> Thanks for doing this! Disappointing to see the poor amd64 results of > course, but this is what we need to see.
You're welcome. I don't think the performance is poor in general, I think I hit a bug in g++. (C development could be a different story, for example.) I had a similar issue on x86 earlier in the g++ 3.4 release series: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15320 I've also seen a report that could be related on powerpc: http://www.cliftonlabs.com.com/savant/list-archives/savant-users/2004-December/000013.html I'm in the process of filing a report against g++ right now. Take care, Dale -- Dale E. Martin - [EMAIL PROTECTED] http://the-martins.org/~dmartin -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
