Commits · 37d080b11d61372cdc7ec1541ab037a0ea392ad6 · Roger Ferrer / llvm-epi · GitLab

Nov 20, 2006
- We should perform extra load for dllimported calls · 37d080b1
  Anton Korobeynikov authored Nov 20, 2006
  37d080b1
- Fixed typo · 84252caf
  Tanya Lattner authored Nov 20, 2006
  84252caf
- Adding 1.9 release and removing references to cfrontend · da2d1e23
  Tanya Lattner authored Nov 20, 2006
  da2d1e23
- Changing cfrontend dir to llvm-gcc3.4 as the example. · b0babee6
  Tanya Lattner authored Nov 20, 2006
  b0babee6
- For PR950: · d9436b68
  Reid Spencer authored Nov 20, 2006
  d9436b68
- For PR950: · 9f4448a2
  Reid Spencer authored Nov 19, 2006
  9f4448a2
Nov 19, 2006
- Corrections from review. · f69acf3e
  Reid Spencer authored Nov 19, 2006
  f69acf3e
Nov 18, 2006
- Don't transform in another bad case: if the block is empty, it should be · ea017f66
  Chris Lattner authored Nov 18, 2006
  ea017f66
- Fix another case we *don't* want to do this xform. · 56ec81ff
  Chris Lattner authored Nov 18, 2006
  56ec81ff
- Update a reference to the old setcc instruction. · e008326b
  Reid Spencer authored Nov 18, 2006
  e008326b
- For PR950: · c828a0e3
  Reid Spencer authored Nov 18, 2006
  c828a0e3
- MachineInstr includes this · 995a0175
  Chris Lattner authored Nov 18, 2006
  995a0175
- Implement operator<< for machine basic blocks to make it easier to dump them. · ace542fa
  Chris Lattner authored Nov 18, 2006
  ace542fa
- make the previous change more aggressive, moving any block with no fallthrough. · 7acdc17f
  Chris Lattner authored Nov 18, 2006
  7acdc17f
- Minor code layout tweak: if we have something like this: · 47ce2615
  Chris Lattner authored Nov 18, 2006
  47ce2615
- Do not convert massive blocks on phi nodes into select statements. Instead · 95adf8f1
  Chris Lattner authored Nov 18, 2006
  95adf8f1
- converting massive blocks of phis into selects like this is silly. · bfbc7ace
  Chris Lattner authored Nov 18, 2006
  bfbc7ace
- Make the awk filter print out which files its eliminating. · 4924c53b
  Reid Spencer authored Nov 18, 2006
  4924c53b
- This is a utility for preparing patches against LLVM. It runs cvs diff with · 2ca613bc
  Reid Spencer authored Nov 18, 2006
  2ca613bc
- Provide a list of gccld optimization switches. · 170ff027
  Reid Spencer authored Nov 18, 2006
  170ff027
- Fix bugs in this. What was I thinking??? :) · e7c6fcdb
  Reid Spencer authored Nov 18, 2006
  e7c6fcdb
- Movind newly-intorudced dll* linkage stuff & windows CC to "supported" · e7cbbfcb
  Anton Korobeynikov authored Nov 18, 2006
  e7cbbfcb
- fix bug · ef22b579
  Chris Lattner authored Nov 18, 2006
  ef22b579
- first draft of 1.9 release notes · b81f10ef
  Chris Lattner authored Nov 18, 2006
  b81f10ef
- Fail even if opt doesn't print anything. · d8bd8def
  Reid Spencer authored Nov 18, 2006
  d8bd8def
- Add a simple test to make sure getModRefInfo is 1/2 way sane. · 178afb1f
  Reid Spencer authored Nov 18, 2006
  178afb1f
- Have ConstantExprs upgrade opcodes the same way as instructions. · ceeed00b
  Reid Spencer authored Nov 18, 2006
  ceeed00b
- on ppc64, float arguments take 8-byte stack slots not 4-byte stack slots. · 2cca385f
  Chris Lattner authored Nov 18, 2006
  2cca385f
- make sure to safe LR8 in the right stack slot for PPC64 · 572e238c
  Chris Lattner authored Nov 18, 2006
  572e238c
- Pretty print 'rldicr r2, r2, 2, 61' as 'sldi r2, r2, 2'. · 9ca15c89
  Chris Lattner authored Nov 18, 2006
  9ca15c89
- Rewrite the branch selector to be correct in the face of large functions. · 542dfd55
  Chris Lattner authored Nov 18, 2006
  542dfd55
- add encoding for BCC, after finally wrestling strange ppc/tblgen endianness · 33fc1d45
  Chris Lattner authored Nov 17, 2006
  33fc1d45
Nov 17, 2006
- convert PPC::BCC to use the 'pred' operand instead of separate predicate · be9377a1
  Chris Lattner authored Nov 17, 2006
  be9377a1
- rename PPC::COND_BRANCH to PPC::BCC · e0263794
  Chris Lattner authored Nov 17, 2006
  e0263794
- start using PPC predicates more consistently. · 8c6a41ea
  Chris Lattner authored Nov 17, 2006
  8c6a41ea
- For unsigned 8-bit division. Use movzbw to set the lower 8 bits of AX while · 9e8093ae
  Evan Cheng authored Nov 17, 2006
  9e8093ae
- Hopefully a good crack at making debugging work on intel -disable-fp-elim. · de5fa025
  Jim Laskey authored Nov 17, 2006
  de5fa025
- Add a x86-64 test case: movb %sil, %ah is illegal. · bae95013
  Evan Cheng authored Nov 17, 2006
  bae95013
- Assert unhandled case. · 73106b5e
  Jim Laskey authored Nov 17, 2006
  73106b5e
- Case sensitive not. · 716f64c3
  Jim Laskey authored Nov 17, 2006
  716f64c3