Commits · d1bea693e20b470f0c2410b68bd867c7203f8b19 · Raul Torres / llvm-target-spread · GitLab

Jul 04, 2014
- InstCombine: Strength reduce sadd.with.overflow into a regular nsw add if we... · a420df29
  Benjamin Kramer authored Jul 04, 2014
  a420df29
Jul 02, 2014
- InstCombine: Optimize x/INT_MIN to x==INT_MIN · f28e2a42
  David Majnemer authored Jul 02, 2014
  f28e2a42
- InstCombine: Don't turn -(x/INT_MIN) -> x/INT_MIN · bdeef602
  David Majnemer authored Jul 02, 2014
  bdeef602
Jul 01, 2014
- Optimize InstCombine stack memory consumption · 813dab2f
  Reid Kleckner authored Jul 01, 2014
  813dab2f
Jun 27, 2014
- Added instruction combine to transform few more negative values addition to subtraction (Part 3) · adc07739
  Dinesh Dwivedi authored Jun 27, 2014
  adc07739
Jun 26, 2014
- This patch removed duplicate code for matching patterns · 99281a06
  Dinesh Dwivedi authored Jun 26, 2014
  99281a06
- Added instruction combine to transform few more negative values addition to subtraction (Part 2) · a7161735
  Dinesh Dwivedi authored Jun 26, 2014
  a7161735
Jun 24, 2014
- InstCombine: Disable umul.with.overflow recognition for vectors. · c96a7f88
  Benjamin Kramer authored Jun 24, 2014
  c96a7f88
- InstCombine: Don't try to reorder shuffles where the mask is a ConstantExpr. · 6de78666
  Benjamin Kramer authored Jun 24, 2014
  6de78666
Jun 19, 2014
- Added instruction combine to transform few more negative values addition to subtraction (Part 1) · 562fd753
  Dinesh Dwivedi authored Jun 19, 2014
  562fd753
- Refactored and updated SimplifyUsingDistributiveLaws() to · b62e52e1
  Dinesh Dwivedi authored Jun 19, 2014
  b62e52e1
- InstCombine: Stop two transforms dueling · 6cf6c053
  David Majnemer authored Jun 19, 2014
  6cf6c053
- Move optimization of some cases of (A & C1)|(B & C2) from instcombine to... · 8561a49c
  Nick Lewycky authored Jun 19, 2014
  8561a49c
- Remove redundant code in InstCombineShift, no functionality change because... · 802df524
  Nick Lewycky authored Jun 19, 2014
  802df524
- R600/SI: Add intrinsics for various math instructions. · a0050b09
  Matt Arsenault authored Jun 19, 2014
  a0050b09
Jun 17, 2014
- [InstCombine] mark ADD with nuw if no unsigned overflow · 33bd53df
  Jingyue Wu authored Jun 17, 2014
  33bd53df
Jun 15, 2014
- Canonicalize addrspacecast ConstExpr between different pointer types · baabe509
  Jingyue Wu authored Jun 15, 2014
  baabe509
Jun 12, 2014
- This removes TODO added in http://reviews.llvm.org/D3658 · 95f0d51b
  Dinesh Dwivedi authored Jun 12, 2014
  95f0d51b
Jun 09, 2014
- Look through addrspacecasts when turning ptr comparisons into · 44f60d0a
  Matt Arsenault authored Jun 09, 2014
  44f60d0a
Jun 07, 2014
- Revert 209903 and 210040. · 4ba22f08
  Rafael Espindola authored Jun 07, 2014
  4ba22f08
Jun 06, 2014
- InstCombine: Canonicalize addrspacecast between different element types · 77145d94
  Jingyue Wu authored Jun 06, 2014
  77145d94
- Added select flavour for ABS and NEG(ABS) · 3217b6c6
  Dinesh Dwivedi authored Jun 06, 2014
  3217b6c6
Jun 05, 2014
- [PPC64LE] Correct vperm -> shuffle transform for little endian · a1184635
  Bill Schmidt authored Jun 05, 2014
  a1184635
Jun 04, 2014
- Add a Constant version of stripPointerCasts. · 78598d9a
  Rafael Espindola authored Jun 04, 2014
  78598d9a
- Clauses in a landingpad are always Constant. Use a stricter type. · 4dc5dfc5
  Rafael Espindola authored Jun 04, 2014
  4dc5dfc5
- InstCombine: Improvement to check if signed addition overflows. · 04c22586
  Rafael Espindola authored Jun 04, 2014
  04c22586
Jun 03, 2014
- Add back commit r210029. · d1a2c2d9
  Rafael Espindola authored Jun 02, 2014
  d1a2c2d9
Jun 02, 2014
- Revert "Add the nsw flag when we detect that an add will not signed overflow." · 582c890f
  Rafael Espindola authored Jun 02, 2014
  582c890f
- Added support to optimize comparisons with "lshr exact" of a constant. · 6b04ef78
  Rafael Espindola authored Jun 02, 2014
  6b04ef78
- Add the nsw flag when we detect that an add will not signed overflow. · 82899feb
  Rafael Espindola authored Jun 02, 2014
  82899feb
- Added inst combine tarnsform for (1 << X) & C pattrens where C is (some PowerOf2 - 1) · ce5d35a9
  Dinesh Dwivedi authored Jun 02, 2014
  ce5d35a9
- Added inst combine transforms for single bit tests from Chris's note · 43e127bd
  Dinesh Dwivedi authored Jun 02, 2014
  43e127bd
May 30, 2014
- PR19753: Optimize comparisons with "ashr exact" of a constanst. · c323952c
  Rafael Espindola authored May 30, 2014
  c323952c
- And fix my fix to sink down through the type at the right time. My · fdc0e0b4
  Chandler Carruth authored May 29, 2014
  fdc0e0b4
- Fix one bug in the latest incarnation of r209843 -- combining GEPs · 3012a1b4
  Chandler Carruth authored May 29, 2014
  3012a1b4
May 29, 2014
- Add support for combining GEPs across PHI nodes · c6b506a0
  Louis Gerbarg authored May 29, 2014
  c6b506a0
- Revert "Revert "Revert "InstCombine: Improvement to check if signed addition overflows.""" · a248f536
  Rafael Espindola authored May 29, 2014
  a248f536
May 28, 2014
- Revert "Revert "InstCombine: Improvement to check if signed addition overflows."" · 6196b743
  Rafael Espindola authored May 28, 2014
  6196b743
- Revert "Add support for combining GEPs across PHI nodes" · 910528a3
  Rafael Espindola authored May 28, 2014
  910528a3
- Revert "InstCombine: Improvement to check if signed addition overflows." · fb59b05c
  Rafael Espindola authored May 28, 2014
  fb59b05c