git/linux-net
1
0
Fork 0
mirror of https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net.git/ synced 2026-04-23 08:55:56 -04:00
Code Issues Packages Projects Releases Wiki Activity

drm/amd/amdgpu embed hw_fence into amdgpu_job

Browse Source 
Why: Previously hw fence is alloced separately with job.
It caused historical lifetime issues and corner cases.
The ideal situation is to take fence to manage both job
and fence's lifetime, and simplify the design of gpu-scheduler.

How:
We propose to embed hw_fence into amdgpu_job.
1. We cover the normal job submission by this method.
2. For ib_test, and submit without a parent job keep the
legacy way to create a hw fence separately.
v2:
use AMDGPU_FENCE_FLAG_EMBED_IN_JOB_BIT to show that the fence is
embedded in a job.
v3:
remove redundant variable ring in amdgpu_job
v4:
add tdr sequence support for this feature. Add a job_run_counter to
indicate whether this job is a resubmit job.
v5
add missing handling in amdgpu_fence_enable_signaling

Signed-off-by: Jingwen Chen <Jingwen.Chen2@amd.com>
Signed-off-by: Jack Zhang <Jack.Zhang7@hotmail.com>
Reviewed-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Reviewed by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This commit is contained in:
Jack Zhang
2021-05-12 15:06:35 +08:00
committed by Alex Deucher
parent 554594567b
commit c530b02f39
9 changed files with 119 additions and 37 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
View File

@@ -4449,7 +4449,7 @@ int amdgpu_device_mode1_reset(struct amdgpu_device *adev)
int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
struct amdgpu_reset_context *reset_context)
{
int i, r = 0;
int i, j, r = 0;
struct amdgpu_job *job = NULL;
bool need_full_reset =
test_bit(AMDGPU_NEED_FULL_RESET, &reset_context->flags);
@@ -4473,6 +4473,17 @@ int amdgpu_device_pre_asic_reset(struct amdgpu_device *adev,
if (!ring || !ring->sched.thread)
continue;
/*clear job fence from fence drv to avoid force_completion
*leave NULL and vm flush fence in fence drv */
for (j = 0; j <= ring->fence_drv.num_fences_mask; j++) {
struct dma_fence *old, **ptr;
ptr = &ring->fence_drv.fences[j];
old = rcu_dereference_protected(*ptr, 1);
if (old && test_bit(AMDGPU_FENCE_FLAG_EMBED_IN_JOB_BIT, &old->flags)) {
RCU_INIT_POINTER(*ptr, NULL);
}
}
/* after all hw jobs are reset, hw fence is meaningless, so force_completion */
amdgpu_fence_driver_force_completion(ring);
}