| Commit message (Expand) | Author | Age | Files | Lines |
* | InstCombine: Disable umul.with.overflow recognition for vectors. | Benjamin Kramer | 2014-06-24 | 1 | -1/+5 |
* | InstCombine: Don't try to reorder shuffles where the mask is a ConstantExpr. | Benjamin Kramer | 2014-06-24 | 1 | -1/+3 |
* | Added instruction combine to transform few more negative values addition to s... | Dinesh Dwivedi | 2014-06-19 | 1 | -0/+45 |
* | Refactored and updated SimplifyUsingDistributiveLaws() to | Dinesh Dwivedi | 2014-06-19 | 2 | -105/+142 |
* | InstCombine: Stop two transforms dueling | David Majnemer | 2014-06-19 | 1 | -2/+5 |
* | Move optimization of some cases of (A & C1)|(B & C2) from instcombine to inst... | Nick Lewycky | 2014-06-19 | 1 | -23/+0 |
* | Remove redundant code in InstCombineShift, no functionality change because in... | Nick Lewycky | 2014-06-19 | 1 | -5/+0 |
* | R600/SI: Add intrinsics for various math instructions. | Matt Arsenault | 2014-06-19 | 1 | -0/+14 |
* | [InstCombine] mark ADD with nuw if no unsigned overflow | Jingyue Wu | 2014-06-17 | 2 | -0/+23 |
* | Canonicalize addrspacecast ConstExpr between different pointer types | Jingyue Wu | 2014-06-15 | 1 | -2/+4 |
* | This removes TODO added in http://reviews.llvm.org/D3658 | Dinesh Dwivedi | 2014-06-12 | 1 | -2/+9 |
* | Look through addrspacecasts when turning ptr comparisons into | Matt Arsenault | 2014-06-09 | 1 | -5/+21 |
* | Revert 209903 and 210040. | Rafael Espindola | 2014-06-07 | 1 | -40/+0 |
* | InstCombine: Canonicalize addrspacecast between different element types | Jingyue Wu | 2014-06-06 | 1 | -1/+23 |
* | Added select flavour for ABS and NEG(ABS) | Dinesh Dwivedi | 2014-06-06 | 2 | -20/+51 |
* | [PPC64LE] Correct vperm -> shuffle transform for little endian | Bill Schmidt | 2014-06-05 | 1 | -1/+10 |
* | Add a Constant version of stripPointerCasts. | Rafael Espindola | 2014-06-04 | 1 | -3/+3 |
* | Clauses in a landingpad are always Constant. Use a stricter type. | Rafael Espindola | 2014-06-04 | 1 | -4/+4 |
* | InstCombine: Improvement to check if signed addition overflows. | Rafael Espindola | 2014-06-04 | 1 | -7/+46 |
* | Add back commit r210029. | Rafael Espindola | 2014-06-02 | 1 | -4/+19 |
* | Revert "Add the nsw flag when we detect that an add will not signed overflow." | Rafael Espindola | 2014-06-02 | 1 | -5/+0 |
* | Added support to optimize comparisons with "lshr exact" of a constant. | Rafael Espindola | 2014-06-02 | 1 | -6/+29 |
* | Add the nsw flag when we detect that an add will not signed overflow. | Rafael Espindola | 2014-06-02 | 1 | -0/+5 |
* | Added inst combine tarnsform for (1 << X) & C pattrens where C is (some Power... | Dinesh Dwivedi | 2014-06-02 | 1 | -8/+24 |
* | Added inst combine transforms for single bit tests from Chris's note | Dinesh Dwivedi | 2014-06-02 | 1 | -1/+28 |
* | PR19753: Optimize comparisons with "ashr exact" of a constanst. | Rafael Espindola | 2014-05-30 | 1 | -0/+17 |
* | And fix my fix to sink down through the type at the right time. My | Chandler Carruth | 2014-05-29 | 1 | -8/+9 |
* | Fix one bug in the latest incarnation of r209843 -- combining GEPs | Chandler Carruth | 2014-05-29 | 1 | -8/+13 |
* | Add support for combining GEPs across PHI nodes | Louis Gerbarg | 2014-05-29 | 1 | -0/+79 |
* | Revert "Revert "Revert "InstCombine: Improvement to check if signed addition ... | Rafael Espindola | 2014-05-29 | 1 | -44/+6 |
* | Revert "Revert "InstCombine: Improvement to check if signed addition overflow... | Rafael Espindola | 2014-05-28 | 1 | -6/+44 |
* | Revert "Add support for combining GEPs across PHI nodes" | Rafael Espindola | 2014-05-28 | 1 | -79/+0 |
* | Revert "InstCombine: Improvement to check if signed addition overflows." | Rafael Espindola | 2014-05-28 | 1 | -44/+6 |
* | Add support for combining GEPs across PHI nodes | Louis Gerbarg | 2014-05-28 | 1 | -0/+79 |
* | InstCombine: Improvement to check if signed addition overflows. | Rafael Espindola | 2014-05-28 | 1 | -6/+44 |
* | Post-commit fixes for r209643 | Filipe Cabecinhas | 2014-05-27 | 1 | -3/+7 |
* | Fix bad assert. | Daniel Jasper | 2014-05-27 | 1 | -1/+2 |
* | Convert some X86 blendv* intrinsics into IR. | Filipe Cabecinhas | 2014-05-27 | 1 | -0/+35 |
* | AArch64/ARM64: move ARM64 into AArch64's place | Tim Northover | 2014-05-24 | 1 | -3/+3 |
* | Added inst-combine for 'MIN(MIN(A, 97), 23)' and 'MAX(MAX(A, 23), 97)' | Dinesh Dwivedi | 2014-05-19 | 1 | -5/+13 |
* | Revert r209049 and r209065, "Add support for combining GEPs across PHI nodes" | NAKAMURA Takumi | 2014-05-17 | 1 | -66/+0 |
* | Fix for sanitizer crash introduced in r209049 | Louis Gerbarg | 2014-05-17 | 1 | -1/+8 |
* | Add support for combining GEPs across PHI nodes | Louis Gerbarg | 2014-05-16 | 1 | -0/+59 |
* | Reverting r208848, reason: build failure: sanitizer-x86_64-linux-bootstrap/bu... | Dinesh Dwivedi | 2014-05-15 | 1 | -37/+5 |
* | Added instcombine for 'MIN(MIN(A, 27), 93)' and 'MAX(MAX(A, 93), 27)' | Dinesh Dwivedi | 2014-05-15 | 1 | -2/+18 |
* | Added inst combine transforms for single bit tests from Chris's note | Dinesh Dwivedi | 2014-05-15 | 1 | -5/+37 |
* | InstCombine: Optimize -x s< cst | David Majnemer | 2014-05-15 | 1 | -0/+10 |
* | Rename ComputeMaskedBits to computeKnownBits. "Masked" has been | Jay Foad | 2014-05-14 | 6 | -30/+30 |
* | Fix the case when reordering shuffle and binop produces a constant. | Serge Pavlov | 2014-05-14 | 1 | -13/+13 |
* | Optimize integral reciprocal (udiv 1, x and sdiv 1, x) to not use division. T... | Nick Lewycky | 2014-05-14 | 1 | -1/+20 |