Date: Mon, 14 Jun 2004 14:58:02 -0400 From: John Baldwin <jhb@FreeBSD.org> To: Robert Watson <rwatson@FreeBSD.org> Cc: "David E. O'Brien" <obrien@FreeBSD.org> Subject: Re: cvs commit: src/sys/amd64/conf GENERIC Message-ID: <200406141458.02468.jhb@FreeBSD.org> In-Reply-To: <Pine.NEB.3.96L.1040614141907.34947B-100000@fledge.watson.org> References: <Pine.NEB.3.96L.1040614141907.34947B-100000@fledge.watson.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Monday 14 June 2004 02:21 pm, Robert Watson wrote: > On Mon, 14 Jun 2004, John Baldwin wrote: > > On Sunday 13 June 2004 07:03 pm, David E. O'Brien wrote: > > > obrien 2004-06-13 23:03:57 UTC > > > > > > FreeBSD src repository > > > > > > Modified files: > > > sys/amd64/conf GENERIC > > > Log: > > > The majority of FreeBSD/amd64 machines are SMP, so use > > > ADAPTIVE_MUTEXES by default to improve performance. > > > > We don't know if it improves performance yet. :) > > Yeah, as I was mentioning to David over the weekend, it's believed to help > performance anecdotally in a number of interesting cases, and if we can > show it helps in a majority of interesting cases (and doesn't > substantially hurt the others) then I think this should actually be in > GENERIC across all platforms, not just amd64. I also found I > misunderstood his commit message as it seems to suggest it helps SMP at > the cost of UP, but in fact does not say that. > > Kris reported that he experienced some apparent hangs related to running > with ADAPTIVE_MUTEXES and we should resolve if it was actually the cause > or whether it just triggered an existing race. I'm betting it is just triggering a race. When I first did the adaptive mutexes I stress tested it (-j <bignum> buildworld loops) on an ultra60, an alpha ds20, and a quad pii-xeon w/o any lockups. -- John Baldwin <jhb@FreeBSD.org> <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve" = http://www.FreeBSD.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200406141458.02468.jhb>