diff options
author | Konsta Holtta <kholtta@nvidia.com> | 2017-05-16 06:47:58 -0400 |
---|---|---|
committer | mobile promotions <svcmobile_promotions@nvidia.com> | 2017-06-14 19:33:32 -0400 |
commit | 7680fd689ecf7d11bf2dfdba41dc2f33cde2bbe7 (patch) | |
tree | b6df6640eaaa7e46deae7035572e7019f4311f7d /drivers/gpu/nvgpu/gk20a/gk20a.c | |
parent | 3c3c39dfe0d1122efeead871eec7c37617404850 (diff) |
gpu: nvgpu: hold power ref for deterministic channels
To support deterministic channels even with platforms where railgating
is supported, have each deterministic-marked channel hold a power
reference during their lifetime, and skip taking power refs for jobs in
submit path for those.
Previously, railgating blocked deterministic submits in general because
of gk20a_busy()/gk20a_idle() calls in submit path possibly taking time
and more significantly because the gpu may need turning on which takes a
nondeterministic and long amount of time.
As an exception, gk20a_do_idle() can still block deterministic submits
until gk20a_do_unidle() is called. Add a rwsem to guard this. VPR resize
needs do_idle, which conflicts with deterministic channels' requirement
to keep the GPU on. This is documented in the ioctl header now.
Make NVGPU_GPU_FLAGS_SUPPORT_DETERMINISTIC_SUBMIT_NO_JOBTRACKING always
set in the gpu characteristics now that it's supported. The only thing
left now blocking NVGPU_GPU_FLAGS_SUPPORT_DETERMINISTIC_SUBMIT_FULL is
the sync framework.
Make the channel debug dump show which channels are deterministic.
Bug 200291300
Jira NVGPU-70
Change-Id: I47b6f3a8517cd6e4255f6ca2855e3dd912e4f5f3
Signed-off-by: Konsta Holtta <kholtta@nvidia.com>
Reviewed-on: http://git-master/r/1483038
Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
Diffstat (limited to 'drivers/gpu/nvgpu/gk20a/gk20a.c')
-rw-r--r-- | drivers/gpu/nvgpu/gk20a/gk20a.c | 19 |
1 files changed, 8 insertions, 11 deletions
diff --git a/drivers/gpu/nvgpu/gk20a/gk20a.c b/drivers/gpu/nvgpu/gk20a/gk20a.c index 8624d601..13635706 100644 --- a/drivers/gpu/nvgpu/gk20a/gk20a.c +++ b/drivers/gpu/nvgpu/gk20a/gk20a.c | |||
@@ -456,22 +456,19 @@ int gk20a_init_gpu_characteristics(struct gk20a *g) | |||
456 | gpu->flags |= NVGPU_GPU_FLAGS_HAS_SYNCPOINTS; | 456 | gpu->flags |= NVGPU_GPU_FLAGS_HAS_SYNCPOINTS; |
457 | 457 | ||
458 | /* | 458 | /* |
459 | * Railgating needs job tracking which prevents fast submits. They're | 459 | * Fast submits are supported as long as the user doesn't request |
460 | * supported otherwise, provided that the user doesn't request anything | 460 | * anything that depends on job tracking. (Here, fast means strictly no |
461 | * that depends on job tracking. (Here, fast means strictly no | ||
462 | * metadata, just the gpfifo contents are copied and gp_put updated). | 461 | * metadata, just the gpfifo contents are copied and gp_put updated). |
463 | */ | 462 | */ |
464 | if (!g->can_railgate) | 463 | gpu->flags |= NVGPU_GPU_FLAGS_SUPPORT_DETERMINISTIC_SUBMIT_NO_JOBTRACKING; |
465 | gpu->flags |= NVGPU_GPU_FLAGS_SUPPORT_DETERMINISTIC_SUBMIT_NO_JOBTRACKING; | ||
466 | 464 | ||
467 | /* | 465 | /* |
468 | * Railgating and sync framework require deferred job cleanup which | 466 | * Sync framework requires deferred job cleanup, wrapping syncs in FDs, |
469 | * prevents deterministic submits. They're supported otherwise, | 467 | * and other heavy stuff, which prevents deterministic submits. This is |
470 | * provided that the user doesn't request anything that depends on | 468 | * supported otherwise, provided that the user doesn't request anything |
471 | * deferred cleanup. | 469 | * that depends on deferred cleanup. |
472 | */ | 470 | */ |
473 | if (!g->can_railgate | 471 | if (!gk20a_channel_sync_needs_sync_framework(g)) |
474 | && !gk20a_channel_sync_needs_sync_framework(g)) | ||
475 | gpu->flags |= NVGPU_GPU_FLAGS_SUPPORT_DETERMINISTIC_SUBMIT_FULL; | 472 | gpu->flags |= NVGPU_GPU_FLAGS_SUPPORT_DETERMINISTIC_SUBMIT_FULL; |
476 | 473 | ||
477 | gpu->flags |= NVGPU_GPU_FLAGS_SUPPORT_USERSPACE_MANAGED_AS; | 474 | gpu->flags |= NVGPU_GPU_FLAGS_SUPPORT_USERSPACE_MANAGED_AS; |