------- Comment #9 from pinskia at gcc dot gnu dot org  2005-11-07 17:38 -------
(In reply to comment #8)
> Andrew, in some sense you are right, given the current reference-counted 
> nature
> of std::string. However, let's not close this PR: submitter has a point that
> the user should be enabled to disable the use of atomics when configuring for
> single thread. I leave to you the exact categorization... Thanks.
I should note I closed this based on a conversation with the submitter.
I also filed PR 24715 for the real issue.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=24704

Reply via email to