Commit c5f74f78 authored by Christian König's avatar Christian König Committed by Alex Deucher

drm/amdgpu: fix and cleanup job destruction

Remove the job reference counting and just properly destroy it from a
work item which blocks on any potential running timeout handler.
Signed-off-by: default avatarChristian König <christian.koenig@amd.com>
Reviewed-by: default avatarMonk.Liu <monk.liu@amd.com>
Signed-off-by: default avatarAlex Deucher <alexander.deucher@amd.com>
parent f42d20a9
...@@ -755,7 +755,6 @@ int amdgpu_job_alloc_with_ib(struct amdgpu_device *adev, unsigned size, ...@@ -755,7 +755,6 @@ int amdgpu_job_alloc_with_ib(struct amdgpu_device *adev, unsigned size,
struct amdgpu_job **job); struct amdgpu_job **job);
void amdgpu_job_free(struct amdgpu_job *job); void amdgpu_job_free(struct amdgpu_job *job);
void amdgpu_job_free_func(struct kref *refcount);
int amdgpu_job_submit(struct amdgpu_job *job, struct amdgpu_ring *ring, int amdgpu_job_submit(struct amdgpu_job *job, struct amdgpu_ring *ring,
struct amd_sched_entity *entity, void *owner, struct amd_sched_entity *entity, void *owner,
struct fence **f); struct fence **f);
......
...@@ -838,8 +838,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p, ...@@ -838,8 +838,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
p->job = NULL; p->job = NULL;
r = amd_sched_job_init(&job->base, &ring->sched, r = amd_sched_job_init(&job->base, &ring->sched,
entity, amdgpu_job_free_func, entity, p->filp, &fence);
p->filp, &fence);
if (r) { if (r) {
amdgpu_job_free(job); amdgpu_job_free(job);
return r; return r;
......
...@@ -28,12 +28,6 @@ ...@@ -28,12 +28,6 @@
#include "amdgpu.h" #include "amdgpu.h"
#include "amdgpu_trace.h" #include "amdgpu_trace.h"
static void amdgpu_job_free_handler(struct work_struct *ws)
{
struct amdgpu_job *job = container_of(ws, struct amdgpu_job, base.work_free_job);
amd_sched_job_put(&job->base);
}
static void amdgpu_job_timedout(struct amd_sched_job *s_job) static void amdgpu_job_timedout(struct amd_sched_job *s_job)
{ {
struct amdgpu_job *job = container_of(s_job, struct amdgpu_job, base); struct amdgpu_job *job = container_of(s_job, struct amdgpu_job, base);
...@@ -42,8 +36,6 @@ static void amdgpu_job_timedout(struct amd_sched_job *s_job) ...@@ -42,8 +36,6 @@ static void amdgpu_job_timedout(struct amd_sched_job *s_job)
job->base.sched->name, job->base.sched->name,
atomic_read(&job->ring->fence_drv.last_seq), atomic_read(&job->ring->fence_drv.last_seq),
job->ring->fence_drv.sync_seq); job->ring->fence_drv.sync_seq);
amd_sched_job_put(&job->base);
} }
int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs, int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs,
...@@ -64,7 +56,6 @@ int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs, ...@@ -64,7 +56,6 @@ int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs,
(*job)->vm = vm; (*job)->vm = vm;
(*job)->ibs = (void *)&(*job)[1]; (*job)->ibs = (void *)&(*job)[1];
(*job)->num_ibs = num_ibs; (*job)->num_ibs = num_ibs;
INIT_WORK(&(*job)->base.work_free_job, amdgpu_job_free_handler);
amdgpu_sync_create(&(*job)->sync); amdgpu_sync_create(&(*job)->sync);
...@@ -103,9 +94,10 @@ static void amdgpu_job_free_resources(struct amdgpu_job *job) ...@@ -103,9 +94,10 @@ static void amdgpu_job_free_resources(struct amdgpu_job *job)
amdgpu_sync_free(&job->sync); amdgpu_sync_free(&job->sync);
} }
void amdgpu_job_free_func(struct kref *refcount) void amdgpu_job_free_cb(struct amd_sched_job *s_job)
{ {
struct amdgpu_job *job = container_of(refcount, struct amdgpu_job, base.refcount); struct amdgpu_job *job = container_of(s_job, struct amdgpu_job, base);
kfree(job); kfree(job);
} }
...@@ -126,8 +118,7 @@ int amdgpu_job_submit(struct amdgpu_job *job, struct amdgpu_ring *ring, ...@@ -126,8 +118,7 @@ int amdgpu_job_submit(struct amdgpu_job *job, struct amdgpu_ring *ring,
if (!f) if (!f)
return -EINVAL; return -EINVAL;
r = amd_sched_job_init(&job->base, &ring->sched, r = amd_sched_job_init(&job->base, &ring->sched, entity, owner, &fence);
entity, amdgpu_job_free_func, owner, &fence);
if (r) if (r)
return r; return r;
...@@ -198,4 +189,5 @@ const struct amd_sched_backend_ops amdgpu_sched_ops = { ...@@ -198,4 +189,5 @@ const struct amd_sched_backend_ops amdgpu_sched_ops = {
.dependency = amdgpu_job_dependency, .dependency = amdgpu_job_dependency,
.run_job = amdgpu_job_run, .run_job = amdgpu_job_run,
.timedout_job = amdgpu_job_timedout, .timedout_job = amdgpu_job_timedout,
.free_job = amdgpu_job_free_cb
}; };
...@@ -319,19 +319,13 @@ static bool amd_sched_entity_in(struct amd_sched_job *sched_job) ...@@ -319,19 +319,13 @@ static bool amd_sched_entity_in(struct amd_sched_job *sched_job)
return added; return added;
} }
static void amd_sched_free_job(struct fence *f, struct fence_cb *cb) {
struct amd_sched_job *job = container_of(cb, struct amd_sched_job,
cb_free_job);
schedule_work(&job->work_free_job);
}
/* job_finish is called after hw fence signaled, and /* job_finish is called after hw fence signaled, and
* the job had already been deleted from ring_mirror_list * the job had already been deleted from ring_mirror_list
*/ */
static void amd_sched_job_finish(struct amd_sched_job *s_job) static void amd_sched_job_finish(struct work_struct *work)
{ {
struct amd_sched_job *next; struct amd_sched_job *s_job = container_of(work, struct amd_sched_job,
finish_work);
struct amd_gpu_scheduler *sched = s_job->sched; struct amd_gpu_scheduler *sched = s_job->sched;
unsigned long flags; unsigned long flags;
...@@ -339,19 +333,26 @@ static void amd_sched_job_finish(struct amd_sched_job *s_job) ...@@ -339,19 +333,26 @@ static void amd_sched_job_finish(struct amd_sched_job *s_job)
spin_lock_irqsave(&sched->job_list_lock, flags); spin_lock_irqsave(&sched->job_list_lock, flags);
list_del_init(&s_job->node); list_del_init(&s_job->node);
if (sched->timeout != MAX_SCHEDULE_TIMEOUT) { if (sched->timeout != MAX_SCHEDULE_TIMEOUT) {
if (cancel_delayed_work(&s_job->work_tdr)) struct amd_sched_job *next;
amd_sched_job_put(s_job);
cancel_delayed_work_sync(&s_job->work_tdr);
/* queue TDR for next job */ /* queue TDR for next job */
next = list_first_entry_or_null(&sched->ring_mirror_list, next = list_first_entry_or_null(&sched->ring_mirror_list,
struct amd_sched_job, node); struct amd_sched_job, node);
if (next) { if (next)
amd_sched_job_get(next);
schedule_delayed_work(&next->work_tdr, sched->timeout); schedule_delayed_work(&next->work_tdr, sched->timeout);
}
} }
spin_unlock_irqrestore(&sched->job_list_lock, flags); spin_unlock_irqrestore(&sched->job_list_lock, flags);
sched->ops->free_job(s_job);
}
static void amd_sched_job_finish_cb(struct fence *f, struct fence_cb *cb)
{
struct amd_sched_job *job = container_of(cb, struct amd_sched_job,
finish_cb);
schedule_work(&job->finish_work);
} }
static void amd_sched_job_begin(struct amd_sched_job *s_job) static void amd_sched_job_begin(struct amd_sched_job *s_job)
...@@ -364,10 +365,7 @@ static void amd_sched_job_begin(struct amd_sched_job *s_job) ...@@ -364,10 +365,7 @@ static void amd_sched_job_begin(struct amd_sched_job *s_job)
if (sched->timeout != MAX_SCHEDULE_TIMEOUT && if (sched->timeout != MAX_SCHEDULE_TIMEOUT &&
list_first_entry_or_null(&sched->ring_mirror_list, list_first_entry_or_null(&sched->ring_mirror_list,
struct amd_sched_job, node) == s_job) struct amd_sched_job, node) == s_job)
{
amd_sched_job_get(s_job);
schedule_delayed_work(&s_job->work_tdr, sched->timeout); schedule_delayed_work(&s_job->work_tdr, sched->timeout);
}
spin_unlock_irqrestore(&sched->job_list_lock, flags); spin_unlock_irqrestore(&sched->job_list_lock, flags);
} }
...@@ -390,9 +388,9 @@ void amd_sched_entity_push_job(struct amd_sched_job *sched_job) ...@@ -390,9 +388,9 @@ void amd_sched_entity_push_job(struct amd_sched_job *sched_job)
{ {
struct amd_sched_entity *entity = sched_job->s_entity; struct amd_sched_entity *entity = sched_job->s_entity;
fence_add_callback(&sched_job->s_fence->base,
&sched_job->cb_free_job, amd_sched_free_job);
trace_amd_sched_job(sched_job); trace_amd_sched_job(sched_job);
fence_add_callback(&sched_job->s_fence->base, &sched_job->finish_cb,
amd_sched_job_finish_cb);
wait_event(entity->sched->job_scheduled, wait_event(entity->sched->job_scheduled,
amd_sched_entity_in(sched_job)); amd_sched_entity_in(sched_job));
} }
...@@ -401,20 +399,17 @@ void amd_sched_entity_push_job(struct amd_sched_job *sched_job) ...@@ -401,20 +399,17 @@ void amd_sched_entity_push_job(struct amd_sched_job *sched_job)
int amd_sched_job_init(struct amd_sched_job *job, int amd_sched_job_init(struct amd_sched_job *job,
struct amd_gpu_scheduler *sched, struct amd_gpu_scheduler *sched,
struct amd_sched_entity *entity, struct amd_sched_entity *entity,
void (*free_cb)(struct kref *refcount),
void *owner, struct fence **fence) void *owner, struct fence **fence)
{ {
INIT_LIST_HEAD(&job->node);
kref_init(&job->refcount);
job->sched = sched; job->sched = sched;
job->s_entity = entity; job->s_entity = entity;
job->s_fence = amd_sched_fence_create(entity, owner); job->s_fence = amd_sched_fence_create(entity, owner);
if (!job->s_fence) if (!job->s_fence)
return -ENOMEM; return -ENOMEM;
job->s_fence->s_job = job; INIT_WORK(&job->finish_work, amd_sched_job_finish);
INIT_LIST_HEAD(&job->node);
INIT_DELAYED_WORK(&job->work_tdr, amd_sched_job_timedout); INIT_DELAYED_WORK(&job->work_tdr, amd_sched_job_timedout);
job->free_callback = free_cb;
if (fence) if (fence)
*fence = &job->s_fence->base; *fence = &job->s_fence->base;
...@@ -468,9 +463,6 @@ static void amd_sched_process_job(struct fence *f, struct fence_cb *cb) ...@@ -468,9 +463,6 @@ static void amd_sched_process_job(struct fence *f, struct fence_cb *cb)
struct amd_gpu_scheduler *sched = s_fence->sched; struct amd_gpu_scheduler *sched = s_fence->sched;
atomic_dec(&sched->hw_rq_count); atomic_dec(&sched->hw_rq_count);
amd_sched_job_finish(s_fence->s_job);
amd_sched_fence_signal(s_fence); amd_sched_fence_signal(s_fence);
trace_amd_sched_process_job(s_fence); trace_amd_sched_process_job(s_fence);
......
...@@ -74,19 +74,16 @@ struct amd_sched_fence { ...@@ -74,19 +74,16 @@ struct amd_sched_fence {
struct amd_gpu_scheduler *sched; struct amd_gpu_scheduler *sched;
spinlock_t lock; spinlock_t lock;
void *owner; void *owner;
struct amd_sched_job *s_job;
}; };
struct amd_sched_job { struct amd_sched_job {
struct kref refcount;
struct amd_gpu_scheduler *sched; struct amd_gpu_scheduler *sched;
struct amd_sched_entity *s_entity; struct amd_sched_entity *s_entity;
struct amd_sched_fence *s_fence; struct amd_sched_fence *s_fence;
struct fence_cb cb_free_job; struct fence_cb finish_cb;
struct work_struct work_free_job; struct work_struct finish_work;
struct list_head node; struct list_head node;
struct delayed_work work_tdr; struct delayed_work work_tdr;
void (*free_callback)(struct kref *refcount);
}; };
extern const struct fence_ops amd_sched_fence_ops; extern const struct fence_ops amd_sched_fence_ops;
...@@ -109,6 +106,7 @@ struct amd_sched_backend_ops { ...@@ -109,6 +106,7 @@ struct amd_sched_backend_ops {
struct fence *(*dependency)(struct amd_sched_job *sched_job); struct fence *(*dependency)(struct amd_sched_job *sched_job);
struct fence *(*run_job)(struct amd_sched_job *sched_job); struct fence *(*run_job)(struct amd_sched_job *sched_job);
void (*timedout_job)(struct amd_sched_job *sched_job); void (*timedout_job)(struct amd_sched_job *sched_job);
void (*free_job)(struct amd_sched_job *sched_job);
}; };
enum amd_sched_priority { enum amd_sched_priority {
...@@ -154,18 +152,5 @@ void amd_sched_fence_signal(struct amd_sched_fence *fence); ...@@ -154,18 +152,5 @@ void amd_sched_fence_signal(struct amd_sched_fence *fence);
int amd_sched_job_init(struct amd_sched_job *job, int amd_sched_job_init(struct amd_sched_job *job,
struct amd_gpu_scheduler *sched, struct amd_gpu_scheduler *sched,
struct amd_sched_entity *entity, struct amd_sched_entity *entity,
void (*free_cb)(struct kref* refcount),
void *owner, struct fence **fence); void *owner, struct fence **fence);
static inline void amd_sched_job_get(struct amd_sched_job *job)
{
if (job)
kref_get(&job->refcount);
}
static inline void amd_sched_job_put(struct amd_sched_job *job)
{
if (job)
kref_put(&job->refcount, job->free_callback);
}
#endif #endif
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment