Commit Graph

Select branches

Hide Pull Requests

master

Mono Color

• 8ace7f0d8e refine basic test master colin.liang 2021-08-18 14:17:48 +0800
• 84e7697c6a refine test and tensor colin.liang 2021-08-17 11:28:17 +0800
• 975a47f7b2 modify build cc code,add AddOp in test colin.liang 2021-08-16 15:35:37 +0800
• 4a9b201c4e add more types and try to build pass colin.liang 2021-08-13 15:05:10 +0800
• e88366b851 add basic test colin.liang 2021-08-11 15:41:34 +0800
• 95fc37ffa8 add integer attribute convert colin.liang 2021-08-11 10:46:07 +0800
• 9d5166684c modify code gen of build function colin.liang 2021-08-04 20:24:07 +0800
• 898eb732de add mlir tblgen builder colin.liang 2021-07-23 11:38:34 +0800
• b0dd7a7518 add mul in test colin.liang 2021-07-15 16:26:17 +0800
• e62fa9311f add tensor from host to jit runner,and get output colin.liang 2021-07-13 20:10:52 +0800
• 62e7b883c7 add cpu runner in llvm colin.liang 2021-07-12 21:03:29 +0800
• 20ff8b4c93 Integrate LLVM at llvm/llvm-project@b3634d3e88 A. Unique TensorFlower 2021-06-21 17:21:35 -0700
• 8ff5f9d94a Integrate LLVM at llvm/llvm-project@186f2ac612 A. Unique TensorFlower 2021-06-21 13:07:29 -0700
• 03d2cb606d [mhlo] Make sure reifyResultTypes returns a tensor of index Benjamin Kramer 2021-06-21 10:41:29 -0700
• a6b8882739 Integrate LLVM at llvm/llvm-project@b650778dc4 A. Unique TensorFlower 2021-06-21 06:39:17 -0700
• 0162057815 Integrate LLVM at llvm/llvm-project@134723edd5 A. Unique TensorFlower 2021-06-21 03:36:41 -0700
• 785d7042f0 Integrate LLVM at llvm/llvm-project@c9889c44ec A. Unique TensorFlower 2021-06-21 00:47:18 -0700
• 05ad9ddf84 Integrate LLVM at llvm/llvm-project@93183a41b9 A. Unique TensorFlower 2021-06-18 06:02:37 -0700
• 4c282fb542 Avoid Broadcast op if all shapes are (known to) be equal. Adrian Kuegel 2021-06-18 00:21:34 -0700
• d4a7901284 Integrate LLVM at llvm/llvm-project@366df11a35 A. Unique TensorFlower 2021-06-17 17:28:04 -0700
• da6593e960 PR #50073: [MLIR] Add GatherOp lowering from lmhlo to Affine. Abhishek Varma 2021-06-17 14:54:39 -0700
• 2ab16024cf Add tests for lowering HLO_ExpOp for complex types to Linalg. Adrian Kuegel 2021-06-17 06:33:13 -0700
• 470ac45f45 [MLIR][HLO] Remove unused pass `TransformUnrankedHloPass` A. Unique TensorFlower 2021-06-17 05:19:54 -0700
• 10634ca3a6 Integrate LLVM at llvm/llvm-project@854ef875b9 A. Unique TensorFlower 2021-06-17 04:06:57 -0700
• 9f47ff607b [MLIR][KernelGen] Better rank specialization for clusters containing `mhlo.select` A. Unique TensorFlower 2021-06-17 04:05:09 -0700
• 376da8592f Add MLIR generated SignOp GPU kernel for complex types. Adrian Kuegel 2021-06-17 03:55:59 -0700
• 73ed8cbf82 Add MLIR generated NegOp GPU kernel for complex types. Adrian Kuegel 2021-06-17 01:30:08 -0700
• 8c8e81cb69 Fix pass definition to inherit from the TableGen generated base class (NFC) Mehdi Amini 2021-06-16 19:04:23 -0700
• 2e08c246e9 Integrate LLVM at llvm/llvm-project@7fff39d9b0 A. Unique TensorFlower 2021-06-16 15:17:21 -0700
• 88cc0c6c46 PR #50271: [MLIR][DISC] Bufferize GatherOp and DynamicGatherOp Wenyi Zhao 2021-06-16 13:44:21 -0700
• 34dc5f2a79 PR #50020: [MLIR][DISC] support fusion on buffer Wenyi Zhao 2021-06-16 09:50:41 -0700
• 82696f8598 [MLIR][HLO] Annotate `mhlo.clamp` and `mhlo.select` as element-wise broadcasting A. Unique TensorFlower 2021-06-16 07:58:09 -0700
• a65cf627c4 Integrate LLVM at llvm/llvm-project@662f9bff33 A. Unique TensorFlower 2021-06-16 05:33:25 -0700
• 06ce0852d1 Integrate LLVM at llvm/llvm-project@cc8d32ae7d A. Unique TensorFlower 2021-06-16 02:37:51 -0700
• b44ab8ad49 Add support for lowering DataMovementOp ops to Linalg on unsigned types. Hanhan Wang 2021-06-15 10:57:15 -0700
• 3afbe312f8 PR #49919: [MLIR][DISC] pattern conversion from tf2mhlo: ConvertUnpackOpDynamic, ConvertSignOpDynamic, ConvertSigmoidGradOpDynamic Feiwen 2021-06-15 10:32:32 -0700
• 5fbdac34a9 [XLA:GPU] Add AllReduce{Start,Done} to MLIR LHLO dialect. Chris Jones 2021-06-15 03:54:24 -0700
• 399dae666d Add MLIR generated ExpOp GPU kernel for complex types. Adrian Kuegel 2021-06-15 00:44:45 -0700
• f9f7a63870 Add missing dep on RAL pass generation Geoffrey Martin-Noble 2021-06-14 17:01:34 -0700
• 6f5a440031 Integrate LLVM at llvm/llvm-project@56ae4f23b2 A. Unique TensorFlower 2021-06-14 14:16:55 -0700
• 7f94bd923b PR #50236: [MLIR][DISC] Bufferize TransposeOp and ConcatenateOp Wenyi Zhao 2021-06-14 12:35:47 -0700
• 23ebbb28d1 PR #50191: [MLIR][DISC] Add RAL (Runtime abstraction layer) Dialect Wenyi Zhao 2021-06-14 11:26:41 -0700
• a6011d0279 [HLO] Add AllReduceScatter to MHLO and LMHLO dialects. Rahul Joshi 2021-06-14 09:36:23 -0700
• dbfa4b1537 Integrate LLVM at llvm/llvm-project@b90f9bea96 A. Unique TensorFlower 2021-06-14 04:13:14 -0700
• 07c92f0ad8 Integrate LLVM at llvm/llvm-project@e0b469ffa1 A. Unique TensorFlower 2021-06-11 19:33:55 -0700
• 89faaa6575 Integrate LLVM at llvm/llvm-project@82a3b606b0 A. Unique TensorFlower 2021-06-11 16:49:50 -0700
• 8388303fd2 PR #50211: [MLIR][DISC] Bufferize RealDynamicSliceOp and ReduceOp Wenyi Zhao 2021-06-11 16:31:53 -0700
• 95ba03534f Allow variadic operands/result in MHLO while Jacques Pienaar 2021-06-11 13:07:19 -0700
• 33f95eecc7 Integrate LLVM at llvm/llvm-project@f3f904563e A. Unique TensorFlower 2021-06-11 08:46:33 -0700
• bd5752f0bf [MLIR][HLO] Find shape equivalences and use them for better rank specialization A. Unique TensorFlower 2021-06-11 03:59:02 -0700
• 5cca8a14e3 Integrate LLVM at llvm/llvm-project@4f6ec382c8 A. Unique TensorFlower 2021-06-10 22:52:43 -0700
• ad7bc780b9 Integrate LLVM at llvm/llvm-project@ff81a2c95d A. Unique TensorFlower 2021-06-10 15:13:18 -0700
• 6660234d80 PR #50100: [MLIR][DISC] Bufferize DynamicIotaOp and DynamicPadOp Wenyi Zhao 2021-06-10 14:19:55 -0700
• 642ca86a3f Integrate LLVM at llvm/llvm-project@ad6a84f82c A. Unique TensorFlower 2021-06-10 11:05:55 -0700
• 14093b7906 [XLA:GPU] Add AllReduce{Start,Done} to MLIR LHLO dialect. A. Unique TensorFlower 2021-06-10 10:26:35 -0700
• 968226b9d7 [XLA:GPU] Add AllReduce{Start,Done} to MLIR LHLO dialect. Chris Jones 2021-06-10 06:53:58 -0700
• 6088eb697c Fix Cosh approximation for F16. Adrian Kuegel 2021-06-10 06:16:00 -0700
• 837a1de7c5 Integrate LLVM at llvm/llvm-project@e11b5b87be A. Unique TensorFlower 2021-06-10 00:16:46 -0700
• 9f67417b41 [MLIR][HLO] Avoid duplicate cluster operands when merging A. Unique TensorFlower 2021-06-09 10:53:51 -0700
• b580722041 [MLIR][KernelGen] Merge rank specialization clusters A. Unique TensorFlower 2021-06-09 10:06:47 -0700
• b6d8160611 Add Broadcasting and BroadcastingElementwise traits to ConstantLikeOp. Adrian Kuegel 2021-06-09 05:08:36 -0700
• b9e45007d5 [MLIR][HLO] Extend broadcast propagation pass to enable more fusion A. Unique TensorFlower 2021-06-09 03:02:17 -0700
• d828b457b3 Handle empty tensors in SimplifyConcatSlice. A. Unique TensorFlower 2021-06-09 02:14:22 -0700
• 1770ed455f Remove unnecessary duplicated source from "lhlo" (NFC) Mehdi Amini 2021-06-08 18:06:26 -0700
• 4134923d4f Integrate LLVM at llvm/llvm-project@f96b5e801d A. Unique TensorFlower 2021-06-08 06:23:20 -0700
• 9a8c254526 Support complex types for Sinh. Adrian Kuegel 2021-06-08 04:22:15 -0700
• c47869f931 [MLIR][HLO] Rename `move-up-dynamic-broadcasts-for-fusion` to `broadcast-propagation` A. Unique TensorFlower 2021-06-08 01:50:02 -0700
• b2839c735b Integrate LLVM at llvm/llvm-project@8344e215ec A. Unique TensorFlower 2021-06-07 17:31:44 -0700
• c11de49300 Integrate LLVM at llvm/llvm-project@7ed7d4ccb8 A. Unique TensorFlower 2021-06-07 12:03:15 -0700
• d1c60df2fe [MHLO:linalg] Be more aggressive about turning mhlo.const into std.constant Benjamin Kramer 2021-06-07 11:56:55 -0700
• 25b93c8d66 Add support for lowering mhlo.iota/dynamic_iota to Linalg on unsigned types. Hanhan Wang 2021-06-07 10:58:21 -0700
• 5315997402 Fix Sinh approximation for F16. Adrian Kuegel 2021-06-07 06:37:20 -0700
• fc723380e6 Update lhlo to use the new structured op interface. Tobias Gysi 2021-06-07 03:10:05 -0700
• ade873a5e0 PR #49970: [MLIR][DISC] bufferize DynamicReshape and DynamicBroadcastInDim Wenyi Zhao 2021-06-04 15:35:08 -0700
• 8b3a75ea25 Integrate LLVM at llvm/llvm-project@b109172d99 A. Unique TensorFlower 2021-06-04 11:10:10 -0700
• 9c895af2f1 Integrate LLVM at llvm/llvm-project@23a116c8c4 A. Unique TensorFlower 2021-06-04 06:44:25 -0700
• f1f4c903df Integrate LLVM at llvm/llvm-project@fcf8827a98 A. Unique TensorFlower 2021-06-04 04:29:47 -0700
• db05388a3c Integrate LLVM at llvm/llvm-project@da3ed58b97 A. Unique TensorFlower 2021-06-03 20:44:08 -0700
• aba16adfa5 Add `mhlo.all_gather` op to MHLO dialect. A. Unique TensorFlower 2021-06-03 10:44:36 -0700
• 4fc2e87a42 Add mhlo python binding generator target Jacques Pienaar 2021-06-03 10:25:38 -0700
• fe42a08fc9 Use channel_handle for ChannelHandles in MHLO ops. This makes the naming of these properties consistent across these ops. A. Unique TensorFlower 2021-06-03 09:48:47 -0700
• 063086dd78 Integrate LLVM at llvm/llvm-project@c89dff5855 A. Unique TensorFlower 2021-06-02 19:35:13 -0700
• 4620410f18 Integrate LLVM at llvm/llvm-project@b25546a4b4 A. Unique TensorFlower 2021-06-02 09:31:57 -0700
• 75a1c450ea [MLIR][KernelGen] Fix Windows build failure A. Unique TensorFlower 2021-06-02 05:33:45 -0700
• 557e56362e [MLIR][KernelGen] Simplify rank specialization tests with smaller target rank A. Unique TensorFlower 2021-06-02 03:47:08 -0700
• 968d4b8709 PR #49598: [MLIR][DISC] legalize tensor_load inserted during hlo-to-lhlo conversion wyzhao 2021-06-01 16:27:07 -0700
• 5baf6e7709 Integrate LLVM at llvm/llvm-project@97d234935f A. Unique TensorFlower 2021-06-01 08:36:58 -0700
• d1828625ab [MLIR][KernelGen] Make maximum supported rank in rank specialization configurable A. Unique TensorFlower 2021-06-01 06:53:30 -0700
• c7c245eaf1 [MLIR][KernelGen] Add MLIR-generated Xlogy kernel A. Unique TensorFlower 2021-06-01 04:47:21 -0700
• 6a570502b6 Integrate LLVM at llvm/llvm-project@f000c4cfb6 A. Unique TensorFlower 2021-06-01 03:49:46 -0700
• f16e5a3a67 [MLIR][HLO] Use canonicalization patterns in broadcast propagation pass A. Unique TensorFlower 2021-06-01 03:13:39 -0700
• adff130b12 Integrate LLVM at llvm/llvm-project@41d7909368 A. Unique TensorFlower 2021-06-01 02:08:44 -0700
• 31536431e0 [MLIR][HLO] Eliminate duplicate broadcastable constraints A. Unique TensorFlower 2021-05-31 13:49:21 -0700
• 0f341012c6 [MLIR][HLO] Eliminate duplicate broadcastable constraints A. Unique TensorFlower 2021-05-31 13:07:16 -0700
• 511a1db4f3 [MLIR][HLO] Use canonicalization patterns in broadcast propagation pass A. Unique TensorFlower 2021-05-31 12:00:41 -0700
• 5f5db13715 [MLIR][HLO] Use canonicalization patterns in broadcast propagation pass A. Unique TensorFlower 2021-05-31 11:43:37 -0700
• 1f786eb934 Integrate LLVM at llvm/llvm-project@c36ff6424f A. Unique TensorFlower 2021-05-31 08:15:15 -0700
• cc1b22e618 [HLO][Linalg] Support scalar broadcasts in point-wise converter A. Unique TensorFlower 2021-05-31 03:49:28 -0700
• 832f39b871 Integrate LLVM at llvm/llvm-project@577fea4e1a A. Unique TensorFlower 2021-05-31 01:53:12 -0700
• 8110683c0b Integrate LLVM at llvm/llvm-project@9968896cd6 A. Unique TensorFlower 2021-05-28 12:08:17 -0700