mlir-hlo

2d818c4fd9 Use mlir::OpState::operator->() to get to methods of mlir::Operation. Christian Sigg 2021-02-28 09:01:26 -0800
006b58c476 Integrate LLVM at llvm/llvm-project@5077d42cfa A. Unique TensorFlower 2021-02-26 22:17:57 -0800
a8f99ee0f5 Fix the shape of linalg.init_tensor in conv op lowering. Hanhan Wang 2021-02-26 09:33:21 -0800
c616963501 Integrate LLVM at llvm/llvm-project@72b18a86e1 A. Unique TensorFlower 2021-02-26 08:19:52 -0800
c68b71b5dc Integrate LLVM at llvm/llvm-project@cb81135f94 A. Unique TensorFlower 2021-02-26 06:45:15 -0800
c63ac91206 Integrate LLVM at llvm/llvm-project@bf6380c096 A. Unique TensorFlower 2021-02-26 04:03:59 -0800
fe4234bcff Integrate LLVM at llvm/llvm-project@5d64dd8e3c A. Unique TensorFlower 2021-02-25 17:35:27 -0800
e7a3ec18ad Integrate LLVM at llvm/llvm-project@7f6e331645 A. Unique TensorFlower 2021-02-25 13:10:45 -0800
bb22553219 Integrate LLVM at llvm/llvm-project@f4d78a5e3a A. Unique TensorFlower 2021-02-25 09:39:28 -0800
90f0d7f935 Add support for lowering mhlo.conv to Linalg on tensors. Hanhan Wang 2021-02-25 05:58:12 -0800
c5f5d13930 [MLIR] Add dependency from "lhlo_ops_structs_inc_gen" to "lhlo" target. Rahul Joshi 2021-02-25 03:50:11 -0800
45a1249fe2 Add support for lowering mhlo.pad to linalg.pad_tensor Hanhan Wang 2021-02-25 02:59:55 -0800
459362b206 Integrate LLVM at llvm/llvm-project@d8a8e5d624 A. Unique TensorFlower 2021-02-25 02:17:32 -0800
7dc1543bb2 Integrate LLVM at llvm/llvm-project@96a3dfeb93 A. Unique TensorFlower 2021-02-24 16:16:16 -0800
d9d6ab105e Integrate LLVM at llvm/llvm-project@5c74c6be3c A. Unique TensorFlower 2021-02-24 13:17:49 -0800
89f7f2bd65 Lower integer matmuls to linalg Geoffrey Martin-Noble 2021-02-24 09:43:54 -0800
475b4a06a5 Add support for lowering mhlo.slice to subtensor. Hanhan Wang 2021-02-24 09:04:32 -0800
b478bdf00e Integrate LLVM at llvm/llvm-project@de40423c85 A. Unique TensorFlower 2021-02-24 07:31:13 -0800
3e2ff6e253 Integrate LLVM at llvm/llvm-project@861dbe1a02 A. Unique TensorFlower 2021-02-23 11:26:24 -0800
ac0552f127 [MLIR][HLO] Remove duplicate `PopulateTransformUnrankedHloPatterns` A. Unique TensorFlower 2021-02-23 07:49:54 -0800
2df1bb3d6b Integrate LLVM at llvm/llvm-project@6c9541d4dd A. Unique TensorFlower 2021-02-23 04:37:31 -0800
855cd484fc Integrate LLVM at llvm/llvm-project@557d2ade01 A. Unique TensorFlower 2021-02-22 16:03:55 -0800
b45eaf8d33 Integrate LLVM at llvm/llvm-project@12edddafac A. Unique TensorFlower 2021-02-22 13:19:23 -0800
0da7ea2545 [MLIR][HLO] Cleanup CMakeLists and comments. Rahul Joshi 2021-02-22 11:35:38 -0800
5adb7c6e12 [MLIR:LHLO] Add optional call target arg mapping to LMHLO CustomCall operations. Rahul Joshi 2021-02-22 08:41:59 -0800
a9cc1dcfa0 [mlir][hlo] Add basic rank-specialization for select Benjamin Kramer 2021-02-22 02:39:56 -0800
909574e393 Pass result element type to XlaBuilder for `mhlo.dot_general` and `mhlo.convolution` ops. Prakalp Srivastava 2021-02-20 07:06:24 -0800
f63c93399a Integrate LLVM at llvm/llvm-project@7e54d7304b A. Unique TensorFlower 2021-02-19 17:45:39 -0800
5cc9fa9d9d Integrate LLVM at llvm/llvm-project@cd8cc00b9e A. Unique TensorFlower 2021-02-18 21:08:32 -0800
5400287d69 Integrate LLVM at llvm/llvm-project@b7e05c874b A. Unique TensorFlower 2021-02-18 16:18:05 -0800
27a8b18526 Integrate LLVM at llvm/llvm-project@f70cdc5b5c A. Unique TensorFlower 2021-02-18 12:38:16 -0800
72d3eee118 Integrate LLVM at llvm/llvm-project@5b094bfeb3 A. Unique TensorFlower 2021-02-18 10:50:02 -0800
b42def4612 [mlir][hlo] Refactor rank specialization to allow an arbitrary number of inputs Benjamin Kramer 2021-02-18 09:52:03 -0800
ca4034b56e [mlir][hlo] Make select ready for dynamic shapes (ranked only for now) Benjamin Kramer 2021-02-18 08:07:36 -0800
a6f03aecfb Integrate LLVM at llvm/llvm-project@624fccba87 A. Unique TensorFlower 2021-02-18 06:48:41 -0800
37e31f8b26 Lower Expm1 kernel to math.ExpM1. Adrian Kuegel 2021-02-18 04:53:34 -0800
90b222cf6e Integrate LLVM at llvm/llvm-project@55756f32f7 A. Unique TensorFlower 2021-02-17 20:31:14 -0800
b579bd5d9e Support dynamic-shaped operand in verification of BroadcastInDim. Richard Uhler 2021-02-17 16:16:32 -0800
dd237d4267 Integrate LLVM at llvm/llvm-project@892d2822b6 A. Unique TensorFlower 2021-02-17 14:51:51 -0800
220deb3709 [MLIR][CHLO] Add legalization for `chlo.polygamma` to MHLO A. Unique TensorFlower 2021-02-17 08:31:46 -0800
c06de24f6c [MLIR][CHLO] Generalize lowering with upcast to n-ary operation A. Unique TensorFlower 2021-02-16 09:46:33 -0800
81abaf364d [MLIR][MHLO] Add polygamma op to the CHLO dialect A. Unique TensorFlower 2021-02-16 08:31:19 -0800
02dab94054 Integrate LLVM at llvm/llvm-project@9f581815ae A. Unique TensorFlower 2021-02-16 03:31:15 -0800
9447374676 Integrate LLVM at llvm/llvm-project@715dc556b7 Benjamin Kramer 2021-02-15 13:34:11 -0800
af5343ff86 Integrate LLVM at llvm/llvm-project@2dbe88db58 A. Unique TensorFlower 2021-02-15 09:57:02 -0800
b594254c79 [mhlo] Lower int->bool to a comparison with zero Adrian Kuegel 2021-02-15 06:37:06 -0800
240a44de82 [mhlo] Lower int->int cast to sign extension instead of zero extension Benjamin Kramer 2021-02-15 05:57:54 -0800
8672735e9a [mhlo] Lower float->bool to a comparison with zero Adrian Kuegel 2021-02-15 04:35:43 -0800
2fe0c33083 Integrate LLVM at llvm/llvm-project@16428a8d91 A. Unique TensorFlower 2021-02-15 04:17:20 -0800
89d81adf6d [mhlo] Lower float->bool to a comparison with zero A. Unique TensorFlower 2021-02-15 03:11:12 -0800
3e80d91e73 [mhlo] Lower float->bool to a comparison with zero Benjamin Kramer 2021-02-15 02:16:35 -0800
824bc9c425 Improve broadcast transformation to treat dynamic shapes with 1 element as scalar. Adrian Kuegel 2021-02-14 23:24:45 -0800
4060a86fe2 Integrate LLVM at llvm/llvm-project@2bfe27da17 A. Unique TensorFlower 2021-02-12 08:30:51 -0800
e993082b97 Integrate LLVM at llvm/llvm-project@6f04addc8b A. Unique TensorFlower 2021-02-12 05:19:24 -0800
f7ac575dd9 Integrate LLVM at llvm/llvm-project@0c118831a3 A. Unique TensorFlower 2021-02-12 03:30:28 -0800
a7365aadbe Integrate LLVM at llvm/llvm-project@8151c1b442 A. Unique TensorFlower 2021-02-11 16:15:59 -0800
077df4f8cc Integrate LLVM at llvm/llvm-project@418c218efa A. Unique TensorFlower 2021-02-11 10:48:15 -0800
6fa6974e8d [XLA/GPU] Plumb through Bitcast op for LMHLO. Tim Shen 2021-02-10 19:44:15 -0800
4e202135d1 Integrate LLVM at llvm/llvm-project@9db6e97a86 A. Unique TensorFlower 2021-02-10 09:31:02 -0800
36e04d92c0 [KERNEL_GEN] Add a pattern to bufferize `mhlo.reshape(<unranked_tensor>)`. Alexander Belyaev 2021-02-10 06:30:50 -0800
54c2a49866 Integrate LLVM at llvm/llvm-project@a5222aa085 A. Unique TensorFlower 2021-02-09 09:12:25 -0800
7446b6022d Integrate LLVM at llvm/llvm-project@2fa4186d4e A. Unique TensorFlower 2021-02-08 16:48:41 -0800
4a29ca3b1d Add layout to mhlo::InfeedOp td. A. Unique TensorFlower 2021-02-08 09:45:40 -0800
80d753c1fe Integrate LLVM at llvm/llvm-project@f89f6d1e5d A. Unique TensorFlower 2021-02-08 08:01:22 -0800
d086b8a0ec Correct HLO atan2 lowering in cases of -inf and -0 inputs. Tres Popp 2021-02-08 06:57:16 -0800
bd0856578f Integrate LLVM at llvm/llvm-project@a4fa667dee A. Unique TensorFlower 2021-02-05 15:51:53 -0800
2aa8a90c69 Integrate LLVM at llvm/llvm-project@a1a1d338e9 A. Unique TensorFlower 2021-02-05 14:19:24 -0800
b251712b1d [XLA:GPU] Add conversion from HLO -> MLIR LMHLO for TriangularSolve - Also add layout attributes for inputs and output for error checking. Rahul Joshi 2021-02-05 09:16:49 -0800
1c4521cc42 Integrate LLVM at llvm/llvm-project@d1978fa4bf Adrian Kuegel 2021-02-05 07:40:43 -0800
99bc05f2e4 Integrate LLVM at llvm/llvm-project@91e7a17133 A. Unique TensorFlower 2021-02-04 13:41:18 -0800
60e1b6882c Add kernel definition for zeta operation. Stephan Herhut 2021-02-04 01:26:44 -0800
945bf8768d Integrate LLVM at llvm/llvm-project@56fcd4ea8d A. Unique TensorFlower 2021-02-03 18:46:55 -0800
3380d0a761 Integrate LLVM at llvm/llvm-project@ea5b75de49 A. Unique TensorFlower 2021-02-03 15:19:30 -0800
44d0464d16 Use linalg.fill on tensors instead of tensor.generate in MHLO -> Linalg conversion. Mahesh Ravishankar 2021-02-03 15:02:21 -0800
10cd797d6d Integrate LLVM at llvm/llvm-project@bbb7555403 A. Unique TensorFlower 2021-02-03 12:55:19 -0800
6cd1875ee4 Implement lowering of chlo::zeta to mhlo dialect. Stephan Herhut 2021-02-03 07:49:12 -0800
9d8a53c452 Integrate LLVM at llvm/llvm-project@e635feb15a Adrian Kuegel 2021-02-03 07:31:04 -0800
0472837e50 Integrate LLVM at llvm/llvm-project@54afcade3b Adrian Kuegel 2021-02-03 05:48:48 -0800
1b63101a38 Integrate LLVM at llvm/llvm-project@fae6d129da Adrian Kuegel 2021-02-03 03:19:19 -0800
39aa1235d1 Integrate LLVM at llvm/llvm-project@2668714747 A. Unique TensorFlower 2021-02-02 17:56:58 -0800
7cbf8d8878 Integrate LLVM at llvm/llvm-project@83d705adb2 A. Unique TensorFlower 2021-02-02 14:52:39 -0800
04110a4b1c Integrate LLVM at llvm/llvm-project@67dfe9c8d7 A. Unique TensorFlower 2021-02-02 13:07:39 -0800
3b67b207c4 [MLIR][CHLO] Use CHLO lowering for `is_inf` op A. Unique TensorFlower 2021-02-02 09:52:11 -0800
0726ab20fa Integrate LLVM at llvm/llvm-project@4d904776a7 Adrian Kuegel 2021-02-02 05:09:12 -0800
ae722a883f Improve performance of lowered chlo.pow with integers Tres Popp 2021-02-02 03:27:38 -0800
0458ae9a22 [MLIR][KernelGen] Add `tf.Digamma` kernels A. Unique TensorFlower 2021-02-02 03:06:54 -0800
f40ccc5b4b [MLIR][CHLO] Add `chlo.digamma` and lowering to MHLO A. Unique TensorFlower 2021-02-02 02:09:04 -0800
c2115f56c7 Integrate LLVM at llvm/llvm-project@8f7f2c4211 Adrian Kuegel 2021-02-02 01:53:06 -0800
96f8771ed7 Add MLIR generated kernel for Angle kernel. Adrian Kuegel 2021-02-02 00:45:39 -0800
9d682343a9 Integrate LLVM at llvm/llvm-project@68f66f37d7 A. Unique TensorFlower 2021-02-01 18:15:37 -0800
6c8ad0895b Integrate LLVM at llvm/llvm-project@75347ba1fa A. Unique TensorFlower 2021-02-01 14:50:03 -0800
16bb35009b Integrate LLVM at llvm/llvm-project@78c22fbce9 A. Unique TensorFlower 2021-02-01 12:26:39 -0800
816d279be3 [MLIR][CHLO] Simplify conversions with upcast A. Unique TensorFlower 2021-02-01 10:46:41 -0800
8e3890e8e8 [MLIR:HLO] Add AllGather and AllToAll operations to LMHLO dialect. Rahul Joshi 2021-02-01 10:22:48 -0800
7a251aaed4 Integrate LLVM at llvm/llvm-project@49a6502cd5 Adrian Kuegel 2021-02-01 07:41:58 -0800
2da9d6947d Integrate LLVM at llvm/llvm-project@e90e455d2a Adrian Kuegel 2021-02-01 05:55:50 -0800
1be1123c70 [XLA:GPU] Add support for PartitionId Rahul Joshi 2021-01-29 13:30:59 -0800
b1ce05cfc9 Integrate LLVM at llvm/llvm-project@1c762a81d2 A. Unique TensorFlower 2021-01-29 11:02:17 -0800
2b72ddc6b2 [MLIR][KernelGen] Add `lgamma` kernels A. Unique TensorFlower 2021-01-29 06:13:20 -0800
e61ef86fdb Add zeta and broadcasting_zeta to chlo dialect. Stephan Herhut 2021-01-29 03:21:59 -0800

Commit Graph Select branches Hide Pull Requests master Mono Color

Commit Graph

Select branches

Hide Pull Requests

master