Commits · c586166fe3499fbe0965aa1d67e81289f38ea175 · Roger Ferrer / llvm-epi · GitLab

May 13, 2011
- Since I can't reproduce the failures from 131261, re-trying with a · aa02c084
  Stuart Hastings authored May 13, 2011
  aa02c084
- Revert 131266 and 131261 due to buildbot complaints. · 8d57d8ea
  Stuart Hastings authored May 13, 2011
  8d57d8ea
- Non-fast-isel followup to 129634; correctly handle branches controlled · 89f1b47e
  Stuart Hastings authored May 12, 2011
  89f1b47e
May 11, 2011
- · 8a7beb80
  Nadav Rotem authored May 11, 2011
  8a7beb80
- Give the 'eh.sjlj.dispatchsetup' intrinsic call the value coming from the setjmp · 50117f81
  Bill Wendling authored May 11, 2011
  50117f81
May 10, 2011
- Disable my little CopyToReg argument hack with fast-isel. rdar://problem/9413587 . · 768de0a0
  Eli Friedman authored May 10, 2011
  768de0a0
- Correctly walk through nested and adjacent CALLSEQ_START nodes. No · 999fa3bf
  Stuart Hastings authored May 10, 2011
  999fa3bf
May 09, 2011
- Look through struct wrapped types for inline asm statments. · 44804284
  Eric Christopher authored May 09, 2011
  44804284
- Indent properly, no functionality change. · 6be291a2
  Duncan Sands authored May 09, 2011
  6be291a2
May 06, 2011
- 80 col violations. · d26fc5e0
  Evan Cheng authored May 06, 2011
  d26fc5e0
- Make the logic for determining function alignment more explicit. No functionality change. · 2518f837
  Eli Friedman authored May 06, 2011
  2518f837
- Use array_lengthof. No functional change. · 7a78f661
  Eli Friedman authored May 06, 2011
  7a78f661
May 05, 2011
- Allow FastISel of three-register-operand instructions. · 68b6b0ef
  Owen Anderson authored May 05, 2011
  68b6b0ef
- Avoid extra vreg copies for arguments passed in registers. Specifically, this... · 441a01a2
  Eli Friedman authored May 05, 2011
  441a01a2
- Small syntax cleanup; we don't need to #define constants in C++. No functionality change intended. · fd8c6adf
  Eli Friedman authored May 05, 2011
  fd8c6adf
May 03, 2011
- Other parts of the SelectionDAG framework assume that targets use their... · 66fd0739
  Owen Anderson authored May 02, 2011
  66fd0739
Apr 30, 2011
- Make FastEmit_ri_ try a bit harder to succeed for supported operations;... · 4105ed15
  Eli Friedman authored Apr 29, 2011
  4105ed15
Apr 28, 2011
- Fix a silly mistake in r130338. · 33c13391
  Eli Friedman authored Apr 28, 2011
  33c13391
- Make the fast-isel code for literal 0.0 a bit shorter/faster, since 0.0 is... · 406c471b
  Eli Friedman authored Apr 27, 2011
  406c471b
- Remove unused function. · 121d27e9
  Eli Friedman authored Apr 27, 2011
  121d27e9
Apr 26, 2011
- Be careful about scheduling nodes above previous calls. It increase usages of · 1355bbdd
  Evan Cheng authored Apr 26, 2011
  1355bbdd
- Fast-isel support for simple inline asms. · 7da91aee
  Dan Gohman authored Apr 26, 2011
  7da91aee
- Fix typo · 2f647540
  Evan Cheng authored Apr 26, 2011
  2f647540
Apr 25, 2011
- A dbg.declare may not be in entry block, even if it is referring to an... · 734f2218
  Devang Patel authored Apr 25, 2011
  734f2218
Apr 23, 2011
- Remove unused STL header includes. · 1a180156
  Jay Foad authored Apr 23, 2011
  1a180156
- Teach FastISel to deal with instructions that have two immediate operands. · dd450b86
  Owen Anderson authored Apr 22, 2011
  dd450b86
Apr 22, 2011
- Recommit the fix for rdar://9289512 with a couple tweaks to · 6d277517
  Chris Lattner authored Apr 22, 2011
  6d277517
- DAGCombine: fold "(zext x) == C" into "x == (trunc C)" if the trunc is lossless. · 341c11da
  Benjamin Kramer authored Apr 22, 2011
  341c11da
Apr 21, 2011
- Revert r1296656, "Fix rdar://9289512 - not folding load into compare at -O0...", · 63098282
  Daniel Dunbar authored Apr 21, 2011
  63098282
Apr 20, 2011
- Rewrite the expander for umulo/smulo to remember to sign extend the input · bcaedb5c
  Eric Christopher authored Apr 20, 2011
  bcaedb5c
Apr 19, 2011
- Delete unnecessary variable. <rdar://problem/7662569 > · 468086d5
  Stuart Hastings authored Apr 19, 2011
  468086d5
- SelectBasicBlock is rather slow even when it doesn't do anything; skip the · bcd09b3a
  Eli Friedman authored Apr 19, 2011
  bcd09b3a
- Support nested CALLSEQ_BEGIN/END; necessary for ARM byval support. <rdar://problem/7662569 > · 0b68c121
  Stuart Hastings authored Apr 19, 2011
  0b68c121
- Implement support for x86 fastisel of small fixed-sized memcpys, which are generated · 91328b31
  Chris Lattner authored Apr 19, 2011
  91328b31
Apr 18, 2011
- while we're at it, handle 'sdiv exact' of a power of 2 also, · 48f75ad6
  Chris Lattner authored Apr 18, 2011
  48f75ad6
- fix rdar://9297011 - udiv by power of two causing fast-isel rejects · 562d6e82
  Chris Lattner authored Apr 18, 2011
  562d6e82
Apr 17, 2011
- 1. merge fast-isel-shift-imm.ll into fast-isel-x86-64.ll · b53ccb8e
  Chris Lattner authored Apr 17, 2011
  b53ccb8e
- fix an oversight which caused us to compile the testcase (and other · 4832660b
  Chris Lattner authored Apr 17, 2011
  4832660b
- Fix rdar://9289512 - not folding load into compare at -O0 · 045c4385
  Chris Lattner authored Apr 17, 2011
  045c4385
- split a complex predicate out to a helper function. Simplify two for loops, · d70ff0d8
  Chris Lattner authored Apr 17, 2011
  d70ff0d8