Cache object refs in loop partitioner instead of object pointers #6004
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
Loop partitioner modifies the IR, which can cause TIR objects to become dead and be destroyed. To avoid working on junk data cache object references instead of object pointers.
|
cc @ZihengJiang |
tqchen
approved these changes
Jul 8, 2020
|
Thanks @kparzysz-quic ! |
pingsutw
pushed a commit
to pingsutw/incubator-tvm
that referenced
this pull request
Jul 8, 2020
…he#5099) * [TOPI][Tensor Core] Optimization of CNNs on Tensor Core apache#6004 * update conv2d test * # pylint: dense_tensorcore.py * modify * modify conv2d * modify the unclear comment,add shape assertion in conv2d compute,combine general gemm intrinsic * add shape assertion in conv2d compute, combine general gemm intrinsic Co-authored-by: libaihong <[email protected]> Co-authored-by: libaihong <[email protected]>
trevor-m
pushed a commit
to trevor-m/tvm
that referenced
this pull request
Jul 14, 2020
…che#6004) * Cache object refs in loop partitioner instead of object pointers Loop partitioner modifies the IR, which can cause TIR objects to become dead and be destroyed. To avoid working on junk data cache object references instead of object pointers. * Fix format/lint errors
trevor-m
pushed a commit
to neo-ai/tvm
that referenced
this pull request
Jul 14, 2020
…che#6004) * Cache object refs in loop partitioner instead of object pointers Loop partitioner modifies the IR, which can cause TIR objects to become dead and be destroyed. To avoid working on junk data cache object references instead of object pointers. * Fix format/lint errors
kevinthesun
pushed a commit
to kevinthesun/tvm
that referenced
this pull request
Sep 17, 2020
…he#5099) * [TOPI][Tensor Core] Optimization of CNNs on Tensor Core apache#6004 * update conv2d test * # pylint: dense_tensorcore.py * modify * modify conv2d * modify the unclear comment,add shape assertion in conv2d compute,combine general gemm intrinsic * add shape assertion in conv2d compute, combine general gemm intrinsic Co-authored-by: libaihong <[email protected]> Co-authored-by: libaihong <[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.
Loop partitioner modifies the IR, which can cause TIR objects to become dead and be destroyed. To avoid working on junk data cache object references instead of object pointers.