[Intel-gfx] [PATCH 23/28] drm: use new iterator in drm_gem_fence_array_add_implicit v3
Daniel Vetter
daniel at ffwll.ch
Tue Oct 19 13:59:18 UTC 2021
On Tue, Oct 19, 2021 at 02:54:04PM +0200, Christian König wrote:
> Am 13.10.21 um 16:21 schrieb Daniel Vetter:
> > On Tue, Oct 05, 2021 at 01:37:37PM +0200, Christian König wrote:
> > > Simplifying the code a bit.
> > >
> > > v2: add missing rcu_read_lock()/unlock()
> > > v3: switch to locked version
> > >
> > > Signed-off-by: Christian König <christian.koenig at amd.com>
> > > Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
> > Please make sure you also apply this to the new copy of this code in
> > drm/sched. This one here is up for deletion, once I get all the driver
> > conversions I have landed ...
>
> Yeah, I do have that. Only added this patch here for completeness so that I
> could at least consider dropping the old access functions.
>
> Put I will hold it back, just ping me when the code in question is removed.
Imo go ahead and push this, the rebasing is trivial since I just delete
code. That way I don't have to remember to do anything :-)
-Daniel
>
> Christian.
>
> > -Daniel
> >
> > > ---
> > > drivers/gpu/drm/drm_gem.c | 26 +++++---------------------
> > > 1 file changed, 5 insertions(+), 21 deletions(-)
> > >
> > > diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c
> > > index 09c820045859..4dcdec6487bb 100644
> > > --- a/drivers/gpu/drm/drm_gem.c
> > > +++ b/drivers/gpu/drm/drm_gem.c
> > > @@ -1340,31 +1340,15 @@ int drm_gem_fence_array_add_implicit(struct xarray *fence_array,
> > > struct drm_gem_object *obj,
> > > bool write)
> > > {
> > > - int ret;
> > > - struct dma_fence **fences;
> > > - unsigned int i, fence_count;
> > > -
> > > - if (!write) {
> > > - struct dma_fence *fence =
> > > - dma_resv_get_excl_unlocked(obj->resv);
> > > -
> > > - return drm_gem_fence_array_add(fence_array, fence);
> > > - }
> > > + struct dma_resv_iter cursor;
> > > + struct dma_fence *fence;
> > > + int ret = 0;
> > > - ret = dma_resv_get_fences(obj->resv, NULL,
> > > - &fence_count, &fences);
> > > - if (ret || !fence_count)
> > > - return ret;
> > > -
> > > - for (i = 0; i < fence_count; i++) {
> > > - ret = drm_gem_fence_array_add(fence_array, fences[i]);
> > > + dma_resv_for_each_fence(&cursor, obj->resv, write, fence) {
> > > + ret = drm_gem_fence_array_add(fence_array, fence);
> > > if (ret)
> > > break;
> > > }
> > > -
> > > - for (; i < fence_count; i++)
> > > - dma_fence_put(fences[i]);
> > > - kfree(fences);
> > > return ret;
> > > }
> > > EXPORT_SYMBOL(drm_gem_fence_array_add_implicit);
> > > --
> > > 2.25.1
> > >
>
--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
More information about the Intel-gfx
mailing list