[PATCH 5/5] amdgpu: use drm sync objects for shared semaphores (v5)
Christian König
deathsimple at vodafone.de
Mon May 29 12:04:20 UTC 2017
Am 29.05.2017 um 09:30 schrieb Dave Airlie:
> From: Dave Airlie <airlied at redhat.com>
>
> This creates a new command submission chunk for amdgpu
> to add in and out sync objects around the submission.
>
> Sync objects are managed via the drm syncobj ioctls.
>
> The command submission interface is enhanced with two new
> chunks, one for syncobj pre submission dependencies,
> and one for post submission sync obj signalling,
> and just takes a list of handles for each.
>
> This is based on work originally done by David Zhou at AMD,
> with input from Christian Konig on what things should look like.
>
> In theory VkFences could be backed with sync objects and
> just get passed into the cs as syncobj handles as well.
>
> NOTE: this interface addition needs a version bump to expose
> it to userspace.
>
> TODO: update to dep_sync when rebasing onto amdgpu master.
> (with this - r-b from Christian)
>
> v1.1: keep file reference on import.
> v2: move to using syncobjs
> v2.1: change some APIs to just use p pointer.
> v3: make more robust against CS failures, we now add the
> wait sems but only remove them once the CS job has been
> submitted.
> v4: rewrite names of API and base on new syncobj code.
> v5: move post deps earlier, rename some apis
>
> Signed-off-by: Dave Airlie <airlied at redhat.com>
> ---
> drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c | 87 ++++++++++++++++++++++++++++++++-
> drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c | 2 +-
> include/uapi/drm/amdgpu_drm.h | 6 +++
> 3 files changed, 93 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
> index 30225d7..3cbd547 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
> @@ -27,6 +27,7 @@
> #include <linux/pagemap.h>
> #include <drm/drmP.h>
> #include <drm/amdgpu_drm.h>
> +#include <drm/drm_syncobj.h>
> #include "amdgpu.h"
> #include "amdgpu_trace.h"
>
> @@ -226,6 +227,8 @@ int amdgpu_cs_parser_init(struct amdgpu_cs_parser *p, void *data)
> break;
>
> case AMDGPU_CHUNK_ID_DEPENDENCIES:
> + case AMDGPU_CHUNK_ID_SYNCOBJ_IN:
> + case AMDGPU_CHUNK_ID_SYNCOBJ_OUT:
> break;
>
> default:
> @@ -1040,6 +1043,40 @@ static int amdgpu_cs_process_fence_dep(struct amdgpu_cs_parser *p,
> return 0;
> }
>
> +static int amdgpu_syncobj_lookup_and_add_to_sync(struct amdgpu_cs_parser *p,
> + uint32_t handle)
> +{
> + int r;
> + struct dma_fence *fence;
> + r = drm_syncobj_fence_get(p->filp, handle, &fence);
> + if (r)
> + return r;
> +
> + r = amdgpu_sync_fence(p->adev, &p->job->sync, fence);
> + dma_fence_put(fence);
> +
> + return r;
> +}
> +
> +static int amdgpu_cs_process_syncobj_in_dep(struct amdgpu_cs_parser *p,
> + struct amdgpu_cs_chunk *chunk)
> +{
> + unsigned num_deps;
> + int i, r;
> + struct drm_amdgpu_cs_chunk_sem *deps;
> +
> + deps = (struct drm_amdgpu_cs_chunk_sem *)chunk->kdata;
> + num_deps = chunk->length_dw * 4 /
> + sizeof(struct drm_amdgpu_cs_chunk_sem);
> +
> + for (i = 0; i < num_deps; ++i) {
> + r = amdgpu_syncobj_lookup_and_add_to_sync(p, deps[i].handle);
> + if (r)
> + return r;
> + }
> + return 0;
> +}
> +
> static int amdgpu_cs_dependencies(struct amdgpu_device *adev,
> struct amdgpu_cs_parser *p)
> {
> @@ -1054,12 +1091,54 @@ static int amdgpu_cs_dependencies(struct amdgpu_device *adev,
> r = amdgpu_cs_process_fence_dep(p, chunk);
> if (r)
> return r;
> + } else if (chunk->chunk_id == AMDGPU_CHUNK_ID_SYNCOBJ_IN) {
> + r = amdgpu_cs_process_syncobj_in_dep(p, chunk);
> + if (r)
> + return r;
> }
> }
>
> return 0;
> }
>
> +static int amdgpu_cs_process_syncobj_out_dep(struct amdgpu_cs_parser *p,
> + struct amdgpu_cs_chunk *chunk)
> +{
> + unsigned num_deps;
> + int i, r;
> + struct drm_amdgpu_cs_chunk_sem *deps;
> +
> + deps = (struct drm_amdgpu_cs_chunk_sem *)chunk->kdata;
> + num_deps = chunk->length_dw * 4 /
> + sizeof(struct drm_amdgpu_cs_chunk_sem);
> +
> + for (i = 0; i < num_deps; ++i) {
> + r = drm_syncobj_replace_fence(p->filp, deps[i].handle,
> + p->fence);
> + if (r)
> + return r;
Thinking more about it, here is still a problem with the handling.
Imagine you have multiple semaphores to signal and only the last handle
is invalid.
So if you have n handles then the sync objects 1..(n-1) will get a fence
which is never signaled because n aborts the command submission.
I think the only valid approach to handle this is to resolve the handles
upfront during the initial parsing of the chunks.
Christian.
> + }
> + return 0;
> +}
> +
> +static int amdgpu_cs_post_dependencies(struct amdgpu_cs_parser *p)
> +{
> + int i, r;
> +
> + for (i = 0; i < p->nchunks; ++i) {
> + struct amdgpu_cs_chunk *chunk;
> +
> + chunk = &p->chunks[i];
> +
> + if (chunk->chunk_id == AMDGPU_CHUNK_ID_SYNCOBJ_OUT) {
> + r = amdgpu_cs_process_syncobj_out_dep(p, chunk);
> + if (r)
> + return r;
> + }
> + }
> + return 0;
> +}
> +
> static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
> union drm_amdgpu_cs *cs)
> {
> @@ -1080,6 +1159,13 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
> job->owner = p->filp;
> job->fence_ctx = entity->fence_context;
> p->fence = dma_fence_get(&job->base.s_fence->finished);
> +
> + r = amdgpu_cs_post_dependencies(p);
> + if (r) {
> + amdgpu_job_free(job);
> + return r;
> + }
> +
> cs->out.handle = amdgpu_ctx_add_fence(p->ctx, ring, p->fence);
> job->uf_sequence = cs->out.handle;
> amdgpu_job_free_resources(job);
> @@ -1087,7 +1173,6 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
>
> trace_amdgpu_cs_ioctl(job);
> amd_sched_entity_push_job(&job->base);
> -
> return 0;
> }
>
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
> index f2d705e..5329e7f 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
> @@ -719,7 +719,7 @@ static struct drm_driver kms_driver = {
> .driver_features =
> DRIVER_USE_AGP |
> DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED | DRIVER_GEM |
> - DRIVER_PRIME | DRIVER_RENDER | DRIVER_MODESET,
> + DRIVER_PRIME | DRIVER_RENDER | DRIVER_MODESET | DRIVER_SYNCOBJ,
> .load = amdgpu_driver_load_kms,
> .open = amdgpu_driver_open_kms,
> .postclose = amdgpu_driver_postclose_kms,
> diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
> index 6c249e5..cc702ca 100644
> --- a/include/uapi/drm/amdgpu_drm.h
> +++ b/include/uapi/drm/amdgpu_drm.h
> @@ -416,6 +416,8 @@ struct drm_amdgpu_gem_va {
> #define AMDGPU_CHUNK_ID_IB 0x01
> #define AMDGPU_CHUNK_ID_FENCE 0x02
> #define AMDGPU_CHUNK_ID_DEPENDENCIES 0x03
> +#define AMDGPU_CHUNK_ID_SYNCOBJ_IN 0x04
> +#define AMDGPU_CHUNK_ID_SYNCOBJ_OUT 0x05
>
> struct drm_amdgpu_cs_chunk {
> __u32 chunk_id;
> @@ -483,6 +485,10 @@ struct drm_amdgpu_cs_chunk_fence {
> __u32 offset;
> };
>
> +struct drm_amdgpu_cs_chunk_sem {
> + __u32 handle;
> +};
> +
> struct drm_amdgpu_cs_chunk_data {
> union {
> struct drm_amdgpu_cs_chunk_ib ib_data;
More information about the amd-gfx
mailing list