-
Notifications
You must be signed in to change notification settings - Fork 4.8k
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
Fix missing zero-offset sequences and add checking #64805
Fix missing zero-offset sequences and add checking #64805
Conversation
Tagging subscribers to this area: @JulieLeeMSFT |
There is a typo in this change, but the proper version found two cases where we miss field sequences. So we'll leave this in the draft state until fixes land. |
e176e4b
to
35ed939
Compare
This is now blocked on #64501. Edit: no more. |
35ed939
to
27d922b
Compare
27d922b
to
ae6c6e5
Compare
Will assume OSX timeouts are not related (happening everywhere currently...). Diffs are cases where we were numbering a real-world array of @dotnet/jit-contrib |
ae6c6e5
to
7ce21ab
Compare
/azp run Antigen, Fuzzlyn |
Azure Pipelines successfully started running 2 pipeline(s). |
@SingleAccretion It looks like the fuzzlyn run hit an assert that was changed in #65387, can you investigate? |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, thanks!
Yes, of course. |
I believe Fuzzlyn/Antigen failures are preexisting. |
A couple changes:
ExtendPtrVN
(or equivalent) when necessary.ADDR(LCL_VAR)
(no tests because reproducing bad codegen for this issue is "hard", and the checker asserts anyway).NotAField
zero-offset sequences. These are important to preserve because they carry the fact we could have had an overlapping field in the sequence. Tests added.NotAField
sequences, to make the dumps more informative.Ref: #64773.