Deadly embrace between pthread_cond_wait and pthread_cond_signal

[Robert Collins]

----- Original Message -----
From: "Christopher Faylor" <cgf@redhat.com>
> >>I bet it would improve even more if we replaced the VirtualQuery
> >>in path.cc, too.
> >
> >With Rob's new patch that does this,
> >there actually isn't very much difference
> >
> >real    7m10.729s
> >user    2m50.963s
> >sys     1m11.721s
> >
> >Thanks !
>
> Huh.  I wonder why it makes such a big difference for pthreads.
>

Volume of calls :}. openening files doesn't occur quite as often as (say)
locking a mutex.

I'd like to publicly thank Greg Smith for his excellent profiling work in
identifying the bottleneck here. I know how time consuming that can be :}.

Thanks Greg!

Rob