Commit Graph

137 Commits

Author SHA1 Message Date
Wenyi Zhao 6660234d80 PR #50100: [MLIR][DISC] Bufferize DynamicIotaOp and DynamicPadOp
Imported from GitHub PR https://github.com/tensorflow/tensorflow/pull/50100

support hlo-to-lhlo conversion for DynamicIotaOp and DynamicPadOp
Copybara import of the project:

--
c3aae94954e35d3f8ad265f619ef9765665a5115 by Wenyi Zhao <reyizero@gmail.com>:

[MLIR][DISC] Bufferize DynamicIotaOp and DynamicPadOp

--
adc6996d70b804d61310d56a33fac975d70c8636 by Wenyi Zhao <reyizero@gmail.com>:

minor

PiperOrigin-RevId: 378733284
2021-06-10 14:20:45 -07:00
A. Unique TensorFlower 14093b7906 [XLA:GPU] Add AllReduce{Start,Done} to MLIR LHLO dialect.
PiperOrigin-RevId: 378681070
2021-06-10 10:27:22 -07:00
Chris Jones 968226b9d7 [XLA:GPU] Add AllReduce{Start,Done} to MLIR LHLO dialect.
PiperOrigin-RevId: 378640706
2021-06-10 06:54:42 -07:00
A. Unique TensorFlower d828b457b3 Handle empty tensors in SimplifyConcatSlice.
If the result of the slice is an empty tensor, do nothing.
This fixes a crash: we can't create a `concat` with an
empty operand range.

PiperOrigin-RevId: 378354956
2021-06-09 02:15:47 -07:00
Wenyi Zhao ade873a5e0 PR #49970: [MLIR][DISC] bufferize DynamicReshape and DynamicBroadcastInDim
Imported from GitHub PR https://github.com/tensorflow/tensorflow/pull/49970

1, add hlo-to-lhlo support for DynamicReshape and DynamicBroadcastInDim

2, add a flag `convert-to-lmhlo-only` to seperate following two case:
   - hlo-to-lhlo only. Simply lowers all mhlo ops to their lmhlo
     counterparts, do not apply any optimization (e.g. elide any
     buffer copy). Buffer optimization is not easy in dynamic
     shape world especially when involving control flow, thus we
     leave this to another dedicated pass.

   - hlo-to-lhlo-or-memref-directly. Lowers some metadata-only mhlo
     ops (e.g. reshape) to memref dialect directly and Lowers others
     to their lmhlo counterparts.
Copybara import of the project:

--
562bd65a368f6194405c4ae6900e3b4388a5ec03 by Wenyi Zhao <reyizero@gmail.com>:

[MLIR][DISC] bufferize DynamicReshape and DynamicBroadcastInDim

1, add hlo-to-lhlo support for DynamicReshape and DynamicBroadcastInDim

2, add a flag `convert-to-lmhlo-only` to seperate following two case:
   - hlo-to-lhlo only. Simply lowers all mhlo ops to their lmhlo
     counterparts, do not apply any optimization (e.g. elide any
     buffer copy). Buffer optimization is not easy in dynamic
     shape world especially when involving control flow, thus we
     leave this to another dedicated pass.

   - hlo-to-lhlo-or-memref-directly. Lowers some metadata-only mhlo
     ops (e.g. reshape) to memref dialect directly and Lowers others
     to their lmhlo counterparts.

PiperOrigin-RevId: 377603395
2021-06-04 15:36:03 -07:00
A. Unique TensorFlower aba16adfa5 Add `mhlo.all_gather` op to MHLO dialect.
Adds import/export/verifier support as well.
Also makes `channel_handle` uniform across mhlo.all_reduce and mhlo.all-gather.

PiperOrigin-RevId: 377323468
2021-06-03 10:45:29 -07:00
Adrian Kuegel a4fa6afa07 [mlir][hlo] Avoid dyn_cast_or_null when called with getDefiningOp result (NFC)
PiperOrigin-RevId: 376110457
2021-05-27 00:20:42 -07:00
wyzhao b93e54d8a4 PR #49454: [MLIR][DISC] Upgrade to use the new `reifyReturnTypeShapes` interface.
Imported from GitHub PR https://github.com/tensorflow/tensorflow/pull/49454

The new interface is more safe to be used during dialect conversion
(e.g. converting from tensor world to buffer world).
Copybara import of the project:

--
a6968072d59bec3c3bbaef0121d297e807c37c91 by Wenyi Zhao <reyizero@gmail.com>:

[MLIR][DISC] Upgrade to use the new `reifyReturnTypeShapes` interface.

The new interface is more safe to be used during dialect conversion
(e.g. converting from tensor world to buffer world).

--
55e7c6b7f2f99b99e226645a57e2433fae3e90ed by Wenyi Zhao <reyizero@gmail.com>:

minor fix

PiperOrigin-RevId: 375500273
2021-05-24 10:11:55 -07:00
Feiwen a7884196f5 PR #49228: [MLIR][DISC] porting dynamic shape related OPs to mhlo and lmhlo dialect
Imported from GitHub PR https://github.com/tensorflow/tensorflow/pull/49228

We are porting our MLIR-based dynamic shape compiler to tf community (From OP def, Patttern, to Optimization pass, etc).
This is the first PR, which including some dynamic shape OPs def in mhlo and lmhlo dialect.
For mhlo dialect, we add:
- HLO_RealDynamicSliceOp
- HLO_DynamicPadOp
- HLO_DynamicGatherOp
- HLO_DynamicConvOp

For lmhlo dialect, we add:
- LHLO_RealDynamicSliceOp
- LHLO_DynamicBroadcastInDimOp
- LHLO_DynamicGatherOp
- LHLO_DynamicPadOp
- LHLO_DynamicBitcastOp
- LHLO_DynamicConvOp
- LHLO_DynamicIotaOp
- LHLO_DynamicReshapeOp
- LHLO_DotGeneralOp
- LHLO_BitcastOp

Rest Ops to add:
* We will send a separate PR containing LHLO_DynamicWhileOp and LHLO_DynamicCaseOp for control flow.
* We will add a separate dedicated dialect like mhlo_ral, which including D2HOp/H2DOp/DebugPrintOp/TopKOp, etc.

Previous discussions:[RFC](https://groups.google.com/a/tensorflow.org/g/mlir/c/_X48poNcbDI/m/jCC8BWIICQAJ), [discussion_1](https://llvm.discourse.group/t/updates-on-mlir-based-dynamic-shape-compiler/2384), [Recording of meeting](https://drive.google.com/file/d/1_uEISlV5MUWdG9faKAdKlCWnPtGjRC-D/view?usp=sharing).
Copybara import of the project:

--
e22d9e61106e00a1a1c6f368cc4a03e3bd1f414c by azazhu <azazhu@gmail.com>:

[DISC]fea: porting mhlo and lmhlo OPs

--
9ec3e76290da07cbd53d7da5fa86ff67179441a1 by azazhu <azazhu@gmail.com>:

[DISC][MLIR] 1. add summary and description for dynamic OPs in mhlo and lmhlo; 2. rm InferOutputTypes; 3. add verify for RealDynamicSliceOp and DynamicPadOp

--
0d68cd135555fd935991c12456b21329e628f23f by azazhu <azazhu@gmail.com>:

[DISC][MLIR] 1.remove D2H,H2D and DebugPrint Ops from mhlo/lmhlo dialect; 2. add type constraint to DynamicPadOp and RealDynamicSliceOp; 3.refine lmhlo type constraint; 4.rename RealDynamicSliceOp as name conflict.

--
698762a77d60f6a844cb1ab3f32740d4ef3c5843 by azazhu <azazhu@gmail.com>:

[DISC][MLIR] 1. replace dyn_cast to cast 2. refine code

PiperOrigin-RevId: 375022260
2021-05-20 23:16:47 -07:00
Rahul Joshi 41f663ce47 [HLO] Adopt custom syntax for convolution dimensions and window attributes (HLO)
PiperOrigin-RevId: 374923250
2021-05-20 12:13:50 -07:00
Rahul Joshi fc88cf1ff4 [HLO] Adopt custom syntax for convolution dims and window attributes for LMHLO_GPU
PiperOrigin-RevId: 374889917
2021-05-20 09:41:48 -07:00
A. Unique TensorFlower 57aeb5ab16 Integrate LLVM at llvm/llvm-project@0316f3e649
Updates LLVM usage to match
[0316f3e64972](https://github.com/llvm/llvm-project/commit/0316f3e64972)

PiperOrigin-RevId: 374855085
2021-05-20 06:09:40 -07:00
Rahul Joshi a361253e4f [HLO] Add custom print/parse for window attributes of convolutions (in LMHLO)
PiperOrigin-RevId: 373807616
2021-05-14 09:47:25 -07:00
A. Unique TensorFlower d2cc74317c Implement constant folding for mhlo.Sign.
PiperOrigin-RevId: 373550014
2021-05-13 03:54:04 -07:00
A. Unique TensorFlower 420c42a0a1 [MLIR][HLO] Support CHLO unary operations in rank specialization clustering
PiperOrigin-RevId: 373397321
2021-05-12 10:20:43 -07:00
Rahul Joshi e260aa771c [HLO] Add custom print/parse for convolution dimension numbers (in LMHLO)
PiperOrigin-RevId: 373379227
2021-05-12 08:52:46 -07:00
A. Unique TensorFlower 7f7a86ad0d [MLIR][HLO] Implement `RegionBranchOpInterface` for rank specialization cluster
PiperOrigin-RevId: 373163196
2021-05-11 09:03:05 -07:00
A. Unique TensorFlower 96a47345cc [MLIR][HLO] Add `rank_specialization_cluster` op to CHLO
The operation will be used to cluster compatible operations that can be rank-
specialized collectively.

PiperOrigin-RevId: 373128557
2021-05-11 05:17:42 -07:00
A. Unique TensorFlower 7f86dd9f7e Constant fold compare EQ if one of the operands is true and compare NE if one of the operands is false.
PiperOrigin-RevId: 373058030
2021-05-10 18:53:49 -07:00
dfki-jugr 6bc854f5d9 PR #48667: [mlir-hlo] Added RegionBranchOpInterfaces to lmhlo operations.
Imported from GitHub PR https://github.com/tensorflow/tensorflow/pull/48667

Added RegionBranchOpInterfaces to lmhlo operations that use regions.
This is needed, since the bufferization features in MLIR have to reason about the control flow within these operations.
Copybara import of the project:

--
572fd7d850a46630b812da84e9094280f89f259e by Julian Gross <julian.gross@dfki.de>:

Added RegionBranchOpInterfaces to lmhlo operations.

PiperOrigin-RevId: 372070825
2021-05-05 00:27:56 -07:00
A. Unique TensorFlower e500ab37a1 Introduce constant folds for ReduceOp with single LogicalAnd or LogicalOr op.
PiperOrigin-RevId: 370551483
2021-04-26 15:11:27 -07:00
Adrian Kuegel 0e2b255f01 Lower LHLO::AbsOp to complex dialect.
Also fix the traits for LHLO::AbsOp to allow different types and add a
verifier.

PiperOrigin-RevId: 370438790
2021-04-26 05:44:03 -07:00
A. Unique TensorFlower 8db96f54d3 [mhlo] Add a folder for mhlo.map which does nothing but return one of the arguments.
Add a folder for maps whose body returns only one of the arguments. When this arises the fold replaces the map output with one of the operand tensors.

PiperOrigin-RevId: 369304322
2021-04-19 14:36:08 -07:00
Rahul Joshi c75cbf4ac7 [MLIR][NFC] Rename ReduceOp operands() => inputs().
- Rename to avoid confusion as operands generally includes all operands of an operation

PiperOrigin-RevId: 368479524
2021-04-14 12:08:23 -07:00
Jacques Pienaar fdd75daed6 Add shape function for MHLO RngNormal and RngUniform
PiperOrigin-RevId: 368276963
2021-04-13 12:59:42 -07:00
A. Unique TensorFlower 6d2209e301 [MLIR][HLO] Canonicalize chained broadcasts
Compose two subsequent `dynamic_broadcast_in_dim` ops into one.

PiperOrigin-RevId: 367630360
2021-04-09 07:35:34 -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 af3bc47a8b Integrate LLVM at llvm/llvm-project@8396aeb07c
Updates LLVM usage to match
[8396aeb07cdd](https://github.com/llvm/llvm-project/commit/8396aeb07cdd)

PiperOrigin-RevId: 366034463
2021-03-31 08:01:34 -07:00
Geoffrey Martin-Noble 5d65758e8c Canonicalize MHLO Case and If Ops with constant conditions
ReplaceOpWithRegion was taken directly from ScfOps. We should maybe put that somewhere common in core.

PiperOrigin-RevId: 365936724
2021-03-30 17:58:01 -07:00
Geoffrey Martin-Noble 2fb2a92c6e Verify mhlo.if region return types match op
This matches the behavior of mhlo.case. Additionally, fix the verification of CaseOp in the case of nested ops with mhlo.return-containing regions.

PiperOrigin-RevId: 365936672
2021-03-30 17:57:20 -07:00
Geoffrey Martin-Noble 7a9394dca5 Restrict MHLO control flow ops to single-block regions
PiperOrigin-RevId: 365935824
2021-03-30 17:51:03 -07:00
Geoffrey Martin-Noble a2b6060c0c Add folder for HLO NotOp
PiperOrigin-RevId: 364989658
2021-03-25 02:08:38 -07:00
A. Unique TensorFlower 0c4a89e52c [MLIR][MHLO] Implement shape reification for `dynamic_broadcast_in_dim`
PiperOrigin-RevId: 363622714
2021-03-18 03:39:15 -07:00
Jacques Pienaar a58e62590e Restrict canonicalization to avoid changing type
Issue #47516

PiperOrigin-RevId: 363300979
2021-03-16 16:54:05 -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
Jacques Pienaar 3de2024a9b Avoid creating tuple type only for verification
Make the error message a bit more verbose & it is cheaper to verify the elements rather than creating a (potentially) new type.

PiperOrigin-RevId: 363073909
2021-03-15 17:58:19 -07:00
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
Rahul Joshi 9902e6ee32 [HLO] Add LMHLO CollectivePermute verification.
- Extract verification of source target pairs attached to collective permute into a common
  helper function and use that to verify both MHLO and LMHLO variants.
- Change MlirGpuTestBase::ParseMlirModule to allow returning back a failure, and use
  that to update the mlir_gpu_compile_test to check the new behavior.

PiperOrigin-RevId: 362156962
2021-03-10 15:37:12 -08:00
Stephan Herhut cabd4d9a06 Canonicalize dynamic_broadcast_in_dim to own shape with rank narrowing on the shape to a corresponding tensor.cast.
PiperOrigin-RevId: 362028291
2021-03-10 05:43:54 -08:00
A. Unique TensorFlower 55eda81407 [MLIR][HLO] Reify shape extents as `index` values
PiperOrigin-RevId: 361519167
2021-03-08 02:42:47 -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
Adrian Kuegel e6a1f5f0f9 Add MinimumBroadcastShapesOp to chlo dialect.
This op is useful for rank specialization of broadcasts. Kernel Generator
needs to generate one kernel for each rank, so if we can minimize the rank
of the broadcast shape, we can support more cases with the same number of
special-cased kernels.

PiperOrigin-RevId: 360137827
2021-03-01 02:23:52 -08:00
Rahul Joshi 5adb7c6e12 [MLIR:LHLO] Add optional call target arg mapping to LMHLO CustomCall operations.
- XLA:HLO -> LMHLO conversion drops all token arguments and return values, however
  custom calls that users write still expect to get buffer pointers for these token types.
- To be able to support this, add an optional call target argument mapping attribute to
  LMHLO custom calls. When this attribute is present, it indicates the number of
  arguments and returns that the custom call expects and also indicates which LMHLO
  arg() or output() maps to which arg or result number of the custom call.

PiperOrigin-RevId: 358826664
2021-02-22 08:43:00 -08:00
Benjamin Kramer ca4034b56e [mlir][hlo] Make select ready for dynamic shapes (ranked only for now)
Move tf.SelectV2 broadcast lowering to a chlo.broadcast_select op, and lower it
to broadcasts on mhlo from there.

PiperOrigin-RevId: 358179975
2021-02-18 08:08:40 -08:00
Richard Uhler b579bd5d9e Support dynamic-shaped operand in verification of BroadcastInDim.
Verification of HLO_BroadcastInDimOp was previously failing or crashing if the
operand had a dynamic shape or was unranked. Update the verification code to
allow the operand to be unranked or have dynamic shape.

PiperOrigin-RevId: 358056793
2021-02-17 16:18:09 -08:00
A. Unique TensorFlower 81abaf364d [MLIR][MHLO] Add polygamma op to the CHLO dialect
PiperOrigin-RevId: 357724465
2021-02-16 08:32:33 -08:00
Stephan Herhut 6cd1875ee4 Implement lowering of chlo::zeta to mhlo dialect.
PiperOrigin-RevId: 355395581
2021-02-03 07:50:05 -08:00
Adrian Kuegel 96f8771ed7 Add MLIR generated kernel for Angle kernel.
This also requires a canonicalization pattern to remove a redundant dynamic
reshape from rank 1 to rank 1.

PiperOrigin-RevId: 355113135
2021-02-02 00:47:20 -08:00
Rahul Joshi 8e3890e8e8 [MLIR:HLO] Add AllGather and AllToAll operations to LMHLO dialect.
- Use a common base class to for AllReduce, AllGather, and AllToAll in the ODS spec.
- Add basic verification for replica groups attribute.

PiperOrigin-RevId: 354969654
2021-02-01 10:23:46 -08:00
Stephan Herhut e61ef86fdb Add zeta and broadcasting_zeta to chlo dialect.
PiperOrigin-RevId: 354500879
2021-01-29 03:22:52 -08:00