| Message ID | 20241114100113.150647-5-pierre-eric.pelloux-prayer@amd.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | Improve gpu_scheduler trace events + uAPI | expand |
On 14/11/2024 10:01, Pierre-Eric Pelloux-Prayer wrote: > A fence uniquely identify a job, so this commits updates the places > where a kernel pointer was used as an identifier by: > > "fence=(context:%llu, seqno:%lld)" Any special reason for %lld? Planning to use like -1 for unknown or something? Btw in i915 we use fence=%llu:%llu which is more compact and IMO still readable so perhaps something to consider. Regards, Tvrtko > Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> > --- > .../gpu/drm/scheduler/gpu_scheduler_trace.h | 39 +++++++++++-------- > 1 file changed, 22 insertions(+), 17 deletions(-) > > diff --git a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h > index c4ec28540656..24358c4d5bbe 100644 > --- a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h > +++ b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h > @@ -36,28 +36,29 @@ 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(uint64_t, fence_context) > + __field(uint64_t, fence_seqno) > ), > > 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; > + > ), > - TP_printk("dev=%s, entity=%p, id=%llu, fence=%p, ring=%s, job count:%u, hw job count:%d", > - __get_str(dev), __entry->entity, __entry->id, > - __entry->fence, __get_str(name), > + TP_printk("dev=%s, id=%llu, fence=(context:%llu, seqno:%lld), ring=%s, job count:%u, hw job count:%d", > + __get_str(dev), __entry->id, > + __entry->fence_context, __entry->fence_seqno, __get_str(name), > __entry->job_count, __entry->hw_job_count) > ); > > @@ -75,37 +76,41 @@ TRACE_EVENT(drm_sched_process_job, > TP_PROTO(struct drm_sched_fence *fence), > TP_ARGS(fence), > TP_STRUCT__entry( > - __field(struct dma_fence *, fence) > + __field(uint64_t, fence_context) > + __field(uint64_t, 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=(context:%llu, seqno:%lld) signaled", > + __entry->fence_context, __entry->fence_seqno) > ); > > TRACE_EVENT(drm_sched_job_wait_dep, > 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, fence_context) > + __field(uint64_t, fence_seqno) > __field(uint64_t, id) > __field(struct dma_fence *, fence) > __field(uint64_t, ctx) > - __field(unsigned, seqno) > + __field(uint64_t, seqno) > ), > > TP_fast_assign( > - __assign_str(name); > + __entry->fence_context = sched_job->s_fence->finished.context; > + __entry->fence_seqno = sched_job->s_fence->finished.seqno; > __entry->id = sched_job->id; > __entry->fence = fence; > __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=(context:%llu, seqno:%lld), id=%llu, dependencies:{(context:%llu, seqno:%lld)}", > + __entry->fence_context, __entry->fence_seqno, __entry->id, > + __entry->ctx, __entry->seqno) > ); > > #endif
On Thu, 14 Nov 2024 11:01:07 +0100 Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> wrote: > A fence uniquely identify a job, so this commits updates the places > where a kernel pointer was used as an identifier by: > > "fence=(context:%llu, seqno:%lld)" > > Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> > --- > .../gpu/drm/scheduler/gpu_scheduler_trace.h | 39 +++++++++++-------- > 1 file changed, 22 insertions(+), 17 deletions(-) > > diff --git a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h > index c4ec28540656..24358c4d5bbe 100644 > --- a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h > +++ b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h > @@ -36,28 +36,29 @@ 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) The __string() is a 4 byte meta field in the trace event structure. This means if you have a 64bit (8 byte) field following it, there will likely be a 4 byte hole between the two fields. I would suggest swapping the name field with the id field, then you will have four 4 byte fields in a row (name, job_count, hw_job_count, dev), which should fill all the holes. -- Steve > __field(uint64_t, id) > __field(u32, job_count) > __field(int, hw_job_count) > __string(dev, dev_name(sched_job->sched->dev)) > + __field(uint64_t, fence_context) > + __field(uint64_t, fence_seqno) > ), > > 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; > + > ), > - TP_printk("dev=%s, entity=%p, id=%llu, fence=%p, ring=%s, job count:%u, hw job count:%d", > - __get_str(dev), __entry->entity, __entry->id, > - __entry->fence, __get_str(name), > + TP_printk("dev=%s, id=%llu, fence=(context:%llu, seqno:%lld), ring=%s, job count:%u, hw job count:%d", > + __get_str(dev), __entry->id, > + __entry->fence_context, __entry->fence_seqno, __get_str(name), > __entry->job_count, __entry->hw_job_count) > ); >
On Thu, 2024-11-14 at 11:01 +0100, Pierre-Eric Pelloux-Prayer wrote: > A fence uniquely identify a job, so this commits updates the places s/identify/identifies > where a kernel pointer was used as an identifier by: But better sth like this: "Currently, the scheduler's tracing infrastructure uses the job's dma_fence and the drm_sched_entity the job belongs to. The dma_fence alone, however, already uniquely identifies a job. Use the dma_fence's context and sequence number exclusively identify a job in debug prints like so: > > "fence=(context:%llu, seqno:%lld)" > > Signed-off-by: Pierre-Eric Pelloux-Prayer > <pierre-eric.pelloux-prayer@amd.com> > --- > .../gpu/drm/scheduler/gpu_scheduler_trace.h | 39 +++++++++++------ > -- > 1 file changed, 22 insertions(+), 17 deletions(-) > > diff --git a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h > b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h > index c4ec28540656..24358c4d5bbe 100644 > --- a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h > +++ b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h > @@ -36,28 +36,29 @@ 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(uint64_t, fence_context) > + __field(uint64_t, fence_seqno) > ), > > 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; > + > ), > - TP_printk("dev=%s, entity=%p, id=%llu, fence=%p, > ring=%s, job count:%u, hw job count:%d", > - __get_str(dev), __entry->entity, __entry->id, > - __entry->fence, __get_str(name), > + TP_printk("dev=%s, id=%llu, fence=(context:%llu, > seqno:%lld), ring=%s, job count:%u, hw job count:%d", > + __get_str(dev), __entry->id, > + __entry->fence_context, __entry->fence_seqno, > __get_str(name), > __entry->job_count, __entry->hw_job_count) > ); > > @@ -75,37 +76,41 @@ TRACE_EVENT(drm_sched_process_job, > TP_PROTO(struct drm_sched_fence *fence), > TP_ARGS(fence), > TP_STRUCT__entry( > - __field(struct dma_fence *, fence) > + __field(uint64_t, fence_context) > + __field(uint64_t, 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=(context:%llu, seqno:%lld) signaled", > + __entry->fence_context, __entry->fence_seqno) > ); > > TRACE_EVENT(drm_sched_job_wait_dep, > 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, fence_context) > + __field(uint64_t, fence_seqno) > __field(uint64_t, id) > __field(struct dma_fence *, fence) > __field(uint64_t, ctx) > - __field(unsigned, seqno) > + __field(uint64_t, seqno) > ), > > TP_fast_assign( > - __assign_str(name); > + __entry->fence_context = sched_job- > >s_fence->finished.context; > + __entry->fence_seqno = sched_job- > >s_fence->finished.seqno; > __entry->id = sched_job->id; > __entry->fence = fence; > __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=(context:%llu, seqno:%lld), id=%llu, > dependencies:{(context:%llu, seqno:%lld)}", > + __entry->fence_context, __entry->fence_seqno, > __entry->id, > + __entry->ctx, __entry->seqno) > ); > > #endif
diff --git a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h index c4ec28540656..24358c4d5bbe 100644 --- a/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h +++ b/drivers/gpu/drm/scheduler/gpu_scheduler_trace.h @@ -36,28 +36,29 @@ 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(uint64_t, fence_context) + __field(uint64_t, fence_seqno) ), 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; + ), - TP_printk("dev=%s, entity=%p, id=%llu, fence=%p, ring=%s, job count:%u, hw job count:%d", - __get_str(dev), __entry->entity, __entry->id, - __entry->fence, __get_str(name), + TP_printk("dev=%s, id=%llu, fence=(context:%llu, seqno:%lld), ring=%s, job count:%u, hw job count:%d", + __get_str(dev), __entry->id, + __entry->fence_context, __entry->fence_seqno, __get_str(name), __entry->job_count, __entry->hw_job_count) ); @@ -75,37 +76,41 @@ TRACE_EVENT(drm_sched_process_job, TP_PROTO(struct drm_sched_fence *fence), TP_ARGS(fence), TP_STRUCT__entry( - __field(struct dma_fence *, fence) + __field(uint64_t, fence_context) + __field(uint64_t, 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=(context:%llu, seqno:%lld) signaled", + __entry->fence_context, __entry->fence_seqno) ); TRACE_EVENT(drm_sched_job_wait_dep, 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, fence_context) + __field(uint64_t, fence_seqno) __field(uint64_t, id) __field(struct dma_fence *, fence) __field(uint64_t, ctx) - __field(unsigned, seqno) + __field(uint64_t, seqno) ), TP_fast_assign( - __assign_str(name); + __entry->fence_context = sched_job->s_fence->finished.context; + __entry->fence_seqno = sched_job->s_fence->finished.seqno; __entry->id = sched_job->id; __entry->fence = fence; __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=(context:%llu, seqno:%lld), id=%llu, dependencies:{(context:%llu, seqno:%lld)}", + __entry->fence_context, __entry->fence_seqno, __entry->id, + __entry->ctx, __entry->seqno) ); #endif
A fence uniquely identify a job, so this commits updates the places where a kernel pointer was used as an identifier by: "fence=(context:%llu, seqno:%lld)" Signed-off-by: Pierre-Eric Pelloux-Prayer <pierre-eric.pelloux-prayer@amd.com> --- .../gpu/drm/scheduler/gpu_scheduler_trace.h | 39 +++++++++++-------- 1 file changed, 22 insertions(+), 17 deletions(-)