7.2-stable review patch. If anyone has any objections, please let me know. ------------------ From: Varun Gupta [ Upstream commit 20fce5b34b21a995839743b4917a1edd2fd503ba ] During VM teardown, the VM's runtime PM reference is dropped asynchronously, allowing the device to autosuspend while stale page faults belonging to the now-dead VM are still queued. When the page-fault worker later tries to ack one of these, it calls into guc_ct_send_locked() on an already-suspended device, tripping:   Assertion `!xe_pm_runtime_suspended(xe)` failed!   WARNING at xe_device.c:1267 xe_device_assert_mem_access+0x11c/0x140 [xe] A live VM/exec queue always holds a PM reference while it has outstanding work, so if the device is suspended at ack time, the owning context is already gone and the fault is stale. Take a runtime PM reference across the entire pagefault queue worker to safely deliver acks for torn-down VMs. v3: - Move PM ref to the generic xe_pagefault_queue_work using guard(xe_pm_runtime)(xe) instead of tracking it in the GuC backend(Matt Brost). v2: - Hold PM ref across the entire batch (begin/end) instead of per-ack. This prevents the device from autosuspending mid-batch, which would leave write_only acks written but the end flush skipped, and skip counter++, desyncing the cadence check.(Himal) - Add a comment explaining stale faults.(Himal) Fixes: f289f7807119 ("drm/xe: Add xe_guc_pagefault layer") Signed-off-by: Varun Gupta Reviewed-by: Matthew Brost Reviewed-by: Tejas Upadhyay Link: https://patch.msgid.link/20260907050011.497181-2-varun.gupta@intel.com Signed-off-by: Tejas Upadhyay (cherry picked from commit fcc2431d2213dc4d04250c4f1ae87d9c3ae0d455) Signed-off-by: Rodrigo Vivi [Rodrigo: Added xe_device struct for compatibility while cherry-picking] Signed-off-by: Sasha Levin --- drivers/gpu/drm/xe/xe_pagefault.c | 10 ++++++++++ drivers/gpu/drm/xe/xe_pagefault_types.h | 3 +++ 2 files changed, 13 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_pagefault.c b/drivers/gpu/drm/xe/xe_pagefault.c index dbf8f71d3328a..a4986df8328d6 100644 --- a/drivers/gpu/drm/xe/xe_pagefault.c +++ b/drivers/gpu/drm/xe/xe_pagefault.c @@ -16,6 +16,7 @@ #include "xe_hw_engine.h" #include "xe_pagefault.h" #include "xe_pagefault_types.h" +#include "xe_pm.h" #include "xe_svm.h" #include "xe_trace_bo.h" #include "xe_vm.h" @@ -292,9 +293,17 @@ static void xe_pagefault_queue_work(struct work_struct *w) { struct xe_pagefault_queue *pf_queue = container_of(w, typeof(*pf_queue), worker); + struct xe_device *xe = pf_queue->xe; struct xe_pagefault pf; unsigned long threshold; + /* + * A live VM holds a PM reference, but a torn-down VM does not. + * Guard the entire worker loop to safely drain stale faults and + * prevent autosuspends from desyncing batched CT flushes. + */ + guard(xe_pm_runtime)(xe); + #define USM_QUEUE_MAX_RUNTIME_MS 20 threshold = jiffies + msecs_to_jiffies(USM_QUEUE_MAX_RUNTIME_MS); @@ -365,6 +374,7 @@ static int xe_pagefault_queue_init(struct xe_device *xe, drm_dbg(&xe->drm, "xe_pagefault_entry_size=%d, total_num_eus=%d, pf_queue->size=%u", xe_pagefault_entry_size(), total_num_eus, pf_queue->size); + pf_queue->xe = xe; spin_lock_init(&pf_queue->lock); INIT_WORK(&pf_queue->worker, xe_pagefault_queue_work); diff --git a/drivers/gpu/drm/xe/xe_pagefault_types.h b/drivers/gpu/drm/xe/xe_pagefault_types.h index c4ee625b93ddd..f63a12aa0d4f7 100644 --- a/drivers/gpu/drm/xe/xe_pagefault_types.h +++ b/drivers/gpu/drm/xe/xe_pagefault_types.h @@ -8,6 +8,7 @@ #include +struct xe_device; struct xe_gt; struct xe_pagefault; @@ -118,6 +119,8 @@ struct xe_pagefault { * queue to absorb the device’s worst-case number of outstanding faults. */ struct xe_pagefault_queue { + /** @xe: Back-pointer to the Xe device */ + struct xe_device *xe; /** * @data: Data in queue containing struct xe_pagefault, protected by * @lock -- 2.53.0