experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
STT-tensorflow/tensorflow/compiler/xla/service/llvm_ir
History
Rahul Joshi 35fc7c9443 [XLA:GPU] Migrate convolution thunk emission to MLIR 
- Map MLIR LHLO Conv operations to ConvolutionThunk.
- Refactor GetGpuConvConfig() to take in a conv descriptor, so that the same function
  can be used to emit a thunk from XLA HLO and MLIR representations.
- Change XlaConvLayoutsToStreamExecutorLayouts() to check shape<->layout
  compatibility instead of layout equality, since some XLA layouts with unit dimensions
   are not preserved in XLA HLO Shape -> MLIR -> XLA HLO Shape path.
- Note that window reversal is not yet representable in HLO/LHLO, so fall back to
   XLA HLO based thunk emission for that case.

PiperOrigin-RevId: 347529995
Change-Id: I883ed8efe3d73fee2e4cbba4fa1173d5f67be7d8
2020-12-14 20:55:56 -08:00
..
alias_analysis_test.cc
Change XLA's default to disable cpu_fast_math options with the exception of min_max behavior.
2020-05-15 13:11:47 -07:00
alias_analysis.cc
…
alias_analysis.h
…
buffer_assignment_util.cc
[XLA/GPU] Migrate all unnested elementwise emitters.
2020-12-11 12:48:29 -08:00
buffer_assignment_util.h
[XLA/GPU] Migrate all unnested elementwise emitters.
2020-12-11 12:48:29 -08:00
BUILD
[MLIR] Migrate loop fusion to take LMHLO.
2020-10-22 13:59:50 -07:00
dynamic_update_slice_util.cc
[XLA] Merge PrepareGeneratorRecursively and GetGenerator. Now GetGenerator will lazily create generators that are requested. NFC.
2020-10-29 18:55:13 -07:00
dynamic_update_slice_util.h
[XLA] Simplify FusedIrEmitter (NFC):
2020-10-29 12:32:59 -07:00
fused_ir_emitter.cc
[XLA] Merge PrepareGeneratorRecursively and GetGenerator. Now GetGenerator will lazily create generators that are requested. NFC.
2020-10-29 18:55:13 -07:00
fused_ir_emitter.h
[XLA] Merge PrepareGeneratorRecursively and GetGenerator. Now GetGenerator will lazily create generators that are requested. NFC.
2020-10-29 18:55:13 -07:00
ir_array_test.cc
[MLIR] Migrate loop fusion to take LMHLO.
2020-10-22 13:59:50 -07:00
ir_array.cc
[XLA:GPU] Migrate convolution thunk emission to MLIR
2020-12-14 20:55:56 -08:00
ir_array.h
[XLA:GPU] Migrate convolution thunk emission to MLIR
2020-12-14 20:55:56 -08:00
ir_builder_mixin.h
[XLA] Remove deprecated function. It's unused.
2020-04-14 04:15:47 -07:00
kernel_support_library.cc
…
kernel_support_library.h
…
llvm_loop.cc
Update paths to LLVM.
2019-11-04 13:53:39 -08:00
llvm_loop.h
…
llvm_util.cc
Integrate LLVM at llvm/llvm-project@8cdf4920c4
2020-11-30 19:11:11 -08:00
llvm_util.h
Roll forward XLA GPU LHLO sort op migration
2020-08-13 12:00:17 -07:00
loop_emitter.cc
Merge pull request #42683 from nouiz:upstream_master_grid_size
2020-10-06 04:04:24 -07:00
loop_emitter.h
Merge pull request #42683 from nouiz:upstream_master_grid_size
2020-10-06 04:04:24 -07:00
math_ops.cc
Enable XLA:CPU fast math for min/max by default to be similar to TF's behavior.
2020-06-17 01:27:08 -07:00
math_ops.h
…
README.md
Fixed typos in documentation & READMEs (#10365)
2017-06-06 16:57:40 -07:00
sort_util.cc
Some refactoring groundwork towards an LHLO backend for XLA GPU; NFC
2020-06-25 20:22:18 -07:00
sort_util.h
Some refactoring groundwork towards an LHLO backend for XLA GPU; NFC
2020-06-25 20:22:18 -07:00
tuple_ops.cc
[XLA/GPU] Add a gold CHECK file for sorting. NFC.
2020-08-05 15:39:42 -07:00
tuple_ops.h
…

README.md

Common utilities and abstractions for handling and emitting LLVM IR for XLA backends.