mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2024-12-22 19:43:07 +07:00
30514decb2
The problem is that executing the jobs in the right order doesn't give you the right result because consecutive jobs executed on the same engine are pipelined. In other words job B does it buffer read before job A has written it's result. Signed-off-by: Chunming Zhou <David1.Zhou@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
194 lines
4.9 KiB
C
194 lines
4.9 KiB
C
/*
|
|
* Copyright 2015 Advanced Micro Devices, Inc.
|
|
*
|
|
* Permission is hereby granted, free of charge, to any person obtaining a
|
|
* copy of this software and associated documentation files (the "Software"),
|
|
* to deal in the Software without restriction, including without limitation
|
|
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
|
|
* and/or sell copies of the Software, and to permit persons to whom the
|
|
* Software is furnished to do so, subject to the following conditions:
|
|
*
|
|
* The above copyright notice and this permission notice shall be included in
|
|
* all copies or substantial portions of the Software.
|
|
*
|
|
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
|
|
* THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
|
|
* OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
|
|
* ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
|
|
* OTHER DEALINGS IN THE SOFTWARE.
|
|
*
|
|
*
|
|
*/
|
|
#include <linux/kthread.h>
|
|
#include <linux/wait.h>
|
|
#include <linux/sched.h>
|
|
#include <drm/drmP.h>
|
|
#include "amdgpu.h"
|
|
#include "amdgpu_trace.h"
|
|
|
|
static void amdgpu_job_timedout(struct amd_sched_job *s_job)
|
|
{
|
|
struct amdgpu_job *job = container_of(s_job, struct amdgpu_job, base);
|
|
|
|
DRM_ERROR("ring %s timeout, last signaled seq=%u, last emitted seq=%u\n",
|
|
job->base.sched->name,
|
|
atomic_read(&job->ring->fence_drv.last_seq),
|
|
job->ring->fence_drv.sync_seq);
|
|
amdgpu_gpu_reset(job->adev);
|
|
}
|
|
|
|
int amdgpu_job_alloc(struct amdgpu_device *adev, unsigned num_ibs,
|
|
struct amdgpu_job **job, struct amdgpu_vm *vm)
|
|
{
|
|
size_t size = sizeof(struct amdgpu_job);
|
|
|
|
if (num_ibs == 0)
|
|
return -EINVAL;
|
|
|
|
size += sizeof(struct amdgpu_ib) * num_ibs;
|
|
|
|
*job = kzalloc(size, GFP_KERNEL);
|
|
if (!*job)
|
|
return -ENOMEM;
|
|
|
|
(*job)->adev = adev;
|
|
(*job)->vm = vm;
|
|
(*job)->ibs = (void *)&(*job)[1];
|
|
(*job)->num_ibs = num_ibs;
|
|
(*job)->need_pipeline_sync = false;
|
|
|
|
amdgpu_sync_create(&(*job)->sync);
|
|
|
|
return 0;
|
|
}
|
|
|
|
int amdgpu_job_alloc_with_ib(struct amdgpu_device *adev, unsigned size,
|
|
struct amdgpu_job **job)
|
|
{
|
|
int r;
|
|
|
|
r = amdgpu_job_alloc(adev, 1, job, NULL);
|
|
if (r)
|
|
return r;
|
|
|
|
r = amdgpu_ib_get(adev, NULL, size, &(*job)->ibs[0]);
|
|
if (r)
|
|
kfree(*job);
|
|
|
|
return r;
|
|
}
|
|
|
|
void amdgpu_job_free_resources(struct amdgpu_job *job)
|
|
{
|
|
struct dma_fence *f;
|
|
unsigned i;
|
|
|
|
/* use sched fence if available */
|
|
f = job->base.s_fence ? &job->base.s_fence->finished : job->fence;
|
|
|
|
for (i = 0; i < job->num_ibs; ++i)
|
|
amdgpu_ib_free(job->adev, &job->ibs[i], f);
|
|
}
|
|
|
|
static void amdgpu_job_free_cb(struct amd_sched_job *s_job)
|
|
{
|
|
struct amdgpu_job *job = container_of(s_job, struct amdgpu_job, base);
|
|
|
|
dma_fence_put(job->fence);
|
|
amdgpu_sync_free(&job->sync);
|
|
kfree(job);
|
|
}
|
|
|
|
void amdgpu_job_free(struct amdgpu_job *job)
|
|
{
|
|
amdgpu_job_free_resources(job);
|
|
|
|
dma_fence_put(job->fence);
|
|
amdgpu_sync_free(&job->sync);
|
|
kfree(job);
|
|
}
|
|
|
|
int amdgpu_job_submit(struct amdgpu_job *job, struct amdgpu_ring *ring,
|
|
struct amd_sched_entity *entity, void *owner,
|
|
struct dma_fence **f)
|
|
{
|
|
int r;
|
|
job->ring = ring;
|
|
|
|
if (!f)
|
|
return -EINVAL;
|
|
|
|
r = amd_sched_job_init(&job->base, &ring->sched, entity, owner);
|
|
if (r)
|
|
return r;
|
|
|
|
job->owner = owner;
|
|
job->fence_ctx = entity->fence_context;
|
|
*f = dma_fence_get(&job->base.s_fence->finished);
|
|
amdgpu_job_free_resources(job);
|
|
amd_sched_entity_push_job(&job->base);
|
|
|
|
return 0;
|
|
}
|
|
|
|
static struct dma_fence *amdgpu_job_dependency(struct amd_sched_job *sched_job)
|
|
{
|
|
struct amdgpu_job *job = to_amdgpu_job(sched_job);
|
|
struct amdgpu_vm *vm = job->vm;
|
|
|
|
struct dma_fence *fence = amdgpu_sync_get_fence(&job->sync);
|
|
|
|
while (fence == NULL && vm && !job->vm_id) {
|
|
struct amdgpu_ring *ring = job->ring;
|
|
int r;
|
|
|
|
r = amdgpu_vm_grab_id(vm, ring, &job->sync,
|
|
&job->base.s_fence->finished,
|
|
job);
|
|
if (r)
|
|
DRM_ERROR("Error getting VM ID (%d)\n", r);
|
|
|
|
fence = amdgpu_sync_get_fence(&job->sync);
|
|
}
|
|
|
|
if (amd_sched_dependency_optimized(fence, sched_job->s_entity))
|
|
job->need_pipeline_sync = true;
|
|
|
|
return fence;
|
|
}
|
|
|
|
static struct dma_fence *amdgpu_job_run(struct amd_sched_job *sched_job)
|
|
{
|
|
struct dma_fence *fence = NULL;
|
|
struct amdgpu_job *job;
|
|
int r;
|
|
|
|
if (!sched_job) {
|
|
DRM_ERROR("job is null\n");
|
|
return NULL;
|
|
}
|
|
job = to_amdgpu_job(sched_job);
|
|
|
|
BUG_ON(amdgpu_sync_peek_fence(&job->sync, NULL));
|
|
|
|
trace_amdgpu_sched_run_job(job);
|
|
r = amdgpu_ib_schedule(job->ring, job->num_ibs, job->ibs, job, &fence);
|
|
if (r)
|
|
DRM_ERROR("Error scheduling IBs (%d)\n", r);
|
|
|
|
/* if gpu reset, hw fence will be replaced here */
|
|
dma_fence_put(job->fence);
|
|
job->fence = dma_fence_get(fence);
|
|
amdgpu_job_free_resources(job);
|
|
return fence;
|
|
}
|
|
|
|
const struct amd_sched_backend_ops amdgpu_sched_ops = {
|
|
.dependency = amdgpu_job_dependency,
|
|
.run_job = amdgpu_job_run,
|
|
.timedout_job = amdgpu_job_timedout,
|
|
.free_job = amdgpu_job_free_cb
|
|
};
|