Added JIT to MLIR lowering #724
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
henrytwo
force-pushed
the
jit_to_mlir_lowering
branch
from
7114ed3 to
79b183d
Compare
silvasean
approved these changes
Mar 31, 2022
There was a problem hiding this comment.
what's your general testing strategy for the backend? Are you going to write unittests with FileCheck? something like this perhaps? https://github.com/llvm/torch-mlir/blob/main/test/python/importer/jit_ir/node_import/prim.py
I don't think we have solid plans for writing tests yet, but I think |
ke1337
approved these changes
Apr 1, 2022
antoniojkim
requested changes
Apr 1, 2022
python/torch_mlir/csrc/base_lazy_backend/mlir_lowering_context.cpp
Outdated
Show resolved
Hide resolved
henrytwo
force-pushed
the
jit_to_mlir_lowering
branch
from
2de2f11 to
4b83a47
Compare
antoniojkim
approved these changes
Apr 1, 2022
henrytwo
force-pushed
the
jit_to_mlir_lowering
branch
2 times, most recently
from
f2c6e82 to
9a41d6c
Compare
Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a jit::Graph is constructed, it gets converted to a jit::Function, which is fed into the existing utility to generate an MlirModule in torch-mlir.
henrytwo
force-pushed
the
jit_to_mlir_lowering
branch
from
9a41d6c to
f882273
Compare
henrytwo
force-pushed
the
jit_to_mlir_lowering
branch
from
f882273 to
c802a94
Compare
antoniojkim
pushed a commit
that referenced
this pull request
May 26, 2022
* Added JIT to MLIR lowering Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a jit::Graph is constructed, it gets converted to a jit::Function, which is fed into the existing utility to generate an MlirModule in torch-mlir. * Renamed `csrc/backend` to `csrc/base_lazy_backend`
antoniojkim
pushed a commit
that referenced
this pull request
Jun 30, 2022
* Added JIT to MLIR lowering Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a jit::Graph is constructed, it gets converted to a jit::Function, which is fed into the existing utility to generate an MlirModule in torch-mlir. * Renamed `csrc/backend` to `csrc/base_lazy_backend`
antoniojkim
pushed a commit
that referenced
this pull request
Jun 30, 2022
* Added JIT to MLIR lowering Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a jit::Graph is constructed, it gets converted to a jit::Function, which is fed into the existing utility to generate an MlirModule in torch-mlir. * Renamed `csrc/backend` to `csrc/base_lazy_backend`
Merged
antoniojkim
pushed a commit
that referenced
this pull request
Jul 5, 2022
* Added JIT to MLIR lowering Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a jit::Graph is constructed, it gets converted to a jit::Function, which is fed into the existing utility to generate an MlirModule in torch-mlir. * Renamed `csrc/backend` to `csrc/base_lazy_backend`
antoniojkim
pushed a commit
that referenced
this pull request
Jul 7, 2022
* Added JIT to MLIR lowering Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a jit::Graph is constructed, it gets converted to a jit::Function, which is fed into the existing utility to generate an MlirModule in torch-mlir. * Renamed `csrc/backend` to `csrc/base_lazy_backend`
henrytwo
added a commit
that referenced
this pull request
Jul 8, 2022
* Added JIT to MLIR lowering Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a jit::Graph is constructed, it gets converted to a jit::Function, which is fed into the existing utility to generate an MlirModule in torch-mlir. * Renamed `csrc/backend` to `csrc/base_lazy_backend`
henrytwo
added a commit
that referenced
this pull request
Jul 8, 2022
* Added JIT to MLIR lowering Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a jit::Graph is constructed, it gets converted to a jit::Function, which is fed into the existing utility to generate an MlirModule in torch-mlir. * Renamed `csrc/backend` to `csrc/base_lazy_backend`
henrytwo
added a commit
that referenced
this pull request
Jul 12, 2022
* Added JIT to MLIR lowering Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a jit::Graph is constructed, it gets converted to a jit::Function, which is fed into the existing utility to generate an MlirModule in torch-mlir. * Renamed `csrc/backend` to `csrc/base_lazy_backend`
antoniojkim
pushed a commit
that referenced
this pull request
Jul 15, 2022
* Added JIT to MLIR lowering Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a jit::Graph is constructed, it gets converted to a jit::Function, which is fed into the existing utility to generate an MlirModule in torch-mlir. * Renamed `csrc/backend` to `csrc/base_lazy_backend`
antoniojkim
pushed a commit
that referenced
this pull request
Jul 19, 2022
* Added JIT to MLIR lowering Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a jit::Graph is constructed, it gets converted to a jit::Function, which is fed into the existing utility to generate an MlirModule in torch-mlir. * Renamed `csrc/backend` to `csrc/base_lazy_backend`
antoniojkim
pushed a commit
that referenced
this pull request
Jul 22, 2022
* Added JIT to MLIR lowering Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a jit::Graph is constructed, it gets converted to a jit::Function, which is fed into the existing utility to generate an MlirModule in torch-mlir. * Renamed `csrc/backend` to `csrc/base_lazy_backend`
henrytwo
added a commit
that referenced
this pull request
Jul 29, 2022
* Added JIT to MLIR lowering Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a jit::Graph is constructed, it gets converted to a jit::Function, which is fed into the existing utility to generate an MlirModule in torch-mlir. * Renamed `csrc/backend` to `csrc/base_lazy_backend`
henrytwo
added a commit
that referenced
this pull request
Jul 29, 2022
* Added JIT to MLIR lowering Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a jit::Graph is constructed, it gets converted to a jit::Function, which is fed into the existing utility to generate an MlirModule in torch-mlir. * Renamed `csrc/backend` to `csrc/base_lazy_backend`
henrytwo
added a commit
that referenced
this pull request
Jul 30, 2022
* Added JIT to MLIR lowering Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a jit::Graph is constructed, it gets converted to a jit::Function, which is fed into the existing utility to generate an MlirModule in torch-mlir. * Renamed `csrc/backend` to `csrc/base_lazy_backend`
qedawkins
pushed a commit
to nod-ai/torch-mlir
that referenced
this pull request
Oct 3, 2022
* Build JNI on Windows This change makes JNI buildable on Windows. It also cleans up the logic for determining when to build JNI and adds the necessary dependency to MainUtils Signed-off-by: Stella Stamenova <[email protected]> * Undo a change made by clang-format locally because clang-format in the builds is complaining about it Signed-off-by: Stella Stamenova <[email protected]> * Check for null Signed-off-by: Stella Stamenova <[email protected]> * Make clang-format happy Signed-off-by: Stella Stamenova <[email protected]> Co-authored-by: Kevin O'Brien <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR adds the ability to generate MLIR using a custom LTC backend.
Lowering to JIT is performed in a way similar to how it's done in the TS LTC backend. After a
jit::Graphis constructed, it gets converted to ajit::Function, which is fed into the existing utility to generate an MLIR in torch-mlir.Note: This PR is based on #723, but GitHub does not allow for setting the base to a fork while keeping this PR in the upstream repo, so there are several shared commits. Only the last one is specifically part of this PR.
A future PR will include an example implementation and usage of the custom backend.
Marked as draft while waiting for a dependency PR to land: #723
cc: @antoniojkim @ke1337