https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115987
--- Comment #22 from Dan Urosu ---
Thank you for your help and explanations!
For now, I found an acceptable solution for my real project.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115987
--- Comment #19 from Dan Urosu ---
(In reply to Marek Polacek from comment #18)
> But is unwrap_1 called in such a way that 'r' binds to a temporary?
Did you mean 'u' binds to a temporary?
In this case r1 is a temporary object and a reference
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115987
--- Comment #17 from Dan Urosu ---
I just tried something after you explained how this warning works.
The most common cause of a real dangling ref is not caught unfortunately.
If you replace the unwrap_1 impl with the below you get no warning.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115987
--- Comment #15 from Dan Urosu ---
(In reply to Dan Urosu from comment #14)
> Thanks for clarifying (1)
> I think I can work-around with the pragma. Will try something shortly in the
> real code.
>
>
> Can you please elaborate on this "Right,
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115987
Dan Urosu changed:
What|Removed |Added
Status|ASSIGNED|RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115987
--- Comment #11 from Dan Urosu ---
1) "We don't analyze the body of the function"
unwrap_1 and unwrap_2 are functionally equivalent. Yet only unwrap_2 triggers
the warning.
2) If the unwrap_2 is called from a macro, then the pragma suppression d
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115987
--- Comment #8 from Dan Urosu ---
Why does not warn for unwarp_1?
The mere addition of an unused template parameter triggers this warning.
Also "decorating the unwrap_2 function to tell the compiler not to warn there."
does not work if we have
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115987
--- Comment #7 from Dan Urosu ---
Why does not warn for unwarp_1?
The mere addition of an unused template parameter triggers this warning.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115987
--- Comment #6 from Dan Urosu ---
Using the macro with the unwrap_1 does not error.
The GCC behavior is very weird.
On Thursday, July 18, 2024 at 12:44:21 PM EDT, Dan Urosu
wrote:
Ah indeed, but...
In the real use case we call it fr
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115987
--- Comment #5 from Dan Urosu ---
Ah indeed, but...
In the real use case we call it from macro.
#pragma GCC diagnostic push
#pragma GCC diagnostic ignored "-Wdangling-reference"
template
const T& unwrap_2(const Wrapper& r, FUNC&&) { // n
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115987
--- Comment #3 from Dan Urosu ---
The 109642 is not a duplicate of the attached code.
The error is the same but the code that causes it is completely different.
Also, the "solution" to disable the warning/error is not actually resolving the
iss
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115987
--- Comment #2 from Dan Urosu ---
Hi Andrew,
I cannot use "#pragma GCC diagnostic ignored" because we have thousands of call
sites where we plan to use this kind of code.
And we don't want to disable globally this warning either.
Would "no_da
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=115987
Bug ID: 115987
Summary: False "possibly dangling reference" false positive
when having an extra template parameter
Product: gcc
Version: 13.3.1
Status: UNCONFIRMED
13 matches
Mail list logo
