Date: Thu, 1 Jan 2009 10:55:11 -0800 From: Mayur <mayur.shardul@gmail.com> To: "Ed Schouten" <ed@80386.nl> Cc: Perforce Change Reviews <perforce@freebsd.org>, Alan Cox <alc@cs.rice.edu>, Mayur Shardul <mayur@freebsd.org> Subject: Re: PERFORCE change 155554 for review Message-ID: <2ac427a0901011055x7e561f12w338efac38e6c5e71@mail.gmail.com> In-Reply-To: <20090101141833.GF1176@hoeg.nl> References: <200901011408.n01E8GYU036190@repoman.freebsd.org> <20090101141833.GF1176@hoeg.nl>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi Ed, This work is not targeted to reduce the contention but this seems an interesting problem. Will look in to it. -- Mayur On Thu, Jan 1, 2009 at 6:18 AM, Ed Schouten <ed@80386.nl> wrote: > Hello Mayur, > > I don't know a lot about our VM subsystem, but I've got a question about > your work. Some time ago I ran a `buildkernel' with lock profiling > enabled and saw the following results (sorted by `wait_total'): > > http://80386.nl/files/lock-profiling-buildkernel.txt > > It seems to me that the "vm page queue mutex" is clearly one of the most > contended locks inside the kernel. Is your work also targeted to reduce > this contention? > > Happy 2009! > > -- > Ed Schouten <ed@80386.nl> > WWW: http://80386.nl/ >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?2ac427a0901011055x7e561f12w338efac38e6c5e71>