From: Harish Kasiviswanathan Date: Tue, 26 Mar 2024 19:32:46 +0000 (-0400) Subject: drm/amdkfd: Reset GPU on queue preemption failure X-Git-Tag: io_uring-6.10-20240523~68^2~23^2~123 X-Git-Url: https://git.kernel.dk/?a=commitdiff_plain;h=d7e8ddc392a3659206e7973e75c47849f2c8b72a;p=linux-2.6-block.git drm/amdkfd: Reset GPU on queue preemption failure Currently, with F32 HWS GPU reset is only when unmap queue fails. However, if compute queue doesn't repond to preemption request in time unmap will return without any error. In this case, only preemption error is logged and Reset is not triggered. Call GPU reset in this case also. Reviewed-by: Alex Deucher Signed-off-by: Harish Kasiviswanathan Reviewed-by: Mukul Joshi Signed-off-by: Alex Deucher --- diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 151fabf84040..c08b6ee25289 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -2000,6 +2000,7 @@ static int unmap_queues_cpsch(struct device_queue_manager *dqm, if (mqd_mgr->check_preemption_failed(mqd_mgr, dqm->packet_mgr.priv_queue->queue->mqd)) { while (halt_if_hws_hang) schedule(); + kfd_hws_hang(dqm); return -ETIME; }