Commits

Branch selector

 penglai

User selector

d0k

Datepicker

All time

Commit History

Commits on Mar 11, 2021

• [HLO:MLIR] Make binary op type reification emit shape_of instead of tensor ops

  Show description for be34b71
  

  d0k

  authored and

  tensorflower-gardener

  committed
  be34b71

  Copy full SHA for be34b71

Commits on Mar 10, 2021

• [TF:XLA:MLIR] Lower SelectV1 to HLO with support for dynamic shapes

  Show description for 339ceec
  

  d0k

  authored and

  tensorflower-gardener

  committed
  339ceec

  Copy full SHA for 339ceec

Commits on Mar 9, 2021

• [TF:XLA:MLIR] Fix TF_RoundOp lowering for negative numbers

  Show description for 6fb2b8b
  

  d0k

  authored and

  tensorflower-gardener

  committed
  6fb2b8b

  Copy full SHA for 6fb2b8b

• Integrate LLVM at llvm/llvm-project@3f3f88fb9503

  Show description for b734c4a
  

  d0k

  authored and

  tensorflower-gardener

  committed
  b734c4a

  Copy full SHA for b734c4a

Commits on Mar 8, 2021

• [XLA:MLIR] Specify return types for SplitAt

  Show description for 8955cf8
  

  d0k

  authored and

  tensorflower-gardener

  committed
  8955cf8

  Copy full SHA for 8955cf8

Commits on Mar 5, 2021

• Integrate LLVM at llvm/llvm-project@9b302513f6d8

  Show description for 62d0787
  

  d0k

  authored and

  tensorflower-gardener

  committed
  62d0787

  Copy full SHA for 62d0787

Commits on Mar 4, 2021

• Integrate LLVM at llvm/llvm-project@b3a33553aec7

  Show description for 6983b30
  

  d0k

  authored and

  tensorflower-gardener

  committed
  6983b30

  Copy full SHA for 6983b30

• Integrate LLVM at llvm/llvm-project@c907681b077c

  Show description for 503babe
  

  d0k

  authored and

  tensorflower-gardener

  committed
  503babe

  Copy full SHA for 503babe

Commits on Mar 3, 2021

• Integrate LLVM at llvm/llvm-project@5d7e0a23c6f2

  Show description for 7ee2c03
  

  d0k

  authored and

  tensorflower-gardener

  committed
  7ee2c03

  Copy full SHA for 7ee2c03

• Integrate LLVM at llvm/llvm-project@8da090381d56

  Show description for 7a2f2a4
  

  d0k

  authored and

  tensorflower-gardener

  committed
  7a2f2a4

  Copy full SHA for 7a2f2a4

• Integrate LLVM at llvm/llvm-project@1a4990a4f71a

  Show description for c180f35
  

  d0k

  authored and

  tensorflower-gardener

  committed
  c180f35

  Copy full SHA for c180f35

Commits on Mar 2, 2021

• Integrate LLVM at llvm/llvm-project@7f086d74c347

  Show description for 613a751
  

  d0k

  authored and

  tensorflower-gardener

  committed
  613a751

  Copy full SHA for 613a751

Commits on Mar 1, 2021

• Integrate LLVM at llvm/llvm-project@99c24f7aa8cc

  Show description for 69759a8
  

  d0k

  authored and

  tensorflower-gardener

  committed
  69759a8

  Copy full SHA for 69759a8

• Filter static dimensions from dynamic_broadcast_in_dim's init_tensor

  Show description for 989b315
  

  d0k

  authored and

  tensorflower-gardener

  committed
  989b315

  Copy full SHA for 989b315

Commits on Feb 22, 2021

• [mlir][hlo] Add basic rank-specialization for select

  Show description for b299b26
  

  d0k

  authored and

  tensorflower-gardener

  committed
  b299b26

  Copy full SHA for b299b26

Commits on Feb 18, 2021

• [mlir][hlo] Refactor rank specialization to allow an arbitrary number of inputs

  Show description for bc2f30e
  

  d0k

  authored and

  tensorflower-gardener

  committed
  bc2f30e

  Copy full SHA for bc2f30e

• [mlir][hlo] Make select ready for dynamic shapes (ranked only for now)

  Show description for e7f2428
  

  d0k

  authored and

  tensorflower-gardener

  committed
  e7f2428

  Copy full SHA for e7f2428

Commits on Feb 15, 2021

• Integrate LLVM at llvm/llvm-project@715dc556b782

  Show description for 1ecfc7e
  

  d0k

  authored and

  tensorflower-gardener

  committed
  1ecfc7e

  Copy full SHA for 1ecfc7e

• [mhlo] Lower int->int cast to sign extension instead of zero extension

  Show description for ecf6728
  

  d0k

  authored and

  tensorflower-gardener

  committed
  ecf6728

  Copy full SHA for ecf6728

• [XLA:CPU] Remove uses of deprecated CreateAlignedLoad/Store builders

  Show description for 011e64e
  

  d0k

  authored and

  tensorflower-gardener

  committed
  011e64e

  Copy full SHA for 011e64e

• [mhlo] Lower float->bool to a comparison with zero

  Show description for 88e1026
  

  d0k

  authored and

  tensorflower-gardener

  committed
  88e1026

  Copy full SHA for 88e1026

Commits on Feb 10, 2021

• Qualify calls to isnan

  Show description for 95efe6e
  

  d0k

  authored and

  tensorflower-gardener

  committed
  95efe6e

  Copy full SHA for 95efe6e

Commits on Feb 9, 2021

• [XLA:CPU] Wire up support for Aarch64 on linux

  Show description for 942f315
  

  d0k

  authored and

  tensorflower-gardener

  committed
  942f315

  Copy full SHA for 942f315

• [mlir][kernelgen] Test ops that can return NaN with the default set of inputs

  Show description for cb9da9e
  

  d0k

  authored and

  tensorflower-gardener

  committed
  cb9da9e

  Copy full SHA for cb9da9e

• Make NaNs equal for testing

  Show description for 1b41b5a
  

  d0k

  authored and

  tensorflower-gardener

  committed
  1b41b5a

  Copy full SHA for 1b41b5a

Commits on Feb 4, 2021

• [MLIR:TF] Add a tf2tf lowering for SqrtGrad

  Show description for 5eb029f
  

  d0k

  authored and

  tensorflower-gardener

  committed
  5eb029f

  Copy full SHA for 5eb029f

Commits on Feb 1, 2021

• Migrate off deprecated llvm::VectorType::getNumElements

  Show description for b195261
  

  d0k

  authored and

  tensorflower-gardener

  committed
  b195261

  Copy full SHA for b195261

Commits on Jan 29, 2021

• [XLA:CPU] Turn off MLIR matmul codegen while we investigate an upstream regression

  Show description for b2e90ec
  

  d0k

  authored and

  tensorflower-gardener

  committed
  b2e90ec

  Copy full SHA for b2e90ec

Commits on Jan 25, 2021

• [kernelgen] Register and test minimum/maximum

  Show description for dcaa971
  

  d0k

  authored and

  tensorflower-gardener

  committed
  dcaa971

  Copy full SHA for dcaa971

• [mlir][hlo] Make min/max always propagate NaNs

  Show description for 07e726f
  

  d0k

  authored and

  tensorflower-gardener

  committed
  07e726f

  Copy full SHA for 07e726f

• [mlir][hlo] Make min/max always propagate NaNs

  Show description for 5933c33
  

  d0k

  authored and

  tensorflower-gardener

  committed
  5933c33

  Copy full SHA for 5933c33

Commits on Jan 13, 2021

• [XLA:CPU] Move elemental conv emitter to generic code path so it can share the multiply-add logic with dot

  Show description for 15e0f25
  

  d0k

  authored and

  tensorflower-gardener

  committed
  15e0f25

  Copy full SHA for 15e0f25

Commits on Jan 7, 2021

• [XLA:CPU] Move bf16->f32 conversion before fusion

  Show description for 7fa2309
  

  d0k

  authored and

  tensorflower-gardener

  committed
  7fa2309

  Copy full SHA for 7fa2309

• [XLA:CPU] Allow more types to be used in dot

  Show description for c21f7e2
  

  d0k

  authored and

  tensorflower-gardener

  committed
  c21f7e2

  Copy full SHA for c21f7e2

Commits on Dec 11, 2020

• [mlir][hlo] Fix lowering of NE comparison. It should return true if either side is NaN

  Show description for d1cf3a2
  

  d0k

  authored and

  tensorflower-gardener

  committed
  d1cf3a2

  Copy full SHA for d1cf3a2

Pagination

Previous

Next