Skip to content

[WIP]: try fixing self-referential initializers from Sema #85000

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.

Sign up for GitHub

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

Jump to bottom
Closed
wants to merge 3 commits into from
Closed

[WIP]: try fixing self-referential initializers from Sema #85000

wants to merge 3 commits into from

Conversation

@jamieQ
Copy link
Contributor

@jamieQ jamieQ commented Oct 19, 2025

like #84951 but with diagnosis attempted in Sema rather than SILGen or DI

jamieQ added 3 commits October 16, 2025 06:08
@jamieQ
[SILGen]: ensure DI checks address-only let temporary allocations
d0e519f 
Previously it was possible to reference uninitialized memory if a
closure that was part of a variable binding initializer referenced the
name of the uninitialized value. DI was ignoring these cases before
because no mark_uninitialized instruction was produced. Instead of
trying to skip emitting that instruction sometimes, just always emit it
if a temporary allocation is used so DI will catch liveness issues.
@jamieQ
[WIP]: try to fix things & add many debug logs
681e226
@jamieQ
try fixing from Sema instead
e0f16ca
@jamieQ
Copy link
Contributor Author

jamieQ commented Oct 19, 2025

@swift-ci please smoke test macos

@jamieQ jamieQ mentioned this pull request Oct 19, 2025
Open
slavapestov
slavapestov reviewed Oct 20, 2025
View reviewed changes
Copy link
Contributor

@slavapestov slavapestov left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Detecting these problems in Sema might be difficult because of how local functions can forward-reference captures (and local functions can also be mutually recursive), but if you think you can sort it out, go ahead :)

@jamieQ
Copy link
Contributor Author

jamieQ commented Oct 20, 2025

Detecting these problems in Sema might be difficult because of how local functions can forward-reference captures (and local functions can also be mutually recursive), but if you think you can sort it out, go ahead :)

ugh, i did not realize this was allowed:

func weird(_ a: Any) {
    func local() -> Any { bad }
    let bad = local()
    print("bad: \(bad)")
}

@jamieQ jamieQ closed this Oct 23, 2025
@jamieQ jamieQ deleted the fix-address-only-DI-Sema branch October 23, 2025 01:48
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@slavapestov slavapestov slavapestov left review comments

@hborla hborla Awaiting requested review from hborla hborla will be requested when the pull request is marked ready for review hborla is a code owner

@xedin xedin Awaiting requested review from xedin xedin will be requested when the pull request is marked ready for review xedin is a code owner

@eeckstein eeckstein Awaiting requested review from eeckstein eeckstein will be requested when the pull request is marked ready for review eeckstein is a code owner

@jckarter jckarter Awaiting requested review from jckarter jckarter will be requested when the pull request is marked ready for review jckarter is a code owner

@kavon kavon Awaiting requested review from kavon kavon will be requested when the pull request is marked ready for review kavon is a code owner

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@jamieQ @slavapestov