Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Merge branch 'draw-instanced' | Brian Paul | 2011-01-15 | 3 | -1/+6 |
|\ | | | | | | | | | | | | | | | Conflicts: src/gallium/auxiliary/draw/draw_llvm.c src/gallium/drivers/llvmpipe/lp_state_fs.c src/glsl/ir_set_program_inouts.cpp src/mesa/tnl/t_vb_program.c | ||||
| * | llvmpipe: enable instanced drawing cap | Brian Paul | 2010-12-08 | 1 | -0/+2 |
| | | |||||
| * | softpipe: enable instanced drawing cap | Brian Paul | 2010-12-08 | 1 | -0/+2 |
| | | |||||
| * | gallivm/llvmpipe: implement system values and instanceID | Brian Paul | 2010-12-08 | 1 | -1/+2 |
| | | |||||
* | | nvc0: try to swap immediates to first source too | Christoph Bumiller | 2011-01-15 | 1 | -0/+7 |
| | | |||||
* | | nvc0: make sure all sources of the BIND op are distinct | Christoph Bumiller | 2011-01-15 | 3 | -0/+45 |
| | | | | | | | | | | They're supposed to be assigned consecutive registers so they can't contain the same SSA value more than once. | ||||
* | | nvc0: update user vbufs on each draw call | Christoph Bumiller | 2011-01-15 | 4 | -37/+119 |
| | | | | | | | | This is required in case set_vertex_buffers is not called again. | ||||
* | | nvc0: enable early fragment tests where possible | Christoph Bumiller | 2011-01-15 | 3 | -2/+13 |
| | | |||||
* | | nvc0: upload small buffers through the command buffer | Christoph Bumiller | 2011-01-15 | 1 | -0/+6 |
| | | |||||
* | | r600g: compiler helper opcode fixes for evergreen | Alex Deucher | 2011-01-14 | 2 | -120/+269 |
| | | | | | | | | Signed-off-by: Alex Deucher <[email protected]> | ||||
* | | r600g: pass r600_bc to some addition compiler helper functions | Alex Deucher | 2011-01-14 | 1 | -55/+62 |
| | | | | | | | | | | | | needed for asic specific opcodes Signed-off-by: Alex Deucher <[email protected]> | ||||
* | | r600g: Disable V_SQ_ALU_WORD1_OP2_SQ_OP2_INST_FLT_TO_INT_FLOOR case. | Vinson Lee | 2011-01-14 | 1 | -0/+2 |
| | | | | | | | | | | | | The usage of macro V_SQ_ALU_WORD1_OP2_SQ_OP2_INST_FLT_TO_INT_FLOOR was introduced by commit 323ef3a1f07ba4333dadebab571ddcd49d95f45c but the macro is undefined. Disable this case to fix the build for now. | ||||
* | | r600g: add more missing instructions to r600_bc_get_num_operands | Christian König | 2011-01-14 | 1 | -1/+5 |
| | | |||||
* | | r600g: Move declaration before code in r600_asm.c. | Vinson Lee | 2011-01-13 | 1 | -1/+3 |
| | | | | | | | | Fixes SCons build. | ||||
* | | r600g: rework literal handling | Christian König | 2011-01-13 | 5 | -277/+151 |
| | | |||||
* | | r600g: merge alu groups | Christian König | 2011-01-13 | 2 | -37/+150 |
| | | |||||
* | | r600g: implement replacing gpr with pv and ps | Christian König | 2011-01-13 | 3 | -5/+63 |
| | | |||||
* | | r600g: add missing RECIPSQRT_CLAMPED to r600_bc_get_num_operands | Christian König | 2011-01-13 | 1 | -0/+1 |
| | | |||||
* | | r600g: rework bank swizzle code | Christian König | 2011-01-13 | 2 | -183/+174 |
| | | |||||
* | | r600g: fix alu slot assignment | Christian König | 2011-01-13 | 1 | -15/+167 |
| | | |||||
* | | r600g: optimize away CF ALU instructions even if type doesn't match | Christian König | 2011-01-13 | 1 | -3/+16 |
| | | |||||
* | | nvc0: identify POINT_RASTER_RULES, add POINT_SMOOTH state | Christoph Bumiller | 2011-01-13 | 4 | -19/+42 |
| | | | | | | | | | | | | Point smoothing requires rasterization rules to be set to OGL. Sorry for the extra noise caused by the header update. | ||||
* | | r600g: Silence uninitialized variable warnings. | Vinson Lee | 2011-01-13 | 1 | -4/+4 |
| | | |||||
* | | nvc0: disable calling of sw methods we don't implement | Ben Skeggs | 2011-01-13 | 1 | -0/+4 |
| | | | | | | | | | | | | | | Left in the code as a marker of what NVIDIA do, just in case we need to do this some day. Signed-off-by: Ben Skeggs <[email protected]> | ||||
* | | nvc0: fix mp_stack_bo relocation | Ben Skeggs | 2011-01-13 | 1 | -2/+2 |
| | | | | | | | | | | | | | | | | | | | | Fixes a PT_NOT_PRESENT error cause by: - allocating in VRAM - emitting GART relocs to 0x17bc/0x17c0, moving the buffer - telling the bufmgr that the buffer should be in VRAM when we use it, but not correcting the value sent to 0x17bc/0x17c0. Signed-off-by: Ben Skeggs <[email protected]> | ||||
* | | noop: change var type to silence warning | Brian Paul | 2011-01-12 | 1 | -1/+1 |
| | | |||||
* | | r600g: also look at tex inst when for maximum gpu count | Christian König | 2011-01-12 | 1 | -1/+7 |
| | | |||||
* | | r600g: implement output modifiers and use them to further optimize LRP | Christian König | 2011-01-12 | 4 | -0/+33 |
| | | |||||
* | | r600g: use special constants for 0, 1, -1, 1.0f, 0.5f etc | Christian König | 2011-01-12 | 1 | -12/+44 |
| | | |||||
* | | r600g: optimize temp register handling for LRP | Christian König | 2011-01-12 | 1 | -34/+38 |
| | | |||||
* | | r600g: optimize away CF_INST_POP | Christian König | 2011-01-12 | 3 | -3/+29 |
| | | | | | | | | | | If last instruction is an CF_INST_ALU we don't need to emit an additional CF_INST_POP for stack clean up after an IF ELSE ENDIF. | ||||
* | | r600g: make dumping of shaders an option | Christian König | 2011-01-12 | 1 | -4/+14 |
| | | |||||
* | | r600g: fix alu dumping | Christian König | 2011-01-12 | 1 | -19/+13 |
| | | |||||
* | | r600g: improve r600_bc_dump | Christian König | 2011-01-12 | 1 | -28/+131 |
| | | |||||
* | | r600g: texture instructions also work fine with TGSI_FILE_INPUT | Christian König | 2011-01-12 | 1 | -1/+3 |
| | | |||||
* | | r600g: DP4 also supports writemasking | Christian König | 2011-01-12 | 1 | -8/+6 |
| | | |||||
* | | r600g: Why all this fiddling with tgsi_helper_copy? | Christian König | 2011-01-12 | 1 | -21/+41 |
| | | | | | | | | | | | | | | | | tgsi_helper_copy is used on several occasions to copy a temporary result into the real destination register to emulate writemasks for OP3 and reduction operations. According to R600 ISA that's unnecessary. This patch fixes this use for MAD, CMP and DP4. | ||||
* | | r600g: fix tex and vtx joining | Christian König | 2011-01-12 | 1 | -2/+2 |
| | | |||||
* | | r600g: Fixed SIN/COS/SCS for the case where the operand is a literal. | Tilman Sauerbeck | 2011-01-11 | 1 | -2/+15 |
| | | | | | | | | | | Signed-off-by: Tilman Sauerbeck <[email protected]> Reviewed-by: Alex Deucher <[email protected]> | ||||
* | | r600g: move user fence into base radeon structure | Jerome Glisse | 2011-01-11 | 1 | -3/+0 |
| | | | | | | | | | | | | | | This avoid any issue when context is free and we still try to access fence through radeon structure. Signed-off-by: Jerome Glisse <[email protected]> | ||||
* | | r300g: add debug option for buffer upload logging | Marek Olšák | 2011-01-10 | 3 | -0/+9 |
| | | |||||
* | | noop: make noop useable like trace or rbug | Jerome Glisse | 2011-01-09 | 2 | -47/+35 |
| | | | | | | | | | | | | | | | | If you want to enable noop set GALLIUM_NOOP=1 as an env variable. You need first to enable noop wrapping for your driver see change to src/gallium/targets/dri-r600/ in this commit as an example. Signed-off-by: Jerome Glisse <[email protected]> | ||||
* | | r300g: do not upload the same user buffer several times | Marek Olšák | 2011-01-09 | 1 | -1/+3 |
| | | | | | | | | Performance++. | ||||
* | | nvc0: implement queries | Christoph Bumiller | 2011-01-09 | 10 | -23/+432 |
| | | |||||
* | | i965g: fix warnings | Dave Airlie | 2011-01-09 | 2 | -2/+1 |
| | | |||||
* | | i965g: update intel_decode from upstream. | Dave Airlie | 2011-01-09 | 4 | -137/+432 |
| | | |||||
* | | i965g: update disassembler code from classic. | Dave Airlie | 2011-01-09 | 8 | -34/+95 |
| | | | | | | | | still a bit of work to do, the winsys gen setting is a bit of a hack. | ||||
* | | i965g: update brw_defines.h from classic driver | Dave Airlie | 2011-01-09 | 3 | -37/+397 |
| | | |||||
* | | i965g: update brw_structs.h from classic driver. | Dave Airlie | 2011-01-09 | 3 | -88/+288 |
| | | |||||
* | | i965g: update to similiar gen stuff as i965 | Dave Airlie | 2011-01-09 | 33 | -151/+173 |
| | |