New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[pytorch/tensorexpr] Create LLJIT instance with an ObjectLinkingLayer #103824
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/103824
Note: Links to docs will display an error until the docs builds have been completed. ✅ 3 Unrelated FailuresAs of commit 05fbe78: UNSTABLE - The following jobs failed but were likely due to flakiness present on trunk and has been marked as unstable:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Upstream LLVM switched LLJIT's default JIT linker for ELF/x86-64 to JITLink. This patch updates PytorchLLVMJITImpl to set ObjectLinkingLayer on LLJIT creation.
@davidberard98 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
@pytorchbot merge |
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
…pytorch#103824) - Upstream LLVM switched LLJIT's default JIT linker for ELF/x86-64 to JITLink: [commit](llvm/llvm-project@b92839c). This commit mandates clients to use JITLink plugins, following the example in "llvm/examples/OrcV2Examples/LLJITWithCustomObjectLinkingLayer". - Current change updates PytorchLLVMJITImpl to set ObjectLinkingLayer on LLJIT creation. - If setObjectLinkingLayerCreator not set, RTDyldObjectLinkingLayer will be constructed. This is currently causing "Symbols not found: [ llvm_orc_registerEHFrameSectionWrapper ]" error for tests in test_quantization.py when pytorch is built to use latest LLVM. Pull Request resolved: pytorch#103824 Approved by: https://github.com/jeffdaily, https://github.com/davidberard98 (cherry picked from commit d2d3394)
…pytorch#103824) - Upstream LLVM switched LLJIT's default JIT linker for ELF/x86-64 to JITLink: [commit](llvm/llvm-project@b92839c). This commit mandates clients to use JITLink plugins, following the example in "llvm/examples/OrcV2Examples/LLJITWithCustomObjectLinkingLayer". - Current change updates PytorchLLVMJITImpl to set ObjectLinkingLayer on LLJIT creation. - If setObjectLinkingLayerCreator not set, RTDyldObjectLinkingLayer will be constructed. This is currently causing "Symbols not found: [ llvm_orc_registerEHFrameSectionWrapper ]" error for tests in test_quantization.py when pytorch is built to use latest LLVM. Pull Request resolved: pytorch#103824 Approved by: https://github.com/jeffdaily, https://github.com/davidberard98 (cherry picked from commit d2d3394)
…pytorch#103824) - Upstream LLVM switched LLJIT's default JIT linker for ELF/x86-64 to JITLink: [commit](llvm/llvm-project@b92839c). This commit mandates clients to use JITLink plugins, following the example in "llvm/examples/OrcV2Examples/LLJITWithCustomObjectLinkingLayer". - Current change updates PytorchLLVMJITImpl to set ObjectLinkingLayer on LLJIT creation. - If setObjectLinkingLayerCreator not set, RTDyldObjectLinkingLayer will be constructed. This is currently causing "Symbols not found: [ llvm_orc_registerEHFrameSectionWrapper ]" error for tests in test_quantization.py when pytorch is built to use latest LLVM. Pull Request resolved: pytorch#103824 Approved by: https://github.com/jeffdaily, https://github.com/davidberard98 (cherry picked from commit d2d3394)
…pytorch#103824) - Upstream LLVM switched LLJIT's default JIT linker for ELF/x86-64 to JITLink: [commit](llvm/llvm-project@b92839c). This commit mandates clients to use JITLink plugins, following the example in "llvm/examples/OrcV2Examples/LLJITWithCustomObjectLinkingLayer". - Current change updates PytorchLLVMJITImpl to set ObjectLinkingLayer on LLJIT creation. - If setObjectLinkingLayerCreator not set, RTDyldObjectLinkingLayer will be constructed. This is currently causing "Symbols not found: [ llvm_orc_registerEHFrameSectionWrapper ]" error for tests in test_quantization.py when pytorch is built to use latest LLVM. Pull Request resolved: pytorch#103824 Approved by: https://github.com/jeffdaily, https://github.com/davidberard98 (cherry picked from commit d2d3394)
…pytorch#103824) - Upstream LLVM switched LLJIT's default JIT linker for ELF/x86-64 to JITLink: [commit](llvm/llvm-project@b92839c). This commit mandates clients to use JITLink plugins, following the example in "llvm/examples/OrcV2Examples/LLJITWithCustomObjectLinkingLayer". - Current change updates PytorchLLVMJITImpl to set ObjectLinkingLayer on LLJIT creation. - If setObjectLinkingLayerCreator not set, RTDyldObjectLinkingLayer will be constructed. This is currently causing "Symbols not found: [ llvm_orc_registerEHFrameSectionWrapper ]" error for tests in test_quantization.py when pytorch is built to use latest LLVM. Pull Request resolved: pytorch#103824 Approved by: https://github.com/jeffdaily, https://github.com/davidberard98 (cherry picked from commit d2d3394)
Upstream LLVM switched LLJIT's default JIT linker for ELF/x86-64 to JITLink: commit. This commit mandates clients to use JITLink plugins, following the example in "llvm/examples/OrcV2Examples/LLJITWithCustomObjectLinkingLayer".
Current change updates PytorchLLVMJITImpl to set ObjectLinkingLayer on LLJIT creation.
If setObjectLinkingLayerCreator not set, RTDyldObjectLinkingLayer will be constructed. This is currently causing "Symbols not found: [ llvm_orc_registerEHFrameSectionWrapper ]" error for tests in test_quantization.py when pytorch is built to use latest LLVM.
cc @EikanWang @jgong5 @wenzhe-nrv @sanchitintel @jeffdaily