b375526b50
llvmpipe: Be more conservative with the supported formats.
...
We'll likely support much more formats, but doing this allows to run
more testsuites without immediately hit assertion failures.
2009-11-24 14:25:20 +00:00
c6d663e6dc
wgl: Fix copy'n'paste typo in comment.
2009-11-24 14:25:20 +00:00
52df532b02
llvmpipe: Fix typo in comparison operator.
2009-11-24 14:25:20 +00:00
3c46bbee1b
tgsi: Document Declaration Semantic token and FACE semantic name.
2009-11-24 11:58:01 +01:00
bae9ece685
slang: Fix allocation size.
...
We don't need 16K+ to store a single pointer.
2009-11-24 11:22:03 +01:00
53d9b7d361
mesa: Fix pointer arithmetic.
2009-11-24 11:17:16 +01:00
0c54d76f37
tgsi: Implement predicated instructions in exec.
2009-11-24 09:05:31 +01:00
59a70c364d
tgsi: Add ureg_DECL_loop().
2009-11-24 09:05:29 +01:00
5ee0d9f632
tgsi: Account for gallium shader token representation changes.
2009-11-24 09:05:28 +01:00
57221c5438
gallium: Refactor the instruction predicate TGSI token.
...
Rename it to tgsi_instruction_predicate -- it's no longer an extended
token. Its presence is indicated by a new flag in tgsi_instruction that
indicates whether an instruction is predicated.
Also, change predicate index representation to match the other
tokens that specify register indices.
2009-11-24 09:05:26 +01:00
2176b3ed9a
r300: fix swtcl bo leak problem.
...
We can get a lot of swtcl bo allocations - need to probably abstract
this a bit further.
Signed-off-by: Dave Airlie <airlied@redhat.com >
2009-11-24 11:56:45 +10:00
bd13e6e5e2
radeon/r200/r300/r600: make bo mapping be explicit
...
This moves the bo mapping outside the DMA layer and makes it explicit,
this should in theory make it simpler to split the clean up the dma/cmdbuf
linkage that I created before that is broken.
Tested on: r600, rv380 (tcl/no-tcl), rv200 (tcl/no-tcl)
Signed-off-by: Dave Airlie <airlied@redhat.com >
2009-11-24 11:38:36 +10:00
863ad9a683
mesa: use gcc __builtin_popcount()
2009-11-23 18:09:46 -07:00
8d80b5400a
r300g: use util_bitcount()
2009-11-23 18:06:19 -07:00
0a27c7b963
egl: use util_bitcount()
2009-11-23 18:04:47 -07:00
15740eb03c
gallium/util: added util_bitcount()
2009-11-23 18:04:22 -07:00
5173d14cb5
Merge commit 'origin/mesa_7_7_branch'
2009-11-23 22:14:48 +01:00
960464e42d
radeon: fix errors in miptree related function
...
- typo
- memory leak
- off by one (spotted by airlied)
2009-11-23 21:59:08 +01:00
635823d267
Merge branch 'mesa_7_7_branch' of http://anongit.freedesktop.org/git/mesa/mesa into mesa_7_7_branch
2009-11-23 21:28:22 +01:00
86710c3334
svga: Scrub Makefiles a bit
...
Remove x86 specific hacks. Not that they will ever be used on
none x86 arches, but they are built by default. And the way the
flags where added was a hack.
2009-11-23 20:09:00 +01:00
18384af749
slang: Check return value from emit_instruction().
2009-11-23 13:22:04 +01:00
c511e0b844
tgsi: Clamp the source argument in micro_exp2() to avoid Inf.
2009-11-23 11:32:58 +01:00
eacdd8fa75
tgsi: Remove code that actually had no effect.
2009-11-23 11:29:29 +01:00
cc93fa3527
softpipe: Initialise TGSI machine's Face.
2009-11-23 10:51:07 +01:00
cc35a454da
tgsi: Fix POSITION and FACE fragment shader inputs.
2009-11-23 10:49:41 +01:00
b7590cde4a
tgsi: Bring BGNFOR/ENDFOR implementation up to spec.
2009-11-23 10:40:05 +01:00
cde758a2b5
tgsi: Fake TXD implementation in exec.
2009-11-23 10:29:08 +01:00
f359ac5486
tgsi: Add execution debugging facilities to exec.
2009-11-23 10:29:07 +01:00
f61865799d
slang: Be more robust with memory in concat_shaders().
2009-11-23 10:29:05 +01:00
67125c7f9a
mesa: Initialize variable in _mesa_get_texel_fetch_func.
2009-11-23 01:52:59 -05:00
4b0b250aae
glx: Prevent potential null pointer deference in driCreateContext.
2009-11-23 01:30:32 -05:00
5b925b7daa
glu/sgi: Fix memory leak in bitmapBuild2DMipmaps.
2009-11-23 01:09:06 -05:00
b611f639b4
glu/sgi: Fix memory leak in gluScaleImage3D.
2009-11-23 00:57:37 -05:00
a12b468d00
r600 : add support for shader instruction trunc and discard.
2009-11-22 21:37:44 -05:00
c3c8c40cab
r600: hopefully fix segfault.
2009-11-23 06:44:29 +10:00
f9b0f1dfa1
r600: fix inline issues
2009-11-23 06:31:29 +10:00
1f8c23d9db
r600 : add stack depth calculation, enable CF pop.
2009-11-22 13:03:20 -05:00
a02938a842
r600 : use cf for all pop now, left optimization for future.
2009-11-22 13:03:19 -05:00
e0fda04013
r300: fix VP source conflict resolution on 64-bit machines
...
On 32bit machines we were lucky because the sizeof(reg) == sizeof(rc_src_register).
On 64bit machines pointers are 8 bytes long, so we were overwriting other data.
2009-11-22 15:08:46 +01:00
dfe440c856
r300: fix SIN/COS/SCS instructions for R300 fp
2009-11-22 14:13:18 +01:00
a9c540f5de
glu/sgi: Fix memory leak in gluScaleImage.
2009-11-22 01:57:35 -05:00
d3b4c99c70
glu: Fix memory leak in __gl_meshMakeEdge.
2009-11-22 01:45:53 -05:00
57d389aab5
svga: Prevent potential null pointer deference in vmw_surface.c.
2009-11-22 01:26:32 -05:00
b62a74d3b9
svga: Fix memory leak in vmw_screen_ioctl.c
2009-11-22 01:20:07 -05:00
88aab56a26
st/xorg: Prevent potential null pointer deference in xorg_xv.c.
2009-11-22 01:10:38 -05:00
46feb7db71
st/xorg: Reorder cases in switch statement.
...
Silences missing break statement warning.
2009-11-22 00:55:52 -05:00
2b07b64061
r300g: Build fix.
...
Oops.
2009-11-21 19:13:26 -08:00
44c0aaf990
r300g: do not reallocate r300_vertex_info and r300_rs_block all the time
2009-11-21 14:52:26 -08:00
3a2cd66af8
r300g: clean up vs/fs tabs
...
Instead of vs_tab, we use vs_output_tab and it's local now. fs_tab hasn't
been used anywhere, so I removed it and r300_update_fs_tab too.
2009-11-21 14:52:26 -08:00
563fe6e8f6
radeon: fix compressed mipmapped textures
...
Tested on r300 only, other cards may require adjusting texture_compressed_row_align.
2009-11-21 16:49:18 +01:00
José Fonseca