diff options
Diffstat (limited to 'drivers/gpu/drm/scheduler')
| -rw-r--r-- | drivers/gpu/drm/scheduler/gpu_scheduler_trace.h | 103 | ||||
| -rw-r--r-- | drivers/gpu/drm/scheduler/sched_entity.c | 48 | ||||
| -rw-r--r-- | drivers/gpu/drm/scheduler/sched_fence.c | 6 | ||||
| -rw-r--r-- | drivers/gpu/drm/scheduler/sched_internal.h | 2 | ||||
| -rw-r--r-- | drivers/gpu/drm/scheduler/sched_main.c | 203 | ||||
| -rw-r--r-- | drivers/gpu/drm/scheduler/tests/mock_scheduler.c | 109 | ||||
| -rw-r--r-- | drivers/gpu/drm/scheduler/tests/sched_tests.h | 3 | ||||
| -rw-r--r-- | drivers/gpu/drm/scheduler/tests/tests_basic.c | 93 |
8 files changed, 380 insertions, 187 deletions
diff --git a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h index f56e77e7f6d0..261713dd7d5a 100644 --- a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h +++ b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h @@ -32,78 +32,123 @@ #define TRACE_SYSTEM gpu_scheduler #define TRACE_INCLUDE_FILE gpu_scheduler_trace +/** + * DOC: uAPI trace events + * + * ``drm_sched_job_queue``, ``drm_sched_job_run``, ``drm_sched_job_add_dep``, + * ``drm_sched_job_done`` and ``drm_sched_job_unschedulable`` are considered + * stable uAPI. + * + * Common trace events attributes: + * + * * ``dev`` - the dev_name() of the device running the job. + * + * * ``ring`` - the hardware ring running the job. Together with ``dev`` it + * uniquely identifies where the job is going to be executed. + * + * * ``fence`` - the &struct dma_fence.context and the &struct dma_fence.seqno of + * &struct drm_sched_fence.finished + * + * All the events depends on drm_sched_job_arm() having been called already for + * the job because they use &struct drm_sched_job.sched or + * &struct drm_sched_job.s_fence. + */ + DECLARE_EVENT_CLASS(drm_sched_job, TP_PROTO(struct drm_sched_job *sched_job, struct drm_sched_entity *entity), TP_ARGS(sched_job, entity), TP_STRUCT__entry( - __field(struct drm_sched_entity *, entity) - __field(struct dma_fence *, fence) __string(name, sched_job->sched->name) - __field(uint64_t, id) __field(u32, job_count) __field(int, hw_job_count) + __string(dev, dev_name(sched_job->sched->dev)) + __field(u64, fence_context) + __field(u64, fence_seqno) + __field(u64, client_id) ), TP_fast_assign( - __entry->entity = entity; - __entry->id = sched_job->id; - __entry->fence = &sched_job->s_fence->finished; __assign_str(name); __entry->job_count = spsc_queue_count(&entity->job_queue); __entry->hw_job_count = atomic_read( &sched_job->sched->credit_count); + __assign_str(dev); + __entry->fence_context = sched_job->s_fence->finished.context; + __entry->fence_seqno = sched_job->s_fence->finished.seqno; + __entry->client_id = sched_job->s_fence->drm_client_id; ), - TP_printk("entity=%p, id=%llu, fence=%p, ring=%s, job count:%u, hw job count:%d", - __entry->entity, __entry->id, - __entry->fence, __get_str(name), - __entry->job_count, __entry->hw_job_count) + TP_printk("dev=%s, fence=%llu:%llu, ring=%s, job count:%u, hw job count:%d, client_id:%llu", + __get_str(dev), + __entry->fence_context, __entry->fence_seqno, __get_str(name), + __entry->job_count, __entry->hw_job_count, __entry->client_id) ); -DEFINE_EVENT(drm_sched_job, drm_sched_job, +DEFINE_EVENT(drm_sched_job, drm_sched_job_queue, TP_PROTO(struct drm_sched_job *sched_job, struct drm_sched_entity *entity), TP_ARGS(sched_job, entity) ); -DEFINE_EVENT(drm_sched_job, drm_run_job, +DEFINE_EVENT(drm_sched_job, drm_sched_job_run, TP_PROTO(struct drm_sched_job *sched_job, struct drm_sched_entity *entity), TP_ARGS(sched_job, entity) ); -TRACE_EVENT(drm_sched_process_job, +TRACE_EVENT(drm_sched_job_done, TP_PROTO(struct drm_sched_fence *fence), TP_ARGS(fence), TP_STRUCT__entry( - __field(struct dma_fence *, fence) + __field(u64, fence_context) + __field(u64, fence_seqno) ), TP_fast_assign( - __entry->fence = &fence->finished; + __entry->fence_context = fence->finished.context; + __entry->fence_seqno = fence->finished.seqno; ), - TP_printk("fence=%p signaled", __entry->fence) + TP_printk("fence=%llu:%llu signaled", + __entry->fence_context, __entry->fence_seqno) ); -TRACE_EVENT(drm_sched_job_wait_dep, +TRACE_EVENT(drm_sched_job_add_dep, + TP_PROTO(struct drm_sched_job *sched_job, struct dma_fence *fence), + TP_ARGS(sched_job, fence), + TP_STRUCT__entry( + __field(u64, fence_context) + __field(u64, fence_seqno) + __field(u64, ctx) + __field(u64, seqno) + ), + + TP_fast_assign( + __entry->fence_context = sched_job->s_fence->finished.context; + __entry->fence_seqno = sched_job->s_fence->finished.seqno; + __entry->ctx = fence->context; + __entry->seqno = fence->seqno; + ), + TP_printk("fence=%llu:%llu depends on fence=%llu:%llu", + __entry->fence_context, __entry->fence_seqno, + __entry->ctx, __entry->seqno) +); + +TRACE_EVENT(drm_sched_job_unschedulable, TP_PROTO(struct drm_sched_job *sched_job, struct dma_fence *fence), TP_ARGS(sched_job, fence), TP_STRUCT__entry( - __string(name, sched_job->sched->name) - __field(uint64_t, id) - __field(struct dma_fence *, fence) - __field(uint64_t, ctx) - __field(unsigned, seqno) + __field(u64, fence_context) + __field(u64, fence_seqno) + __field(u64, ctx) + __field(u64, seqno) ), TP_fast_assign( - __assign_str(name); - __entry->id = sched_job->id; - __entry->fence = fence; + __entry->fence_context = sched_job->s_fence->finished.context; + __entry->fence_seqno = sched_job->s_fence->finished.seqno; __entry->ctx = fence->context; __entry->seqno = fence->seqno; ), - TP_printk("job ring=%s, id=%llu, depends fence=%p, context=%llu, seq=%u", - __get_str(name), __entry->id, - __entry->fence, __entry->ctx, - __entry->seqno) + TP_printk("fence=%llu:%llu depends on unsignalled fence=%llu:%llu", + __entry->fence_context, __entry->fence_seqno, + __entry->ctx, __entry->seqno) ); #endif /* _GPU_SCHED_TRACE_H_ */ diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm/scheduler/sched_entity.c index bd39db7bb240..8867b95ab089 100644 --- a/drivers/gpu/drm/scheduler/sched_entity.c +++ b/drivers/gpu/drm/scheduler/sched_entity.c @@ -21,7 +21,7 @@ * */ -#include <linux/kthread.h> +#include <linux/export.h> #include <linux/slab.h> #include <linux/completion.h> @@ -176,6 +176,7 @@ static void drm_sched_entity_kill_jobs_work(struct work_struct *wrk) { struct drm_sched_job *job = container_of(wrk, typeof(*job), work); + drm_sched_fence_scheduled(job->s_fence, NULL); drm_sched_fence_finished(job->s_fence, -ESRCH); WARN_ON(job->s_fence->parent); job->sched->ops->free_job(job); @@ -354,17 +355,6 @@ void drm_sched_entity_destroy(struct drm_sched_entity *entity) } EXPORT_SYMBOL(drm_sched_entity_destroy); -/* drm_sched_entity_clear_dep - callback to clear the entities dependency */ -static void drm_sched_entity_clear_dep(struct dma_fence *f, - struct dma_fence_cb *cb) -{ - struct drm_sched_entity *entity = - container_of(cb, struct drm_sched_entity, cb); - - entity->dependency = NULL; - dma_fence_put(f); -} - /* * drm_sched_entity_wakeup - callback to clear the entity's dependency and * wake up the scheduler @@ -375,7 +365,8 @@ static void drm_sched_entity_wakeup(struct dma_fence *f, struct drm_sched_entity *entity = container_of(cb, struct drm_sched_entity, cb); - drm_sched_entity_clear_dep(f, cb); + entity->dependency = NULL; + dma_fence_put(f); drm_sched_wakeup(entity->rq->sched); } @@ -428,13 +419,6 @@ static bool drm_sched_entity_add_dependency_cb(struct drm_sched_entity *entity) fence = dma_fence_get(&s_fence->scheduled); dma_fence_put(entity->dependency); entity->dependency = fence; - if (!dma_fence_add_callback(fence, &entity->cb, - drm_sched_entity_clear_dep)) - return true; - - /* Ignore it when it is already scheduled */ - dma_fence_put(fence); - return false; } if (!dma_fence_add_callback(entity->dependency, &entity->cb, @@ -477,10 +461,10 @@ struct drm_sched_job *drm_sched_entity_pop_job(struct drm_sched_entity *entity) while ((entity->dependency = drm_sched_job_dependency(sched_job, entity))) { - trace_drm_sched_job_wait_dep(sched_job, entity->dependency); - - if (drm_sched_entity_add_dependency_cb(entity)) + if (drm_sched_entity_add_dependency_cb(entity)) { + trace_drm_sched_job_unschedulable(sched_job, entity->dependency); return NULL; + } } /* skip jobs from entity that marked guilty */ @@ -545,10 +529,10 @@ void drm_sched_entity_select_rq(struct drm_sched_entity *entity) return; /* - * Only when the queue is empty are we guaranteed that the scheduler - * thread cannot change ->last_scheduled. To enforce ordering we need - * a read barrier here. See drm_sched_entity_pop_job() for the other - * side. + * Only when the queue is empty are we guaranteed that + * drm_sched_run_job_work() cannot change entity->last_scheduled. To + * enforce ordering we need a read barrier here. See + * drm_sched_entity_pop_job() for the other side. */ smp_rmb(); @@ -586,7 +570,15 @@ void drm_sched_entity_push_job(struct drm_sched_job *sched_job) bool first; ktime_t submit_ts; - trace_drm_sched_job(sched_job, entity); + trace_drm_sched_job_queue(sched_job, entity); + + if (trace_drm_sched_job_add_dep_enabled()) { + struct dma_fence *entry; + unsigned long index; + + xa_for_each(&sched_job->dependencies, index, entry) + trace_drm_sched_job_add_dep(sched_job, entry); + } atomic_inc(entity->rq->sched->score); WRITE_ONCE(entity->last_user, current->group_leader); diff --git a/drivers/gpu/drm/scheduler/sched_fence.c b/drivers/gpu/drm/scheduler/sched_fence.c index e971528504a5..9391d6f0dc01 100644 --- a/drivers/gpu/drm/scheduler/sched_fence.c +++ b/drivers/gpu/drm/scheduler/sched_fence.c @@ -21,7 +21,7 @@ * */ -#include <linux/kthread.h> +#include <linux/export.h> #include <linux/module.h> #include <linux/sched.h> #include <linux/slab.h> @@ -206,7 +206,8 @@ struct drm_sched_fence *to_drm_sched_fence(struct dma_fence *f) EXPORT_SYMBOL(to_drm_sched_fence); struct drm_sched_fence *drm_sched_fence_alloc(struct drm_sched_entity *entity, - void *owner) + void *owner, + u64 drm_client_id) { struct drm_sched_fence *fence = NULL; @@ -215,6 +216,7 @@ struct drm_sched_fence *drm_sched_fence_alloc(struct drm_sched_entity *entity, return NULL; fence->owner = owner; + fence->drm_client_id = drm_client_id; spin_lock_init(&fence->lock); return fence; diff --git a/drivers/gpu/drm/scheduler/sched_internal.h b/drivers/gpu/drm/scheduler/sched_internal.h index 599cf6e1bb74..7ea5a6736f98 100644 --- a/drivers/gpu/drm/scheduler/sched_internal.h +++ b/drivers/gpu/drm/scheduler/sched_internal.h @@ -24,7 +24,7 @@ void drm_sched_entity_select_rq(struct drm_sched_entity *entity); struct drm_sched_job *drm_sched_entity_pop_job(struct drm_sched_entity *entity); struct drm_sched_fence *drm_sched_fence_alloc(struct drm_sched_entity *s_entity, - void *owner); + void *owner, u64 drm_client_id); void drm_sched_fence_init(struct drm_sched_fence *fence, struct drm_sched_entity *entity); void drm_sched_fence_free(struct drm_sched_fence *fence); diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c index 829579c41c6b..e2cda28a1af4 100644 --- a/drivers/gpu/drm/scheduler/sched_main.c +++ b/drivers/gpu/drm/scheduler/sched_main.c @@ -66,6 +66,7 @@ * This implies waiting for previously executed jobs. */ +#include <linux/export.h> #include <linux/wait.h> #include <linux/sched.h> #include <linux/completion.h> @@ -83,12 +84,6 @@ #define CREATE_TRACE_POINTS #include "gpu_scheduler_trace.h" -#ifdef CONFIG_LOCKDEP -static struct lockdep_map drm_sched_lockdep_map = { - .name = "drm_sched_lockdep_map" -}; -#endif - int drm_sched_policy = DRM_SCHED_POLICY_FIFO; /** @@ -268,38 +263,14 @@ drm_sched_rq_select_entity_rr(struct drm_gpu_scheduler *sched, entity = rq->current_entity; if (entity) { list_for_each_entry_continue(entity, &rq->entities, list) { - if (drm_sched_entity_is_ready(entity)) { - /* If we can't queue yet, preserve the current - * entity in terms of fairness. - */ - if (!drm_sched_can_queue(sched, entity)) { - spin_unlock(&rq->lock); - return ERR_PTR(-ENOSPC); - } - - rq->current_entity = entity; - reinit_completion(&entity->entity_idle); - spin_unlock(&rq->lock); - return entity; - } + if (drm_sched_entity_is_ready(entity)) + goto found; } } list_for_each_entry(entity, &rq->entities, list) { - if (drm_sched_entity_is_ready(entity)) { - /* If we can't queue yet, preserve the current entity in - * terms of fairness. - */ - if (!drm_sched_can_queue(sched, entity)) { - spin_unlock(&rq->lock); - return ERR_PTR(-ENOSPC); - } - - rq->current_entity = entity; - reinit_completion(&entity->entity_idle); - spin_unlock(&rq->lock); - return entity; - } + if (drm_sched_entity_is_ready(entity)) + goto found; if (entity == rq->current_entity) break; @@ -308,6 +279,22 @@ drm_sched_rq_select_entity_rr(struct drm_gpu_scheduler *sched, spin_unlock(&rq->lock); return NULL; + +found: + if (!drm_sched_can_queue(sched, entity)) { + /* + * If scheduler cannot take more jobs signal the caller to not + * consider lower priority queues. + */ + entity = ERR_PTR(-ENOSPC); + } else { + rq->current_entity = entity; + reinit_completion(&entity->entity_idle); + } + + spin_unlock(&rq->lock); + + return entity; } /** @@ -379,11 +366,16 @@ static void drm_sched_run_free_queue(struct drm_gpu_scheduler *sched) { struct drm_sched_job *job; - spin_lock(&sched->job_list_lock); job = list_first_entry_or_null(&sched->pending_list, struct drm_sched_job, list); if (job && dma_fence_is_signaled(&job->s_fence->finished)) __drm_sched_run_free_queue(sched); +} + +static void drm_sched_run_free_queue_unlocked(struct drm_gpu_scheduler *sched) +{ + spin_lock(&sched->job_list_lock); + drm_sched_run_free_queue(sched); spin_unlock(&sched->job_list_lock); } @@ -391,7 +383,7 @@ static void drm_sched_run_free_queue(struct drm_gpu_scheduler *sched) * drm_sched_job_done - complete a job * @s_job: pointer to the job which is done * - * Finish the job's fence and wake up the worker thread. + * Finish the job's fence and resubmit the work items. */ static void drm_sched_job_done(struct drm_sched_job *s_job, int result) { @@ -401,7 +393,7 @@ static void drm_sched_job_done(struct drm_sched_job *s_job, int result) atomic_sub(s_job->credits, &sched->credit_count); atomic_dec(sched->score); - trace_drm_sched_process_job(s_fence); + trace_drm_sched_job_done(s_fence); dma_fence_get(&s_fence->finished); drm_sched_fence_finished(s_fence, result); @@ -536,11 +528,37 @@ static void drm_sched_job_begin(struct drm_sched_job *s_job) spin_unlock(&sched->job_list_lock); } +/** + * drm_sched_job_reinsert_on_false_timeout - reinsert the job on a false timeout + * @sched: scheduler instance + * @job: job to be reinserted on the pending list + * + * In the case of a "false timeout" - when a timeout occurs but the GPU isn't + * hung and is making progress, the scheduler must reinsert the job back into + * @sched->pending_list. Otherwise, the job and its resources won't be freed + * through the &struct drm_sched_backend_ops.free_job callback. + * + * This function must be used in "false timeout" cases only. + */ +static void drm_sched_job_reinsert_on_false_timeout(struct drm_gpu_scheduler *sched, + struct drm_sched_job *job) +{ + spin_lock(&sched->job_list_lock); + list_add(&job->list, &sched->pending_list); + + /* After reinserting the job, the scheduler enqueues the free-job work + * again if ready. Otherwise, a signaled job could be added to the + * pending list, but never freed. + */ + drm_sched_run_free_queue(sched); + spin_unlock(&sched->job_list_lock); +} + static void drm_sched_job_timedout(struct work_struct *work) { struct drm_gpu_scheduler *sched; struct drm_sched_job *job; - enum drm_gpu_sched_stat status = DRM_GPU_SCHED_STAT_NOMINAL; + enum drm_gpu_sched_stat status = DRM_GPU_SCHED_STAT_RESET; sched = container_of(work, struct drm_gpu_scheduler, work_tdr.work); @@ -551,9 +569,10 @@ static void drm_sched_job_timedout(struct work_struct *work) if (job) { /* - * Remove the bad job so it cannot be freed by concurrent - * drm_sched_cleanup_jobs. It will be reinserted back after sched->thread - * is parked at which point it's safe. + * Remove the bad job so it cannot be freed by a concurrent + * &struct drm_sched_backend_ops.free_job. It will be + * reinserted after the scheduler's work items have been + * cancelled, at which point it's safe. */ list_del_init(&job->list); spin_unlock(&sched->job_list_lock); @@ -568,6 +587,9 @@ static void drm_sched_job_timedout(struct work_struct *work) job->sched->ops->free_job(job); sched->free_guilty = false; } + + if (status == DRM_GPU_SCHED_STAT_NO_HANG) + drm_sched_job_reinsert_on_false_timeout(sched, job); } else { spin_unlock(&sched->job_list_lock); } @@ -590,6 +612,10 @@ static void drm_sched_job_timedout(struct work_struct *work) * This function is typically used for reset recovery (see the docu of * drm_sched_backend_ops.timedout_job() for details). Do not call it for * scheduler teardown, i.e., before calling drm_sched_fini(). + * + * As it's only used for reset recovery, drivers must not call this function + * in their &struct drm_sched_backend_ops.timedout_job callback when they + * skip a reset using &enum drm_gpu_sched_stat.DRM_GPU_SCHED_STAT_NO_HANG. */ void drm_sched_stop(struct drm_gpu_scheduler *sched, struct drm_sched_job *bad) { @@ -599,10 +625,10 @@ void drm_sched_stop(struct drm_gpu_scheduler *sched, struct drm_sched_job *bad) /* * Reinsert back the bad job here - now it's safe as - * drm_sched_get_finished_job cannot race against us and release the + * drm_sched_get_finished_job() cannot race against us and release the * bad job at this point - we parked (waited for) any in progress - * (earlier) cleanups and drm_sched_get_finished_job will not be called - * now until the scheduler thread is unparked. + * (earlier) cleanups and drm_sched_get_finished_job() will not be + * called now until the scheduler's work items are submitted again. */ if (bad && bad->sched == sched) /* @@ -615,7 +641,8 @@ void drm_sched_stop(struct drm_gpu_scheduler *sched, struct drm_sched_job *bad) * Iterate the job list from later to earlier one and either deactive * their HW callbacks or remove them from pending list if they already * signaled. - * This iteration is thread safe as sched thread is stopped. + * This iteration is thread safe as the scheduler's work items have been + * cancelled. */ list_for_each_entry_safe_reverse(s_job, tmp, &sched->pending_list, list) { @@ -674,15 +701,19 @@ EXPORT_SYMBOL(drm_sched_stop); * drm_sched_backend_ops.timedout_job() for details). Do not call it for * scheduler startup. The scheduler itself is fully operational after * drm_sched_init() succeeded. + * + * As it's only used for reset recovery, drivers must not call this function + * in their &struct drm_sched_backend_ops.timedout_job callback when they + * skip a reset using &enum drm_gpu_sched_stat.DRM_GPU_SCHED_STAT_NO_HANG. */ void drm_sched_start(struct drm_gpu_scheduler *sched, int errno) { struct drm_sched_job *s_job, *tmp; /* - * Locking the list is not required here as the sched thread is parked - * so no new jobs are being inserted or removed. Also concurrent - * GPU recovers can't run in parallel. + * Locking the list is not required here as the scheduler's work items + * are currently not running, so no new jobs are being inserted or + * removed. Also concurrent GPU recovers can't run in parallel. */ list_for_each_entry_safe(s_job, tmp, &sched->pending_list, list) { struct dma_fence *fence = s_job->s_fence->parent; @@ -764,6 +795,8 @@ EXPORT_SYMBOL(drm_sched_resubmit_jobs); * @credits: the number of credits this job contributes to the schedulers * credit limit * @owner: job owner for debugging + * @drm_client_id: &struct drm_file.client_id of the owner (used by trace + * events) * * Refer to drm_sched_entity_push_job() documentation * for locking considerations. @@ -784,7 +817,8 @@ EXPORT_SYMBOL(drm_sched_resubmit_jobs); */ int drm_sched_job_init(struct drm_sched_job *job, struct drm_sched_entity *entity, - u32 credits, void *owner) + u32 credits, void *owner, + uint64_t drm_client_id) { if (!entity->rq) { /* This will most likely be followed by missing frames @@ -810,7 +844,7 @@ int drm_sched_job_init(struct drm_sched_job *job, job->entity = entity; job->credits = credits; - job->s_fence = drm_sched_fence_alloc(entity, owner); + job->s_fence = drm_sched_fence_alloc(entity, owner, drm_client_id); if (!job->s_fence) return -ENOMEM; @@ -850,7 +884,6 @@ void drm_sched_job_arm(struct drm_sched_job *job) job->sched = sched; job->s_priority = entity->priority; - job->id = atomic64_inc_return(&sched->job_id_count); drm_sched_fence_init(job->s_fence, job->entity); } @@ -1193,7 +1226,7 @@ static void drm_sched_free_job_work(struct work_struct *w) if (job) sched->ops->free_job(job); - drm_sched_run_free_queue(sched); + drm_sched_run_free_queue_unlocked(sched); drm_sched_run_job_queue(sched); } @@ -1229,7 +1262,7 @@ static void drm_sched_run_job_work(struct work_struct *w) atomic_add(sched_job->credits, &sched->credit_count); drm_sched_job_begin(sched_job); - trace_drm_run_job(sched_job, entity); + trace_drm_sched_job_run(sched_job, entity); /* * The run_job() callback must by definition return a fence whose * refcount has been incremented for the scheduler already. @@ -1256,6 +1289,25 @@ static void drm_sched_run_job_work(struct work_struct *w) drm_sched_run_job_queue(sched); } +static struct workqueue_struct *drm_sched_alloc_wq(const char *name) +{ +#if (IS_ENABLED(CONFIG_LOCKDEP)) + static struct lockdep_map map = { + .name = "drm_sched_lockdep_map" + }; + + /* + * Avoid leaking a lockdep map on each drm sched creation and + * destruction by using a single lockdep map for all drm sched + * allocated submit_wq. + */ + + return alloc_ordered_workqueue_lockdep_map(name, WQ_MEM_RECLAIM, &map); +#else + return alloc_ordered_workqueue(name, WQ_MEM_RECLAIM); +#endif +} + /** * drm_sched_init - Init a gpu scheduler instance * @@ -1296,13 +1348,7 @@ int drm_sched_init(struct drm_gpu_scheduler *sched, const struct drm_sched_init_ sched->submit_wq = args->submit_wq; sched->own_submit_wq = false; } else { -#ifdef CONFIG_LOCKDEP - sched->submit_wq = alloc_ordered_workqueue_lockdep_map(args->name, - WQ_MEM_RECLAIM, - &drm_sched_lockdep_map); -#else - sched->submit_wq = alloc_ordered_workqueue(args->name, WQ_MEM_RECLAIM); -#endif + sched->submit_wq = drm_sched_alloc_wq(args->name); if (!sched->submit_wq) return -ENOMEM; @@ -1348,6 +1394,18 @@ Out_check_own: } EXPORT_SYMBOL(drm_sched_init); +static void drm_sched_cancel_remaining_jobs(struct drm_gpu_scheduler *sched) +{ + struct drm_sched_job *job, *tmp; + + /* All other accessors are stopped. No locking necessary. */ + list_for_each_entry_safe_reverse(job, tmp, &sched->pending_list, list) { + sched->ops->cancel_job(job); + list_del(&job->list); + sched->ops->free_job(job); + } +} + /** * drm_sched_fini - Destroy a gpu scheduler * @@ -1355,19 +1413,11 @@ EXPORT_SYMBOL(drm_sched_init); * * Tears down and cleans up the scheduler. * - * This stops submission of new jobs to the hardware through - * drm_sched_backend_ops.run_job(). Consequently, drm_sched_backend_ops.free_job() - * will not be called for all jobs still in drm_gpu_scheduler.pending_list. - * There is no solution for this currently. Thus, it is up to the driver to make - * sure that: - * - * a) drm_sched_fini() is only called after for all submitted jobs - * drm_sched_backend_ops.free_job() has been called or that - * b) the jobs for which drm_sched_backend_ops.free_job() has not been called - * after drm_sched_fini() ran are freed manually. - * - * FIXME: Take care of the above problem and prevent this function from leaking - * the jobs in drm_gpu_scheduler.pending_list under any circumstances. + * This stops submission of new jobs to the hardware through &struct + * drm_sched_backend_ops.run_job. If &struct drm_sched_backend_ops.cancel_job + * is implemented, all jobs will be canceled through it and afterwards cleaned + * up through &struct drm_sched_backend_ops.free_job. If cancel_job is not + * implemented, memory could leak. */ void drm_sched_fini(struct drm_gpu_scheduler *sched) { @@ -1397,11 +1447,18 @@ void drm_sched_fini(struct drm_gpu_scheduler *sched) /* Confirm no work left behind accessing device structures */ cancel_delayed_work_sync(&sched->work_tdr); + /* Avoid memory leaks if supported by the driver. */ + if (sched->ops->cancel_job) + drm_sched_cancel_remaining_jobs(sched); + if (sched->own_submit_wq) destroy_workqueue(sched->submit_wq); sched->ready = false; kfree(sched->sched_rq); sched->sched_rq = NULL; + + if (!list_empty(&sched->pending_list)) + dev_warn(sched->dev, "Tearing down scheduler while jobs are pending!\n"); } EXPORT_SYMBOL(drm_sched_fini); diff --git a/drivers/gpu/drm/scheduler/tests/mock_scheduler.c b/drivers/gpu/drm/scheduler/tests/mock_scheduler.c index f999c8859cf7..65acffc3fea8 100644 --- a/drivers/gpu/drm/scheduler/tests/mock_scheduler.c +++ b/drivers/gpu/drm/scheduler/tests/mock_scheduler.c @@ -63,8 +63,8 @@ static void drm_mock_sched_job_complete(struct drm_mock_sched_job *job) lockdep_assert_held(&sched->lock); job->flags |= DRM_MOCK_SCHED_JOB_DONE; - list_move_tail(&job->link, &sched->done_list); - dma_fence_signal(&job->hw_fence); + list_del(&job->link); + dma_fence_signal_locked(&job->hw_fence); complete(&job->done); } @@ -117,13 +117,13 @@ drm_mock_sched_job_new(struct kunit *test, ret = drm_sched_job_init(&job->base, &entity->base, 1, - NULL); + NULL, + 1); KUNIT_ASSERT_EQ(test, ret, 0); job->test = test; init_completion(&job->done); - spin_lock_init(&job->lock); INIT_LIST_HEAD(&job->link); hrtimer_setup(&job->timer, drm_mock_sched_job_signal_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS); @@ -169,7 +169,7 @@ static struct dma_fence *mock_sched_run_job(struct drm_sched_job *sched_job) dma_fence_init(&job->hw_fence, &drm_mock_sched_hw_fence_ops, - &job->lock, + &sched->lock, sched->hw_timeline.context, atomic_inc_return(&sched->hw_timeline.next_seqno)); @@ -200,38 +200,82 @@ static struct dma_fence *mock_sched_run_job(struct drm_sched_job *sched_job) return &job->hw_fence; } +/* + * Normally, drivers would take appropriate measures in this callback, such as + * killing the entity the faulty job is associated with, resetting the hardware + * and / or resubmitting non-faulty jobs. + * + * For the mock scheduler, there are no hardware rings to be resetted nor jobs + * to be resubmitted. Thus, this function merely ensures that + * a) timedout fences get signaled properly and removed from the pending list + * b) the mock scheduler framework gets informed about the timeout via a flag + * c) The drm_sched_job, not longer needed, gets freed + */ static enum drm_gpu_sched_stat mock_sched_timedout_job(struct drm_sched_job *sched_job) { + struct drm_mock_scheduler *sched = drm_sched_to_mock_sched(sched_job->sched); struct drm_mock_sched_job *job = drm_sched_job_to_mock_job(sched_job); + unsigned long flags; + + if (job->flags & DRM_MOCK_SCHED_JOB_DONT_RESET) { + job->flags &= ~DRM_MOCK_SCHED_JOB_DONT_RESET; + return DRM_GPU_SCHED_STAT_NO_HANG; + } - job->flags |= DRM_MOCK_SCHED_JOB_TIMEDOUT; + spin_lock_irqsave(&sched->lock, flags); + if (!dma_fence_is_signaled_locked(&job->hw_fence)) { + list_del(&job->link); + job->flags |= DRM_MOCK_SCHED_JOB_TIMEDOUT; + dma_fence_set_error(&job->hw_fence, -ETIMEDOUT); + dma_fence_signal_locked(&job->hw_fence); + } + spin_unlock_irqrestore(&sched->lock, flags); + + dma_fence_put(&job->hw_fence); + drm_sched_job_cleanup(sched_job); + /* Mock job itself is freed by the kunit framework. */ - return DRM_GPU_SCHED_STAT_NOMINAL; + return DRM_GPU_SCHED_STAT_RESET; } static void mock_sched_free_job(struct drm_sched_job *sched_job) { - struct drm_mock_scheduler *sched = - drm_sched_to_mock_sched(sched_job->sched); struct drm_mock_sched_job *job = drm_sched_job_to_mock_job(sched_job); - unsigned long flags; - /* Remove from the scheduler done list. */ - spin_lock_irqsave(&sched->lock, flags); - list_del(&job->link); - spin_unlock_irqrestore(&sched->lock, flags); dma_fence_put(&job->hw_fence); - drm_sched_job_cleanup(sched_job); /* Mock job itself is freed by the kunit framework. */ } +static void mock_sched_cancel_job(struct drm_sched_job *sched_job) +{ + struct drm_mock_scheduler *sched = drm_sched_to_mock_sched(sched_job->sched); + struct drm_mock_sched_job *job = drm_sched_job_to_mock_job(sched_job); + unsigned long flags; + + hrtimer_cancel(&job->timer); + + spin_lock_irqsave(&sched->lock, flags); + if (!dma_fence_is_signaled_locked(&job->hw_fence)) { + list_del(&job->link); + dma_fence_set_error(&job->hw_fence, -ECANCELED); + dma_fence_signal_locked(&job->hw_fence); + } + spin_unlock_irqrestore(&sched->lock, flags); + + /* + * The GPU Scheduler will call drm_sched_backend_ops.free_job(), still. + * Mock job itself is freed by the kunit framework. + */ +} + static const struct drm_sched_backend_ops drm_mock_scheduler_ops = { .run_job = mock_sched_run_job, .timedout_job = mock_sched_timedout_job, - .free_job = mock_sched_free_job + .free_job = mock_sched_free_job, + .cancel_job = mock_sched_cancel_job, }; /** @@ -265,7 +309,6 @@ struct drm_mock_scheduler *drm_mock_sched_new(struct kunit *test, long timeout) sched->hw_timeline.context = dma_fence_context_alloc(1); atomic_set(&sched->hw_timeline.next_seqno, 0); INIT_LIST_HEAD(&sched->job_list); - INIT_LIST_HEAD(&sched->done_list); spin_lock_init(&sched->lock); return sched; @@ -280,38 +323,6 @@ struct drm_mock_scheduler *drm_mock_sched_new(struct kunit *test, long timeout) */ void drm_mock_sched_fini(struct drm_mock_scheduler *sched) { - struct drm_mock_sched_job *job, *next; - unsigned long flags; - LIST_HEAD(list); - - drm_sched_wqueue_stop(&sched->base); - - /* Force complete all unfinished jobs. */ - spin_lock_irqsave(&sched->lock, flags); - list_for_each_entry_safe(job, next, &sched->job_list, link) - list_move_tail(&job->link, &list); - spin_unlock_irqrestore(&sched->lock, flags); - - list_for_each_entry(job, &list, link) - hrtimer_cancel(&job->timer); - - spin_lock_irqsave(&sched->lock, flags); - list_for_each_entry_safe(job, next, &list, link) - drm_mock_sched_job_complete(job); - spin_unlock_irqrestore(&sched->lock, flags); - - /* - * Free completed jobs and jobs not yet processed by the DRM scheduler - * free worker. - */ - spin_lock_irqsave(&sched->lock, flags); - list_for_each_entry_safe(job, next, &sched->done_list, link) - list_move_tail(&job->link, &list); - spin_unlock_irqrestore(&sched->lock, flags); - - list_for_each_entry_safe(job, next, &list, link) - mock_sched_free_job(&job->base); - drm_sched_fini(&sched->base); } diff --git a/drivers/gpu/drm/scheduler/tests/sched_tests.h b/drivers/gpu/drm/scheduler/tests/sched_tests.h index 27caf8285fb7..63d4f2ac7074 100644 --- a/drivers/gpu/drm/scheduler/tests/sched_tests.h +++ b/drivers/gpu/drm/scheduler/tests/sched_tests.h @@ -49,7 +49,6 @@ struct drm_mock_scheduler { spinlock_t lock; struct list_head job_list; - struct list_head done_list; struct { u64 context; @@ -98,6 +97,7 @@ struct drm_mock_sched_job { #define DRM_MOCK_SCHED_JOB_DONE 0x1 #define DRM_MOCK_SCHED_JOB_TIMEDOUT 0x2 +#define DRM_MOCK_SCHED_JOB_DONT_RESET 0x4 unsigned long flags; struct list_head link; @@ -106,7 +106,6 @@ struct drm_mock_sched_job { unsigned int duration_us; ktime_t finish_at; - spinlock_t lock; struct dma_fence hw_fence; struct kunit *test; diff --git a/drivers/gpu/drm/scheduler/tests/tests_basic.c b/drivers/gpu/drm/scheduler/tests/tests_basic.c index 7230057e0594..55eb142bd7c5 100644 --- a/drivers/gpu/drm/scheduler/tests/tests_basic.c +++ b/drivers/gpu/drm/scheduler/tests/tests_basic.c @@ -5,6 +5,8 @@ #include "sched_tests.h" +#define MOCK_TIMEOUT (HZ / 5) + /* * DRM scheduler basic tests should check the basic functional correctness of * the scheduler, including some very light smoke testing. More targeted tests, @@ -28,7 +30,7 @@ static void drm_sched_basic_exit(struct kunit *test) static int drm_sched_timeout_init(struct kunit *test) { - test->priv = drm_mock_sched_new(test, HZ); + test->priv = drm_mock_sched_new(test, MOCK_TIMEOUT); return 0; } @@ -204,6 +206,47 @@ static struct kunit_suite drm_sched_basic = { .test_cases = drm_sched_basic_tests, }; +static void drm_sched_basic_cancel(struct kunit *test) +{ + struct drm_mock_sched_entity *entity; + struct drm_mock_scheduler *sched; + struct drm_mock_sched_job *job; + bool done; + + /* + * Check that drm_sched_fini() uses the cancel_job() callback to cancel + * jobs that are still pending. + */ + + sched = drm_mock_sched_new(test, MAX_SCHEDULE_TIMEOUT); + entity = drm_mock_sched_entity_new(test, DRM_SCHED_PRIORITY_NORMAL, + sched); + + job = drm_mock_sched_job_new(test, entity); + + drm_mock_sched_job_submit(job); + + done = drm_mock_sched_job_wait_scheduled(job, HZ); + KUNIT_ASSERT_TRUE(test, done); + + drm_mock_sched_entity_free(entity); + drm_mock_sched_fini(sched); + + KUNIT_ASSERT_EQ(test, job->hw_fence.error, -ECANCELED); +} + +static struct kunit_case drm_sched_cancel_tests[] = { + KUNIT_CASE(drm_sched_basic_cancel), + {} +}; + +static struct kunit_suite drm_sched_cancel = { + .name = "drm_sched_basic_cancel_tests", + .init = drm_sched_basic_init, + .exit = drm_sched_basic_exit, + .test_cases = drm_sched_cancel_tests, +}; + static void drm_sched_basic_timeout(struct kunit *test) { struct drm_mock_scheduler *sched = test->priv; @@ -227,14 +270,14 @@ static void drm_sched_basic_timeout(struct kunit *test) done = drm_mock_sched_job_wait_scheduled(job, HZ); KUNIT_ASSERT_TRUE(test, done); - done = drm_mock_sched_job_wait_finished(job, HZ / 2); + done = drm_mock_sched_job_wait_finished(job, MOCK_TIMEOUT / 2); KUNIT_ASSERT_FALSE(test, done); KUNIT_ASSERT_EQ(test, job->flags & DRM_MOCK_SCHED_JOB_TIMEDOUT, 0); - done = drm_mock_sched_job_wait_finished(job, HZ); + done = drm_mock_sched_job_wait_finished(job, MOCK_TIMEOUT); KUNIT_ASSERT_FALSE(test, done); KUNIT_ASSERT_EQ(test, @@ -244,8 +287,51 @@ static void drm_sched_basic_timeout(struct kunit *test) drm_mock_sched_entity_free(entity); } +static void drm_sched_skip_reset(struct kunit *test) +{ + struct drm_mock_scheduler *sched = test->priv; + struct drm_mock_sched_entity *entity; + struct drm_mock_sched_job *job; + unsigned int i; + bool done; + + /* + * Submit a single job against a scheduler with the timeout configured + * and verify that if the job is still running, the timeout handler + * will skip the reset and allow the job to complete. + */ + + entity = drm_mock_sched_entity_new(test, + DRM_SCHED_PRIORITY_NORMAL, + sched); + job = drm_mock_sched_job_new(test, entity); + + job->flags = DRM_MOCK_SCHED_JOB_DONT_RESET; + + drm_mock_sched_job_submit(job); + + done = drm_mock_sched_job_wait_scheduled(job, HZ); + KUNIT_ASSERT_TRUE(test, done); + + done = drm_mock_sched_job_wait_finished(job, 2 * MOCK_TIMEOUT); + KUNIT_ASSERT_FALSE(test, done); + + KUNIT_ASSERT_EQ(test, + job->flags & DRM_MOCK_SCHED_JOB_DONT_RESET, + 0); + + i = drm_mock_sched_advance(sched, 1); + KUNIT_ASSERT_EQ(test, i, 1); + + done = drm_mock_sched_job_wait_finished(job, HZ); + KUNIT_ASSERT_TRUE(test, done); + + drm_mock_sched_entity_free(entity); +} + static struct kunit_case drm_sched_timeout_tests[] = { KUNIT_CASE(drm_sched_basic_timeout), + KUNIT_CASE(drm_sched_skip_reset), {} }; @@ -471,6 +557,7 @@ static struct kunit_suite drm_sched_credits = { kunit_test_suites(&drm_sched_basic, &drm_sched_timeout, + &drm_sched_cancel, &drm_sched_priority, &drm_sched_modify_sched, &drm_sched_credits); |