Commit Graph

198 Commits (306ed62eddd3b806386d6495fb248bf4a9849802)

Author SHA1 Message Date
Maksim Levental ac1f03e6f7
add jit,no-jit release matrix (#1916) 2023-03-05 22:13:33 -08:00
Maksim Levental 415265a64c
Add `torch-mlir-no-jit-importer` build case for mac os wheels (#1902)
* add flags to setup.py for out-of-tree build

* - fix build_ext bug
- add wheels script cases for mac wheels
2023-03-05 12:23:43 -06:00
Ashay Rane 987d5ab335
CI: use `sudo` to remove Docker-created files (#1905) 2023-02-27 17:44:50 -06:00
Ashay Rane ea00371d85
CI: clear workspace directory before checkout (#1900)
We have recently started seeing errors like:

```
  Synchronizing submodule url for 'externals/llvm-project'
  Synchronizing submodule url for 'externals/mlir-hlo'
  /usr/bin/git -c protocol.version=2 submodule update --init --force --depth=1
  Error: fatal: Unable to create '/home/anush/actions-runner/_work/torch-mlir/torch-mlir/.git/modules/externals/llvm-project/index.lock': File exists.
```

As a workaround, this patch removes the workspace directory before the
checkout step.
2023-02-24 14:44:35 -06:00
Ashay Rane 268364e061
CI: install `unzip` before using it (#1893)
The RollPyTorch action needs the `unzip` command to peek into WHL files
for fetching metadata.  This patch makes sure that the command is
installed before referencing it.
2023-02-19 17:49:08 -06:00
powderluv 5710871f4f
Update buildAndTest.yml (#1881)
* Update buildAndTest.yml

* Update oneshotSnapshotPackage.yml

* Update buildRelease.yml

* Update RollPyTorch.yml

* Update oneshotSnapshotPackage.yml

* Update buildAndTest.yml
2023-02-15 09:17:12 -08:00
Ashay Rane 67ab708b63
python: separate build- and test-related pip dependencies (#1874)
We want to ensure that pip packages required for building torch-mlir
should be included in the dependencies of torch-mlir, but we don't want
the pip packages required for _testing_ of torch-mlir to be included
among the dependencies.  To be able to specify and install one set of
dependencies and not the other, this patch separates the pip packages
into two files: build-requirements.txt and test-requirements.txt.

This patch also updates references to the requirements.txt file so that
CI builds that run end-to-end tests install test-related pip
dependencies while everything else (including WHL builds) sticks to just
the build-related pip dependencies.

Despite this change, this patch should not affect a torch-mlir
developer's workflow.  More precisely, since this patch makes the
top-level requirements.txt file refer to both build-requirements.txt and
test-requirements.txt files, a torch-mlir developer should be able to
continue referring to the requirements.txt file without any impact.
2023-02-13 21:22:09 -06:00
powderluv 320e67ff34
Python 3.11 support (#1848)
* Python 3.11 support

* test without torchvision

* Update pytorch-requirements.txt

* Update buildRelease.yml

* Update action.yml

* Update install_macos_deps.sh

* Update build_macos_packages.sh
2023-02-10 07:16:37 -08:00
Ashay Rane 711646d095
mhlo: migrate conversion to stablehlo (#1840)
This patch replaces all MHLO operations with their StableHLO
counterparts and adds a validation pass to ensure that no MHLO operations
remain before translating all Stablehlo operations to the MHLO dialect
for further lowering to the Linalg dialect.

This patch also updates all lit tests so that they refer to the
`convert-torch-to-stablehlo` pass and so that they check for StableHLO
operations.
2023-02-02 07:29:47 -06:00
Ashay Rane a897c49803
CI: miscellaneous fixes for Release builds (#1781)
- Use v3 of actions/checkout, since the version we use (v2) uses
   Node.js 12, which is deprecated by GitHub.

 - Source the PowerShell venv sctipt (instead of the bash sript) since
   the calling script is a PowerShell script.  Without this, the build
   doesn't use venv at all.

 - Make the build dependencies in whl-requirements.txt (used by
   setup.py) match those in requirements.txt.  To that end, this patch
   creates a build-requirements.txt that is referenced by
   requirements.txt and whl-requirements.txt.
2023-01-06 20:41:43 -06:00
Ashay Rane f6b6069a34
ci: post comment on RollPyTorch tracker issue upon build failure (#1730)
Now that the RollPyTorch tracker issue exists, we can automate the job
of notifying folks of failures instead of having to do it manually.
This patch adds a step to the workflow to post such a message.
2022-12-18 13:45:30 -06:00
powderluv cd90c0aaf5
Update buildAndTest.yml (#1723) 2022-12-15 05:42:01 -08:00
Ashay Rane 64f9a0e978
ci: print ccache statistics and configuration at end of CI run (#1719)
There appear to be two problems with the caching layer in our CI runs:
(a) the sizes of some of the caches have grown to multiples of the
300 MB limit and (b) caching on Windows seems to be provide little to no
benefit.

To help understand the reasons for these problems, this patch adds a
line item to the list of steps run in CI to dump the ccache
configuration and statistics just prior to uploading the cache artifact.
2022-12-14 09:50:43 -06:00
Ashay Rane 731c313231
ci: run `git pull` before committing pytorch version updates (#1716)
The RollPyTorch action often takes more than 1.5 hours to finish.
During this time, if another PR is merged, then the RollPyTorch action
needs to first pull the merged changes before committing the updates to
the PyTorch commit hash and version files.  This patch adds the required
`git pull` statement, without which, the subsequent `git push` statement
fails, causing the RollPyTorch action to fail as well.
2022-12-13 13:41:41 -06:00
Daniel Ellis 07a65961dd
Disable pypi publishing.
See https://github.com/llvm/torch-mlir/issues/1709
2022-12-13 11:45:41 -05:00
Ramiro Leal-Cavazos a710237437
[custom op] Generalize shape library logic to work with dtypes (#1594)
* [custom op] Generalize shape library logic to work with dtypes

This commit generalizes the shape library logic, so that dtype rules
for ops can also be expressed using the same mechanism. In other
words, each op can now have a shape function and a dtype function
specified in Python that is imported during lowering to calculate the
shapes and dtypes throught a program. For more information about how
to specify a dtype function, see the updated
`docs/adding_a_shape_and_dtype_function.md`.

For those not familiar with how the shape library works, the file
`docs/calculations_lib.md` provides an overview.
2022-12-13 08:25:41 -08:00
Sambhav Jain 109c91ae9b
[CI] Verify bazel buildifier is run and changes committed (#1700)
Ensures the buildifier (linter for bazel build files) is run and changes are pushed.
2022-12-08 15:56:57 -08:00
Daniel Ellis 98d80a642a
Publish releases to PyPI after build 2022-12-07 10:01:55 -05:00
Ashay Rane b43965d8d3
build: fetch PyTorch version using downloaded WHL file (#1632)
Until recently, the metadata file in the torchvision package included
the nightly version of the torch package, but since that is no longer
the case, our RollPyTorch workflow is broken.

As a workaround, this patch uses the `pip download` command's ability to
fetch the dependent torch package for the specified version of
torchvision, before peeking into the WHL file for the torch package to
determine the release version and the commit hash.
2022-11-23 13:54:54 -06:00
Ashay Rane 4eead74232
ci: delay RollPyTorch action by 1 hour to use latest torchvision package (#1603)
The upload timestamp of the nightly torchvision package has drifted
beyond the scheduled time of the RollPyTorch action because of the time
change due to daylight saving.  As a result, the RollPyTorch action now
picks the torchvision package from a day earlier instead of the most
recent package.

This patch schedules the RollPyTorch action to start one hour later than
before so that it continues to pick the most recent nightly package.
2022-11-23 11:31:02 -06:00
Sambhav Jain ba5b90ee27
Enable bazel LIT tests in CI (#1596)
Bazel LIT test support was added in https://github.com/llvm/torch-mlir/pull/1585. This PR enables the tests in CI.

```
INFO: Build completed successfully, 254 total actions
@torch-mlir//test/Conversion:TorchToArith/basic.mlir.test                PASSED in 0.3s
@torch-mlir//test/Conversion:TorchToLinalg/basic.mlir.test               PASSED in 0.5s
@torch-mlir//test/Conversion:TorchToLinalg/elementwise.mlir.test         PASSED in 0.3s
@torch-mlir//test/Conversion:TorchToLinalg/flatten.mlir.test             PASSED in 0.3s
@torch-mlir//test/Conversion:TorchToLinalg/pooling.mlir.test             PASSED in 0.3s
@torch-mlir//test/Conversion:TorchToLinalg/unsqueeze.mlir.test           PASSED in 0.2s
@torch-mlir//test/Conversion:TorchToLinalg/view.mlir.test                PASSED in 0.3s
@torch-mlir//test/Conversion:TorchToMhlo/basic.mlir.test                 PASSED in 0.5s
@torch-mlir//test/Conversion:TorchToMhlo/elementwise.mlir.test           PASSED in 0.9s
@torch-mlir//test/Conversion:TorchToMhlo/gather.mlir.test                PASSED in 0.3s
@torch-mlir//test/Conversion:TorchToMhlo/linear.mlir.test                PASSED in 0.6s
@torch-mlir//test/Conversion:TorchToMhlo/pooling.mlir.test               PASSED in 0.3s
@torch-mlir//test/Conversion:TorchToMhlo/reduction.mlir.test             PASSED in 0.4s
@torch-mlir//test/Conversion:TorchToMhlo/view_like.mlir.test             PASSED in 0.6s
@torch-mlir//test/Conversion:TorchToSCF/basic.mlir.test                  PASSED in 0.2s
@torch-mlir//test/Conversion:TorchToTosa/basic.mlir.test                 PASSED in 1.1s
@torch-mlir//test/Dialect:Torch/GlobalizeObjectGraph/basic.mlir.test     PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/GlobalizeObjectGraph/error.mlir.test     PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/GlobalizeObjectGraph/free-functions.mlir.test PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/GlobalizeObjectGraph/initializers.mlir.test PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/GlobalizeObjectGraph/methods.mlir.test   PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/GlobalizeObjectGraph/module-uses-error.mlir.test PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/GlobalizeObjectGraph/module-uses.mlir.test PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/GlobalizeObjectGraph/multiple-instances-error.mlir.test PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/GlobalizeObjectGraph/multiple-instances-multiple-module-args.mlir.test PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/GlobalizeObjectGraph/multiple-instances.mlir.test PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/GlobalizeObjectGraph/submodules.mlir.test PASSED in 0.3s
@torch-mlir//test/Dialect:Torch/GlobalizeObjectGraph/visibility.mlir.test PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/adjust-calling-conventions.mlir.test     PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/canonicalize.mlir.test                   PASSED in 0.4s
@torch-mlir//test/Dialect:Torch/decompose-complex-ops-legal.mlir.test    PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/decompose-complex-ops.mlir.test          PASSED in 0.9s
@torch-mlir//test/Dialect:Torch/drop-shape-calculations.mlir.test        PASSED in 0.4s
@torch-mlir//test/Dialect:Torch/erase-module-initializer.mlir.test       PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/inline-global-slots-analysis.mlir.test   PASSED in 0.3s
@torch-mlir//test/Dialect:Torch/inline-global-slots-transform.mlir.test  PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/invalid.mlir.test                        PASSED in 0.4s
@torch-mlir//test/Dialect:Torch/lower-to-backend-contract-error.mlir.test PASSED in 17.3s
@torch-mlir//test/Dialect:Torch/maximize-value-semantics.mlir.test       PASSED in 0.3s
@torch-mlir//test/Dialect:Torch/ops.mlir.test                            PASSED in 0.3s
@torch-mlir//test/Dialect:Torch/prepare-for-globalize-object-graph.mlir.test PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/promote-types.mlir.test                  PASSED in 0.3s
@torch-mlir//test/Dialect:Torch/reduce-op-variants-error.mlir.test       PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/reduce-op-variants.mlir.test             PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/refine-public-return.mlir.test           PASSED in 0.2s
@torch-mlir//test/Dialect:Torch/refine-types-branch.mlir.test            PASSED in 0.3s
@torch-mlir//test/Dialect:Torch/refine-types-ops.mlir.test               PASSED in 0.6s
@torch-mlir//test/Dialect:Torch/refine-types.mlir.test                   PASSED in 0.4s
@torch-mlir//test/Dialect:Torch/reify-shape-calculations.mlir.test       PASSED in 2.9s
@torch-mlir//test/Dialect:Torch/simplify-shape-calculations.mlir.test    PASSED in 0.3s
@torch-mlir//test/Dialect:Torch/torch-function-to-torch-backend-pipeline.mlir.test PASSED in 0.6s
@torch-mlir//test/Dialect:TorchConversion/canonicalize.mlir.test         PASSED in 0.2s
@torch-mlir//test/Dialect:TorchConversion/finalizing-backend-type-conversion.mlir.test PASSED in 0.3s
@torch-mlir//test/Dialect:TorchConversion/func-backend-type-conversion.mlir.test PASSED in 0.2s
@torch-mlir//test/Dialect:TorchConversion/ops.mlir.test                  PASSED in 0.3s
@torch-mlir//test/Dialect:TorchConversion/verify-linalg-on-tensors-backend-contract.mlir.test PASSED in 0.3s
@torch-mlir//test/Dialect:TorchConversion/verify-tosa-backend-contract.mlir.test PASSED in 0.2s
@torch-mlir//test/RefBackend:insert-rng-globals.mlir.test                PASSED in 0.2s
INFO: Build completed successfully, 2[54](https://github.com/sjain-stanford/torch-mlir/actions/runs/3476816449/jobs/5812368489#step:7:55) total actions
@torch-mlir//test/RefBackend:munge-calling-conventions.mlir.test         PASSED in 0.2s

Executed [59](https://github.com/sjain-stanford/torch-mlir/actions/runs/3476816449/jobs/5812368489#step:7:60) out of 59 tests: 59 tests pass.
```

GHA workflow: https://github.com/sjain-stanford/torch-mlir/actions/runs/3476816449/jobs/5812368489
2022-11-16 11:59:33 -08:00
Sambhav Jain 4aa1e90b34
Fix cache bug with Bazel builds in CI (#1593)
Some time ago, bazel builds in CI were being sped up fine with caching. However, over time the cache got stale because `actions/cache@v3` apparently doesn't update caches when it "hits" unless it is configured to do so specifically. This requires using a uniqued per-commit `key` (to force it to update cache after each successful run) and a relaxed `restore-keys` which is not unique per-commit so newer commits can restore from the nearest hit.

Test GHA run 1 (no cache hit): [1h 1m 52s](https://github.com/sjain-stanford/torch-mlir/actions/runs/3474770334/usage)
Test GHA run 2 (cache hit, same commit): [5m 14s](https://github.com/sjain-stanford/torch-mlir/actions/runs/3475132135/usage)
Test GHA run 3 (cache hit, different commit): [6m 6s](https://github.com/sjain-stanford/torch-mlir/actions/runs/3475161009/usage)
2022-11-15 18:48:31 -08:00
Sambhav Jain 99ec6039f6
Fix bazel CI (#1591)
I accidentally broke bazel CI by forgetting to update the GHA workflow in my [previous PR](https://github.com/llvm/torch-mlir/pull/1587). This should get it back to green, my apologies.

Qualifying CI run: https://github.com/sjain-stanford/torch-mlir/actions/runs/3472523982
2022-11-15 09:51:52 -08:00
Ashay Rane f1ef5681cc
build: pin torchvision to latest nightly (#1584)
We currently pin the `torch` package to the latest nightly version, but
since `torchvision` depends on the `torch` package, the pip resolver
then has to run through an extensive list of `torchvision` packages that
can be installed with the pinned `torch` package.  This search fails in
the RollPyTorch action, causing pip to settle on an old version of
`torchvision` that does not work with our tests.  In reality, we are
only interested in a specific version of the `torchvision` package.

To make the dependency explicit and to prevent test failures because of
incorrect package installations, this patch makes two key changes:

1. `torchvision` is now pinned to the latest nightly release in
   pytorch-requirements.txt along with the version of `torch` that is
   necessary to install the requested `torchvision` package

2. The RollPyTorch action now looks for the latest `torchvision` package
   instead of the latest `torch` package before writing the version
   numbers for pinning in pytorch-requirements.txt
2022-11-14 15:56:02 -06:00
Ashay Rane 2846776897
ci: enable ccache on Windows (#1548)
This patch makes a few small, but key, changes to enable ccache on
Windows.  First, it replaces the hendrikmuhs/ccache-action action with
command line invocations to the ccache binary, since the action has two
bugs, one of which causes CI to refer to different ccache artifacts
before versus after the build on Windows whereas the other bug can
sometimes cause the action to incorrectly infer that the cache is empty.

Second, this patch slightly alters the cache key, so that our old cache
artifacts, which have grown too big, are eventually discarded in favor
of the new, smaller cache artifacts.  Along the way, this patch also
keeps the RollPyTorch's cache artifact separate from the regular build's
cache artifact so as to keep these artifacts small, and also because the
RollPyTorch action is off the critical path for most contributors.

Finally, this patch makes small changes to the CMake file so that on
Windows, the ccache binary is added as a prefix, as recommended on the
[ccache Wiki](https://github.com/ccache/ccache/wiki/MS-Visual-Studio).
2022-11-03 12:17:22 -05:00
Ashay Rane f847642495
CI script improvements (#1547)
* ci: update versions of external actions

Node.js 12 actions are deprecated and will eventually go away, so this
patch bumps the old actions to their latest versions that use Node.js
16.

* ci: replace deprecated action with bash commands

The llvm/actions/install-ninja action uses Node.js 12, which is
deprecated.  Since that action is not updated to work with Node.js 16,
this patch replaces that action with equivalent bash commands to install
Ninja.

* ci: use smaller ccache artifacts to reduce evictions

Over time, our ccache sizes have grown quite large (some as large as
1.3 GB), which results in us routinely exceeding GitHub's limits, thus
triggering frequent cache evictions.  As a result, cache downloads and
uploads take unnecessary long, in addition to fewer cache entries being
available.

Based on experiments on a clean cache state, it appears that we need
less than 300 MB of (compressed) ccache artifacts for each build type.
Anything larger than that will accrue changes from the past that aren't
needed.

To alleviate the cache burden, this patch sets the maximum ccache size
to be 300 MB.  This change should not affect the success or failure of
our builds.  I will monitor the build times to check whether this change
causes any performance degradation.

* ci: use consistent platform identifiers

Prior to this patch, some of our builds ran on `ubuntu-latest`, while
some others ran on `ubuntu-20.04` and others ran on `ubuntu-22.04`, with
similar situations for macOS and windows.  This patch instead sets all
Linux builds to run on `ubuntu-latest`, all macOS builds to run on
`macos-latest`, and all Windows builds to run on `windows-latest`, to
make debugging future CI failures a little easier.
2022-11-02 21:37:01 -05:00
Ashay Rane 031d127940
ci: introduce read-only and read-write PyTorch build caches (#1546)
Until recently, we had to either risk feature branches creating PyTorch
build caches (which were unusable by the main branch or other parallel
feature branches because of GitHub's rules around sharing caches among
branches) or we had to limit the PyTorch build caches to only the main
branch, causing CI runs on feature branches to be terribly slow because
they had to rebuild PyTorch each time.

This patch enables the best of both worlds, by using a fork
(github.com/ashay/cache) of the GitHub's cache action, where the fork
adds an option (called `save`) which, when set, uploads a new cache
entry.  We thus set this `save` flag only when we're building PyTorch
from source in Torch-MLIR's main branch, whereas all other builds set
this `save` flag to `false`.

The ability to conditionally update the cache has been an oft-requested
feature on the original (github.com/actions/cache) repository and
multiple unmerged PRs exist to allow conditional cache updates, so it is
likely that using the fork is only a temporary solution.
2022-11-01 23:26:17 -07:00
powderluv 1a33577860
remove spurious ref in publish pages (#1536)
We don't need to pass in optional tag information.
2022-11-01 09:42:21 -07:00
Ashay Rane a8970101dc
pytorch: rename pytorch-version.txt to pytorch-hash.txt (#1541)
This patch is part of a larger set of improvements to the CI/build
system.  In the code, we refer to the version as the string that
contains the release identifier such as 1.14.0.dev20221028, so calling
the file that contains the commit hash as pytorch-version.txt creates
confusion.  For the sake of simplicity, this patch renames that file to
be pytorch-hash.txt.
2022-10-31 22:03:05 -05:00
Ashay Rane 2cf1092d4d
ci: restrict PyTorch cache to just the main branch (#1540)
If PyTorch build caches are created on a branch other than the main
branch, then GitHub does not share those caches with the main branch,
making every CI run that runs for each PR slow.  This patch resolves the
problem by letting only the main branch create and use PyTorch build
caches.
2022-10-31 15:14:53 -05:00
powderluv 87ab714ed6
Update buildRelease.yml (#1535) 2022-10-30 00:14:54 -07:00
powderluv bbde4e163f
Add Windows Builder (#1521)
Add a powershell script to build windows .whl packages
Disable LTC as it doesn't build on Windows.
Add GHA hooks
Use Python 3.10.8
2022-10-25 16:13:31 -07:00
Ashay Rane 801452b2f4
ci: make RollPyTorch run only on the Torch-MLIR repo (#1516) 2022-10-25 17:56:59 -05:00
Ashay Rane a9942f343a
Cache PyTorch source builds to reduce CI time (#1500)
* ci: cache PyTorch source builds

This patch reduces the time spent in regular CI builds by caching
PyTorch source builds.  Specifically, this patch:

1. Makes CI lookup the cache entry for the PyTorch commit hash in
   pytorch-version.txt
2. If lookup was successful, CI fetches the previously-generated WHL
   file into the build_tools/python/wheelhouse directory
3. CI sets the `TM_PYTORCH_INSTALL_WITHOUT_REBUILD` variable to `true`
4. The build_libtorch.sh script then uses the downloaded WHL file
   instead of rebuilding PyTorch

* ci: warm up PyTorch source cache during daily RollPyTorch action

This patch makes the RollPyTorch action write the updated WHL file to
the cache, so that it can be later retrieved by CI that runs for each
PR.  We deliberately add the caching step to the end of the action since
the RollPyTorch action never needs to read from the cache, although
executing this step earlier in the process should not cause problems
either.
2022-10-18 00:42:42 -05:00
Ashay Rane 0374a6da4e
ci: re-enable auto execution of the RollPyTorch action (#1488)
Now that the RollPyTorch action seems to have become stable, this patch
enables that action to be run at around 4am Pacific Time every day.
2022-10-12 19:18:54 -05:00
Daniel Ellis 67a0fb14ef Fix build release workflow. 2022-10-11 15:19:53 -04:00
Daniel Ellis 2e0d806bf7 Publish releases page after both mac and linux builds.
Mac was finishing first, causing linux releases to be lagged a day behind.
2022-10-10 10:37:02 -04:00
Ashay Rane 8a8e779529
Disable auto-update of PyTorch version until CI script stabilizes (#1456)
Instead of letting the auto-update script either fail because of script
errors or letting it commit bad versions, this patch makes the update
process manual, for now.  Once the script stabilizes, I will its
re-enable periodic execution.
2022-10-04 03:02:44 -05:00
Ashay Rane da02390188
build: update ODS and shape library when updating PyTorch (#1450)
Updating the PyTorch version may break the Torch-MLIR build, as it did
recently, since the PyTorch update caused the shape library to change,
but the shape library was not updated in the commit for updating
PyTorch.

This patch introduces a new default-off environment variable to the
build_linux_packages.sh script called `TM_UPDATE_ODS_AND_SHAPE_LIB`
which instructs the script to run the update_torch_ods.sh and
update_shape_lib.sh scripts.

However, running these scripts requires an in-tree build and the tests
that run for an in-tree build of Torch-MLIR are more comprehensive than
those that run for an out-of-tree build, so this patch also swaps out
the out-of-tree build for an in-tree build.
2022-10-02 18:02:34 -05:00
Ashay Rane 005d40f4d7
build: check exit code without causing script to fail (#1447)
A bug in the CI script caused the entire script to fail if the exit code
of the command for comparing with the existing hash returned a non-zero
exit status.  The non-zero exit status for this comparison does not
imply failed execution, since it only indicates that the hash has
changed.
2022-10-02 16:04:26 -05:00
Ashay Rane b3345e69e2
build: miscellaneous performance improvements (#1443) 2022-09-30 12:47:43 -05:00
Ashay Rane cf41a2582e
Final changes necessary to auto-update PyTorch version (#1438)
* build: push directly from CI to main branch

This avoids the need to create, approve, and merge a separate PR, in
addition to avoiding unnecessary CI runs for the PyTorch version update.

* build: schedule cronjob to run RollPyTorch action

This patch schedules the RollPyTorch action to be run at noon UTC, which
roughly corresponds to 4am Pacific Time.  We pick this time since the
commit for PyTorch nightly releases are picked just after midnight
Pacific Time and the nightly release artifacts are produced in about 2
to 3 hours after the commit is picked.
2022-09-29 17:15:32 -05:00
powderluv da584fbb73
Update releases pages after release builds (#1432)
* Update buildRelease.yml

Update Releases right after a Release build.

* Move gh-page update after release builds

This removes the periodic update and updates after a release build.
2022-09-29 12:49:41 -07:00
Ashay Rane 8f608c048d
build: use Github Actions for creating PR (#1433) 2022-09-29 07:09:16 -05:00
Ashay Rane 53e76b8ab6
build: create RollPyTorch to update PyTorch version in Torch-MLIR (#1419)
This patch fetches the most recent nightly (binary) build of PyTorch,
before pinning it in pytorch-requirements.txt, which is referenced in
the top-level requirements.txt file.  This way, end users will continue
to be able to run `pip -r requirements.txt` without worrying whether
doing so will break their Torch-MLIR build.

This patch also fetches the git commit hash that corresponds to the
nightly release, and this hash is passed to the out-of-tree build so
that it can build PyTorch from source.

If we were to sort the torch versions as numbers (in the usual
descending order), then 1.9 appears before 1.13.  To fix this problem,
we use the `--version-sort` flag (along with `--reverse` for specifying
a descending order).  We also filter out lines that don't contain
version numbers by only considering lines that start with a digit.

As a matter of slight clarity, this patch renames the variable
`torch_from_src` to `torch_from_bin`, since that variable is initialized
to `TM_USE_PYTORCH_BINARY`.

Co-authored-by: powderluv <powderluv@users.noreply.github.com>
2022-09-28 15:38:30 -05:00
Daniel Ellis 1dfe5efe9e Create github action for creating pip-compatible releases index 2022-09-23 15:26:19 -04:00
powderluv e6528f701a
Move CIs to use docker builds (#1316)
* Move CIs to use docker builds

Now that #1234 has landed and anyone can run CI / Release builds locally move GHA to use the same flow.

* update names

* Update comments
2022-09-02 18:35:40 -07:00
powderluv 7769eb88f8
Set ccache logging to verbose temporarily (#1326)
This is to debug what is causing the exactly ccache look up failures etc.
2022-08-31 16:09:46 -07:00
Sean Silva e16b43e20b Remove "torchscript" association from the e2e framework.
We use it for more than TorchScript testing now. This is a purely
mechanical change to adjust some file paths to remove "torchscript".

The most perceptible change here is that now e2e tests are run with

```
./tools/e2e_test.sh
instead of:
./tools/torchscript_e2e_test.sh
```
2022-08-29 14:10:03 -07:00
Sean Silva bcccf41d96 Add CI for generated files.
This ensures that they are always up to date.

This also updates the shape lib to make the new CI actually pass :)
2022-08-29 12:07:16 -07:00
powderluv c0630da678
Disable LTC by default until upstream revert relands (#1303)
* Disable LTC by default until upstream revert relands

Tracked with the WIP https://github.com/llvm/torch-mlir/pull/1292

* Disable LTC e2e tests temporarily

* Update setup.py

Disable LTC in setup.py temporarily until upstream is fixed.
2022-08-28 19:11:40 -07:00
Tanyo Kwok 2374098d71
[MHLO] Init end to end unit tests (#1223) 2022-08-23 16:47:21 +08:00
Henry Tu ba17a4d6c0
Reenable LTC in out-of-tree build (for real this time) (#1205)
* Fix OOT LTC CI build failure

* Disable LTC during macOS package gen

* Add more details about static TorchMLIRJITIRImporter library
2022-08-19 15:25:00 -04:00
Sambhav Jain 114f48e96c
[Bazel] Check cache directory exists before changing owners (#1241)
This fixes a seeding issue with the [previous PR](https://github.com/llvm/torch-mlir/pull/1240) where bazel build's GHA cache is not present to begin with and one of the commands (chown) fails on it. Should get the Bazel build back to green.
2022-08-17 17:04:50 -07:00
Sambhav Jain 9c8b962720
Dockerize and Cache Bazel {Local, CI} Builds (#1240)
This PR adds:

- A minimal docker wrapper to the bazel GHA workflow to make it reproducible locally
- Bazel cache to speed up GHA workflows (down to ~5 minutes from ~40+minutes)

This is a no-op for non-bazel workflows and an incremental improvement.
2022-08-17 12:46:17 -07:00
Ashay Rane 606f4d2c0e
build: streamline options for enabling LTC and MHLO (#1221) 2022-08-12 23:49:28 -07:00
Sambhav Jain 34478ab1c7
[Build] Add concurrency groups to address long queue times (#1219)
We're seeing large CI queue times ([example](https://discord.com/channels/636084430946959380/742573221882364009/1007631811184164944)) especially with MacOS VMs on GHA. Part of the problem is follow-on commits to the same branch which trigger new runs while the previous runs are still in-progress, hogging on the scarce VMs.

This PR adds concurrency groups to the GHA workflow which ensures that only a single job or workflow using the same concurrency group will run at a time. This would cancel any in-progress jobs in the same github workflow and github ref (e.g. `refs/heads/main` or `refs/pull/<pr_number>/merge`).

As discussed on discord [thread](https://discord.com/channels/636084430946959380/1007787336848912386/1007787338895740928), once this lands we may have to closely monitor the workflows to see this didn't introduce unintended consequences. If so, we could either revert, or decide to selectively cancel particular runs (e.g. macos only which is the main bottleneck right now) instead of entire workflow.

This will also require some expectation management. As in, if you see an  on the main branch, it may not necessarily mean things broke, it could mean the run was killed by a more recent run. Making it a bit harder to traceback a failure to a commit in a sequence of commits (requiring to run those builds again).

Thanks @powderluv for the proposal and pointer to this! It should help with the scarce VMs on GHA and save on queue time. 

References:
* https://docs.github.com/en/actions/using-jobs/using-concurrency#example-only-cancel-in-progress-jobs-or-runs-for-the-current-workflow
* https://docs.github.com/en/actions/using-workflows/workflow-syntax-for-github-actions#example-only-cancel-in-progress-jobs-or-runs-for-the-current-workflow
2022-08-12 17:38:48 -07:00
Ashay Rane 1581d6a84c
build: fix typo in path (#1218)
When we renamed the directory containing submodules from `external` to
`externals`, we accidentally left the original name in the Github
workflow.  This patch fixes the problem.
2022-08-12 15:38:25 -07:00
Sambhav Jain aed0ec3a2c
Merge matrix runs to fail fast globally (#1216)
My earlier[ PR](https://github.com/llvm/torch-mlir/pull/1213) had (among other things) decoupled ubuntu and macos builds into separate matrix runs. This is not working well due to limited number of MacOS GHA VMs causing long queue times and backlog. There are two reasons causing this backlog: 

1. macos arm64 builds with pytorch source are getting erratically cancelled due to resource / network constraints. This is addressed with this: https://github.com/llvm/torch-mlir/pull/1215

> "macos-arm64 (in-tree, OFF) The hosted runner: GitHub Actions 3 lost communication with the server. Anything in your workflow that terminates the runner process, starves it for CPU/Memory, or blocks its network access can cause this error."

2. macos runs don't fail-fast when ubuntu runs fail due to being in separate matrix setups. This PR couples them again.
2022-08-12 11:30:09 -07:00
Sambhav Jain b8bd0a46cc
use pytorch binary for macos-arm64 builds (#1215) 2022-08-12 06:33:57 -07:00
Sambhav Jain f00ca91db0
Simplify matrix configuration for CI workflows (#1213)
Addresses https://github.com/llvm/torch-mlir/issues/1207. 

#### Provisioned jobs:
```
# ubuntu - x86_64 - llvm in-tree     - pytorch binary - build+test    # most used dev flow and fastest signal
# ubuntu - x86_64 - llvm out-of-tree - pytorch source - build+test    # most elaborate build
# macos  - arm64  - llvm in-tree     - pytorch source - build only    # cross compile, can't test arm64
```

#### Main changes
- [x] Spawn macos builds from a separate matrix (in the same workflow). It made sense to do this as they are fairly different from ubuntu (cross compile, use a different cmake configuration). This simplifies the matrix configuration and exclusions quite a bit, and makes the workflow a bit more tractable and maintenance friendly.
- [x] Remove the submodule md5sum step for ccache config. This was [broken](https://github.com/llvm/torch-mlir/runs/7779288734?check_suite_focus=true#step:3:145) for a while now.
- [x] Removes unused matrix options - `os`, `targetarch`, `python-version`, `llvmtype`.
- [x] Address ZSTD [comment](https://github.com/llvm/torch-mlir/pull/1204#discussion_r942349282) on @powderluv's cross compile [PR](https://github.com/llvm/torch-mlir/pull/1204). 

#### Further improvements (to be addressed in follow-on):
* ubuntu-x86_64 out-of-tree integration tests fail ([error](https://github.com/sjain-stanford/torch-mlir/runs/7781264029?check_suite_focus=true)); only run unit tests for now (tests are excluded in current CI too)

#### Passing workflow:
https://github.com/sjain-stanford/torch-mlir/actions/runs/2840676309
![image](https://user-images.githubusercontent.com/19234106/184194535-f3807991-401a-4cb9-b030-0ee8c334eba3.png)
2022-08-11 16:35:15 -07:00
powderluv 2342456356
mac m1 cross compile (#1204)
* mac m1 cross compile

Add support for M1 cross compile

* Remove redundant ExecutionEngine

It is registered as part of RegisterEverything

* nuke non-universal zstd

disable LTC
2022-08-10 08:48:39 -07:00
powderluv 9cf0b6e8ff
Disable out-of-tree and PyTorch binary (#1206) 2022-08-09 18:18:12 -07:00
Sambhav Jain d41c7becf5
[Bazel] Allow workflow_dispatch manual trigger on bazel workflow (#1203)
At the moment we don't gate torch-mlir PRs with bazel builds. This means bazel builds don't get run on open PRs, and so there's no good way to validate a fix PR which is meant to fix a broken bazel build. This option allows a bazel build to be manually triggered as needed on open PRs.
2022-08-09 13:28:21 -07:00
Sambhav Jain b696362b7d
Enable OOT builds in CI (#1188) 2022-08-09 12:13:16 -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
Henry Tu f85ae9c685
Reenable LTC in out-of-tree build (#1177) 2022-08-08 17:35:22 -04:00
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
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
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
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 dfcc26556a Added e2e LTC tests (#916)
* Added e2e LTC Torch MLIR tests

* Fix seed for reproducability

* Check if computation is None before getting debug string

* Updated unit tests, and added numeric tests

* Print name of the model layer that fails numeric validation

* Run LTC e2e test with CI/CD

* Set seed in main function, instead of beginning of execution

* Add comment to specify number of digits of precision

* Fixed typo

* Remove tests for LTC example models

* Added LTC option to torchscript e2e

* Implement compile and run for LTC e2e test

* xfail all tests that use ops that aren't currently supported
2022-07-30 09:40:02 -04:00
Jae Hoon (Antonio) Kim 2f22e2ef40 Add initial LTC backend (#610)
* Add initial LTC backend skeleton

* Disable CI build and move TorchMLIRPyTorch.cmake
2022-07-30 09:40:02 -04:00
powderluv db4a6991a0
buildAndTest.yml for matrix builds (#1098)
* Update buildAndTest.yml

test with fast-fail matrix builds

* Remove redundant and statement

* Downgrade to 20.04

Until upstream PyTorch FBGEMM is fixed to compile with clang+14+ https://github.com/pytorch/pytorch/pull/82396

* Update buildAndTest.yml

run tests on only the binary config.
2022-07-29 10:52:46 -07:00
powderluv 31fd812acf
Add linux and macOS source builds in CI (#1070)
This enables building Pytorch from source in the CI.
The build should mostly hit the ccache.
Release builds will follow once we have some runtime on the CI.
2022-07-21 14:16:03 -07:00
Ziheng Jiang c61c99e887
[MHLO] Init MHLO integration. (#1083)
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>
2022-07-20 16:18:16 -07:00
powderluv a1947c7bd1
Update oneshotSnapshotPackage.yml 2022-07-02 10:00:52 -07:00
powderluv 2f0b1d0b08
bump macOS builds to Python 3.10 2022-06-04 22:44:32 -07:00
powderluv b14c5d619d
Build the nightly package only once a day/night
No need to be shipping two releases a day, our supported packages and binaries have grown.
2022-06-04 22:40:53 -07:00
Maksim Levental cec5aeedb0
add ci tests (#754) 2022-05-25 14:59:59 -05:00
powderluv cfc1a6515c
build only Python3.9 to avoid timeout
GH runner times out when building 3.9 and 3.10 on macOS.
2022-05-13 00:07:55 -07:00
powderluv 2877a37ac6
Update buildRelease.yml
Fix filename changed missed in Code Review.
2022-04-25 17:00:31 -07:00
powderluv 6d09c98b2f
Fix version information in Release builds (#788)
env vars seems to be lost in manylinux docker.
Use a version file like IREE does.
2022-04-25 14:13:17 -07:00
Ahmed S. Taei 6b3d0b7e7a
Add bazel build support (2/N) (#744)
- Add bazel GitHub actions.
2022-04-25 12:33:15 -07:00
powderluv 0f751498a7
Update releaseSnapshotPackage.yml 2022-04-22 15:38:36 -07:00
powderluv d789aee11e
Only upload torch*.whl (#786)
only upload torch*.whl to unblock OSX build failures during upload. We have to move to svenstaro/upload-release-action
2022-04-22 15:17:09 -07:00
powderluv cbf158f069
Update buildRelease.yml
Update artifact directory to ./build_tools/python_deploy/wheelhouse/*.whl
2022-04-21 19:57:27 -07:00
powderluv 9f2184da98
Update oneshotSnapshotPackage.yml
remove now deprecated inputs to build and test
2022-04-21 19:12:42 -07:00
powderluv 8003b92fa7
Delete releasePackage.yml 2022-04-21 18:54:01 -07:00
powderluv c1026fa95b
Switch to using the new Release builds (#780) 2022-04-21 18:46:34 -07:00
powderluv 4ef61aa27f
Minor buildsystem fixes (#778)
Sets up auto-pinning of latest torch-nightly
2022-04-21 15:53:00 -07:00
powderluv 0257d91a21
Update buildManylinux.yml
use sudo for mac OS
2022-04-21 11:06:02 -07:00
powderluv 299c1bbe6d
Update buildManylinux.yml
fix build naming
2022-04-21 10:55:40 -07:00
powderluv b03eac4224
Enable OSX (Intel, Apple Silicon Builds) (#776)
Update pinned pytorch version. Will submit a follow on PR to bump.
Also update artifacts directory
2022-04-21 10:47:28 -07:00
powderluv cc3a4a58ef
Add oneshot release snapshot for test/ondemand (#768)
* Add oneshot release snapshot for test/ondemand

Add some build scripts to test new release flow based on IREE.
Wont affect current builds, once this works well we can plumb it
in.

Build with manylinux docker

* Fixes a few issues found when debugging powderluv's setup.

* It is optional to link against Python3_LIBRARIES. Check that and don't do it if they don't exist for this config.
* Clean and auditwheel need to operate on sanitized package names. So "torch_mlir" vs "torch-mlir".
* Adds a pyproject.toml file that pins the build dependencies needed to detect both Torch and Python (the MLIR Python build was failing to detect because Numpy wasn't in the pip venv).
* Commented out auditwheel: These wheels are not PyPi compliant since they weak link to libtorch at runtime. However, they should be fine to deploy to users.
* Adds the --extra-index-url to the pip wheel command, allowing PyTorch to be found.
* Hack setup.py to remove the _mlir_libs dir before building. This keeps back-to-back versions from accumulating in the wheels for subsequent versions. IREE has a more principled way of doing this, but what I have here should work.

Co-authored-by: Stella Laurenzo <stellaraccident@gmail.com>
2022-04-21 02:19:12 -07:00
Clément Fournier 578d0ec292 Review comments 2022-04-19 15:11:17 -07:00
Clément Fournier 3e0c1cf6af Change cache suffix to not invalidate existing caches 2022-04-19 15:11:17 -07:00
Clément Fournier 566650c5ae Use distinct ccaches
Since they run in distinct jobs, using the same ccache would
cause one job to overwrite the cache of the other.

See https://github.com/ljfitz/torch-mlir/pull/16 for a proof
that this works. The first build takes a long time but ccache
takes over in the dummy commit.
2022-04-19 15:11:17 -07:00
Clément Fournier f9d5201ae6 address PR review 2022-04-19 15:11:17 -07:00