Commits · dee12d8110e83075346841322f60c51d6ea04ad2 · Roger Ferrer / llvm-epi · GitLab

Sep 23, 2010
- Avoid some Mach-O specific alignment being done on ELF. · dee12d81
  Rafael Espindola authored Sep 22, 2010
  dee12d81
Sep 22, 2010
- Correctly align bss. · 9bac6cb5
  Rafael Espindola authored Sep 22, 2010
  9bac6cb5
- llvmc: Support '-emit-llvm -S' with -opt. · 922f13ce
  Mikhail Glushenkov authored Sep 22, 2010
  922f13ce
- fix rdar://8456371 - Handle commutable instructions written backward. · 415624cf
  Chris Lattner authored Sep 22, 2010
  415624cf
- Fix an inconsistency in the x86 backend that led it to reject "calll foo" on · bd857253
  Chris Lattner authored Sep 22, 2010
  bd857253
- fix rdar://8456412 - llvm-mc crash in encoder on "mov %rdx, %cr8" · 37fc469f
  Chris Lattner authored Sep 22, 2010
  37fc469f
- fix rdar://8456417 - llvm-mc can't do basic math · 2bb9504d
  Chris Lattner authored Sep 22, 2010
  2bb9504d
- add the missing aliases for fp stack cmovs, rdar://8456391 · 9dfd2e35
  Chris Lattner authored Sep 22, 2010
  9dfd2e35
- reimplement elf TLS support in terms of addressing modes, eliminating SegmentBaseAddress. · 8a236b63
  Chris Lattner authored Sep 22, 2010
  8a236b63
- Fix rdar://8456364 - llvm-mc rejects '%CS' · 1261b81e
  Chris Lattner authored Sep 22, 2010
  1261b81e
- fix rdar://8456389 - llvm-mc mismatch with 'as' on 'fstp' · 39cf8ded
  Chris Lattner authored Sep 22, 2010
  39cf8ded
- fix rdar://8456361 - llvm-mc rejects 'rep movsd' · a91e510f
  Chris Lattner authored Sep 22, 2010
  a91e510f
- linux has a different stack alignment than the mac, relax this a bit. · 505af598
  Chris Lattner authored Sep 22, 2010
  505af598
- give VZEXT_LOAD a memory operand, it now works with segment registers. · 54e53295
  Chris Lattner authored Sep 22, 2010
  54e53295
- revert r114386 now that address modes work correctly, we get a nice · 07827ba9
  Chris Lattner authored Sep 22, 2010
  07827ba9
- give LCMPXCHG_DAG[8] a memory operand, allowing it to work with addrspace 256/257 · e479e964
  Chris Lattner authored Sep 21, 2010
  e479e964
- filecheckize · 0cefa511
  Chris Lattner authored Sep 21, 2010
  0cefa511
- OptimizeCompareInstr should avoid iterating pass the beginning of the MBB when... · d757c88b
  Evan Cheng authored Sep 21, 2010
  d757c88b
- Enable target-specific mul-lowering on ARM, even at -Os. Remove a test that this makes · 61158f98
  Owen Anderson authored Sep 21, 2010
  61158f98
Sep 21, 2010
- Use FileCheck · d92f42d1
  Devang Patel authored Sep 21, 2010
  d92f42d1
- When adding the carry bit to another value on X86, exploit the fact that the carry-materialization · f4b1a5bd
  Owen Anderson authored Sep 21, 2010
  f4b1a5bd
- llvmc: Allow multiple output languages. · ed79d5f2
  Mikhail Glushenkov authored Sep 21, 2010
  ed79d5f2
- fix rdar://8453210 , a crash handling a call through a GS relative load. · bb0a1c44
  Chris Lattner authored Sep 21, 2010
  bb0a1c44
- Implement support for .local and its "interesting" interactions with .comm. · f0591c16
  Rafael Espindola authored Sep 21, 2010
  f0591c16
- Enable machine sinking critical edge splitting. e.g. · f3e9a485
  Evan Cheng authored Sep 20, 2010
  f3e9a485
Sep 20, 2010
- CombinerAA is now reordering these stores. · 5e473424
  Owen Anderson authored Sep 20, 2010
  5e473424
- When TCO is turned on, it is possible to end up with aliasing FrameIndex's. Therefore, · 272ff949
  Owen Anderson authored Sep 20, 2010
  272ff949
- Simplify ARM callee-saved register handling by removing the distinction · 94dfd6fc
  Jim Grosbach authored Sep 20, 2010
  94dfd6fc
- Produce a R_X86_64_32 when the value is >=0. · 32ab4558
  Rafael Espindola authored Sep 20, 2010
  32ab4558
Sep 19, 2010
- test/CodeGen/X86: Add explicit triplet -mtriple=i686-linux to 3 tests... · b912c27f
  NAKAMURA Takumi authored Sep 19, 2010
  b912c27f
Sep 18, 2010
- Add the exit instruction to the PTX target. · dbb199d8
  Eric Christopher authored Sep 18, 2010
  dbb199d8
- Make sure the STT_FILE symbol is the first one in the symbol table. · 57e599a8
  Rafael Espindola authored Sep 18, 2010
  57e599a8
- Invert the logic of reachesChainWithoutSideEffects(). What we want to check is that there is · b92b13d8
  Owen Anderson authored Sep 18, 2010
  b92b13d8
- Add test that was missing in my previous commit. · 15829c6c
  Rafael Espindola authored Sep 18, 2010
  15829c6c
- Attempt to XFAIL this test on arm-linux, which is inexplicably failing. · 49c15c0f
  Dan Gohman authored Sep 18, 2010
  49c15c0f
- Fix vmov.f64 disassembly on targets where sizeof(long) != 8. · de636ca9
  Benjamin Kramer authored Sep 17, 2010
  de636ca9
- Add target-specific DAG combiner for BUILD_VECTOR and VMOVRRD. An i64 · cb6db988
  Bob Wilson authored Sep 17, 2010
  cb6db988
- Teach the (non-MC) instruction printer to use the cannonical names for push/pop, · 7a6c37d3
  Jim Grosbach authored Sep 17, 2010
  7a6c37d3
- Teach machine sink to · e53ab6df
  Evan Cheng authored Sep 17, 2010
  e53ab6df
Sep 17, 2010
- Update tests to handle MC-inst instruction printing of shift operations. The · 6d800f88
  Jim Grosbach authored Sep 17, 2010
  6d800f88