----- Original Message ----- > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On 05/19/2011 08:14 PM, Adam Jackson wrote: > > Despite that negative values aren't sensible here, making this > > unsigned > > is dangerous. Consider get_pointer_generic, which computes a value > > of > > the form: > > > > void *base + (int x * int stride + int y) * unsigned bpp > > > > The usual arithmetic conversions will coerce the (x*stride + y) > > subexpression to unsigned. Since stride can be negative, this is > > disastrous. > > > > Fixes at least the following piglit tests on Ironlake: > > > > fbo/fbo-blit-d24s8 > > spec/ARB_depth_texture/fbo-clear-formats > > spec/EXT_packed_depth_stencil/fbo-clear-formats > > > > Signed-off-by: Adam Jackson <a...@redhat.com> > > I don't care which approach we go with. This will likely fix it > everywhere, but I'd like there to be a comment explaining why the > return > type must not be unsigned. Otherwise someone will come along in the > future and make it unsigned in the name of "code cleanup."
I agree with Ian: Adam's patch is good, but a comment would be useful, as the implicit coercion to unsigned is both unfortunate and unintuitive. Jose _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
