summaryrefslogtreecommitdiffstats
path: root/src/gallium
Commit message (Collapse)AuthorAgeFilesLines
* r600g: Remove unused variable.Vinson Lee2010-09-251-1/+0
| | | | | | Fixes this GCC warning. radeon_bo_pb.c: In function 'radeon_bo_pb_create_buffer': radeon_bo_pb.c:178: warning: unused variable 'domain'
* softpipe: minor asst. clean-upsBrian Paul2010-09-252-12/+11
|
* softpipe: make clip state functions staticBrian Paul2010-09-253-26/+26
|
* softpipe: make stream out state functions staticBrian Paul2010-09-253-24/+24
|
* softpipe: make rasterizer state functions staticBrian Paul2010-09-253-16/+19
|
* softpipe: make vertex state functions staticBrian Paul2010-09-253-22/+24
|
* softpipe: make sampler state functions staticBrian Paul2010-09-253-59/+31
|
* softpipe: make blend/stencil/depth functions staticBrian Paul2010-09-253-50/+44
|
* softpipe: make shader-related functions staticBrian Paul2010-09-253-63/+58
|
* softpipe: rename sp_state_fs.c -> sp_state_shader.cBrian Paul2010-09-253-2/+2
|
* st/dri: Remove unnecessary header.Vinson Lee2010-09-251-1/+0
|
* gallivm: fix repeat() function for NPOT texturesBrian Paul2010-09-252-10/+17
| | | | | The trick of casting the coord to an unsigned value only works for POT textures. Add a bias instead. This fixes a few piglit texwrap failures.
* softpipe: fix repeat() function for NPOT texturesBrian Paul2010-09-251-4/+4
| | | | | The trick of casting the coord to an unsigned value only works for POT textures. Add a bias instead. This fixes a few piglit texwrap failures.
* r600g: Disable unused variables.Vinson Lee2010-09-251-1/+1
| | | | | | | | | The variables are used only in currently disabled code. Fixes this GCC warning. r600_context.c: In function 'r600_flush': r600_context.c:76: warning: unused variable 'dname' r600_context.c:75: warning: unused variable 'dc'
* r600g: Remove unused variable.Vinson Lee2010-09-251-1/+1
| | | | | | Fixes this GCC warning. r600_draw.c: In function 'r600_draw_common': r600_draw.c:71: warning: unused variable 'format'
* r600g: Remove unused variable.Vinson Lee2010-09-251-1/+0
| | | | | | Fixes this GCC warning. r600_screen.c: In function 'r600_screen_create': r600_screen.c:239: warning: unused variable 'family'
* nv50: fix/handle a few more PIPE_CAPsChristoph Bumiller2010-09-251-1/+6
|
* nv50: use CLEAR_BUFFERS for surface fillsChristoph Bumiller2010-09-253-22/+94
| | | | The 2D engine's fill doesn't seem suited for RGBA32F or ZS buffers.
* nv50: use formats table in nv50_surface.cChristoph Bumiller2010-09-252-52/+31
|
* r600g: fix vertex resource & polygon offsetJerome Glisse2010-09-253-34/+127
| | | | Signed-off-by: Jerome Glisse <[email protected]>
* r600g: add eg db count control register.Dave Airlie2010-09-253-9/+12
|
* r600g: make index bias fix for evergreenDave Airlie2010-09-251-1/+1
|
* gallivm: Remove dead experimental code.José Fonseca2010-09-252-78/+46
|
* llvmpipe: handle up to 8 planes in triangle binnerKeith Whitwell2010-09-251-7/+8
|
* llvmpipe: handle FACING interpolants in line and point setupKeith Whitwell2010-09-252-3/+19
|
* gallivm: Fetch the lod from the dynamic state when min_lod == max_lod.José Fonseca2010-09-252-5/+4
|
* draw: Fullfil the new min_lod/max_lod/lod_bias/border_color dynamic stateJosé Fonseca2010-09-253-0/+33
|
* gallivm: optimize yuv decodingRoland Scheidegger2010-09-251-9/+55
| | | | | | | | | | | this is more a proof to show vector shifts on x86 with per-element shift count are evil. Since we can avoid the shift with a single compare/select, use that instead. Replaces more than 20 instructions (and slow ones at that) with about 3, and cuts compiled shader size with mesa's yuvsqure demo by over 10% (no performance measurements done - but selection is blazing fast). Might want to revisit that for future cpus - unfortunately AVX won't have vector shifts neither, but AMD's XOP will, but even in that case using selection here is probably not slower.
* gallivm: don't use URem/UDiv when calculating offsets for blocksRoland Scheidegger2010-09-251-1/+11
| | | | | | | While it's true that llvm can and will indeed replace this with bit arithmetic (since block height/width is POT), it does so (llvm 2.7) by element and hence extracts/shifts/reinserts each element individually. This costs about 16 instructions (and extract is not really fast) vs. 1...
* gallivm: fix copy&paste bugRoland Scheidegger2010-09-251-1/+1
| | | | | looks like pot_depth should be used, not pot_height (found by accident, not verified)
* r600g: add eg poly mode code.Dave Airlie2010-09-252-1/+23
|
* r600g: Silence 'control reaches end of non-void function' warning.Vinson Lee2010-09-241-0/+3
| | | | | | Fixes this GCC warning. r600_hw_states.c: In function 'r600_translate_fill': r600_state_inlines.h:136: warning: control reaches end of non-void function
* r600g: Remove unused variable.Vinson Lee2010-09-241-1/+0
| | | | | | Fixes this GCC warning. eg_hw_states.c: In function 'eg_resource': eg_hw_states.c:525: warning: unused variable 'r'
* r600g: Disable unused variables.Vinson Lee2010-09-241-1/+1
| | | | | | | | | The variables are only used in currently disabled code. Fixes this GCC warning. r600_state2.c: In function 'r600_flush2': r600_state2.c:613: warning: unused variable 'dname' r600_state2.c:612: warning: unused variable 'dc'
* r600g: Remove unused variable.Vinson Lee2010-09-241-1/+0
| | | | | | Fixes this GCC warning. r600_buffer.c: In function 'r600_buffer_transfer_map': r600_buffer.c:141: warning: unused variable 'rctx'
* r600g: Remove unnecessary header.Vinson Lee2010-09-241-1/+0
|
* r300g: Silence uninitialized variable warning.Vinson Lee2010-09-241-0/+1
| | | | | | | Silence this GCC warning. r300_state_derived.c: In function 'r300_update_derived_state': r300_state_derived.c:578: warning: 'r' may be used uninitialized in this function r300_state_derived.c:578: note: 'r' was declared here
* util: fix util_pack_color for B4G4R4A4Marek Olšák2010-09-251-1/+1
| | | | NOTE: This is a candidate for the 7.9 branch.
* r600g: bring over fix from old path to new pathJerome Glisse2010-09-244-34/+212
| | | | | | | | | | Up to 2010-09-19: r600g: fix tiling support for ddx supplied buffers 9b146eae2521d8e5f6d3cbefa4f6f7737666313a user buffer seems to be broken... new to fix that. Signed-off-by: Jerome Glisse <[email protected]>
* r600g: fix evergreen new pathJerome Glisse2010-09-243-11/+16
| | | | | | glxgears seems to work, had somelockup but now they seems to have vanish. Signed-off-by: Jerome Glisse <[email protected]>
* r600g: fix reg definitionJerome Glisse2010-09-241-1/+3
| | | | | | Doesn't bother fixing old path code, just disable that reg. Signed-off-by: Jerome Glisse <[email protected]>
* r600g: fix evergreen new pathJerome Glisse2010-09-244-178/+271
| | | | Signed-off-by: Jerome Glisse <[email protected]>
* r600g: fixup some evergreen register definitionsJerome Glisse2010-09-242-56/+56
| | | | Signed-off-by: Jerome Glisse <[email protected]>
* llvmpipe: make texture border_color dynamic stateBrian Paul2010-09-247-27/+63
|
* llvmpipe: make min/max lod and lod bias dynamic stateBrian Paul2010-09-2410-24/+79
| | | | | | Before, changing any of these sampler values triggered generation of new JIT code. Added a new flag for the special case of min_lod == max_lod which is hit during auto mipmap generation.
* r600g: fix compilation after change to evergreend.hJerome Glisse2010-09-241-2/+2
| | | | Signed-off-by: Jerome Glisse <[email protected]>
* r600g: evergreen fix for new designJerome Glisse2010-09-246-81/+59
| | | | Signed-off-by: Jerome Glisse <[email protected]>
* r600g: move use_mem_constants flags for new designs structure alignmentJerome Glisse2010-09-241-1/+1
| | | | Signed-off-by: Jerome Glisse <[email protected]>
* r600g: fix typo in evergreen define (resource are in [0x30000;0x34000] range)Jerome Glisse2010-09-241-1/+1
| | | | Signed-off-by: Jerome Glisse <[email protected]>
* d3d1x: CRLF -> LF in progsLuca Barbieri2010-09-2426-5300/+5300
|
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-28 21:46:22 +0000