From owner-freebsd-current Sun Apr 5 17:07:49 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id RAA26239 for freebsd-current-outgoing; Sun, 5 Apr 1998 17:07:49 -0700 (PDT) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from wankers.net (root@host-32-96-40-98.atl.bellsouth.net [32.96.40.98]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id RAA26226; Sun, 5 Apr 1998 17:07:36 -0700 (PDT) (envelope-from dex@wankers.net) Received: from localhost (dex@localhost [127.0.0.1]) by wankers.net (8.8.8/8.8.7) with SMTP id UAA08237; Sun, 5 Apr 1998 20:07:20 -0400 (EDT) (envelope-from dex@wankers.net) Date: Sun, 5 Apr 1998 20:07:19 -0400 (EDT) From: Dexnation Holodream X-Sender: dex@localhost To: Dag-Erling Coidan =?iso-8859-1?Q?Sm=F8rgrav?= cc: dyson@FreeBSD.ORG, Dan Swartzendruber , dg@root.com, stable@FreeBSD.ORG, current@FreeBSD.ORG Subject: Re: swap-leak in 2.2.5 ? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from QUOTED-PRINTABLE to 8bit by hub.freebsd.org id RAA26228 Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Well, folks, as was previously mentioned, the way to judge whether you need more ram is your paging (see vmstat(8)), and, if you want to see how much ram is being used, and how, and how much swap is being used, etc see top(1). both will provide you with real-time figures, and you can watch what processes are doing. for example: last pid: 8019; load averages: 0.35, 0.27, 0.20 19:54:08 47 processes: 1 running, 46 sleeping CPU states: 2.7% user, 0.0% nice, 2.7% system, 0.8% interrupt, 93.8% idle Mem: 65M Active, 10M Inact, 14M Wired, 8335K Buf, 35M Free Swap: 256M Total, 64K Used, 256M Free the above is a paste from top running on my P133 w/ 128M of ram and 256M of swap or: % vmstat 5 procs memory page disks faults cpu r b w avm fre flt re pi po fr sr s0 c0 in sy cs us sy id 1 2 0 4144276 35408 5 0 0 0 5 0 1 0 239 297 45 2 1 98 0 2 0 4137596 35404 2 0 0 0 0 0 0 0 232 281 42 0 0 99 0 2 0 4136312 35404 1 0 0 0 0 0 0 0 237 332 55 1 1 98 this is a 15 second long vmstat (5 second updates) pi is page in, and po is page out. As you can see, my machine is doing very little as this is taken. Now, watch what happens when I run two instances of GIMP on top of this: % vmstat 5 procs memory page disks faults cpu r b w avm fre flt re pi po fr sr s0 c0 in sy cs us sy id 2 2 0 4142396 35396 5 0 0 0 5 0 1 0 239 297 45 2 1 98 0 3 0 4169616 32216 129 0 9 0 60 0 40 0 350 1249 220 7 7 86 3 2 0 4191424 23416 1120 10 0 0 58 9623 126 0 431 2038 394 19 31 50 1 5 0 4540 17792 638 13 4 0 18 19144 77 0 314 1076 258 33 26 40 1 3 0 7220 18352 140 0 2 1 0 4824 24 0 253 27107 1840 61 29 10 1 2 0 7940 16368 104 23 4 0 14 0 7 0 250 2068 363 23 7 70 0 2 0 3136 16804 24 2 0 0 24 4791 0 0 267 631 109 2 4 94 0 2 0 2412 16712 2 3 0 0 0 0 2 0 238 397 43 1 2 98 0 2 0 3096 16712 0 0 0 0 0 0 0 0 236 281 42 0 1 99 As you can see, there has been some activity (granted, not much) last pid: 8089; load averages: 0.61, 0.44, 0.29 19:58:21 52 processes: 1 running, 51 sleeping CPU states: 4.0% user, 0.0% nice, 4.4% system, 0.0% interrupt, 91.6% idle Mem: 93M Active, 32K Inact, 15M Wired, 16M Cache, 8348K Buf, 616K Free Swap: 256M Total, 104K Used, 256M Free % swapinfo Device 1K-blocks Used Avail Capacity Type /dev/sd0b 262144 13976 248104 5% Interleaved Some swap has been used, here. closing GIMP: % !vm vmstat 5 procs memory page disks faults cpu r b w avm fre flt re pi po fr sr s0 c0 in sy cs us sy id 2 2 0 4180640 16728 6 0 0 0 5 2 1 0 239 300 45 2 1 98 0 2 0 4143644 54188 72 11 1 0 1952 0 1 0 282 851 157 5 5 90 0 2 0 4144368 54188 1 0 0 0 0 0 2 0 250 341 58 1 1 98 last pid: 8148; load averages: 0.16, 0.31, 0.25 20:00:51 48 processes: 2 running, 46 sleeping CPU states: 1.2% user, 0.0% nice, 1.9% system, 0.8% interrupt, 96.1% idle Mem: 51M Active, 9160K Inact, 15M Wired, 15M Cache, 8340K Buf, 34M Free Swap: 256M Total, 104K Used, 256M Free as you can see, top still reflects data cached in swap, but look at swapinfo: % swapinfo Device 1K-blocks Used Avail Capacity Type /dev/sd0b 262144 40 262040 0% Interleaved My interpretation, here, is that the data is still cached, but that space will be reused if I don't run GIMP again, but will stay cached in case I need it (OR there's a bug hidden somewhere). -Jon On 6 Apr 1998, Dag-Erling Coidan [iso-8859-1] Smørgrav wrote: > "John S. Dyson" writes: > > On my workstation, I run with 1.2GB of available swap space, and anybody > > can afford that, can't they? (BTW, I seldom use more than 30-40MB, but > > with the price of disk, who cares?) > > I only have 512 MB (out of 9 GB of disk space) but then again I have > 128 MB RAM, so I practically never use any swap at all, except > possibly while making world. On my laptop, however, I only have 16 MB > RAM (should have been 32, but Big Three-Letter Computer Company (tm) > screwed up and I'm still waiting for the missing RAM) so swap space > gets eaten up PDQ. > > -- > fprintf(stderr, "I have a closed mind. It helps keeping the rain out.\n"); > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-stable" in the body of the message > > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message