[13:32:06] [PASSED] SECAM
[13:32:06] [PASSED] MONO_525
[13:32:06] [PASSED] MONO_625
[13:32:06] =========== [PASSED] drm_test_cmdline_tv_options ===========
[13:32:06] =============== [PASSED] drm_cmdline_parser ================
[13:32:06] ========== drmm_connector_hdmi_init (20 subtests) ==========
[13:32:06] [PASSED] drm_test_connector_hdmi_init_valid
[13:32:06] [PASSED] drm_test_connector_hdmi_init_bpc_8
[13:32:06] [PASSED] drm_test_connector_hdmi_init_bpc_10
[13:32:06] [PASSED] drm_test_connector_hdmi_init_bpc_12
[13:32:06] [PASSED] drm_test_connector_hdmi_init_bpc_invalid
[13:32:06] [PASSED] drm_test_connector_hdmi_init_bpc_null
[13:32:06] [PASSED] drm_test_connector_hdmi_init_formats_empty
[13:32:06] [PASSED] drm_test_connector_hdmi_init_formats_no_rgb
[13:32:06] === drm_test_connector_hdmi_init_formats_yuv420_allowed ===
[13:32:06] [PASSED] supported_formats=0x9 yuv420_allowed=1
[13:32:06] [PASSED] supported_formats=0x9 yuv420_allowed=0
[13:32:06] [PASSED] supported_formats=0x3 yuv420_allowed=1
[13:32:06] [PASSED] supported_formats=0x3 yuv420_allowed=0
[13:32:06] === [PASSED] drm_test_connector_hdmi_init_formats_yuv420_allowed ===
[13:32:06] [PASSED] drm_test_connector_hdmi_init_null_ddc
[13:32:06] [PASSED] drm_test_connector_hdmi_init_null_product
[13:32:06] [PASSED] drm_test_connector_hdmi_init_null_vendor
[13:32:06] [PASSED] drm_test_connector_hdmi_init_product_length_exact
[13:32:06] [PASSED] drm_test_connector_hdmi_init_product_length_too_long
[13:32:06] [PASSED] drm_test_connector_hdmi_init_product_valid
[13:32:06] [PASSED] drm_test_connector_hdmi_init_vendor_length_exact
[13:32:06] [PASSED] drm_test_connector_hdmi_init_vendor_length_too_long
[13:32:06] [PASSED] drm_test_connector_hdmi_init_vendor_valid
[13:32:06] ========= drm_test_connector_hdmi_init_type_valid =========
[13:32:06] [PASSED] HDMI-A
[13:32:06] [PASSED] HDMI-B
[13:32:06] ===== [PASSED] drm_test_connector_hdmi_init_type_valid =====
[13:32:06] ======== drm_test_connector_hdmi_init_type_invalid ========
[13:32:06] [PASSED] Unknown
[13:32:06] [PASSED] VGA
[13:32:06] [PASSED] DVI-I
[13:32:06] [PASSED] DVI-D
[13:32:06] [PASSED] DVI-A
[13:32:06] [PASSED] Composite
[13:32:06] [PASSED] SVIDEO
[13:32:06] [PASSED] LVDS
[13:32:06] [PASSED] Component
[13:32:06] [PASSED] DIN
[13:32:06] [PASSED] DP
[13:32:06] [PASSED] TV
[13:32:06] [PASSED] eDP
[13:32:06] [PASSED] Virtual
[13:32:06] [PASSED] DSI
[13:32:06] [PASSED] DPI
[13:32:06] [PASSED] Writeback
[13:32:06] [PASSED] SPI
[13:32:06] [PASSED] USB
[13:32:06] ==== [PASSED] drm_test_connector_hdmi_init_type_invalid ====
[13:32:06] ============ [PASSED] drmm_connector_hdmi_init =============
[13:32:06] ============= drmm_connector_init (3 subtests) =============
[13:32:06] [PASSED] drm_test_drmm_connector_init
[13:32:06] [PASSED] drm_test_drmm_connector_init_null_ddc
[13:32:06] ========= drm_test_drmm_connector_init_type_valid =========
[13:32:06] [PASSED] Unknown
[13:32:06] [PASSED] VGA
[13:32:06] [PASSED] DVI-I
[13:32:06] [PASSED] DVI-D
[13:32:06] [PASSED] DVI-A
[13:32:06] [PASSED] Composite
[13:32:06] [PASSED] SVIDEO
[13:32:06] [PASSED] LVDS
[13:32:06] [PASSED] Component
[13:32:06] [PASSED] DIN
[13:32:06] [PASSED] DP
[13:32:06] [PASSED] HDMI-A
[13:32:06] [PASSED] HDMI-B
[13:32:06] [PASSED] TV
[13:32:06] [PASSED] eDP
[13:32:06] [PASSED] Virtual
[13:32:06] [PASSED] DSI
[13:32:06] [PASSED] DPI
[13:32:06] [PASSED] Writeback
[13:32:06] [PASSED] SPI
[13:32:06] [PASSED] USB
[13:32:06] ===== [PASSED] drm_test_drmm_connector_init_type_valid =====
[13:32:06] =============== [PASSED] drmm_connector_init ===============
[13:32:06] ========= drm_connector_dynamic_init (6 subtests) ==========
[13:32:06] [PASSED] drm_test_drm_connector_dynamic_init
[13:32:06] [PASSED] drm_test_drm_connector_dynamic_init_null_ddc
[13:32:06] [PASSED] drm_test_drm_connector_dynamic_init_not_added
[13:32:06] [PASSED] drm_test_drm_connector_dynamic_init_properties
[13:32:06] ===== drm_test_drm_connector_dynamic_init_type_valid ======
[13:32:06] [PASSED] Unknown
[13:32:06] [PASSED] VGA
[13:32:06] [PASSED] DVI-I
[13:32:06] [PASSED] DVI-D
[13:32:06] [PASSED] DVI-A
[13:32:06] [PASSED] Composite
[13:32:06] [PASSED] SVIDEO
[13:32:06] [PASSED] LVDS
[13:32:06] [PASSED] Component
[13:32:06] [PASSED] DIN
[13:32:06] [PASSED] DP
[13:32:06] [PASSED] HDMI-A
[13:32:06] [PASSED] HDMI-B
[13:32:06] [PASSED] TV
[13:32:06] [PASSED] eDP
[13:32:06] [PASSED] Virtual
[13:32:06] [PASSED] DSI
[13:32:06] [PASSED] DPI
[13:32:06] [PASSED] Writeback
[13:32:06] [PASSED] SPI
[13:32:06] [PASSED] USB
[13:32:06] = [PASSED] drm_test_drm_connector_dynamic_init_type_valid ==
[13:32:06] ======== drm_test_drm_connector_dynamic_init_name =========
[13:32:06] [PASSED] Unknown
[13:32:06] [PASSED] VGA
[13:32:06] [PASSED] DVI-I
[13:32:06] [PASSED] DVI-D
[13:32:06] [PASSED] DVI-A
[13:32:06] [PASSED] Composite
[13:32:06] [PASSED] SVIDEO
[13:32:06] [PASSED] LVDS
[13:32:06] [PASSED] Component
[13:32:06] [PASSED] DIN
[13:32:06] [PASSED] DP
[13:32:06] [PASSED] HDMI-A
[13:32:06] [PASSED] HDMI-B
[13:32:06] [PASSED] TV
[13:32:06] [PASSED] eDP
[13:32:06] [PASSED] Virtual
[13:32:06] [PASSED] DSI
[13:32:06] [PASSED] DPI
[13:32:06] [PASSED] Writeback
[13:32:06] [PASSED] SPI
[13:32:06] [PASSED] USB
[13:32:06] ==== [PASSED] drm_test_drm_connector_dynamic_init_name =====
[13:32:06] =========== [PASSED] drm_connector_dynamic_init ============
[13:32:06] ==== drm_connector_dynamic_register_early (4 subtests) =====
[13:32:06] [PASSED] drm_test_drm_connector_dynamic_register_early_on_list
[13:32:06] [PASSED] drm_test_drm_connector_dynamic_register_early_defer
[13:32:06] [PASSED] drm_test_drm_connector_dynamic_register_early_no_init
[13:32:06] [PASSED] drm_test_drm_connector_dynamic_register_early_no_mode_object
[13:32:06] ====== [PASSED] drm_connector_dynamic_register_early =======
[13:32:06] ======= drm_connector_dynamic_register (7 subtests) ========
[13:32:06] [PASSED] drm_test_drm_connector_dynamic_register_on_list
[13:32:06] [PASSED] drm_test_drm_connector_dynamic_register_no_defer
[13:32:06] [PASSED] drm_test_drm_connector_dynamic_register_no_init
[13:32:06] [PASSED] drm_test_drm_connector_dynamic_register_mode_object
[13:32:06] [PASSED] drm_test_drm_connector_dynamic_register_sysfs
[13:32:06] [PASSED] drm_test_drm_connector_dynamic_register_sysfs_name
[13:32:06] [PASSED] drm_test_drm_connector_dynamic_register_debugfs
[13:32:06] ========= [PASSED] drm_connector_dynamic_register ==========
[13:32:06] = drm_connector_attach_broadcast_rgb_property (2 subtests) =
[13:32:06] [PASSED] drm_test_drm_connector_attach_broadcast_rgb_property
[13:32:06] [PASSED] drm_test_drm_connector_attach_broadcast_rgb_property_hdmi_connector
[13:32:06] === [PASSED] drm_connector_attach_broadcast_rgb_property ===
[13:32:06] ========== drm_get_tv_mode_from_name (2 subtests) ==========
[13:32:06] ========== drm_test_get_tv_mode_from_name_valid ===========
[13:32:06] [PASSED] NTSC
[13:32:06] [PASSED] NTSC-443
[13:32:06] [PASSED] NTSC-J
[13:32:06] [PASSED] PAL
[13:32:06] [PASSED] PAL-M
[13:32:06] [PASSED] PAL-N
[13:32:06] [PASSED] SECAM
[13:32:06] [PASSED] Mono
[13:32:06] ====== [PASSED] drm_test_get_tv_mode_from_name_valid =======
[13:32:06] [PASSED] drm_test_get_tv_mode_from_name_truncated
[13:32:06] ============ [PASSED] drm_get_tv_mode_from_name ============
[13:32:06] = drm_test_connector_hdmi_compute_mode_clock (12 subtests) =
[13:32:06] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb
[13:32:06] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_10bpc
[13:32:06] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_10bpc_vic_1
[13:32:06] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_12bpc
[13:32:06] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_12bpc_vic_1
[13:32:06] [PASSED] drm_test_drm_hdmi_compute_mode_clock_rgb_double
[13:32:06] = drm_test_connector_hdmi_compute_mode_clock_yuv420_valid =
[13:32:06] [PASSED] VIC 96
[13:32:06] [PASSED] VIC 97
[13:32:06] [PASSED] VIC 101
[13:32:06] [PASSED] VIC 102
[13:32:06] [PASSED] VIC 106
[13:32:06] [PASSED] VIC 107
[13:32:06] === [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_valid ===
[13:32:06] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_10_bpc
[13:32:06] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv420_12_bpc
[13:32:06] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_8_bpc
[13:32:06] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_10_bpc
[13:32:06] [PASSED] drm_test_connector_hdmi_compute_mode_clock_yuv422_12_bpc
[13:32:06] === [PASSED] drm_test_connector_hdmi_compute_mode_clock ====
[13:32:06] == drm_hdmi_connector_get_broadcast_rgb_name (2 subtests) ==
[13:32:06] === drm_test_drm_hdmi_connector_get_broadcast_rgb_name ====
[13:32:06] [PASSED] Automatic
[13:32:06] [PASSED] Full
[13:32:06] [PASSED] Limited 16:235
[13:32:06] === [PASSED] drm_test_drm_hdmi_connector_get_broadcast_rgb_name ===
[13:32:06] [PASSED] drm_test_drm_hdmi_connector_get_broadcast_rgb_name_invalid
[13:32:06] ==== [PASSED] drm_hdmi_connector_get_broadcast_rgb_name ====
[13:32:06] == drm_hdmi_connector_get_output_format_name (2 subtests) ==
[13:32:06] === drm_test_drm_hdmi_connector_get_output_format_name ====
[13:32:06] [PASSED] RGB
[13:32:06] [PASSED] YUV 4:2:0
[13:32:06] [PASSED] YUV 4:2:2
[13:32:06] [PASSED] YUV 4:4:4
[13:32:06] === [PASSED] drm_test_drm_hdmi_connector_get_output_format_name ===
[13:32:06] [PASSED] drm_test_drm_hdmi_connector_get_output_format_name_invalid
[13:32:06] ==== [PASSED] drm_hdmi_connector_get_output_format_name ====
[13:32:06] ============= drm_damage_helper (21 subtests) ==============
[13:32:06] [PASSED] drm_test_damage_iter_no_damage
[13:32:06] [PASSED] drm_test_damage_iter_no_damage_fractional_src
[13:32:06] [PASSED] drm_test_damage_iter_no_damage_src_moved
[13:32:06] [PASSED] drm_test_damage_iter_no_damage_fractional_src_moved
[13:32:06] [PASSED] drm_test_damage_iter_no_damage_not_visible
[13:32:06] [PASSED] drm_test_damage_iter_no_damage_no_crtc
[13:32:06] [PASSED] drm_test_damage_iter_no_damage_no_fb
[13:32:06] [PASSED] drm_test_damage_iter_simple_damage
[13:32:06] [PASSED] drm_test_damage_iter_single_damage
[13:32:06] [PASSED] drm_test_damage_iter_single_damage_intersect_src
[13:32:06] [PASSED] drm_test_damage_iter_single_damage_outside_src
[13:32:06] [PASSED] drm_test_damage_iter_single_damage_fractional_src
[13:32:06] [PASSED] drm_test_damage_iter_single_damage_intersect_fractional_src
[13:32:06] [PASSED] drm_test_damage_iter_single_damage_outside_fractional_src
[13:32:06] [PASSED] drm_test_damage_iter_single_damage_src_moved
[13:32:06] [PASSED] drm_test_damage_iter_single_damage_fractional_src_moved
[13:32:06] [PASSED] drm_test_damage_iter_damage
[13:32:06] [PASSED] drm_test_damage_iter_damage_one_intersect
[13:32:06] [PASSED] drm_test_damage_iter_damage_one_outside
[13:32:06] [PASSED] drm_test_damage_iter_damage_src_moved
[13:32:06] [PASSED] drm_test_damage_iter_damage_not_visible
[13:32:06] ================ [PASSED] drm_damage_helper ================
[13:32:06] ============== drm_dp_mst_helper (3 subtests) ==============
[13:32:06] ============== drm_test_dp_mst_calc_pbn_mode ==============
[13:32:06] [PASSED] Clock 154000 BPP 30 DSC disabled
[13:32:06] [PASSED] Clock 234000 BPP 30 DSC disabled
[13:32:06] [PASSED] Clock 297000 BPP 24 DSC disabled
[13:32:06] [PASSED] Clock 332880 BPP 24 DSC enabled
[13:32:06] [PASSED] Clock 324540 BPP 24 DSC enabled
[13:32:06] ========== [PASSED] drm_test_dp_mst_calc_pbn_mode ==========
[13:32:06] ============== drm_test_dp_mst_calc_pbn_div ===============
[13:32:06] [PASSED] Link rate 2000000 lane count 4
[13:32:06] [PASSED] Link rate 2000000 lane count 2
[13:32:06] [PASSED] Link rate 2000000 lane count 1
[13:32:06] [PASSED] Link rate 1350000 lane count 4
[13:32:06] [PASSED] Link rate 1350000 lane count 2
[13:32:06] [PASSED] Link rate 1350000 lane count 1
[13:32:06] [PASSED] Link rate 1000000 lane count 4
[13:32:06] [PASSED] Link rate 1000000 lane count 2
[13:32:06] [PASSED] Link rate 1000000 lane count 1
[13:32:06] [PASSED] Link rate 810000 lane count 4
[13:32:06] [PASSED] Link rate 810000 lane count 2
[13:32:06] [PASSED] Link rate 810000 lane count 1
[13:32:06] [PASSED] Link rate 540000 lane count 4
[13:32:06] [PASSED] Link rate 540000 lane count 2
[13:32:06] [PASSED] Link rate 540000 lane count 1
[13:32:06] [PASSED] Link rate 270000 lane count 4
[13:32:06] [PASSED] Link rate 270000 lane count 2
[13:32:06] [PASSED] Link rate 270000 lane count 1
[13:32:06] [PASSED] Link rate 162000 lane count 4
[13:32:06] [PASSED] Link rate 162000 lane count 2
[13:32:06] [PASSED] Link rate 162000 lane count 1
[13:32:06] ========== [PASSED] drm_test_dp_mst_calc_pbn_div ===========
[13:32:06] ========= drm_test_dp_mst_sideband_msg_req_decode =========
[13:32:06] [PASSED] DP_ENUM_PATH_RESOURCES with port number
[13:32:06] [PASSED] DP_POWER_UP_PHY with port number
[13:32:06] [PASSED] DP_POWER_DOWN_PHY with port number
[13:32:06] [PASSED] DP_ALLOCATE_PAYLOAD with SDP stream sinks
[13:32:06] [PASSED] DP_ALLOCATE_PAYLOAD with port number
[13:32:06] [PASSED] DP_ALLOCATE_PAYLOAD with VCPI
[13:32:06] [PASSED] DP_ALLOCATE_PAYLOAD with PBN
[13:32:06] [PASSED] DP_QUERY_PAYLOAD with port number
[13:32:06] [PASSED] DP_QUERY_PAYLOAD with VCPI
[13:32:06] [PASSED] DP_REMOTE_DPCD_READ with port number
[13:32:06] [PASSED] DP_REMOTE_DPCD_READ with DPCD address
[13:32:06] [PASSED] DP_REMOTE_DPCD_READ with max number of bytes
[13:32:06] [PASSED] DP_REMOTE_DPCD_WRITE with port number
[13:32:06] [PASSED] DP_REMOTE_DPCD_WRITE with DPCD address
[13:32:06] [PASSED] DP_REMOTE_DPCD_WRITE with data array
[13:32:06] [PASSED] DP_REMOTE_I2C_READ with port number
[13:32:06] [PASSED] DP_REMOTE_I2C_READ with I2C device ID
[13:32:06] [PASSED] DP_REMOTE_I2C_READ with transactions array
[13:32:06] [PASSED] DP_REMOTE_I2C_WRITE with port number
[13:32:06] [PASSED] DP_REMOTE_I2C_WRITE with I2C device ID
[13:32:06] [PASSED] DP_REMOTE_I2C_WRITE with data array
[13:32:06] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream ID
[13:32:06] [PASSED] DP_QUERY_STREAM_ENC_STATUS with client ID
[13:32:06] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream event
[13:32:06] [PASSED] DP_QUERY_STREAM_ENC_STATUS with valid stream event
[13:32:06] [PASSED] DP_QUERY_STREAM_ENC_STATUS with stream behavior
[13:32:06] [PASSED] DP_QUERY_STREAM_ENC_STATUS with a valid stream behavior
[13:32:06] ===== [PASSED] drm_test_dp_mst_sideband_msg_req_decode =====
[13:32:06] ================ [PASSED] drm_dp_mst_helper ================
[13:32:06] ================== drm_exec (7 subtests) ===================
[13:32:06] [PASSED] sanitycheck
[13:32:06] [PASSED] test_lock
[13:32:06] [PASSED] test_lock_unlock
[13:32:06] [PASSED] test_duplicates
[13:32:06] [PASSED] test_prepare
[13:32:06] [PASSED] test_prepare_array
[13:32:06] [PASSED] test_multiple_loops
[13:32:06] ==================== [PASSED] drm_exec =====================
[13:32:06] =========== drm_format_helper_test (17 subtests) ===========
[13:32:06] ============== drm_test_fb_xrgb8888_to_gray8 ==============
[13:32:06] [PASSED] single_pixel_source_buffer
[13:32:06] [PASSED] single_pixel_clip_rectangle
[13:32:06] [PASSED] well_known_colors
[13:32:06] [PASSED] destination_pitch
[13:32:06] ========== [PASSED] drm_test_fb_xrgb8888_to_gray8 ==========
[13:32:06] ============= drm_test_fb_xrgb8888_to_rgb332 ==============
[13:32:06] [PASSED] single_pixel_source_buffer
[13:32:06] [PASSED] single_pixel_clip_rectangle
[13:32:06] [PASSED] well_known_colors
[13:32:06] [PASSED] destination_pitch
[13:32:06] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb332 ==========
[13:32:06] ============= drm_test_fb_xrgb8888_to_rgb565 ==============
[13:32:06] [PASSED] single_pixel_source_buffer
[13:32:06] [PASSED] single_pixel_clip_rectangle
[13:32:06] [PASSED] well_known_colors
[13:32:06] [PASSED] destination_pitch
[13:32:06] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb565 ==========
[13:32:06] ============ drm_test_fb_xrgb8888_to_xrgb1555 =============
[13:32:06] [PASSED] single_pixel_source_buffer
[13:32:06] [PASSED] single_pixel_clip_rectangle
[13:32:06] [PASSED] well_known_colors
[13:32:06] [PASSED] destination_pitch
[13:32:06] ======== [PASSED] drm_test_fb_xrgb8888_to_xrgb1555 =========
[13:32:06] ============ drm_test_fb_xrgb8888_to_argb1555 =============
[13:32:06] [PASSED] single_pixel_source_buffer
[13:32:06] [PASSED] single_pixel_clip_rectangle
[13:32:06] [PASSED] well_known_colors
[13:32:06] [PASSED] destination_pitch
[13:32:06] ======== [PASSED] drm_test_fb_xrgb8888_to_argb1555 =========
[13:32:06] ============ drm_test_fb_xrgb8888_to_rgba5551 =============
[13:32:06] [PASSED] single_pixel_source_buffer
[13:32:06] [PASSED] single_pixel_clip_rectangle
[13:32:06] [PASSED] well_known_colors
[13:32:06] [PASSED] destination_pitch
[13:32:06] ======== [PASSED] drm_test_fb_xrgb8888_to_rgba5551 =========
[13:32:06] ============= drm_test_fb_xrgb8888_to_rgb888 ==============
[13:32:06] [PASSED] single_pixel_source_buffer
[13:32:06] [PASSED] single_pixel_clip_rectangle
[13:32:06] [PASSED] well_known_colors
[13:32:06] [PASSED] destination_pitch
[13:32:06] ========= [PASSED] drm_test_fb_xrgb8888_to_rgb888 ==========
[13:32:06] ============= drm_test_fb_xrgb8888_to_bgr888 ==============
[13:32:06] [PASSED] single_pixel_source_buffer
[13:32:06] [PASSED] single_pixel_clip_rectangle
[13:32:06] [PASSED] well_known_colors
[13:32:06] [PASSED] destination_pitch
[13:32:06] ========= [PASSED] drm_test_fb_xrgb8888_to_bgr888 ==========
[13:32:06] ============ drm_test_fb_xrgb8888_to_argb8888 =============
[13:32:06] [PASSED] single_pixel_source_buffer
[13:32:06] [PASSED] single_pixel_clip_rectangle
[13:32:06] [PASSED] well_known_colors
[13:32:06] [PASSED] destination_pitch
[13:32:06] ======== [PASSED] drm_test_fb_xrgb8888_to_argb8888 =========
[13:32:06] =========== drm_test_fb_xrgb8888_to_xrgb2101010 ===========
[13:32:06] [PASSED] single_pixel_source_buffer
[13:32:06] [PASSED] single_pixel_clip_rectangle
[13:32:06] [PASSED] well_known_colors
[13:32:06] [PASSED] destination_pitch
[13:32:06] ======= [PASSED] drm_test_fb_xrgb8888_to_xrgb2101010 =======
[13:32:06] =========== drm_test_fb_xrgb8888_to_argb2101010 ===========
[13:32:06] [PASSED] single_pixel_source_buffer
[13:32:06] [PASSED] single_pixel_clip_rectangle
[13:32:06] [PASSED] well_known_colors
[13:32:06] [PASSED] destination_pitch
[13:32:06] ======= [PASSED] drm_test_fb_xrgb8888_to_argb2101010 =======
[13:32:06] ============== drm_test_fb_xrgb8888_to_mono ===============
[13:32:06] [PASSED] single_pixel_source_buffer
[13:32:06] [PASSED] single_pixel_clip_rectangle
[13:32:06] [PASSED] well_known_colors
[13:32:06] [PASSED] destination_pitch
[13:32:06] ========== [PASSED] drm_test_fb_xrgb8888_to_mono ===========
[13:32:06] ==================== drm_test_fb_swab =====================
[13:32:06] [PASSED] single_pixel_source_buffer
[13:32:06] [PASSED] single_pixel_clip_rectangle
[13:32:06] [PASSED] well_known_colors
[13:32:06] [PASSED] destination_pitch
[13:32:06] ================ [PASSED] drm_test_fb_swab =================
[13:32:06] ============ drm_test_fb_xrgb8888_to_xbgr8888 =============
[13:32:06] [PASSED] single_pixel_source_buffer
[13:32:06] [PASSED] single_pixel_clip_rectangle
[13:32:06] [PASSED] well_known_colors
[13:32:06] [PASSED] destination_pitch
[13:32:06] ======== [PASSED] drm_test_fb_xrgb8888_to_xbgr8888 =========
[13:32:06] ============ drm_test_fb_xrgb8888_to_abgr8888 =============
[13:32:06] [PASSED] single_pixel_source_buffer
[13:32:06] [PASSED] single_pixel_clip_rectangle
[13:32:06] [PASSED] well_known_colors
[13:32:06] [PASSED] destination_pitch
[13:32:06] ======== [PASSED] drm_test_fb_xrgb8888_to_abgr8888 =========
[13:32:06] ================= drm_test_fb_clip_offset =================
[13:32:06] [PASSED] pass through
[13:32:06] [PASSED] horizontal offset
[13:32:06] [PASSED] vertical offset
[13:32:06] [PASSED] horizontal and vertical offset
[13:32:06] [PASSED] horizontal offset (custom pitch)
[13:32:06] [PASSED] vertical offset (custom pitch)
[13:32:06] [PASSED] horizontal and vertical offset (custom pitch)
[13:32:06] ============= [PASSED] drm_test_fb_clip_offset =============
[13:32:06] =================== drm_test_fb_memcpy ====================
[13:32:06] [PASSED] single_pixel_source_buffer: XR24 little-endian (0x34325258)
[13:32:06] [PASSED] single_pixel_source_buffer: XRA8 little-endian (0x38415258)
[13:32:06] [PASSED] single_pixel_source_buffer: YU24 little-endian (0x34325559)
[13:32:06] [PASSED] single_pixel_clip_rectangle: XB24 little-endian (0x34324258)
[13:32:06] [PASSED] single_pixel_clip_rectangle: XRA8 little-endian (0x38415258)
[13:32:06] [PASSED] single_pixel_clip_rectangle: YU24 little-endian (0x34325559)
[13:32:06] [PASSED] well_known_colors: XB24 little-endian (0x34324258)
[13:32:06] [PASSED] well_known_colors: XRA8 little-endian (0x38415258)
[13:32:06] [PASSED] well_known_colors: YU24 little-endian (0x34325559)
[13:32:06] [PASSED] destination_pitch: XB24 little-endian (0x34324258)
[13:32:06] [PASSED] destination_pitch: XRA8 little-endian (0x38415258)
[13:32:06] [PASSED] destination_pitch: YU24 little-endian (0x34325559)
[13:32:06] =============== [PASSED] drm_test_fb_memcpy ================
[13:32:06] ============= [PASSED] drm_format_helper_test ==============
[13:32:06] ================= drm_format (18 subtests) =================
[13:32:06] [PASSED] drm_test_format_block_width_invalid
[13:32:06] [PASSED] drm_test_format_block_width_one_plane
[13:32:06] [PASSED] drm_test_format_block_width_two_plane
[13:32:06] [PASSED] drm_test_format_block_width_three_plane
[13:32:06] [PASSED] drm_test_format_block_width_tiled
[13:32:06] [PASSED] drm_test_format_block_height_invalid
[13:32:06] [PASSED] drm_test_format_block_height_one_plane
[13:32:06] [PASSED] drm_test_format_block_height_two_plane
[13:32:06] [PASSED] drm_test_format_block_height_three_plane
[13:32:06] [PASSED] drm_test_format_block_height_tiled
[13:32:06] [PASSED] drm_test_format_min_pitch_invalid
[13:32:06] [PASSED] drm_test_format_min_pitch_one_plane_8bpp
[13:32:06] [PASSED] drm_test_format_min_pitch_one_plane_16bpp
[13:32:06] [PASSED] drm_test_format_min_pitch_one_plane_24bpp
[13:32:06] [PASSED] drm_test_format_min_pitch_one_plane_32bpp
[13:32:06] [PASSED] drm_test_format_min_pitch_two_plane
[13:32:06] [PASSED] drm_test_format_min_pitch_three_plane_8bpp
[13:32:06] [PASSED] drm_test_format_min_pitch_tiled
[13:32:06] =================== [PASSED] drm_format ====================
[13:32:06] ============== drm_framebuffer (10 subtests) ===============
[13:32:06] ========== drm_test_framebuffer_check_src_coords ==========
[13:32:06] [PASSED] Success: source fits into fb
[13:32:06] [PASSED] Fail: overflowing fb with x-axis coordinate
[13:32:06] [PASSED] Fail: overflowing fb with y-axis coordinate
[13:32:06] [PASSED] Fail: overflowing fb with source width
[13:32:06] [PASSED] Fail: overflowing fb with source height
[13:32:06] ====== [PASSED] drm_test_framebuffer_check_src_coords ======
[13:32:06] [PASSED] drm_test_framebuffer_cleanup
[13:32:06] =============== drm_test_framebuffer_create ===============
[13:32:06] [PASSED] ABGR8888 normal sizes
[13:32:06] [PASSED] ABGR8888 max sizes
[13:32:06] [PASSED] ABGR8888 pitch greater than min required
[13:32:06] [PASSED] ABGR8888 pitch less than min required
[13:32:06] [PASSED] ABGR8888 Invalid width
[13:32:06] [PASSED] ABGR8888 Invalid buffer handle
[13:32:06] [PASSED] No pixel format
[13:32:06] [PASSED] ABGR8888 Width 0
[13:32:06] [PASSED] ABGR8888 Height 0
[13:32:06] [PASSED] ABGR8888 Out of bound height * pitch combination
[13:32:06] [PASSED] ABGR8888 Large buffer offset
[13:32:06] [PASSED] ABGR8888 Buffer offset for inexistent plane
[13:32:06] [PASSED] ABGR8888 Invalid flag
[13:32:06] [PASSED] ABGR8888 Set DRM_MODE_FB_MODIFIERS without modifiers
[13:32:06] [PASSED] ABGR8888 Valid buffer modifier
[13:32:06] [PASSED] ABGR8888 Invalid buffer modifier(DRM_FORMAT_MOD_SAMSUNG_64_32_TILE)
[13:32:06] [PASSED] ABGR8888 Extra pitches without DRM_MODE_FB_MODIFIERS
[13:32:06] [PASSED] ABGR8888 Extra pitches with DRM_MODE_FB_MODIFIERS
[13:32:06] [PASSED] NV12 Normal sizes
[13:32:06] [PASSED] NV12 Max sizes
[13:32:06] [PASSED] NV12 Invalid pitch
[13:32:06] [PASSED] NV12 Invalid modifier/missing DRM_MODE_FB_MODIFIERS flag
[13:32:06] [PASSED] NV12 different modifier per-plane
[13:32:06] [PASSED] NV12 with DRM_FORMAT_MOD_SAMSUNG_64_32_TILE
[13:32:06] [PASSED] NV12 Valid modifiers without DRM_MODE_FB_MODIFIERS
[13:32:06] [PASSED] NV12 Modifier for inexistent plane
[13:32:06] [PASSED] NV12 Handle for inexistent plane
[13:32:06] [PASSED] NV12 Handle for inexistent plane without DRM_MODE_FB_MODIFIERS
[13:32:06] [PASSED] YVU420 DRM_MODE_FB_MODIFIERS set without modifier
[13:32:06] [PASSED] YVU420 Normal sizes
[13:32:06] [PASSED] YVU420 Max sizes
[13:32:06] [PASSED] YVU420 Invalid pitch
[13:32:06] [PASSED] YVU420 Different pitches
[13:32:06] [PASSED] YVU420 Different buffer offsets/pitches
[13:32:06] [PASSED] YVU420 Modifier set just for plane 0, without DRM_MODE_FB_MODIFIERS
[13:32:06] [PASSED] YVU420 Modifier set just for planes 0, 1, without DRM_MODE_FB_MODIFIERS
[13:32:06] [PASSED] YVU420 Modifier set just for plane 0, 1, with DRM_MODE_FB_MODIFIERS
[13:32:06] [PASSED] YVU420 Valid modifier
[13:32:06] [PASSED] YVU420 Different modifiers per plane
[13:32:06] [PASSED] YVU420 Modifier for inexistent plane
[13:32:06] [PASSED] YUV420_10BIT Invalid modifier(DRM_FORMAT_MOD_LINEAR)
[13:32:06] [PASSED] X0L2 Normal sizes
[13:32:06] [PASSED] X0L2 Max sizes
[13:32:06] [PASSED] X0L2 Invalid pitch
[13:32:06] [PASSED] X0L2 Pitch greater than minimum required
[13:32:06] [PASSED] X0L2 Handle for inexistent plane
[13:32:06] [PASSED] X0L2 Offset for inexistent plane, without DRM_MODE_FB_MODIFIERS set
[13:32:06] [PASSED] X0L2 Modifier without DRM_MODE_FB_MODIFIERS set
[13:32:06] [PASSED] X0L2 Valid modifier
[13:32:06] [PASSED] X0L2 Modifier for inexistent plane
[13:32:06] =========== [PASSED] drm_test_framebuffer_create ===========
[13:32:06] [PASSED] drm_test_framebuffer_free
[13:32:06] [PASSED] drm_test_framebuffer_init
[13:32:06] [PASSED] drm_test_framebuffer_init_bad_format
[13:32:06] [PASSED] drm_test_framebuffer_init_dev_mismatch
[13:32:06] [PASSED] drm_test_framebuffer_lookup
[13:32:06] [PASSED] drm_test_framebuffer_lookup_inexistent
[13:32:06] [PASSED] drm_test_framebuffer_modifiers_not_supported
[13:32:06] ================= [PASSED] drm_framebuffer =================
[13:32:06] ================ drm_gem_shmem (8 subtests) ================
[13:32:06] [PASSED] drm_gem_shmem_test_obj_create
[13:32:06] [PASSED] drm_gem_shmem_test_obj_create_private
[13:32:06] [PASSED] drm_gem_shmem_test_pin_pages
[13:32:06] [PASSED] drm_gem_shmem_test_vmap
[13:32:06] [PASSED] drm_gem_shmem_test_get_sg_table
[13:32:06] [PASSED] drm_gem_shmem_test_get_pages_sgt
[13:32:06] [PASSED] drm_gem_shmem_test_madvise
[13:32:06] [PASSED] drm_gem_shmem_test_purge
[13:32:06] ================== [PASSED] drm_gem_shmem ==================
[13:32:06] === drm_atomic_helper_connector_hdmi_check (27 subtests) ===
[13:32:06] [PASSED] drm_test_check_broadcast_rgb_auto_cea_mode
[13:32:06] [PASSED] drm_test_check_broadcast_rgb_auto_cea_mode_vic_1
[13:32:06] [PASSED] drm_test_check_broadcast_rgb_full_cea_mode
[13:32:06] [PASSED] drm_test_check_broadcast_rgb_full_cea_mode_vic_1
[13:32:06] [PASSED] drm_test_check_broadcast_rgb_limited_cea_mode
[13:32:06] [PASSED] drm_test_check_broadcast_rgb_limited_cea_mode_vic_1
[13:32:06] ====== drm_test_check_broadcast_rgb_cea_mode_yuv420 =======
[13:32:06] [PASSED] Automatic
[13:32:06] [PASSED] Full
[13:32:06] [PASSED] Limited 16:235
[13:32:06] == [PASSED] drm_test_check_broadcast_rgb_cea_mode_yuv420 ===
[13:32:06] [PASSED] drm_test_check_broadcast_rgb_crtc_mode_changed
[13:32:06] [PASSED] drm_test_check_broadcast_rgb_crtc_mode_not_changed
[13:32:06] [PASSED] drm_test_check_disable_connector
[13:32:06] [PASSED] drm_test_check_hdmi_funcs_reject_rate
[13:32:06] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_rgb
[13:32:06] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_yuv420
[13:32:06] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_ignore_yuv422
[13:32:06] [PASSED] drm_test_check_max_tmds_rate_bpc_fallback_ignore_yuv420
[13:32:06] [PASSED] drm_test_check_driver_unsupported_fallback_yuv420
[13:32:06] [PASSED] drm_test_check_output_bpc_crtc_mode_changed
[13:32:06] [PASSED] drm_test_check_output_bpc_crtc_mode_not_changed
[13:32:06] [PASSED] drm_test_check_output_bpc_dvi
[13:32:06] [PASSED] drm_test_check_output_bpc_format_vic_1
[13:32:06] [PASSED] drm_test_check_output_bpc_format_display_8bpc_only
[13:32:06] [PASSED] drm_test_check_output_bpc_format_display_rgb_only
[13:32:06] [PASSED] drm_test_check_output_bpc_format_driver_8bpc_only
[13:32:06] [PASSED] drm_test_check_output_bpc_format_driver_rgb_only
[13:32:06] [PASSED] drm_test_check_tmds_char_rate_rgb_8bpc
[13:32:06] [PASSED] drm_test_check_tmds_char_rate_rgb_10bpc
[13:32:06] [PASSED] drm_test_check_tmds_char_rate_rgb_12bpc
[13:32:06] ===== [PASSED] drm_atomic_helper_connector_hdmi_check ======
[13:32:06] === drm_atomic_helper_connector_hdmi_reset (6 subtests) ====
[13:32:06] [PASSED] drm_test_check_broadcast_rgb_value
[13:32:06] [PASSED] drm_test_check_bpc_8_value
[13:32:06] [PASSED] drm_test_check_bpc_10_value
[13:32:06] [PASSED] drm_test_check_bpc_12_value
[13:32:06] [PASSED] drm_test_check_format_value
[13:32:06] [PASSED] drm_test_check_tmds_char_value
[13:32:06] ===== [PASSED] drm_atomic_helper_connector_hdmi_reset ======
[13:32:06] = drm_atomic_helper_connector_hdmi_mode_valid (4 subtests) =
[13:32:06] [PASSED] drm_test_check_mode_valid
[13:32:06] [PASSED] drm_test_check_mode_valid_reject
[13:32:06] [PASSED] drm_test_check_mode_valid_reject_rate
[13:32:06] [PASSED] drm_test_check_mode_valid_reject_max_clock
[13:32:06] === [PASSED] drm_atomic_helper_connector_hdmi_mode_valid ===
[13:32:06] = drm_atomic_helper_connector_hdmi_infoframes (5 subtests) =
[13:32:06] [PASSED] drm_test_check_infoframes
[13:32:06] [PASSED] drm_test_check_reject_avi_infoframe
[13:32:06] [PASSED] drm_test_check_reject_hdr_infoframe_bpc_8
[13:32:06] [PASSED] drm_test_check_reject_hdr_infoframe_bpc_10
[13:32:06] [PASSED] drm_test_check_reject_audio_infoframe
[13:32:06] === [PASSED] drm_atomic_helper_connector_hdmi_infoframes ===
[13:32:06] ================= drm_managed (2 subtests) =================
[13:32:06] [PASSED] drm_test_managed_release_action
[13:32:06] [PASSED] drm_test_managed_run_action
[13:32:06] =================== [PASSED] drm_managed ===================
[13:32:06] =================== drm_mm (6 subtests) ====================
[13:32:06] [PASSED] drm_test_mm_init
[13:32:06] [PASSED] drm_test_mm_debug
[13:32:06] [PASSED] drm_test_mm_align32
[13:32:06] [PASSED] drm_test_mm_align64
[13:32:06] [PASSED] drm_test_mm_lowest
[13:32:06] [PASSED] drm_test_mm_highest
[13:32:06] ===================== [PASSED] drm_mm ======================
[13:32:06] ============= drm_modes_analog_tv (5 subtests) =============
[13:32:06] [PASSED] drm_test_modes_analog_tv_mono_576i
[13:32:06] [PASSED] drm_test_modes_analog_tv_ntsc_480i
[13:32:06] [PASSED] drm_test_modes_analog_tv_ntsc_480i_inlined
[13:32:06] [PASSED] drm_test_modes_analog_tv_pal_576i
[13:32:06] [PASSED] drm_test_modes_analog_tv_pal_576i_inlined
[13:32:06] =============== [PASSED] drm_modes_analog_tv ===============
[13:32:06] ============== drm_plane_helper (2 subtests) ===============
[13:32:06] =============== drm_test_check_plane_state ================
[13:32:06] [PASSED] clipping_simple
[13:32:06] [PASSED] clipping_rotate_reflect
[13:32:06] [PASSED] positioning_simple
[13:32:06] [PASSED] upscaling
[13:32:06] [PASSED] downscaling
[13:32:06] [PASSED] rounding1
[13:32:06] [PASSED] rounding2
[13:32:06] [PASSED] rounding3
[13:32:06] [PASSED] rounding4
[13:32:06] =========== [PASSED] drm_test_check_plane_state ============
[13:32:06] =========== drm_test_check_invalid_plane_state ============
[13:32:06] [PASSED] positioning_invalid
[13:32:06] [PASSED] upscaling_invalid
[13:32:06] [PASSED] downscaling_invalid
[13:32:06] ======= [PASSED] drm_test_check_invalid_plane_state ========
[13:32:06] ================ [PASSED] drm_plane_helper =================
[13:32:06] ====== drm_connector_helper_tv_get_modes (1 subtest) =======
[13:32:06] ====== drm_test_connector_helper_tv_get_modes_check =======
[13:32:06] [PASSED] None
[13:32:06] [PASSED] PAL
[13:32:06] [PASSED] NTSC
[13:32:06] [PASSED] Both, NTSC Default
[13:32:06] [PASSED] Both, PAL Default
[13:32:06] [PASSED] Both, NTSC Default, with PAL on command-line
[13:32:06] [PASSED] Both, PAL Default, with NTSC on command-line
[13:32:06] == [PASSED] drm_test_connector_helper_tv_get_modes_check ===
[13:32:06] ======== [PASSED] drm_connector_helper_tv_get_modes ========
[13:32:06] ================== drm_rect (9 subtests) ===================
[13:32:06] [PASSED] drm_test_rect_clip_scaled_div_by_zero
[13:32:06] [PASSED] drm_test_rect_clip_scaled_not_clipped
[13:32:06] [PASSED] drm_test_rect_clip_scaled_clipped
[13:32:06] [PASSED] drm_test_rect_clip_scaled_signed_vs_unsigned
[13:32:06] ================= drm_test_rect_intersect =================
[13:32:06] [PASSED] top-left x bottom-right: 2x2+1+1 x 2x2+0+0
[13:32:06] [PASSED] top-right x bottom-left: 2x2+0+0 x 2x2+1-1
[13:32:06] [PASSED] bottom-left x top-right: 2x2+1-1 x 2x2+0+0
[13:32:06] [PASSED] bottom-right x top-left: 2x2+0+0 x 2x2+1+1
[13:32:06] [PASSED] right x left: 2x1+0+0 x 3x1+1+0
[13:32:06] [PASSED] left x right: 3x1+1+0 x 2x1+0+0
[13:32:06] [PASSED] up x bottom: 1x2+0+0 x 1x3+0-1
[13:32:06] [PASSED] bottom x up: 1x3+0-1 x 1x2+0+0
[13:32:06] [PASSED] touching corner: 1x1+0+0 x 2x2+1+1
[13:32:06] [PASSED] touching side: 1x1+0+0 x 1x1+1+0
[13:32:06] [PASSED] equal rects: 2x2+0+0 x 2x2+0+0
[13:32:06] [PASSED] inside another: 2x2+0+0 x 1x1+1+1
[13:32:06] [PASSED] far away: 1x1+0+0 x 1x1+3+6
[13:32:06] [PASSED] points intersecting: 0x0+5+10 x 0x0+5+10
[13:32:06] [PASSED] points not intersecting: 0x0+0+0 x 0x0+5+10
stty: 'standard input': Inappropriate ioctl for device
[13:32:06] ============= [PASSED] drm_test_rect_intersect =============
[13:32:06] ================ drm_test_rect_calc_hscale ================
[13:32:06] [PASSED] normal use
[13:32:06] [PASSED] out of max range
[13:32:06] [PASSED] out of min range
[13:32:06] [PASSED] zero dst
[13:32:06] [PASSED] negative src
[13:32:06] [PASSED] negative dst
[13:32:06] ============ [PASSED] drm_test_rect_calc_hscale ============
[13:32:06] ================ drm_test_rect_calc_vscale ================
[13:32:06] [PASSED] normal use
[13:32:06] [PASSED] out of max range
[13:32:06] [PASSED] out of min range
[13:32:06] [PASSED] zero dst
[13:32:06] [PASSED] negative src
[13:32:06] [PASSED] negative dst
[13:32:06] ============ [PASSED] drm_test_rect_calc_vscale ============
[13:32:06] ================== drm_test_rect_rotate ===================
[13:32:06] [PASSED] reflect-x
[13:32:06] [PASSED] reflect-y
[13:32:06] [PASSED] rotate-0
[13:32:06] [PASSED] rotate-90
[13:32:06] [PASSED] rotate-180
[13:32:06] [PASSED] rotate-270
[13:32:06] ============== [PASSED] drm_test_rect_rotate ===============
[13:32:06] ================ drm_test_rect_rotate_inv =================
[13:32:06] [PASSED] reflect-x
[13:32:06] [PASSED] reflect-y
[13:32:06] [PASSED] rotate-0
[13:32:06] [PASSED] rotate-90
[13:32:06] [PASSED] rotate-180
[13:32:06] [PASSED] rotate-270
[13:32:06] ============ [PASSED] drm_test_rect_rotate_inv =============
[13:32:06] ==================== [PASSED] drm_rect =====================
[13:32:06] ============ drm_sysfb_modeset_test (1 subtest) ============
[13:32:06] ============ drm_test_sysfb_build_fourcc_list =============
[13:32:06] [PASSED] no native formats
[13:32:06] [PASSED] XRGB8888 as native format
[13:32:06] [PASSED] remove duplicates
[13:32:06] [PASSED] convert alpha formats
[13:32:06] [PASSED] random formats
[13:32:06] ======== [PASSED] drm_test_sysfb_build_fourcc_list =========
[13:32:06] ============= [PASSED] drm_sysfb_modeset_test ==============
[13:32:06] ================== drm_fixp (2 subtests) ===================
[13:32:06] [PASSED] drm_test_int2fixp
[13:32:06] [PASSED] drm_test_sm2fixp
[13:32:06] ==================== [PASSED] drm_fixp =====================
[13:32:06] ============================================================
[13:32:06] Testing complete. Ran 630 tests: passed: 630
[13:32:06] Elapsed time: 27.816s total, 1.711s configuring, 25.636s building, 0.424s running
+ /kernel/tools/testing/kunit/kunit.py run --kunitconfig /kernel/drivers/gpu/drm/ttm/tests/.kunitconfig
[13:32:06] Configuring KUnit Kernel ...
Regenerating .config ...
Populating config with:
$ make ARCH=um O=.kunit olddefconfig
[13:32:08] 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
[13:32:18] Starting KUnit Kernel (1/1)...
[13:32:18] ============================================================
Running tests with:
$ .kunit/linux kunit.enable=1 mem=1G console=tty kunit_shutdown=halt
[13:32:18] ================= ttm_device (5 subtests) ==================
[13:32:18] [PASSED] ttm_device_init_basic
[13:32:18] [PASSED] ttm_device_init_multiple
[13:32:18] [PASSED] ttm_device_fini_basic
[13:32:18] [PASSED] ttm_device_init_no_vma_man
[13:32:18] ================== ttm_device_init_pools ==================
[13:32:18] [PASSED] No DMA allocations, no DMA32 required
[13:32:18] [PASSED] DMA allocations, DMA32 required
[13:32:18] [PASSED] No DMA allocations, DMA32 required
[13:32:18] [PASSED] DMA allocations, no DMA32 required
[13:32:18] ============== [PASSED] ttm_device_init_pools ==============
[13:32:18] =================== [PASSED] ttm_device ====================
[13:32:18] ================== ttm_pool (8 subtests) ===================
[13:32:18] ================== ttm_pool_alloc_basic ===================
[13:32:18] [PASSED] One page
[13:32:18] [PASSED] More than one page
[13:32:18] [PASSED] Above the allocation limit
[13:32:18] [PASSED] One page, with coherent DMA mappings enabled
[13:32:18] [PASSED] Above the allocation limit, with coherent DMA mappings enabled
[13:32:18] ============== [PASSED] ttm_pool_alloc_basic ===============
[13:32:18] ============== ttm_pool_alloc_basic_dma_addr ==============
[13:32:18] [PASSED] One page
[13:32:18] [PASSED] More than one page
[13:32:18] [PASSED] Above the allocation limit
[13:32:18] [PASSED] One page, with coherent DMA mappings enabled
[13:32:18] [PASSED] Above the allocation limit, with coherent DMA mappings enabled
[13:32:18] ========== [PASSED] ttm_pool_alloc_basic_dma_addr ==========
[13:32:18] [PASSED] ttm_pool_alloc_order_caching_match
[13:32:18] [PASSED] ttm_pool_alloc_caching_mismatch
[13:32:18] [PASSED] ttm_pool_alloc_order_mismatch
[13:32:18] [PASSED] ttm_pool_free_dma_alloc
[13:32:18] [PASSED] ttm_pool_free_no_dma_alloc
[13:32:18] [PASSED] ttm_pool_fini_basic
[13:32:18] ==================== [PASSED] ttm_pool =====================
[13:32:18] ================ ttm_resource (8 subtests) =================
[13:32:18] ================= ttm_resource_init_basic =================
[13:32:18] [PASSED] Init resource in TTM_PL_SYSTEM
[13:32:18] [PASSED] Init resource in TTM_PL_VRAM
[13:32:18] [PASSED] Init resource in a private placement
[13:32:18] [PASSED] Init resource in TTM_PL_SYSTEM, set placement flags
[13:32:18] ============= [PASSED] ttm_resource_init_basic =============
[13:32:18] [PASSED] ttm_resource_init_pinned
[13:32:18] [PASSED] ttm_resource_fini_basic
[13:32:18] [PASSED] ttm_resource_manager_init_basic
[13:32:18] [PASSED] ttm_resource_manager_usage_basic
[13:32:18] [PASSED] ttm_resource_manager_set_used_basic
[13:32:18] [PASSED] ttm_sys_man_alloc_basic
[13:32:18] [PASSED] ttm_sys_man_free_basic
[13:32:18] ================== [PASSED] ttm_resource ===================
[13:32:18] =================== ttm_tt (15 subtests) ===================
[13:32:18] ==================== ttm_tt_init_basic ====================
[13:32:18] [PASSED] Page-aligned size
[13:32:18] [PASSED] Extra pages requested
[13:32:18] ================ [PASSED] ttm_tt_init_basic ================
[13:32:18] [PASSED] ttm_tt_init_misaligned
[13:32:18] [PASSED] ttm_tt_fini_basic
[13:32:18] [PASSED] ttm_tt_fini_sg
[13:32:18] [PASSED] ttm_tt_fini_shmem
[13:32:18] [PASSED] ttm_tt_create_basic
[13:32:18] [PASSED] ttm_tt_create_invalid_bo_type
[13:32:18] [PASSED] ttm_tt_create_ttm_exists
[13:32:18] [PASSED] ttm_tt_create_failed
[13:32:18] [PASSED] ttm_tt_destroy_basic
[13:32:18] [PASSED] ttm_tt_populate_null_ttm
[13:32:18] [PASSED] ttm_tt_populate_populated_ttm
[13:32:18] [PASSED] ttm_tt_unpopulate_basic
[13:32:18] [PASSED] ttm_tt_unpopulate_empty_ttm
[13:32:18] [PASSED] ttm_tt_swapin_basic
[13:32:18] ===================== [PASSED] ttm_tt ======================
[13:32:18] =================== ttm_bo (14 subtests) ===================
[13:32:18] =========== ttm_bo_reserve_optimistic_no_ticket ===========
[13:32:18] [PASSED] Cannot be interrupted and sleeps
[13:32:18] [PASSED] Cannot be interrupted, locks straight away
[13:32:18] [PASSED] Can be interrupted, sleeps
[13:32:18] ======= [PASSED] ttm_bo_reserve_optimistic_no_ticket =======
[13:32:18] [PASSED] ttm_bo_reserve_locked_no_sleep
[13:32:18] [PASSED] ttm_bo_reserve_no_wait_ticket
[13:32:18] [PASSED] ttm_bo_reserve_double_resv
[13:32:18] [PASSED] ttm_bo_reserve_interrupted
[13:32:18] [PASSED] ttm_bo_reserve_deadlock
[13:32:18] [PASSED] ttm_bo_unreserve_basic
[13:32:18] [PASSED] ttm_bo_unreserve_pinned
[13:32:18] [PASSED] ttm_bo_unreserve_bulk
[13:32:18] [PASSED] ttm_bo_fini_basic
[13:32:18] [PASSED] ttm_bo_fini_shared_resv
[13:32:18] [PASSED] ttm_bo_pin_basic
[13:32:18] [PASSED] ttm_bo_pin_unpin_resource
[13:32:18] [PASSED] ttm_bo_multiple_pin_one_unpin
[13:32:18] ===================== [PASSED] ttm_bo ======================
[13:32:18] ============== ttm_bo_validate (21 subtests) ===============
[13:32:18] ============== ttm_bo_init_reserved_sys_man ===============
[13:32:18] [PASSED] Buffer object for userspace
[13:32:18] [PASSED] Kernel buffer object
[13:32:18] [PASSED] Shared buffer object
[13:32:18] ========== [PASSED] ttm_bo_init_reserved_sys_man ===========
[13:32:18] ============== ttm_bo_init_reserved_mock_man ==============
[13:32:18] [PASSED] Buffer object for userspace
[13:32:18] [PASSED] Kernel buffer object
[13:32:18] [PASSED] Shared buffer object
[13:32:18] ========== [PASSED] ttm_bo_init_reserved_mock_man ==========
[13:32:18] [PASSED] ttm_bo_init_reserved_resv
[13:32:18] ================== ttm_bo_validate_basic ==================
[13:32:18] [PASSED] Buffer object for userspace
[13:32:18] [PASSED] Kernel buffer object
[13:32:18] [PASSED] Shared buffer object
[13:32:18] ============== [PASSED] ttm_bo_validate_basic ==============
[13:32:18] [PASSED] ttm_bo_validate_invalid_placement
[13:32:18] ============= ttm_bo_validate_same_placement ==============
[13:32:18] [PASSED] System manager
[13:32:18] [PASSED] VRAM manager
[13:32:18] ========= [PASSED] ttm_bo_validate_same_placement ==========
[13:32:18] [PASSED] ttm_bo_validate_failed_alloc
[13:32:18] [PASSED] ttm_bo_validate_pinned
[13:32:18] [PASSED] ttm_bo_validate_busy_placement
[13:32:18] ================ ttm_bo_validate_multihop =================
[13:32:18] [PASSED] Buffer object for userspace
[13:32:18] [PASSED] Kernel buffer object
[13:32:18] [PASSED] Shared buffer object
[13:32:18] ============ [PASSED] ttm_bo_validate_multihop =============
[13:32:18] ========== ttm_bo_validate_no_placement_signaled ==========
[13:32:18] [PASSED] Buffer object in system domain, no page vector
[13:32:18] [PASSED] Buffer object in system domain with an existing page vector
[13:32:18] ====== [PASSED] ttm_bo_validate_no_placement_signaled ======
[13:32:18] ======== ttm_bo_validate_no_placement_not_signaled ========
[13:32:18] [PASSED] Buffer object for userspace
[13:32:18] [PASSED] Kernel buffer object
[13:32:18] [PASSED] Shared buffer object
[13:32:18] ==== [PASSED] ttm_bo_validate_no_placement_not_signaled ====
[13:32:18] [PASSED] ttm_bo_validate_move_fence_signaled
[13:32:18] ========= ttm_bo_validate_move_fence_not_signaled =========
[13:32:18] [PASSED] Waits for GPU
[13:32:18] [PASSED] Tries to lock straight away
[13:32:18] ===== [PASSED] ttm_bo_validate_move_fence_not_signaled =====
[13:32:18] [PASSED] ttm_bo_validate_happy_evict
[13:32:18] [PASSED] ttm_bo_validate_all_pinned_evict
[13:32:18] [PASSED] ttm_bo_validate_allowed_only_evict
[13:32:18] [PASSED] ttm_bo_validate_deleted_evict
[13:32:18] [PASSED] ttm_bo_validate_busy_domain_evict
[13:32:18] [PASSED] ttm_bo_validate_evict_gutting
[13:32:18] [PASSED] ttm_bo_validate_recrusive_evict
stty: 'standard input': Inappropriate ioctl for device
[13:32:18] ================= [PASSED] ttm_bo_validate =================
[13:32:18] ============================================================
[13:32:18] Testing complete. Ran 101 tests: passed: 101
[13:32:18] Elapsed time: 11.462s total, 1.700s configuring, 9.546s building, 0.184s running
+ cleanup
++ stat -c %u:%g /kernel
+ chown -R 1003:1003 /kernel
^ permalink raw reply [flat|nested] 10+ messages in thread
* ✓ Xe.CI.BAT: success for Fix fs_reclaim deadlock caused by CCS save/restore
2026-01-29 12:51 [PATCH 0/2] Fix fs_reclaim deadlock caused by CCS save/restore Satyanarayana K V P
` (2 preceding siblings ...)
2026-01-29 13:32 ` ✓ CI.KUnit: success for Fix fs_reclaim deadlock caused by CCS save/restore Patchwork
@ 2026-01-29 14:13 ` Patchwork
3 siblings, 0 replies; 10+ messages in thread
From: Patchwork @ 2026-01-29 14:13 UTC (permalink / raw)
To: Satyanarayana K V P; +Cc: intel-xe
[-- Attachment #1: Type: text/plain, Size: 964 bytes --]
== Series Details ==
Series: Fix fs_reclaim deadlock caused by CCS save/restore
URL : https://patchwork.freedesktop.org/series/160841/
State : success
== Summary ==
CI Bug Log - changes from xe-4469-7d2c5f861ae9bbed1464e1a6aa39a9055139a4b0_BAT -> xe-pw-160841v1_BAT
====================================================
Summary
-------
**SUCCESS**
No regressions found.
Participating hosts (11 -> 11)
------------------------------
No changes in participating hosts
Changes
-------
No changes found
Build changes
-------------
* Linux: xe-4469-7d2c5f861ae9bbed1464e1a6aa39a9055139a4b0 -> xe-pw-160841v1
IGT_8724: 6a10d1136a4e856cc4ceb5b899c32b03ed8ddc5e @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
xe-4469-7d2c5f861ae9bbed1464e1a6aa39a9055139a4b0: 7d2c5f861ae9bbed1464e1a6aa39a9055139a4b0
xe-pw-160841v1: 160841v1
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/intel-xe/xe-pw-160841v1/index.html
[-- Attachment #2: Type: text/html, Size: 1512 bytes --]
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 1/2] drm/xe/vf: Fix fs_reclaim warning with CCS save/restore BB allocation
2026-01-29 12:51 ` [PATCH 1/2] drm/xe/vf: Fix fs_reclaim warning with CCS save/restore BB allocation Satyanarayana K V P
@ 2026-01-29 15:39 ` Thomas Hellström
2026-01-29 18:03 ` Matthew Brost
0 siblings, 1 reply; 10+ messages in thread
From: Thomas Hellström @ 2026-01-29 15:39 UTC (permalink / raw)
To: Satyanarayana K V P, intel-xe
Cc: Matthew Brost, Michal Wajdeczko, Matthew Auld
Hi.
On Thu, 2026-01-29 at 12:51 +0000, Satyanarayana K V P wrote:
> CCS save/restore batch buffers are attached during BO allocation and
> detached during BO teardown. The shrinker triggers xe_bo_move(),
> which is
> used for both allocation and deletion paths.
>
> When BO allocation and shrinking occur concurrently, a circular
> locking
> dependency involving fs_reclaim and swap_guard can occur, leading to
> a
> deadlock such as:
>
> ======================================================
> WARNING: possible circular locking dependency detected
> ------------------------------------------------------
>
> CPU0 CPU1
> ---- ----
> lock(fs_reclaim);
> lock(&sa_manager->swap_guard);
> lock(fs_reclaim);
> lock(&sa_manager->swap_guard);
>
> *** DEADLOCK ***
> =====================================================
>
> To avoid this, allocate CCS save/restore BB BOs using GFP_ATOMIC,
> preventing reclaim from being invoked in this context.
>
> Fixes: 864690cf4dd62 ("drm/xe/vf: Attach and detach CCS copy commands
> with BO")
> Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
> Suggested-by: Matthew Brost <matthew.brost@intel.com>
> Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
> Cc: Matthew Auld <matthew.auld@intel.com>
If shrinking and allocation is indeed happening concurrently, then
GFP_ATOMIC is highly likely to fail. In fact it shouldn't be used if we
can't gracefully recover from a failure or if the failure doesn't
matter at all, like in debugging cases where we can lose data.
I'm trying to wrap my head around the sa manager shadow approach and it
seems to me like external code putting the sa manager in a particular
state and the internal lock is accessed from outside the sa code with
no clearly defined locking rules / asserts? IMO it's a very odd and
fragile construct, in particular when used with guard() where it's
unclear exactly what scope is needing locking.
I'm trying to find some documentation to explain why this is used and
the only thing I can find is
"Directly clearing the BB lacks atomicity and can lead to undefined
behavior if the vCPU is halted mid-operation...."
But what if the vCPU is halted during xe_sa_bo_sync_shadow()? What is
exactly is the atomicity in this case?
Thanks,
Thomas
> ---
> drivers/gpu/drm/xe/xe_bb.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_bb.c b/drivers/gpu/drm/xe/xe_bb.c
> index 8b678297aaa2..355365625df9 100644
> --- a/drivers/gpu/drm/xe/xe_bb.c
> +++ b/drivers/gpu/drm/xe/xe_bb.c
> @@ -62,7 +62,7 @@ struct xe_bb *xe_bb_new(struct xe_gt *gt, u32
> dwords, bool usm)
> struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt, u32 dwords,
> enum xe_sriov_vf_ccs_rw_ctxs ctx_id)
> {
> - struct xe_bb *bb = kmalloc(sizeof(*bb), GFP_KERNEL);
> + struct xe_bb *bb = kmalloc(sizeof(*bb), GFP_ATOMIC);
> struct xe_device *xe = gt_to_xe(gt);
> struct xe_sa_manager *bb_pool;
> int err;
> @@ -78,7 +78,7 @@ struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt, u32
> dwords,
> */
>
> bb_pool = xe->sriov.vf.ccs.contexts[ctx_id].mem.ccs_bb_pool;
> - bb->bo = xe_sa_bo_new(bb_pool, 4 * (dwords + 1));
> + bb->bo = __xe_sa_bo_new(bb_pool, 4 * (dwords + 1),
> GFP_ATOMIC);
>
> if (IS_ERR(bb->bo)) {
> err = PTR_ERR(bb->bo);
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 1/2] drm/xe/vf: Fix fs_reclaim warning with CCS save/restore BB allocation
2026-01-29 15:39 ` Thomas Hellström
@ 2026-01-29 18:03 ` Matthew Brost
2026-01-29 18:30 ` Thomas Hellström
0 siblings, 1 reply; 10+ messages in thread
From: Matthew Brost @ 2026-01-29 18:03 UTC (permalink / raw)
To: Thomas Hellström
Cc: Satyanarayana K V P, intel-xe, Michal Wajdeczko, Matthew Auld
On Thu, Jan 29, 2026 at 04:39:37PM +0100, Thomas Hellström wrote:
> Hi.
>
> On Thu, 2026-01-29 at 12:51 +0000, Satyanarayana K V P wrote:
> > CCS save/restore batch buffers are attached during BO allocation and
> > detached during BO teardown. The shrinker triggers xe_bo_move(),
> > which is
> > used for both allocation and deletion paths.
> >
> > When BO allocation and shrinking occur concurrently, a circular
> > locking
> > dependency involving fs_reclaim and swap_guard can occur, leading to
> > a
> > deadlock such as:
> >
> > ======================================================
> > WARNING: possible circular locking dependency detected
> > ------------------------------------------------------
> >
> > CPU0 CPU1
> > ---- ----
> > lock(fs_reclaim);
> > lock(&sa_manager->swap_guard);
> > lock(fs_reclaim);
> > lock(&sa_manager->swap_guard);
> >
> > *** DEADLOCK ***
> > =====================================================
> >
> > To avoid this, allocate CCS save/restore BB BOs using GFP_ATOMIC,
> > preventing reclaim from being invoked in this context.
> >
> > Fixes: 864690cf4dd62 ("drm/xe/vf: Attach and detach CCS copy commands
> > with BO")
> > Signed-off-by: Satyanarayana K V P <satyanarayana.k.v.p@intel.com>
> > Suggested-by: Matthew Brost <matthew.brost@intel.com>
> > Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
> > Cc: Matthew Auld <matthew.auld@intel.com>
>
> If shrinking and allocation is indeed happening concurrently, then
> GFP_ATOMIC is highly likely to fail. In fact it shouldn't be used if we
> can't gracefully recover from a failure or if the failure doesn't
> matter at all, like in debugging cases where we can lose data.
>
This is a good point. We might need to rethink this one. Would
GFP_NOWAIT be better here? Also as you have hit on - only the allocation
path can fail which is handled gracefully (i.e., the shrinker path
doesn't rely on allocations).
> I'm trying to wrap my head around the sa manager shadow approach and it
> seems to me like external code putting the sa manager in a particular
> state and the internal lock is accessed from outside the sa code with
> no clearly defined locking rules / asserts? IMO it's a very odd and
> fragile construct, in particular when used with guard() where it's
> unclear exactly what scope is needing locking.
The lock protects:
- xe_sa_bo_swap_shadow through xe_sa_bo_sync_shadow, so this includes
xe_bb_ccs_new as it picks the correct SA based on shadow state.
>
> I'm trying to find some documentation to explain why this is used and
> the only thing I can find is
>
> "Directly clearing the BB lacks atomicity and can lead to undefined
> behavior if the vCPU is halted mid-operation...."
This could be improved.
>
> But what if the vCPU is halted during xe_sa_bo_sync_shadow()? What is
> exactly is the atomicity in this case?
>
The vCPU is updating the shadow buffer which isn't programmed on the GPU
so if it interrupted mid-instruction writing, the GPU doesn't hang on a
partially written instruction.
Some back ground, I had suggested a AVX based solution to write out /
clear instructions solution [1] but it was rejected in favor of a shadow
buffer solution which appears to have lockdep issues.
Any ideas here would be helpful.
Matt
[1] https://patchwork.freedesktop.org/series/156482/
> Thanks,
> Thomas
>
>
> > ---
> > drivers/gpu/drm/xe/xe_bb.c | 4 ++--
> > 1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_bb.c b/drivers/gpu/drm/xe/xe_bb.c
> > index 8b678297aaa2..355365625df9 100644
> > --- a/drivers/gpu/drm/xe/xe_bb.c
> > +++ b/drivers/gpu/drm/xe/xe_bb.c
> > @@ -62,7 +62,7 @@ struct xe_bb *xe_bb_new(struct xe_gt *gt, u32
> > dwords, bool usm)
> > struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt, u32 dwords,
> > enum xe_sriov_vf_ccs_rw_ctxs ctx_id)
> > {
> > - struct xe_bb *bb = kmalloc(sizeof(*bb), GFP_KERNEL);
> > + struct xe_bb *bb = kmalloc(sizeof(*bb), GFP_ATOMIC);
> > struct xe_device *xe = gt_to_xe(gt);
> > struct xe_sa_manager *bb_pool;
> > int err;
> > @@ -78,7 +78,7 @@ struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt, u32
> > dwords,
> > */
> >
> > bb_pool = xe->sriov.vf.ccs.contexts[ctx_id].mem.ccs_bb_pool;
> > - bb->bo = xe_sa_bo_new(bb_pool, 4 * (dwords + 1));
> > + bb->bo = __xe_sa_bo_new(bb_pool, 4 * (dwords + 1),
> > GFP_ATOMIC);
> >
> > if (IS_ERR(bb->bo)) {
> > err = PTR_ERR(bb->bo);
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 1/2] drm/xe/vf: Fix fs_reclaim warning with CCS save/restore BB allocation
2026-01-29 18:03 ` Matthew Brost
@ 2026-01-29 18:30 ` Thomas Hellström
2026-01-29 19:35 ` Matthew Brost
0 siblings, 1 reply; 10+ messages in thread
From: Thomas Hellström @ 2026-01-29 18:30 UTC (permalink / raw)
To: Matthew Brost
Cc: Satyanarayana K V P, intel-xe, Michal Wajdeczko, Matthew Auld
On Thu, 2026-01-29 at 10:03 -0800, Matthew Brost wrote:
> On Thu, Jan 29, 2026 at 04:39:37PM +0100, Thomas Hellström wrote:
> > Hi.
> >
> > On Thu, 2026-01-29 at 12:51 +0000, Satyanarayana K V P wrote:
> > > CCS save/restore batch buffers are attached during BO allocation
> > > and
> > > detached during BO teardown. The shrinker triggers xe_bo_move(),
> > > which is
> > > used for both allocation and deletion paths.
> > >
> > > When BO allocation and shrinking occur concurrently, a circular
> > > locking
> > > dependency involving fs_reclaim and swap_guard can occur, leading
> > > to
> > > a
> > > deadlock such as:
> > >
> > > ======================================================
> > > WARNING: possible circular locking dependency detected
> > > ------------------------------------------------------
> > >
> > > CPU0 CPU1
> > > ---- ----
> > > lock(fs_reclaim);
> > > lock(&sa_manager->swap_guard);
> > > lock(fs_reclaim);
> > > lock(&sa_manager->swap_guard);
> > >
> > > *** DEADLOCK ***
> > > =====================================================
> > >
> > > To avoid this, allocate CCS save/restore BB BOs using GFP_ATOMIC,
> > > preventing reclaim from being invoked in this context.
> > >
> > > Fixes: 864690cf4dd62 ("drm/xe/vf: Attach and detach CCS copy
> > > commands
> > > with BO")
> > > Signed-off-by: Satyanarayana K V P
> > > <satyanarayana.k.v.p@intel.com>
> > > Suggested-by: Matthew Brost <matthew.brost@intel.com>
> > > Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
> > > Cc: Matthew Auld <matthew.auld@intel.com>
> >
> > If shrinking and allocation is indeed happening concurrently, then
> > GFP_ATOMIC is highly likely to fail. In fact it shouldn't be used
> > if we
> > can't gracefully recover from a failure or if the failure doesn't
> > matter at all, like in debugging cases where we can lose data.
> >
>
> This is a good point. We might need to rethink this one. Would
> GFP_NOWAIT be better here? Also as you have hit on - only the
> allocation
> path can fail which is handled gracefully (i.e., the shrinker path
> doesn't rely on allocations).
I think if the lock is needed, the typical approach would be to split
bb_new into alloc() outside the lock and init() inside the lock if
needed. It looks like tlb_inval is using GFP_ATOMIC as well, so should
be able to benefit. That touches drm code, though, but I think
GFP_KERNEL should be used throughout unless we have a backup path or
don't care about failures.
>
> > I'm trying to wrap my head around the sa manager shadow approach
> > and it
> > seems to me like external code putting the sa manager in a
> > particular
> > state and the internal lock is accessed from outside the sa code
> > with
> > no clearly defined locking rules / asserts? IMO it's a very odd and
> > fragile construct, in particular when used with guard() where it's
> > unclear exactly what scope is needing locking.
>
> The lock protects:
>
> - xe_sa_bo_swap_shadow through xe_sa_bo_sync_shadow, so this
> includes
> xe_bb_ccs_new as it picks the correct SA based on shadow state.
So then basically all (or most) calls that touch a shadow-enabled sa
manager need to be called under the lock?
>
> >
> > I'm trying to find some documentation to explain why this is used
> > and
> > the only thing I can find is
> >
> > "Directly clearing the BB lacks atomicity and can lead to undefined
> > behavior if the vCPU is halted mid-operation...."
>
> This could be improved.
>
> >
> > But what if the vCPU is halted during xe_sa_bo_sync_shadow()? What
> > is
> > exactly is the atomicity in this case?
> >
>
> The vCPU is updating the shadow buffer which isn't programmed on the
> GPU
> so if it interrupted mid-instruction writing, the GPU doesn't hang on
> a
> partially written instruction.
Oh, so we're updating a buffer simultaneously executing on the GPU?
>
> Some back ground, I had suggested a AVX based solution to write out /
> clear instructions solution [1] but it was rejected in favor of a
> shadow
> buffer solution which appears to have lockdep issues.
>
> Any ideas here would be helpful.
I still don't fully get the flow, I must admit, and the interaction
with the CPU-buffer we already have there on DGFX...
Thanks,
Thomas
>
> Matt
>
> [1] https://patchwork.freedesktop.org/series/156482/
>
> > Thanks,
> > Thomas
> >
> >
> > > ---
> > > drivers/gpu/drm/xe/xe_bb.c | 4 ++--
> > > 1 file changed, 2 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/drivers/gpu/drm/xe/xe_bb.c
> > > b/drivers/gpu/drm/xe/xe_bb.c
> > > index 8b678297aaa2..355365625df9 100644
> > > --- a/drivers/gpu/drm/xe/xe_bb.c
> > > +++ b/drivers/gpu/drm/xe/xe_bb.c
> > > @@ -62,7 +62,7 @@ struct xe_bb *xe_bb_new(struct xe_gt *gt, u32
> > > dwords, bool usm)
> > > struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt, u32 dwords,
> > > enum xe_sriov_vf_ccs_rw_ctxs ctx_id)
> > > {
> > > - struct xe_bb *bb = kmalloc(sizeof(*bb), GFP_KERNEL);
> > > + struct xe_bb *bb = kmalloc(sizeof(*bb), GFP_ATOMIC);
> > > struct xe_device *xe = gt_to_xe(gt);
> > > struct xe_sa_manager *bb_pool;
> > > int err;
> > > @@ -78,7 +78,7 @@ struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt,
> > > u32
> > > dwords,
> > > */
> > >
> > > bb_pool = xe-
> > > >sriov.vf.ccs.contexts[ctx_id].mem.ccs_bb_pool;
> > > - bb->bo = xe_sa_bo_new(bb_pool, 4 * (dwords + 1));
> > > + bb->bo = __xe_sa_bo_new(bb_pool, 4 * (dwords + 1),
> > > GFP_ATOMIC);
> > >
> > > if (IS_ERR(bb->bo)) {
> > > err = PTR_ERR(bb->bo);
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 1/2] drm/xe/vf: Fix fs_reclaim warning with CCS save/restore BB allocation
2026-01-29 18:30 ` Thomas Hellström
@ 2026-01-29 19:35 ` Matthew Brost
2026-01-31 1:28 ` Matthew Brost
0 siblings, 1 reply; 10+ messages in thread
From: Matthew Brost @ 2026-01-29 19:35 UTC (permalink / raw)
To: Thomas Hellström
Cc: Satyanarayana K V P, intel-xe, Michal Wajdeczko, Matthew Auld
On Thu, Jan 29, 2026 at 07:30:01PM +0100, Thomas Hellström wrote:
> On Thu, 2026-01-29 at 10:03 -0800, Matthew Brost wrote:
> > On Thu, Jan 29, 2026 at 04:39:37PM +0100, Thomas Hellström wrote:
> > > Hi.
> > >
> > > On Thu, 2026-01-29 at 12:51 +0000, Satyanarayana K V P wrote:
> > > > CCS save/restore batch buffers are attached during BO allocation
> > > > and
> > > > detached during BO teardown. The shrinker triggers xe_bo_move(),
> > > > which is
> > > > used for both allocation and deletion paths.
> > > >
> > > > When BO allocation and shrinking occur concurrently, a circular
> > > > locking
> > > > dependency involving fs_reclaim and swap_guard can occur, leading
> > > > to
> > > > a
> > > > deadlock such as:
> > > >
> > > > ======================================================
> > > > WARNING: possible circular locking dependency detected
> > > > ------------------------------------------------------
> > > >
> > > > CPU0 CPU1
> > > > ---- ----
> > > > lock(fs_reclaim);
> > > > lock(&sa_manager->swap_guard);
> > > > lock(fs_reclaim);
> > > > lock(&sa_manager->swap_guard);
> > > >
> > > > *** DEADLOCK ***
> > > > =====================================================
> > > >
> > > > To avoid this, allocate CCS save/restore BB BOs using GFP_ATOMIC,
> > > > preventing reclaim from being invoked in this context.
> > > >
> > > > Fixes: 864690cf4dd62 ("drm/xe/vf: Attach and detach CCS copy
> > > > commands
> > > > with BO")
> > > > Signed-off-by: Satyanarayana K V P
> > > > <satyanarayana.k.v.p@intel.com>
> > > > Suggested-by: Matthew Brost <matthew.brost@intel.com>
> > > > Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
> > > > Cc: Matthew Auld <matthew.auld@intel.com>
> > >
> > > If shrinking and allocation is indeed happening concurrently, then
> > > GFP_ATOMIC is highly likely to fail. In fact it shouldn't be used
> > > if we
> > > can't gracefully recover from a failure or if the failure doesn't
> > > matter at all, like in debugging cases where we can lose data.
> > >
> >
> > This is a good point. We might need to rethink this one. Would
> > GFP_NOWAIT be better here? Also as you have hit on - only the
> > allocation
> > path can fail which is handled gracefully (i.e., the shrinker path
> > doesn't rely on allocations).
>
> I think if the lock is needed, the typical approach would be to split
> bb_new into alloc() outside the lock and init() inside the lock if
> needed. It looks like tlb_inval is using GFP_ATOMIC as well, so should
> be able to benefit. That touches drm code, though, but I think
> GFP_KERNEL should be used throughout unless we have a backup path or
> don't care about failures.
>
+1. I talked this over with Thomas a agree spliting alloc() and init()
is the best solution + likely can use this going forward or in some
existing cases.
> >
> > > I'm trying to wrap my head around the sa manager shadow approach
> > > and it
> > > seems to me like external code putting the sa manager in a
> > > particular
> > > state and the internal lock is accessed from outside the sa code
> > > with
> > > no clearly defined locking rules / asserts? IMO it's a very odd and
> > > fragile construct, in particular when used with guard() where it's
> > > unclear exactly what scope is needing locking.
> >
> > The lock protects:
> >
> > - xe_sa_bo_swap_shadow through xe_sa_bo_sync_shadow, so this
> > includes
> > xe_bb_ccs_new as it picks the correct SA based on shadow state.
>
> So then basically all (or most) calls that touch a shadow-enabled sa
> manager need to be called under the lock?
>
Correct. So I think the guards in the correct places.
Matt
> >
> > >
> > > I'm trying to find some documentation to explain why this is used
> > > and
> > > the only thing I can find is
> > >
> > > "Directly clearing the BB lacks atomicity and can lead to undefined
> > > behavior if the vCPU is halted mid-operation...."
> >
> > This could be improved.
> >
> > >
> > > But what if the vCPU is halted during xe_sa_bo_sync_shadow()? What
> > > is
> > > exactly is the atomicity in this case?
> > >
> >
> > The vCPU is updating the shadow buffer which isn't programmed on the
> > GPU
> > so if it interrupted mid-instruction writing, the GPU doesn't hang on
> > a
> > partially written instruction.
>
> Oh, so we're updating a buffer simultaneously executing on the GPU?
>
> >
> > Some back ground, I had suggested a AVX based solution to write out /
> > clear instructions solution [1] but it was rejected in favor of a
> > shadow
> > buffer solution which appears to have lockdep issues.
> >
> > Any ideas here would be helpful.
>
> I still don't fully get the flow, I must admit, and the interaction
> with the CPU-buffer we already have there on DGFX...
>
> Thanks,
> Thomas
>
>
> >
> > Matt
> >
> > [1] https://patchwork.freedesktop.org/series/156482/
> >
> > > Thanks,
> > > Thomas
> > >
> > >
> > > > ---
> > > > drivers/gpu/drm/xe/xe_bb.c | 4 ++--
> > > > 1 file changed, 2 insertions(+), 2 deletions(-)
> > > >
> > > > diff --git a/drivers/gpu/drm/xe/xe_bb.c
> > > > b/drivers/gpu/drm/xe/xe_bb.c
> > > > index 8b678297aaa2..355365625df9 100644
> > > > --- a/drivers/gpu/drm/xe/xe_bb.c
> > > > +++ b/drivers/gpu/drm/xe/xe_bb.c
> > > > @@ -62,7 +62,7 @@ struct xe_bb *xe_bb_new(struct xe_gt *gt, u32
> > > > dwords, bool usm)
> > > > struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt, u32 dwords,
> > > > enum xe_sriov_vf_ccs_rw_ctxs ctx_id)
> > > > {
> > > > - struct xe_bb *bb = kmalloc(sizeof(*bb), GFP_KERNEL);
> > > > + struct xe_bb *bb = kmalloc(sizeof(*bb), GFP_ATOMIC);
> > > > struct xe_device *xe = gt_to_xe(gt);
> > > > struct xe_sa_manager *bb_pool;
> > > > int err;
> > > > @@ -78,7 +78,7 @@ struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt,
> > > > u32
> > > > dwords,
> > > > */
> > > >
> > > > bb_pool = xe-
> > > > >sriov.vf.ccs.contexts[ctx_id].mem.ccs_bb_pool;
> > > > - bb->bo = xe_sa_bo_new(bb_pool, 4 * (dwords + 1));
> > > > + bb->bo = __xe_sa_bo_new(bb_pool, 4 * (dwords + 1),
> > > > GFP_ATOMIC);
> > > >
> > > > if (IS_ERR(bb->bo)) {
> > > > err = PTR_ERR(bb->bo);
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 1/2] drm/xe/vf: Fix fs_reclaim warning with CCS save/restore BB allocation
2026-01-29 19:35 ` Matthew Brost
@ 2026-01-31 1:28 ` Matthew Brost
0 siblings, 0 replies; 10+ messages in thread
From: Matthew Brost @ 2026-01-31 1:28 UTC (permalink / raw)
To: Thomas Hellström
Cc: Satyanarayana K V P, intel-xe, Michal Wajdeczko, Matthew Auld
On Thu, Jan 29, 2026 at 11:35:44AM -0800, Matthew Brost wrote:
> On Thu, Jan 29, 2026 at 07:30:01PM +0100, Thomas Hellström wrote:
> > On Thu, 2026-01-29 at 10:03 -0800, Matthew Brost wrote:
> > > On Thu, Jan 29, 2026 at 04:39:37PM +0100, Thomas Hellström wrote:
> > > > Hi.
> > > >
> > > > On Thu, 2026-01-29 at 12:51 +0000, Satyanarayana K V P wrote:
> > > > > CCS save/restore batch buffers are attached during BO allocation
> > > > > and
> > > > > detached during BO teardown. The shrinker triggers xe_bo_move(),
> > > > > which is
> > > > > used for both allocation and deletion paths.
> > > > >
> > > > > When BO allocation and shrinking occur concurrently, a circular
> > > > > locking
> > > > > dependency involving fs_reclaim and swap_guard can occur, leading
> > > > > to
> > > > > a
> > > > > deadlock such as:
> > > > >
> > > > > ======================================================
> > > > > WARNING: possible circular locking dependency detected
> > > > > ------------------------------------------------------
> > > > >
> > > > > CPU0 CPU1
> > > > > ---- ----
> > > > > lock(fs_reclaim);
> > > > > lock(&sa_manager->swap_guard);
> > > > > lock(fs_reclaim);
> > > > > lock(&sa_manager->swap_guard);
> > > > >
> > > > > *** DEADLOCK ***
> > > > > =====================================================
> > > > >
> > > > > To avoid this, allocate CCS save/restore BB BOs using GFP_ATOMIC,
> > > > > preventing reclaim from being invoked in this context.
> > > > >
> > > > > Fixes: 864690cf4dd62 ("drm/xe/vf: Attach and detach CCS copy
> > > > > commands
> > > > > with BO")
> > > > > Signed-off-by: Satyanarayana K V P
> > > > > <satyanarayana.k.v.p@intel.com>
> > > > > Suggested-by: Matthew Brost <matthew.brost@intel.com>
> > > > > Cc: Michal Wajdeczko <michal.wajdeczko@intel.com>
> > > > > Cc: Matthew Auld <matthew.auld@intel.com>
> > > >
> > > > If shrinking and allocation is indeed happening concurrently, then
> > > > GFP_ATOMIC is highly likely to fail. In fact it shouldn't be used
> > > > if we
> > > > can't gracefully recover from a failure or if the failure doesn't
> > > > matter at all, like in debugging cases where we can lose data.
> > > >
> > >
> > > This is a good point. We might need to rethink this one. Would
> > > GFP_NOWAIT be better here? Also as you have hit on - only the
> > > allocation
> > > path can fail which is handled gracefully (i.e., the shrinker path
> > > doesn't rely on allocations).
> >
> > I think if the lock is needed, the typical approach would be to split
> > bb_new into alloc() outside the lock and init() inside the lock if
> > needed. It looks like tlb_inval is using GFP_ATOMIC as well, so should
> > be able to benefit. That touches drm code, though, but I think
> > GFP_KERNEL should be used throughout unless we have a backup path or
> > don't care about failures.
> >
>
> +1. I talked this over with Thomas a agree spliting alloc() and init()
> is the best solution + likely can use this going forward or in some
> existing cases.
>
Thomas - since this is a Fixes patch should be start with this patch as
is (maybe s/ATOMIC/NOWAIT) for easy backporting then following patch in
the series does alloc() / init() split? I have feeling the split will a
fairly big diff given it will likely touch 4 layers.
Matt
> > >
> > > > I'm trying to wrap my head around the sa manager shadow approach
> > > > and it
> > > > seems to me like external code putting the sa manager in a
> > > > particular
> > > > state and the internal lock is accessed from outside the sa code
> > > > with
> > > > no clearly defined locking rules / asserts? IMO it's a very odd and
> > > > fragile construct, in particular when used with guard() where it's
> > > > unclear exactly what scope is needing locking.
> > >
> > > The lock protects:
> > >
> > > - xe_sa_bo_swap_shadow through xe_sa_bo_sync_shadow, so this
> > > includes
> > > xe_bb_ccs_new as it picks the correct SA based on shadow state.
> >
> > So then basically all (or most) calls that touch a shadow-enabled sa
> > manager need to be called under the lock?
> >
>
> Correct. So I think the guards in the correct places.
>
> Matt
>
> > >
> > > >
> > > > I'm trying to find some documentation to explain why this is used
> > > > and
> > > > the only thing I can find is
> > > >
> > > > "Directly clearing the BB lacks atomicity and can lead to undefined
> > > > behavior if the vCPU is halted mid-operation...."
> > >
> > > This could be improved.
> > >
> > > >
> > > > But what if the vCPU is halted during xe_sa_bo_sync_shadow()? What
> > > > is
> > > > exactly is the atomicity in this case?
> > > >
> > >
> > > The vCPU is updating the shadow buffer which isn't programmed on the
> > > GPU
> > > so if it interrupted mid-instruction writing, the GPU doesn't hang on
> > > a
> > > partially written instruction.
> >
> > Oh, so we're updating a buffer simultaneously executing on the GPU?
> >
> > >
> > > Some back ground, I had suggested a AVX based solution to write out /
> > > clear instructions solution [1] but it was rejected in favor of a
> > > shadow
> > > buffer solution which appears to have lockdep issues.
> > >
> > > Any ideas here would be helpful.
> >
> > I still don't fully get the flow, I must admit, and the interaction
> > with the CPU-buffer we already have there on DGFX...
> >
> > Thanks,
> > Thomas
> >
> >
> > >
> > > Matt
> > >
> > > [1] https://patchwork.freedesktop.org/series/156482/
> > >
> > > > Thanks,
> > > > Thomas
> > > >
> > > >
> > > > > ---
> > > > > drivers/gpu/drm/xe/xe_bb.c | 4 ++--
> > > > > 1 file changed, 2 insertions(+), 2 deletions(-)
> > > > >
> > > > > diff --git a/drivers/gpu/drm/xe/xe_bb.c
> > > > > b/drivers/gpu/drm/xe/xe_bb.c
> > > > > index 8b678297aaa2..355365625df9 100644
> > > > > --- a/drivers/gpu/drm/xe/xe_bb.c
> > > > > +++ b/drivers/gpu/drm/xe/xe_bb.c
> > > > > @@ -62,7 +62,7 @@ struct xe_bb *xe_bb_new(struct xe_gt *gt, u32
> > > > > dwords, bool usm)
> > > > > struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt, u32 dwords,
> > > > > enum xe_sriov_vf_ccs_rw_ctxs ctx_id)
> > > > > {
> > > > > - struct xe_bb *bb = kmalloc(sizeof(*bb), GFP_KERNEL);
> > > > > + struct xe_bb *bb = kmalloc(sizeof(*bb), GFP_ATOMIC);
> > > > > struct xe_device *xe = gt_to_xe(gt);
> > > > > struct xe_sa_manager *bb_pool;
> > > > > int err;
> > > > > @@ -78,7 +78,7 @@ struct xe_bb *xe_bb_ccs_new(struct xe_gt *gt,
> > > > > u32
> > > > > dwords,
> > > > > */
> > > > >
> > > > > bb_pool = xe-
> > > > > >sriov.vf.ccs.contexts[ctx_id].mem.ccs_bb_pool;
> > > > > - bb->bo = xe_sa_bo_new(bb_pool, 4 * (dwords + 1));
> > > > > + bb->bo = __xe_sa_bo_new(bb_pool, 4 * (dwords + 1),
> > > > > GFP_ATOMIC);
> > > > >
> > > > > if (IS_ERR(bb->bo)) {
> > > > > err = PTR_ERR(bb->bo);
^ permalink raw reply [flat|nested] 10+ messages in thread
end of thread, other threads:[~2026-01-31 1:28 UTC | newest]
Thread overview: 10+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-01-29 12:51 [PATCH 0/2] Fix fs_reclaim deadlock caused by CCS save/restore Satyanarayana K V P
2026-01-29 12:51 ` [PATCH 1/2] drm/xe/vf: Fix fs_reclaim warning with CCS save/restore BB allocation Satyanarayana K V P
2026-01-29 15:39 ` Thomas Hellström
2026-01-29 18:03 ` Matthew Brost
2026-01-29 18:30 ` Thomas Hellström
2026-01-29 19:35 ` Matthew Brost
2026-01-31 1:28 ` Matthew Brost
2026-01-29 12:51 ` [PATCH 2/2] drm/xe/sa: Add lockdep annotations for SA manager swap_guard Satyanarayana K V P
2026-01-29 13:32 ` ✓ CI.KUnit: success for Fix fs_reclaim deadlock caused by CCS save/restore Patchwork
2026-01-29 14:13 ` ✓ Xe.CI.BAT: " Patchwork
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox