Nicolai Haehnle
826815a5d2
r300: Dump fragment program after translation if RADEON_DEBUG=pixel is set
2007-03-19 20:02:19 +01:00
Nicolai Haehnle
7b992d024b
r300: Whitespace cleanup (remove trailing spaces)
2007-03-19 19:49:06 +01:00
Nicolai Haehnle
c4bf863f4c
r300: Fix WRITEMASK handling when writing to result.depth
...
This is a necessary change to emit the right instructions when writing
to result.depth.
However, even with this test, Z-write doesn't work properly, and I don't
fully understand why. In addition to this, we'll at least have to disable
early-Z, but even that doesn't seem to be enough.
2007-03-19 19:48:58 +01:00
Nicolai Haehnle
b645e8c96d
r300: Streamlined fragment program LIT implementation
...
Fix a bug in the LIT implementation (clamp exponent to 128, not 0.5)
and change the implementation around. In theory, the new implementation
needs as little as 5 instruction slots. Unfortunately, the dependency
analysis in find_and_replace_slot is not strong enough to look at
individual components of a register yet.
2007-03-19 19:10:21 +01:00
Nicolai Haehnle
ff6ab9b45b
r300: Fix fragment program reordering
...
Do not move an instruction that writes to a temp forward past an instruction
that reads the same temporary.
2007-03-19 18:38:07 +01:00
Nicolai Haehnle
ec1a77c864
r300: Fragment program dumps format tex instructions
2007-03-19 18:38:07 +01:00
Nicolai Haehnle
a8e65a010c
r300: Fix hw fragment program dump
...
Dumps of fragment programs were incorrect when the program consisted of multiple
nodes.
Also, improved the formatting a bit.
2007-03-19 18:38:07 +01:00
Nicolai Haehnle
7b430acd71
r300: Fix fragment program instruction pairing and register allocation
...
There were a number of bugs related to the pairing of vector and scalar
operations where swizzles ended up using the wrong source register,
or an instruction was moved forward and ended up overwriting an aliased
register.
The new algorithm for register allocation is quite conservative and may
run out of registers before necessary. On the plus side, It Just Works.
Pairing is done whenever possible, and in more cases than before, so
in practice this change should be a net win.
2007-03-19 18:38:07 +01:00
Michel Dänzer
07db8c9115
i915tex: The intended triple buffering fix.
...
Making modifications while the editor spawned by git-commit was suspended
didn't have the intended effect.
2007-03-19 18:34:27 +01:00
Michel Dänzer
9b42100c04
i915tex: Fix triple buffering after recent Mesa core changes.
...
Remove superfluous _mesa_resize_framebuffer call which is now harmful because
it causes the third renderbuffer to have width/height 0, so Mesa refuses to
render to it.
In the long term, it would be nice to remove the hack in
intel_alloc_window_storage in favour of a proper Mesa interface for flipping
between more than two colour buffers.
2007-03-19 17:23:44 +01:00
Nian Wu
e01ee3da57
Merge git://proxy01.pd.intel.com:9419/git/mesa/mesa into crestline
2007-03-19 17:00:19 +08:00
Keith Whitwell
77544d7b7d
fix off-by-one in load_state_immediate
2007-03-18 20:13:45 +00:00
Keith Whitwell
c9e39aeaef
fix typo in subrect_disable packet
2007-03-18 20:13:45 +00:00
Keith Whitwell
07265280a9
fix typo in subrect_disable packet
2007-03-18 20:13:45 +00:00
Miguel Marte
c41d6ab6f0
screen offset changes, bug 9965
2007-03-18 11:08:29 -06:00
Nicolai Haehnle
441f038748
fix assertion in read_depth_pixels()
2007-03-18 11:02:35 -06:00
Oliver McFadden
dd868b4100
r300: Removed the deprecated $XFree86$ CVS keywords.
2007-03-18 16:57:45 +00:00
Oliver McFadden
1062106521
r300: Corrected the string for polygon mode; it wasn't updated from unkXXX.
2007-03-18 16:45:32 +00:00
Ben Skeggs
bcf4f4de76
nouveau: avoid using uninitialised TexSrcUnit.
2007-03-19 00:13:41 +11:00
Xiang, Haihao
bb02092d74
mesa: SWAP_BUFF support when calling DrawPixels(DEPTH_COMPONENT)
...
or TexImage(DEPTH_COMPONENT)
2007-03-18 18:44:51 +08:00
Xiang, Haihao
b6fe1bdd4d
mesa: enhance fxt1_quantize_ALPHA1
...
If possible, let minCol != maxCol
2007-03-18 18:34:21 +08:00
Xiang, Haihao
84081774e6
i965: fix for FXT1 & S3TC texture format
...
choose the right mesa texformat for FXT1 & S3TC
2007-03-18 18:26:16 +08:00
Nian Wu
fd1b1fce3f
Merge git://proxy01.pd.intel.com:9419/git/mesa/mesa into crestline
2007-03-18 17:00:18 +08:00
Oliver McFadden
cfe984dbd0
Use _mesa_copy_instructions rather than memcpy in _mesa_insert_mvp_code.
2007-03-18 07:35:45 +00:00
Haihao Xiang
ee5b7f0574
fix some format conversion bugs in glGetTexImage(), bug 10288
2007-03-17 09:42:36 -06:00
Nian Wu
38889f5221
Merge git://proxy01.pd.intel.com:9419/git/mesa/mesa into crestline
2007-03-17 17:00:25 +08:00
Dave Airlie
6a9b0cd0b4
r300: change vendor string to DRI R300 Project
2007-03-17 16:46:24 +11:00
Brian
e5070bc3ca
Assorted fixes for dealing with zero-size frame/renderbuffers.
...
In xmesa_check_and_update_buffer_size() handle xmctx==NULL correctly: still
call _mesa_resize_framebufer(). If we don't we can wind up in a situation
where the framebuffer size is non-zero but an attached renderbuffer size
is still initialized to zero. This inconsistancy can later cause problems.
Check for zero-size renderbuffers in update_color_draw_buffers() and
update_color_read_buffer().
See bug 7205.
2007-03-16 11:00:07 -06:00
Brian
7573b58db6
Colortable re-org.
...
The pixel transfer path has three color table lookups.
Use an array [3] to store that info, rather than separate variables.
2007-03-16 09:36:12 -06:00
Brian
d7049431a0
added a renderbuffer comment
2007-03-16 08:36:22 -06:00
Mathias Hopf
f04979ae48
added null xmctx check to XMesaResizeBuffers(), bug 7205
2007-03-16 08:28:34 -06:00
Nian Wu
a02870f4f6
Merge git://proxy01.pd.intel.com:9419/git/mesa/mesa into crestline
2007-03-16 17:00:24 +08:00
Brian
e348016253
silently ignore DeleteProgram/Shader(id=0)
2007-03-15 16:07:39 -06:00
Brian
7ed292a4e9
in _mesa_GetColorTable, return silently if table size is 0
2007-03-15 16:06:58 -06:00
Oliver McFadden
95764262a7
Fixed a minor typo in the comment near _mesa_copy_instructions.
2007-03-15 21:57:31 +00:00
Oliver McFadden
0c25d9ab19
r300: Added _mesa_copy_instructions.
2007-03-15 20:55:30 +00:00
Oliver McFadden
1195caa274
r300: Renamed r300_vertexprog.c to r300_vertprog.c
2007-03-15 20:39:07 +00:00
Oliver McFadden
0e9ada1087
r300: Use _mesa_alloc_instructions/_mesa_init_instructions instead of malloc.
...
Note that insert_wpos in r300_vertexprog.c is still a little flaky and could be
improved.
2007-03-15 19:49:10 +00:00
Oliver McFadden
0d6d80ef3d
r300: Updated R300 to use the new SWIZZLE macros.
2007-03-15 19:09:10 +00:00
Oliver McFadden
51693b22cd
r300: Fixed an unused variable warning and removed some cruft, too.
2007-03-15 18:44:29 +00:00
Oliver McFadden
b3a9a90cdf
r300: Fixed a printf conversion warning.
2007-03-15 18:30:33 +00:00
Oliver McFadden
7f08dd3fc6
r300: Fixed "no previous prototype for 'r300RefillCurrentDmaRegion'" warning.
2007-03-15 18:30:33 +00:00
Brian
fc6b8f2d30
added SWIZZLE_XXXX, YYYY, etc. macros
2007-03-15 11:51:54 -06:00
Oliver McFadden
4e4ab2a62b
Committed Rune Petersen's fragment.position patch (Bug #10024 ) plus a few small
...
corrections.
2007-03-15 17:35:34 +00:00
Brian
4d2eb637a2
no-op clear if buffer width or height is zero (bug 7205)
2007-03-15 11:16:41 -06:00
Brian
038e981cac
add some rb->Data null ptr checks (bug 7205)
2007-03-15 11:11:41 -06:00
Brian
90563d39cb
Merge branch 'master' of git+ssh://brianp@git.freedesktop.org/git/mesa/mesa
2007-03-15 09:03:03 -06:00
Brian
0cfdf432e4
implement byteswapping for all multi-byte types in _mesa_pack_rgba_span_float(), bug 10298
2007-03-15 09:02:14 -06:00
Keith Whitwell
4a7fe4fcfa
Fix off by one error in immediate state packet size.
2007-03-15 10:29:38 +00:00
Nian Wu
d63eef4b86
Merge git://proxy01.pd.intel.com:9419/git/mesa/mesa into crestline
2007-03-15 17:00:22 +08:00
