------- Comment #11 from paolo dot carlini at oracle dot com 2010-09-08 09:59 ------- (In reply to comment #8) > But a 9-10x difference doesn't sound reasonable to me. The synced mode is not > unbuffered, before or after my suggested change, it uses the internal buffer > in > glibc.
So? We are not changing glibc here. The C++ library does *not* use buffering in the synced mode, and it does otherwise, for fstreams in particular. Where do you think the performance difference is essentially coming from? -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45574