Commits

Branch selector

 master

User selector

mcinally

Datepicker

All time

Commit History

Commits on Sep 21, 2016

• [AVX512] Fix return types on int_x86_avx512_gatherXXX_di intrinsics

  Show description for 3d9ce2b
  mcinally

  committed
  3d9ce2b

  Copy full SHA for 3d9ce2b

Commits on Feb 16, 2015

• [AVX512] Make 512b vector floating point rounds legal on AVX512.

  Show description for cdddfe0
  mcinally

  committed
  cdddfe0

  Copy full SHA for cdddfe0

Commits on Dec 11, 2014

• [AVX512] Add support for 512b variable bit shift intrinsics.

  Show description for 14273ae
  mcinally

  committed
  14273ae

  Copy full SHA for 14273ae

Commits on Nov 25, 2014

• [AVX512] Add 512b integer shift by variable intrinsics and patterns.

  Show description for 9f4bb04
  mcinally

  committed
  9f4bb04

  Copy full SHA for 9f4bb04

Commits on Nov 14, 2014

• [AVX512] Add 512b masked integer shift by immediate patterns.

  Show description for b3625eb
  mcinally

  committed
  b3625eb

  Copy full SHA for b3625eb

Commits on Nov 12, 2014

• [AVX512] Add integer shift by immediate intrinsics.

  Show description for be30336
  mcinally

  committed
  be30336

  Copy full SHA for be30336

Commits on Jul 15, 2014

• Revert r213070. It's breaking the build in MCELFStreamer::EmitInstToData(...).

  Show description for ef88687
  mcinally

  committed
  ef88687

  Copy full SHA for ef88687

• Add x86 patterns to match a specific add-with-carry.

  Show description for 160c4cb
  mcinally

  committed
  160c4cb

  Copy full SHA for 160c4cb

Commits on Jul 9, 2014

• Update ReleaseNotes to mention Atomic NAND semantic changes.

  Show description for 9753da1
  mcinally

  committed
  9753da1

  Copy full SHA for 9753da1

Commits on Jun 18, 2014

• Add pattern for unsigned v4i32->v4f64 convert on AVX512.

  Show description for c52345c
  mcinally

  committed
  c52345c

  Copy full SHA for c52345c

Commits on Jun 16, 2014

• Hook up vector int_ctlz for AVX512.

  Show description for 5d57928
  mcinally

  committed
  5d57928

  Copy full SHA for 5d57928

Commits on Jun 13, 2014

• Fix bad copy-and-paste from r210652. AVX512 masked leading zero intrinsics.

  Show description for 1d5e379
  mcinally

  committed
  1d5e379

  Copy full SHA for 1d5e379

• Add HasCDI predicate to AVX512 VPBROADCASTM*.

  Show description for a411d0e
  mcinally

  committed
  a411d0e

  Copy full SHA for a411d0e

Commits on Jun 11, 2014

• Add AVX512 masked leadz instrinsic support.

  Show description for 998d8f5
  mcinally

  committed
  998d8f5

  Copy full SHA for 998d8f5

Commits on Mar 26, 2014

• Fix AVX512 Gather and Scatter execution domains.

  Show description for 4de1039
  mcinally

  committed
  4de1039

  Copy full SHA for 4de1039

Commits on Mar 25, 2014

• Fix AVX2 Gather execution domains.

  Show description for 3ec862b
  mcinally

  committed
  3ec862b

  Copy full SHA for 3ec862b

Commits on Mar 5, 2014

• Lower AVX v4i64->v4i32 truncate to one shuffle.

  Show description for f3ff7c3
  mcinally

  committed
  f3ff7c3

  Copy full SHA for f3ff7c3

Commits on Feb 19, 2014

• Fix AVX512 vector sqrt assembly strings.

  Show description for 35f15e5
  mcinally

  committed
  35f15e5

  Copy full SHA for 35f15e5

Commits on Jan 13, 2014

• Clean up RUN command for Assembler/getInt.ll.

  Show description for 9ad2161
  mcinally

  committed
  9ad2161

  Copy full SHA for 9ad2161

• Fix uninitialized warning in llvm/lib/IR/DataLayout.cpp.

  Show description for 486fdf2
  mcinally

  committed
  486fdf2

  Copy full SHA for 486fdf2

Commits on Jan 7, 2014

• Fix uninitialized variable warning in DataLayout.

  Show description for 67cc1dc
  mcinally

  committed
  67cc1dc

  Copy full SHA for 67cc1dc

Commits on Dec 6, 2013

• Update AVX512 vector blend intrinsic names.

  Show description for febc28b
  mcinally

  committed
  febc28b

  Copy full SHA for febc28b

Commits on Dec 5, 2013

• Add FileCheck statements for r196435.

  Show description for f6770bc
  mcinally

  committed
  f6770bc

  Copy full SHA for f6770bc

• Add AVX512 patterns for v16i32 broadcast and v2i64 zero extend load.

  Show description for 6c8fadd
  mcinally

  committed
  6c8fadd

  Copy full SHA for 6c8fadd

Commits on Dec 4, 2013

• Fix assembly syntax for AVX512 vector blend instructions.

  Show description for 6d3d93c
  mcinally

  committed
  6d3d93c

  Copy full SHA for 6d3d93c

• Suppress '(x < y) ? a : 0 -> (x < y) & a' transform on X86 architectures with dedicated mask registers.

  Show description for 8095580
  mcinally

  committed
  8095580

  Copy full SHA for 8095580

Commits on Nov 26, 2013

• Add an intrinsic for the SSE2 PAUSE instruction.

  Show description for 0e6ec12
  mcinally

  committed
  0e6ec12

  Copy full SHA for 0e6ec12

Commits on Nov 19, 2013

• Fix assembly operands for the SSE2 cvtsd2ss instruction.

  Show description for c5a925c
  mcinally

  committed
  c5a925c

  Copy full SHA for c5a925c

Commits on Nov 15, 2013

• Add AVX512 unmasked FMA intrinsics and support.

  Show description for 28e12e9
  mcinally

  committed
  28e12e9

  Copy full SHA for 28e12e9

Commits on Nov 4, 2013

• Add support for AVX512 masked vector blend intrinsics.

  Show description for 2e58f1d
  mcinally

  committed
  2e58f1d

  Copy full SHA for 2e58f1d

Commits on Oct 31, 2013

• Add AVX512 unmasked integer broadcast intrinsics and support.

  Show description for 424f197
  mcinally

  committed
  424f197

  Copy full SHA for 424f197

Commits on Oct 30, 2013

• Refactor the AVX512 intrinsics. Cluster the intrinsics into the appropriate vector extension class within the .td file.

  Show description for 95efb03
  mcinally

  committed
  95efb03

  Copy full SHA for 95efb03