Commits

Branch selector

 master

User selector

aschwaighofer

Datepicker

All time

Commit History

Commits on May 8, 2015

• ScheduleDAGInstrs: In functions with tail calls PseudoSourceValues are not non-aliasing distinct objects

  Show description for 75e36e8
  aschwaighofer

  committed
  75e36e8

  Copy full SHA for 75e36e8

Commits on Oct 26, 2014

• Add an option to the LTO code generator to disable vectorization during LTO

  Show description for b26fb77
  aschwaighofer

  committed
  b26fb77

  Copy full SHA for b26fb77

Commits on Oct 10, 2014

• SimplifyCFG: Don't convert phis into selects if we could remove undef behavior

  Show description for dd8c386
  aschwaighofer

  committed
  dd8c386

  Copy full SHA for dd8c386

Commits on Aug 7, 2014

• SLPVectorizer: Use the type of the value loaded/stored to get the ABI alignment

  Show description for 2158dec
  aschwaighofer

  committed
  2158dec

  Copy full SHA for 2158dec

Commits on Jun 26, 2014

• GVN: Preserve invariant.load metadata

  Show description for c2d93c4
  aschwaighofer

  committed
  c2d93c4

  Copy full SHA for c2d93c4

Commits on Jun 22, 2014

• LoopVectorizer: Fix a dominance issue

  Show description for 78f173a
  aschwaighofer

  committed
  78f173a

  Copy full SHA for 78f173a

Commits on Jun 18, 2014

• Add a triple so that right syntax is choosen on mac osx systems

  Show description for 5d5ddf9
  aschwaighofer

  committed
  5d5ddf9

  Copy full SHA for 5d5ddf9

Commits on May 29, 2014

• LoopVectorizer: Add a check that the backedge taken count + 1 does not overflow

  Show description for 06413cd
  aschwaighofer

  committed
  06413cd

  Copy full SHA for 06413cd

Commits on May 4, 2014

• SLPVectorizer: Bring back the insertelement patch (r205965) with fixes

  Show description for 28a739b
  aschwaighofer

  committed
  28a739b

  Copy full SHA for 28a739b

Commits on Apr 10, 2014

• Reapply "SLPVectorizer: Ignore users that are insertelements we can reschedule them"

  Show description for e2d124d
  aschwaighofer

  committed
  e2d124d

  Copy full SHA for e2d124d

Commits on Apr 9, 2014

• SLPVectorizer: Only vectorize intrinsics whose operands are widened equally

  Show description for b0ee237
  aschwaighofer

  committed
  b0ee237

  Copy full SHA for b0ee237

Commits on Mar 31, 2014

• Revert "SLPVectorizer: Ignore users that are insertelements we can reschedule them"

  Show description for 694365f
  aschwaighofer

  committed
  694365f

  Copy full SHA for 694365f

Commits on Mar 28, 2014

• SLPVectorizer: Take credit for free extractelement instructions

  Show description for ce36237
  aschwaighofer

  committed
  ce36237

  Copy full SHA for ce36237

• SLPVectorizer: Fix typos

  Show description for aa0a2a3
  aschwaighofer

  committed
  aa0a2a3

  Copy full SHA for aa0a2a3

• SLPVectorizer: Ignore users that are insertelements we can reschedule them

  Show description for b48eb2c
  aschwaighofer

  committed
  b48eb2c

  Copy full SHA for b48eb2c

Commits on Mar 26, 2014

• PR15967 Fix in basicaa for faulty returning no alias.

  Show description for 23463c9
  aschwaighofer

  committed
  23463c9

  Copy full SHA for 23463c9

Commits on Mar 12, 2014

• Fix whitespace in vectorizer example

  Show description for 8d46932
  aschwaighofer

  committed
  8d46932

  Copy full SHA for 8d46932

• Fix vectorizer docs.

  Show description for 6a7d263
  aschwaighofer

  committed
  6a7d263

  Copy full SHA for 6a7d263

Commits on Mar 7, 2014

• ISel: Make VSELECT selection terminate in cases where the condition type has to

  Show description for aa5b17b
  aschwaighofer

  committed
  aa5b17b

  Copy full SHA for aa5b17b

Commits on Mar 5, 2014

• LoopVectorizer: Preserve fast-math flags

  Show description for 9d84b4d
  aschwaighofer

  committed
  9d84b4d

  Copy full SHA for 9d84b4d

Commits on Feb 24, 2014

• SLPVectorizer: Try vectorizing 'splat' stores

  Show description for 68085c7
  aschwaighofer

  committed
  68085c7

  Copy full SHA for 68085c7

• LTO: Add the loop vectorizer to the LTO pipeline.

  Show description for 137342f
  aschwaighofer

  committed
  137342f

  Copy full SHA for 137342f

Commits on Feb 16, 2014

• SCEVExpander: Try hard not to create derived induction variables in other loops

  Show description for 2ced338
  aschwaighofer

  committed
  2ced338

  Copy full SHA for 2ced338

Commits on Feb 15, 2014

• Revert "SCEVExpander: Try hard not to create derived induction variables in other loops"

  Show description for a9db46b
  aschwaighofer

  committed
  a9db46b

  Copy full SHA for a9db46b

• SCEVExpander: Try hard not to create derived induction variables in other loops

  Show description for e672548
  aschwaighofer

  committed
  e672548

  Copy full SHA for e672548

Commits on Feb 8, 2014

• LoopVectorizer: Keep track of conditional store basic blocks

  Show description for 846acbe
  aschwaighofer

  committed
  846acbe

  Copy full SHA for 846acbe

Commits on Feb 2, 2014

• LoopVectorizer: Enable unrolling of conditional stores and the load/store

  Show description for a16c1b5
  aschwaighofer

  committed
  a16c1b5

  Copy full SHA for a16c1b5

Commits on Feb 1, 2014

• ARMTTI: We don't have 16 allocatable scalar registers

  Show description for 991dd3b
  aschwaighofer

  committed
  991dd3b

  Copy full SHA for 991dd3b

Commits on Jan 29, 2014

• LoopVectorizer: Add a test case for unrolling of small loops that need a runtime

  Show description for fb70e11
  aschwaighofer

  committed
  fb70e11

  Copy full SHA for fb70e11

• LoopVectorizer: Don't count the induction variable multiple times

  Show description for 8dc253e
  aschwaighofer

  committed
  8dc253e

  Copy full SHA for 8dc253e

Commits on Jan 28, 2014

• LoopVectorize: Support conditional stores by scalarizing

  Show description for a47aa4b
  aschwaighofer

  committed
  a47aa4b

  Copy full SHA for a47aa4b

Commits on Jan 19, 2014

• LoopVectorizer: A reduction that has multiple uses of the reduction value is not

  Show description for 2becaaf
  aschwaighofer

  committed
  2becaaf

  Copy full SHA for 2becaaf

Commits on Jan 16, 2014

• BasicAA: We need to check both access sizes when comparing a gep and an

  Show description for 8963071
  aschwaighofer

  committed
  8963071

  Copy full SHA for 8963071

Commits on Jan 15, 2014

• LoopVectorize: Only strip casts from integer types when replacing symbolic

  Show description for e96fec2
  aschwaighofer

  committed
  e96fec2

  Copy full SHA for e96fec2

Commits on Jan 11, 2014

• LoopVectorizer: Enable strided memory accesses versioning per default

  Show description for 73c9559
  aschwaighofer

  committed
  73c9559

  Copy full SHA for 73c9559

Pagination

Previous

Next