Commits · 206cf6487c0759dd5e1c943690b830543d4d415a · Raul Torres / llvm-target-spread · GitLab

Oct 30, 2012
- Reformat and 80-column this. It's not strictly conforming · 206cf648
  Eric Christopher authored Oct 30, 2012
  206cf648
- Fix PR14212: For some strange reason I treated vectors differently from · 1296b595
  Chandler Carruth authored Oct 30, 2012
  1296b595
- [inline asm] Get the mayLoad/mayStore directly from the MIOp_ExtraInfo operand. · 909f6a03
  Chad Rosier authored Oct 30, 2012
  909f6a03
- BBVectorize: Cache fixed-order pairs instead of recomputing pointer info. · 08f34ac9
  Hal Finkel authored Oct 30, 2012
  08f34ac9
- [mips] Allow tail-call optimization for vararg functions and functions which · 9c962c02
  Akira Hatanaka authored Oct 30, 2012
  9c962c02
- Add a comment for r167040. · 86f6050c
  Chad Rosier authored Oct 30, 2012
  86f6050c
- LoopIdiom: Fix a serious missed optimization: we only turned top-level loops into memmove. · 48a64782
  Benjamin Kramer authored Oct 30, 2012
  48a64782
- BBVectorize: Fix a small bug introduced in r167042. · 2eaadd1a
  Hal Finkel authored Oct 30, 2012
  2eaadd1a
- Add code for saving formal argument information to MipsFunctionInfo. This · 4866fe14
  Akira Hatanaka authored Oct 30, 2012
  4866fe14
- BBVectorize: Simplify how input swapping is handled. · f3848909
  Hal Finkel authored Oct 30, 2012
  f3848909
- Add definition of function MipsTargetLowering::passArgOnStack which emits nodes · 6233cf56
  Akira Hatanaka authored Oct 30, 2012
  6233cf56
- [inline asm] Implement mayLoad and mayStore for inline assembly. In general, · 9e1274fb
  Chad Rosier authored Oct 30, 2012
  9e1274fb
- Do not do tail-call optimization if target is mips16. · 8e50aba5
  Akira Hatanaka authored Oct 30, 2012
  8e50aba5
- BBVectorize: Don't make calls to SE when the result is unused. · eac28871
  Hal Finkel authored Oct 30, 2012
  eac28871
- 80-col · d3df6651
  Nadav Rotem authored Oct 30, 2012
  d3df6651
- LoopVectorize: Add support for write-only loops when the write destination is a single pointer. · bc21aceb
  Nadav Rotem authored Oct 30, 2012
  bc21aceb
- PowerPC: Expand FSRQT for vector types · 5c043aeb
  Adhemerval Zanella authored Oct 30, 2012
  5c043aeb
- LoopVectorize: Fix a bug in the initialization of reduction variables. AND... · b3e8e688
  Nadav Rotem authored Oct 30, 2012
  b3e8e688
- Set %defaultjit to use MCJIT for PowerPC targets. · 7db44294
  Ulrich Weigand authored Oct 30, 2012
  7db44294
- Fix grammar. · 10e0e2ec
  Bill Wendling authored Oct 30, 2012
  10e0e2ec
- Enable ELF machine type to be specified explicitly in X86 backend · 83a77c32
  Michael Liao authored Oct 30, 2012
  83a77c32
- Change ForceSizeOpt attribute into MinSize attribute · 5799e9f6
  Quentin Colombet authored Oct 30, 2012
  5799e9f6
- Fix isEliminableCastPair to work correctly in the presence of pointers · e2395dc2
  Duncan Sands authored Oct 30, 2012
  e2395dc2
- switch_to_lookup_table.ll: Remove some unnecessary lines, comments, · e0cf14fa
  Hans Wennborg authored Oct 30, 2012
  e0cf14fa
- PowerPC: More support for Altivec compare operations · 56775e0f
  Adhemerval Zanella authored Oct 30, 2012
  56775e0f
- Add a helper for telling whether a type is a pointer or vector of pointer type. · 3ce427c0
  Duncan Sands authored Oct 30, 2012
  3ce427c0
- Enable some additional constant folding for PPCDoubleDouble. · 6a9bb51a
  Ulrich Weigand authored Oct 30, 2012
  6a9bb51a
- Use TargetTransformInfo to control switch-to-lookup table transformation · f3254838
  Hans Wennborg authored Oct 30, 2012
  f3254838
- Remove an invalid assert in TargetTransformImpl · d0b95b09
  Hal Finkel authored Oct 30, 2012
  d0b95b09
- * Add e_flags enum for Hexagon · 67211fb6
  Sid Manning authored Oct 30, 2012
  67211fb6
- ARM: Better disassembly for pc-relative LDR. · 4739f2eb
  Jim Grosbach authored Oct 30, 2012
  4739f2eb
- Change mips16 delay slot jumps to non delay slot forms by default. · a8117537
  Reed Kotler authored Oct 30, 2012
  a8117537
- LoopVectorizer: change debug prints: Print the module identifier when deciding... · 73ddcfe0
  Nadav Rotem authored Oct 30, 2012
  73ddcfe0
- Re-commit r166971. I reverted it to quickly, when buildbots didn't have a chance · a3d8e997
  Jakub Staszak authored Oct 30, 2012
  a3d8e997
- Fix ARM's b.w instruction for thumb 2 and the encoding T4. The branch target · 6fd96248
  Kevin Enderby authored Oct 29, 2012
  6fd96248
- Revert r166971. It causes buildbot failure. To be investigated. · d74cb61d
  Jakub Staszak authored Oct 29, 2012
  d74cb61d
Oct 29, 2012
- llvm/test/CodeGen/X86/vec_shuffle-30.ll: Try to unbreak builds - assuming +avx. · 382df5eb
  NAKAMURA Takumi authored Oct 29, 2012
  382df5eb
- Remove unused variable. · c3a92131
  Jakub Staszak authored Oct 29, 2012
  c3a92131
- Simplify code. No functionality change. · 9c361bdf
  Jakub Staszak authored Oct 29, 2012
  9c361bdf
- Allow to fold vector load if there is more than one bitcast, so in the case: · c8f4825b
  Jakub Staszak authored Oct 29, 2012
  c8f4825b