From owner-freebsd-stable Fri May 4 10:19:14 2001 Delivered-To: freebsd-stable@freebsd.org Received: from smtp.bmi.net (smtp.bmi.net [204.57.191.31]) by hub.freebsd.org (Postfix) with ESMTP id 4C7A737B424 for ; Fri, 4 May 2001 10:19:08 -0700 (PDT) (envelope-from jmcoopr@webmail.bmi.net) Received: from webmail.bmi.net (dsl-154.bmi.net [207.173.60.230]) by smtp.bmi.net (Pro-8.9.3/Pro-8.9.3) with ESMTP id KAA02944; Fri, 4 May 2001 10:19:02 -0700 Message-ID: <3AF2E488.F5549020@webmail.bmi.net> Date: Fri, 04 May 2001 10:19:04 -0700 From: John Merryweather Cooper X-Mailer: Mozilla 4.77 [en] (X11; U; Linux 2.2.12 i386) X-Accept-Language: en-US, en, en-GB MIME-Version: 1.0 To: Matt Dillon Cc: David Petrou , Antoine.Beaupre@ericsson.ca, freebsd-stable@FreeBSD.ORG Subject: Re: Should /sys/i386/conf/GENERIC's SysV shared memory settings+defaults be re-thought? References: <20010503221951.G47670@amant.pdl.cs.cmu.edu> <200105041614.f44GEsb25774@earth.backplane.com> <3AF2DBFD.37AA2124@webmail.bmi.net> <200105041707.f44H7Rw26684@earth.backplane.com> Content-Type: multipart/mixed; boundary="------------0DCA0C72B7F7E0D0CB092BA9" Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG This is a multi-part message in MIME format. --------------0DCA0C72B7F7E0D0CB092BA9 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Matt Dillon wrote: > > :> > :> In general there is no downside. Shared memory is swap-backed > :> (though the in-kernel control structures are not). > :> > :> I think it's high time that the system defaults be raised. I'll do > :> it later today. You should also be able to raise the defaults > :> using appropriate sysctl's, e.g.: > :> > :> sysctl -a | fgrep kern.ipc > :> sysctl -w kern.ipc.shmmax=33554432 > :> > :> And, of course, with kernel conf variables. > :> > :> -Matt > : > :Excellent! Other than larger swap usage, I haven't seen a downside to > :my settings (although I think they're a little on the large end) > :Performance with graphics, particularly with xscreensaver seems much > :improved though. > : > :jmc > > Could you do me a favor and post your 'ipcs -a' output for the case > of your nominally heavy X load? See attached. jmc ========================= > I am looking to raise the SHMMAXPGS default from 1024 to 8192, > SHMMNI from 96 to 192, and SHMSEG from 64 to 128. The question is > whether I have to raise SHMMNI and SHMSEG even higher -- I'm guessing > that your problems were mostly due to the too-low SHMMAXPGS default. > > -Matt --------------0DCA0C72B7F7E0D0CB092BA9 Content-Type: text/plain; charset=us-ascii; name="heavy.txt" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="heavy.txt" Message Queues: T ID KEY MODE OWNER GROUP CREATOR CGROUP CBYTES QNUM QBYTES LSPID LRPID STIME RTIME CTIME Shared Memory: T ID KEY MODE OWNER GROUP CREATOR CGROUP NATTCH SEGSZ CPID LPID ATIME DTIME CTIME m 262144 0 --rwarwarwa jmcoopr operator jmcoopr operator 2 393216 503 49816:00:42 16:01:09 16:00:42 m 4063233 0 --rwarwarwa jmcoopr operator jmcoopr operator 2 393216 579 49816:00:47 17:07:18 16:00:47 m 5046275 0 --rwarwarwa jmcoopr operator jmcoopr operator 2 393216 572 49816:00:48 16:00:49 16:00:48 m 14680069 0 --rwarwarwa jmcoopr operator jmcoopr operator 2 393216 606 49816:01:10 no-entry 16:01:10 m 3276806 0 --rwarwarwa jmcoopr operator jmcoopr operator 2 163840 601 49816:01:02 no-entry 16:01:02 m 3670023 0 --rwarwarwa jmcoopr operator jmcoopr operator 2 393216 599 49816:00:59 no-entry 16:00:59 m 1507336 0 --rwarwarwa jmcoopr operator jmcoopr operator 2 393216 601 49816:00:59 no-entry 16:00:59 m 37945353 2622055 --rw-r--r-- root wheel root wheel 0 1048576 96500 9687010:15:52 10:15:56 6:52:42 m 2097162 0 --rwarwarwa jmcoopr operator jmcoopr operator 2 393216 608 49816:01:24 no-entry 16:01:24 Semaphores: T ID KEY MODE OWNER GROUP CREATOR CGROUP NSEMS OTIME CTIME s 65536 2622057 --rw-rw-rw- root wheel root wheel 14no-entry 10:15:52 --------------0DCA0C72B7F7E0D0CB092BA9-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message