Windows python interop.h #1397
Closed
Windows python interop.h #1397
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
|
Oh hey @oroppas! Don't close this PR, since it still has useful changes! I just wanted to make sure that we could share the work involved. No need to apologize, and thanks for your contributions. They are much appreciated! |
|
My apologies, I forgot to answer your question about #1310 versus #1376. You're right that #1376 does not include the changes in #1310, so we should definitely leave #1310 open. However, are changes in this PR covered by #1376 and #1310? If so, we should close this PR. Sorry for not grasping that earlier. |
qedawkins
pushed a commit
to nod-ai/torch-mlir
that referenced
this pull request
Oct 3, 2022
According to https://llvm.org/docs/LangRef.html#alloca-instruction, the `alloca` instruction allocates memory on the stack frame of the currently executing function, to be automatically released when this function returns to its caller. We can avoid over allocing by placing `alloca` instruction at the function scope and outside of any loops. This change is motivated by onnx/onnx-mlir#1061, where models segfault at runtime when compiling at `-O0`, due to run out of stack space. Signed-off-by: Whitney Tsang <whitneyt@ca.ibm.com>
qedawkins
pushed a commit
to nod-ai/torch-mlir
that referenced
this pull request
Oct 3, 2022
According to https://llvm.org/docs/LangRef.html#alloca-instruction, the alloca instruction allocates memory on the stack frame of the currently executing function, to be automatically released when this function returns to its caller. We can avoid over allocing by placing alloca instruction at the function scope and outside of any loops. llvm#1397 and llvm#1437 attempted to create the allocas outside of loops directly. However, it is not always possible to create outside of the whole loop nest. For example, when lowering an operation that is already in a loop nest, or when the outer loop is only created after the placement of the alloca. This PR adds BufferLoopHoistingPass after LowerAffinePass to hoist allocations out of loop nests to avoid stack overflow. Signed-off-by: Whitney Tsang whitneyt@ca.ibm.com
powderluv
added a commit
to powderluv/iree
that referenced
this pull request
Oct 10, 2022
It breaks windows builds when it is included. Similar fix went into torch-mlir llvm/torch-mlir#1397
powderluv
added a commit
to powderluv/iree
that referenced
this pull request
Oct 10, 2022
It breaks windows builds when it is included. Similar fix went into torch-mlir llvm/torch-mlir#1397
hanhanW
pushed a commit
to iree-org/iree
that referenced
this pull request
Oct 14, 2022
It breaks windows builds when it is included. Similar fix went into torch-mlir llvm/torch-mlir#1397
PhaneeshB
pushed a commit
to PhaneeshB/iree
that referenced
this pull request
Oct 31, 2022
It breaks windows builds when it is included. Similar fix went into torch-mlir llvm/torch-mlir#1397
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.
Fixes #1301