experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
rei/fork-r2.3
STT-tensorflow/tensorflow/compiler/mlir/xla/tests
History
Tim Shen 3980537192 [MLIR] Remove TupleSelectOp from LHLO. 
LHLO uses output-parameter, but TupleSelectOp outputs into a tuple on the device. The current type constraints are wrong, and there is not enough expressiveness in LHLO to define a device-memory representation that can be passed in between kernel launches.

PiperOrigin-RevId: 317749542
Change-Id: I6d3350cbf9decf006f239a7208f6bbef0175ac61
2020-06-22 16:08:30 -07:00
..
hlo_to_lhlo_with_xla Remove --dump-input=fail flag passed into FileCheck in FileCheck tests. 2020-06-10 12:53:21 -07:00
translate Change HLO importer to set visibility when importing. 2020-06-15 16:37:44 -07:00
BUILD Use Bazel target names aligned with upstream library names 2020-06-10 19:57:02 -07:00
canonicalize.mlir Remove the canonicalize pattern for folding a pad op into the following conv op. 2020-06-19 07:14:26 -07:00
chlo_infer_shape_type_methods.mlir Integrate LLVM at https://github.com/llvm/llvm-project/commit/ec02635d104c 2020-06-12 17:34:02 -07:00
chlo_legalize_to_hlo_broadcasts.mlir Add shape constraints to CHLO->HLO lowering. 2020-06-15 07:54:29 -07:00
concatenate.mlir Add a new xla-opt binary tool to exercise only the XLA related passes 2020-04-13 20:08:11 -07:00
convert.mlir Integrate LLVM at https://github.com/llvm/llvm-project/commit/92cb0ce8f814 2020-06-08 09:38:15 -07:00
hlo-legalize-to-lhlo.mlir Extend lowering from hlo to lhlo to also support buffer allocation with escaping result buffers. This is now a flag to the pass (defaults to the current preallocation behavior). 2020-06-16 05:24:07 -07:00
hlo-legalize-to-linalg.mlir Remove special handling of reshape of unit-dims. 2020-06-12 13:15:55 -07:00
inlining.mlir Remove --dump-input=fail flag passed into FileCheck in FileCheck tests. 2020-06-10 12:53:21 -07:00
legalize_tanh_to_approximation.mlir Add lowering for TanhOp that uses an approximation instead of lowering to intrinsics. 2020-06-17 01:31:07 -07:00
legalize-control-flow.mlir Rename xla_hlo.conditional to xla_hlo.if. 2020-05-19 10:39:47 -07:00
legalize-tf-BatchMatMulV2.mlir Integrate LLVM at https://github.com/llvm/llvm-project/commit/ec02635d104c 2020-06-12 17:34:02 -07:00
legalize-tf-binary-elementwise.mlir Add shape constraints to CHLO->HLO lowering. 2020-06-15 07:54:29 -07:00
legalize-tf-control-flow.mlir Drop --dump-input-on-failure from FileCheck commands 2020-06-10 12:00:09 -07:00
legalize-tf-full-conversion.mlir Emit error messages for all missing legalizations in TF to XLA full legalization pass. 2020-05-07 10:31:53 -07:00
legalize-tf-with-tf2xla.mlir Fix bug in xla-legalize-tf-with-tf2xla pass by handling non-tensor operands 2020-06-18 16:06:15 -07:00
legalize-tf.mlir Legalize tf.CheckNumerics op as No-op in HLO 2020-06-17 23:34:01 -07:00
legalize-to-std.mlir Drop --dump-input-on-failure from FileCheck commands 2020-06-10 12:00:09 -07:00
lhlo_ops.mlir [MLIR] Remove TupleSelectOp from LHLO. 2020-06-22 16:08:30 -07:00
lhlo-copy-removal.mlir Drop --dump-input-on-failure from FileCheck commands 2020-06-10 12:00:09 -07:00
lhlo-fuse-linalg.mlir Support escaping result memrefs in lhlo_fuse_linalg. 2020-06-16 07:40:59 -07:00
lhlo-legalize-select-and-scatter.mlir Integrate LLVM at https://github.com/llvm/llvm-project/commit/9b72b47ed633 2020-06-16 05:43:03 -07:00
lhlo-legalize-to-affine.mlir [MLIR]Add conversions dot op from LHLO to Affine 2020-06-10 08:06:14 +00:00
lhlo-legalize-to-gpu.mlir Remove --dump-input=fail flag passed into FileCheck in FileCheck tests. 2020-06-10 12:53:21 -07:00
lhlo-legalize-to-linalg.mlir Remove special handling of reshape of unit-dims. 2020-06-12 13:15:55 -07:00
lhlo-legalize-to-llvm.mlir Drop --dump-input-on-failure from FileCheck commands 2020-06-10 12:00:09 -07:00
lhlo-legalize-to-parallel-loops.mlir Integrate LLVM at https://github.com/llvm/llvm-project/commit/9b72b47ed633 2020-06-16 05:43:03 -07:00
lower-complex.mlir Drop --dump-input-on-failure from FileCheck commands 2020-06-10 12:00:09 -07:00
lower-general-dot.mlir Add a new xla-opt binary tool to exercise only the XLA related passes 2020-04-13 20:08:11 -07:00
materialize-broadcasts.mlir Remove --dump-input=fail flag passed into FileCheck in FileCheck tests. 2020-06-10 12:53:21 -07:00
mlir_hlo_builder_test.cc Add MlirHloBuilder op implementations 2020-04-20 14:12:14 -07:00
ops.mlir Updated concatenate op for unranked tensors 2020-06-12 15:05:59 -07:00
reduce.mlir Add a new xla-opt binary tool to exercise only the XLA related passes 2020-04-13 20:08:11 -07:00
reshape.mlir Add a new xla-opt binary tool to exercise only the XLA related passes 2020-04-13 20:08:11 -07:00
reverse.mlir Add a new xla-opt binary tool to exercise only the XLA related passes 2020-04-13 20:08:11 -07:00
sink-constants-to-control-flow.mlir Remove --dump-input=fail flag passed into FileCheck in FileCheck tests. 2020-06-10 12:53:21 -07:00
transpose.mlir Remove --dump-input=fail flag passed into FileCheck in FileCheck tests. 2020-06-10 12:53:21 -07:00
tuple.mlir Add a new xla-opt binary tool to exercise only the XLA related passes 2020-04-13 20:08:11 -07:00
unfuse_batch_norm.mlir Integrate LLVM at https://github.com/llvm/llvm-project/commit/52cae05e087b 2020-06-10 21:39:41 -07:00
xla-hlo-fusion.mlir Drop more --dump-input-on-failure from FileCheck commands 2020-06-10 12:32:15 -07:00
xla-transform-unranked-hlo.mlir Integrate LLVM at https://github.com/llvm/llvm-project/commit/ec02635d104c 2020-06-12 17:34:02 -07:00