Commit Graph

1264 Commits (bb47c166a047ddbc913950cb0f2571594c0fe261)
 

Author SHA1 Message Date
Ashay Rane bb47c166a0
llvm: update tag to 061e0189 (#1180)
Summary of changes:
 - Switch to C++17 (similar to https://reviews.llvm.org/D131348)
 - Update MHLO to build with LLVM commit hash 061e0189
 - Replace deprecated `hasValue()` and `getValue()` with `has_value()`
   and `value()` respectively (https://reviews.llvm.org/D131349)
 - Use `TypedAttr` (https://reviews.llvm.org/D130092)
 - Use updated assembly format of `mhlo.compare` op (commit
   d03ef01e70fbf9afd0fa1976fbb7ed31838929b3 in MHLO repo)
2022-08-08 20:17:35 -07:00
Henry Tu 3e97a33c80
Revert "Reenable LTC in out-of-tree build (#1177)" (#1183)
This reverts commit f85ae9c685.
2022-08-08 18:58:35 -07:00
武家伟 351f15424e
[MHLO] Add transposed convolution conversion pattern (#1171)
Co-authored-by: Bairen Yi <yibairen.byron@bytedance.com>
Co-authored-by: Jiawei Wu <xremold@gmail.com>
Co-authored-by: Tianyou Guo <tianyou.gty@alibaba-inc.com>
Co-authored-by: Xu Yan <yancey.yx@alibaba-inc.com>
Co-authored-by: Ziheng Jiang <ziheng.jiang@bytedance.com>
2022-08-09 09:50:07 +08:00
Sean Silva 504de5e701 Rework how global slot initializers work.
Rather than a per-global-slot initializer region, we now have one for
the whole module. For example, it might look like this:

```
torch.global_slot "private" @tensor : !torch.tensor
torch.global_slot "private" @list : !torch.list<tensor>
torch.global_slot.module_initializer {
  %0 = torch.tensor.literal(dense<0.0> : tensor<f32>) : !torch.tensor
  %1 = torch.prim.ListConstruct %0 : (!torch.tensor) -> !torch.list<tensor>
  torch.initialize.global_slots [
    @tensor(%0 : !torch.tensor)
    @list(%1 : !torch.list<tensor>)
  ]
}
```

This new structure allows GlobalizeObjectGraph to create the initializer in a
much simpler way, avoiding the need to reason about whether different slots
alias each other. Reasoning about whether slots alias each other now is the
responsibility of InlineGlobalSlots, which has to do a much more complicated
analysis, implemented using MLIR's dataflow analysis framework.

Recommended review order:
- Check out the new IR constructs in the .mlir files of various passes
- Op definitions (*.td)
- Changes to GlobalizeObjectGraph pass.
- InlineGlobalSlots pass (~total rewrite)
- Misc changes:
  - Moving torchMlirAdjustStaticInformation for sharing with C++ code.
  - EraseModuleInitializer pass

To make this a bit nicer, it would be good to have a `torch.module` op
with an initializer region attached. That would be more invasive though.

This change has highlighted certain aspects of our project layering
which are worth calling out. None of our backends can handle global
slots, so we enforce that there are no global slots before backend
lowering. At an earlier stage in the project, we had aspirations of
transparently handling mutable global state and such, but for reasons
described below, that is no longer a goal. So really global slots should
be seen as a progressive lowering step as part of inlining all the
IValue's in the original program (GlobalizeObjectGraph is also one such
step).

Over time, with insights from work like IREE-JAX, it has become clear
that there isn't a reliable programming model we can compile for users
where we just transparently handle mutable global state (and some other
things, like lists and dictionaries). There is a need for an "outer
program" that orchestrates more restricted subroutines of the kind we
can handle in our compile flow here. The benefit of that is that it
decouples considerations like shapes, dtypes, etc. from the program
constructs used in the outer program. As long as the outer program can
efficiently invoke (pipelining/async/etc.) high-performance
data-parallel numerical subroutines of the kind we compile in our flow
here, then there is a complete programming model. This is also
consistent with the direction of upstream PyTorch which is becoming more
tracing-based (which inherently loses a lot of program structure, which
then has to be applied back with an "outer program" orchestrating the
traced subroutines).
2022-08-08 18:12:06 -07:00
Vidush Singhal 34e207eeb5
E2E support for AtenRemainderScalarOp (#1119)
* E2E support for AtenRemainderScalarOp
2022-08-08 20:02:52 -04:00
Vidush Singhal b70548edff
Add decomposition and E2E support for Aten_EmbeddingBag (#1137)
* Add decomposition and E2E support for Aten_EmbeddingBag
2022-08-08 18:56:49 -04:00
Henry Tu f85ae9c685
Reenable LTC in out-of-tree build (#1177) 2022-08-08 17:35:22 -04:00
Tanyo Kwok 290d7755fb
importer: add initial support for loading Float16 tensors (#1169)
follow up #761:

    This patch updates the `torch_mlir::convertTensorToMlirElementsAttr()`
    method to enable the creation of tensors whose base type is Float16.
    This patch also adds a test to validate the IR generation, and it
    updates the test for importing tensors of various types.
2022-08-08 12:37:31 +08:00
Tanyo Kwok 1ee865983b
[MHLO] fix tensor mode aten.div op pattern (#1160)
* [MHLO] fix tensor mode aten.div op pattern

See RFC #999
Co-authored-by: Bairen Yi <yibairen.byron@bytedance.com>
Co-authored-by: Jiawei Wu <xremold@gmail.com>
Co-authored-by: Tianyou Guo <tianyou.gty@alibaba-inc.com>
Co-authored-by: Xu Yan <yancey.yx@alibaba-inc.com>
Co-authored-by: Ziheng Jiang <ziheng.jiang@bytedance.com>
2022-08-06 23:38:06 +08:00
Sean Silva 5618890ca0 development.md: Avoid name collisions with PYTORCH_ variables 2022-08-05 19:41:08 -07:00
Sean Silva 1fdaf2faa0 development.md: How to enable ASan 2022-08-05 17:37:09 -07:00
Henry Tu e322f6a878
Update LTC CMake hack documentation (#1155)
* Update CMakeLists.txt

* Update CMakeLists.txt

* Update CMakeLists.txt

* Update CMakeLists.txt

* Update buildAndTest.yml

* Update setup.py

* Address review comments
2022-08-05 14:12:20 -04:00
Sean Silva 8ce5d3f12c E2E framework: Report tensor dtype in summary
This helps to triage issues related to backends that don't support all
dtypes.
2022-08-05 10:05:18 -07:00
Vivek Khandelwal c129a6de93 [MLIR][TORCH] Add support for dim=None to Aten[Var|Std]DimOp
PyTorch recently added support for `dim=None` in the `torch.var`
(5ca9b2b6fa)
and `torch.std`op (eb0e30e0bc).
This commit adds the corresponding support in torch-mlir.

Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2022-08-05 20:28:56 +05:30
Sean Silva 31727f81d8 torch_mlir.compile: Allow ignoring traced shapes
In some cases, users know that a traced graph is valid for a wider set
of shapes than they originally traced it with. Provide an option for
users to ignore the shapes in the traced graph when they know it is
legal.

Fixes #997
2022-08-04 10:18:34 -07:00
Sean Silva 6484776a25 Make numerical stability test more perverse
To test the summation stability of `torch.aten.var`, add a large
constant to it, which increases the effective precision requirements.
2022-08-04 10:04:38 -07:00
武家伟 c94431f71c
[MHLO] Add convolution op pattern (#1152)
Co-authored-by: Bairen Yi <yibairen.byron@bytedance.com>
Co-authored-by: Jiawei Wu <xremold@gmail.com>
Co-authored-by: Tianyou Guo <tianyou.gty@alibaba-inc.com>
Co-authored-by: Xu Yan <yancey.yx@alibaba-inc.com>
Co-authored-by: Ziheng Jiang <ziheng.jiang@bytedance.com>
2022-08-04 00:41:35 -07:00
gpetters94 08fc2d89bb
Add non-unit groups support to aten.convolution (#858) 2022-08-04 02:18:38 -04:00
武家伟 d030591df9
[MHLO] Init MHLO pooling-like op conversion (#1141)
* [MHLO] Init MHLO pooling-like op conversion and remove 'op' suffix in filenames

Co-authored-by: Bairen Yi <yibairen.byron@bytedance.com>
Co-authored-by: Jiawei Wu <xremold@gmail.com>
Co-authored-by: Tianyou Guo tianyou.gty@alibaba-inc.com
Co-authored-by: Xu Yan <yancey.yx@alibaba-inc.com>
Co-authored-by: Ziheng Jiang <ziheng.jiang@bytedance.com>

See RFC #999
2022-08-04 12:34:22 +08:00
Tanyo Kwok f0a24f59f6
[MHLO] Init MHLO linear op patterns (#1132)
See RFC https://github.com/llvm/torch-mlir/issues/999

Co-authored-by: Bairen Yi yibairen.byron@bytedance.com
Co-authored-by: Jiawei Wu xremold@gmail.com
Co-authored-by: Tianyou Guo tianyou.gty@alibaba-inc.com
Co-authored-by: Xu Yan yancey.yx@alibaba-inc.com
Co-authored-by: Ziheng Jiang ziheng.jiang@bytedance.com
2022-08-03 19:10:54 -07:00
Ahmed S. Taei 48ec300586
[Fix bazel] Add recently added torch->mhlo conversion pass to bazel (#1148) 2022-08-03 14:12:07 -07:00
powderluv 37a229cffc
Update buildAndTest.yml (#1145) 2022-08-03 12:50:54 -07:00
powderluv 0d25b6f10e
Fix cache-suffix name bug (#1138)
This should enabling better caching of builds.
2022-08-03 07:53:01 -07:00
Vivek Khandelwal f2a0e32127 [MLIR][TORCH] Fix CI failure
This commit fixes the CI failure by temporarily adding the failing
test to xfail set.

Signed-Off By: Vivek Khandelwal<vivek@nod-labs.com>
2022-08-03 20:07:56 +05:30
武家伟 636f5acb10
[MHLO] Init MHLO reduce-like op conversion (#1133)
* [MHLO] init reduce-like op conversion from Torch to MHLO
Co-authored-by: Bairen Yi <yibairen.byron@bytedance.com>
Co-authored-by: Jiawei Wu <xremold@gmail.com>
Co-authored-by: Tianyou Guo <tianyou.gty@alibaba-inc.com>
Co-authored-by: Xu Yan <yancey.yx@alibaba-inc.com>
Co-authored-by: Ziheng Jiang <ziheng.jiang@bytedance.com>
2022-08-03 10:47:52 +08:00
Tanyo Kwok 0b23af27d3
[MHLO] support non-constant torch scalar in BasicOps (#1134)
See RFC https://github.com/llvm/torch-mlir/issues/999

Co-authored-by: Bairen Yi yibairen.byron@bytedance.com
Co-authored-by: Jiawei Wu xremold@gmail.com
Co-authored-by: Tianyou Guo tianyou.gty@alibaba-inc.com
Co-authored-by: Xu Yan yancey.yx@alibaba-inc.com
Co-authored-by: Ziheng Jiang ziheng.jiang@bytedance.com
2022-08-03 08:16:31 +08:00
Ahmed S. Taei 82af44da2f
Fix mhlo bazel rule name (#1136) 2022-08-02 12:28:53 -07:00
Ramiro Leal-Cavazos a7af1fd873
Add support for `dim=None` to `AtenMeanDimOp` (#1129)
PyTorch recently added support for `dim=None` in the `torch.mean`
op (2bfae07a79). This
commit adds the corresponding support in torch-mlir.
2022-08-02 16:08:06 +00:00
Quinn Dawkins 38d8498b21
add e2e support for aten.atan2 (#1117)
- Includes math-to-libm pass in refbackend for math::atan2 support
2022-08-02 11:39:41 -04:00
Yan Xu 704efdc259
[MHLO] add aten::gelu op pattern (#1127)
add aten::gelu op pattern, and moved some unit tests from basic.mlir to elementwise.mlir
2022-08-02 15:01:30 +08:00
武家伟 76c976682c
[MHLO] Support for dynamic shape in basic op conversion by introducing CHLO dialect (#1123)
* [MHLO] Support for dynamic shape in basic op conversion by introducing CHLO dialect
Co-authored-by: Bairen Yi <yibairen.byron@bytedance.com>
Co-authored-by: Jiawei Wu <xremold@gmail.com>
Co-authored-by: Tianyou Guo <tianyou.gty@alibaba-inc.com>
Co-authored-by: Xu Yan <yancey.yx@alibaba-inc.com>
Co-authored-by: Ziheng Jiang <ziheng.jiang@bytedance.com>

* [MHLO] Support I32 as shape tensor dtype

* [NFC] Add a 'TODO' annotation
2022-08-02 12:53:24 +08:00
Tanyo Kwok 3772e0bd91
[NFC][MHLO] move util funcs to MhloLegalizeUtils.h/cpp (#1128)
See RFC: https://github.com/llvm/torch-mlir/issues/999

Co-authored-by: Bairen Yi yibairen.byron@bytedance.com
Co-authored-by: Jiawei Wu xremold@gmail.com
Co-authored-by: Tianyou Guo tianyou.gty@alibaba-inc.com
Co-authored-by: Xu Yan yancey.yx@alibaba-inc.com
Co-authored-by: Ziheng Jiang ziheng.jiang@bytedance.com
2022-08-02 09:21:37 +08:00
Vidush Singhal fe3c9f5208
Add embedding Bag e2e case in xfail set (#1130) 2022-08-01 18:23:45 -04:00
Vidush Singhal ed13ebfd8d
E2E support for AtenEmbeddingBagPaddingIdxOp SUM Mode (#1066) 2022-08-01 16:44:11 -04:00
Alec 554570f3ab Implemented a decomposition of aten::narrow 2022-08-01 18:32:14 +05:30
powderluv 30017af5b5
Disable LTC on Release builds (#1125) 2022-07-31 20:58:29 -07:00
Ashay Rane 55bbbec144
llvm: update tag to 02b3a35 (#1124)
As part of this update, this patch also updates MHLO tag to ad54b43.
2022-07-31 15:26:24 -07:00
Henry Tu 2c3b3606d0 Resolve remaining LTC CI failures (#1110)
* Replace CHECK_EQ with TORCH_CHECK_EQ

* Check value of TORCH_MLIR_USE_INSTALLED_PYTORCH during LTC build

* Update LTC XFAIL with NewZerosModule ops

* Explicitly blacklist _like ops

* Automatically blacklist new_/_like ops

* Prune away unused Python dependencies from LTC

* Add flag to disable LTC

* Autogen dummy _REFERENCE_LAZY_BACKEND library when LTC is disabled

* Implement compute_shape_var

* Removed Var tests from XFAIL Set

* XFAIL tests using _local_scalar_dense or index.Tensor

* Add StdDim tests to XFAIL set

* Autogen aten::cat
2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim 425362263b Clean up Autogen (#1112)
* Remove unnecessary sed in autogen

* Remove .pyc files frrom VCS
2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim 368963243e Export LTC Headers (#1108) 2022-07-30 09:40:02 -04:00
Henry Tu 70395de197 Resolve CI testing failure for Lazy Tensor Core (#1088)
* Xfail unsupported ops

* Register FuncDialect

* Include dynamic_ir in build

* Code reformat

* Enable LTC tests for macOS and Source Build
2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim 0d16a91656 Add support for lift_fresh op (#1101) 2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim e37891b997 Default Device Ordinal API (#1079)
* Add default device ordinal API

* Fix reference backend
2022-07-30 09:40:02 -04:00
Antonio Kim de6c135dc3 Fix LTC autogen for CI with nightly PyTorch
- Update llvm-project pin to match main
2022-07-30 09:40:02 -04:00
Henry Tu cec74b8d37 Blacklist _convolution op (#1048)
* Blacklist _convolution op in LTC

* Removed duplicate Torch_AtenSelectScatterOp instance from autogen .td

* Removed duplicate Torch_AtenSliceScatterOp instance from autogen .td
2022-07-30 09:40:02 -04:00
Henry Tu 47bb38d180 Reference Lazy Backend (#1045)
* Changed Example MLIR backend to Reference MLIR backend

* Moved reference_ltc_backend into csrc

* Merged sys_utils.h

* Renamed reference_ltc_backend to reference_lazy_backend

* Addressed review comments

* Update docs with new library name

* Removed _REFERENCE_LAZY_BACKEND from .gitignore

* Added reference_lazy_backend to the TorchMLIRPythonModules dependency list

Fixed typo in `ltc_examples.md`

Missed instance where `ltc_backend` was used instead of `lazy_backend`.
2022-07-30 09:40:02 -04:00
Henry Tu f5acad8512 Prune xfail e2e LTC tests & fix bugs from functionalization pass (#1044)
- Pruned number of xfailed e2e LTC tests from 305 to 134
  - Reviewed every failure to ensure the error genuinely warrants an xfail
- Fixed bug where non-tensor outputs of LTC computation had `.to('cpu')` called, which caused a failure and inflated the xfail count
- Fixed bug with `HBC_basic` test where a constant tensor was created in its constructor without being declared as a buffer, which prevented the device from being updated when the parent `torch.nn.Module` got moved to the `lazy` device
  - Note that this test is still xfail'd due to some unsupported ops. Left a comment about some potential issues that may arise if it gets reenabled in the future
- Updated autogen `GeneratedTorchOps.td` to reflect the latest set of supported ops
- Renamed `aten.zero.functionalization` to `aten.zero` to reflect upstream PyTorch changes
2022-07-30 09:40:02 -04:00
Henry Tu 9de06f3ebd Update Torch MLIR readme 2022-07-30 09:40:02 -04:00
Henry Tu 61db88c9ec LTC Documentation (#1021)
* Create ltc_backend.md

* Added introduction and examples

* Added descriptions for files from autogen

* Added remaining file descriptions

* architecture overview

* Added subheadings to architecture section

* Description for implementing custom backend

* Add graphics

* Future expansion

* Remove "location" in architecture section

* Updated LTC in readme

* Remove extra space in example

* Fix typo

* Reworded vendor compilation process

* Address PR review comments

* Updated diagrams

* Add kernel registration to diagram

* Address PR comments

* Create separate ltc_examples file

* PyBind -> pybind
2022-07-30 09:40:02 -04:00
Henry Tu a3244b6a44 Update README.md to reflect LTC landing in main branch (#1000)
* Update README.md

* Update README.md

Co-authored-by: Jae Hoon (Antonio) Kim <17433012+antoniojkim@users.noreply.github.com>

* Update README.md

Co-authored-by: Jae Hoon (Antonio) Kim <17433012+antoniojkim@users.noreply.github.com>

* Update README.md

Co-authored-by: Jae Hoon (Antonio) Kim <17433012+antoniojkim@users.noreply.github.com>
2022-07-30 09:40:02 -04:00