public inbox for gcc-prs@sourceware.org help / color / mirror / Atom feed
From: shuki_duv@yahoo.com To: gcc-gnats@gcc.gnu.org Subject: c++/6564: -frepo with function template in anonymous namespace fails to compile Date: Sun, 05 May 2002 10:16:00 -0000 [thread overview] Message-ID: <20020505171043.15707.qmail@sources.redhat.com> (raw) >Number: 6564 >Category: c++ >Synopsis: -frepo with function template in anonymous namespace fails to compile >Confidential: no >Severity: serious >Priority: medium >Responsible: unassigned >State: open >Class: rejects-legal >Submitter-Id: net >Arrival-Date: Sun May 05 10:16:00 PDT 2002 >Closed-Date: >Last-Modified: >Originator: Shimshon Duvdevan >Release: gcc version 3.0.4, Thread model: posix >Organization: >Environment: sparc-sun-solaris2.6 (bug irrelevant to platform) >Description: A function template is defined in anonymous namespace, and is called from global namespace. When trying to compile with -frepo, the recompiling/relinking goes on indefinitely, until done 17 times (regardless of -ftemplate-depth-n): [these are the commands for the first example, second is similar] $ g++ -frepo -Wall -c -o anon.o anon.C $ g++ -frepo -Wall -o anon anon.o collect: recompiling anon.C collect: relinking collect: recompiling anon.C collect: relinking collect: recompiling anon.C collect: relinking collect: recompiling anon.C collect: relinking collect: recompiling anon.C collect: relinking collect: recompiling anon.C collect: relinking collect: recompiling anon.C collect: relinking collect: recompiling anon.C collect: relinking collect: recompiling anon.C collect: relinking collect: recompiling anon.C collect: relinking collect: recompiling anon.C collect: relinking collect: recompiling anon.C collect: relinking collect: recompiling anon.C collect: relinking collect: recompiling anon.C collect: relinking collect: recompiling anon.C collect: relinking collect: recompiling anon.C collect: relinking collect: recompiling anon.C collect: relinking Undefined first referenced symbol in file void (anonymous namespace)::foo<int>(int)anon.o ld: fatal: Symbol referencing errors. No output written to anon collect2: ld returned 1 exit status >How-To-Repeat: [anon.C - first example] namespace { template <typename Tp> void foo(Tp) {} } int main() { foo(int()); } [anon2.C - second example] namespace { struct Foo {}; } template <typename Tp> void foo(Tp) {} int main() { foo(Foo()); } >Fix: Inlining the function explicitly or implicitly (via -O3) makes the problem go away, but only because the function is, well, inlined. Using static qualifier instead of the anonymous namespace fixes the problem in the first case, but then there are scenarios that are not fixable in this way, such as a function template (not in anonymous namespace), whose parameter is a class which is defined in an anonymous namespace, see second example. >Release-Note: >Audit-Trail: >Unformatted:
next reply other threads:[~2002-05-05 17:16 UTC|newest] Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top 2002-05-05 10:16 shuki_duv [this message] 2002-07-06 7:50 lerdsuwa
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=20020505171043.15707.qmail@sources.redhat.com \ --to=shuki_duv@yahoo.com \ --cc=gcc-gnats@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).