Commits · e2d437148ae37c9fbafe74b61cb4a9d81388af1b · Roger Ferrer / llvm-epi · GitLab

Mar 11, 2014
- GVN: merge overflow intrinsics with non-overflow instructions. · e2d43714
  Erik Verbruggen authored Mar 11, 2014
  e2d43714
Mar 10, 2014
- Use discriminator information in sample profiles. · 92aa8c22
  Diego Novillo authored Mar 10, 2014
  92aa8c22
- MemCpyOpt: When merging memsets also merge the trivial case of two memsets... · 3ef5e46b
  Benjamin Kramer authored Mar 10, 2014
  3ef5e46b
- For functions with ARM target specific calling convention, when simplify-libcall · 0e8f4612
  Evan Cheng authored Mar 10, 2014
  0e8f4612
Mar 09, 2014
- Revert r203230, "CodeGenPrep: sink extends of illegal types into use block." · 1783e1e9
  NAKAMURA Takumi authored Mar 09, 2014
  1783e1e9
- IR: Change inalloca's grammar a bit · c4ab61cb
  David Majnemer authored Mar 09, 2014
  c4ab61cb
Mar 07, 2014
- CodeGenPrep: sink extends of illegal types into use block. · ad3d81d3
  Tim Northover authored Mar 07, 2014
  ad3d81d3
- InstCombine: form shuffles from wider range of insert/extractelements · fad2761c
  Tim Northover authored Mar 07, 2014
  fad2761c
- Allow constant folding of round function whenever feasible · b67688a8
  Karthik Bhat authored Mar 07, 2014
  b67688a8
Mar 06, 2014
- Allow constant folding of copysign · daa8cd10
  Karthik Bhat authored Mar 06, 2014
  daa8cd10
- Change math intrinsic attributes from readonly to readnone. These · b741b945
  Raul E. Silvera authored Mar 06, 2014
  b741b945
Mar 05, 2014
- LoopVectorizer: Preserve fast-math flags · ab12363c
  Arnold Schwaighofer authored Mar 05, 2014
  ab12363c
- ConstantFolding: Also fold the vector overloads of our math intrinsics. · 061d147f
  Benjamin Kramer authored Mar 05, 2014
  061d147f
- Trivial test commit. · 18ebc7cd
  Raul E. Silvera authored Mar 05, 2014
  18ebc7cd
- Allow constant folding of fma and fmuladd · 8377858c
  Matt Arsenault authored Mar 05, 2014
  8377858c
Mar 03, 2014
- Pass to emit DWARF path discriminators. · f5041ce5
  Diego Novillo authored Mar 03, 2014
  f5041ce5
Feb 27, 2014
- Add a debug info code generation level to the compile unit metadata · 75d49db1
  Eric Christopher authored Feb 27, 2014
  75d49db1
Feb 26, 2014
- Fix broken FileCheck prefixes · 0a0c674b
  Nico Rieck authored Feb 26, 2014
  0a0c674b
- GlobalOpt: Apply fastcc to internal x86_thiscallcc functions · 22869378
  Reid Kleckner authored Feb 26, 2014
  22869378
- Fix broken FileCheck prefix · 5645b363
  Nico Rieck authored Feb 26, 2014
  5645b363
- Fix PR18165: LSR must avoid scaling factors that exceed the limit on truncated use. · 429e9edd
  Andrew Trick authored Feb 26, 2014
  429e9edd
- [SROA] Use the correct index integer size in GEPs through non-default · dfb2efd0
  Chandler Carruth authored Feb 26, 2014
  dfb2efd0
- [SROA] Teach SROA how to handle pointers from address spaces other than · 286d87ed
  Chandler Carruth authored Feb 26, 2014
  286d87ed
- [SROA] Split the alignment computation complete for the memcpy rewriting · aa72b93a
  Chandler Carruth authored Feb 26, 2014
  aa72b93a
- [SROA] Fix PR18615 with some long overdue simplifications to the bounds · 6aedc106
  Chandler Carruth authored Feb 26, 2014
  6aedc106
Feb 25, 2014
- [SROA] Fix another instability in SROA with respect to the slice · 3bf18ed5
  Chandler Carruth authored Feb 25, 2014
  3bf18ed5
Feb 24, 2014
- SLPVectorizer: Try vectorizing 'splat' stores · 9611d23d
  Arnold Schwaighofer authored Feb 24, 2014
  9611d23d
Feb 21, 2014
- Make sure that value handle users see the transformation of an indirect call... · 75080ff2
  Nick Lewycky authored Feb 20, 2014
  75080ff2
Feb 19, 2014
- X86: move test requiring X86TargetLowering info into its own directory · d495642c
  Tim Northover authored Feb 19, 2014
  d495642c
- Try addding datalayout in case that's what Hexagon doesn't like. · d496355b
  Tim Northover authored Feb 19, 2014
  d496355b
- X86 CodeGenPrep: sink shufflevectors before shifts · aeb8e06d
  Tim Northover authored Feb 19, 2014
  aeb8e06d
Feb 17, 2014
- fix for null VectorizedValue assertion in the SLP Vectorizer (in function... · 7a463d06
  Gerolf Hoflehner authored Feb 17, 2014
  7a463d06
Feb 16, 2014
- SCEVExpander: Try hard not to create derived induction variables in other loops · 26f567d8
  Arnold Schwaighofer authored Feb 16, 2014
  26f567d8
- Fix broken CHECK lines · 76471787
  Nico Rieck authored Feb 16, 2014
  76471787
Feb 15, 2014
- Revert "SCEVExpander: Try hard not to create derived induction variables in other loops" · 847d9614
  Arnold Schwaighofer authored Feb 15, 2014
  847d9614
- SCEVExpander: Try hard not to create derived induction variables in other loops · 1e12f856
  Arnold Schwaighofer authored Feb 15, 2014
  1e12f856
Feb 14, 2014
- Do more addrspacecast transforms that happen for bitcast. · aa689f50
  Matt Arsenault authored Feb 14, 2014
  aa689f50
Feb 13, 2014
- Re-commit: Demote EmitRawText call in AsmPrinter::EmitInlineAsm() and remove... · 753e1762
  Daniel Sanders authored Feb 13, 2014
  753e1762
- GlobalOpt: Aliases don't have sections, don't copy them when replacing · 22b19da9
  Reid Kleckner authored Feb 13, 2014
  22b19da9
- Remove a very old instcombine where we would turn sequences of selects into · 883b5add
  Owen Anderson authored Feb 12, 2014
  883b5add