On Mar 14 12:47, Mojmir Svoboda wrote: > morning, Corinna > > > > the thing is that cygwin's pthread_mutex_lock performs quite lazily it > > > takes about seven times more than using native critical section. > > > Bookkeeping overhead? > > excuse my unfamiliarity with the matter, but i have no exact idea what > bookkeping could be. could you elaborate the answer more, please? > or point me to some article/documentation where i can read more?
Only in the source, sorry. I'm also not exaclty familiar with the pthread code in Cygwin. I only look into it when a serious problem crops up. > > Just have a look into the Cygwin source. > > yes, that's first place i tried to look, but got little disoriented > with the _ and __ versions of the pthread_mutex_lock. i've eventually > finished in __pthread_lock. am i on the right track? Er... probably. > > If you want to speed up Cygwin's pthread implementation, feel free to > > contribute: http://cygwin.com/contrib.html > > well i'm using windows and cygwin for few months so i quite don't feel > ready... yet ;) No worries. Everyone of us started with moaning and hair pulling over the existing code ;) Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Project Co-Leader cygwin AT cygwin DOT com Red Hat -- Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple Problem reports: http://cygwin.com/problems.html Documentation: http://cygwin.com/docs.html FAQ: http://cygwin.com/faq/
