colin
/
mlir-hlo
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Commit Graph

  • 843af36e05 [MLIR] Add e2e test for unranked unary TF op, lowered and run with CPU runner. Alexander Belyaev 2020-08-09 02:36:32 -0700
  • 417e9e9c34 Integrate LLVM at llvm/llvm-project@38537307e5 Benjamin Kramer 2020-08-08 05:03:25 -0700
  • e6fa003bf2 Integrate LLVM at llvm/llvm-project@b6d9add71b A. Unique TensorFlower 2020-08-08 04:34:20 -0700
  • 53fdda7f3e Update mhlo.constant to use a custom assembly format instead of a custom printer and parser (NFC). Andy Ly 2020-08-07 21:33:04 -0700
  • 2b0a244a6b Add license header to lit.site.cfg.py.in Geoffrey Martin-Noble 2020-08-07 15:03:00 -0700
  • d742477c02 Verify that MHLO DynamicUpdateSlice start indices have matching element types. Lucy Fox 2020-08-07 10:46:06 -0700
  • 5f8da992f2 Fix mlir-hlo CMakefiles after internal change Mehdi Amini 2020-08-06 20:34:30 -0700
  • e00565b544 Integrate LLVM at llvm/llvm-project@9dbdaea9a0 A. Unique TensorFlower 2020-08-06 18:41:46 -0700
  • a6978cf4ab [MLIR][NFC] Adopt FuncOp/Region argument API's. Rahul Joshi 2020-08-06 18:31:33 -0700
  • cd22ecd136 Relax DynamicBroadcastInDim verifier when dimensions are dynamic. Lucy Fox 2020-08-06 17:29:29 -0700
  • c78e144f82 Integrate LLVM at llvm/llvm-project@633e3dacf2 A. Unique TensorFlower 2020-08-06 08:42:35 -0700
  • a68a16cdc7 [MLIR][XLA] Allow for choice of safe/unsafe variant in broadcast utils A. Unique TensorFlower 2020-08-06 07:46:30 -0700
  • bc3293a05f [MLIR] Remove Affine->STD and SCF->STD patterns from lhlo->llvm pass. Alexander Belyaev 2020-08-06 06:39:01 -0700
  • 701312720c Add CMake files and lit configurations, enough for `ninja check-mlir-hlo` to pass on all the tests Mehdi Amini 2020-08-05 23:27:48 -0700
  • c340367702 Add canonicalization for unpacking and repacking the same tuple (e.g. tuple -> get_tuple_element -> tuple). Andy Ly 2020-08-05 21:36:20 -0700
  • ad12e06ceb Integrate LLVM at llvm/llvm-project@acb66b9111 A. Unique TensorFlower 2020-08-05 16:29:14 -0700
  • 4372124362 [MLIR][XLA] Allow for choice of safe/unsafe variant in broadcast utils A. Unique TensorFlower 2020-08-05 12:42:49 -0700
  • 5d3cc2105e [MLIR][HLO] Remove redundant casts from unranked to ranked transformation A. Unique TensorFlower 2020-08-05 11:10:20 -0700
  • 37c36a4389 [MLIR][XLA] Allow for choice of safe/unsafe variant in broadcast utils A. Unique TensorFlower 2020-08-05 11:08:20 -0700
  • d26e863865 Integrate LLVM at llvm/llvm-project@c558c22cab Benjamin Kramer 2020-08-05 04:30:28 -0700
  • 6584c2ab1f Remove optional static registration for hlo dialects again. Stephan Herhut 2020-08-05 03:52:27 -0700
  • 9a97cffbb5 Integrate LLVM at llvm/llvm-project@28e322ea93 A. Unique TensorFlower 2020-08-04 22:30:46 -0700
  • 7bff244b8b Allow index typed memrefs in reshape_memref_cast. Stephan Herhut 2020-08-04 13:33:31 -0700
  • 66f023a057 Add build helper to mlir-hlo Mehdi Amini 2020-08-04 10:45:09 -0700
  • bf8836effb Integrate LLVM at llvm/llvm-project@b5059b7140 A. Unique TensorFlower 2020-08-03 23:04:41 -0700
  • 6c7d1a7c7e Internal change Mehdi Amini 2020-08-03 22:59:23 -0700
  • f044cf780c Introduce early documentation on mlir-hlo Mehdi Amini 2020-08-03 19:23:23 -0700
  • 4c8fead3e0 Add support for token operands to mhlo.tuple. Andy Ly 2020-08-03 10:17:46 -0700
  • 3fe9a7d2db Legalize TensorFlow NonMaxSuppressionV4 and SelfAdjointEigV2Op ops to HLO Smit Hinsu 2020-07-31 23:10:33 -0700
  • 7809320a5e Integrate LLVM at llvm/llvm-project@b7cfa6ca92 A. Unique TensorFlower 2020-07-31 18:23:00 -0700
  • 577a81a66d Sink standard dialect constants in sink_constants_to_control_flow pass Smit Hinsu 2020-07-31 16:01:37 -0700
  • 1c535f1718 Restrict GetDimensionSize HLO op result type to 32 bit integer Smit Hinsu 2020-07-31 14:37:20 -0700
  • fcb91fb0b9 [MLIR] Add conversion between XLA Fusion and MHLO FusionOp. Tim Shen 2020-07-31 13:58:01 -0700
  • 9cbe5f2285 Constrain mhlo.const to static shaped tensors. Stephan Herhut 2020-07-31 11:50:22 -0700
  • 734b9b25fd Add optional static registration for mhlo/lmhlo passes. Stephan Herhut 2020-07-31 05:05:18 -0700
  • 735ae2838c Integrate LLVM at llvm/llvm-project@cd4e8d7f6f Thomas Joerg 2020-07-31 02:08:18 -0700
  • d83d437588 Add chlo.acos and legalization Jacques Pienaar 2020-07-30 12:07:36 -0700
  • cce4bddf4b Remove unnecessary conversions between Shape and ExtentTensor. Tres Popp 2020-07-30 04:59:33 -0700
  • ffef8d6593 Support CHLO->LHLO lowering for broadcasting operations with both inputs unranked. Tres Popp 2020-07-30 01:56:40 -0700
  • b09bf2a4dc Move hlo end to end tests to the hlo directory tree. Stephan Herhut 2020-07-30 01:04:37 -0700
  • de5ddaf7c9 HLO Random operations should match shape constraints. Robert Suderman 2020-07-29 12:31:15 -0700
  • 1b0eb4baa7 Do not mandate the result type of shape computations but have it be inferred from context. Stephan Herhut 2020-07-29 07:35:52 -0700
  • cd01bb4c4e More cleanup in mlir-hlo to prepare for the standalone build Mehdi Amini 2020-07-28 16:12:08 -0700
  • effd3fb4f9 Extend unranked to ranked pattern for hlo operations to all unary and binary ops. Stephan Herhut 2020-07-28 00:55:58 -0700
  • b7c4314e7f Allow running tests with FILECHECK_OPTS=-enable-var-scope Benjamin Kramer 2020-07-27 12:49:58 -0700
  • 739758f9cc Integrate LLVM at llvm/llvm-project@eed333149d Thomas Joerg 2020-07-27 07:13:38 -0700
  • 8023baa959 Modified HLOAbsOp lowering for differing types. Robert Suderman 2020-07-24 15:17:48 -0700
  • 882468da13 PR #41662: NFC - minor spelling tweaks Kazuaki Ishizaki 2020-07-24 01:30:44 -0700
  • 8f262ae8f5 Add support for lowering mhlo.iota to Linalg. Hanhan Wang 2020-07-23 16:18:01 +0000
  • 4251630426 Support CHLO broadcasting operations between scalar and unranked tensors. Tres Popp 2020-07-22 12:25:26 +0000
  • 63d62b7952 Change cast to dyn_cast in hlo::ReshapeOp's verification. Tres Popp 2020-07-21 09:06:21 +0000
  • c23ad602c8 Add a transform for Gathers to torch_index_select. Robert Suderman 2020-07-20 23:58:31 +0000
  • cc776071fe Fix namespace for complex lowerings Robert Suderman 2020-07-20 17:58:56 +0000
  • 30fa3db949 Update comments to reflect the new names of `GenericOp` constructor parameters A. Unique TensorFlower 2020-07-17 16:43:03 +0000
  • c44e08351d Fix mhlo to lmhlo conversion for ReduceOp. Stephan Herhut 2020-07-16 11:40:32 +0000
  • c8bb0ff54d Integrate LLVM at https://github.com/llvm/llvm-project/commit/f233b92f92a6 A. Unique TensorFlower 2020-07-15 22:48:16 +0000
  • 98a1e3b108 Add an optimization that converts some Gathers to Slices. Robert Suderman 2020-07-15 17:56:59 +0000
  • 7a6adc6a84 Add canonicalization patterns for dynamic_broadcast_in_dim where the target shape is the shape of the operand. Stephan Herhut 2020-07-15 07:37:58 +0000
  • 86f290896d Implement lowering of lmhlo.reshape_memref_cast to LLVM for unknown length shape operand. Stephan Herhut 2020-07-13 15:26:53 +0000
  • d166b66cba PR #40925: [MLIR] Update lhlo.const to linalg lowering to use affine.store inste… Uday Bondhugula 2020-07-10 17:03:44 +0000
  • 6eaccefdab [MLIR][LHLO] Lower ReshapeMemRefCastOp to LLVM. Alexander Belyaev 2020-07-10 09:55:17 +0000
  • 06ae59074f Fold xla iota across a 1-length dimension into a zero value Robert Suderman 2020-07-09 18:58:28 +0000
  • c3be2474dd Integrate LLVM at https://github.com/llvm/llvm-project/commit/305b500eaf8c Stephan Herhut 2020-07-09 14:54:46 +0000
  • de0578b4f9 PR #40745: [MLIR] Add constant folder for xla_hlo.broadcast_in_dim op Uday Bondhugula 2020-07-09 10:36:37 +0000
  • 506ddd9c4a Cleanup build rule names in compiler/mlir/hlo to remove the redundant/obsolete xla_ prefix Mehdi Amini 2020-07-09 03:32:16 +0000
  • f4303855c4 Add mlir_cpu_runner tests infra for CHLO->LHLO->LLVM lowering. Alexander Belyaev 2020-07-08 17:42:56 +0000
  • a575636862 Rename XlaHloDialect class into MhloDialect following the recent dialect namespace renaming Mehdi Amini 2020-07-08 17:19:13 +0000
  • 94dcb90d38 Rename xla_chlo dialect into chlo Mehdi Amini 2020-07-08 17:12:48 +0000
  • 7c4a5d62b5 Rename xla_lhlo dialect into lmhlo Mehdi Amini 2020-07-08 17:05:32 +0000
  • b076e018a8 [MLIR][LHLO] Legalize CallOp that call funcs with tensor args/results. Alexander Belyaev 2020-07-08 12:59:45 +0000
  • e8cfdee592 [MLIR][LHLO] Convert mhlo.dynamic_reshape -> lhlo.reshape_memref_cast. Alexander Belyaev 2020-07-08 09:11:30 +0000
  • 8692fde3f9 [MLIR] Convert FuncOp signature with unranked types in HLO->LHLO conversion. Alexander Belyaev 2020-07-08 08:43:30 +0000
  • e1651b6090 Canonicalize multidimensional iota to use broadcast Robert Suderman 2020-07-08 00:35:27 +0000
  • 8900222fed Rename `xla_hlo` dialect to `mhlo` Mehdi Amini 2020-07-07 04:51:24 +0000
  • fa057cc0bc Move HLO tests to `third_party/tensorflow/compiler/mlir/hlo/` Mehdi Amini 2020-07-06 23:28:26 +0000
  • 31dc1b21eb Move XLA-independent transforms to the new MLIR-HLO directory Mehdi Amini 2020-07-06 20:57:00 +0000
  • 72010faaa7 [MLIR][LHLO] Add ReshapeMemrefCastOp to LHLO. Alexander Belyaev 2020-07-06 16:56:07 +0000
  • fcf3df1541 Move the HLO/LHLO dialects to a new directory: tensorflow/compiler/mlir/hlo Mehdi Amini 2020-07-01 19:18:52 +0000
  • 5fe5c39ccc Initial Revision Mehdi Amini 2020-07-29 22:52:21 +0000