Am Dienstag, den 05.06.2018, 12:03 -0700 schrieb Eric Anholt:
> Between creation and queueing of a job, you need to prevent any other
> job from being created and queued.  Otherwise the scheduler's fences
> may be signaled out of seqno order.
>
> > Signed-off-by: Eric Anholt <eric at anholt.net>
> Fixes: 57692c94dcbe ("drm/v3d: Introduce a new DRM driver for Broadcom V3D V3.x+")
> ---
>
> ccing amd-gfx due to interaction of this series with the scheduler.
>
> Â drivers/gpu/drm/v3d/v3d_drv.h |Â Â 5 +++++
> Â drivers/gpu/drm/v3d/v3d_gem.c | 11 +++++++++--
> Â 2 files changed, 14 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/v3d/v3d_drv.h b/drivers/gpu/drm/v3d/v3d_drv.h
> index a043ac3aae98..26005abd9c5d 100644
> --- a/drivers/gpu/drm/v3d/v3d_drv.h
> +++ b/drivers/gpu/drm/v3d/v3d_drv.h
> @@ -85,6 +85,11 @@ struct v3d_dev {
> > Â Â */
> > Â struct mutex reset_lock;
> Â
> > + /* Lock taken when creating and pushing the GPU scheduler
> > + Â * jobs, to keep the sched-fence seqnos in order.
> > + Â */
> > + struct mutex sched_lock;
> +
> > Â struct {
> > Â u32 num_allocated;
> > Â u32 pages_allocated;
> diff --git a/drivers/gpu/drm/v3d/v3d_gem.c b/drivers/gpu/drm/v3d/v3d_gem.c
> index b513f9189caf..9ea83bdb9a30 100644
> --- a/drivers/gpu/drm/v3d/v3d_gem.c
> +++ b/drivers/gpu/drm/v3d/v3d_gem.c
> @@ -550,13 +550,16 @@ v3d_submit_cl_ioctl(struct drm_device *dev, void *data,
> > Â if (ret)
> > Â goto fail;
> Â
> > + mutex_lock(&v3d->sched_lock);
> > Â if (exec->bin.start != exec->bin.end) {
> > Â ret = drm_sched_job_init(&exec->bin.base,
> > Â Â &v3d->queue[V3D_BIN].sched,
> > Â Â &v3d_priv->sched_entity[V3D_BIN],
> > Â Â v3d_priv);
> > - if (ret)
> > + if (ret) {
> > + mutex_unlock(&v3d->sched_lock);
> Â goto fail_unreserve;
I don't see any path where you would go to fail_unreserve with the
mutex not yet locked, so you could just fold the mutex_unlock into this
error path for a bit less code duplication.
Otherwise this looks fine.
Regards,
Lucas
> + }
> Â
> > Â exec->bin_done_fence =
> > Â dma_fence_get(&exec->bin.base.s_fence->finished);
> @@ -570,12 +573,15 @@ v3d_submit_cl_ioctl(struct drm_device *dev, void *data,
> > Â Â &v3d->queue[V3D_RENDER].sched,
> > Â Â &v3d_priv->sched_entity[V3D_RENDER],
> > Â Â v3d_priv);
> > - if (ret)
> > + if (ret) {
> > + mutex_unlock(&v3d->sched_lock);
> > Â goto fail_unreserve;
> > + }
> Â
> > Â kref_get(&exec->refcount); /* put by scheduler job completion */
> > Â drm_sched_entity_push_job(&exec->render.base,
> > Â Â Â &v3d_priv->sched_entity[V3D_RENDER]);
> > + mutex_unlock(&v3d->sched_lock);
> Â
> > Â v3d_attach_object_fences(exec);
> Â
> @@ -615,6 +621,7 @@ v3d_gem_init(struct drm_device *dev)
> > Â spin_lock_init(&v3d->job_lock);
> > Â mutex_init(&v3d->bo_lock);
> > Â mutex_init(&v3d->reset_lock);
> > + mutex_init(&v3d->sched_lock);
> Â
> >  /* Note: We don't allocate address 0.  Various bits of HW
> > Â Â * treat 0 as special, such as the occlusion query counters
