On Wed, 14 Dec 2011 14:38:52 -0700, Brian Paul <[email protected]> wrote: > On 12/14/2011 02:13 PM, Eric Anholt wrote: > > On Tue, 13 Dec 2011 18:07:05 -0700, Brian Paul<[email protected]> > > wrote: > >> From: Brian Paul<[email protected]> > >> > >> This fixes two things: > >> 1. If the texture object was created with glTexStorage2D, the call > >> to _mesa_TexImage2D() would generate INVALID_OPERATION since the > >> texture is marked as immutable. > >> 2. _mesa_TexImage2D() always frees any existing texture image memory > >> before allocating new memory. That's inefficient since the existing > >> image is usually the right size already. Now we only make the call > >> when necessary. > >> > >> v2: use _mesa_TexImage() in prepare_dest_image() to make sure side-effects > >> of changing a texture image are observed (like FBO completeness). > > > > Should this live in main/mipmap.c? It looks like > > generate_mipmap_uncompressed() needs it. > > I think the broader question is are we updating fbo completeness when > we change any texture image by mipmap generation, right? > > We're probably missing that in a few places. > > How about I look into that after this meta fix? R-b?
If you put the function in main/mipmap.c where you're about to move it to when you look at it after the meta fix :)
pgp9mfh48V3j9.pgp
Description: PGP signature
_______________________________________________ mesa-dev mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/mesa-dev
