| Commit message (Expand) | Author | Age | Files | Lines |
* | R600/SI: Add pattern for AMDGPUurecip | Michel Danzer | 2013-04-10 | 1 | -2/+10 |
* | R600/SI: remove image sample writemask | Christian Konig | 2013-04-10 | 1 | -13/+12 |
* | R600/SI: Add support for buffer stores v2 | Tom Stellard | 2013-04-05 | 1 | -2/+8 |
* | R600/SI: add SETO/SETUO patterns | Christian Konig | 2013-03-27 | 1 | -0/+14 |
* | R600/SI: add cummuting of rev instructions | Christian Konig | 2013-03-27 | 1 | -15/+12 |
* | R600/SI: add mulhu/mulhs patterns | Christian Konig | 2013-03-27 | 1 | -1/+18 |
* | R600/SI: add srl/sha patterns for SI | Christian Konig | 2013-03-27 | 1 | -2/+6 |
* | R600/SI: avoid unecessary subreg extraction in IMAGE_SAMPLE | Christian Konig | 2013-03-26 | 1 | -13/+7 |
* | R600/SI: replace WQM intrinsic | Christian Konig | 2013-03-26 | 1 | -11/+0 |
* | R600: Use legacy (0 * anything = 0) MUL instructions for pow intrinsics | Michel Danzer | 2013-03-22 | 1 | -2/+1 |
* | R600/SI: implement indirect adressing for SI | Christian Konig | 2013-03-18 | 1 | -0/+69 |
* | R600/SI: add float vector types | Christian Konig | 2013-03-18 | 1 | -10/+71 |
* | R600/SI: add shl pattern | Christian Konig | 2013-03-18 | 1 | -1/+3 |
* | R600/SI: add BUFFER_LOAD_DWORD pattern | Christian Konig | 2013-03-18 | 1 | -3/+9 |
* | R600/SI: implement SI.load.const intrinsic | Christian Konig | 2013-03-18 | 1 | -0/+12 |
* | R600/SI: enable all S_LOAD and S_BUFFER_LOAD opcodes | Christian Konig | 2013-03-18 | 1 | -11/+25 |
* | R600/SI: rework input interpolation v2 | Christian Konig | 2013-03-07 | 1 | -61/+6 |
* | R600/SI: remove SGPR address space v2 | Christian Konig | 2013-03-07 | 1 | -3/+2 |
* | R600/SI: switch types of SGPRs to v*i8 | Christian Konig | 2013-03-07 | 1 | -2/+2 |
* | R600/SI: remove S_MOV immediate patterns | Christian Konig | 2013-03-01 | 1 | -12/+2 |
* | R600/SI: add some more instruction flags | Christian Konig | 2013-02-26 | 1 | -14/+79 |
* | R600/SI: add post ISel folding for SI v2 | Christian Konig | 2013-02-26 | 1 | -11/+11 |
* | R600/SI: fix VOP3b encoding v2 | Christian Konig | 2013-02-26 | 1 | -6/+8 |
* | R600/SI: fix and cleanup SI register definition v2 | Christian Konig | 2013-02-26 | 1 | -2/+3 |
* | R600/SI: Add pattern for sign extension of i1 to i32. | Michel Danzer | 2013-02-22 | 1 | -0/+5 |
* | R600/SI: Add pattern for logical or of i1 values. | Michel Danzer | 2013-02-22 | 1 | -0/+4 |
* | R600/SI: Add pattern for fceil. | Michel Danzer | 2013-02-22 | 1 | -1/+3 |
* | R600/SI: inline V_ADD|SUB_F32 patterns | Christian Konig | 2013-02-21 | 1 | -9/+5 |
* | R600/SI: replace IMPLICIT_DEF with SIOperand.ZERO | Christian Konig | 2013-02-21 | 1 | -1/+1 |
* | R600/SI: replace SI_V_CNDLT with a pattern | Christian Konig | 2013-02-21 | 1 | -7/+5 |
* | R600/SI: use patterns for clamp, fabs, fneg | Christian Konig | 2013-02-21 | 1 | -4/+22 |
* | R600/SI: add all the other missing asm operands v2 | Christian Konig | 2013-02-21 | 1 | -22/+25 |
* | R600/SI: simplify VOPC_* pattern v2 | Christian Konig | 2013-02-21 | 1 | -248/+196 |
* | R600/SI: Make sure M0 is loaded for V_INTERP_MOV_F32 | Michel Danzer | 2013-02-21 | 1 | -1/+2 |
* | R600/SI: Use MULADD_IEEE/V_MAD_F32 instruction for mad pattern | Vincent Lejeune | 2013-02-18 | 1 | -2/+2 |
* | R600/SI: Add pattern to simplify i64 loading | Christian Konig | 2013-02-16 | 1 | -0/+5 |
* | R600/SI: nuke SReg_1 v3 | Christian Konig | 2013-02-16 | 1 | -31/+19 |
* | R600/SI: cleanup literal handling v3 | Christian Konig | 2013-02-16 | 1 | -44/+37 |
* | R600/SI: replace AllReg_* with [SV]Src_* v2 | Christian Konig | 2013-02-16 | 1 | -64/+64 |
* | R600/SI: Fix int_SI_fs_interp_constant | Michel Danzer | 2013-02-14 | 1 | -11/+18 |
* | R600/SI: Add pattern for mul. | Tom Stellard | 2013-02-07 | 1 | -0/+4 |
* | R600/SI: simplify and fix SMRD encoding | Tom Stellard | 2013-02-07 | 1 | -3/+36 |
* | R600/SI: add proper 64bit immediate support v2 | Tom Stellard | 2013-02-07 | 1 | -11/+8 |
* | R600/SI: Use proper instructions for array/shadow samplers. | Tom Stellard | 2013-02-07 | 1 | -3/+39 |
* | R600/SI: Make sample intrinsic address parameter type overloaded. | Tom Stellard | 2013-02-07 | 1 | -18/+34 |
* | R600/SI: Add basic support for more integer vector types. | Tom Stellard | 2013-02-07 | 1 | -6/+10 |
* | R600/SI: Add pattern for flog2 | Michel Danzer | 2013-02-07 | 1 | -1/+3 |
* | R600: Consolidate sub register indices. | Tom Stellard | 2013-02-07 | 1 | -24/+24 |
* | R600/SI: Add patterns for fcos and fsin. | Tom Stellard | 2013-02-05 | 1 | -0/+10 |
* | R600/SI: Use unnormalized coordinates for sampling with the RECT target. | Tom Stellard | 2013-01-21 | 1 | -0/+6 |