In the Linux kernel, the following vulnerability has been resolved:

drm/xe: Guard page-fault worker with runtime PM check

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)

(cherry picked from commit fcc2431d2213dc4d04250c4f1ae87d9c3ae0d455) [Rodrigo: Added xe_device struct for compatibility while cherry-picking]

In the Linux kernel, the following vulnerability has been resolved:

drm/xe: Guard page-fault worker with runtime PM check

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)

(cherry picked from commit fcc2431d2213dc4d04250c4f1ae87d9c3ae0d455) [Rodrigo: Added xe_device struct for compatibility while cherry-picking]

Affected: Ubuntu:16.04:LTS: linux-hwe-edge, Ubuntu:18.04:LTS: linux-aws-5.0, Ubuntu:18.04:LTS: linux-aws-5.3, Ubuntu:18.04:LTS: linux-azure, Ubuntu:18.04:LTS: linux-azure-5.3, Ubuntu:18.04:LTS: linux-azure-edge, Ubuntu:18.04:LTS: linux-gcp, Ubuntu:18.04:LTS: linux-gcp-5.3, Ubuntu:18.04:LTS: linux-gke-4.15, Ubuntu:18.04:LTS: linux-gke-5.4 and 118 more

In the Linux kernel, the following vulnerability has been resolved: drm/xe: Guard page-fault worker with runtime PM check 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) (cherry picked from commit fcc2431d2213dc4d04250c4f1ae87d9c3ae0d455) [Rodrigo: Added xe_device struct for compatibility while cherry-picking]