Commits · ff659b5b8635ead41d68dfad27c4f5fbe5f17fd5 · Roger Ferrer / llvm-epi · GitLab

Mar 07, 2009
- Arithmetic instructions don't set EFLAGS bits OF and CF bits · ff659b5b
  Dan Gohman authored Mar 07, 2009
  ff659b5b
Mar 06, 2009
- Add a comment. · 211f3c0f
  Mikhail Glushenkov authored Mar 06, 2009
  211f3c0f
- Trailing whitespace. · 0658cac5
  Mikhail Glushenkov authored Mar 06, 2009
  0658cac5
- Trailing whitespace. · 6a64f907
  Mikhail Glushenkov authored Mar 06, 2009
  6a64f907
- While thinking about the one-definition-rule and trying · ed722831
  Duncan Sands authored Mar 06, 2009
  ed722831
- While converting an aggregate to scalare, ignore and remove aggregate's debug info. · 25b62516
  Devang Patel authored Mar 06, 2009
  25b62516
- While hoisting instruction to speculatively execute simple bb, ignore dbg intrinsics. · 5aed7765
  Devang Patel authored Mar 06, 2009
  5aed7765
- Skip DbgInfoIntrinsic. · e8c6d310
  Devang Patel authored Mar 06, 2009
  e8c6d310
- Fix ScheduleDAGRRList::CopyAndMoveSuccessors' handling of nodes · 15af5524
  Dan Gohman authored Mar 06, 2009
  15af5524
- Revert 66224. · fc507a1f
  Devang Patel authored Mar 06, 2009
  fc507a1f
- Revert rev. 66167. · d926aaa2
  Devang Patel authored Mar 06, 2009
  d926aaa2
- Do not let debug info prevert globalopt from shriking a global vars to boolean. · ab16577a
  Devang Patel authored Mar 06, 2009
  ab16577a
Mar 05, 2009
- Fix the "test" optimization to recognize "dec" as an add of · 2c2f192c
  Dan Gohman authored Mar 05, 2009
  2c2f192c
- Make this test more thorough. Not only should there be no %esi, · 14cc4a45
  Dan Gohman authored Mar 05, 2009
  14cc4a45
- GlobalOpt only process non constant local GVs while optimizing global vars. · 709d6ac4
  Devang Patel authored Mar 05, 2009
  709d6ac4
- Do not split edges to EH landing pads. It will cause code size explosion. · b7922dee
  Evan Cheng authored Mar 05, 2009
  b7922dee
- Ignore the debug info intrinsics when looking for dependency through basic block. · abe41924
  Zhou Sheng authored Mar 05, 2009
  abe41924
Mar 04, 2009
- Should have XFAILed this test. · 19ce8619
  Bill Wendling authored Mar 04, 2009
  19ce8619
- Temporarily revert r65994. It was causing rdar://6646455 . · 803da0db
  Bill Wendling authored Mar 04, 2009
  803da0db
- Re-apply 66008, now that the unfoldMemoryOperand bug is fixed. · 55d7b2ac
  Dan Gohman authored Mar 04, 2009
  55d7b2ac
- Fix PR3720 by properly propagating alignment information from memcpy/memmove · 5c204c92
  Chris Lattner authored Mar 04, 2009
  5c204c92
- Add a restore folder, which shaves a dozen or so machineinstrs off oggenc. ... · 0dedc114
  Owen Anderson authored Mar 04, 2009
  0dedc114
- Fix PR3666: isel calls to constant addresses. · 9edd616b
  Evan Cheng authored Mar 04, 2009
  9edd616b
- PR3686: make the legalizer handle bitcast from i80 to x86 long double. · 7604d377
  Eli Friedman authored Mar 04, 2009
  7604d377
- Revert r66004 for now; it's causing a variety of test failures. · 6728f892
  Dan Gohman authored Mar 04, 2009
  6728f892
- Rename test. · e08f4cb9
  Evan Cheng authored Mar 04, 2009
  e08f4cb9
- Teach the x86 backend to eliminate "test" instructions by using the EFLAGS · fe8d71f4
  Dan Gohman authored Mar 04, 2009
  fe8d71f4
- Fix PR3701. 1. X86 target renamed eflags register to flags. This matches what... · b8905c4e
  Evan Cheng authored Mar 04, 2009
  b8905c4e
- If a global constant is dead then global's debug info should not prevent the... · 81245961
  Devang Patel authored Mar 04, 2009
  81245961
- Make my earlier patch to skip debug intrinsics · 0365d3b8
  Dale Johannesen authored Mar 04, 2009
  0365d3b8
- The DAG combiner was performing a BT combine. The BT combine had a value of -1, · 6d271473
  Bill Wendling authored Mar 04, 2009
  6d271473
Mar 03, 2009
- Instruction counters must skip the bitcasts that · 09c3e8ec
  Dale Johannesen authored Mar 03, 2009
  09c3e8ec
- When removing a store to an alloca that has only one · 77456b7a
  Dale Johannesen authored Mar 03, 2009
  77456b7a
- Remove accidental check-ins in r65960. :-( · 7fcd6148
  Bill Wendling authored Mar 03, 2009
  7fcd6148
- Use > instead of >=. We want to promote aggregates of 128-bytes. · a68fc7af
  Bill Wendling authored Mar 03, 2009
  a68fc7af
- Ignore the debug info intrinsics when adding instructions into alias sets. · 50603510
  Zhou Sheng authored Mar 03, 2009
  50603510
- Testcase for line number sinking in InstCombine. · fcf4ecec
  Dale Johannesen authored Mar 03, 2009
  fcf4ecec
- If branch conditions' one successor is dominating another non-latch successor... · d50ebbdf
  Devang Patel authored Mar 02, 2009
  d50ebbdf
Mar 02, 2009
- Fix PR3694: add an instcombine micro-optimization that helps · 5795a609
  Duncan Sands authored Mar 02, 2009
  5795a609
- Fix a problem with DAGCombine on 64b targets where folding · a9e98122
  Nate Begeman authored Mar 01, 2009
  a9e98122