Commits

Branch selector

 master

User selector

mssimpso

Datepicker

All time

Commit History

Commits on Apr 26, 2017

• [LV] Handle external uses of floating-point induction variables

  Show description for 8619795
  mssimpso

  committed
  8619795

  Copy full SHA for 8619795

Commits on Apr 21, 2017

• [LV] Model if-converted phi node costs

  Show description for 915a4bc
  mssimpso

  committed
  915a4bc

  Copy full SHA for 915a4bc

Commits on Apr 10, 2017

• [ARM/AArch64] Ensure valid vector element types for interleaved accesses

  Show description for 42df88e
  mssimpso

  committed
  42df88e

  Copy full SHA for 42df88e

Commits on Apr 7, 2017

• Reapply r298620: [LV] Vectorize GEPs

  Show description for ccc38cc
  mssimpso

  committed
  ccc38cc

  Copy full SHA for ccc38cc

Commits on Apr 5, 2017

• [LV] Make test case more robust

  Show description for a9eefb0
  mssimpso

  committed
  a9eefb0

  Copy full SHA for a9eefb0

Commits on Mar 29, 2017

• [InstCombine] Correct the check for vector GEPs

  Show description for 9467021
  mssimpso

  committed
  9467021

  Copy full SHA for 9467021

Commits on Mar 27, 2017

• [LV] Transform truncations of non-primary induction variables

  Show description for fcdf36f
  mssimpso

  committed
  fcdf36f

  Copy full SHA for fcdf36f

Commits on Mar 23, 2017

• [LV] Vectorize GEPs

  Show description for 77a842c
  mssimpso

  committed
  77a842c

  Copy full SHA for 77a842c

• [LV] Delete unneeded scalar GEP creation code

  Show description for 49901e7
  mssimpso

  committed
  49901e7

  Copy full SHA for 49901e7

Commits on Mar 8, 2017

• [LV] Select legal insert point when fixing first-order recurrences

  Show description for 8ca15c2
  mssimpso

  committed
  8ca15c2

  Copy full SHA for 8ca15c2

• [LV] Make the test case for PR30183 less fragile

  Show description for b9f3ad9
  mssimpso

  committed
  b9f3ad9

  Copy full SHA for b9f3ad9

• [LV] Add missing check labels to tests and reformat

  Show description for d20a0da
  mssimpso

  committed
  d20a0da

  Copy full SHA for d20a0da

Commits on Mar 7, 2017

• [LV] Consider users that are memory accesses in uniforms expansion step

  Show description for 30bc56b
  mssimpso

  committed
  30bc56b

  Copy full SHA for 30bc56b

Commits on Mar 2, 2017

• [ARM/AArch64] Update costs for interleaved accesses with wide types

  Show description for 201896c
  mssimpso

  committed
  201896c

  Copy full SHA for 201896c

• [ARM/AArch64] Support wide interleaved accesses

  Show description for 6523b45
  mssimpso

  committed
  6523b45

  Copy full SHA for 6523b45

• [LV] Considier non-consecutive but vectorizable accesses for VF selection

  Show description for 465f5a1
  mssimpso

  committed
  465f5a1

  Copy full SHA for 465f5a1

Commits on Feb 24, 2017

• [LV] Merge floating-point and integer induction widening code

  Show description for aceef6f
  mssimpso

  committed
  aceef6f

  Copy full SHA for aceef6f

Commits on Feb 22, 2017

• [LV] Update floating-point induction test checks (NFC)

  Show description for 322ee60
  mssimpso

  committed
  322ee60

  Copy full SHA for 322ee60

• [LV] Add scalar floating-point induction test (NFC)

  Show description for ef6eb7f
  mssimpso

  committed
  ef6eb7f

  Copy full SHA for ef6eb7f

Commits on Feb 17, 2017

• [LAA] Remove unused code (NFC)

  Show description for b82f5b8
  mssimpso

  committed
  b82f5b8

  Copy full SHA for b82f5b8

• [LV] Remove constant restriction for vector phi creation

  Show description for 132a62c
  mssimpso

  committed
  132a62c

  Copy full SHA for 132a62c

Commits on Feb 14, 2017

• Reapply "[LV] Extend trunc optimization to all IVs with constant integer steps"

  Show description for 06f1a4b
  mssimpso

  committed
  06f1a4b

  Copy full SHA for 06f1a4b

Commits on Feb 13, 2017

• Revert "[LV] Extend trunc optimization to all IVs with constant integer steps"

  Show description for d1fc544
  mssimpso

  committed
  d1fc544

  Copy full SHA for d1fc544

• [LV] Extend trunc optimization to all IVs with constant integer steps

  Show description for f32a3fd
  mssimpso

  committed
  f32a3fd

  Copy full SHA for f32a3fd

Commits on Feb 10, 2017

• [LV] Remove type restriction for vector phi creation

  Show description for 21785b8
  mssimpso

  committed
  21785b8

  Copy full SHA for 21785b8

Commits on Feb 7, 2017

• [LV] Add new ARM/AArch64 interleaved access cost model tests (NFC)

  Show description for 9e8cc9c
  mssimpso

  committed
  9e8cc9c

  Copy full SHA for 9e8cc9c

• [LV] Simplify ARM/AArch64 interleaved access cost model tests (NFC)

  Show description for 85dcf5e
  mssimpso

  committed
  85dcf5e

  Copy full SHA for 85dcf5e

Commits on Feb 1, 2017

• [LV] Move interleaved access helper functions to VectorUtils (NFC)

  Show description for aa96bc9
  mssimpso

  committed
  aa96bc9

  Copy full SHA for aa96bc9

Commits on Jan 31, 2017

• Fix VectorUtils include guard name (NFC)

  Show description for 91af052
  mssimpso

  committed
  91af052

  Copy full SHA for 91af052

Commits on Jan 27, 2017

• [ARM/AArch64] Relocate and update InterleavedAccessPass tests (NFC)

  Show description for 6ea1012
  mssimpso

  committed
  6ea1012

  Copy full SHA for 6ea1012

Commits on Jan 17, 2017

• [LV] Add requires asserts to test case

  Show description for 1b2cd63
  mssimpso

  committed
  1b2cd63

  Copy full SHA for 1b2cd63

• [LV] Mark non-consecutive-like pointers non-uniform

  Show description for caab3a8
  mssimpso

  committed
  caab3a8

  Copy full SHA for caab3a8

Commits on Jan 9, 2017

• [LV] Fix-up external IV users after updating dominator tree

  Show description for 0a3fcf0
  mssimpso

  committed
  0a3fcf0

  Copy full SHA for 0a3fcf0

Commits on Dec 16, 2016

• Reapply "[LV] Enable vectorization of loops with conditional stores by default"

  Show description for 3499b6e
  mssimpso

  committed
  3499b6e

  Copy full SHA for 3499b6e

• [LV] Don't attempt to type-shrink scalarized instructions

  Show description for 48a6409
  mssimpso

  committed
  48a6409

  Copy full SHA for 48a6409

Pagination

Previous

Next