Hello, This is a set of fixes for bugs I ran into while looking at [1]. Hopefully that's enough to recover from AS_ACTIVE bit stuck situations, but it'd be good to understand why the MMU block is completely blocked in some cases and try to come up with better mitigations than a full GPU reset.
Check each patch for a detailed changelog. Regards, Boris Boris Brezillon (6): drm/panthor: Always wait after sending a command to an AS drm/panthor: Kill lock_region() drm/panthor: Recover from panthor_gpu_flush_caches() failures drm/panthor: Add support for atomic page table updates drm/panthor: Make panthor_vm_[un]map_pages() more robust drm/panthor: Relax a check in panthor_sched_pre_reset() drivers/gpu/drm/panthor/panthor_gpu.c | 19 +- drivers/gpu/drm/panthor/panthor_mmu.c | 277 +++++++++++++----------- drivers/gpu/drm/panthor/panthor_sched.c | 2 - 3 files changed, 164 insertions(+), 134 deletions(-) -- 2.51.1
