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

Commit Graph

  • 8d051723c0 Use InferTypeOpInterface for HLO AbsOp and fix result shape inference Smit Hinsu 2020-12-17 17:36:05 -0800
  • 43ede42ce1 Integrate LLVM at llvm/llvm-project@e113317958 A. Unique TensorFlower 2020-12-17 16:34:05 -0800
  • 737d15ded5 Handle operands with zero elements in HLO PadOp folder Smit Hinsu 2020-12-17 09:25:26 -0800
  • 8134bff98d [XLA:GPU] Add layout attributes to LHLO_GPU Convolution operations. Rahul Joshi 2020-12-17 09:25:03 -0800
  • dc7e63f74c Use mlir::OpState::operator->() to get to methods of mlir::Operation. Christian Sigg 2020-12-17 06:28:06 -0800
  • 1f244c3e2c Fix SignOp lowering for floating point values. Adrian Kuegel 2020-12-17 01:44:54 -0800
  • 099c130daf Fix MLIR include paths. Christian Sigg 2020-12-17 00:55:03 -0800
  • 5da9190dd9 Integrate LLVM at llvm/llvm-project@0cf7e4b252 A. Unique TensorFlower 2020-12-16 20:29:15 -0800
  • 5a080ad795 Add Bazel build configuration to MLIR HLO standalone repo. Phoenix Meadowlark 2020-12-16 16:22:14 -0800
  • 65222893ae [KERNEL_GEN] Convert LHLO AddOp, SubOp (ComplexType) to complex ops. Alexander Belyaev 2020-12-16 05:44:38 -0800
  • 61244b136c Try to avoid a segfault if we don't support a lowering. Adrian Kuegel 2020-12-16 04:57:35 -0800
  • e6e8920921 [KERNEL_GEN] Switch the pipeline to Linalg-on-Tensors. Alexander Belyaev 2020-12-16 01:50:12 -0800
  • 5ab94a00a7 [mlir][NFC] Replace usages or mlir/IR/StandardTypes.h with mlir/IR/BuiltinTypes.h River Riddle 2020-12-15 21:27:33 -0800
  • ccbf39842f Integrate LLVM at llvm/llvm-project@7aeb3804c4 A. Unique TensorFlower 2020-12-15 17:39:43 -0800
  • 6d7812bbc5 Correct non ConversionRewriter transformation in LegalizeToLinalg Tres Popp 2020-12-15 08:56:01 -0800
  • f0c2695d31 Fix SignOp lowering for floating point values. A. Unique TensorFlower 2020-12-15 06:49:01 -0800
  • ddda2699fb [KERNEL_GEN] Switch the pipeline to Linalg-on-Tensors. Alexander Belyaev 2020-12-15 06:31:28 -0800
  • 79fa36bcbc Fix SignOp lowering for floating point values. Adrian Kuegel 2020-12-15 05:12:08 -0800
  • 9540e51617 [mlir][NFC] Replace usages or mlir/IR/StandardTypes.h with mlir/IR/BuiltinTypes.h River Riddle 2020-12-15 00:58:42 -0800
  • 9df327d88f Forward listeners in LhloLegalizeToParallelLoops builders Tres Popp 2020-12-15 00:12:32 -0800
  • 1a58f19664 [NFC] Make function names follow style guide. Hanhan Wang 2020-12-14 10:46:04 -0800
  • 8b35a75d4a [KERNEL_GEN] Switch the pipeline to Linalg-on-Tensors. Alexander Belyaev 2020-12-14 05:46:09 -0800
  • 6b439f7eee [mlir][NFC] Replace usages or mlir/IR/StandardTypes.h with mlir/IR/BuiltinTypes.h River Riddle 2020-12-11 19:00:36 -0800
  • 3abdd556de [mlir][NFC] Replace usages or mlir/IR/StandardTypes.h with mlir/IR/BuiltinTypes.h River Riddle 2020-12-11 18:12:14 -0800
  • 22b91d62ff [XLA/GPU] Migrate all unnested elementwise emitters. Tim Shen 2020-12-11 12:44:38 -0800
  • 21bb9e8c9a Integrate LLVM at llvm/llvm-project@38d32e4fd7 A. Unique TensorFlower 2020-12-11 07:27:49 -0800
  • 9930c20c31 [mlir][hlo] Fix lowering of NE comparison. It should return true if either side is NaN Benjamin Kramer 2020-12-11 06:45:22 -0800
  • ab6ee11813 Fix folding of HLO SliceOp with zero elements Smit Hinsu 2020-12-10 20:21:49 -0800
  • f232da1f9d [MLIR:HLO] Add window_reversal attribute to convolution attributes. - Add this attribute to match the corresponding XLA HLO attribute on convolution operations. - A true value indicates a reversal of the corresponding kernel spatial dimension. - Since XLA builder does not support this attribute, use a custom HLO converted to map from mlir::mhlo::ConvOp to XLA. Rahul Joshi 2020-12-10 16:38:26 -0800
  • 3442ac270d Integrate LLVM at llvm/llvm-project@997a719d5a A. Unique TensorFlower 2020-12-10 11:57:30 -0800
  • 6a05893169 [XLA/GPU] Migrate all unnested elementwise emitters. A. Unique TensorFlower 2020-12-10 00:21:01 -0800
  • 51f535454d Use OpState::operator->() to get to member functions in Operation so we can remove the corresponding methods from OpState. Christian Sigg 2020-12-09 23:53:05 -0800
  • c36afd275e [HLO] Add a pattern for HLO ConstOp to HLO -> Linalg conversion. Alexander Belyaev 2020-12-09 23:24:23 -0800
  • cfcf741932 [XLA/GPU] Migrate all unnested elementwise emitters. Tim Shen 2020-12-09 23:07:25 -0800
  • 65ebd85563 [XLA/GPU] Migrate all unnested elementwise emitters. A. Unique TensorFlower 2020-12-09 16:04:43 -0800
  • 1c10e1fec6 [XLA/GPU] Migrate all unnested elementwise emitters. Tim Shen 2020-12-09 13:31:16 -0800
  • afea1e1897 Integrate LLVM at llvm/llvm-project@6883042528 A. Unique TensorFlower 2020-12-09 12:05:36 -0800
  • 1a5a1b5f41 [XLA/GPU] Migrate all unnested elementwise emitters. A. Unique TensorFlower 2020-12-09 09:53:38 -0800
  • 3c33fe4b9e [XLA/GPU] Migrate all unnested elementwise emitters. Tim Shen 2020-12-09 08:42:11 -0800
  • 6cb9cc53ec Integrate LLVM at llvm/llvm-project@d553243fe4 A. Unique TensorFlower 2020-12-08 13:08:24 -0800
  • e9057cbd93 [mlir] Enable conversion of HLO SignOp to Linalg. Alexander Belyaev 2020-12-08 12:07:39 -0800
  • 4e6367e9d6 Integrate LLVM at llvm/llvm-project@a1344779ab A. Unique TensorFlower 2020-12-08 06:46:02 -0800
  • c3790af758 Add plumbing for or and xor to hlo to lhlo and linalg lowerings. Stephan Herhut 2020-12-08 06:38:26 -0800
  • dd5895d083 Extend unranked hlo transformations to also support and, or and xor. Stephan Herhut 2020-12-08 00:59:43 -0800
  • 812221db97 Integrate LLVM at llvm/llvm-project@2ac4d0f45a A. Unique TensorFlower 2020-12-07 16:20:12 -0800
  • 5235eceea0 Lower mhlo shifts to linalg Benjamin Kramer 2020-12-07 13:01:25 -0800
  • eaa21130e8 Integrate LLVM at llvm/llvm-project@ecaff13fc0 A. Unique TensorFlower 2020-12-07 06:59:56 -0800
  • d327fc5737 [kernel_gen] Lower max rank specialization from 6 to 5 Tres Popp 2020-12-07 02:17:57 -0800
  • 4c0c9ab119 Integrate LLVM at llvm/llvm-project@db226cdf4c A. Unique TensorFlower 2020-12-07 02:12:28 -0800
  • bc7b6374c8 Fix handling of negative seeds in random number generator op kernels for XLA Smit Hinsu 2020-12-05 18:54:37 -0800
  • 55268f9ee8 Integrate LLVM at llvm/llvm-project@7f6f9f4cf9 A. Unique TensorFlower 2020-12-04 13:32:48 -0800
  • c33bdcbd03 Remove fold of `mhlo.compare(%arg0, %arg0)` for floating types. Phoenix Meadowlark 2020-12-04 12:14:29 -0800
  • 9bd1995f90 Legalize XlaReplicaId to HLO replica-id op Smit Hinsu 2020-12-04 11:04:02 -0800
  • e48881af81 [MLIR:LHLO_GPU] Add fused convolution operation without any side inputs. Rahul Joshi 2020-12-04 10:08:37 -0800
  • 3691e39f62 Integrate LLVM at llvm/llvm-project@f5d52916ce A. Unique TensorFlower 2020-12-04 07:23:19 -0800
  • e87d53742b Fix handling of negative seeds in random number generator op kernels for XLA A. Unique TensorFlower 2020-12-04 00:03:15 -0800
  • 9456af5880 Fix handling of negative seeds in random number generator op kernels for XLA Smit Hinsu 2020-12-03 22:09:03 -0800
  • a7506fab4f Integrate LLVM at llvm/llvm-project@55db6ec1cc A. Unique TensorFlower 2020-12-03 16:54:20 -0800
  • 73cb134cca [MLIR] Add XLA HLO -> LMHLO conversion for all elementwise ops. Tim Shen 2020-12-03 16:11:22 -0800
  • 4b6a38da54 Integrate LLVM at llvm/llvm-project@1860331932 A. Unique TensorFlower 2020-12-03 12:44:00 -0800
  • 7c3f049c8e [kernel_gen] Lower max rank specialization from 6 to 5 Tres Popp 2020-12-03 09:18:39 -0800
  • d2e3797d7d Integrate LLVM at llvm/llvm-project@6627a3c287 A. Unique TensorFlower 2020-12-03 05:09:10 -0800
  • 1c6c04f76b Integrate LLVM at llvm/llvm-project@baa005c96c A. Unique TensorFlower 2020-12-02 15:40:41 -0800
  • 446e25ed7d Integrate LLVM at llvm/llvm-project@b40b3196b3 A. Unique TensorFlower 2020-12-02 12:24:02 -0800
  • dbbdfea95b [MLIR:HLO] Generate enum decls for HLO and LHLO GPU dialects. Rahul Joshi 2020-12-02 11:38:26 -0800
  • d7bd5233ab [XLA:GPU] Migrate GEMM Thunk emission to MLIR. Rahul Joshi 2020-12-02 09:42:26 -0800
  • 1b711670bc Fix handling of negative seeds in random number generator op kernels for XLA A. Unique TensorFlower 2020-12-02 08:40:44 -0800
  • 733fc6d032 Fix handling of negative seeds in random number generator op kernels for XLA Smit Hinsu 2020-12-02 07:22:51 -0800
  • 2f6ced90fc Integrate LLVM at llvm/llvm-project@c266c56d54 A. Unique TensorFlower 2020-12-02 06:40:30 -0800
  • 27b10932cb Integrate LLVM at llvm/llvm-project@aafb366210 A. Unique TensorFlower 2020-12-01 16:09:57 -0800
  • f89244381d [mlir][NFC] Replace usages of Function.h and Module.h with BuiltinOps.h River Riddle 2020-12-01 13:17:12 -0800
  • eb9015a2b0 Integrate LLVM at llvm/llvm-project@bb993b1d9d A. Unique TensorFlower 2020-12-01 13:08:13 -0800
  • 7138157ed9 Integrate LLVM at llvm/llvm-project@523775f967 A. Unique TensorFlower 2020-12-01 06:45:51 -0800
  • fd286b11f1 Integrate LLVM at llvm/llvm-project@40659cd2c6 A. Unique TensorFlower 2020-12-01 04:41:24 -0800
  • 00ab6722ac Integrate LLVM at llvm/llvm-project@d928dfc6f9 A. Unique TensorFlower 2020-11-30 21:44:31 -0800
  • 37b1110ae0 Integrate LLVM at llvm/llvm-project@8cdf4920c4 A. Unique TensorFlower 2020-11-30 19:05:44 -0800
  • 3f815138b6 Integrate LLVM at llvm/llvm-project@750049d78b A. Unique TensorFlower 2020-11-30 14:31:27 -0800
  • 5350b8080d [NFC] Add default value for attributes in lmhlo::FusionOp custom builder. Rahul Joshi 2020-11-30 11:46:37 -0800
  • a4a3e9b73c Integrate LLVM at llvm/llvm-project@234a5297aa A. Unique TensorFlower 2020-11-30 09:16:49 -0800
  • dd15c6cd84 [MLIR][KernelGen] Generate assertion message in `transform_unranked_hlo` pass A. Unique TensorFlower 2020-11-30 01:40:08 -0800
  • 1fe04c0aff Integrate LLVM at llvm/llvm-project@564628014c Benjamin Kramer 2020-11-27 10:04:13 -0800
  • d14c63da54 Add a canonicalization pattern to remove redundant dynamic_reshapes. Adrian Kuegel 2020-11-27 04:46:02 -0800
  • f183e95e69 Register the new branch op rewrite patterns in hlo to lhlo conversion for good measure. Stephan Herhut 2020-11-27 03:43:21 -0800
  • d617f3edb6 Integrate LLVM at llvm/llvm-project@b33fbbaa34 Benjamin Kramer 2020-11-27 03:27:14 -0800
  • df244927d3 Integrate LLVM at llvm/llvm-project@20c926e079 Benjamin Kramer 2020-11-26 10:11:24 -0800
  • 6a71a84302 Support different input/output type for TransformUnrankedHlo. Adrian Kuegel 2020-11-26 04:19:51 -0800
  • 1b98bf5fab Integrate LLVM at llvm/llvm-project@54ec9bb551 Benjamin Kramer 2020-11-26 04:05:21 -0800
  • 651afab874 Integrate LLVM at llvm/llvm-project@dc35368ccf A. Unique TensorFlower 2020-11-25 16:15:35 -0800
  • 5583c63cab [KERNEL_GEN] Add unranked Conj kernel. Alexander Belyaev 2020-11-25 06:36:42 -0800
  • a6948f6b41 Integrate LLVM at llvm/llvm-project@a38d13ed36 Benjamin Kramer 2020-11-25 05:33:02 -0800
  • 42345f41f7 Integrate LLVM at llvm/llvm-project@e0f4dea0d0 A. Unique TensorFlower 2020-11-25 00:26:18 -0800
  • e970a97efb Integrate LLVM at llvm/llvm-project@42eaf4fe0a A. Unique TensorFlower 2020-11-24 17:11:57 -0800
  • 3e01448481 [MLIR] Move some walk()ing functions to the lmhlo::FusionOp API. Tim Shen 2020-11-24 12:21:49 -0800
  • 85f92a1651 [KernelGen] Lower tf.Erf and tf.Erfc ops to CHLO. Lucy Fox 2020-11-24 10:55:01 -0800
  • 9f20ef0581 Integrate LLVM at llvm/llvm-project@5ce85e6635 Benjamin Kramer 2020-11-24 09:52:54 -0800
  • 80d7342ae3 Integrate LLVM at llvm/llvm-project@fd92c5dbee Benjamin Kramer 2020-11-24 05:13:46 -0800
  • b016b5a219 Fix constant folding of mhlo.convert op with i1 element types Smit Hinsu 2020-11-23 14:17:47 -0800
  • 2ac41d8cd2 Integrate LLVM at llvm/llvm-project@76bd4444e3 Benjamin Kramer 2020-11-23 10:17:24 -0800
  • ac54c5ccfa [XLA:GPU] Convert Cholesky custom call in XLA HLO to LHLO GPU Dialect. Rahul Joshi 2020-11-23 10:05:34 -0800
  • aa4d33149a Integrate LLVM at llvm/llvm-project@3c696a212b Benjamin Kramer 2020-11-23 04:55:06 -0800