Commits · d97b83e320819eb26723b02544a081e2abce9a3f · Roger Ferrer / llvm-epi · GitLab

Mar 22, 2012
- Remove -enable-lsr-nested in time for 3.1. · d97b83e3
  Andrew Trick authored Mar 22, 2012
  d97b83e3
- Refactor the code for visiting instructions out into helper functions. · 817a7c6f
  Dan Gohman authored Mar 22, 2012
  817a7c6f
- Remove unused simplifyIVUsers · 06549890
  Andrew Trick authored Mar 22, 2012
  06549890
- Remove -enable-iv-rewrite, which has been unsupported since 3.0. · f47d0af5
  Andrew Trick authored Mar 22, 2012
  f47d0af5
- don't use "signed", just something I noticed in patches flying by. · 7d7dba3c
  Chris Lattner authored Mar 22, 2012
  7d7dba3c
Mar 21, 2012
- [asan] fix one more bug related to long double · 84a7f2e8
  Kostya Serebryany authored Mar 21, 2012
  84a7f2e8
- Zap some dead code pointed out by Chandler. · 7d522f16
  Eric Christopher authored Mar 20, 2012
  7d522f16
Mar 20, 2012
- LoopSimplify bug fix. Handle indirect loop back edges. · f7711010
  Andrew Trick authored Mar 20, 2012
  f7711010
- whitespace · bb01cbb3
  Andrew Trick authored Mar 20, 2012
  bb01cbb3
Mar 19, 2012
- [asan] don't emit __asan_mapping_offset/__asan_mapping_scale by default --... · c58dc9fc
  Kostya Serebryany authored Mar 19, 2012
  c58dc9fc
Mar 16, 2012
- Revert r152907. · 55b6b2b6
  Bill Wendling authored Mar 16, 2012
  55b6b2b6
- The alignment of the pointer part of the store instruction may have an · a2a26b54
  Bill Wendling authored Mar 16, 2012
  a2a26b54
- Rip out support for 'llvm.noinline'. This thing has a strange history... · b37fc13a
  Chandler Carruth authored Mar 16, 2012
  b37fc13a
- Start removing the use of an ad-hoc 'never inline' set and instead · d7a5f2ad
  Chandler Carruth authored Mar 16, 2012
  d7a5f2ad
- LSR fix: Add isSimplifiedLoopNest to IVUsers analysis. · 070e540a
  Andrew Trick authored Mar 16, 2012
  070e540a
- In InstCombiner::visitOr, make sure we reverse the operand swap used for... · e06535b2
  Eli Friedman authored Mar 16, 2012
  e06535b2
Mar 15, 2012
- Short term fix for pr12270 before we change dominates to handle unreachable · f5892785
  Rafael Espindola authored Mar 15, 2012
  f5892785
- Use an iterator instead of calling .size() on the worklist every time, which is wasteful. · 7fa1be77
  Bill Wendling authored Mar 15, 2012
  7fa1be77
- Remove the basic inliner. This was added in 2007, and hasn't really · be2ccf01
  Chandler Carruth authored Mar 15, 2012
  be2ccf01
- This pass didn't want the inline cost per-se, it just wants generic code · 3904590b
  Chandler Carruth authored Mar 15, 2012
  3904590b
- Fixed a transform crash when setting a negative size value for memset. Fixes PR12202. · a733297f
  Aaron Ballman authored Mar 15, 2012
  a733297f
- [tsan] use FunctionBlackList · abad002d
  Kostya Serebryany authored Mar 14, 2012
  abad002d
- [asan] rename class BlackList to FunctionBlackList and move it into a separate... · 01401cec
  Kostya Serebryany authored Mar 14, 2012
  01401cec
- When an invoke is marked with metadata indicating its unwind edge · 532fb813
  Dan Gohman authored Mar 14, 2012
  532fb813
Mar 14, 2012
- Change where we enable the heuristic that delays inlining into functions · 30b8416d
  Chandler Carruth authored Mar 14, 2012
  30b8416d
Mar 13, 2012
- Target override to allow CodeGenPrepare to sink address operands to intrinsics... · 615fd897
  Pete Cooper authored Mar 13, 2012
  615fd897
- enhance jump threading to preserve TBAA information when PRE'ing loads, · 87fa77bd
  Chris Lattner authored Mar 13, 2012
  87fa77bd
- Teach globalopt how to evaluate an invoke with a non-void return type. · eab06fa3
  Dan Gohman authored Mar 13, 2012
  eab06fa3
Mar 12, 2012
- When inlining a function and adding its inner call sites to the · 595fda84
  Chandler Carruth authored Mar 12, 2012
  595fda84
Mar 11, 2012
- llvm::SwitchInst · 97b02fc1
  Stepan Dyatkovskiy authored Mar 11, 2012
  97b02fc1
Mar 09, 2012
- Add statistics on removed switch cases, and fix the phi statistic · 14eb1758
  Duncan Sands authored Mar 09, 2012
  14eb1758
- When identifying exit nodes for the reverse-CFG reverse-post-order · 500b598c
  Dan Gohman authored Mar 09, 2012
  500b598c
- Eliminate switch cases that can never match, for example removes all · cca89124
  Duncan Sands authored Mar 09, 2012
  cca89124
Mar 08, 2012
- Taken into account Duncan's comments for r149481 dated by 2nd Feb 2012: · 5b648afb
  Stepan Dyatkovskiy authored Mar 08, 2012
  5b648afb
Mar 05, 2012
- fix typos · 5ce71b18
  Sebastian Pop authored Mar 05, 2012
  5ce71b18
- remove spaces on empty lines · 8844e224
  Sebastian Pop authored Mar 05, 2012
  8844e224
- This is not a common case, in fact it never happens! · 3eb32857
  Duncan Sands authored Mar 05, 2012
  3eb32857
- Switch mem2reg to use the new hashing infrastructure. · d95357a1
  Chandler Carruth authored Mar 05, 2012
  d95357a1
- Replace the ad-hoc hashing in GVN with the new hashing infrastructure. · e134d1a3
  Chandler Carruth authored Mar 05, 2012
  e134d1a3
Mar 04, 2012
- Nick pointed out on IRC that GVN's propagateEquality wasn't propagating · 4d928e7d
  Duncan Sands authored Mar 04, 2012
  4d928e7d