diff options
| author | David Yat Sin <David.YatSin@amd.com> | 2025-07-16 22:04:28 +0000 |
|---|---|---|
| committer | Alex Deucher <alexander.deucher@amd.com> | 2025-08-04 15:38:49 -0400 |
| commit | f6c0f3d24478a0792e50a64c2eba9f34d65519f2 (patch) | |
| tree | ef3152a04e6f4f39405355fb0a42b7f03d3d1704 /drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c | |
| parent | 796ff8a7e01bd18738d3bb4111f9d6f963145d29 (diff) | |
drm/amdkfd: Fix checkpoint-restore on multi-xcc
GPUs with multi-xcc have multiple MQDs per queue. This patch saves and
restores all the MQDs within the partition.
Signed-off-by: David Yat Sin <David.YatSin@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit a578f2a58c3ab38f0643b1b6e7534af860233cb1)
Cc: stable@vger.kernel.org
Diffstat (limited to 'drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c')
| -rw-r--r-- | drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c | 20 |
1 files changed, 15 insertions, 5 deletions
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c index c643e0ccec52..7fbb5c274ccc 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c @@ -914,7 +914,10 @@ static int criu_checkpoint_queues_device(struct kfd_process_device *pdd, q_data = (struct kfd_criu_queue_priv_data *)q_private_data; - /* data stored in this order: priv_data, mqd, ctl_stack */ + /* + * data stored in this order: + * priv_data, mqd[xcc0], mqd[xcc1],..., ctl_stack[xcc0], ctl_stack[xcc1]... + */ q_data->mqd_size = mqd_size; q_data->ctl_stack_size = ctl_stack_size; @@ -963,7 +966,7 @@ int kfd_criu_checkpoint_queues(struct kfd_process *p, } static void set_queue_properties_from_criu(struct queue_properties *qp, - struct kfd_criu_queue_priv_data *q_data) + struct kfd_criu_queue_priv_data *q_data, uint32_t num_xcc) { qp->is_interop = false; qp->queue_percent = q_data->q_percent; @@ -976,7 +979,11 @@ static void set_queue_properties_from_criu(struct queue_properties *qp, qp->eop_ring_buffer_size = q_data->eop_ring_buffer_size; qp->ctx_save_restore_area_address = q_data->ctx_save_restore_area_address; qp->ctx_save_restore_area_size = q_data->ctx_save_restore_area_size; - qp->ctl_stack_size = q_data->ctl_stack_size; + if (q_data->type == KFD_QUEUE_TYPE_COMPUTE) + qp->ctl_stack_size = q_data->ctl_stack_size / num_xcc; + else + qp->ctl_stack_size = q_data->ctl_stack_size; + qp->type = q_data->type; qp->format = q_data->format; } @@ -1036,12 +1043,15 @@ int kfd_criu_restore_queue(struct kfd_process *p, goto exit; } - /* data stored in this order: mqd, ctl_stack */ + /* + * data stored in this order: + * mqd[xcc0], mqd[xcc1],..., ctl_stack[xcc0], ctl_stack[xcc1]... + */ mqd = q_extra_data; ctl_stack = mqd + q_data->mqd_size; memset(&qp, 0, sizeof(qp)); - set_queue_properties_from_criu(&qp, q_data); + set_queue_properties_from_criu(&qp, q_data, NUM_XCC(pdd->dev->adev->gfx.xcc_mask)); print_queue_properties(&qp); |