Commits · 1a6bb9fcf6ea8e917b7fdcb9cf6ff76b79ec6985 · Raul Torres / llvm-target-spread · GitLab

Mar 24, 2015
- Revert "Remove an InstCombine that seems to have become redundant." · 1a6bb9fc
  David Blaikie authored Mar 24, 2015
  1a6bb9fc
- Remove an InstCombine that seems to have become redundant. · e37e10dc
  David Blaikie authored Mar 24, 2015
  e37e10dc
- [X86, AVX] instcombine vperm2 intrinsics with zero inputs into shuffles · 43a87fdc
  Sanjay Patel authored Mar 24, 2015
  43a87fdc
- Revert r233062 ""float2int": Add a new pass to demote from float to int where possible." · e42c6455
  Hans Wennborg authored Mar 24, 2015
  e42c6455
- [IRCE] Fix a regression introduced in r232444. · 337d46b3
  Sanjoy Das authored Mar 24, 2015
  337d46b3
- "float2int": Add a new pass to demote from float to int where possible. · 408df516
  James Molloy authored Mar 24, 2015
  408df516
Mar 23, 2015
- DebugInfo: Overload get() in DIDescriptor subclasses · 9b9cc2da
  Duncan P. N. Exon Smith authored Mar 23, 2015
  9b9cc2da
Mar 21, 2015
- [SimplifyLibCalls] Fix negative shifts being produced by the memchr -> bitfield transform. · d6aa0ec7
  Benjamin Kramer authored Mar 21, 2015
  d6aa0ec7
- [SimplifyLibCalls] Turn memchr(const, C, const) into a bitfield check. · 7857d723
  Benjamin Kramer authored Mar 21, 2015
  7857d723
- SimplifyLibCalls: Add basic optimization of memchr calls. · 691363e7
  Benjamin Kramer authored Mar 21, 2015
  691363e7
- MemoryDependenceAnalysis: Don't miscompile atomics · e165502e
  David Majnemer authored Mar 21, 2015
  e165502e
Mar 20, 2015
- [X86, AVX] instcombine common cases of vperm2* intrinsics into shuffles · ccf5f24b
  Sanjay Patel authored Mar 20, 2015
  ccf5f24b
- Correctly estimate SROA savings for store operands in inline cost analysis. · 6c428d6f
  Wei Mi authored Mar 20, 2015
  6c428d6f
Mar 19, 2015
- LowerBitSets: Avoid reusing byte set addresses. · 994ba3d2
  Peter Collingbourne authored Mar 19, 2015
  994ba3d2
- [InstCombine] Don't fold a GEP into itself through a PHI node · 5add63f2
  Daniel Jasper authored Mar 19, 2015
  5add63f2
Mar 17, 2015
- TLI: Add addVectorizableFunctionsFromVecLib. · 6d8a2aa9
  Michael Zolotukhin authored Mar 17, 2015
  6d8a2aa9
- TTI: Honour cost model for estimating cost of vector-intrinsic and calls. · c3d60efb
  Michael Zolotukhin authored Mar 17, 2015
  c3d60efb
- [SwitchLowering] Remove incoming values in the reverse order · 24fcae8f
  Michael Liao authored Mar 17, 2015
  24fcae8f
- [IRCE] Re-commit tests cases. · 1ac93a5d
  Sanjoy Das authored Mar 17, 2015
  1ac93a5d
- [IRCE] Delete two tests. · 95d04121
  Sanjoy Das authored Mar 17, 2015
  95d04121
- [IRCE] Support half-range checks. · e2cde6f1
  Sanjoy Das authored Mar 17, 2015
  e2cde6f1
- GCOV: Make the exit block placement from r223193 optional · 3faa76bf
  Justin Bogner authored Mar 16, 2015
  3faa76bf
- LowerBitSets: do not use private aliases at all on Darwin. · ad0bdcd2
  Peter Collingbourne authored Mar 16, 2015
  ad0bdcd2
Mar 16, 2015
- DebugInfo: Fix testcases that fail -verify-debug-info=true · b786572d
  Duncan P. N. Exon Smith authored Mar 16, 2015
  b786572d
- [objc-arc] Make the ARC optimizer more conservative by forcing it to be... · dd60f9bb
  Michael Gottesman authored Mar 16, 2015
  dd60f9bb
Mar 15, 2015
- Verifier: Check debug info intrinsic arguments · 166121ad
  Duncan P. N. Exon Smith authored Mar 15, 2015
  166121ad
Mar 14, 2015
- Update InstCombine to transform aggregate stores into scalar stores. · b344ac9a
  Mehdi Amini authored Mar 14, 2015
  b344ac9a
- Add a bunch of CHECK missing colons in tests. NFC. · 082c5c70
  Ahmed Bougacha authored Mar 14, 2015
  082c5c70
- LowerBitSets: Do not export symbols for bit set referenced globals on Darwin. · c9f277f7
  Peter Collingbourne authored Mar 14, 2015
  c9f277f7
Mar 13, 2015
- Reapply "[Reassociate] Add initial support for vector instructions." · 1858ba76
  Robert Lougher authored Mar 13, 2015
  1858ba76
- instcombine: alloca: Canonicalize scalar allocation array size · be95b4af
  Duncan P. N. Exon Smith authored Mar 13, 2015
  be95b4af
- AsmWriter: Write alloca array size explicitly (and -instcombine fixup) · 720762e2
  Duncan P. N. Exon Smith authored Mar 13, 2015
  720762e2
- Revert: "[Reassociate] Add initial support for vector instructions." · 5e0ea66d
  Robert Lougher authored Mar 13, 2015
  5e0ea66d
- [Reassociate] Add initial support for vector instructions. · 1bad505c
  Robert Lougher authored Mar 13, 2015
  1bad505c
- [opaque pointer type] Add textual IR support for explicit type parameter to gep operator · f72d05bc
  David Blaikie authored Mar 13, 2015
  f72d05bc
- ConstantFold: Fix big shift constant folding · e2a4b856
  David Majnemer authored Mar 13, 2015
  e2a4b856
Mar 12, 2015
- Reapply 'Run LICM pass after loop unrolling pass.' · 49bc7643
  Kevin Qin authored Mar 12, 2015
  49bc7643
Mar 11, 2015
- InstCombine: Don't fold call bitcast into args if callee is byval · d61a6fd8
  David Majnemer authored Mar 11, 2015
  d61a6fd8
- Inliner should not add callgraph edges for intrinsic calls (PR22857) · c04b6f24
  Sanjay Patel authored Mar 11, 2015
  c04b6f24
- Prefer pipes over temporary files in a feeble attempt to stabilize this test on windows. · 23445cad
  Benjamin Kramer authored Mar 11, 2015
  23445cad