diff options
| author | Rob Clark <robdclark@chromium.org> | 2025-06-29 13:13:03 -0700 |
|---|---|---|
| committer | Rob Clark <robin.clark@oss.qualcomm.com> | 2025-07-04 17:48:36 -0700 |
| commit | 5b5582c6b8845cb11e7c62443b0b243953f716da (patch) | |
| tree | 79318bcbdc1c3db3bad9ff3e5112450d8344a6f8 /drivers/gpu/drm/msm/msm_ringbuffer.c | |
| parent | drm/msm: Rename msm_gem_vma_purge() -> _unmap() (diff) | |
| download | linux-5b5582c6b8845cb11e7c62443b0b243953f716da.tar.gz linux-5b5582c6b8845cb11e7c62443b0b243953f716da.zip | |
drm/msm: Drop queued submits on lastclose()
If we haven't written the submit into the ringbuffer yet, then drop it.
The submit still retires through the normal path, to preserve fence
signalling order, but we can skip the IB's to userspace cmdstream.
Signed-off-by: Rob Clark <robdclark@chromium.org>
Signed-off-by: Rob Clark <robin.clark@oss.qualcomm.com>
Tested-by: Antonino Maniscalco <antomani103@gmail.com>
Reviewed-by: Antonino Maniscalco <antomani103@gmail.com>
Patchwork: https://patchwork.freedesktop.org/patch/661489/
Diffstat (limited to 'drivers/gpu/drm/msm/msm_ringbuffer.c')
| -rw-r--r-- | drivers/gpu/drm/msm/msm_ringbuffer.c | 6 |
1 files changed, 6 insertions, 0 deletions
diff --git a/drivers/gpu/drm/msm/msm_ringbuffer.c b/drivers/gpu/drm/msm/msm_ringbuffer.c index 552b8da9e5f7..b2f612e5dc79 100644 --- a/drivers/gpu/drm/msm/msm_ringbuffer.c +++ b/drivers/gpu/drm/msm/msm_ringbuffer.c @@ -17,6 +17,7 @@ static struct dma_fence *msm_job_run(struct drm_sched_job *job) struct msm_fence_context *fctx = submit->ring->fctx; struct msm_gpu *gpu = submit->gpu; struct msm_drm_private *priv = gpu->dev->dev_private; + unsigned nr_cmds = submit->nr_cmds; int i; msm_fence_init(submit->hw_fence, fctx); @@ -36,8 +37,13 @@ static struct dma_fence *msm_job_run(struct drm_sched_job *job) /* TODO move submit path over to using a per-ring lock.. */ mutex_lock(&gpu->lock); + if (submit->queue->ctx->closed) + submit->nr_cmds = 0; + msm_gpu_submit(gpu, submit); + submit->nr_cmds = nr_cmds; + mutex_unlock(&gpu->lock); return dma_fence_get(submit->hw_fence); |