Date: Wed, 1 Jul 2009 17:41:34 -1000 (HST) From: Jeff Roberson <jroberson@jroberson.net> To: Randall Stewart <rrs@lakerest.net> Cc: Pierre Guinoiseau <geekounet@poildetroll.net>, "O. Hartmann" <ohartman@mail.zedat.fu-berlin.de>, Attilio Rao <attilio@freebsd.org>, bf <bf2006a@yahoo.com>, freebsd-current@freebsd.org, Kip Macy <kmacy@freebsd.org> Subject: Re: signifanctly slowdown of FreeBSD 8.0-CURRENT/amd64 Message-ID: <alpine.BSF.2.00.0907011740320.989@desktop> In-Reply-To: <2BE0378C-96A3-4714-A5C3-7B1A6AA0DCE2@lakerest.net> References: <951233.95131.qm@web39108.mail.mud.yahoo.com> <3c1674c90905302055g542cfadarf201cc273639977d@mail.gmail.com> <4A23919F.8050905@mail.zedat.fu-berlin.de> <4A2B3040.7020509@poildetroll.net> <3bbf2fe10906070339k663bace7qe5142702248ce6c9@mail.gmail.com> <4A2BBDC0.6000801@poildetroll.net> <3bbf2fe10906070623o65ce021fkb7f59fe1924cc1ec@mail.gmail.com> <4A353E21.1080001@poildetroll.net> <2BE0378C-96A3-4714-A5C3-7B1A6AA0DCE2@lakerest.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, 23 Jun 2009, Randall Stewart wrote: > One thing I have noticed for a while.. and have not > been able to track down.. > > If one runs > > /usr/src/tools/tools/syscall_timing/syscall_timing > > On a 7.2 kernel and compare it on the same machine to an 8.0 kernel > you will see almost a 3x slow down in 8. > > Yes I have 8.0 without witness and all the debug. The thing > that is most interesting is I even ran 8 on a 7.2 user space... > same result. > > My AMD 64 1.6Gig 8 core machines (in SMP) are showing > > 250ns or so for getuid 1000 times on 7.2 > and > 880ns or so for getuid 1000 times on 8.0 This is due to Kib's segment changes in current. I found it with hwpmc and he has produced a patch which fixes it. It will be in after beta 1. Thank you for reporting it. Jeff > > I started tracking this in 7.1.. > > When I get some more time next week I will do some more digging.. not sure > its related to this issue though > > R > > On Jun 14, 2009, at 2:14 PM, Pierre Guinoiseau wrote: > >> Hi again, >> >> I don't know if it can be of better help or not, but I made 2 other >> dumps, before and after the slowdown, and without X running and all... I >> recompiled some packages several times to make the slowdown appear. >> >> The dumps are here : http://foo.poildetroll.net/freebsd/ktr/ >> >> Thanks again for your help! >> >> Pierre Guinoiseau >> >> >> Attilio Rao wrote: >>> 2009/6/7 Pierre Guinoiseau <geekounet@poildetroll.net>: >>>> Ok, here is a ktr output. This time, the slowdown appeared while >>>> recompiling thunderbird. I have 2 core at 2.2Ghz (and powerd running, I >>>> don't know if it matters). >>>> >>>> BTW, what is the use of py-tkinter ? If it was in order to cause the >>>> bug, it failed, it needed a higher load. :) >>> >>> Ah sorry, I was supposed to let you run schedgraph but I'm going to do >>> that, so you actually didn't need it >>> I will let you know if something cames up. >>> If others can report the same it will be great. >>> >>> Thanks, >>> Attilio >>> _______________________________________________ >>> freebsd-current@freebsd.org mailing list >>> http://lists.freebsd.org/mailman/listinfo/freebsd-current >>> To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org" >> > > ------------------------------ > Randall Stewart > 803-317-4952 (cell) > 803-345-0391(direct) > > _______________________________________________ > freebsd-current@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-current > To unsubscribe, send any mail to "freebsd-current-unsubscribe@freebsd.org"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?alpine.BSF.2.00.0907011740320.989>