> Note that before this patch, someone trying to get a new job for an FB > that already has a job running would actually get the old job object > and might start messing up with it. Not sure that's what we want.
I don't think that's what we want either, and I think I took pains to ensure this would never happen (at the expense of perf). I recall reading the blob will just keep rendering to future frames speculatively "until SurfaceFlinger stops giving us framebuffers", so that matches moreso the behaviour you're adding. On further thought, I think this is R-b, thank you :) > > > > On Fri, Aug 02, 2019 at 12:12:50PM +0200, Boris Brezillon wrote: > > > This guarantees that new draws targetting the same framebuffer will > > > get a new job instance. > > > > > > Signed-off-by: Boris Brezillon <boris.brezil...@collabora.com> > > > --- > > > src/gallium/drivers/panfrost/pan_job.c | 8 ++++++++ > > > 1 file changed, 8 insertions(+) > > > > > > diff --git a/src/gallium/drivers/panfrost/pan_job.c > > > b/src/gallium/drivers/panfrost/pan_job.c > > > index 960c8556e2f0..d2a4c8c3c600 100644 > > > --- a/src/gallium/drivers/panfrost/pan_job.c > > > +++ b/src/gallium/drivers/panfrost/pan_job.c > > > @@ -173,6 +173,14 @@ panfrost_job_submit(struct panfrost_context *ctx, > > > struct panfrost_job *job) > > > > > > if (ret) > > > fprintf(stderr, "panfrost_job_submit failed: %d\n", ret); > > > + > > > + /* Remove the job from the ctx->jobs set so that future > > > + * panfrost_get_job() calls don't see it. > > > + * We must reset the job key to avoid removing another valid > > > entry when > > > + * the job is freed. > > > + */ > > > + _mesa_hash_table_remove_key(ctx->jobs, &job->key); > > > + memset(&job->key, 0, sizeof(job->key)); > > > } > > > > > > void > > > -- > > > 2.21.0 > > > >
signature.asc
Description: PGP signature
_______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
