On Mon, Jul 30, 2012 at 11:17 AM, Christian König <deathsimple@xxxxxxxxxxx> wrote: > On 30.07.2012 16:56, Jerome Glisse wrote: >> >> On Sun, Jul 29, 2012 at 1:04 PM, Marek Olšák <maraeo@xxxxxxxxx> wrote: >>> >>> If we don't need stencil, don't allocate it. >>> If we need only stencil (like PIPE_FORMAT_S8_UINT), don't allocate depth. >>> >>> v2: actually do it correctly >> >> Big NAK >> >> We need to allocate stencil and depth no matter what. Otherwise it >> will lockup. You can take a look by poisonning the surface and see >> that when stencil is disabled or depth is disabled the hw still write >> to it. > > > Really? That bug is new to me, at least on SI that works perfectly > (currently working on it), so on which hardware do you see that behavior? I must have put which GPU are affected in one commit either ddx, mesa or libdrm. From memory all evergreen but not in all case and cayman in all case. Cheers, Jerome > > Anyway, when we have hardware bugs that enabling depth also enables stencil > we should handle that in the hardware specific drivers, not in general > purpose code. > > Cheers, > Christian. At time it was the easiest solution to put it there. Cheers, Jerome _______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/dri-devel
