Commit graph

25784 commits

Author SHA1 Message Date
Eric Anholt
cbde276580 Revert "i965: Use VBOs in the VBO module on 965, now that we have ARB_map_buffer_range."
This reverts commit 00413d8742.  Even with
fixes, using ARB_map_buffer_range in the VBO module isn't showing up as a
significant win, and some cases apparently regressed.

Bug #23624.
2009-09-02 12:20:01 -07:00
Eric Anholt
ec32ef12b3 intel: Add support for FlushMappedBufferRange for ARB_map_buffer_range.
This should help for the usage by the VBO module, where we would upload
the whole remaining chunk of the buffer for a series of range maps that should
cover just a segment of it.
2009-09-02 12:20:00 -07:00
Eric Anholt
82313eed01 intel: Sync a synchronized READ_BIT map buffer range with GL drawing to it.
It's probably uncommon, but would obviously have gone wrong.
2009-09-02 12:20:00 -07:00
Eric Anholt
aecba6b1c8 intel: Move MapBufferRange mesa state setting up to cover the 915 case. 2009-09-02 12:20:00 -07:00
Jakob Bornecrantz
7d8f33590f st/xorg: Guard against realy old versions of Xorg 2009-09-02 19:48:49 +02:00
Brian Paul
5e809216a8 Revert "mesa: fix the non-GNU C bit-field case"
This reverts commit 4b08e74982.

Don't know what I was thinking there.
2009-09-02 10:38:46 -06:00
Brian Paul
4b08e74982 mesa: fix the non-GNU C bit-field case 2009-09-02 10:12:15 -06:00
Gary Wong
d98b9f4edb mesa: silence gcc bit-field warning 2009-09-02 10:11:15 -06:00
Brian Paul
a7a3d3c271 mesa: replace 8 with NUM_UNITS 2009-09-02 09:43:10 -06:00
Brian Paul
8cb651578d mesa: remove accidentally committed printf 2009-09-02 09:25:28 -06:00
Brian Paul
83e4470d9a mesa: added #ifdef __GNUC__ around GLubyte bitfield usage
It would be nice if there were a #pragma or something to disable the warnings:

main/texenvprogram.c:87: warning: type of bit-field ‘Source’ is a GCC extension
main/texenvprogram.c:88: warning: type of bit-field ‘Operand’ is a GCC extension

but there doesn't appear to be a way to do that.
2009-09-02 08:47:47 -06:00
Chris Wilson
a46e32787d mesa: Compact state key for TexEnv program cache
By rearranging the bitfields within the key we can reduce the size
of the key from 644 to 196 bytes, reducing the cost of both the
hashing and equality tests.
2009-09-02 08:47:47 -06:00
Michel Dänzer
c0cf0fd163 st/xorg: Remove code no longer support by xserver Git master.
Shouldn't be necessary with older versions either.
2009-09-02 16:01:21 +02:00
Michel Dänzer
ea55f8ed8f st/xorg: Add support for xserver Git master DRI2 interface.
Reorganize to handle both interfaces with common code.
2009-09-02 16:00:30 +02:00
Michel Dänzer
0b26d97f4d st/dri: Slight consolidation of DRI2 buffer handling.
Always use / assume a combined depth/stencil buffer.

Also update drawable formats from surfaces returned by the driver.
2009-09-02 15:41:41 +02:00
Michel Dänzer
385620e5eb st/xorg: Add EXA UploadToScreen and DownloadFromScreen hooks. 2009-09-02 15:41:40 +02:00
Xiang, Haihao
98a8744e02 i965: CS FENCE in URB_FENCE is 11-bits wide 2009-09-02 13:28:32 +08:00
Xiang, Haihao
f5539b6991 i965: validate sf state 2009-09-02 13:28:23 +08:00
Eric Anholt
60b08eb1fd mesa: Make MultiDrawElements submit multiple primitives at once.
Previously, MultiDrawElements just called DrawElements a bunch of times.
By sending several primitives down the pipeline at once, we avoid a bunch
of validation.  On my GL demo, this improves fps by 2.5% (+/- .41%) and
reduces CPU usage by 70.5% (+/- 2.9%) (n=3).

Reviewed by: Ian Romanick <ian.d.romanick@intel.com>
2009-09-01 20:35:19 -07:00
Ben Skeggs
a82cd55a5c nv50: SWZ is the same as MOV from our perspective 2009-09-02 12:05:34 +10:00
Brian Paul
282f578dbd mesa: skip bitmap drawing code if width==0 or height==0 2009-09-01 17:40:07 -06:00
Brian Paul
29e22059a9 intel: use _mesa_expand_bitmap() to skip an intermediate buffer 2009-09-01 17:39:32 -06:00
Brian Paul
d5b6dd0001 st/mesa: use new _mesa_expand_bitmap() function 2009-09-01 17:39:32 -06:00
Brian Paul
278e76832f mesa: new _mesa_expand_bitmap() function 2009-09-01 17:39:32 -06:00
Brian Paul
921bbfc942 mesa: remove redundant assignments 2009-09-01 17:39:32 -06:00
Brian Paul
457e427e32 mesa: more clean-ups 2009-09-01 17:39:32 -06:00
Brian Paul
84d6bed4d6 mesa: change conditional to match the previous one 2009-09-01 17:39:32 -06:00
Brian Paul
25e5a6f279 mesa: updated #includes 2009-09-01 17:39:32 -06:00
Brian Paul
95136694ab mesa: remove unused texenv_fragment_program::ctx field 2009-09-01 17:39:32 -06:00
Brian Paul
1480bcacfe mesa: remove unused ureg::abs field 2009-09-01 17:39:32 -06:00
Brian Paul
c7ba2b8622 mesa: remove unused ureg:negateabs field 2009-09-01 17:39:32 -06:00
Brian Paul
9ed0315271 mesa: more comments, clean-ups 2009-09-01 17:39:31 -06:00
Brian Paul
b5ec0a6858 mesa: simplify translate_tex_src_bit() 2009-09-01 17:39:31 -06:00
Brian Paul
f337e2c495 mesa: minor code clean-ups, comments 2009-09-01 17:39:31 -06:00
Brian Paul
d55a28e3dc mesa: replace 8 with MAX_TEXTURE_UNITS 2009-09-01 17:39:31 -06:00
Brian Paul
51334d8401 dri: remove unused meta_clear_tris() 2009-09-01 17:39:31 -06:00
Brian Paul
c2542cdaff intel: use BUFFER_BITS_COLOR 2009-09-01 17:39:31 -06:00
Brian Paul
5353961344 intel: fix incorrect parameter type for intel_bufferobj_map_range() 2009-09-01 17:39:31 -06:00
Brian Paul
85d3562385 radeon: trim down #includes 2009-09-01 17:39:31 -06:00
Brian Paul
18916986b4 radeon: use _mesa_meta_clear() 2009-09-01 17:39:30 -06:00
Brian Paul
92d63931e2 mesa: obey stencil write mask in _mesa_meta_draw_pixels() 2009-09-01 17:39:30 -06:00
Brian Paul
a1e869b417 intel: set Length/Offset fields in intel_bufferobj_map() 2009-09-01 17:39:30 -06:00
Brian Paul
15e6d08670 intel: use _mesa_meta_copy_pixels() when do_blit_copypixels() fails
Also, trim down #includes.
2009-09-01 17:39:30 -06:00
Brian Paul
be229d24c4 intel: trim down #includes 2009-09-01 17:39:30 -06:00
Brian Paul
886a0a7150 intel: use _mesa_meta_draw_pixels()
The textured quad path is slightly faster and will work with POT textures
on i945.
2009-09-01 17:39:30 -06:00
Brian Paul
90cba4a839 intel: trim down #includes 2009-09-01 17:39:30 -06:00
Brian Paul
50d8b295f6 intel: use _mesa_meta_clear(), it's a bit faster 2009-09-01 17:39:30 -06:00
Pauli Nieminen
06bf6628aa radeon: Fix OQ to set ful lstate as dirty too. 2009-09-02 02:04:10 +03:00
Pauli Nieminen
9f650e19dd radeon: Fix debug output to filter out less critical messages instead of more critical. 2009-09-02 02:04:10 +03:00
Zack Rusin
7e4487ad5c st/xorg: getting indexes right tends to be helpful 2009-09-01 19:01:05 -04:00