8134bff98d
[XLA:GPU] Add layout attributes to LHLO_GPU Convolution operations.
...
- MLIR MemRefs do not preserve layout information correctly when unit dimensions
are involved. Operations like convolution that use cuDNN however need the correct
layout to be preserved so that we do not end up creating an incompatible combination
of input/filter/output layout that is not supported by cuDNN.
- Add these layouts to convolution attributes in the form of I32ArrayAttr for representing
the layout in "minor_to_major" form similar to XLA.
PiperOrigin-RevId: 348034757
2020-12-17 09:26:28 -08:00
dc7e63f74c
Use mlir::OpState::operator->() to get to methods of mlir::Operation.
...
This is a preparation step to remove those methods from OpState.
PiperOrigin-RevId: 348010582
2020-12-17 06:29:33 -08:00
1f244c3e2c
Fix SignOp lowering for floating point values.
...
It didn't return 0 for 0.0 and -0.0.
Currently we emit -0.0 for -0.0 which is correct according to the HLO dialect.
For the TF_SignOp we should emit 0.0 in that case, we will leave that as a TODO.
Enable the tests which work now, and add another one for Int64.
Also improve the registration code, we should not register the Int32 kernel.
PiperOrigin-RevId: 347981124
2020-12-17 01:45:54 -08:00
099c130daf
Fix MLIR include paths.
...
PiperOrigin-RevId: 347976151
2020-12-17 00:56:04 -08:00
5da9190dd9
Integrate LLVM at llvm/llvm-project@0cf7e4b252
...
Updates LLVM usage to match
[0cf7e4b252fe](https://github.com/llvm/llvm-project/commit/0cf7e4b252fe )
PiperOrigin-RevId: 347948887
2020-12-16 20:30:17 -08:00
5a080ad795
Add Bazel build configuration to MLIR HLO standalone repo.
...
Tested that this works in OSS in tree with `bazel build //...` and as a submodule in IREE.
PiperOrigin-RevId: 347919106
2020-12-16 16:23:31 -08:00
65222893ae
[KERNEL_GEN] Convert LHLO AddOp, SubOp (ComplexType) to complex ops.
...
PiperOrigin-RevId: 347805898
2020-12-16 05:45:06 -08:00
61244b136c
Try to avoid a segfault if we don't support a lowering.
...
It can happen that a lowering for a certain type is not implemented yet.
We should not segfault in such a case, but instead return a failure().
PiperOrigin-RevId: 347801106
2020-12-16 04:58:17 -08:00
e6e8920921
[KERNEL_GEN] Switch the pipeline to Linalg-on-Tensors.
...
PiperOrigin-RevId: 347781190
2020-12-16 01:51:15 -08:00
5ab94a00a7
[mlir][NFC] Replace usages or mlir/IR/StandardTypes.h with mlir/IR/BuiltinTypes.h
...
StandardTypes.h was moved to BuiltinTypes.h and is being removed.
PiperOrigin-RevId: 347751511
2020-12-15 21:28:18 -08:00
ccbf39842f
Integrate LLVM at llvm/llvm-project@7aeb3804c4
...
Updates LLVM usage to match
[7aeb3804c46c](https://github.com/llvm/llvm-project/commit/7aeb3804c46c )
PiperOrigin-RevId: 347727361
2020-12-15 17:40:25 -08:00
6d7812bbc5
Correct non ConversionRewriter transformation in LegalizeToLinalg
...
PiperOrigin-RevId: 347622657
2020-12-15 08:56:58 -08:00
f0c2695d31
Fix SignOp lowering for floating point values.
...
It didn't return 0 for 0.0 and -0.0.
Currently we emit -0.0 for -0.0 which is correct according to the HLO dialect.
For the TF_SignOp we should emit 0.0 in that case, we will leave that as a TODO.
Enable the tests which work now, and add another one for Int64.
Also improve the registration code, we should not register the Int32 kernel.
PiperOrigin-RevId: 347602378
2020-12-15 06:49:48 -08:00
ddda2699fb
[KERNEL_GEN] Switch the pipeline to Linalg-on-Tensors.
...
PiperOrigin-RevId: 347600145
2020-12-15 06:32:25 -08:00
79fa36bcbc
Fix SignOp lowering for floating point values.
...
It didn't return 0 for 0.0 and -0.0.
Currently we emit -0.0 for -0.0 which is correct according to the HLO dialect.
For the TF_SignOp we should emit 0.0 in that case, we will leave that as a TODO.
Enable the tests which work now, and add another one for Int64.
Also improve the registration code, we should not register the Int32 kernel.
PiperOrigin-RevId: 347590340
2020-12-15 05:12:48 -08:00
9540e51617
[mlir][NFC] Replace usages or mlir/IR/StandardTypes.h with mlir/IR/BuiltinTypes.h
...
StandardTypes.h was moved to BuiltinTypes.h and is being removed.
PiperOrigin-RevId: 347559927
2020-12-15 00:59:29 -08:00
9df327d88f
Forward listeners in LhloLegalizeToParallelLoops builders
...
PiperOrigin-RevId: 347554379
2020-12-15 00:13:15 -08:00
1a58f19664
[NFC] Make function names follow style guide.
...
Functions should start with a capital letter and have a capital letter for each
new word. See https://google.github.io/styleguide/cppguide.html#Function_Names
PiperOrigin-RevId: 347420402
2020-12-14 10:46:55 -08:00
8b35a75d4a
[KERNEL_GEN] Switch the pipeline to Linalg-on-Tensors.
...
PiperOrigin-RevId: 347368063
2020-12-14 05:46:47 -08:00
6b439f7eee
[mlir][NFC] Replace usages or mlir/IR/StandardTypes.h with mlir/IR/BuiltinTypes.h
...
StandardTypes.h was moved to BuiltinTypes.h and is being removed.
PiperOrigin-RevId: 347115952
2020-12-11 19:01:25 -08:00
3abdd556de
[mlir][NFC] Replace usages or mlir/IR/StandardTypes.h with mlir/IR/BuiltinTypes.h
...
StandardTypes.h was moved to BuiltinTypes.h and is being removed.
PiperOrigin-RevId: 347111863
2020-12-11 18:12:59 -08:00
22b91d62ff
[XLA/GPU] Migrate all unnested elementwise emitters.
...
Also fix GetHloOutputs/GetHloOperands to treat aliased operands correctly.
PiperOrigin-RevId: 347055290
2020-12-11 12:45:21 -08:00
21bb9e8c9a
Integrate LLVM at llvm/llvm-project@38d32e4fd7
...
Updates LLVM usage to match
[38d32e4fd70c](https://github.com/llvm/llvm-project/commit/38d32e4fd70c )
PiperOrigin-RevId: 346994267
2020-12-11 07:28:49 -08:00
9930c20c31
[mlir][hlo] Fix lowering of NE comparison. It should return true if either side is NaN
...
PiperOrigin-RevId: 346988987
2020-12-11 06:46:14 -08:00
ab6ee11813
Fix folding of HLO SliceOp with zero elements
...
This was causing division by zero in this case.
PiperOrigin-RevId: 346920942
2020-12-10 20:22:48 -08:00
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.
PiperOrigin-RevId: 346891737
2020-12-10 16:39:19 -08:00
3442ac270d
Integrate LLVM at llvm/llvm-project@997a719d5a
...
Updates LLVM usage to match
[997a719d5a70](https://github.com/llvm/llvm-project/commit/997a719d5a70 )
PiperOrigin-RevId: 346834025
2020-12-10 11:58:16 -08:00
6a05893169
[XLA/GPU] Migrate all unnested elementwise emitters.
...
PiperOrigin-RevId: 346725498
2020-12-10 00:21:36 -08:00
51f535454d
Use OpState::operator->() to get to member functions in Operation so we can remove the corresponding methods from OpState.
...
PiperOrigin-RevId: 346721668
2020-12-09 23:54:07 -08:00
c36afd275e
[HLO] Add a pattern for HLO ConstOp to HLO -> Linalg conversion.
...
PiperOrigin-RevId: 346718273
2020-12-09 23:24:57 -08:00
cfcf741932
[XLA/GPU] Migrate all unnested elementwise emitters.
...
PiperOrigin-RevId: 346716519
2020-12-09 23:08:13 -08:00
65ebd85563
[XLA/GPU] Migrate all unnested elementwise emitters.
...
PiperOrigin-RevId: 346658288
2020-12-09 16:05:43 -08:00
1c10e1fec6
[XLA/GPU] Migrate all unnested elementwise emitters.
...
PiperOrigin-RevId: 346624905
2020-12-09 13:31:50 -08:00
afea1e1897
Integrate LLVM at llvm/llvm-project@6883042528
...
Updates LLVM usage to match
[6883042528d0](https://github.com/llvm/llvm-project/commit/6883042528d0 )
PiperOrigin-RevId: 346606080
2020-12-09 12:06:47 -08:00
1a5a1b5f41
[XLA/GPU] Migrate all unnested elementwise emitters.
...
PiperOrigin-RevId: 346574093
2020-12-09 09:54:16 -08:00
3c33fe4b9e
[XLA/GPU] Migrate all unnested elementwise emitters.
...
PiperOrigin-RevId: 346559170
2020-12-09 08:42:51 -08:00
6cb9cc53ec
Integrate LLVM at llvm/llvm-project@d553243fe4
...
Updates LLVM usage to match
[d553243fe4b5](https://github.com/llvm/llvm-project/commit/d553243fe4b5 )
PiperOrigin-RevId: 346393264
2020-12-08 13:09:13 -08:00
e9057cbd93
[mlir] Enable conversion of HLO SignOp to Linalg.
...
PiperOrigin-RevId: 346380514
2020-12-08 12:08:26 -08:00
4e6367e9d6
Integrate LLVM at llvm/llvm-project@a1344779ab
...
Updates LLVM usage to match
[a1344779ab01](https://github.com/llvm/llvm-project/commit/a1344779ab01 )
PiperOrigin-RevId: 346312316
2020-12-08 06:46:57 -08:00
c3790af758
Add plumbing for or and xor to hlo to lhlo and linalg lowerings.
...
PiperOrigin-RevId: 346311314
2020-12-08 06:39:02 -08:00
dd5895d083
Extend unranked hlo transformations to also support and, or and xor.
...
PiperOrigin-RevId: 346270393
2020-12-08 01:00:26 -08:00
812221db97
Integrate LLVM at llvm/llvm-project@2ac4d0f45a
...
Updates LLVM usage to match
[2ac4d0f45a2a](https://github.com/llvm/llvm-project/commit/2ac4d0f45a2a )
PiperOrigin-RevId: 346203743
2020-12-07 16:20:57 -08:00
5235eceea0
Lower mhlo shifts to linalg
...
PiperOrigin-RevId: 346161253
2020-12-07 13:02:32 -08:00
eaa21130e8
Integrate LLVM at llvm/llvm-project@ecaff13fc0
...
Updates LLVM usage to match
[ecaff13fc0bc](https://github.com/llvm/llvm-project/commit/ecaff13fc0bc )
PiperOrigin-RevId: 346083459
2020-12-07 07:00:58 -08:00
d327fc5737
[kernel_gen] Lower max rank specialization from 6 to 5
...
We don't care much about rank 6 broadcasting operations and this lowers compile times significantly.
PiperOrigin-RevId: 346046601
2020-12-07 02:18:38 -08:00
4c0c9ab119
Integrate LLVM at llvm/llvm-project@db226cdf4c
...
Updates LLVM usage to match
[db226cdf4cf9](https://github.com/llvm/llvm-project/commit/db226cdf4cf9 )
PiperOrigin-RevId: 346045832
2020-12-07 02:13:36 -08:00
bc7b6374c8
Fix handling of negative seeds in random number generator op kernels for XLA
...
Casting negative s32 number to u64 directly will have leading 1s in the representation which is not what we want to get a single u64 out of two s32 seeds. Fixed this by first getting unsigned number of the same bit-width.
PiperOrigin-RevId: 345902167
2020-12-05 18:55:41 -08:00
55268f9ee8
Integrate LLVM at llvm/llvm-project@7f6f9f4cf9
...
Updates LLVM usage to match
[7f6f9f4cf966](https://github.com/llvm/llvm-project/commit/7f6f9f4cf966 )
PiperOrigin-RevId: 345745888
2020-12-04 13:33:48 -08:00
c33bdcbd03
Remove fold of `mhlo.compare(%arg0, %arg0)` for floating types.
...
Two tensors having the same SSA-value isn't sufficient for equality for floating types, as `NaN != NaN`. As written this causes `tf.IsNan` to [miscompile](https://github.com/google/iree/issues/4061 ).
PiperOrigin-RevId: 345730640
2020-12-04 12:15:02 -08:00
9bd1995f90
Legalize XlaReplicaId to HLO replica-id op
...
Also, define shape inference function for HLO replica-id op.
PiperOrigin-RevId: 345714342
2020-12-04 11:04:40 -08:00
Rahul Joshi
Christian Sigg
Adrian Kuegel
A. Unique TensorFlower
Phoenix Meadowlark
Alexander Belyaev
River Riddle
Tres Popp
Hanhan Wang
Tim Shen
Benjamin Kramer
Smit Hinsu
Stephan Herhut