On Thu, Aug 18, 2016 at 01:56:56PM +0000, Zanoni, Paulo R wrote: > Em Qui, 2016-08-18 às 09:21 +0100, Chris Wilson escreveu: > > Only fbc1 is tied to using a fence. Later iterations of fbc are more > > flexible and allow operation on unfenced frontbuffers. > > But then we'll lose GTT tracking - which we currently rely on - and I'm > 87.5% sure we'll need to implement some workarounds we skipped.
You've missed the patches that point out that the GTT tracking is broken? :) We don't rely on GTT tracking at all because we don't use the tracked GTT address for pwrite or GTT mmap access (and not for WC or CPU mmap access). And CS uses a completely different address space (caveat those older gen in which it looks like the GGTT). -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
