-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 11/15/2011 06:16 PM, Eric Anholt wrote: > On Sun, 13 Nov 2011 22:32:15 -0800, Chad Versace > <chad.vers...@linux.intel.com> wrote: >> In intel_map_renderbuffer_s8(), detile and copy the stencil buffer into >> the temporary buffer only if the renderbuffer is mapped in read mode. If >> the caller never going to read the bits but just clobber them, then it's >> wasted effort to detile the stencil buffer. >> >> CC: Eric Anholt <e...@anholt.net> >> Signed-off-by: Chad Versace <chad.vers...@linux.intel.com> > > I don't think this is correct. > > Imagine something that isn't writing all of the pixels, but definitely > isn't reading data before writing. It might sensibly ask for > GL_WRITE_BIT but not GL_READ_BIT, and then all the pixels it doesn't > write get trashed when the whole rectangle is written back out. > > If the mode contained GL_MAP_INVALIDATE_RANGE_BIT (basically, > interpreting the mode argument like GL_ARB_map_buffer_range, which was > my intent if not Brian's), this would be reasonable.
You're correct. This patch is nakd. - ---- Chad Versace chad.vers...@linux.intel.com -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQIcBAEBAgAGBQJOxCQDAAoJEAIvNt057x8iMnQQALJW9MK18IHdQ08RX0lxLDkW Tzy5U3zv+j8NxzdQl4aQyA2lXmpNZHzPB6L3AV0Yog+I3ey63A0ADNi94brbu4LS EX/tu+Vb0VsSTC9/aSazDDLQJbLecRo2ZdYueOn/3eKJONmHeCZz8XZrn5wgSE1W Uh4dSR7cORw1vBZBugB5vD55JlRrZtDM45d6s/VRzY2O2ujrBQvr/lNbTZ35HNd7 zIL+67y50RBWMYIiyyzyXHa8lWVwllo1K9CM0LBmebxcX5t6rCn5QyZcT5F0DQAn CyKpKBVDt2mkaOqgnFUiXSTETTOKCla9tzBjNS6OlI65wAmYRZZz0he1rrdl1g6Y kKkI61qWwCOyKMLdRxCWoXYY3QmenIEe4++idtlTauFgKtSZE2v4oAx2X1Z4FvY8 vbcpje3tbt7khMQnFVrVU/nHIyKw6Ko2r+WP32Cr2j044f1VAj+N1zIpCNdPTsd6 ZvJWdAUcPM+vN9fXacmWJAheE2sfzXCJ9o6ffCpRs7LYSv2ZuVlpyZ53a06D5dl/ ya9A9dMbjmFr0MGfELWHKgyRv+Cx/0XxrgvmQNxz6O4TUp6V+xiE6Qiyr5kj90wI bah5zurL3mRVTH0Dp/pQFOxiIIzovCTXGHi35ymRL1/rLWyhgApa2jngVqqeawEi 9Ea+IzVuGlyke/v2kocL =PH5l -----END PGP SIGNATURE----- _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
