diff options
| author | Thomas Zimmermann <tzimmermann@suse.de> | 2023-10-11 09:50:59 +0200 |
|---|---|---|
| committer | Thomas Zimmermann <tzimmermann@suse.de> | 2023-10-11 09:50:59 +0200 |
| commit | 57390019b68b83f96eb98f490367b9df1f2d77cb (patch) | |
| tree | e6d4b6c75efdd2d7fb7d37f980688c491be3ff6a /drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c | |
| parent | e5f9d543419c78ac58f3b3557bc5a76b20ff600b (diff) | |
| parent | 389af786f92ecdff35883551d54bf4e507ffcccb (diff) | |
| download | linux-57390019b68b83f96eb98f490367b9df1f2d77cb.tar.gz linux-57390019b68b83f96eb98f490367b9df1f2d77cb.tar.bz2 linux-57390019b68b83f96eb98f490367b9df1f2d77cb.zip | |
Merge drm/drm-next into drm-misc-next
Updating drm-misc-next to the state of Linux v6.6-rc2.
Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
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 | 35 |
1 files changed, 18 insertions, 17 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 ba9d69054119..adb5e4bdc0b2 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c @@ -123,7 +123,7 @@ int pqm_set_gws(struct process_queue_manager *pqm, unsigned int qid, if (!gws && pdd->qpd.num_gws == 0) return -EINVAL; - if (KFD_GC_VERSION(dev) != IP_VERSION(9, 4, 3)) { + if (KFD_GC_VERSION(dev) != IP_VERSION(9, 4, 3) && !dev->kfd->shared_resources.enable_mes) { if (gws) ret = amdgpu_amdkfd_add_gws_to_process(pdd->process->kgd_process_info, gws, &mem); @@ -136,7 +136,9 @@ int pqm_set_gws(struct process_queue_manager *pqm, unsigned int qid, } else { /* * Intentionally set GWS to a non-NULL value - * for GFX 9.4.3. + * for devices that do not use GWS for global wave + * synchronization but require the formality + * of setting GWS for cooperative groups. */ pqn->q->gws = gws ? ERR_PTR(-ENOMEM) : NULL; } @@ -173,7 +175,8 @@ void pqm_uninit(struct process_queue_manager *pqm) list_for_each_entry_safe(pqn, next, &pqm->queues, process_queue_list) { if (pqn->q && pqn->q->gws && - KFD_GC_VERSION(pqn->q->device) != IP_VERSION(9, 4, 3)) + KFD_GC_VERSION(pqn->q->device) != IP_VERSION(9, 4, 3) && + !pqn->q->device->kfd->shared_resources.enable_mes) amdgpu_amdkfd_remove_gws_from_process(pqm->process->kgd_process_info, pqn->q->gws); kfd_procfs_del_queue(pqn->q); @@ -365,17 +368,20 @@ int pqm_create_queue(struct process_queue_manager *pqm, goto err_create_queue; } - if (q && p_doorbell_offset_in_process) + if (q && p_doorbell_offset_in_process) { /* Return the doorbell offset within the doorbell page * to the caller so it can be passed up to user mode * (in bytes). - * There are always 1024 doorbells per process, so in case - * of 8-byte doorbells, there are two doorbell pages per - * process. + * relative doorbell index = Absolute doorbell index - + * absolute index of first doorbell in the page. */ - *p_doorbell_offset_in_process = - (q->properties.doorbell_off * sizeof(uint32_t)) & - (kfd_doorbell_process_slice(dev->kfd) - 1); + uint32_t first_db_index = amdgpu_doorbell_index_on_bar(pdd->dev->adev, + pdd->qpd.proc_doorbells, + 0); + + *p_doorbell_offset_in_process = (q->properties.doorbell_off + - first_db_index) * sizeof(uint32_t); + } pr_debug("PQM After DQM create queue\n"); @@ -455,7 +461,8 @@ int pqm_destroy_queue(struct process_queue_manager *pqm, unsigned int qid) } if (pqn->q->gws) { - if (KFD_GC_VERSION(pqn->q->device) != IP_VERSION(9, 4, 3)) + if (KFD_GC_VERSION(pqn->q->device) != IP_VERSION(9, 4, 3) && + !dev->kfd->shared_resources.enable_mes) amdgpu_amdkfd_remove_gws_from_process( pqm->process->kgd_process_info, pqn->q->gws); @@ -929,12 +936,6 @@ int kfd_criu_restore_queue(struct kfd_process *p, goto exit; } - if (!pdd->doorbell_index && - kfd_alloc_process_doorbells(pdd->dev->kfd, &pdd->doorbell_index) < 0) { - ret = -ENOMEM; - goto exit; - } - /* data stored in this order: mqd, ctl_stack */ mqd = q_extra_data; ctl_stack = mqd + q_data->mqd_size; |