Fix RhpCallCatchFunclet .text relocation on amd64. #111227
Open
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
dotnet@c8e4f2c changed asm implementation in RhpThrowHwEx on amd64 from a relative jump, using jmp RhpThrowHwEx to a mov rax, RhpThrowHwEx that is later jumped to using jmp rax. This introduce a ADDR64 relocation record in the generated object file: 000004EC ADDR64 00000000 00000000 2A RhpThrowHwEx That will then be left as a relocation that needs to be fixed up by the loader: 73000 RVA, C SizeOfBlock F7C DIR64 0000000140073A90 RhpThrowHwEx 0 ABS Where the RVA 73000 is part of the .text section that is read, execute section. Having relocations done to the .text section is however not supported on all Windows platform configurations leading to loader errors blocking use of nativeaot on these configurations. This commit change the mov to a lea instruction using relative offset to calculate address of RhpThrowHwEx. That will end up with a REL32 relocation record in the generated object file: 000004ED REL32 00000000 2A RhpThrowHwEx Since this uses relative addressing it won't leave any relocation in the final image making sure the generated nativeaot image can be successfully loaded on Windows platforms with configurations preventing relocation in .text section.
|
Tagging subscribers to this area: @agocke, @MichalStrehovsky, @jkotas |
|
/azp run runtime-nativeaot-outerloop |
|
Azure Pipelines successfully started running 1 pipeline(s). |
MichalStrehovsky
approved these changes
Jan 9, 2025
There was a problem hiding this comment.
LGTM assuming CET pipeline in outerloop is green (cc @VSadov in case there are some CET considerations in general). Thanks!
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.
c8e4f2c changed asm implementation in
RhpThrowHwExon amd64 from a relative jump, usingjmp RhpThrowHwExto amov rax, RhpThrowHwExthat is later jumped to usingjmp rax.This introduce an
ADDR64relocation record in the generated object file:000004EC ADDR64 00000000 00000000 2A RhpThrowHwExThat will then be left as a relocation that needs to be fixed up by the loader:
Where the RVA 73000 is part of the .text section that is read, execute section.
Having relocations done to the .text section is however not supported on all Windows platform configurations leading to loader errors blocking use of nativeaot on these configurations.
This commit change the mov to a
leainstruction using relative offset to calculate address ofRhpThrowHwEx. That will end up with aREL32relocation record in the generated object file:000004ED REL32 00000000 2A RhpThrowHwExSince this uses relative addressing it won't leave any relocation in the final image making sure the generated nativeaot image can be successfully loaded on Windows platforms configurations that prevents relocations to .text section.