Commit Graph

249 Commits

Author SHA1 Message Date
Adrian Kuegel cc607bc72d Support up to rank 8 in rank specialization for SelectOp.
PiperOrigin-RevId: 367406557
2021-04-08 04:55:41 -07:00
Hanhan Wang c466f08993 Add checks in ReduceWindowOpOnTensorsConversion.
The pattern does not support ops with non-zero padding config. Add a check to
prevent unexpected lowering.

It is not easy to add tests because other patterns will convert body ops, and
it causes issues like invalid IRs.

PiperOrigin-RevId: 367202450
2021-04-07 05:46:01 -07:00
A. Unique TensorFlower a62382cf1f Integrate LLVM at llvm/llvm-project@0e92cbd6a6
Updates LLVM usage to match
[0e92cbd6a652](https://github.com/llvm/llvm-project/commit/0e92cbd6a652)

PiperOrigin-RevId: 367062980
2021-04-06 12:40:51 -07:00
Rahul Joshi ff2cbfa2ec [MLIR] Add support for representing variadic reduce-window in HLO/LMHLO dialect.
-  Fixed a subset of transformations to handle variadic reduce-window.

PiperOrigin-RevId: 366278650
2021-04-01 10:24:50 -07:00
A. Unique TensorFlower bbe0aa204c [MLIR][MHLO] Merge assuming ops with compatible witnesses
PiperOrigin-RevId: 366018349
2021-03-31 06:11:38 -07:00
Adrian Kuegel c8157ba4df Move code from helper struct to the only user.
We don't need the separate helper struct anymore, because it is now only used
in one place.

PiperOrigin-RevId: 366012639
2021-03-31 05:21:34 -07:00
Adrian Kuegel 4033a56750 Add special cases for SelectOp rank specialization.
We now use the same special cases for all ops with arity >= 2.
For binary ops, we now have only one special case if at least one of the
operands has exactly one element. In that case, we reshape both operands to
rank 1. Before, we had separate special cases whether the left-hand side
or the right-hand side have a scalar shape.

PiperOrigin-RevId: 366005835
2021-03-31 04:28:51 -07:00
A. Unique TensorFlower 9206805c58 [MLIR][MHLO] Do not yield results of ops that were moved out of assuming regions
When an op is moved out of an assuming region we already know statically that it
is independent of the assuming region. Hence, there is no need to yield its
results.

PiperOrigin-RevId: 366001405
2021-03-31 03:50:27 -07:00
A. Unique TensorFlower 8ade5d78c8 [MLIR][MHLO] Move `cstr_broadcastable` and `shape_of` out of `assuming` regions
Add pattern to move operations out of assuming op. This only valid for
constraint-independent ops, like `cstr_broadcastable` and `shape_of`. It will
eventually allow to make assuming regions' constraints independent from each
other so that they can be merged.

PiperOrigin-RevId: 365993145
2021-03-31 02:39:07 -07:00
A. Unique TensorFlower eade942635 [MLIR][MHLO] Add pattern to move ops into the assuming region
This will eventually allow to make assuming regions' constraints independent
from each other.

PiperOrigin-RevId: 365985081
2021-03-31 01:23:31 -07:00
Adrian Kuegel c1a6ae8994 Generalize the HloBinaryElementwiseAdaptor
We can use it also for ternary ops like Select if we change the signature so
that a ValueRange is passed in.
Also remove special casing for HloComplexAdaptor. It can be handled with the
generic adaptor as well.

PiperOrigin-RevId: 365777493
2021-03-30 03:53:53 -07:00
Adrian Kuegel 6388e8d9ee mlir-hlo-opt: set preloadDialectsInContext to false.
This requires specifying dependent dialects in several passes.

PiperOrigin-RevId: 365758084
2021-03-30 01:07:14 -07:00
A. Unique TensorFlower 85a306d356 [MLIR][MHLO] Add pattern to inline broadcasted shapes
Simplify reasoning about `cstr_broadcastable` ops in the
`mhlo-move-up-dynamic-broadcasts-for-fusion` pass.

PiperOrigin-RevId: 365560893
2021-03-29 06:32:32 -07:00
A. Unique TensorFlower fb819c1de8 [MLIR][MHLO] Apply patterns in MoveUpDynamicBroadcastsForFusionPass greedily
PiperOrigin-RevId: 365556488
2021-03-29 06:02:06 -07:00
Adrian Kuegel a34aa699f8 Fix tanh lowering for NaN input.
If the input is NaN, the result should be NaN, too.

PiperOrigin-RevId: 364788902
2021-03-24 06:34:36 -07:00
Stella Laurenzo 7f2bf48b8b Integrate LLVM at llvm/llvm-project@b24436ac96
Updates LLVM usage to match
[b24436ac96bd](https://github.com/llvm/llvm-project/commit/b24436ac96bd)

PiperOrigin-RevId: 364615807
2021-03-23 12:20:17 -07:00
A. Unique TensorFlower 8987dfd1d6 [MLIR][HLO] Move broadcasts over n-ary shape-preserving ops
This will open up more fusion opportunities.

PiperOrigin-RevId: 364577231
2021-03-23 09:38:39 -07:00
A. Unique TensorFlower 54f37abc28 [MHLO] Move broadcasts over elementwise ops
Move up dynamic broadcasts and shape computations to allow for more fusion
opportunities.

PiperOrigin-RevId: 364514158
2021-03-23 02:34:41 -07:00
Benjamin Kramer 59fa7c0ef7 [MHLO:linalg] Lower all dynamic broadcasts of static shapes to linalg.generic
We only need the memref_reinterpret_cast if we don't know whether a dimension
gets expanded or not. With static shapes we know that a dimension can only be
expanded if it's a static 1, so lower it in the same way we lower fully
static broadcasts.

PiperOrigin-RevId: 363859181
2021-03-19 03:52:02 -07:00
Hanhan Wang 2e0ee7759b Add support for lowering mhlo.torch_index_select to Linalg on tensors.
The change upstreams the pattern from IREE repo to MHLO repo.

PiperOrigin-RevId: 363406294
2021-03-17 06:33:41 -07:00
A. Unique TensorFlower c54527fe88 Integrate LLVM at llvm/llvm-project@678241795c
Updates LLVM usage to match
[678241795c95](https://github.com/llvm/llvm-project/commit/678241795c95)

PiperOrigin-RevId: 363257913
2021-03-16 13:33:00 -07:00
A. Unique TensorFlower 2be112a603 [MLIR][MHLO] Approximate `tf.Tanh` as constant +/-1 for small/large values
Fix issue raised in https://github.com/tensorflow/tensorflow/issues/47724

PiperOrigin-RevId: 363210296
2021-03-16 10:14:30 -07:00
Hanhan Wang 4f5e1c51dd Add support for lowering NHWC pooling mhlo.reduce_window to Linalg on tensors.
The change upstreams the pattern from IREE repo to MHLO repo.

PiperOrigin-RevId: 362312573
2021-03-11 09:41:34 -08:00
Hanhan Wang 630cabefb0 Add support for lowering 2D depthwise mhlo.conv to Linalg on tensors.
The change upstreams the pattern from IREE repo to MHLO repo.

PiperOrigin-RevId: 362300550
2021-03-11 08:41:38 -08:00
Benjamin Kramer 94f9740c67 [MLIR][HLO:Linalg] Lower mhlo.dynamic_iota to indexed_generic
This is the same as iota, but instead of taking the dimensions from the result
tensor we use the supplied shape extents tensor.

PiperOrigin-RevId: 362298548
2021-03-11 08:31:29 -08:00
Benjamin Kramer 09f8046816 [MLIR:HLO:LINALG] Fix codegen for mhlo.reshape when one side is rank 0
This is an annoying edge case because the collapse->expand lowering expects at
least R1 or it will produce invalid linalg reshapes. Using the direct lowering
works fine.

PiperOrigin-RevId: 362269199
2021-03-11 05:29:56 -08:00
Benjamin Kramer d77b556822 [MLIR][MHLO] Allow recursion in the shape_of mover
This allows it to push shape_of over a chain of ops all the way to the top.

PiperOrigin-RevId: 362249009
2021-03-11 02:52:21 -08:00
Benjamin Kramer 67a770e4e0 [HLO:MLIR] Make binary op type reification emit shape_of instead of tensor ops
This gives cleaner code and allows shape optimizations to happen on the result.

PiperOrigin-RevId: 362242975
2021-03-11 02:01:35 -08:00
Mahesh Ravishankar b212bd66ae Build fix for missing precision_config.
THe conversion from dot_general to dot fails when trying to retrieve
and use the precision config, since precision_config is optional.

PiperOrigin-RevId: 362095296
2021-03-10 11:10:51 -08:00
A. Unique TensorFlower e199df1dbf [MLIR][MHLO] Declare `shape_of` dynamically legal in move-up-dynamic-broadcasts
This allows shape reification to produce `shape_of` ops while they can still be
moved up.

PiperOrigin-RevId: 362075609
2021-03-10 09:59:17 -08:00
A. Unique TensorFlower c217a6ef61 [MHLO] Add pass to move up dynamic broadcasts for fusion
For now, the pass only reifies the required shape computations. Moving
broadcasts will follow to allow for fusion across them.

PiperOrigin-RevId: 362033715
2021-03-10 06:21:57 -08:00
A. Unique TensorFlower 507d9fb61d [MLIR][KernelGen] Add `tf.Polygamma` kernel
PiperOrigin-RevId: 362002943
2021-03-10 02:22:01 -08:00
A. Unique TensorFlower 218476128e [MLIR][KernelGen] Fix zeta lowering at poles
Return nan at zeta poles or inf where the limit is defined. Also test the kernel
based on the series representation of zeta.

PiperOrigin-RevId: 361993482
2021-03-10 01:09:10 -08:00
Benjamin Kramer 5be8be31b5 Integrate LLVM at llvm/llvm-project@3f3f88fb95
Updates LLVM usage to match
[3f3f88fb9503](https://github.com/llvm/llvm-project/commit/3f3f88fb9503)

PiperOrigin-RevId: 361762801
2021-03-09 02:19:24 -08:00
Marius Brehler 29f70cb892 PR #46723: Adjust types of loop counters
Imported from GitHub PR https://github.com/tensorflow/tensorflow/pull/46723

Reduces some warnings about comparison of integers of different signs.
Copybara import of the project:

--
311f436f77b334f5462127d8cf179cce067969ca by Marius Brehler <marius.brehler@iml.fraunhofer.de>:

Adjust types of loop counters

Reduces some warnings about comparison of integers of different signs.

PiperOrigin-RevId: 360912203
2021-03-04 07:36:12 -08:00
A. Unique TensorFlower 39650a5d5a Remove rank 1 specialization from TransformUnrankedHloPass.
For binary ops, we already special-case rank 0 vs rank 1, and same shape. So we
don't need to special-case a maximum rank of 1.

PiperOrigin-RevId: 360891955
2021-03-04 05:24:53 -08:00
Adrian Kuegel 62b357b601 Remove rank 1 specialization from TransformUnrankedHloPass.
For binary ops, we already special-case rank 0 vs rank 1, and same shape. So we
don't need to special-case a maximum rank of 1.

PiperOrigin-RevId: 360881387
2021-03-04 04:04:11 -08:00
Geoffrey Martin-Noble 8687f3e4cf Lower MHLO Dot to type-polymorphic linalg named ops
The linalg named ops are now type polymorphic, so the type-monomorphic
varieties are redundant (and will be deleted soon).

PiperOrigin-RevId: 360509010
2021-03-02 14:00:58 -08:00
Benjamin Kramer 1facbe9eb5 Integrate LLVM at llvm/llvm-project@7f086d74c3
Updates LLVM usage to match
[7f086d74c347](https://github.com/llvm/llvm-project/commit/7f086d74c347)

PiperOrigin-RevId: 360434104
2021-03-02 08:33:21 -08:00
Adrian Kuegel 0683db3b24 Legalize MinimumBroadcastShapes op.
Use it in TransformUnrankedHloPass, which allows to reduce the maximum
rank for rank specialized broadcast from 6 to 5.

PiperOrigin-RevId: 360415743
2021-03-02 06:39:01 -08:00
Christian Sigg 70ee9369d5 Use mlir::OpState::operator->() to get to Operation::getAttrs().
This is a preparation step to remove getAttrs() from OpState.

PiperOrigin-RevId: 360159716
2021-03-01 04:53:00 -08:00
Benjamin Kramer e19ccf975e Filter static dimensions from dynamic_broadcast_in_dim's init_tensor
Otherwise we'd generate invalid IR for those cases.

PiperOrigin-RevId: 360144122
2021-03-01 03:03:54 -08:00
Christian Sigg 2d818c4fd9 Use mlir::OpState::operator->() to get to methods of mlir::Operation.
This is a preparation step to remove those methods from OpState.

PiperOrigin-RevId: 360043992
2021-02-28 09:02:33 -08:00
Hanhan Wang a8f99ee0f5 Fix the shape of linalg.init_tensor in conv op lowering.
The output spatial dims are not as same as the input spatial dims. Only supports
static output spatial dims for now.

PiperOrigin-RevId: 359775479
2021-02-26 09:34:11 -08:00
Hanhan Wang 90f0d7f935 Add support for lowering mhlo.conv to Linalg on tensors.
This pattern only works for normal convolutions. It does not work for depthwise
convolutions. The Linalg conv ops are defined with static rank, so it only
supports 1d/2d/3d cases, which are the most typical cases.

This also refactors out the same check in lmhlo.conv lowering.

PiperOrigin-RevId: 359503527
2021-02-25 05:59:08 -08:00
Hanhan Wang 45a1249fe2 Add support for lowering mhlo.pad to linalg.pad_tensor
The change upstreams the pattern from IREE repo to MHLO repo.

PiperOrigin-RevId: 359481543
2021-02-25 03:00:39 -08:00
Geoffrey Martin-Noble 89f7f2bd65 Lower integer matmuls to linalg
PiperOrigin-RevId: 359306495
2021-02-24 09:45:07 -08:00
Hanhan Wang 475b4a06a5 Add support for lowering mhlo.slice to subtensor.
PiperOrigin-RevId: 359297978
2021-02-24 09:06:09 -08:00
A. Unique TensorFlower ac0552f127 [MLIR][HLO] Remove duplicate `PopulateTransformUnrankedHloPatterns`
PiperOrigin-RevId: 359046173
2021-02-23 07:50:47 -08:00
Benjamin Kramer a9cc1dcfa0 [mlir][hlo] Add basic rank-specialization for select
This just blows up everything to ranked (up to 6) and is probably quite slow.
This is sufficient to make kernelgen compile SelectV2.

PiperOrigin-RevId: 358777728
2021-02-22 02:41:12 -08:00