Priya Hosur 94e25cb6ab drm/amdkfd: Add TLB flush after MES queue eviction/suspension ... MES (Micro Engine Scheduler) does not perform heavy-weight TLB invalidation after unmapping queues, unlike HWS which does this automatically. This causes a race condition where in-flight DMA descriptors can access memory that has been unmapped, leading to page faults and GPU queue hangs during SVM page migration. The issue manifests as KFDSVMRangeTest.MultiThreadMigrationTest failures on gfx1151 (Strix Point) with XNACK mode 1 enabled - the GPU compute queue hangs with packets submitted but never consumed. Add kfd_flush_tlb() calls after MES queue removal in two locations: - evict_process_queues_cpsch(): after all queues removed during eviction - suspend_queues(): after debug/criu queue suspension (with mem_fence barrier) This ensures all in-flight memory accesses from unmapped queues are flushed before memory is freed or migrated. Signed-off-by: Priya Hosur <Priya.Hosur@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> (cherry picked from commit f5c4f88e0f) Cc: stable@vger.kernel.org		2026-09-02 16:20:48 -04:00
..
drm	drm/amdkfd: Add TLB flush after MES queue eviction/suspension	2026-09-02 16:20:48 -04:00
host1x	Linux 7.2	2026-08-20 10:58:44 +10:00
ipu-v3
nova-core	drm for 7.3-rc1	2026-08-21 08:41:00 -07:00
tests	drm/tests/gpu_buddy: fix interleaving in buffer clearance test	2026-08-06 18:07:10 +05:30
trace
vga
buddy.c	Linux 7.2	2026-08-20 10:58:44 +10:00
Kconfig	gpu: Fix dependencies in CONFIG_GPU_BUDDY_KUNIT_TEST	2026-07-29 13:06:24 +02:00
Makefile