From owner-freebsd-smp Mon Jul 8 10:39:59 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8CCAF37B400 for ; Mon, 8 Jul 2002 10:39:57 -0700 (PDT) Received: from rhenium.btinternet.com (rhenium.btinternet.com [194.73.73.93]) by mx1.FreeBSD.org (Postfix) with ESMTP id 06F7A43E3B for ; Mon, 8 Jul 2002 10:39:57 -0700 (PDT) (envelope-from ximf@btinternet.com) Received: from host217-35-62-204.in-addr.btopenworld.com ([217.35.62.204] helo=IMFSYS) by rhenium.btinternet.com with smtp (Exim 3.22 #8) id 17RcUN-0007jr-00 for freebsd-smp@FreeBSD.org; Mon, 08 Jul 2002 18:39:55 +0100 Message-ID: <000501c226a6$81048960$0200a8c0@IMFSYS> From: "Alistair Cockeram" To: Subject: help Date: Sun, 7 Jul 2002 22:21:05 +0100 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0005_01C22604.95BB25F0" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 Disposition-Notification-To: "Alistair Cockeram" X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org This is a multi-part message in MIME format. ------=_NextPart_000_0005_01C22604.95BB25F0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable test ------=_NextPart_000_0005_01C22604.95BB25F0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

test

------=_NextPart_000_0005_01C22604.95BB25F0-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Jul 10 4:21:58 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 083BE37B400 for ; Wed, 10 Jul 2002 04:21:57 -0700 (PDT) Received: from mta6.snfc21.pbi.net (mta6.snfc21.pbi.net [206.13.28.240]) by mx1.FreeBSD.org (Postfix) with ESMTP id ACB3843E4A for ; Wed, 10 Jul 2002 04:21:56 -0700 (PDT) (envelope-from mtm@pacbell.net) Received: from kokeb.ambesa.net ([64.166.84.33]) by mta6.snfc21.pbi.net (iPlanet Messaging Server 5.1 (built May 7 2001)) with ESMTP id <0GZ100E7D68KG9@mta6.snfc21.pbi.net> for freebsd-smp@freebsd.org; Wed, 10 Jul 2002 04:21:56 -0700 (PDT) Received: from kokeb.ambesa.net (tanstaafl@localhost [IPv6:::1]) by kokeb.ambesa.net (8.12.5/8.12.5) with ESMTP id g6ABLn6V000883 for ; Wed, 10 Jul 2002 04:21:49 -0700 (PDT envelope-from mtm@kokeb.ambesa.net) Received: (from mtm@localhost) by kokeb.ambesa.net (8.12.5/8.12.5/Submit) id g6ABLnXU000882; Wed, 10 Jul 2002 04:21:49 -0700 (PDT) Date: Wed, 10 Jul 2002 04:21:49 -0700 From: Mike Makonnen Subject: Getting resource limits out from under Giant To: freebsd-smp@freebsd.org Message-id: <20020710042149.26b39b62.makonnen@pacbell.net> MIME-version: 1.0 X-Mailer: Sylpheed version 0.7.8 (GTK+ 1.2.10; i386-portbld-freebsd5.0) Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Hello folks, I took a stab at getting resource limits out from under Giant. The patch is at http://home.pacbell.net/makonnen/limits.giant.patch Comments? One thing I would like feedback on is the unconditional allocation of memory in fork1() to avoid allocating memory while holding the proc lock. Resource limits are copy-on-write by default and since the copy doesn't happen very often this leads to most times allocating a new struct plimit and just deallocating it a few lines later. I suspect that's probably not what we want. I can see a couple of ways to solve it, like keeping a preallocated pool or dropping, allocating the memory, and then re-aquiring the proc lock. Cheers, Mike Makonnen To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Jul 11 2: 3:58 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2386C37B400 for ; Thu, 11 Jul 2002 02:03:56 -0700 (PDT) Received: from amsfep12-int.chello.nl (amsfep12-int.chello.nl [213.46.243.17]) by mx1.FreeBSD.org (Postfix) with ESMTP id DDDD043E31 for ; Thu, 11 Jul 2002 02:03:54 -0700 (PDT) (envelope-from Harold@drostp.nl) Received: from dnp-pdc.drostp.nl ([212.142.3.88]) by amsfep12-int.chello.nl (InterMail vM.5.01.03.06 201-253-122-118-106-20010523) with ESMTP id <20020711090354.VIHO28368.amsfep12-int.chello.nl@dnp-pdc.drostp.nl> for ; Thu, 11 Jul 2002 11:03:54 +0200 X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3 content-class: urn:content-classes:message Subject: Date: Thu, 11 Jul 2002 11:12:09 +0200 MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----_=_NextPart_001_01C228BB.08E4472E" Message-ID: <040F28299690B345B86908BFC2631B187036@dnp-pdc.drostp.nl> X-MS-Has-Attach: yes X-MS-TNEF-Correlator: Thread-Index: AcIouwg7IH9HI5auRGqof1PI29mMGA== From: "Harold de Bruijn" To: Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org This is a multi-part message in MIME format. ------_=_NextPart_001_01C228BB.08E4472E Content-Type: multipart/alternative; boundary="----_=_NextPart_002_01C228BB.08E4472E" ------_=_NextPart_002_01C228BB.08E4472E Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable =20 Met vriendelijke groet, Drost & Partners =20 Harold de Bruijn =20 ------_=_NextPart_002_01C228BB.08E4472E Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

Met vriendelijke groet,

Drost & Partners

Harold de Bruijn

------_=_NextPart_002_01C228BB.08E4472E-- ------_=_NextPart_001_01C228BB.08E4472E Content-Type: text/x-vcard; name="Harold de Bruijn.vcf" Content-Transfer-Encoding: base64 Content-Description: Harold de Bruijn.vcf Content-Disposition: attachment; filename="Harold de Bruijn.vcf" QkVHSU46VkNBUkQNClZFUlNJT046Mi4xDQpOOmRlIEJydWlqbjtIYXJvbGQNCkZOOkhhcm9sZCBk ZSBCcnVpam4NCkVNQUlMO1BSRUY7SU5URVJORVQ6SGFyb2xkQGRyb3N0cC5ubA0KUkVWOjIwMDIw MjE5VDEzMjM1OFoNCkVORDpWQ0FSRA0K ------_=_NextPart_001_01C228BB.08E4472E-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Jul 11 7:54:42 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 77BC737B400; Thu, 11 Jul 2002 07:53:57 -0700 (PDT) Received: from excite.com (host134225.metrored.net.ar [200.59.134.225]) by mx1.FreeBSD.org (Postfix) with SMTP id B636A43E54; Thu, 11 Jul 2002 07:53:45 -0700 (PDT) (envelope-from Center_for_Age6414n05@excite.com) Received: from smtp013.mail.yahou.com ([29.16.108.187]) by mailout2-eri1.midmouth.com with esmtp; Thu, 11 Jul 0102 06:54:02 -0100 Received: from unknown (HELO mail.gimmixx.net) (155.22.111.109) by mta21.bigpong.com with SMTP; 11 Jul 0102 05:51:36 +1000 Received: from unknown (HELO rly-yk05.pesdets.com) (187.197.82.227) by n7.groups.huyahoo.com with asmtp; Thu, 11 Jul 0102 15:49:10 -0300 Received: from unknown (146.133.89.190) by symail.kustanai.co.kr with SMTP; 11 Jul 0102 12:46:44 -0000 Received: from 11.215.236.71 ([11.215.236.71]) by q4.quickslow.com with local; 11 Jul 0102 12:44:18 +0200 Reply-To: "Office" Message-ID: <025c34b26e3a$5734d7c4$5cb48da7@ajjqnx> From: "Office" To: Cc: , , , , , Subject: Three types of HGH products Date: Thu, 11 Jul 0102 12:53:21 +0200 MiME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_NextPart_000_00A3_62D23A1C.D7817E20" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2919.6700 Importance: Normal Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org ------=_NextPart_000_00A3_62D23A1C.D7817E20 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: base64 QXJlIEhHSCBwcm9kdWN0cyBkaWZmZXJlbnQ/DQoNCjxodG1sPg0KDQo8aGVh ZD4NCg0KPG1ldGEgaHR0cC1lcXVpdj0iQ29udGVudC1UeXBlIiBjb250ZW50 PSJ0ZXh0L2h0bWw7IGNoYXJzZXQ9d2luZG93cy0xMjUyIj4NCg0KPG1ldGEg bmFtZT0iR0VORVJBVE9SIiBjb250ZW50PSJNaWNyb3NvZnQgRnJvbnRQYWdl IDQuMCI+DQoNCjxtZXRhIG5hbWU9IlByb2dJZCIgY29udGVudD0iRnJvbnRQ YWdlLkVkaXRvci5Eb2N1bWVudCI+DQoNCjx0aXRsZT5UaGVyZSBhcmUgdGhy ZWUgZGlmZmVyZW50IHR5cGVzIG9mIEhHSCBwcm9kdWN0czwvdGl0bGU+DQoN CjwvaGVhZD4NCg0KPGJvZHkgYmFja2dyb3VuZD0iY2xvdWRzLmpwZyI+DQoN CjxwPjxmb250IHNpemU9IjQiPjxmb250IGNvbG9yPSIjODAwMDAwIj48Yj5U aGVyZSBhcmUgdGhyZWUgZGlmZmVyZW50IHR5cGVzIG9mDQoNCkhHSCBwcm9k dWN0cy48L2I+PC9mb250Pjxicj4NCg0KVGhlIGNvbmZ1c2lvbiBpcyB0aGF0 IGFsbCB0aHJlZSBhcmU8YnI+DQoNCmFkdmVydGlzZWQgYXMgaWYgdGhleSB3 ZXJlIHRoZSBzYW1lLjwvZm9udD48YnI+DQoNCiZuYnNwOzxicj4NCg0KJm5i c3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7IDx1PlRo ZSB0aHJlZSB0eXBlcyBhcmU6PC91Pjxicj4NCg0KJm5ic3A7PGJyPg0KDQo8 Yj4xKTwvYj4gLS0tIDxmb250IGNvbG9yPSIjMDAwMEZGIj48Yj5Ib21lb3Bh dGhpYyBIR0g8L2I+PC9mb250Pjxicj4NCg0KPGI+Mik8L2I+IC0tLSA8Zm9u dCBjb2xvcj0iIzAwMDBGRiI+PGI+UHJlLWN1cnNvciBIR0g8L2I+PC9mb250 Pjxicj4NCg0KPGI+Myk8L2I+IC0tLSA8Zm9udCBjb2xvcj0iIzAwMDBGRiI+ PGI+UmVhbCBvciBzeW50aGV0aWMgSEdIPC9iPjwvZm9udD4NCg0KKGRlbGl2 ZXJlZCBieSBpbmplY3Rpb248YnI+DQoNCiZuYnNwOyZuYnNwOyZuYnNwOyZu YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyBvciwgYnkgYW4gb3JhbCBzcHJheSBt ZXRob2QpLjxicj4NCg0KJm5ic3A7PGJyPg0KDQpEbyB5b3Uga25vdyBkaWZm ZXJlbmNlcz88YnI+DQoNCiZuYnNwOzxicj4NCg0KQ2FsbCB1cyBhbmQgd2Un bGwgZXhwbGFpbiB0aGVtIHRvIHlvdS48YnI+DQoNCiZuYnNwOzxicj4NCg0K T3VyIHRvbGwgZnJlZSBudW1iZXIgaXMgPGZvbnQgY29sb3I9IiMwMDAwODAi PjxiPjEtODg4LTYyMS03MzAwPC9iPjwvZm9udD48YnI+DQoNCkFuIEhHSCBz dGFmZiBtZW1iZXIgaXMgYXZhaWxhYmxlPGJyPg0KDQo5IHRvIDUgUGFjaWZp YyBUaW1lLjxicj4NCg0KSWYgYWZ0ZXIgaG91cnMsIHBsZWFzZSBsZWF2ZSB5 b3UgbmFtZTxicj4NCg0KYW5kIGRheSBhbmQgZXZlbmluZyBwaG9uZSBudW1i ZXJzLjxicj4NCg0KV2Ugd2lsbCBjYWxsIHlvdSBiYWNrIGluIGEgbm8gcHJl c3N1cmUsPGJyPg0KDQplZHVjYXRpb25hbCBtYW5uZXIuPGJyPg0KDQpJZiB5 b3UgYXJlIG92ZXJzZWFzIGNhbGwgeW91ciBsb25nIGRpc3RhbmNlPGJyPg0K DQpvcGVyYXRvciBhbmQgYXNrIHRvIGJlIGNvbm5lY3RlZCB0byBvdXI8YnI+ DQoNCnBob25lIG51bWJlci4mbmJzcDsgV2Ugd2lsbCBjYWxsIHlvdSBiYWNr IHNvPGJyPg0KDQp3ZSBjYW4gcGF5IGZvciB0aGUgbG9uZyBkaXN0YW5jZSBj aGFyZ2VzLjxicj4NCg0KJm5ic3A7PGJyPg0KDQo8Zm9udCBjb2xvcj0iI0ZG MDAwMCI+Rm9yIG1vcmUgaW5mb3JtYXRpb24gb24gSEdIIHJlYWQgb24uLi4u Li4uLi4uLi48L2ZvbnQ+PGJyPg0KDQombmJzcDs8YnI+DQoNCkhBVkUgWU9V IEhFQVJEIE9GPGJyPg0KDQpIVU1BTiBHUk9XVEggSE9STU9ORSAoSEdIKT8/ Pzxicj4NCg0KJm5ic3A7PGJyPg0KDQombmJzcDsmbmJzcDsmbmJzcDsmbmJz cDsgUmVsZWFzZWQgYnkgeW91ciBvd24gcGl0dWl0YXJ5IGdsYW5kLCBIR0gg c3RhcnRzDQoNCmRlY2xpbmluZzxicj4NCg0KaW4geW91ciAyMHMsIGV2ZW4g bW9yZSBpbiB5b3VyIDMwcyBhbmQgNDBzLCBldmVudHVhbGx5IHJlc3VsdGlu Zzxicj4NCg0KaW4gdGhlIHNocmlua2FnZSBvZiBtYWpvciBvcmdhbnMgLS0g cGx1cywgYWxsPGJyPg0KDQpvdGhlciBzeW1wdG9tcyByZWxhdGVkIHRvIG9s ZCBhZ2UuPGJyPg0KDQombmJzcDs8YnI+DQoNCiZuYnNwOzxicj4NCg0KSU4g VEhPVVNBTkRTIE9GIENMSU5JQ0FMIFNUVURJRVMsPGJyPg0KDQpIR0ggSEFT IEJFRU4gU0hPV04gVE8gQUNDT01QTElTSCBUSEUgRk9MTE9XSU5HOjxicj4N Cg0KJm5ic3A7PGJyPg0KDQoqIFJlZHVjZSBCb2R5IEZhdCBhbmQgQnVpbGQg TGVhbiBNdXNjbGU8YnI+DQoNCiZuYnNwOyZuYnNwOyBXSVRIT1VUIEVYRVJD SVNFITxicj4NCg0KJm5ic3A7PGJyPg0KDQoqIEVuaGFuY2UgU2V4dWFsIFBl cmZvcm1hbmNlPGJyPg0KDQombmJzcDs8YnI+DQoNCiogUmVtb3ZlIFdyaW5r bGVzIGFuZCBDZWxsdWxpdGU8YnI+DQoNCiZuYnNwOzxicj4NCg0KKiBMb3dl ciBCbG9vZCBQcmVzc3VyZSBhbmQgSW1wcm92ZSBDaG9sZXN0ZXJvbCBQcm9m aWxlPGJyPg0KDQombmJzcDs8YnI+DQoNCiogSW1wcm92ZSBTbGVlcCwgVmlz aW9uIGFuZCBNZW1vcnk8YnI+DQoNCiZuYnNwOzxicj4NCg0KKiBSZXN0b3Jl IEhhaXIgQ29sb3IgYW5kIEdyb3d0aDxicj4NCg0KJm5ic3A7PGJyPg0KDQoq IFN0cmVuZ3RoZW4gdGhlIEltbXVuZSBTeXN0ZW08YnI+DQoNCiZuYnNwOzxi cj4NCg0KKiBJbmNyZWFzZSBFbmVyZ3kgYW5kIENhcmRpYWMgT3V0cHV0PGJy Pg0KDQombmJzcDs8YnI+DQoNCiogVHVybiBiYWNrIHlvdXIgYm9keSdzIEJp b2xvZ2ljYWwgVGltZSBDbG9jayAxMCAtIDIwIHllYXJzPGJyPg0KDQombmJz cDs8YnI+DQoNCiogTGl2ZSBMb25nZXIgQU5EIFN0cm9uZ2VyPGJyPg0KDQom bmJzcDs8YnI+DQoNCkFsbCBuYXR1cmFsIGFuZCBvcmdhbmljIHBsYW50IGJh c2VkPGJyPg0KDQombmJzcDs8YnI+DQoNCjxmb250IGNvbG9yPSIjMDAwMEZG Ij48Yj5GRUVMIDEwIFlFQVJTIFlPVU5HRVIgV0lUSCBPUkFMIFNQUkFZIEhH SC48YnI+DQoNCkdVQVJBTlRFRUQ8L2I+PC9mb250Pjxicj4NCg0KJm5ic3A7 PGJyPg0KDQombmJzcDsmbmJzcDsmbmJzcDsgV2UgYXJlIHRoZSBtYW51ZmFj dHVyZXIgYW5kIHdlIHNlbGwgZGlyZWN0bHkgdG8gRG9jdG9ycyw8YnI+DQoN CkNoaXJvcHJhY3RvcnMsIGFuZCBjb25zdW1lcnMgd29ybGQgd2lkZSB0aGUg aGlnaGVzdCBncmFkZTxicj4NCg0KJm5ic3A7SEdIIE9yYWwgU3ByYXkgYXZh aWxhYmxlLiZuYnNwOzxicj4NCg0KJm5ic3A7PGJyPg0KDQombmJzcDsmbmJz cDsmbmJzcDsmbmJzcDsgV2l0aCBpbnRlcm5ldCBtYXJrZXRpbmcsIHdlIGFy ZSBhYmxlIHRvIHNhdmUNCg0KYWR2ZXJ0aXNpbmc8YnI+DQoNCmNvc3QgYW5k IHBhc3MgdGhvc2Ugc2F2aW5ncyBhbG9uZyB0byB5b3UuPGJyPg0KDQpCdXQg eW91IG11c3QgYWN0IG5vdy4mbmJzcDs8YnI+DQoNCiZuYnNwOzxicj4NCg0K VG8gcmVjZWl2ZSBtb3JlIGluZm9ybWF0aW9uIGNhbGwmbmJzcDsgdXMgbm93 Ljxicj4NCg0KJm5ic3A7PGJyPg0KDQombmJzcDsmbmJzcDsmbmJzcDsmbmJz cDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsg VE9MTCBGUkVFIDxiPjxmb250IGNvbG9yPSIjMDAwMDgwIj4xLTg4OC02MjEt NzMwMDwvZm9udD48L2I+PGJyPg0KDQombmJzcDs8YnI+DQoNCldlIG11c3Qg c3BlYWsgdG8geW91IGluIHBlcnNvbiB0byBxdWFsaWZ5IHlvdXIgdXNhZ2Uu PGJyPg0KDQombmJzcDs8YnI+DQoNCiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNw OyBBbGwgb2YgeW91ciBxdWVzdGlvbnMgd2lsbCBiZSBhZGRyZXNzZWQgYW5k IGFuc3dlcmVkIGluDQoNCmEgZnJpZW5kbHksPGJyPg0KDQpubyBwcmVzc3Vy ZSBtYW5uZXIuJm5ic3A7IE91ciBtYWluIHB1cnBvc2UgaXMgdG8gcHJvdmlk ZSB5b3Ugd2l0aDxicj4NCg0KJm5ic3A7aW5mb3JtYXRpb24gc28geW91IGNh biBtYWtlIGFuIGVkdWNhdGVkIGRlY2lzaW9uLjxicj4NCg0KJm5ic3A7PGJy Pg0KDQombmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsgRm9yIG1vcmUgaW5mb3Jt YXRpb24gY2FsbDxicj4NCg0KJm5ic3A7PGJyPg0KDQombmJzcDsmbmJzcDsm bmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJzcDsmbmJz cDsmbmJzcDsgPGI+PGZvbnQgY29sb3I9IiMwMDAwODAiPjEtODg4LTYyMS03 MzAwPC9mb250PjwvYj48YnI+DQoNCiZuYnNwOzxicj4NCg0KJm5ic3A7SWYg eW91IGFyZSBvbiBsaW5lIHdyaXRlIGRvd24gb3VyPGJyPg0KDQpwaG9uZSBu dW1iZXIgYW5kIGNhbGwgdXMgd2hlbiB5b3UgY2FuLjxicj4NCg0KJm5ic3A7 PGJyPg0KDQpTb29uLCB5b3UgYW5kIHlvdXIgbG92ZWQgb25lcyB3aWxsIGJl IHZlcnkgZ2xhZCB5b3UgZGlkLjxicj4NCg0KJm5ic3A7PGJyPg0KDQpSZWFk IHdoYXQgcGVvcGxlIGFyZSBzYXlpbmc6PGJyPg0KDQombmJzcDs8YnI+DQoN CiZxdW90O1RoZSBlZmZlY3RzIG9mIDYgbW9udGhzIG9mIEdIIG9uPGJyPg0K DQpsZWFuIGJvZHkgbWFzcyBhbmQgZmF0IHdlcmUgZXF1aXZhbGVudDxicj4N Cg0KaW4gbWFnbml0dWRlIHRvIHRoZSBjaGFuZ2VzIGluY3VycmVkPGJyPg0K DQpkdXJpbmcgMTAtMjAgeWVhcnMgb2YgYWdpbmcuJnF1b3Q7PGJyPg0KDQpE ci4gRGFuaWVsIFJ1ZG1hbiwgTUQsPGJyPg0KDQpOZXcgRW5nbGFuZCBKb3Vy bmFsIG9mIE1lZGljaW5lLjxicj4NCg0KJm5ic3A7PGJyPg0KDQomcXVvdDtX aXRoaW4gZm91ciBtb250aHMsIG15IGJvZHkgZmF0IGRlY3JlYXNlZDxicj4N Cg0KJm5ic3A7Zm9ybSAzMCUgZG93biB0byAyMSUhIEkgbm90aWNlZCBteSBz a2luPGJyPg0KDQombmJzcDtpcyBtb3JlIHN1cHBsZSBhbmQgbXkgb3ZlcmFs bCBtZW50YWw8YnI+DQoNCiZuYnNwO291dGxvb2sgaW1wcm92ZWQgc2lnbmlm aWNhbnRseS4mcXVvdDs8YnI+DQoNCiZuYnNwO0QuVy4sIE5ldyBKZXJzZXk8 YnI+DQoNCiZuYnNwOzxicj4NCg0KJnF1b3Q7V2UgaGF2ZSBiZWVuIG9uIHRo ZSBzcHJheSBmb3IganVzdCAzIHdlZWtzPGJyPg0KDQpub3csIGFuZCBiZXNp ZGVzIHRoZSB0cmVtZW5kb3VzIGVuZXJneSB3ZTxicj4NCg0KYm90aCBmZWVs LCBteSBodXNiYW5kcyBhbGxlcmdpZXMgYW5kIHNwZWxsczxicj4NCg0Kb2Yg ZGVwcmVzc2lvbiBoYXZlIGxpZnRlZC4gSSBhbSBoZWFsaW5nPGJyPg0KDQpl eHRyZW1lbHkgZmFzdCBhZnRlciBhbiBhY2NpZGVudCBhbmQgaGF2ZTxicj4N Cg0KbG9zdCA3IGxicy4gd2l0aG91dCB0cnlpbmchJnF1b3Q7PGJyPg0KDQpD LkIuLCBGbGFnc3RhZmYuIEFaPGJyPg0KDQombmJzcDs8YnI+DQoNClRoYW5r cyBmb3IgcmVhZGluZyBvdXIgbGV0dGVyLDxicj4NCg0KVGhlIEhHSCBTdGFm Zjxicj4NCg0KVVNBIERpdmlzaW9uPGJyPg0KDQombmJzcDs8YnI+DQoNClBT OiZuYnNwOyBUaGUgSEdIIFN0YWZmIGd1YXJhbnRlZXMgdGhlPGJyPg0KDQpo aWdoZXN0IHF1YWxpdHkgYW5kIGxvd2VzdCBwcmljZS48YnI+DQoNCiZuYnNw Ozxicj4NCg0KJm5ic3A7V2UgbWFudWZhY3R1cmUgYW5kIHNoaXAgZGlyZWN0 bHkgdG8geW91ciBkb29yLjxicj4NCg0KJm5ic3A7PGJyPg0KDQpDYWxsIHVz IG5vdyA8Yj48Zm9udCBjb2xvcj0iIzAwMDA4MCI+MS04ODgtNjIxLTczMDA8 L2ZvbnQ+PC9iPjxicj4NCg0KJm5ic3A7PGJyPg0KDQo9PT09PT09Jm5ic3A7 Jm5ic3A7IEVuZCBvZiBtZXNzYWdlID09PT09PT09Jm5ic3A7PGJyPg0KDQom bmJzcDs8YnI+DQoNCiZuYnNwOyZuYnNwOyBUaGUgZm9sbG93aW5nIHN0YXRl bWVudCBpcyBwcm92aWRlZCB0byBiZTxicj4NCg0KaW4gY29tcGxpYW5jZSB3 aXRoIGNvbW1lcmNpYWwgZW1haWwgbGF3cy48YnI+DQoNCiZuYnNwOzxicj4N Cg0KJm5ic3A7Jm5ic3A7IElmIHlvdSBkbyBub3Qgd2lzaCB0byByZWNlaXZl IGZ1cnRoZXI8YnI+DQoNCm1haWxpbmdzLCBwbGVhc2UgY2xpY2sgcmVwbHkg dG86ICB0aGVfaGdoX2NsaW5pY0BidGFtYWlsLm5ldC5jbiAgYW5kIHR5cGUg cmVtb3ZlIGluIHRoZSBzdWJqZWN0IGJveC48YnI+DQoNClRoZW4gY2xpY2sg c2VuZC48YnI+DQoNCiZuYnNwOzxicj4NCg0KJm5ic3A7Jm5ic3A7IFRoaXMg bWVzc2FnZSBpcyBpbiBmdWxsIGNvbXBsaWFuY2Ugd2l0aDxicj4NCg0KVS5T LiBGZWRlcmFsIHJlcXVpcmVtZW50cyBmb3IgY29tbWVyY2lhbDxicj4NCg0K ZW1haWwgdW5kZXIgYmlsbCBTLjE2MTggVGl0bGUgbGxsLCBTZWN0aW9uIDMw MSw8YnI+DQoNClBhcmFncmFwaCAoYSkoMikoQykgcGFzc2VkIGJ5IHRoZSAx MDV0aCBVLlMuPGJyPg0KDQpDb25ncmVzcyBhbmQgaXMgbm90IGNvbnNpZGVy ZWQgU1BBTTxicj4NCg0Kc2luY2UgaXQgaW5jbHVkZXMgYSByZW1vdmUgbWVj aGFuaXNtLio8YnI+DQoNClRoaXMgbWVzc2FnZSBpcyBub3QgaW50ZW5kZWQg Zm9yIHJlc2lkZW50cyBpbiB0aGU8YnI+DQoNCnN0YXRlcyBvZiBDQSwgTkMs IE5WLCBSSSwgVE4sIFZBICZhbXA7IFdBLjxicj4NCg0KU2NyZWVuaW5nIG9m IGFkZHJlc3NlcyBoYXMgYmVlbiBkb25lIHRvIHRoZSBiZXN0PGJyPg0KDQpv ZiBvdXIgdGVjaG5pY2FsIGFiaWxpdHkuPGJyPg0KDQombmJzcDs8YnI+DQoN CiZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZu YnNwOyZuYnNwOyZuYnNwOyZuYnNwOyZuYnNwOyBDYWxsIHVzDQoNCm5vdyA8 Yj48Zm9udCBjb2xvcj0iIzAwMDA4MCI+MS04ODgtNjIxLTczMDA8L2ZvbnQ+ PC9iPiBmb3IgeW91cjxicj4NCg0KJm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7 Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5ic3A7Jm5i c3A7IGZyZWUNCg0KSEdIIGNvbnN1bHRhdGlvbi48L3A+DQoNCjxwPjxicj4N Cg0KVGhhbmsgeW91PC9wPg0KDQo8L2JvZHk+DQoNCjwvaHRtbD4NCg0KIA0K IA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0K IA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0K IA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KIA0KDQot LQ0KDQo3NTg0bkJ3ZjItNjU3REd0VzYwMTRxQkd4My0yNjJ6bDI5 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Jul 12 1: 7:46 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 99D3937B400 for ; Fri, 12 Jul 2002 01:07:44 -0700 (PDT) Received: from mta7.pltn13.pbi.net (mta7.pltn13.pbi.net [64.164.98.8]) by mx1.FreeBSD.org (Postfix) with ESMTP id 30C9D43E4A for ; Fri, 12 Jul 2002 01:07:44 -0700 (PDT) (envelope-from mtm@pacbell.net) Received: from kokeb.ambesa.net ([64.173.11.248]) by mta7.pltn13.pbi.net (iPlanet Messaging Server 5.1 (built May 7 2001)) with ESMTP id <0GZ400HQIMKVHB@mta7.pltn13.pbi.net> for freebsd-smp@FreeBSD.ORG; Fri, 12 Jul 2002 01:07:43 -0700 (PDT) Received: from kokeb.ambesa.net (tanstaafl@localhost [IPv6:::1]) by kokeb.ambesa.net (8.12.5/8.12.5) with ESMTP id g6C87YaU020765; Fri, 12 Jul 2002 01:07:34 -0700 (PDT envelope-from mtm@kokeb.ambesa.net) Received: (from mtm@localhost) by kokeb.ambesa.net (8.12.5/8.12.5/Submit) id g6C87YWj020764; Fri, 12 Jul 2002 01:07:34 -0700 (PDT) Date: Fri, 12 Jul 2002 01:07:34 -0700 From: Mike Makonnen Subject: Re: Getting resource limits out from under Giant In-reply-to: <20020710042149.26b39b62.makonnen@pacbell.net> To: Mike Makonnen Cc: freebsd-smp@FreeBSD.ORG Message-id: <20020712010734.40b40b77.makonnen@pacbell.net> MIME-version: 1.0 X-Mailer: Sylpheed version 0.7.8 (GTK+ 1.2.10; i386-portbld-freebsd5.0) Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT References: <20020710042149.26b39b62.makonnen@pacbell.net> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Wed, 10 Jul 2002 04:21:49 -0700 Mike Makonnen wrote: > Hello folks, > > I took a stab at getting resource limits out from under Giant. > The patch is at http://home.pacbell.net/makonnen/limits.giant.patch > > Comments? Ok, I've broken it down into 3 separate diffs. http://home.pacbell.net/makonnen/limits.tar.gz 1. Touches sys/resourcevar.h - add a mutex to struct plimit - define 2 macros: LIM_LOCK and LIM_UNLOCK Touches kern/kern_resource.c - add 3 functions: limget(), limhold(), limfree() - modify limcopy() to take two arguments, the new and the old plimits structure. 2. Touches users of resource limits - These are exit1(), fork1(), proc0_init(), and acct_process() 3. Actually removes lock/unlock of Giant, and GIANT_REQUIRED Cheers, Mike Makonnen To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Jul 12 1:16:38 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id BC63E37B4B3 for ; Fri, 12 Jul 2002 01:16:32 -0700 (PDT) Received: from elvis.mu.org (elvis.mu.org [192.203.228.196]) by mx1.FreeBSD.org (Postfix) with ESMTP id 826CC43E31 for ; Fri, 12 Jul 2002 01:16:32 -0700 (PDT) (envelope-from bright@elvis.mu.org) Received: by elvis.mu.org (Postfix, from userid 1192) id 5EF5AAE23A; Fri, 12 Jul 2002 01:16:32 -0700 (PDT) Date: Fri, 12 Jul 2002 01:16:32 -0700 From: Alfred Perlstein To: Mike Makonnen Cc: freebsd-smp@FreeBSD.ORG Subject: Re: Getting resource limits out from under Giant Message-ID: <20020712081632.GJ97638@elvis.mu.org> References: <20020710042149.26b39b62.makonnen@pacbell.net> <20020712010734.40b40b77.makonnen@pacbell.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20020712010734.40b40b77.makonnen@pacbell.net> User-Agent: Mutt/1.3.27i Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org * Mike Makonnen [020712 01:07] wrote: > On Wed, 10 Jul 2002 04:21:49 -0700 > Mike Makonnen wrote: > > > Hello folks, > > > > I took a stab at getting resource limits out from under Giant. > > The patch is at http://home.pacbell.net/makonnen/limits.giant.patch > > > > Comments? > > Ok, I've broken it down into 3 separate diffs. > http://home.pacbell.net/makonnen/limits.tar.gz > > 1. Touches sys/resourcevar.h > - add a mutex to struct plimit > - define 2 macros: LIM_LOCK and LIM_UNLOCK > Touches kern/kern_resource.c > - add 3 functions: limget(), limhold(), limfree() > - modify limcopy() to take two arguments, the new and the old plimits > structure. > > 2. Touches users of resource limits > - These are exit1(), fork1(), proc0_init(), and acct_process() > > 3. Actually removes lock/unlock of Giant, and GIANT_REQUIRED Cool! Are you sure that no one can come in and mess with some other process's p->p_limit structure? If so you may need to protect it via the proc lock. -- -Alfred Perlstein [alfred@freebsd.org] 'Instead of asking why a piece of software is using "1970s technology," start asking why software is ignoring 30 years of accumulated wisdom.' Tax deductible donations for FreeBSD: http://www.freebsdfoundation.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Jul 12 2: 6:47 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id BD4F237B400 for ; Fri, 12 Jul 2002 02:06:43 -0700 (PDT) Received: from mta7.pltn13.pbi.net (mta7.pltn13.pbi.net [64.164.98.8]) by mx1.FreeBSD.org (Postfix) with ESMTP id 65F1343E4A for ; Fri, 12 Jul 2002 02:06:43 -0700 (PDT) (envelope-from mtm@pacbell.net) Received: from kokeb.ambesa.net ([64.173.11.248]) by mta7.pltn13.pbi.net (iPlanet Messaging Server 5.1 (built May 7 2001)) with ESMTP id <0GZ400H2IPB6LU@mta7.pltn13.pbi.net> for freebsd-smp@FreeBSD.ORG; Fri, 12 Jul 2002 02:06:43 -0700 (PDT) Received: from kokeb.ambesa.net (tanstaafl@localhost [IPv6:::1]) by kokeb.ambesa.net (8.12.5/8.12.5) with ESMTP id g6C96XaU021103; Fri, 12 Jul 2002 02:06:33 -0700 (PDT envelope-from mtm@kokeb.ambesa.net) Received: (from mtm@localhost) by kokeb.ambesa.net (8.12.5/8.12.5/Submit) id g6C96XSe021102; Fri, 12 Jul 2002 02:06:33 -0700 (PDT) Date: Fri, 12 Jul 2002 02:06:33 -0700 From: Mike Makonnen Subject: Re: Getting resource limits out from under Giant In-reply-to: <20020712081632.GJ97638@elvis.mu.org> To: Alfred Perlstein Cc: freebsd-smp@FreeBSD.ORG Message-id: <20020712020633.6a66f187.makonnen@pacbell.net> MIME-version: 1.0 X-Mailer: Sylpheed version 0.7.8 (GTK+ 1.2.10; i386-portbld-freebsd5.0) Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT References: <20020710042149.26b39b62.makonnen@pacbell.net> <20020712010734.40b40b77.makonnen@pacbell.net> <20020712081632.GJ97638@elvis.mu.org> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Fri, 12 Jul 2002 01:16:32 -0700 Alfred Perlstein wrote: > > Cool! Are you sure that no one can come in and mess with some other > process's p->p_limit structure? If so you may need to protect > it via the proc lock. > Hmm... looks like I missed one. The vm system touches it. Other than that, it looks like all other references are just reading from it. In the case where there's a chance I might switch plimits structures I do a proc lock in addition to locking the limits structure. Otherwise, if all I'm doing is changing the value of one of the limits I only lock the limits structure itself. Is that not enough? Cheers, Mike. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Jul 12 11: 9:39 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id BFC1837B400 for ; Fri, 12 Jul 2002 11:09:36 -0700 (PDT) Received: from mail.speakeasy.net (mail12.speakeasy.net [216.254.0.212]) by mx1.FreeBSD.org (Postfix) with ESMTP id 3214143E84 for ; Fri, 12 Jul 2002 11:09:36 -0700 (PDT) (envelope-from jhb@FreeBSD.org) Received: (qmail 4852 invoked from network); 12 Jul 2002 18:09:35 -0000 Received: from unknown (HELO server.baldwin.cx) ([216.27.160.63]) (envelope-sender ) by mail12.speakeasy.net (qmail-ldap-1.03) with DES-CBC3-SHA encrypted SMTP for ; 12 Jul 2002 18:09:35 -0000 Received: from laptop.baldwin.cx (gw1.twc.weather.com [216.133.140.1]) by server.baldwin.cx (8.11.6/8.11.6) with ESMTP id g6CI9Y036705; Fri, 12 Jul 2002 14:09:34 -0400 (EDT) (envelope-from jhb@FreeBSD.org) Message-ID: X-Mailer: XFMail 1.5.2 on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <20020712020633.6a66f187.makonnen@pacbell.net> Date: Fri, 12 Jul 2002 14:09:32 -0400 (EDT) From: John Baldwin To: Mike Makonnen Subject: Re: Getting resource limits out from under Giant Cc: freebsd-smp@FreeBSD.ORG, Alfred Perlstein Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On 12-Jul-2002 Mike Makonnen wrote: > On Fri, 12 Jul 2002 01:16:32 -0700 > Alfred Perlstein wrote: > >> >> Cool! Are you sure that no one can come in and mess with some other >> process's p->p_limit structure? If so you may need to protect >> it via the proc lock. >> > > Hmm... looks like I missed one. The vm system touches it. Other than > that, > it looks like all other references are just reading from it. > > In the case where there's a chance I might switch plimits structures > I do a proc lock in addition to locking the limits structure. Otherwise, > if all I'm doing is changing the value of one of the limits I only lock > the > limits structure itself. Is that not enough? You need to ensure you have a stable reference to the limits structure. If you are dereferencing p_limit you need to use the proc lock to protect reading that pointer. The lock also ensure the value is up to date. -- John Baldwin <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Jul 13 15:23:38 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 7CEBB37B400; Sat, 13 Jul 2002 15:23:35 -0700 (PDT) Received: from mta6.snfc21.pbi.net (mta6.snfc21.pbi.net [206.13.28.240]) by mx1.FreeBSD.org (Postfix) with ESMTP id 21BA643E3B; Sat, 13 Jul 2002 15:23:35 -0700 (PDT) (envelope-from mtm@pacbell.net) Received: from kokeb.ambesa.net ([64.166.85.137]) by mta6.snfc21.pbi.net (iPlanet Messaging Server 5.1 (built May 7 2001)) with ESMTP id <0GZ700F2PKVAG3@mta6.snfc21.pbi.net>; Sat, 13 Jul 2002 15:23:35 -0700 (PDT) Received: from kokeb.ambesa.net (tanstaafl@localhost [IPv6:::1]) by kokeb.ambesa.net (8.12.5/8.12.5) with ESMTP id g6DMNK0Y006510; Sat, 13 Jul 2002 15:23:20 -0700 (PDT envelope-from mtm@kokeb.ambesa.net) Received: (from mtm@localhost) by kokeb.ambesa.net (8.12.5/8.12.5/Submit) id g6DMNKM1006509; Sat, 13 Jul 2002 15:23:20 -0700 (PDT) Date: Sat, 13 Jul 2002 15:23:19 -0700 From: Mike Makonnen Subject: Re: Getting resource limits out from under Giant In-reply-to: To: John Baldwin Cc: freebsd-smp@FreeBSD.ORG, bright@mu.org Message-id: <20020713152319.75c416a2.makonnen@pacbell.net> MIME-version: 1.0 X-Mailer: Sylpheed version 0.7.8 (GTK+ 1.2.10; i386-portbld-freebsd5.0) Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT References: <20020712020633.6a66f187.makonnen@pacbell.net> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Fri, 12 Jul 2002 14:09:32 -0400 (EDT) John Baldwin wrote: > > > You need to ensure you have a stable reference to the limits > structure. If you are dereferencing p_limit you need to use the proc > lock to protect reading that pointer. The lock also ensure the value > is up to date. > I see what you mean: foolimit = p->p_limit->pl_rlimit[FOOLIMIT].rlim_cur; is OK. But plim = p->p_limit; ... ... foolimit = p->p_limit->pl_rlimit[FOOLIMIT].rlim_cur; needs to lock both the proc and the limit. Even better, it should use the first form if it can. The part I am not clear about is when modifying a value in the limits structure. There is one instance in the patch where I do this: a: allocate some memory lock proc lock p->plimit check if we must copy-on-write do so if necessary unlock plimit unlock proc free some memory lock p->plimit if conditions have changed and we must do a copy-on-write goto a modify limits as necessary My reasoning for dropping both locks in the middle and reaquiring only the limit lock is that anyone who might mess with our p->plimit must aquire both the proc and the limit lock. So, as long as I have the limit lock, there isn't any chance that it will go stale on me. I will work on the patch some more this weekend and repost. Cheers, Mike Makonnen To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Jul 13 15:45: 5 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id BCCA537B400; Sat, 13 Jul 2002 15:45:03 -0700 (PDT) Received: from mta7.pltn13.pbi.net (mta7.pltn13.pbi.net [64.164.98.8]) by mx1.FreeBSD.org (Postfix) with ESMTP id 64E6943E42; Sat, 13 Jul 2002 15:45:03 -0700 (PDT) (envelope-from mtm@pacbell.net) Received: from kokeb.ambesa.net ([64.166.85.137]) by mta7.pltn13.pbi.net (iPlanet Messaging Server 5.1 (built May 7 2001)) with ESMTP id <0GZ7007QOLV2LO@mta7.pltn13.pbi.net>; Sat, 13 Jul 2002 15:45:03 -0700 (PDT) Received: from kokeb.ambesa.net (tanstaafl@localhost [IPv6:::1]) by kokeb.ambesa.net (8.12.5/8.12.5) with ESMTP id g6DMim0Y006630; Sat, 13 Jul 2002 15:44:48 -0700 (PDT envelope-from mtm@kokeb.ambesa.net) Received: (from mtm@localhost) by kokeb.ambesa.net (8.12.5/8.12.5/Submit) id g6DMimX4006629; Sat, 13 Jul 2002 15:44:48 -0700 (PDT) Date: Sat, 13 Jul 2002 15:44:48 -0700 From: Mike Makonnen Subject: Re: Getting resource limits out from under Giant In-reply-to: <20020713152319.75c416a2.makonnen@pacbell.net> To: Mike Makonnen Cc: jhb@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG, bright@mu.org Message-id: <20020713154448.37b47d4a.makonnen@pacbell.net> MIME-version: 1.0 X-Mailer: Sylpheed version 0.7.8 (GTK+ 1.2.10; i386-portbld-freebsd5.0) Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT References: <20020712020633.6a66f187.makonnen@pacbell.net> <20020713152319.75c416a2.makonnen@pacbell.net> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Sat, 13 Jul 2002 15:23:19 -0700 Mike Makonnen wrote: > On Fri, 12 Jul 2002 14:09:32 -0400 (EDT) > John Baldwin wrote: > > > > > > > You need to ensure you have a stable reference to the limits > > structure. If you are dereferencing p_limit you need to use the proc > > lock to protect reading that pointer. The lock also ensure the > > value is up to date. > > > > I see what you mean: > > foolimit = p->p_limit->pl_rlimit[FOOLIMIT].rlim_cur; > > is OK. But > > plim = p->p_limit; > ... > ... > foolimit = p->p_limit->pl_rlimit[FOOLIMIT].rlim_cur; uhh... I meant foolimit = plim->pl_rlimit[FOOLIMIT].rlim_cur; To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Jul 13 19:50:28 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 4F79937B400 for ; Sat, 13 Jul 2002 19:50:25 -0700 (PDT) Received: from mail.speakeasy.net (mail12.speakeasy.net [216.254.0.212]) by mx1.FreeBSD.org (Postfix) with ESMTP id D996943E58 for ; Sat, 13 Jul 2002 19:50:24 -0700 (PDT) (envelope-from jhb@FreeBSD.org) Received: (qmail 6448 invoked from network); 14 Jul 2002 02:50:21 -0000 Received: from unknown (HELO server.baldwin.cx) ([216.27.160.63]) (envelope-sender ) by mail12.speakeasy.net (qmail-ldap-1.03) with DES-CBC3-SHA encrypted SMTP for ; 14 Jul 2002 02:50:21 -0000 Received: from laptop.baldwin.cx (laptop.baldwin.cx [192.168.0.4]) by server.baldwin.cx (8.11.6/8.11.6) with ESMTP id g6E2oA043207; Sat, 13 Jul 2002 22:50:10 -0400 (EDT) (envelope-from jhb@FreeBSD.org) Message-ID: X-Mailer: XFMail 1.5.2 on FreeBSD X-Priority: 3 (Normal) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <20020713152319.75c416a2.makonnen@pacbell.net> Date: Sat, 13 Jul 2002 22:50:20 -0400 (EDT) From: John Baldwin To: Mike Makonnen Subject: Re: Getting resource limits out from under Giant Cc: bright@mu.org, freebsd-smp@FreeBSD.ORG Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On 13-Jul-2002 Mike Makonnen wrote: > On Fri, 12 Jul 2002 14:09:32 -0400 (EDT) > John Baldwin wrote: > >> >> >> You need to ensure you have a stable reference to the limits >> structure. If you are dereferencing p_limit you need to use the proc >> lock to protect reading that pointer. The lock also ensure the value >> is up to date. >> > > I see what you mean: > > foolimit = p->p_limit->pl_rlimit[FOOLIMIT].rlim_cur; > > is OK. But > > plim = p->p_limit; > ... > ... > foolimit = p->p_limit->pl_rlimit[FOOLIMIT].rlim_cur; > > needs to lock both the proc and the limit. Even better, it should use > the first form if it can. > > The part I am not clear about is when modifying a value in the limits > structure. > There is one instance in the patch where I do this: > > a: > allocate some memory > lock proc > lock p->plimit > check if we must copy-on-write > do so if necessary > unlock plimit > unlock proc > free some memory > lock p->plimit > if conditions have changed and we must do a copy-on-write > goto a > modify limits as necessary > > My reasoning for dropping both locks in the middle and reaquiring only > the > limit lock is that anyone who might mess with our p->plimit must > aquire both the proc and the limit lock. So, as long as I have the limit > lock, there isn't any chance that it will go stale on me. > > I will work on the patch some more this weekend and repost. Why not just do a simple change: don't drop the locks so quick, but go ahead and do the modify limit step after using your preallocated memory to do the copy-on-write. Then you can drop your locks and free any memory using cached pointers. No need for a loop and much simpler. Basically, all you need to do is move your 'free some memory' step down to the bottom, then you can immediately go from the copy-on-write step to the modify limits step. -- John Baldwin <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Jul 13 22:21:52 2002 Delivered-To: freebsd-smp@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8045C37B400; Sat, 13 Jul 2002 22:21:50 -0700 (PDT) Received: from mta7.pltn13.pbi.net (mta7.pltn13.pbi.net [64.164.98.8]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2992E43E67; Sat, 13 Jul 2002 22:21:50 -0700 (PDT) (envelope-from mtm@pacbell.net) Received: from kokeb.ambesa.net ([64.166.85.137]) by mta7.pltn13.pbi.net (iPlanet Messaging Server 5.1 (built May 7 2001)) with ESMTP id <0GZ800J2348BM4@mta7.pltn13.pbi.net>; Sat, 13 Jul 2002 22:21:48 -0700 (PDT) Received: from kokeb.ambesa.net (tanstaafl@localhost [IPv6:::1]) by kokeb.ambesa.net (8.12.5/8.12.5) with ESMTP id g6E5LW0Y008069; Sat, 13 Jul 2002 22:21:32 -0700 (PDT envelope-from mtm@kokeb.ambesa.net) Received: (from mtm@localhost) by kokeb.ambesa.net (8.12.5/8.12.5/Submit) id g6E5LVQR008068; Sat, 13 Jul 2002 22:21:31 -0700 (PDT) Date: Sat, 13 Jul 2002 22:21:31 -0700 From: Mike Makonnen Subject: Re: Getting resource limits out from under Giant In-reply-to: To: John Baldwin Cc: bright@mu.org, freebsd-smp@FreeBSD.ORG Message-id: <20020713222131.6ebf07c8.makonnen@pacbell.net> MIME-version: 1.0 X-Mailer: Sylpheed version 0.7.8 (GTK+ 1.2.10; i386-portbld-freebsd5.0) Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT References: <20020713152319.75c416a2.makonnen@pacbell.net> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Sat, 13 Jul 2002 22:50:20 -0400 (EDT) John Baldwin wrote: > > Why not just do a simple change: don't drop the locks so quick, but go > ahead and do the modify limit step after using your preallocated > memory to do the copy-on-write. Then you can drop your locks and free > any memory using cached pointers. No need for a loop and much > simpler. Basically, all you need to do is move your 'free some memory' > step down to the bottom, then you can immediately go from the > copy-on-write step to the modify limits step. > I do that in kern/kern_acct.c:acct_process(), which only requires an assignment statemt. But in kern/kern_resource.c:dosetrlimit() there are 2 reasons why I drop the proc lock. I will defer to your judgement as to whether they're valid or not. 1. There is one code path, when changing the stack size limit, where it goes off into vm land. I thought it might be better to drop the proc lock so that we don't unnecesarily block other threads. 2. I chose to use a mtx_pool mutex. My understanding is that I may not aquire another lock while holding one of these. But, as part of the stack size limit code path, we aquire a vm_map_lock. While I suppose I could use a regular mutex, I thought the additional complexity was worth the treadoff in structural/code overhead. Cheers, Mike Makonnen To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message