Commits · e34014d11c8b022735adedfe783ee1445faefd35 · Roger Ferrer / llvm-epi · GitLab

Jan 19, 2015
- IR: Remove isa<MDNodeFwdDecl>, NFC · e34014d1
  Duncan P. N. Exon Smith authored Jan 19, 2015
  e34014d1
- IR: Unify code for MDNode::isResolved(), NFC · 66ed5223
  Duncan P. N. Exon Smith authored Jan 19, 2015
  66ed5223
- IR: Store RAUW support and Context in the same pointer, NFC · 2711ca7c
  Duncan P. N. Exon Smith authored Jan 19, 2015
  2711ca7c
- [Hexagon] Converting intrinsics combine imm/imm, simple shifts and extends. · fcd4569a
  Colin LeMahieu authored Jan 19, 2015
  fcd4569a
- IR: Add isUniqued() and isTemporary() · de03a8b3
  Duncan P. N. Exon Smith authored Jan 19, 2015
  de03a8b3
- IR: Use an enum to describe Metadata storage, NFC · f1340453
  Duncan P. N. Exon Smith authored Jan 19, 2015
  f1340453
- [Hexagon] Converting remaining ALU32/ALU intrinsics. · 9327bdad
  Colin LeMahieu authored Jan 19, 2015
  9327bdad
- [Hexagon] Converting ALU32/ALU intrinsics to new patterns. · 663419b0
  Colin LeMahieu authored Jan 19, 2015
  663419b0
- Remove support for DIVariable's FlagIndirectVariable and expect · 5883af3f
  Adrian Prantl authored Jan 19, 2015
  5883af3f
- [AArch64] Implement GHC calling convention · fa78d086
  Greg Fitzgerald authored Jan 19, 2015
  fa78d086
- [Hexagon] Converting halfword to double accumulating multiply intrinsics. · 310bad8b
  Colin LeMahieu authored Jan 19, 2015
  310bad8b
- Produce errors when an assignment expression would use a common symbol. · c569ac46
  Rafael Espindola authored Jan 19, 2015
  c569ac46
- [ARM] SSAT/USAT with an 'asr #32' shift should result in an undefined encoding... · 3131e85e
  Bradley Smith authored Jan 19, 2015
  3131e85e
- [ARM] Fixup sign extend instruction availability w.r.t. DSP extension · 30057b24
  Bradley Smith authored Jan 19, 2015
  30057b24
- Bring r226038 back. · 12ca34f5
  Rafael Espindola authored Jan 19, 2015
  12ca34f5
- [PM] Replace the Pass argument to SplitEdge with specific analyses used · d450056c
  Chandler Carruth authored Jan 19, 2015
  d450056c
- [PM] Cleanup a dead option to critical edge splitting that I noticed · f8753fc4
  Chandler Carruth authored Jan 19, 2015
  f8753fc4
- [PM] Remove the Pass argument from all of the critical edge splitting · 37df2cfb
  Chandler Carruth authored Jan 19, 2015
  37df2cfb
- [PM] Relax asserts and always try to reconstruct loop simplify form when · ad34d913
  Chandler Carruth authored Jan 19, 2015
  ad34d913
- SLPVectorizer: limit the number of alias checks to reduce the runtime. · 76cb53a8
  Erik Eckstein authored Jan 19, 2015
  76cb53a8
- [PowerPC] Minor correction to r226432 · c3168129
  Hal Finkel authored Jan 19, 2015
  c3168129
- [MIScheduler] Slightly better handling of constrainLocalCopy when both source and dest are local · 54c61ede
  Michael Kuperstein authored Jan 19, 2015
  54c61ede
- [PowerPC] Add r2 as an operand for all calls under both PPC64 ELF V1 and V2 · af51993e
  Hal Finkel authored Jan 19, 2015
  af51993e
- [x86] Change AVX512 intrinsics to take a 8-bit immediate for the comparision... · f4bf9119
  Craig Topper authored Jan 19, 2015
  f4bf9119
- [PM] Lift the analyses into the interface for · 0eae1120
  Chandler Carruth authored Jan 19, 2015
  0eae1120
Jan 18, 2015
- unique_ptrify the RelInfo parameter to TargetRegistry::createMCSymbolizer · 186db431
  David Blaikie authored Jan 18, 2015
  186db431
- std::unique_ptrify the MCStreamer argument to createAsmPrinter · 9459832e
  David Blaikie authored Jan 18, 2015
  9459832e
- [PowerPC] Don't hard-code R2 as register when processing TOC relocations · 58884f9f
  Hal Finkel authored Jan 18, 2015
  58884f9f
- [PowerPC] Add some FIXMEs for fastcc and FPR <-> GPR moves · 8ea446b6
  Hal Finkel authored Jan 18, 2015
  8ea446b6
- [PowerPC] Initial PPC64 calling-convention changes for fastcc · f81b6dd7
  Hal Finkel authored Jan 18, 2015
  f81b6dd7
- [PM] Pull the analyses used for another utility routine into its API · b5797b65
  Chandler Carruth authored Jan 18, 2015
  b5797b65
- [PM] Sink the specific analyses preserved by SplitBlock into its · 32c52c7e
  Chandler Carruth authored Jan 18, 2015
  32c52c7e
- [PM] Replace another Pass argument with specific analyses that are · b5c11535
  Chandler Carruth authored Jan 18, 2015
  b5c11535
- [PM] Refactor how the LoopRotation pass access the DominatorTree. · 94209094
  Chandler Carruth authored Jan 18, 2015
  94209094
- [PM] Lift the actual analyses used into the inferface rather than · 5eee895c
  Chandler Carruth authored Jan 18, 2015
  5eee895c
- [PM] Now that LoopInfo isn't in the Pass type hierarchy, it is much · 691addc2
  Chandler Carruth authored Jan 18, 2015
  691addc2
Jan 17, 2015
- [PM] Cleanup more warnings my refactoring exposed where now we have · bc045a5a
  Chandler Carruth authored Jan 17, 2015
  bc045a5a
- [PM] Remove a dead field. · 24fd029a
  Chandler Carruth authored Jan 17, 2015
  24fd029a
- [PM] Split the LoopInfo object apart from the legacy pass, creating · 4f8f307c
  Chandler Carruth authored Jan 17, 2015
  4f8f307c
- [PowerPC] Don't list R11 as a patchpoint scratch register · c19805a7
  Hal Finkel authored Jan 17, 2015
  c19805a7