public inbox for gcc-bugs@sourceware.org help / color / mirror / Atom feed
From: "jlegg at feralinteractive dot com" <gcc-bugzilla@gcc.gnu.org> To: gcc-bugs@gcc.gnu.org Subject: [Bug c++/102496] New: [11 regression] extern __thread declaration in function scope produces a non-TLS reference Date: Mon, 27 Sep 2021 10:37:26 +0000 [thread overview] Message-ID: <bug-102496-4@http.gcc.gnu.org/bugzilla/> (raw) https://gcc.gnu.org/bugzilla/show_bug.cgi?id=102496 Bug ID: 102496 Summary: [11 regression] extern __thread declaration in function scope produces a non-TLS reference Product: gcc Version: 11.2.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: c++ Assignee: unassigned at gcc dot gnu.org Reporter: jlegg at feralinteractive dot com Target Milestone: --- A variable declaration inside a function body qualified with extern and __thread appears to ignore the __thread qualification when compiling c++. For example: Contents of test1.cc: extern __thread int thing; __thread int thing = 0; Contents of test2.cc: int main() { extern __thread int thing; return thing; } When invoking g++ 11.2.0 with arguments test1.cc test2.cc, I get the following output: /usr/local/bin/ld: thing: TLS definition in /tmp/ccXxZiSY.o section .tbss mismatches non-TLS reference in /tmp/ccYXyoSR.o /usr/local/bin/ld: /tmp/ccYXyoSR.o: error adding symbols: bad value collect2: error: ld returned 1 exit status If the example code was instead in files with a .c extension and compiled with gcc (as opposed to g++) 11.2.0, it does not produce an error. I also could not reproduce the error with g++ 10.2.0. The options I used when configuring GCC are: --prefix=/usr --program-suffix=-11.2 --libexecdir=/usr/lib --with-gxx-include-dir=/usr/include/c++/11.2 --libdir=/usr/lib --with-tune=generic --enable-checking=release --disable-multilib --enable-languages=c,c++ --build=x86_64-linux-gnu --target=x86_64-linux-gnu --enable-linker-build-id --enable-version-specific-runtime-libs It is possible to work around the issue by moving or copying the variable declaration to namespace scope.
next reply other threads:[~2021-09-27 10:37 UTC|newest] Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-09-27 10:37 jlegg at feralinteractive dot com [this message] 2021-09-27 10:43 ` [Bug c++/102496] [11/12 " pinskia at gcc dot gnu.org 2021-09-27 10:52 ` [Bug c++/102496] [11/12 regression] extern __thread and thread_local " pinskia at gcc dot gnu.org 2021-09-28 8:32 ` rguenth at gcc dot gnu.org 2021-09-28 14:31 ` jakub at gcc dot gnu.org 2021-09-28 14:38 ` jakub at gcc dot gnu.org 2021-10-01 8:31 ` cvs-commit at gcc dot gnu.org 2021-10-01 12:39 ` [Bug c++/102496] [11 " jakub at gcc dot gnu.org 2021-10-05 20:30 ` cvs-commit at gcc dot gnu.org 2021-10-12 9:55 ` jakub at gcc dot gnu.org
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=bug-102496-4@http.gcc.gnu.org/bugzilla/ \ --to=gcc-bugzilla@gcc.gnu.org \ --cc=gcc-bugs@gcc.gnu.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).