Commits

Breadcrumbs

History for

onsimd

User selector

All users

All time

Commit History

Commits on Oct 18, 2013

• integrate the SIMD polyfill code

  kripken

  committed
  4cdc5fe

  Copy full SHA for 4cdc5fe

• basic support for uint32x4

  kripken

  committed
  d32d18a

  Copy full SHA for d32d18a

• shufflevector

  kripken

  committed
  b085d4c

  Copy full SHA for b085d4c

• insertelement

  kripken

  committed
  e623c44

  Copy full SHA for e623c44

• vector load

  kripken

  committed
  5708083

  Copy full SHA for 5708083

• add basic float simd calls

  kripken

  committed
  f439b01

  Copy full SHA for f439b01

• define vector literal type

  kripken

  committed
  775e1fa

  Copy full SHA for 775e1fa

• setup to do vector ops in processMathop

  kripken

  committed
  0567f27

  Copy full SHA for 0567f27

• use HEAPF32 for <4 x float>

  kripken

  committed
  e717685

  Copy full SHA for e717685

Commits on Oct 17, 2013

• emit only tokens from tokenize, avoid allocating an object to wrap them

  kripken

  committed
  128137d

  Copy full SHA for 128137d

• remove .item from tokenizer output

  kripken

  committed
  39cb16d

  Copy full SHA for 39cb16d

Commits on Oct 16, 2013

• do not add ',' tokens during param list processing

  kripken

  committed
  87cdaae

  Copy full SHA for 87cdaae

Commits on Oct 12, 2013

• avoid creating unnecessary objects when calling tokenize

  kripken

  committed
  66d334a

  Copy full SHA for 66d334a

Commits on Oct 10, 2013

• do not emit postSets for function indexing etc., bake them into the memory initializer

  kripken

  committed
  894db65

  Copy full SHA for 894db65

Commits on Oct 9, 2013

• optimize parseNumerical

  kripken

  committed
  9bc421c

  Copy full SHA for 9bc421c

Commits on Oct 8, 2013

• use ident in alloca intertype

  kripken

  committed
  0e4520e

  Copy full SHA for 0e4520e

Commits on Oct 3, 2013

• do not tack on .0 to Infinity or NaN

  kripken

  committed
  04732d0

  Copy full SHA for 04732d0

• do not remove '.' from floats in parseNumerical

  kripken

  committed
  d476a58

  Copy full SHA for d476a58

• overflow numbers directly in js compiler

  kripken

  committed
  3f03cd8

  Copy full SHA for 3f03cd8

• optimize Math.x to Math_x

  kripken

  committed
  6e8ba91

  Copy full SHA for 6e8ba91

• if not running js opts, ensure a . in asmEnsureFloat immediately

  kripken

  committed
  d229313

  Copy full SHA for d229313

• add explicit vars for illegal mathops and landingpad

  kripken

  committed
  1a9b9af

  Copy full SHA for 1a9b9af

• register phi variables in function

  kripken

  committed
  07b106e

  Copy full SHA for 07b106e

• chunk label init with the rest

  kripken

  committed
  435f17d

  Copy full SHA for 435f17d

• remove unnecessary function param coercion parens

  kripken

  committed
  911c152

  Copy full SHA for 911c152

• remove whitespace in mathops

  kripken

  committed
  08c73e4

  Copy full SHA for 08c73e4

Commits on Sep 30, 2013

• Merge branch 'incoming' of github.com:ngld/emscripten into incoming

  Show description for b01d44e
  kripken

  committed
  b01d44e

  Copy full SHA for b01d44e

Commits on Sep 28, 2013

• fast-path simple loads

  kripken

  committed
  26cab17

  Copy full SHA for 26cab17

Commits on Sep 27, 2013

• improve processMathOp quoting

  kripken

  committed
  f6eb863

  Copy full SHA for f6eb863

• better check for when things need quoting

  kripken

  committed
  a242ce0

  Copy full SHA for a242ce0

• === in isLocalVar

  kripken

  committed
  076828e

  Copy full SHA for 076828e

• optimize parseLLVMString

  kripken

  committed
  9dae026

  Copy full SHA for 9dae026

• optimize makeSignOp

  kripken

  committed
  94b9d47

  Copy full SHA for 94b9d47

Commits on Sep 26, 2013

• hackish optimization for pre processing of heap inits, avoid pure numeric strings

  kripken

  committed
  c4258fe

  Copy full SHA for c4258fe

• fix getFastValue regression

  kripken

  committed
  2aa7826

  Copy full SHA for 2aa7826