On Wed, Oct 26, 2016 at 12:05:44AM +0100, Chris Wilson wrote: > On Tue, Oct 25, 2016 at 12:19:29AM +0100, Robert Bragg wrote: > > + /* So that we don't have to worry about updating the context ID > > + * in OACONTOL on the fly we make sure to pin the context > > + * upfront for the lifetime of the stream... > > + */ > > + vma = stream->ctx->engine[RCS].state; > > There's a caveat here that suggests I had better wrap up this into its > own function. (We need to flush dirty cachelines to memory on first > binding of the context.)
Not that actually affects hsw. -Chris -- Chris Wilson, Intel Open Source Technology Centre _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx
