This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug target/65780] [5/6 Regression] Uninitialized common handling in executables
- From: "hjl.tools at gmail dot com" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Thu, 16 Apr 2015 19:28:03 +0000
- Subject: [Bug target/65780] [5/6 Regression] Uninitialized common handling in executables
- Auto-submitted: auto-generated
- References: <bug-65780-4 at http dot gcc dot gnu dot org/bugzilla/>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65780
--- Comment #33 from H.J. Lu <hjl.tools at gmail dot com> ---
(In reply to Jeffrey A. Law from comment #32)
> HJ, you're missing my point. I need to understand the meaning of the
> variable defined_locally to move forward with the patch. Is it a must
> property (ie, if true, the symbol is always defined locally), or is it a may
> property (ie, if set, the symbol may be defined locally, but might also be
> defined externally).
>
> Much of the correctness of the patch hinges on that question.
There are 2 variables, defined_locally and resolved_locally.
A symbol can be defined locally, but not resolved locally,
depending on PIC and/or visibility.