A testcase for iwyu crash while analyzing llvm 7+ #601
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
|
No surprise checks are failing. |
|
A link to the clang change. |
|
I ran this crash through a debugger, and this actually seems to hit So I don't think this has anything to do with late-parsed templates, but rather is a more pressing problem with us trying to force-instantiate implicit ctors/dtors. |
|
This is most likely related to #419 (it's crashing on a NULL |
|
This is a good testcase, but we don't have any XFAIL facility in the current infrastructure, so we can |
kimgr
added a commit
to kimgr/include-what-you-use
that referenced
this pull request
May 23, 2021
In some scenarios when looking up and defining implicit methods, Sema assumes that its TUScope member is non-null. It turns out Sema resets TUScope at the end of parsing the translation unit, so the AST visitors are not expected to need a translation unit decl scope. Since our function template parsing and instantiation of implicit methods triggers quite a lot of Sema code, we've hit quite a few crashes on this in more complex scenarios with templates. Explicitly wire TUScope to Sema::getCurScope, to ensure there is a scope available for lookup and method synthesis. This fixes examples posted in include-what-you-use#151, include-what-you-use#419, include-what-you-use#601 and include-what-you-use#820.
kimgr
added a commit
to kimgr/include-what-you-use
that referenced
this pull request
May 23, 2021
In some scenarios when looking up and defining implicit methods, Sema assumes that its TUScope member is non-null. It turns out Sema resets TUScope at the end of parsing the translation unit, so the AST visitors are not expected to need a translation unit decl scope. Since our function template parsing and instantiation of implicit methods triggers quite a lot of Sema code, we've hit quite a few crashes on this in more complex scenarios with templates. Explicitly wire TUScope to Sema::getCurScope, to ensure there is a scope available for lookup and method synthesis. This fixes examples posted in include-what-you-use#151, include-what-you-use#419, include-what-you-use#601 and include-what-you-use#820.
kimgr
added a commit
to kimgr/include-what-you-use
that referenced
this pull request
Jun 6, 2021
In some scenarios when looking up and defining implicit methods, Sema assumes that its TUScope member is non-null. It turns out Sema resets TUScope at the end of parsing the translation unit, so the AST visitors are not expected to need a translation unit decl scope. Since our function template parsing and instantiation of implicit methods triggers quite a lot of Sema code, we've hit quite a few crashes on this in more complex scenarios with templates. Explicitly wire TUScope to Sema::getCurScope, to ensure there is a scope available for lookup and method synthesis. This fixes examples posted in include-what-you-use#151, include-what-you-use#419, include-what-you-use#601 and include-what-you-use#820.
kimgr
added a commit
to kimgr/include-what-you-use
that referenced
this pull request
Jun 6, 2021
In some scenarios when looking up and defining implicit methods, Sema assumes that its TUScope member is non-null. It turns out Sema resets TUScope at the end of parsing the translation unit, so the AST visitors are not expected to need a translation unit decl scope. Since our function template parsing and instantiation of implicit methods triggers quite a lot of Sema code, we've hit quite a few crashes on this in more complex scenarios with templates. Explicitly wire TUScope to Sema::getCurScope, to ensure there is a scope available for lookup and method synthesis. This fixes examples posted in include-what-you-use#151, include-what-you-use#419, include-what-you-use#601, include-what-you-use#673 and include-what-you-use#820. Use minimal repro from include-what-you-use#673 to build a half-hearted testcase.
kimgr
added a commit
that referenced
this pull request
Jun 6, 2021
In some scenarios when looking up and defining implicit methods, Sema assumes that its TUScope member is non-null. It turns out Sema resets TUScope at the end of parsing the translation unit, so the AST visitors are not expected to need a translation unit decl scope. Since our function template parsing and instantiation of implicit methods triggers quite a lot of Sema code, we've hit quite a few crashes on this in more complex scenarios with templates. Explicitly wire TUScope to Sema::getCurScope, to ensure there is a scope available for lookup and method synthesis. This fixes examples posted in #151, #419, #601, #673 and #820. Use minimal repro from #673 to build a half-hearted testcase.
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.
This testcase reproduces a problem with iwyu-ing llvm.
The problem still exists.
I am testing a fix on the clang side.
Irregardless of where the fix happens, a reproducing test is needed.