| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | AMDGPU: Constant fold rcp node | Matt Arsenault | 2017-03-08 | 1 | -2/+12 |
| * | AMDGPU: Support v2i16/v2f16 packed operations | Matt Arsenault | 2017-02-27 | 1 | -3/+8 |
| * | AMDGPU : Replace FMAD with FMA when denormals are enabled. | Wei Ding | 2017-02-24 | 1 | -1/+5 |
| * | AMDGPU: Add cvt.pkrtz intrinsic | Matt Arsenault | 2017-02-22 | 1 | -0/+1 |
| * | AMDGPU: Remove llvm.AMDGPU.clamp intrinsic | Matt Arsenault | 2017-02-21 | 1 | -11/+0 |
| * | AMDGPU: Redefine clamp node as clamp 0.0-1.0 | Matt Arsenault | 2017-02-21 | 1 | -14/+45 |
| * | AMDGPU: Add pass to expand memcpy/memmove/memset | Matt Arsenault | 2017-02-09 | 1 | -4/+5 |
| * | AMDGPU: Fold fneg into fmin/fmax_legacy | Matt Arsenault | 2017-02-03 | 1 | -2/+24 |
| * | AMDGPU: Fold fneg into fminnum/fmaxnum | Matt Arsenault | 2017-02-03 | 1 | -0/+30 |
| * | AMDGPU: Check if users of fneg can fold mods | Matt Arsenault | 2017-02-02 | 1 | -4/+64 |
| * | Revert "In visitSTORE, always use FindBetterChain, rather than only when UseA... | Nirav Dave | 2017-02-02 | 1 | -0/+10 |
| * | In visitSTORE, always use FindBetterChain, rather than only when UseAA is ena... | Nirav Dave | 2017-02-02 | 1 | -10/+0 |
| * | AMDGPU: Use source modifiers with f16->f32 conversions | Matt Arsenault | 2017-02-02 | 1 | -0/+42 |
| * | AMDGPU: Cleanup fmin/fmax legacy function | Matt Arsenault | 2017-02-01 | 1 | -11/+8 |
| * | Re-commit AMDGPU/GlobalISel: Add support for simple shaders | Tom Stellard | 2017-01-30 | 1 | -0/+6 |
| * | Revert "AMDGPU/GlobalISel: Add support for simple shaders" | Tom Stellard | 2017-01-30 | 1 | -6/+0 |
| * | AMDGPU/GlobalISel: Add support for simple shaders | Tom Stellard | 2017-01-30 | 1 | -0/+6 |
| * | Cleanup dump() functions. | Matthias Braun | 2017-01-28 | 1 | -1/+1 |
| * | Revert "In visitSTORE, always use FindBetterChain, rather than only when UseA... | Nirav Dave | 2017-01-26 | 1 | -0/+10 |
| * | In visitSTORE, always use FindBetterChain, rather than only when UseAA is ena... | Nirav Dave | 2017-01-26 | 1 | -10/+0 |
| * | AMDGPU: Fold fneg into round instructions | Matt Arsenault | 2017-01-26 | 1 | -1/+7 |
| * | AMDGPU: Propagate fast math flags in fneg combines | Matt Arsenault | 2017-01-23 | 1 | -3/+3 |
| * | AMDGPU/R600: Serialize vector trunc stores to private AS | Jan Vesely | 2017-01-20 | 1 | -0/+1 |
| * | AMDGPU: Disable some fneg combines unless nsz | Matt Arsenault | 2017-01-19 | 1 | -0/+6 |
| * | AMDGPU: Skip fneg/select combine if it can fold into other | Matt Arsenault | 2017-01-12 | 1 | -29/+40 |
| * | AMDGPU: Fold free fneg into sin | Matt Arsenault | 2017-01-12 | 1 | -1/+5 |
| * | AMDGPU: Fold fneg into fmul_legacy | Matt Arsenault | 2017-01-12 | 1 | -2/+5 |
| * | AMDGPU: Fold fneg into rcp | Matt Arsenault | 2017-01-12 | 1 | -1/+7 |
| * | AMDGPU: Fold fneg into fp_round | Matt Arsenault | 2017-01-12 | 1 | -2/+18 |
| * | AMDGPU: Fold fneg into fp_extend | Matt Arsenault | 2017-01-12 | 1 | -0/+14 |
| * | AMDGPU: Fold fneg into fma or fmad | Matt Arsenault | 2017-01-12 | 1 | -0/+24 |
| * | AMDGPU: Fold fneg into fmul | Matt Arsenault | 2017-01-12 | 1 | -0/+17 |
| * | AMDGPU: Fold fneg into fadd | Matt Arsenault | 2017-01-12 | 1 | -0/+60 |
| * | AMDGPU: Pull fneg/fabs out of a select | Matt Arsenault | 2017-01-11 | 1 | -0/+74 |
| * | AMDGPU: Add tests for HasMultipleConditionRegisters | Matt Arsenault | 2017-01-10 | 1 | -0/+7 |
| * | AMDGPU/R600: Don't use REGISTER_{LOAD,STORE} ISD nodes | Jan Vesely | 2017-01-06 | 1 | -10/+0 |
| * | AMDGPU/SI: Implement sendmsghalt intrinsic | Jan Vesely | 2017-01-04 | 1 | -0/+1 |
| * | AMDGPU: Invert cmp + select with constant | Matt Arsenault | 2016-12-22 | 1 | -0/+19 |
| * | AMDGPU: Update isFPImmLegal for f16 | Matt Arsenault | 2016-12-22 | 1 | -1/+2 |
| * | AMDGPU/SI: Add a MachineMemOperand when lowering llvm.amdgcn.buffer.load.* | Tom Stellard | 2016-12-20 | 1 | -0/+2 |
| * | AMDGPU: Fix asserting on returned tail calls | Matt Arsenault | 2016-12-15 | 1 | -2/+4 |
| * | Revert "In visitSTORE, always use FindBetterChain, rather than only when UseA... | Nirav Dave | 2016-12-14 | 1 | -0/+10 |
| * | In visitSTORE, always use FindBetterChain, rather than only when UseAA is ena... | Nirav Dave | 2016-12-14 | 1 | -10/+0 |
| * | Replace APFloatBase static fltSemantics data members with getter functions | Stephan Bergmann | 2016-12-14 | 1 | -2/+2 |
| * | AMDGPU: Fix i128 mul | Matt Arsenault | 2016-12-09 | 1 | -0/+4 |
| * | Revert "In visitSTORE, always use FindBetterChain, rather than only when UseA... | Nirav Dave | 2016-12-09 | 1 | -0/+10 |
| * | In visitSTORE, always use FindBetterChain, rather than only when UseAA is ena... | Nirav Dave | 2016-12-09 | 1 | -10/+0 |
| * | AMDGPU : Add S_SETREG instructions to fix fdiv precision issues. | Tom Stellard | 2016-12-07 | 1 | -0/+3 |
| * | AMDGPU: Refactor exp instructions | Matt Arsenault | 2016-12-05 | 1 | -0/+2 |
| * | [AMDGPU] Allow hoisting of comparisons out of a loop and eliminate condition ... | Stanislav Mekhanoshin | 2016-11-28 | 1 | -0/+1 |
