Commits

Branch selector

 master

User selector

milseman

Datepicker

All time

Commit History

Commits on Dec 13, 2012

• Pattern matching code for intrinsics.

  Show description for 8ad435f
  milseman

  committed
  8ad435f

  Copy full SHA for 8ad435f

• m_CombineOr and m_CombineAnd pattern combinators

  Show description for 8961924
  milseman

  committed
  8961924

  Copy full SHA for 8961924

Commits on Dec 12, 2012

• Restore the PHI optimization I accidently removed

  Show description for 94993e3
  milseman

  committed
  94993e3

  Copy full SHA for 94993e3

• Remove trailing whitespace

  Show description for 4d96e6f
  milseman

  committed
  4d96e6f

  Copy full SHA for 4d96e6f

• Have SimplifyBinOp call the new FAdd/FSub/FMul helpers, with fast-math flags off

  Show description for d0a0d22
  milseman

  committed
  d0a0d22

  Copy full SHA for d0a0d22

• Remove redunant optimizations from InstCombine, instead call the appropriate functions from SimplifyInstruction

  Show description for c244f38
  milseman

  committed
  c244f38

  Copy full SHA for c244f38

• Added a slew of SimplifyInstruction floating-point optimizations, many of which take advantage of fast-math flags. Test cases included.

  Show description for 09ee250
  milseman

  committed
  09ee250

  Copy full SHA for 09ee250

• Pattern matchers for floating point values

  Show description for 1d4b150
  milseman

  committed
  1d4b150

  Copy full SHA for 1d4b150

• Remove FIXMEs surrounding Constant[Data]Vectors, instead

  Show description for 08b25ce
  milseman

  committed
  08b25ce

  Copy full SHA for 08b25ce

Commits on Dec 9, 2012

• Reorganize FastMathFlags to be a wrapper around unsigned, and streamline some interfaces.

  Show description for 1638b83
  milseman

  committed
  1638b83

  Copy full SHA for 1638b83

• Have the bitcode reader/writer just use FPMathOperator's fast math enum directly

  Show description for 855d025
  milseman

  committed
  855d025

  Copy full SHA for 855d025

Commits on Dec 7, 2012

• Remove trailing whitespace

  Show description for 829c8bd
  milseman

  committed
  829c8bd

  Copy full SHA for 829c8bd

Commits on Dec 6, 2012

• Have CannotBeNegativeZero() be aware of the nsz fast-math flag

  Show description for 85893f4
  milseman

  committed
  85893f4

  Copy full SHA for 85893f4

Commits on Dec 3, 2012

• Minor tweaking to SmallVector static size.

  Show description for cb66b5d
  milseman

  committed
  cb66b5d

  Copy full SHA for cb66b5d

• Since this SmallVector immediately grows on the next line, don't waste stack space. SmallVector is still needed due to existing APIs growing their arguments

  Show description for dbd7b95
  milseman

  committed
  dbd7b95

  Copy full SHA for dbd7b95

Commits on Dec 1, 2012

• Update the emacs mode to recognize fadd, fsum, fmul, fdiv, frem, fcmp, icmp

  Show description for 1854e14
  milseman

  committed
  1854e14

  Copy full SHA for 1854e14

Commits on Nov 30, 2012

• Vim mode updated to recognize fast-math flags

  Show description for 79f09f7
  milseman

  committed
  79f09f7

  Copy full SHA for 79f09f7

• Removed redundancy in wording

  Show description for 5ba39ae
  milseman

  committed
  5ba39ae

  Copy full SHA for 5ba39ae

• Clean up the documentation to have a common description area for fast-math flags.

  Show description for 01b3710
  milseman

  committed
  01b3710

  Copy full SHA for 01b3710

Commits on Nov 29, 2012

• copyFastMathFlags utility and test case

  Show description for 4b896dd
  milseman

  committed
  4b896dd

  Copy full SHA for 4b896dd

Commits on Nov 28, 2012

• Whoops, fixed bad merge

  Show description for 3d15a91
  milseman

  committed
  3d15a91

  Copy full SHA for 3d15a91

• Fixed bad test case

  Show description for 9eec659
  milseman

  committed
  9eec659

  Copy full SHA for 9eec659

• Fast-math: IRBuilder test for creating instructions with fast-math flags

  Show description for f564e9f
  milseman

  committed
  f564e9f

  Copy full SHA for f564e9f

• Fast-math: Extend IRBuilder to have settable FastMathFlags to create instructions with

  Show description for 9992bf7
  milseman

  committed
  9992bf7

  Copy full SHA for 9992bf7

• Fast-math comments and convenience method

  Show description for 0d38424
  milseman

  committed
  0d38424

  Copy full SHA for 0d38424

Commits on Nov 27, 2012

• Fast-math test for SimplifyInstruction: fold multiply by 0

  Show description for bc43fe1
  milseman

  committed
  bc43fe1

  Copy full SHA for bc43fe1

• Fast-math flags documentation added to LangRef

  Show description for a33e3d7
  milseman

  committed
  a33e3d7

  Copy full SHA for a33e3d7

• Fast-math test for SimplifyInstruction: fold multiply by 0

  Show description for bb22ad1
  milseman

  committed
  bb22ad1

  Copy full SHA for bb22ad1

• Fast-math optimization: fold multiply by zero

  Show description for eb61c92
  milseman

  committed
  eb61c92

  Copy full SHA for eb61c92

• Fast-math test case for bitcode and textual reading/writing

  Show description for a52f3ae
  milseman

  committed
  a52f3ae

  Copy full SHA for a52f3ae

• Fast-math flags for the bitcode

  Show description for 495d10a
  milseman

  committed
  495d10a

  Copy full SHA for 495d10a

• Fast-math flags for LLVM IR parsing and printing

  Show description for 15c13d3
  milseman

  committed
  15c13d3

  Copy full SHA for 15c13d3

• Fast-math interfaces for Instructions

  Show description for 125fc7f
  milseman

  committed
  125fc7f

  Copy full SHA for 125fc7f

• Fast-math flags added to FPMathOperator.

  Show description for ab4649b
  milseman

  committed
  ab4649b

  Copy full SHA for ab4649b

Commits on Nov 16, 2012

• Remove trailing whitespace

  Show description for 96952bd
  milseman

  committed
  96952bd

  Copy full SHA for 96952bd

Pagination

Previous

Next