From owner-freebsd-smp Tue Sep 5 17:59: 5 2000 Delivered-To: freebsd-smp@freebsd.org Received: from magnesium.net (toxic.magnesium.net [207.154.84.15]) by hub.freebsd.org (Postfix) with SMTP id 7DF9637B42C for ; Tue, 5 Sep 2000 17:59:02 -0700 (PDT) Received: (qmail 64820 invoked by uid 1142); 6 Sep 2000 00:59:01 -0000 Date: 5 Sep 2000 17:59:01 -0700 Date: Tue, 5 Sep 2000 17:58:32 -0700 From: Jason Evans To: current@freebsd.org, smp@freebsd.org Subject: HEADS UP: SMP code commit iminent Message-ID: <20000905175832.A437@blitz.canonware.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In compliance with the SMP project announcement (archived at http://people.freebsd.org/~jasone/smp/smp_project_announcement.txt), this email is a minimum 24 hour notice that SMP code will be committed to -current. A large patch including major changes to how the kernel works will be checked in sometime after 6 September, 18:00 PST. Also in compliance with the SMP project announcement, a static tag will be created between the above-mentioned time and when the actual commit is done. If you have questions or issues, make sure to first read the original project announcement (URL above) before sending email to the lists. For more extensive information about the SMP project, see the project web page at: http://people.freebsd.org/~jasone/smp/ Let the fun begin! Jason Evans SMP Project Manager To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Sep 5 18:37:43 2000 Delivered-To: freebsd-smp@freebsd.org Received: from dragon.nuxi.com (trang.nuxi.com [209.152.133.57]) by hub.freebsd.org (Postfix) with ESMTP id 9809337B422; Tue, 5 Sep 2000 18:37:40 -0700 (PDT) Received: (from obrien@localhost) by dragon.nuxi.com (8.9.3/8.9.1) id SAA87418; Tue, 5 Sep 2000 18:37:38 -0700 (PDT) (envelope-from obrien) Date: Tue, 5 Sep 2000 18:37:38 -0700 From: "David O'Brien" To: Jason Evans Cc: current@freebsd.org, smp@freebsd.org Subject: Re: HEADS UP: SMP code commit iminent Message-ID: <20000905183738.A87381@dragon.nuxi.com> Reply-To: obrien@freebsd.org References: <20000905175832.A437@blitz.canonware.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i In-Reply-To: <20000905175832.A437@blitz.canonware.com>; from jasone@canonware.com on Tue, Sep 05, 2000 at 05:58:32PM -0700 X-Operating-System: FreeBSD 5.0-CURRENT Organization: The NUXI BSD group X-Pgp-Rsa-Fingerprint: B7 4D 3E E9 11 39 5F A3 90 76 5D 69 58 D9 98 7A X-Pgp-Rsa-Keyid: 1024/34F9F9D5 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, Sep 05, 2000 at 05:58:32PM -0700, Jason Evans wrote: > this email is a minimum 24 hour notice that SMP code will be committed > to -current. What is the status of the Alpha bits? Will we have a working kernel after the commit, or should we site tight for a week while the Alpha bits are tweaked into working status? -- -- David (obrien@FreeBSD.org) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Sep 5 18:59:51 2000 Delivered-To: freebsd-smp@freebsd.org Received: from magnesium.net (toxic.magnesium.net [207.154.84.15]) by hub.freebsd.org (Postfix) with SMTP id 26E2337B424 for ; Tue, 5 Sep 2000 18:59:48 -0700 (PDT) Received: (qmail 67276 invoked by uid 1142); 6 Sep 2000 01:59:47 -0000 Date: 5 Sep 2000 18:59:47 -0700 Date: Tue, 5 Sep 2000 18:59:20 -0700 From: Jason Evans To: David O'Brien Cc: current@freebsd.org, smp@freebsd.org Subject: Re: HEADS UP: SMP code commit iminent Message-ID: <20000905185920.C437@blitz.canonware.com> References: <20000905175832.A437@blitz.canonware.com> <20000905183738.A87381@dragon.nuxi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: <20000905183738.A87381@dragon.nuxi.com>; from obrien@freebsd.org on Tue, Sep 05, 2000 at 06:37:38PM -0700 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, Sep 05, 2000 at 06:37:38PM -0700, David O'Brien wrote: > What is the status of the Alpha bits? Will we have a working kernel > after the commit, or should we site tight for a week while the Alpha bits > are tweaked into working status? It should work (compile and run), though interrupt threads will not be enabled. Jason To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Sep 5 21:57:21 2000 Delivered-To: freebsd-smp@freebsd.org Received: from feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (Postfix) with ESMTP id 0AF4A37B424; Tue, 5 Sep 2000 21:57:18 -0700 (PDT) Received: from beppo.feral.com (beppo [192.67.166.79]) by feral.com (8.9.3/8.9.3) with ESMTP id VAA21352; Tue, 5 Sep 2000 21:57:17 -0700 Date: Tue, 5 Sep 2000 21:57:05 -0700 (PDT) From: Matthew Jacob Reply-To: mjacob@feral.com To: "David O'Brien" Cc: Jason Evans , current@FreeBSD.ORG, smp@FreeBSD.ORG Subject: Re: HEADS UP: SMP code commit iminent In-Reply-To: <20000905183738.A87381@dragon.nuxi.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Progress will be more rapid with things checked in than not, as long as Jason's statement about "the (alpha) system will run" after the checkin. Jason- I think we'd all appreciate a UTC timestamp suitable for -D that we can all use to checkout stuff prior to the big change. -matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Sep 5 22:12:14 2000 Delivered-To: freebsd-smp@freebsd.org Received: from dragon.nuxi.com (trang.nuxi.com [209.152.133.57]) by hub.freebsd.org (Postfix) with ESMTP id EF90E37B422; Tue, 5 Sep 2000 22:12:10 -0700 (PDT) Received: (from obrien@localhost) by dragon.nuxi.com (8.9.3/8.9.1) id WAA88466; Tue, 5 Sep 2000 22:12:07 -0700 (PDT) (envelope-from obrien) Date: Tue, 5 Sep 2000 22:12:07 -0700 From: "David O'Brien" To: Matthew Jacob Cc: current@FreeBSD.ORG, smp@FreeBSD.ORG Subject: Re: HEADS UP: SMP code commit iminent Message-ID: <20000905221206.A88437@dragon.nuxi.com> Reply-To: obrien@FreeBSD.ORG References: <20000905183738.A87381@dragon.nuxi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i In-Reply-To: ; from mjacob@feral.com on Tue, Sep 05, 2000 at 09:57:05PM -0700 X-Operating-System: FreeBSD 5.0-CURRENT Organization: The NUXI BSD group X-Pgp-Rsa-Fingerprint: B7 4D 3E E9 11 39 5F A3 90 76 5D 69 58 D9 98 7A X-Pgp-Rsa-Keyid: 1024/34F9F9D5 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, Sep 05, 2000 at 09:57:05PM -0700, Matthew Jacob wrote: > Jason- I think we'd all appreciate a UTC timestamp suitable for -D that we can > all use to checkout stuff prior to the big change. There will be a TAG for that to make life easier. -- -- David (obrien@FreeBSD.org) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Sep 5 23: 9:37 2000 Delivered-To: freebsd-smp@freebsd.org Received: from thelab.hub.org (CDR27-115.accesscable.net [24.138.27.115]) by hub.freebsd.org (Postfix) with ESMTP id D7F4E37B43E for ; Tue, 5 Sep 2000 23:09:34 -0700 (PDT) Received: from localhost (scrappy@localhost) by thelab.hub.org (8.11.0/8.11.0) with ESMTP id e86698X12399 for ; Wed, 6 Sep 2000 03:09:09 -0300 (ADT) (envelope-from scrappy@hub.org) X-Authentication-Warning: thelab.hub.org: scrappy owned process doing -bs Date: Wed, 6 Sep 2000 03:09:07 -0300 (ADT) From: The Hermit Hacker To: freebsd-smp@freebsd.org Subject: Sept 5th patch ... Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Sweet ... downloaded it tonight and installed it on my Dual-Celeron ... smooth as can be ... Looking at the web site referenced in the status report posted on -current, there are some old '97 benchmarks ... has anyone done anything recent? More curious about before/after these patches ... I read that performance wasn't a key target for these changes, but I'm wondering if there is there expected to be any performance improvements just by the fact of the work that was done (ie. schedualer is now better or some such) ... Nice work folks ... :) Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy Systems Administrator @ hub.org primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Sep 5 23:39:59 2000 Delivered-To: freebsd-smp@freebsd.org Received: from netplex.com.au (adsl-63-207-30-186.dsl.snfc21.pacbell.net [63.207.30.186]) by hub.freebsd.org (Postfix) with ESMTP id 0B69337B43E for ; Tue, 5 Sep 2000 23:39:13 -0700 (PDT) Received: from netplex.com.au (peter@localhost [127.0.0.1]) by netplex.com.au (8.11.0/8.9.3) with ESMTP id e866ckG49084; Tue, 5 Sep 2000 23:38:51 -0700 (PDT) (envelope-from peter@netplex.com.au) Message-Id: <200009060638.e866ckG49084@netplex.com.au> X-Mailer: exmh version 2.1.1 10/15/1999 To: The Hermit Hacker Cc: freebsd-smp@FreeBSD.ORG Subject: Re: Sept 5th patch ... In-Reply-To: Date: Tue, 05 Sep 2000 23:38:46 -0700 From: Peter Wemm Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org The Hermit Hacker wrote: > > Sweet ... downloaded it tonight and installed it on my Dual-Celeron > ... smooth as can be ... > > Looking at the web site referenced in the status report posted on > -current, there are some old '97 benchmarks ... has anyone done anything > recent? More curious about before/after these patches ... > > I read that performance wasn't a key target for these changes, but > I'm wondering if there is there expected to be any performance > improvements just by the fact of the work that was done (ie. schedualer is > now better or some such) ... A buildworld, for example, is quite a bit slower (eg: 33 minutes -> 45 minutes on one particular quad xeon). But keep in mind that this is just about the worst possible way to do this. The reasons for doing it this way: 1: It gives us a validatable, stable, understandable baseline to work from. If something wierd happens, we have a baseline to fall back to and compare against. 2: This implements the basic functionality. Everything else is built on top of this and can be done in parallel from here on. 3: For example, fast interrupts can be done in parallel with other pushdown work (VFS, network, device drivers, etc). Performance is going to suck and stay sucking until the proper lazy context switch interrupt threads are done. We have the BSD/OS code to use as a template, but unlike BSD/OS we have a fallback in place to buy us time to debug the lazy context switch code. Remember, the i386 kernel is now running with *NO SUCH THING AS SPLxxx()* - interrupt handlers have to do a process context switch and compete for the giant lock while before doing their interrupt work. There is a lot of overhead ehere. Considering just how suboptimal this initial checkin will be, its been a pleasant suprise that it wasn't a larger impact than it is. We are expecting this and it will not take long to get it back to at meet or exceed the performance that it was at before. Other reentrancy improvements will offset the overheads imposed by the slow interrupt threads, so overall we will start winning long before we break even with interrupt code efficiency. Once the Alpha's interrupt threads are done, we can remove spl's from the source. On the i386 they will be do-nothing inlines or macros. The actual splxxx()/ splx() pairs will probably remain for a while as they are excellent markers of places that need to be checked for reentrancy. I personally am quite excited. Once it hits the tree, the number of developers who can be directly involved goes from "a few" to "everybody interested". Cheers, -Peter -- Peter Wemm - peter@FreeBSD.org; peter@yahoo-inc.com; peter@netplex.com.au "All of this is for nothing if we don't go to the stars" - JMS/B5 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 6 8:50:23 2000 Delivered-To: freebsd-smp@freebsd.org Received: from thelab.hub.org (CDR27-115.accesscable.net [24.138.27.115]) by hub.freebsd.org (Postfix) with ESMTP id 9821937B423 for ; Wed, 6 Sep 2000 08:50:20 -0700 (PDT) Received: from localhost (scrappy@localhost) by thelab.hub.org (8.11.0/8.11.0) with ESMTP id e86FnTP30806; Wed, 6 Sep 2000 12:49:34 -0300 (ADT) (envelope-from scrappy@hub.org) X-Authentication-Warning: thelab.hub.org: scrappy owned process doing -bs Date: Wed, 6 Sep 2000 12:49:29 -0300 (ADT) From: The Hermit Hacker To: Peter Wemm Cc: freebsd-smp@FreeBSD.ORG Subject: Re: Sept 5th patch ... In-Reply-To: <200009060638.e866ckG49084@netplex.com.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Performance is going to suck and stay sucking until the proper lazy > context switch interrupt threads are done. We have the BSD/OS code to > use as a template, but unlike BSD/OS we have a fallback in place to > buy us time to debug the lazy context switch code. Well, I wouldn't say that performance is 'sucking' right now with the patch installed, so if this is the worst its gonna get ... fantastic job on reducing the effect of the 'suck' :) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 6 10:36:54 2000 Delivered-To: freebsd-smp@freebsd.org Received: from smtp02.iafrica.com (smtp02.iafrica.com [196.7.0.140]) by hub.freebsd.org (Postfix) with ESMTP id B634D37B423 for ; Wed, 6 Sep 2000 10:36:48 -0700 (PDT) Received: from [196.7.18.138] (helo=grimreaper.grondar.za ident=root) by smtp02.iafrica.com with esmtp (Exim 1.92 #1) id 13Wj7p-000EuV-00; Wed, 6 Sep 2000 19:36:42 +0200 Received: from grimreaper.grondar.za (mark@localhost [127.0.0.1]) by grimreaper.grondar.za (8.11.0/8.11.0) with ESMTP id e86Hbqd08340; Wed, 6 Sep 2000 19:37:55 +0200 (SAST) (envelope-from mark@grimreaper.grondar.za) Message-Id: <200009061737.e86Hbqd08340@grimreaper.grondar.za> To: Peter Wemm Cc: freebsd-smp@FreeBSD.ORG Subject: Re: Sept 5th patch ... References: <200009060638.e866ckG49084@netplex.com.au> In-Reply-To: <200009060638.e866ckG49084@netplex.com.au> ; from Peter Wemm "Tue, 05 Sep 2000 23:38:46 MST." Date: Wed, 06 Sep 2000 19:37:51 +0200 From: Mark Murray Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Remember, the i386 kernel is now running with *NO SUCH THING AS SPLxxx()* - > interrupt handlers have to do a process context switch and compete for the > giant lock while before doing their interrupt work. There is a lot of > overhead ehere. So, are there going to be some documented guidelines for us neophyte kernel-hackers? I have just gotten used to the splfoo()/splx() idea, and now its gone. How, now, do we write canonical code that doesn't futz with itself when interrupts/threads/processes happen? M -- Mark Murray Join the anti-SPAM movement: http://www.cauce.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 6 11: 5:31 2000 Delivered-To: freebsd-smp@freebsd.org Received: from tristero.cryptocourier.com (black-3.dsl.speakeasy.net [216.231.56.189]) by hub.freebsd.org (Postfix) with SMTP id 8B54A37B422 for ; Wed, 6 Sep 2000 11:05:20 -0700 (PDT) Received: (qmail 17891 invoked from network); 6 Sep 2000 18:08:03 -0000 Received: from cx435051-d.fed1.sdca.home.com (HELO thirteen.flargh.com) (u6030@24.5.33.77) by tristero.cryptocourier.com with SMTP; 6 Sep 2000 18:08:03 -0000 To: smp@freebsd.org Subject: Re: Sept 5th patch ... In-Reply-To: Your message of "Wed, 06 Sep 2000 19:37:51 +0200." <200009061737.e86Hbqd08340@grimreaper.grondar.za> Date: Wed, 06 Sep 2000 11:00:15 -0700 From: joel boutros Message-Id: <20000906180520.8B54A37B422@hub.freebsd.org> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > So, are there going to be some documented guidelines for us neophyte > kernel-hackers? I have just gotten used to the splfoo()/splx() idea, > and now its gone. How, now, do we write canonical code that doesn't > futz with itself when interrupts/threads/processes happen? the general trick is locking critical sections of code or data. previously when you did an splXXX(), you were assured that you weren't going to get interrupted by anything of specified priority or lower, so it was safe to assume your data probably wasn't going to change out from under you. now, you mark the critical sections as in-use, and if something else does come in, it checks to see if the section is "available" and, if not, figures out some strategy to deal with it. your critical section can be finely-grained (data such as members of a structure or elements of a queue) or coarsely-grained (holding an exclusive lock on an entire subsystem, like the network stack). and your locks can appear in a number of varieties. the strategy employed depends on your implementation. i'd recommend picking up a book on threads programming. the real gist of the change is that instead of telling the system you don't want to be interrupted, you make your code more tolerant of interruptions, which is what multithreading really requires. - joel To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 6 11:32:55 2000 Delivered-To: freebsd-smp@freebsd.org Received: from mta6.snfc21.pbi.net (mta6.snfc21.pbi.net [206.13.28.240]) by hub.freebsd.org (Postfix) with ESMTP id 6175637B423 for ; Wed, 6 Sep 2000 11:32:53 -0700 (PDT) Received: from pacbell.net ([63.194.212.226]) by mta6.snfc21.pbi.net (Sun Internet Mail Server sims.3.5.2000.01.05.12.18.p9) with ESMTP id <0G0H006XTA4XLO@mta6.snfc21.pbi.net> for smp@freebsd.org; Wed, 6 Sep 2000 11:31:45 -0700 (PDT) Date: Wed, 06 Sep 2000 11:26:36 -0700 From: Adagio Vangogh Subject: Re: Sept 5th patch ... To: joel boutros Cc: smp@freebsd.org Reply-To: adagio_v@pacbell.net Message-id: <39B68C5C.3C21C55E@pacbell.net> MIME-version: 1.0 X-Mailer: Mozilla 4.74 [en] (Win98; U) Content-type: text/plain; charset=us-ascii Content-transfer-encoding: 7bit X-Accept-Language: en References: <20000906180520.8B54A37B422@hub.freebsd.org> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org joel boutros wrote: > the general trick is locking critical sections of code or data. > previously when you did an splXXX(), you were assured that you > weren't going to get interrupted by anything of specified priority > or lower, so it was safe to assume your data probably wasn't going > to change out from under you. FreeBSD has always used the notion of a "class" mask for interrupts and not priorities. Ie, all network devices fell under the net class and only a member of this class would block the rest of the members in the same class. How's this handled? --adagio v. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 6 11:38:18 2000 Delivered-To: freebsd-smp@freebsd.org Received: from fw.wintelcom.net (ns1.wintelcom.net [209.1.153.20]) by hub.freebsd.org (Postfix) with ESMTP id 5F87337B423 for ; Wed, 6 Sep 2000 11:38:16 -0700 (PDT) Received: (from bright@localhost) by fw.wintelcom.net (8.10.0/8.10.0) id e86IcFP17675; Wed, 6 Sep 2000 11:38:15 -0700 (PDT) Date: Wed, 6 Sep 2000 11:38:14 -0700 From: Alfred Perlstein To: Adagio Vangogh Cc: joel boutros , smp@FreeBSD.ORG Subject: Re: Sept 5th patch ... Message-ID: <20000906113814.K18862@fw.wintelcom.net> References: <20000906180520.8B54A37B422@hub.freebsd.org> <39B68C5C.3C21C55E@pacbell.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.4i In-Reply-To: <39B68C5C.3C21C55E@pacbell.net>; from adagio_v@pacbell.net on Wed, Sep 06, 2000 at 11:26:36AM -0700 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org * Adagio Vangogh [000906 11:33] wrote: > > > joel boutros wrote: > > > the general trick is locking critical sections of code or data. > > previously when you did an splXXX(), you were assured that you > > weren't going to get interrupted by anything of specified priority > > or lower, so it was safe to assume your data probably wasn't going > > to change out from under you. > > FreeBSD has always used the notion of a "class" mask for interrupts > and not priorities. Ie, all network devices fell under the net class and > > only a member of this class would block the rest of the members in the > same class. How's this handled? It's not going to be. :) We're doing fine grained locks. -- -Alfred Perlstein - [bright@wintelcom.net|alfred@freebsd.org] "I have the heart of a child; I keep it in a jar on my desk." To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 6 11:39:39 2000 Delivered-To: freebsd-smp@freebsd.org Received: from crewsoft.com (ns.aenet.net [157.22.214.1]) by hub.freebsd.org (Postfix) with ESMTP id 2532037B424 for ; Wed, 6 Sep 2000 11:39:37 -0700 (PDT) Received: from [63.197.8.222] (HELO wireless-networks.com) by crewsoft.com (CommuniGate Pro SMTP 3.3b5) with ESMTP id 285138; Wed, 06 Sep 2000 11:41:17 -0700 Message-ID: <39B68F61.6B528751@wireless-networks.com> Date: Wed, 06 Sep 2000 11:39:29 -0700 From: Cedric Berger X-Mailer: Mozilla 4.73 [en] (Windows NT 5.0; U) X-Accept-Language: en MIME-Version: 1.0 To: adagio_v@pacbell.net Cc: smp@freebsd.org Subject: Re: Sept 5th patch ... References: <20000906180520.8B54A37B422@hub.freebsd.org> <39B68C5C.3C21C55E@pacbell.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > FreeBSD has always used the notion of a "class" mask for interrupts > and not priorities. Ie, all network devices fell under the net class and > > only a member of this class would block the rest of the members in the > same class. How's this handled? I guess that you have to figure out why it this "class" of device is protected at some point, (for example, access to the stack input queue), and protect that specific *object* (i.e. the stack input queue) with a mutex. Cedric To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 6 13:13:32 2000 Delivered-To: freebsd-smp@freebsd.org Received: from tele-post-20.mail.demon.net (tele-post-20.mail.demon.net [194.217.242.20]) by hub.freebsd.org (Postfix) with ESMTP id 28DF737B423; Wed, 6 Sep 2000 13:13:26 -0700 (PDT) Received: from nlsys.demon.co.uk ([158.152.125.33] helo=herring.nlsystems.com) by tele-post-20.mail.demon.net with esmtp (Exim 2.12 #2) id 13WlZK-0005Nb-0K; Wed, 6 Sep 2000 20:13:14 +0000 Received: from salmon.nlsystems.com (salmon.nlsystems.com [10.0.0.3]) by herring.nlsystems.com (8.9.3/8.8.8) with ESMTP id VAA04101; Wed, 6 Sep 2000 21:13:53 +0100 (BST) (envelope-from dfr@nlsystems.com) Date: Wed, 6 Sep 2000 21:13:49 +0100 (BST) From: Doug Rabson To: "David O'Brien" Cc: Jason Evans , current@freebsd.org, smp@freebsd.org Subject: Re: HEADS UP: SMP code commit iminent In-Reply-To: <20000905183738.A87381@dragon.nuxi.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, 5 Sep 2000, David O'Brien wrote: > On Tue, Sep 05, 2000 at 05:58:32PM -0700, Jason Evans wrote: > > this email is a minimum 24 hour notice that SMP code will be committed > > to -current. > > What is the status of the Alpha bits? Will we have a working kernel > after the commit, or should we site tight for a week while the Alpha bits > are tweaked into working status? The code should boot - if not, I will follow up with a few cleanup commits. The next job for the alpha is to overhaul the interrupt code to support irq threads. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 20 8348 3944 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 6 13:43:36 2000 Delivered-To: freebsd-smp@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id B560237B424; Wed, 6 Sep 2000 13:43:33 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id NAA73340; Wed, 6 Sep 2000 13:43:21 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009062043.NAA73340@pike.osd.bsdi.com> Subject: Re: HEADS UP: SMP code commit iminent In-Reply-To: from Doug Rabson at "Sep 6, 2000 09:13:49 pm" To: Doug Rabson Date: Wed, 6 Sep 2000 13:43:20 -0700 (PDT) Cc: "David O'Brien" , Jason Evans , current@FreeBSD.ORG, smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Doug Rabson wrote: > On Tue, 5 Sep 2000, David O'Brien wrote: > > > On Tue, Sep 05, 2000 at 05:58:32PM -0700, Jason Evans wrote: > > > this email is a minimum 24 hour notice that SMP code will be committed > > > to -current. > > > > What is the status of the Alpha bits? Will we have a working kernel > > after the commit, or should we site tight for a week while the Alpha bits > > are tweaked into working status? > > The code should boot - if not, I will follow up with a few cleanup > commits. The next job for the alpha is to overhaul the interrupt code to > support irq threads. The current patchset finished a 3-hour buildworld w/o problems. Sometime after the world had finished while I was asleep it had a kernel trap with a memory management fault however. :( 3 hours of reliable uptime in multi- user should be enough stability for development however. > -- > Doug Rabson Mail: dfr@nlsystems.com > Nonlinear Systems Ltd. Phone: +44 20 8348 3944 -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc "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 Wed Sep 6 15:35: 7 2000 Delivered-To: freebsd-smp@freebsd.org Received: from zero.arkaine.com (zero.arkaine.com [206.217.210.40]) by hub.freebsd.org (Postfix) with ESMTP id 2F65637B423; Wed, 6 Sep 2000 15:35:01 -0700 (PDT) Received: from exsrv.arkaine.com (exsrv.arkaine.com [192.168.10.10]) by zero.arkaine.com (8.9.3/8.9.3) with ESMTP id SAA40321; Wed, 6 Sep 2000 18:22:10 -0500 (EST) (envelope-from andre@arkaine.com) Received: by exsrv.arkaine.com with Internet Mail Service (5.5.2650.21) id ; Wed, 6 Sep 2000 18:46:12 -0400 Message-ID: From: Andre Chang To: "'freebsd-smp@freebsd.org'" , "'freebsd-scsi@freebsd.org'" Subject: SMP and AMI RAID controller Date: Wed, 6 Sep 2000 18:46:11 -0400 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2650.21) Content-Type: text/plain; charset="iso-8859-1" Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hello, I'm having a problem installing a SMP kernel with a certain RAID controller. The controller drivers are fairly new and I'm hoping that people from SMP or SCSI lists may shed s ome light :) First I'll list the system configuration and info: hardware: Intel PR440FX MB (onboard SCSI, LAN) 2 Intel PPro 200 CPU's (same _ numbers) 1 128MB EDO DIMM 1 Intel Pro 10/100 NIC 1 matrox SVGA card 1 AMI MEGARAID 466 RAID w 16MB EDO chip 2 2GB HDD (RAID 1 -mirroring- config) 1 floppy 1 2GB HDD for testing with onboard controller (not in final system) software: FreeBSD ZERO.akwc.net 4.1-RELEASE FreeBSD 4.1-RELEASE #0: Thu Jul 27 04:44:16 GMT 2000 root@usw4.freebsd.org:/usr/src/sys/compile/GENERIC i386 I reverted to testing the release where the fixed amr drivers were available. TESTING A: I've tested doing a FTP install onto the mirrored HDD's of the AMI controller, 4.1-RELEASE and booting the machine. This works fine. When I build a kernel that is the same as GENERIC except for commenting out the following two lines: options SMP # Symmetric MultiProcessor Kernel options APIC_IO # Symmetric (APIC) I/O the machine gets to this point and hangs forever, sometimes locking the machine: Waiting 15 seconds for SCSI devices to settle no devsw (majdev=0 bootdev=0xa0200000) Mounting root from ufs:/dev/amrd0s1a TESTING B: I've done the exact same as "TESTING A:" but used the spare 2GB HDD on the onboard controller as the install and boot drive (while leaving the AMI controller in the system) and the system boots. Thoughts: I cant figure out if its the kernel or the drivers for the AMI RAID controller. I'm leaning towards the amr driver but how does SMP affect that? Any help would be greatly appreciated. Thanks :) -- Andre. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 6 16:50:34 2000 Delivered-To: freebsd-smp@freebsd.org Received: from thelab.hub.org (CDR27-115.accesscable.net [24.138.27.115]) by hub.freebsd.org (Postfix) with ESMTP id 54B7E37B422; Wed, 6 Sep 2000 16:50:31 -0700 (PDT) Received: from localhost (scrappy@localhost) by thelab.hub.org (8.11.0/8.11.0) with ESMTP id e86No6H40455; Wed, 6 Sep 2000 20:50:07 -0300 (ADT) (envelope-from scrappy@hub.org) X-Authentication-Warning: thelab.hub.org: scrappy owned process doing -bs Date: Wed, 6 Sep 2000 20:50:06 -0300 (ADT) From: The Hermit Hacker To: freebsd-smp@freebsd.org Cc: freebsd-current@freebsd.org Subject: Installed SMP patch and now mouse doesn't work? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Have nothing else to go on yet ... I updated both the current source code and installed the SMP patch from Sept 5th. Everything installed great, rebooted and I can work in character mode ... go to X, and the mouse gives me no activity, where it did just before the upgrade ... Am still investigating at this end, but if anyone else has experience similar and knows how to fix? Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy Systems Administrator @ hub.org primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 6 16:55:55 2000 Delivered-To: freebsd-smp@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id 69C2A37B422; Wed, 6 Sep 2000 16:55:52 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id QAA81102; Wed, 6 Sep 2000 16:55:37 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009062355.QAA81102@pike.osd.bsdi.com> Subject: Re: Installed SMP patch and now mouse doesn't work? In-Reply-To: from The Hermit Hacker at "Sep 6, 2000 08:50:06 pm" To: The Hermit Hacker Date: Wed, 6 Sep 2000 16:55:37 -0700 (PDT) Cc: freebsd-smp@FreeBSD.ORG, freebsd-current@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org The Hermit Hacker wrote: > > Have nothing else to go on yet ... I updated both the current source code > and installed the SMP patch from Sept 5th. Everything installed great, > rebooted and I can work in character mode ... go to X, and the mouse gives > me no activity, where it did just before the upgrade ... > > Am still investigating at this end, but if anyone else has experience > similar and knows how to fix? Uh, all of the SMPng boxes so far haven't had mice. :P Is your mouse a serial mouse? There seem to be some problems with input on the sio devices. -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc "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 Wed Sep 6 17: 6:57 2000 Delivered-To: freebsd-smp@freebsd.org Received: from thelab.hub.org (CDR27-115.accesscable.net [24.138.27.115]) by hub.freebsd.org (Postfix) with ESMTP id 7042E37B424; Wed, 6 Sep 2000 17:06:50 -0700 (PDT) Received: from localhost (scrappy@localhost) by thelab.hub.org (8.11.0/8.11.0) with ESMTP id e8706Qw41783; Wed, 6 Sep 2000 21:06:26 -0300 (ADT) (envelope-from scrappy@hub.org) X-Authentication-Warning: thelab.hub.org: scrappy owned process doing -bs Date: Wed, 6 Sep 2000 21:06:26 -0300 (ADT) From: The Hermit Hacker To: freebsd-current@freebsd.org Cc: freebsd-smp@freebsd.org Subject: 'interrupt-level buffer overflows' for sio device? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org this corresponds to the device I have the mouse on ... the machine has only been up 18hrs ... not sure where else to look ... I'm referencing /dev/ttyd1, like I've always done for my mouse, and just checked /dev (ran MAKEDEV ttyd1) to make sure nothing has changed there ... Help? sio0 at port 0x3f8-0x3ff irq 4 flags 0x10 on isa0 sio0: type 16550A sio1 at port 0x2f8-0x2ff irq 3 on isa0 sio1: type 16550A sio1: 32 more interrupt-level buffer overflows (total 32) sio1: 123 more interrupt-level buffer overflows (total 155) sio1: 135 more interrupt-level buffer overflows (total 290) sio1: 134 more interrupt-level buffer overflows (total 424) sio1: 30 more interrupt-level buffer overflows (total 454) sio1: 12 more interrupt-level buffer overflows (total 466) sio1: 40 more interrupt-level buffer overflows (total 506) sio1: 133 more interrupt-level buffer overflows (total 639) sio1: 135 more interrupt-level buffer overflows (total 774) sio1: 134 more interrupt-level buffer overflows (total 908) sio1: 135 more interrupt-level buffer overflows (total 1043) sio1: 68 more interrupt-level buffer overflows (total 1111) sio1: 118 more interrupt-level buffer overflows (total 1229) sio1: 14 more interrupt-level buffer overflows (total 1243) sio1: 20 more interrupt-level buffer overflows (total 1263) sio1: 119 more interrupt-level buffer overflows (total 1382) sio1: 134 more interrupt-level buffer overflows (total 1516) sio1: 135 more interrupt-level buffer overflows (total 1651) sio1: 57 more interrupt-level buffer overflows (total 1708) sio1: 3 more interrupt-level buffer overflows (total 1711) sio1: 31 more interrupt-level buffer overflows (total 1742) Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy Systems Administrator @ hub.org primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 6 17:15:14 2000 Delivered-To: freebsd-smp@freebsd.org Received: from thelab.hub.org (CDR27-115.accesscable.net [24.138.27.115]) by hub.freebsd.org (Postfix) with ESMTP id 81FEF37B422; Wed, 6 Sep 2000 17:15:05 -0700 (PDT) Received: from localhost (scrappy@localhost) by thelab.hub.org (8.11.0/8.11.0) with ESMTP id e870Ecd41859; Wed, 6 Sep 2000 21:14:38 -0300 (ADT) (envelope-from scrappy@hub.org) X-Authentication-Warning: thelab.hub.org: scrappy owned process doing -bs Date: Wed, 6 Sep 2000 21:14:38 -0300 (ADT) From: The Hermit Hacker To: John Baldwin Cc: freebsd-smp@FreeBSD.ORG, freebsd-current@FreeBSD.ORG Subject: Re: Installed SMP patch and now mouse doesn't work? In-Reply-To: <200009062355.QAA81102@pike.osd.bsdi.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, 6 Sep 2000, John Baldwin wrote: > The Hermit Hacker wrote: > > > > Have nothing else to go on yet ... I updated both the current source code > > and installed the SMP patch from Sept 5th. Everything installed great, > > rebooted and I can work in character mode ... go to X, and the mouse gives > > me no activity, where it did just before the upgrade ... > > > > Am still investigating at this end, but if anyone else has experience > > similar and knows how to fix? > > Uh, all of the SMPng boxes so far haven't had mice. :P Is your mouse a > serial mouse? There seem to be some problems with input on the sio > devices. That would be her ... :) Its my home machine, which is why I run -current on it ... am willing to try anything/everything towards debugging and repairing it ... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 6 18:35:20 2000 Delivered-To: freebsd-smp@freebsd.org Received: from magnesium.net (toxic.magnesium.net [207.154.84.15]) by hub.freebsd.org (Postfix) with SMTP id C434B37B43E for ; Wed, 6 Sep 2000 18:35:17 -0700 (PDT) Received: (qmail 27704 invoked by uid 1142); 7 Sep 2000 01:35:17 -0000 Date: 6 Sep 2000 18:35:17 -0700 Date: Wed, 6 Sep 2000 18:07:43 -0700 From: Jason Evans To: smp@freebsd.org, current@freebsd.org Subject: SMP mega-commit complete Message-ID: <20000906180743.O437@blitz.canonware.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org A static tag, named PRE_SMPNG, was created at 18:00 PDT. The SMP patches have now been committed. There are a number of known issues, and we are actively tracking them on the SMP web page at: http://people.freebsd.org/~jasone/smp/ If you run into issues that appear related to the SMP changes, and they aren't listed as known issues, please bring them up on the -smp or -current mailing list. Jason To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 6 20:41:28 2000 Delivered-To: freebsd-smp@freebsd.org Received: from web5104.mail.yahoo.com (web5104.mail.yahoo.com [216.115.106.74]) by hub.freebsd.org (Postfix) with SMTP id 886EA37B422 for ; Wed, 6 Sep 2000 20:41:27 -0700 (PDT) Message-ID: <20000907034126.18120.qmail@web5104.mail.yahoo.com> Received: from [203.11.192.243] by web5104.mail.yahoo.com; Thu, 07 Sep 2000 13:41:26 EST Date: Thu, 7 Sep 2000 13:41:26 +1000 (EST) From: =?iso-8859-1?q?Paul=20Jansen?= Subject: New AMD processor SMP mainboards To: smp@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I'm quite interested in these new SMP boards coming out for the Duron/Thunderbird (Athlon). I just recompiled my Kernel last night to enable SMP on my Dual Celeron system. I notice that SMP support in 4.1 is only for Intel APIC systems. What's the status of Multiprocessing in Alpha systems? Since the new AMD SMP boards will be using the alpha EV6 bus is it possible to backport some of the Alpha versions code to allow AMD SMP? Does anyone know about this? Is there any work going on for AMD SMP support currently? Thanks. _____________________________________________________________________________ http://geocities.yahoo.com.au - Yahoo! Australia & NZ GeoCities - Build your own Web Site - for free! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 6 21:27:18 2000 Delivered-To: freebsd-smp@freebsd.org Received: from thelab.hub.org (CDR27-115.accesscable.net [24.138.27.115]) by hub.freebsd.org (Postfix) with ESMTP id D5F0437B423; Wed, 6 Sep 2000 21:27:15 -0700 (PDT) Received: from localhost (scrappy@localhost) by thelab.hub.org (8.11.0/8.11.0) with ESMTP id e874Qpq96386; Thu, 7 Sep 2000 01:26:52 -0300 (ADT) (envelope-from scrappy@hub.org) X-Authentication-Warning: thelab.hub.org: scrappy owned process doing -bs Date: Thu, 7 Sep 2000 01:26:50 -0300 (ADT) From: The Hermit Hacker To: Mike Smith Cc: freebsd-smp@freebsd.org Subject: Re: NF7100R - Quad-Xeon machine ... In-Reply-To: <200009011931.MAA05477@mass.osd.bsdi.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 1 Sep 2000, Mike Smith wrote: > > > > Anyone have any experience with this? The server that we are picking this > > up to upgrade is currently running FreeBSD, and it was pretty much my > > first FreeBSD "slipped through the cracks" ... I'd prefer *not* to have to > > go to Solaris/x86 to support it, so ... > > > > The machine is spec'd out as: > > > > NF7100R/Xeon 700/1024KB/256MB/10 HS HDD/6HS PCI > > + Nefinity 700Mhz/1MB Upgrade II With Pentium III Xeon Processor > > + ServeRAID-4M Ultra 160 SCSI Controller > > > > I know there is a major cleanup/rewrite of the SMP code going on, and very > > much look forward to seeing that when it comes out ... I just want to be > > reasonably comfortable that FreeBSD 4.1-STABLE is going to be STABLE on > > that hardware ... > > You'll need a different RAID controller, since I haven't had the > opportunity to produce support for the ServeRAID, and you might want damn damn damn :( anything that I can do once this box gets in to help move that forward? I work at a "mostly Solaris" shop, and over the next year, the acting director of our dept happens to be someone that, over the past 3 years, I've been able to educate in the benefits of moving towards FreeBSD ... this past week, Sun put a nail in its coffin by letting her know that they don't have much commitment to Solaris/x86, so I have ~6 machines that are going to be converted over to FreeBSD over the next year ... two of them are Netfinity servers, plus this new one coming in. I don't mind dedicating the resources, and time, required to be able to support this, I just don't have the experience :( > to have a known-good ethernet card to hand because I had some funky > experiences with the old NF7000's and their onboard ethernet, but the > box itself should be fine. S'alright, in this, I won't touch anything but Intel cards ... about the only one that I've had very consistent good luck with across various OSs :) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 6 22: 8: 0 2000 Delivered-To: freebsd-smp@freebsd.org Received: from fledge.watson.org (fledge.watson.org [204.156.12.50]) by hub.freebsd.org (Postfix) with ESMTP id 7BDBD37B423 for ; Wed, 6 Sep 2000 22:07:57 -0700 (PDT) Received: from fledge.watson.org (robert@fledge.pr.watson.org [192.0.2.3]) by fledge.watson.org (8.9.3/8.9.3) with SMTP id BAA25787; Thu, 7 Sep 2000 01:07:51 -0400 (EDT) (envelope-from robert@fledge.watson.org) Date: Thu, 7 Sep 2000 01:07:51 -0400 (EDT) From: Robert Watson X-Sender: robert@fledge.watson.org To: joel boutros Cc: smp@freebsd.org Subject: Re: Sept 5th patch ... In-Reply-To: <20000906180520.8B54A37B422@hub.freebsd.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, 6 Sep 2000, joel boutros wrote: > > So, are there going to be some documented guidelines for us neophyte > > kernel-hackers? I have just gotten used to the splfoo()/splx() idea, > > and now its gone. How, now, do we write canonical code that doesn't > > futz with itself when interrupts/threads/processes happen? > > the general trick is locking critical sections of code or data. > previously when you did an splXXX(), you were assured that you > weren't going to get interrupted by anything of specified priority > or lower, so it was safe to assume your data probably wasn't going > to change out from under you. now, you mark the critical sections > as in-use, and if something else does come in, it checks to see if > the section is "available" and, if not, figures out some strategy > to deal with it. Deadlocks seem to be a common problem in the system as it stands (especially the file system) -- usually the answer is a strict locking order. Will there be general kernel policies on locking order, or useful primitives for automatically ordering sets of locks (presumably sorted by address) and acquiring all of them, backing out in-progress operations when the current set of locks is found to be insufficient, etc? Also, we currently have some primitives for rudimentary atomic arithmetic operations. Also, any plans on priorities for kernel threads, so that we can introduce opportunities for priority inversion? :-) Robert N M Watson robert@fledge.watson.org http://www.watson.org/~robert/ PGP key fingerprint: AF B5 5F FF A6 4A 79 37 ED 5F 55 E9 58 04 6A B1 TIS Labs at Network Associates, Safeport Network Services To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Sep 6 23:31:51 2000 Delivered-To: freebsd-smp@freebsd.org Received: from io.yi.org (24.67.218.186.bc.wave.home.com [24.67.218.186]) by hub.freebsd.org (Postfix) with ESMTP id 6788737B422; Wed, 6 Sep 2000 23:31:48 -0700 (PDT) Received: from io.yi.org (localhost.gvcl1.bc.wave.home.com [127.0.0.1]) by io.yi.org (Postfix) with ESMTP id C2598BA6D; Wed, 6 Sep 2000 23:33:02 -0700 (PDT) X-Mailer: exmh version 2.1.1 10/15/1999 To: Robert Watson Cc: joel boutros , smp@FreeBSD.ORG, jake@io.yi.org Subject: Re: Sept 5th patch ... In-Reply-To: Message from Robert Watson of "Thu, 07 Sep 2000 01:07:51 EDT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 06 Sep 2000 23:33:02 -0700 From: Jake Burkholder Message-Id: <20000907063302.C2598BA6D@io.yi.org> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > On Wed, 6 Sep 2000, joel boutros wrote: > > > > So, are there going to be some documented guidelines for us neophyte > > > kernel-hackers? I have just gotten used to the splfoo()/splx() idea, > > > and now its gone. How, now, do we write canonical code that doesn't > > > futz with itself when interrupts/threads/processes happen? > > > > the general trick is locking critical sections of code or data. > > previously when you did an splXXX(), you were assured that you > > weren't going to get interrupted by anything of specified priority > > or lower, so it was safe to assume your data probably wasn't going > > to change out from under you. now, you mark the critical sections > > as in-use, and if something else does come in, it checks to see if > > the section is "available" and, if not, figures out some strategy > > to deal with it. > > Deadlocks seem to be a common problem in the system as it stands > (especially the file system) -- usually the answer is a strict locking > order. Will there be general kernel policies on locking order, or useful > primitives for automatically ordering sets of locks (presumably sorted by > address) and acquiring all of them, backing out in-progress operations > when the current set of locks is found to be insufficient, etc? Also, we The general policy is outlined in the slides prepared by Chuck Paterson for the SMP meeting. Roughly: If lock "A" is held while lock "B" is acquired, then "B" can not be held while "A" is acquired. Further, if lock "C" is ever acquired while holding lock "B" then "A" can not be acquired after "C". (these slides are available from Jason's page, worth checking out) The witness code is meant to impose this lock ordering, now in kern_mutex.c :), but it may be a while before we have more than just the 2 main locks to think about, Giant and sched_lock. As far as primitives operating on sets of locks, I can definitely see a potential need, but its still remains to be seen how things will progress. Anyone interested is encouraged to get involved. > currently have some primitives for rudimentary atomic arithmetic > operations. > > Also, any plans on priorities for kernel threads, so that we can introduce > opportunities for priority inversion? :-) Well, interrupt threads are basically just high priority kernel threads. There is code to deal with priority inversion, propogate_priority() in sys/i386/i386/synch_machdep.c, but it is believed to still have bugs that need to be worked out and is currently unused. Jake > > Robert N M Watson > > robert@fledge.watson.org http://www.watson.org/~robert/ > PGP key fingerprint: AF B5 5F FF A6 4A 79 37 ED 5F 55 E9 58 04 6A B1 > TIS Labs at Network Associates, Safeport Network Services > > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-smp" in the body of the message To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 7 4: 6:37 2000 Delivered-To: freebsd-smp@freebsd.org Received: from tkc.att.ne.jp (tkc.att.ne.jp [165.76.16.7]) by hub.freebsd.org (Postfix) with ESMTP id 4748C37B424; Thu, 7 Sep 2000 04:06:33 -0700 (PDT) Received: from work.mzaki.nom (191.pool9.ipctokyo.att.ne.jp [165.76.205.191]) by tkc.att.ne.jp (8.8.8+Spin/3.6W-CONS(10/24/99)) id UAA22859; Thu, 7 Sep 2000 20:06:21 +0900 (JST) Date: Thu, 07 Sep 2000 20:06:20 +0900 Message-ID: <86ya14ihqr.wl@tkc.att.ne.jp> From: Motomichi Matsuzaki To: jasone@canonware.com Cc: smp@freebsd.org, current@freebsd.org Subject: Re: SMP mega-commit complete In-Reply-To: In your message of "6 Sep 2000 18:35:17 -0700" <20000906180743.O437@blitz.canonware.com> References: <20000906180743.O437@blitz.canonware.com> X-Mailer: Wanderlust/2.2.12 (Joyride) XEmacs/21.1 (Bryce Canyon) MIME-Version: 1.0 (generated by WEMI 1.13.7 - "Shimada") Content-Type: text/plain; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org At 6 Sep 2000 18:35:17 -0700, Jason Evans wrote: > If you run into issues that appear related to the SMP changes, and they > aren't listed as known issues, please bring them up on the -smp or -current > mailing list. this breaks building GENERIC kernel. cc -c -O -pipe -Wall -Wredundant-decls -Wnested-externs -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Winline -Wcast-qual -fformat-extensions -ansi -nostdinc -I- -I. -I/opt/usr/src/sys -I/opt/usr/src/sys/../include -D_KERNEL -include opt_global.h -elf -mpreferred-stack-boundary=2 /opt/usr/src/sys/i386/i386/genassym.c In file included from /opt/usr/src/sys/sys/signalvar.h:42, from /opt/usr/src/sys/sys/user.h:59, from /opt/usr/src/sys/i386/i386/genassym.c:66: machine/smp.h:19: #error SMP not supported with I386_CPU *** Error code 1 -- Motomichi Matsuzaki Dept. of Biological Sciences, Grad. School of Science, Univ. of Tokyo, Japan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 7 6:52:14 2000 Delivered-To: freebsd-smp@freebsd.org Received: from mail-sp.viplink.com.br (baco.viplink.com.br [200.211.188.87]) by hub.freebsd.org (Postfix) with ESMTP id 108E737B422 for ; Thu, 7 Sep 2000 06:52:10 -0700 (PDT) Received: from ROMAMK (200.211.188.76 [200.211.188.76]) by mail-sp.viplink.com.br with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2448.0) id RZ8QAAMT; Thu, 7 Sep 2000 10:39:19 -0300 Message-ID: <00b301c018d3$7a9009c0$0a06030a@visionmis.com.br> Reply-To: "Samuka" From: "Samuka" To: Subject: Old ALR-QSMP-Help Date: Thu, 7 Sep 2000 10:56:49 -0300 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2919.6600 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2919.6600 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, I have a ALR(Gateway today) QSMP w/ 4 processor pentium p54 166 Mhz, bus EISA/ISA, and I have try install FBsd r3.3 3.4 and 4.1 but it's failing every time when i enable the smp. The spec smp not is v1.4, maybe 1.1, i'm not sure.Somebody have some experience about this machine or could help me? Tks Samuka To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 7 9: 2:29 2000 Delivered-To: freebsd-smp@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id ACAF737B423 for ; Thu, 7 Sep 2000 09:02:27 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id JAA03075; Thu, 7 Sep 2000 09:02:16 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009071602.JAA03075@pike.osd.bsdi.com> Subject: Re: New AMD processor SMP mainboards In-Reply-To: <20000907034126.18120.qmail@web5104.mail.yahoo.com> from "[Paul Jansen]" at "Sep 7, 2000 01:41:26 pm" To: "[Paul Jansen]" Date: Thu, 7 Sep 2000 09:02:16 -0700 (PDT) Cc: smp@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org [Paul Jansen] wrote: [Charset iso-8859-1 unsupported, filtering to ASCII...] > I'm quite interested in these new SMP boards coming > out for the Duron/Thunderbird (Athlon). I just > recompiled my Kernel last night to enable SMP on my > Dual Celeron system. I notice that SMP support in 4.1 > is only for Intel APIC systems. > What's the status of Multiprocessing in Alpha systems? > Since the new AMD SMP boards will be using the alpha > EV6 bus is it possible to backport some of the Alpha > versions code to allow AMD SMP? > Does anyone know about this? > Is there any work going on for AMD SMP support > currently? We will work to support them as soon as we can get hardware. -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc "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 Thu Sep 7 12:31:17 2000 Delivered-To: freebsd-smp@freebsd.org Received: from volatile.chemicals.tacorp.com (ci391991-a.grnvle1.sc.home.com [24.9.31.75]) by hub.freebsd.org (Postfix) with ESMTP id CE2B637B424; Thu, 7 Sep 2000 12:31:12 -0700 (PDT) Received: (from morganw@localhost) by volatile.chemicals.tacorp.com (8.11.0/8.11.0) id e87JV7f31509; Thu, 7 Sep 2000 15:31:07 -0400 (EDT)?g (envelope-from morganw)œ Date: Thu, 7 Sep 2000 15:31:06 -0400 (EDT) From: Wesley Morgan To: Jason Evans Cc: smp@FreeBSD.ORG, current@FreeBSD.ORG Subject: Re: SMP mega-commit complete In-Reply-To: <20000906180743.O437@blitz.canonware.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, 6 Sep 2000, Jason Evans wrote: > A static tag, named PRE_SMPNG, was created at 18:00 PDT. The SMP patches > have now been committed. Is it safe to follow src-sys from PRE_SMPNG and everything else as -current? -- _ __ ___ ____ ___ ___ ___ Wesley N Morgan _ __ ___ | _ ) __| \ wesleymorgan@home.com _ __ | _ \._ \ |) | FreeBSD: The Power To Serve _ |___/___/___/ 6bone: 3ffe:1ce3:7::b4ff:fe53:c297 Hi! I'm a .signature virus! Copy me into your ~/.signature to help me spread! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 7 13:42:11 2000 Delivered-To: freebsd-smp@freebsd.org Received: from thelab.hub.org (CDR27-115.accesscable.net [24.138.27.115]) by hub.freebsd.org (Postfix) with ESMTP id 7A8BE37B422 for ; Thu, 7 Sep 2000 13:42:09 -0700 (PDT) Received: from localhost (scrappy@localhost) by thelab.hub.org (8.11.0/8.11.0) with ESMTP id e87KfUj90772; Thu, 7 Sep 2000 17:41:33 -0300 (ADT) (envelope-from scrappy@hub.org) X-Authentication-Warning: thelab.hub.org: scrappy owned process doing -bs Date: Thu, 7 Sep 2000 17:41:30 -0300 (ADT) From: The Hermit Hacker To: Samuka Cc: freebsd-smp@FreeBSD.ORG Subject: Re: Old ALR-QSMP-Help In-Reply-To: <00b301c018d3$7a9009c0$0a06030a@visionmis.com.br> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Thu, 7 Sep 2000, Samuka wrote: > Hi, > I have a ALR(Gateway today) QSMP w/ 4 processor pentium p54 166 Mhz, bus > EISA/ISA, and I have try install FBsd r3.3 3.4 and 4.1 but it's failing > every time when i enable the smp. The spec smp not is v1.4, maybe 1.1, i'm > not sure.Somebody have some experience about this machine or could help me? I've got a bunch of dual's laying about at work that I'm running FreeBSD/UP on right now ... would also be most interested in this ... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 7 14:29:21 2000 Delivered-To: freebsd-smp@freebsd.org Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (Postfix) with ESMTP id 20BC537B422; Thu, 7 Sep 2000 14:29:14 -0700 (PDT) Received: from harmony.village.org (harmony.village.org [10.0.0.6]) by rover.village.org (8.9.3/8.9.3) with ESMTP id PAA39022; Thu, 7 Sep 2000 15:29:12 -0600 (MDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (localhost.village.org [127.0.0.1]) by harmony.village.org (8.9.3/8.8.3) with ESMTP id PAA06098; Thu, 7 Sep 2000 15:29:03 -0600 (MDT) Message-Id: <200009072129.PAA06098@harmony.village.org> To: The Hermit Hacker Subject: Re: 'interrupt-level buffer overflows' for sio device? Cc: freebsd-current@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG In-reply-to: Your message of "Wed, 06 Sep 2000 21:06:26 -0300." References: Date: Thu, 07 Sep 2000 15:29:02 -0600 From: Warner Losh Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In message The Hermit Hacker writes: : sio1: 32 more interrupt-level buffer overflows (total 32) From the man page: sio%d: interrupt-level buffer overflow. Problem in the bottom half of the driver. This likely means that the bottom half of sio isn't running fast enough to harvest the characters in the interrupt level buffers. Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 7 16: 0:29 2000 Delivered-To: freebsd-smp@freebsd.org Received: from berserker.bsdi.com (berserker.twistedbit.com [199.79.183.1]) by hub.freebsd.org (Postfix) with ESMTP id A35F237B423; Thu, 7 Sep 2000 16:00:26 -0700 (PDT) Received: from berserker.bsdi.com (cp@LOCALHOST [127.0.0.1]) by berserker.bsdi.com (8.9.3/8.9.3) with ESMTP id RAA24702; Thu, 7 Sep 2000 17:00:15 -0600 (MDT) Message-Id: <200009072300.RAA24702@berserker.bsdi.com> To: Warner Losh Cc: The Hermit Hacker , freebsd-current@freebsd.org, freebsd-smp@freebsd.org Subject: Re: 'interrupt-level buffer overflows' for sio device? From: Chuck Paterson Date: Thu, 07 Sep 2000 17:00:15 -0600 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org FYI, this is very likely not caused by the heavy weight interrupt threads, but rather because the interrupt threads can't be run because the giant lock is held by a process running in the kernel. Once we get drivers to have their own locking and pulled out from under the giant lock this problem should deminish greatly. Before we can do this there are various infrastructure pieces which must be made mp safe, such as the lockmanger. Chuck To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 7 16:40:37 2000 Delivered-To: freebsd-smp@freebsd.org Received: from pike.osd.bsdi.com (pike.osd.bsdi.com [204.216.28.222]) by hub.freebsd.org (Postfix) with ESMTP id A515437B423; Thu, 7 Sep 2000 16:40:35 -0700 (PDT) Received: (from jhb@localhost) by pike.osd.bsdi.com (8.9.3/8.9.3) id QAA24513; Thu, 7 Sep 2000 16:40:18 -0700 (PDT) (envelope-from jhb) From: John Baldwin Message-Id: <200009072340.QAA24513@pike.osd.bsdi.com> Subject: Re: Sept 5th patch ... In-Reply-To: <20000907063302.C2598BA6D@io.yi.org> from Jake Burkholder at "Sep 6, 2000 11:33:02 pm" To: Jake Burkholder Date: Thu, 7 Sep 2000 16:40:18 -0700 (PDT) Cc: Robert Watson , joel boutros , smp@FreeBSD.ORG, jake@io.yi.org X-Mailer: ELM [version 2.4ME+ PL68 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Jake Burkholder wrote: > Well, interrupt threads are basically just high priority kernel threads. > There is code to deal with priority inversion, propogate_priority() in > sys/i386/i386/synch_machdep.c, but it is believed to still have bugs > that need to be worked out and is currently unused. > > Jake Which is to say: when we have tried turning it on the system locks up with interrupts disabled when entering multiuser mode. :P -- John Baldwin -- http://www.FreeBSD.org/~jhb/ PGP Key: http://www.cslab.vt.edu/~jobaldwi/pgpkey.asc "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 Thu Sep 7 17:23:32 2000 Delivered-To: freebsd-smp@freebsd.org Received: from wantadilla.lemis.com (wantadilla.lemis.com [192.109.197.80]) by hub.freebsd.org (Postfix) with ESMTP id 602E937B422; Thu, 7 Sep 2000 17:23:25 -0700 (PDT) Received: (from grog@localhost) by wantadilla.lemis.com (8.11.0/8.9.3) id e880MsN95987; Fri, 8 Sep 2000 09:52:54 +0930 (CST) (envelope-from grog) Date: Fri, 8 Sep 2000 09:52:54 +0930 From: Greg Lehey To: Chuck Paterson Cc: Warner Losh , The Hermit Hacker , freebsd-current@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: 'interrupt-level buffer overflows' for sio device? Message-ID: <20000908095254.E83632@wantadilla.lemis.com> References: <200009072300.RAA24702@berserker.bsdi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0i In-Reply-To: <200009072300.RAA24702@berserker.bsdi.com>; from cp@bsdi.com on Thu, Sep 07, 2000 at 05:00:15PM -0600 Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia Phone: +61-8-8388-8286 Fax: +61-8-8388-8725 Mobile: +61-418-838-708 WWW-Home-Page: http://www.lemis.com/~grog X-PGP-Fingerprint: 6B 7B C3 8C 61 CD 54 AF 13 24 52 F8 6D A4 95 EF Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Thursday, 7 September 2000 at 17:00:15 -0600, Chuck Paterson wrote: > FYI, this is very likely not caused by the heavy weight > interrupt threads, but rather because the interrupt threads can't > be run because the giant lock is held by a process running in the > kernel. Once we get drivers to have their own locking and pulled out > from under the giant lock this problem should deminish greatly. Before > we can do this there are various infrastructure pieces which must > be made mp safe, such as the lockmanger. We're running sio as a fast interrupt, so it's definitely not because of a heavyweight thread :-) I think fast interrupts also completely bypass mutexes, though something might have changed since I last looked. Greg -- Finger grog@lemis.com for PGP public key See complete headers for address and phone numbers To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 7 17:50:59 2000 Delivered-To: freebsd-smp@freebsd.org Received: from gidora.zeta.org.au (gidora.zeta.org.au [203.26.10.25]) by hub.freebsd.org (Postfix) with SMTP id 7132737B424 for ; Thu, 7 Sep 2000 17:50:54 -0700 (PDT) Received: (qmail 2174 invoked from network); 8 Sep 2000 00:50:50 -0000 Received: from unknown (HELO bde.zeta.org.au) (203.2.228.102) by gidora.zeta.org.au with SMTP; 8 Sep 2000 00:50:50 -0000 Date: Fri, 8 Sep 2000 11:50:45 +1100 (EST) From: Bruce Evans X-Sender: bde@besplex.bde.org To: Warner Losh Cc: The Hermit Hacker , freebsd-current@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: 'interrupt-level buffer overflows' for sio device? In-Reply-To: <200009072129.PAA06098@harmony.village.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Thu, 7 Sep 2000, Warner Losh wrote: > In message The Hermit Hacker writes: > : sio1: 32 more interrupt-level buffer overflows (total 32) > > >From the man page: > sio%d: interrupt-level buffer overflow. Problem in the bottom half of > the driver. > > This likely means that the bottom half of sio isn't running fast > enough to harvest the characters in the interrupt level buffers. This means that fast interrupt handlers aren't actually fast. Especially if it occurs for slow devices like mouses. Even slow interrupt handlers had about 10 times lower latency than necessary for 1200 bps serial mouses. Bruce To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 7 20: 2:27 2000 Delivered-To: freebsd-smp@freebsd.org Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (Postfix) with ESMTP id 9217837B424; Thu, 7 Sep 2000 20:02:24 -0700 (PDT) Received: from harmony.village.org (harmony.village.org [10.0.0.6]) by rover.village.org (8.9.3/8.9.3) with ESMTP id VAA40597; Thu, 7 Sep 2000 21:02:22 -0600 (MDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (localhost.village.org [127.0.0.1]) by harmony.village.org (8.9.3/8.8.3) with ESMTP id VAA50459; Thu, 7 Sep 2000 21:02:11 -0600 (MDT) Message-Id: <200009080302.VAA50459@harmony.village.org> To: Bruce Evans Subject: Re: 'interrupt-level buffer overflows' for sio device? Cc: The Hermit Hacker , freebsd-current@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG In-reply-to: Your message of "Fri, 08 Sep 2000 11:50:45 +1100." References: Date: Thu, 07 Sep 2000 21:02:10 -0600 From: Warner Losh Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Also, this is interrupt level overflows. We can run the fast interrupts fast enough to harvest characters from the hardware. What's not happening is that sio's bottom half isn't being run fast enough so the interrupt level buffers overflow. Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 7 20:10: 7 2000 Delivered-To: freebsd-smp@freebsd.org Received: from thelab.hub.org (CDR27-115.accesscable.net [24.138.27.115]) by hub.freebsd.org (Postfix) with ESMTP id BB3B537B422; Thu, 7 Sep 2000 20:10:03 -0700 (PDT) Received: from localhost (scrappy@localhost) by thelab.hub.org (8.11.0/8.11.0) with ESMTP id e8839AD04714; Fri, 8 Sep 2000 00:09:10 -0300 (ADT) (envelope-from scrappy@hub.org) X-Authentication-Warning: thelab.hub.org: scrappy owned process doing -bs Date: Fri, 8 Sep 2000 00:09:10 -0300 (ADT) From: The Hermit Hacker To: Warner Losh Cc: Bruce Evans , freebsd-current@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: 'interrupt-level buffer overflows' for sio device? In-Reply-To: <200009080302.VAA50459@harmony.village.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Thu, 7 Sep 2000, Warner Losh wrote: > Also, this is interrupt level overflows. We can run the fast > interrupts fast enough to harvest characters from the hardware. > What's not happening is that sio's bottom half isn't being run fast > enough so the interrupt level buffers overflow. Okay, I'm a little confused here ... from what I'm reading/following, this isn't a new problem ... or is it? If not, why has it suddenly manifested itself with the new SMP code? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 7 20:15:14 2000 Delivered-To: freebsd-smp@freebsd.org Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (Postfix) with ESMTP id 689CF37B424; Thu, 7 Sep 2000 20:15:10 -0700 (PDT) Received: from harmony.village.org (harmony.village.org [10.0.0.6]) by rover.village.org (8.9.3/8.9.3) with ESMTP id VAA40667; Thu, 7 Sep 2000 21:15:08 -0600 (MDT) (envelope-from imp@harmony.village.org) Received: from harmony.village.org (localhost.village.org [127.0.0.1]) by harmony.village.org (8.9.3/8.8.3) with ESMTP id VAA50701; Thu, 7 Sep 2000 21:14:58 -0600 (MDT) Message-Id: <200009080314.VAA50701@harmony.village.org> To: The Hermit Hacker Subject: Re: 'interrupt-level buffer overflows' for sio device? Cc: Bruce Evans , freebsd-current@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG In-reply-to: Your message of "Fri, 08 Sep 2000 00:09:10 -0300." References: Date: Thu, 07 Sep 2000 21:14:58 -0600 From: Warner Losh Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In message The Hermit Hacker writes: : Okay, I'm a little confused here ... from what I'm reading/following, this : isn't a new problem ... or is it? If not, why has it suddenly manifested : itself with the new SMP code? It seems to be new with the SMP code. At least that's what my reading of the original message is. Warner To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 7 20:29:33 2000 Delivered-To: freebsd-smp@freebsd.org Received: from thelab.hub.org (CDR27-115.accesscable.net [24.138.27.115]) by hub.freebsd.org (Postfix) with ESMTP id DEE8437B424; Thu, 7 Sep 2000 20:29:28 -0700 (PDT) Received: from localhost (scrappy@localhost) by thelab.hub.org (8.11.0/8.11.0) with ESMTP id e883SrK05070; Fri, 8 Sep 2000 00:28:53 -0300 (ADT) (envelope-from scrappy@hub.org) X-Authentication-Warning: thelab.hub.org: scrappy owned process doing -bs Date: Fri, 8 Sep 2000 00:28:53 -0300 (ADT) From: The Hermit Hacker To: Warner Losh Cc: Bruce Evans , freebsd-current@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: 'interrupt-level buffer overflows' for sio device? In-Reply-To: <200009080314.VAA50701@harmony.village.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Thu, 7 Sep 2000, Warner Losh wrote: > In message The Hermit Hacker writes: > : Okay, I'm a little confused here ... from what I'm reading/following, this > : isn't a new problem ... or is it? If not, why has it suddenly manifested > : itself with the new SMP code? > > It seems to be new with the SMP code. At least that's what my reading > of the original message is. That was what I thought when I wrote the original message ... it just scares me sometimes how quickly things get acknowledged as a problem in the Open Source community, especially after dealign with tech support at Sun and getting the complete run around ... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 8 0: 5:46 2000 Delivered-To: freebsd-smp@freebsd.org Received: from mailman.zeta.org.au (mailman.zeta.org.au [203.26.10.16]) by hub.freebsd.org (Postfix) with ESMTP id 18E9D37B422; Fri, 8 Sep 2000 00:05:41 -0700 (PDT) Received: from bde.zeta.org.au (bde.zeta.org.au [203.2.228.102]) by mailman.zeta.org.au (8.8.7/8.8.7) with ESMTP id SAA25388; Fri, 8 Sep 2000 18:05:22 +1100 Date: Fri, 8 Sep 2000 18:05:18 +1100 (EST) From: Bruce Evans X-Sender: bde@besplex.bde.org To: Warner Losh Cc: The Hermit Hacker , freebsd-current@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: 'interrupt-level buffer overflows' for sio device? In-Reply-To: <200009080302.VAA50459@harmony.village.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Thu, 7 Sep 2000, Warner Losh wrote: > Also, this is interrupt level overflows. We can run the fast > interrupts fast enough to harvest characters from the hardware. > What's not happening is that sio's bottom half isn't being run fast > enough so the interrupt level buffers overflow. Not surprising, since sio's bottom half scheduler (schedsofttty()) is now null. The top half apparently only gets scheduled if there is an output interrupt, a fake output interrupt (flags 0x8) a modem status change, or urgent data. Urgent data probably keeps it going for slip and pppd, but there is no urgent data in the termios line discipline. Bruce To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 8 8: 2:28 2000 Delivered-To: freebsd-smp@freebsd.org Received: from charles.dircon.net (charles.dircon.net [195.157.2.129]) by hub.freebsd.org (Postfix) with ESMTP id EC28937B42C for ; Fri, 8 Sep 2000 08:02:23 -0700 (PDT) Received: from diablo.dircon.net (desk108.ch.dircon.net [195.157.3.108]) by charles.dircon.net with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2650.21) id RWA3Q39C; Fri, 8 Sep 2000 16:02:56 +0100 Received: by diablo.dircon.net (Postfix, from userid 1000) id 80F011B222; Fri, 8 Sep 2000 16:02:22 +0100 (BST) Date: Fri, 8 Sep 2000 16:02:22 +0100 From: Mark Blackman To: "Potter, Jeff" Cc: "'freebsd-smp@freebsd.org'" , sam.yeung@dircon.net Subject: Re: FreeBSD 4.0 Release SMP kernel stalls during bootup"SMP: CPU1 api c_initialize()" (See PR19338 for details) Message-ID: <20000908160222.A37468@diablo.dircon.net> References: <2D348570FE47D0118C3F00805FEA318104BC29FA@EXCHOU-CA0901> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2i In-Reply-To: <2D348570FE47D0118C3F00805FEA318104BC29FA@EXCHOU-CA0901>; from Jeff.Potter@Compaq.com on Fri, Aug 04, 2000 at 07:39:29PM -0500 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org We're having a bit of trouble with a Compaq DL380 here (with SMP). hangs just after lint0: 0x00000700 lint1: 0x00010400 TPR: 0x00000010 SVR: 0x000001ff Could I persuade you to send a kernel config file that works with DL380 SMP and/or any BIOS settings required, please. This is for a big customer so speed will be gratefully appreciated. - Mark (Blackman) On Fri, Aug 04, 2000 at 07:39:29PM -0500, Potter, Jeff wrote: > > http://www.freebsd.org/cgi/query-pr.cgi?pr=19338 > > Does anyone have any advice or a patch that possibly could fix this issue? > The SMP kernel works fine on the Compaq DL380, but fails on the Compaq > DL360. The most obvious difference in the hardware of the two units is the > BSP/AP APIC ID numbers are wired differently [per Intel spec]. MP Table for > the Compaq DL360 is listed in PR 19338. We don't suspect ill behaved > hardware because RedHat 6.2, and other SMP capable OSes work fine on both > units. Copied 'most' of the boot message below from the customer issue > report. If more information is needed for resolution, please let me know. > Thanks in advance! > > Regards, > JP > > > Output from boot -hv DL360 > -------------------------- > connected > Copyright (c) 1992-2000 The FreeBSD Project. > Copyright (c) 1982, 1986, 1989, 1991, 1993 > The Regents of the University of California. All rights reserved. > FreeBSD 4.0-20000531-STABLE #0: Thu Jun 1 16:03:10 BST 2000 > root@fubar.noc.demon.net:/usr/src/sys/compile/CPQ > Calibrating clock(s) ... TSC clock: 797430926 Hz, i8254 clock: 1193118 Hz > CLK_USE_I8254_CALIBRATION not specified - using default frequency > Timecounter "i8254" frequency 1193182 Hz > CLK_USE_TSC_CALIBRATION not specified - using old calibration method > CPU: Pentium III/Pentium III Xeon (797.48-MHz 686-class CPU) > Origin = "GenuineIntel" Id = 0x683 Stepping = 3 > > Features=0x383fbff CMOV,PAT,PSE36,MMX,FXSR,XMM> > real memory = 1207943168 (1179632K bytes) > Physical memory chunk(s): > 0x00001000 - 0x0009efff, 647168 bytes (158 pages) > 0x002c3000 - 0x47ff3fff, 1205014528 bytes (294193 pages) > avail memory = 1170939904 (1143496K bytes) > Programming 35 pins in IOAPIC #0 > IOAPIC #0 intpin 2 -> irq 0 > IOAPIC #0 intpin 24 -> irq 2 > SMP: CPU0 apic_initialize(): > lint0: 0x00000700 lint1: 0x00010400 TPR: 0x00000010 SVR: 0x000001ff > FreeBSD/SMP: Multiprocessor motherboard > cpu0 (BSP): apic id: 3, version: 0x00040011, at 0xfee00000 > cpu1 (AP): apic id: 0, version: 0x00040011, at 0xfee00000 > io0 (APIC): apic id: 8, version: 0x00220011, at 0xfec00000 > bios32: Found BIOS32 Service Directory header at 0xc00ffee0 > bios32: Entry = 0xf0000 (c00f0000) Rev = 0 Len = 1 > pcibios: PCI BIOS entry at 0x94 > Other BIOS signatures found: > ACPI: 000f4f90 > Preloaded elf kernel "kernel" at 0xc02a7000. > Pentium Pro MTRR support enabled > SMP: CPU0 bsp_apic_configure(): > lint0: 0x00010700 lint1: 0x00000400 TPR: 0x00000010 SVR: 0x000001ff > ... > ... "snip" > ... > Device configuration finished. > APIC_IO: routing 8254 via IOAPIC #0 intpin 2 > bpf: lo0 attached > SMP: AP CPU #1 Launched! > SMP: CPU1 apic_initialize(): > lint0: 0x00010700 lint1: 0x00010400 TPR: 0x00000010 SVR: 0x000001ff > > --------- > > A three fingered salute then adds the following > > boot() called on cpu#1 > Uptime: 0s > Rebooting... > cpu_reset called on cpu#1 > cpu_reset: Stopping other CPUs > cpu_reset: Restarting BSP > pu_reset_proxy: Grabbed mp locck uf_orr sBeSP: > BSP did not grab mp lock > > --------------------------------- > > > > > > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-smp" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 8 10:25:12 2000 Delivered-To: freebsd-smp@freebsd.org Received: from fledge.watson.org (fledge.watson.org [204.156.12.50]) by hub.freebsd.org (Postfix) with ESMTP id E14B937B440 for ; Fri, 8 Sep 2000 10:25:09 -0700 (PDT) Received: from fledge.watson.org (robert@fledge.pr.watson.org [192.0.2.3]) by fledge.watson.org (8.9.3/8.9.3) with SMTP id NAA49343 for ; Fri, 8 Sep 2000 13:25:07 -0400 (EDT) (envelope-from robert@fledge.watson.org) Date: Fri, 8 Sep 2000 13:25:06 -0400 (EDT) From: Robert Watson X-Sender: robert@fledge.watson.org To: freebsd-smp@FreeBSD.org Subject: SMPng: ``Giving up on 1 buffer'' Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I recently cvsup'd to a recent -CURRENT, past the SMPng commit. Everything seems hunky dory, except for the fact that during my last shutdown, I was surprised to see that the disk sync'ing failed, with one buffer not committed. On bootup, the file system was not clean, requiring a fsck which showed no problems. I assume that this is related somehow. Robert N M Watson robert@fledge.watson.org http://www.watson.org/~robert/ PGP key fingerprint: AF B5 5F FF A6 4A 79 37 ED 5F 55 E9 58 04 6A B1 TIS Labs at Network Associates, Safeport Network Services To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 8 10:43:31 2000 Delivered-To: freebsd-smp@freebsd.org Received: from piranha.amis.net (piranha.amis.net [212.18.32.3]) by hub.freebsd.org (Postfix) with ESMTP id 7663B37B440; Fri, 8 Sep 2000 10:43:29 -0700 (PDT) Received: from titanic.medinet.si (titanic.medinet.si [212.18.32.66]) by piranha.amis.net (Postfix) with ESMTP id BF7005D27; Fri, 8 Sep 2000 19:43:26 +0200 (CEST) Date: Fri, 8 Sep 2000 19:43:26 +0200 (CEST) From: Blaz Zupan X-Sender: blaz@titanic.medinet.si To: Robert Watson Cc: freebsd-smp@FreeBSD.org Subject: Re: SMPng: ``Giving up on 1 buffer'' In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I recently cvsup'd to a recent -CURRENT, past the SMPng commit. > Everything seems hunky dory, except for the fact that during my last > shutdown, I was surprised to see that the disk sync'ing failed, with one > buffer not committed. On bootup, the file system was not clean, requiring > a fsck which showed no problems. I assume that this is related somehow. I saw this as well. I have a UP machine, and on every single reboot it fails on syncing buffers and the filesystem is dirty after the reboot. Blaz Zupan, Medinet d.o.o, Linhartova 21, 2000 Maribor, Slovenia E-mail: blaz@amis.net, Tel: +386-2-320-6320, Fax: +386-2-320-6325 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 8 11:59:13 2000 Delivered-To: freebsd-smp@freebsd.org Received: from fw.wintelcom.net (ns1.wintelcom.net [209.1.153.20]) by hub.freebsd.org (Postfix) with ESMTP id 4D21137B446 for ; Fri, 8 Sep 2000 11:59:10 -0700 (PDT) Received: (from bright@localhost) by fw.wintelcom.net (8.10.0/8.10.0) id e88IxAL10017 for smp@freebsd.org; Fri, 8 Sep 2000 11:59:10 -0700 (PDT) Date: Fri, 8 Sep 2000 11:59:09 -0700 From: Alfred Perlstein To: smp@freebsd.org Subject: SMPng: catching signals and mutexes. Message-ID: <20000908115909.F12231@fw.wintelcom.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.4i Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org There seems to be no way to specify that a signal is to interrupt a wait on a mutex, I'm wondering if we should make it possible to do so. code in question: sb_lock() uipc_socket2.c line 313 of 1012: while (sb->sb_flags & SB_LOCK) { sb->sb_flags |= SB_WANT; error = tsleep((caddr_t)&sb->sb_flags, (sb->sb_flags & SB_NOINTR) ? PSOCK : PSOCK|PCATCH, "sblock", 0); if (error) return (error); } sb->sb_flags |= SB_LOCK; return (0); has to made into: int sb_lock(sb) register struct sockbuf *sb; { int error; while (mtx_try_enter(sb->sb_mtx, MTX_DEF) == 0) { error = tsleep((caddr_t)&sb->sb_flags, (sb->sb_flags & SB_NOINTR) ? PSOCK : PSOCK|PCATCH, "sblock", 0); if (error) return (error); } return (0); } I'm not sure I like this at all, shouldn't there be an mtx_enter with some option to return an error if a signal arrives? Thanks, -- -Alfred Perlstein - [bright@wintelcom.net|alfred@freebsd.org] "I have the heart of a child; I keep it in a jar on my desk." To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 8 12:48:19 2000 Delivered-To: freebsd-smp@freebsd.org Received: from anchor-post-31.mail.demon.net (anchor-post-31.mail.demon.net [194.217.242.89]) by hub.freebsd.org (Postfix) with ESMTP id 3560337B440; Fri, 8 Sep 2000 12:48:16 -0700 (PDT) Received: from nlsys.demon.co.uk ([158.152.125.33] helo=herring.nlsystems.com) by anchor-post-31.mail.demon.net with esmtp (Exim 2.12 #1) id 13XU8D-000EeV-0V; Fri, 8 Sep 2000 20:48:13 +0100 Received: from salmon.nlsystems.com (salmon.nlsystems.com [10.0.0.3]) by herring.nlsystems.com (8.9.3/8.8.8) with ESMTP id UAA12959; Fri, 8 Sep 2000 20:49:42 +0100 (BST) (envelope-from dfr@nlsystems.com) Date: Fri, 8 Sep 2000 20:49:07 +0100 (BST) From: Doug Rabson To: John Baldwin Cc: The Hermit Hacker , freebsd-smp@freebsd.org, freebsd-current@freebsd.org Subject: Re: Installed SMP patch and now mouse doesn't work? In-Reply-To: <200009062355.QAA81102@pike.osd.bsdi.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, 6 Sep 2000, John Baldwin wrote: > The Hermit Hacker wrote: > > > > Have nothing else to go on yet ... I updated both the current source code > > and installed the SMP patch from Sept 5th. Everything installed great, > > rebooted and I can work in character mode ... go to X, and the mouse gives > > me no activity, where it did just before the upgrade ... > > > > Am still investigating at this end, but if anyone else has experience > > similar and knows how to fix? > > Uh, all of the SMPng boxes so far haven't had mice. :P Is your mouse a > serial mouse? There seem to be some problems with input on the sio > devices. This is because the schedsofttty() call is commented out so the high-level interrupt routine in sio is never entered. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 20 8348 3944 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 8 12:48:49 2000 Delivered-To: freebsd-smp@freebsd.org Received: from pcnet1.pcnet.com (pcnet1.pcnet.com [204.213.232.3]) by hub.freebsd.org (Postfix) with ESMTP id 76D8537B446 for ; Fri, 8 Sep 2000 12:48:46 -0700 (PDT) Received: (from eischen@localhost) by pcnet1.pcnet.com (8.8.7/PCNet) id PAA10762; Fri, 8 Sep 2000 15:48:18 -0400 (EDT) Date: Fri, 8 Sep 2000 15:48:18 -0400 (EDT) From: Daniel Eischen To: Alfred Perlstein Cc: smp@FreeBSD.ORG Subject: Re: SMPng: catching signals and mutexes. In-Reply-To: <20000908115909.F12231@fw.wintelcom.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 8 Sep 2000, Alfred Perlstein wrote: > There seems to be no way to specify that a signal is to interrupt > a wait on a mutex, I'm wondering if we should make it possible to > do so. [...] > I'm not sure I like this at all, shouldn't there be an > mtx_enter with some option to return an error if a signal > arrives? IMHO, yes. Solaris has cv_wait_sig() for just that. -- Dan Eischen To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 8 13:19: 4 2000 Delivered-To: freebsd-smp@freebsd.org Received: from magnesium.net (toxic.magnesium.net [207.154.84.15]) by hub.freebsd.org (Postfix) with SMTP id 5C43B37B505 for ; Fri, 8 Sep 2000 13:19:01 -0700 (PDT) Received: (qmail 44052 invoked by uid 1142); 8 Sep 2000 20:19:00 -0000 Date: 8 Sep 2000 13:19:00 -0700 Date: Fri, 8 Sep 2000 13:18:52 -0700 From: Jason Evans To: Robert Watson Cc: freebsd-smp@FreeBSD.org Subject: Re: SMPng: ``Giving up on 1 buffer'' Message-ID: <20000908131852.B19447@blitz.canonware.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0.1i In-Reply-To: ; from rwatson@FreeBSD.org on Fri, Sep 08, 2000 at 01:25:06PM -0400 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, Sep 08, 2000 at 01:25:06PM -0400, Robert Watson wrote: > I recently cvsup'd to a recent -CURRENT, past the SMPng commit. > Everything seems hunky dory, except for the fact that during my last > shutdown, I was surprised to see that the disk sync'ing failed, with one > buffer not committed. On bootup, the file system was not clean, requiring > a fsck which showed no problems. I assume that this is related somehow. Jake Burkholder also experiences this behavior, but it has never happened to me. So, it seems to be a hit and miss thing. You get hit. Jason To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 8 13:27:49 2000 Delivered-To: freebsd-smp@freebsd.org Received: from mass.osd.bsdi.com (mass.osd.bsdi.com [204.216.28.234]) by hub.freebsd.org (Postfix) with ESMTP id 80FB537B449 for ; Fri, 8 Sep 2000 13:27:48 -0700 (PDT) Received: from mass.osd.bsdi.com (localhost [127.0.0.1]) by mass.osd.bsdi.com (8.9.3/8.9.3) with ESMTP id NAA03987; Fri, 8 Sep 2000 13:39:00 -0700 (PDT) (envelope-from msmith@mass.osd.bsdi.com) Message-Id: <200009082039.NAA03987@mass.osd.bsdi.com> X-Mailer: exmh version 2.1.1 10/15/1999 To: The Hermit Hacker Cc: freebsd-smp@freebsd.org Subject: Re: NF7100R - Quad-Xeon machine ... In-reply-to: Your message of "Thu, 07 Sep 2000 01:26:50 -0300." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 08 Sep 2000 13:38:59 -0700 From: Mike Smith Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > You'll need a different RAID controller, since I haven't had the > > opportunity to produce support for the ServeRAID, and you might want > > damn damn damn :( anything that I can do once this box gets in to help > move that forward? Send it to me. Seriously; I need one of these controllers (and preferably an entire box with disks) to do the work. One of the older P6-based NF7000's would be fine, or even one of the NF5000's with the embedded ServeRAID would do it. -- ... every activity meets with opposition, everyone who acts has his rivals and unfortunately opponents also. But not because people want to be opponents, rather because the tasks and relationships force people to take different points of view. [Dr. Fritz Todt] To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 8 15:48:33 2000 Delivered-To: freebsd-smp@freebsd.org Received: from thelab.hub.org (CDR27-115.accesscable.net [24.138.27.115]) by hub.freebsd.org (Postfix) with ESMTP id 3E92937B50F; Fri, 8 Sep 2000 15:48:30 -0700 (PDT) Received: from localhost (scrappy@localhost) by thelab.hub.org (8.11.0/8.11.0) with ESMTP id e88Mm0601620; Fri, 8 Sep 2000 19:48:00 -0300 (ADT) (envelope-from scrappy@hub.org) X-Authentication-Warning: thelab.hub.org: scrappy owned process doing -bs Date: Fri, 8 Sep 2000 19:48:00 -0300 (ADT) From: The Hermit Hacker To: freebsd-current@freebsd.org Cc: freebsd-smp@freebsd.org Subject: mouse works!! Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org serial mouse works with currrent sources ... thanks guys Marc G. Fournier ICQ#7615664 IRC Nick: Scrappy Systems Administrator @ hub.org primary: scrappy@hub.org secondary: scrappy@{freebsd|postgresql}.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 8 16:32:34 2000 Delivered-To: freebsd-smp@freebsd.org Received: from io.yi.org (24.67.218.186.bc.wave.home.com [24.67.218.186]) by hub.freebsd.org (Postfix) with ESMTP id B3F5537B443 for ; Fri, 8 Sep 2000 16:32:31 -0700 (PDT) Received: from io.yi.org (localhost.gvcl1.bc.wave.home.com [127.0.0.1]) by io.yi.org (Postfix) with ESMTP id D27EBBA6D; Fri, 8 Sep 2000 16:33:01 -0700 (PDT) X-Mailer: exmh version 2.1.1 10/15/1999 To: Daniel Eischen Cc: Alfred Perlstein , smp@FreeBSD.ORG Subject: Re: SMPng: catching signals and mutexes. In-Reply-To: Message from Daniel Eischen of "Fri, 08 Sep 2000 15:48:18 EDT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 08 Sep 2000 16:33:01 -0700 From: Jake Burkholder Message-Id: <20000908233301.D27EBBA6D@io.yi.org> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > On Fri, 8 Sep 2000, Alfred Perlstein wrote: > > There seems to be no way to specify that a signal is to interrupt > > a wait on a mutex, I'm wondering if we should make it possible to > > do so. > > [...] > > > I'm not sure I like this at all, shouldn't there be an > > mtx_enter with some option to return an error if a signal > > arrives? > > IMHO, yes. Solaris has cv_wait_sig() for just that. > Mutexes are meant to serve a different purpose than tsleep or condition variables. The next thing in line is to add a mutex argument to tsleep, which will be atomicly released and re-acquired upon resume. The code in question might then look like: mtx_enter(&sb->sb_mtx, MTX_DEF); while (sb->sb_flags & SB_LOCK) { sb->sb_flags |= SB_WANT; error = tsleep((caddr_t)&sb->sb_flags, sb->sb_mtx, (sb->sb_flags & SB_NOINTR) ? PSOCK : PSOCK|PCATCH, "sblock", 0); if (error) { mtx_exit(sb->sb_mtx, MTX_DEF); return (error); } } sb->sb_flags |= SB_LOCK; mtx_exit(sb->sb_mtx, MTX_DEF); Jake To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 8 22:50:29 2000 Delivered-To: freebsd-smp@freebsd.org Received: from hawk.prod.itd.earthlink.net (hawk.prod.itd.earthlink.net [207.217.120.22]) by hub.freebsd.org (Postfix) with ESMTP id 1CB3137B424; Fri, 8 Sep 2000 22:50:26 -0700 (PDT) Received: from earthlink.net (dialup-63.209.93.125.LosAngeles1.Level3.net [63.209.93.125]) by hawk.prod.itd.earthlink.net (8.9.3-EL_1_3/8.9.3) with ESMTP id WAA09839; Fri, 8 Sep 2000 22:46:16 -0700 (PDT) Message-ID: <39B9CEAC.E2F5AF0A@earthlink.net> Date: Fri, 08 Sep 2000 22:46:20 -0700 From: Wes Warner X-Mailer: Mozilla 4.73 [en] (X11; U; Linux 2.4.0-test8 i686) X-Accept-Language: en MIME-Version: 1.0 To: freebsd-scsi@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: adaptec 2940uw with smp kernel on a bp6 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, I have a abit bp6 and an adaptec 2940 uw scsi card. If I compile a SMP kernel with cupport for my scsi card I get this error... Waiting 15 seconds for SCSI devices to settle Timeout SCB handled by another timeout (probe1:ahc0:0:1:0)SCB 0xd timedout ahc0 Issued channel A bus reset 14 SCBs Aborted I can run SMP fine without the card or run with the card and no SMP, but I can't do them both together. Is there a fix for this? Thanks, Wes To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 8 22:54:23 2000 Delivered-To: freebsd-smp@freebsd.org Received: from pcnet1.pcnet.com (pcnet1.pcnet.com [204.213.232.3]) by hub.freebsd.org (Postfix) with ESMTP id 74DAD37B42C for ; Fri, 8 Sep 2000 22:54:20 -0700 (PDT) Received: (from eischen@localhost) by pcnet1.pcnet.com (8.8.7/PCNet) id BAA25707; Sat, 9 Sep 2000 01:51:20 -0400 (EDT) Date: Sat, 9 Sep 2000 01:51:20 -0400 (EDT) From: Daniel Eischen To: Jake Burkholder Cc: Alfred Perlstein , smp@FreeBSD.ORG Subject: Re: SMPng: catching signals and mutexes. In-Reply-To: <20000908233301.D27EBBA6D@io.yi.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 8 Sep 2000, Jake Burkholder wrote: > > On Fri, 8 Sep 2000, Alfred Perlstein wrote: > > > There seems to be no way to specify that a signal is to interrupt > > > a wait on a mutex, I'm wondering if we should make it possible to > > > do so. > > > > [...] > > > > > I'm not sure I like this at all, shouldn't there be an > > > mtx_enter with some option to return an error if a signal > > > arrives? > > > > IMHO, yes. Solaris has cv_wait_sig() for just that. > > > > Mutexes are meant to serve a different purpose than tsleep or > condition variables. > > The next thing in line is to add a mutex argument to tsleep, > which will be atomicly released and re-acquired upon resume. So how far are we from removing tsleep in favor of cv_wait()/ cv_wait_sig() or similar? Let's get rid of tsleep if at all possible. -- Dan Eischen To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Sep 9 7:27:25 2000 Delivered-To: freebsd-smp@freebsd.org Received: from fire.starkreality.com (fire.starkreality.com [198.78.158.210]) by hub.freebsd.org (Postfix) with ESMTP id F357437B422; Sat, 9 Sep 2000 07:27:16 -0700 (PDT) Received: from palehorse.starkreality.com (palehorse.starkreality.com [198.78.158.217]) by fire.starkreality.com (8.9.3/8.9.2) with ESMTP id JAA13254; Sat, 9 Sep 2000 09:27:06 -0500 (CDT) Message-Id: <4.3.2.7.2.20000909091918.039a0f70@mail.starkreality.com> X-Sender: caesar@mail.starkreality.com X-Mailer: QUALCOMM Windows Eudora Version 4.3.2 Date: Sat, 09 Sep 2000 09:27:05 -0500 To: Wes Warner , freebsd-scsi@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG From: "William S. Duncanson" Subject: Re: adaptec 2940uw with smp kernel on a bp6 In-Reply-To: <39B9CEAC.E2F5AF0A@earthlink.net> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hm, I have a Tyan Tiger 100 with dual processors running in SMP mode with a 2940UW. I would suggest making sure that the BIOS for the motherboard and the card are up to date. Does any other MP capable OS work? What are the outputs of dmesg and mptable? At 00:46 9/9/2000, Wes Warner wrote: >Hi, I have a abit bp6 and an adaptec 2940 uw scsi card. If I compile a >SMP >kernel with cupport for my scsi card I get this error... > >Waiting 15 seconds for SCSI devices to settle >Timeout SCB handled by another timeout >(probe1:ahc0:0:1:0)SCB 0xd timedout >ahc0 Issued channel A bus reset 14 SCBs Aborted > >I can run SMP fine without the card or run with the card and no SMP, but >I can't do them both together. Is there a fix for this? > > Thanks, > Wes > > >To Unsubscribe: send mail to majordomo@FreeBSD.org >with "unsubscribe freebsd-smp" in the body of the message -- William S. Duncanson caesar@starkreality.com Smash forehead on keyboard to continue... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Sep 9 8: 2:49 2000 Delivered-To: freebsd-smp@freebsd.org Received: from finch-post-12.mail.demon.net (finch-post-12.mail.demon.net [194.217.242.41]) by hub.freebsd.org (Postfix) with ESMTP id 084BA37B424 for ; Sat, 9 Sep 2000 08:02:46 -0700 (PDT) Received: from nlsys.demon.co.uk ([158.152.125.33] helo=herring.nlsystems.com) by finch-post-12.mail.demon.net with esmtp (Exim 2.12 #1) id 13Xm9U-000IxB-0C; Sat, 9 Sep 2000 15:02:45 +0000 Received: from salmon.nlsystems.com (salmon.nlsystems.com [10.0.0.3]) by herring.nlsystems.com (8.9.3/8.8.8) with ESMTP id QAA18224; Sat, 9 Sep 2000 16:04:16 +0100 (BST) (envelope-from dfr@nlsystems.com) Date: Sat, 9 Sep 2000 16:03:28 +0100 (BST) From: Doug Rabson To: Daniel Eischen Cc: Jake Burkholder , Alfred Perlstein , smp@freebsd.org Subject: Re: SMPng: catching signals and mutexes. In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, 9 Sep 2000, Daniel Eischen wrote: > On Fri, 8 Sep 2000, Jake Burkholder wrote: > > > On Fri, 8 Sep 2000, Alfred Perlstein wrote: > > > > There seems to be no way to specify that a signal is to interrupt > > > > a wait on a mutex, I'm wondering if we should make it possible to > > > > do so. > > > > > > [...] > > > > > > > I'm not sure I like this at all, shouldn't there be an > > > > mtx_enter with some option to return an error if a signal > > > > arrives? > > > > > > IMHO, yes. Solaris has cv_wait_sig() for just that. > > > > > > > Mutexes are meant to serve a different purpose than tsleep or > > condition variables. > > > > The next thing in line is to add a mutex argument to tsleep, > > which will be atomicly released and re-acquired upon resume. > > So how far are we from removing tsleep in favor of cv_wait()/ > cv_wait_sig() or similar? Let's get rid of tsleep if at all > possible. Removing tsleep would be a large job and is reasonably separate from adding mutexes to replace spls for data protection. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 20 8348 3944 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Sep 9 9:51:21 2000 Delivered-To: freebsd-smp@freebsd.org Received: from gull.prod.itd.earthlink.net (gull.prod.itd.earthlink.net [207.217.121.85]) by hub.freebsd.org (Postfix) with ESMTP id 62BDF37B422; Sat, 9 Sep 2000 09:51:18 -0700 (PDT) Received: from earthlink.net (dialup-209.245.67.138.LosAngeles1.Level3.net [209.245.67.138]) by gull.prod.itd.earthlink.net (8.9.3-EL_1_3/8.9.3) with ESMTP id JAA24551; Sat, 9 Sep 2000 09:51:16 -0700 (PDT) Message-ID: <39BA6A83.8B800FD9@earthlink.net> Date: Sat, 09 Sep 2000 09:51:15 -0700 From: Wes Warner X-Mailer: Mozilla 4.73 [en] (X11; U; Linux 2.4.0-test8 i686) X-Accept-Language: en MIME-Version: 1.0 To: freebsd-scsi@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: adaptec 2940uw with smp kernel on a bp6 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I"m running 4.1. Wes Tom Samplonius wrote: > > On Fri, 8 Sep 2000, Wes Warner wrote: > > > Hi, I have a abit bp6 and an adaptec 2940 uw scsi card. If I compile a > > SMP > > kernel with cupport for my scsi card I get this error... > > > > Waiting 15 seconds for SCSI devices to settle > > Timeout SCB handled by another timeout > > (probe1:ahc0:0:1:0)SCB 0xd timedout > > ahc0 Issued channel A bus reset 14 SCBs Aborted > > > > I can run SMP fine without the card or run with the card and no SMP, but > > I can't do them both together. Is there a fix for this? > > Perhaps. What version of FreeBSD are you using? > > Tom To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Sep 9 9:55:27 2000 Delivered-To: freebsd-smp@freebsd.org Received: from gull.prod.itd.earthlink.net (gull.prod.itd.earthlink.net [207.217.121.85]) by hub.freebsd.org (Postfix) with ESMTP id 3AFAE37B42C; Sat, 9 Sep 2000 09:55:22 -0700 (PDT) Received: from earthlink.net (dialup-209.245.67.138.LosAngeles1.Level3.net [209.245.67.138]) by gull.prod.itd.earthlink.net (8.9.3-EL_1_3/8.9.3) with ESMTP id JAA03817; Sat, 9 Sep 2000 09:55:20 -0700 (PDT) Message-ID: <39BA6B77.CA1D2C5F@earthlink.net> Date: Sat, 09 Sep 2000 09:55:19 -0700 From: Wes Warner X-Mailer: Mozilla 4.73 [en] (X11; U; Linux 2.4.0-test8 i686) X-Accept-Language: en MIME-Version: 1.0 To: freebsd-scsi@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: adaptec 2940uw with smp kernel on a bp6 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I updated both the scsi card bios and my mobo bios to see if that helped already. I didn't get any improvement. I currently am running slackware and it runs fine both smp and the scsi card. It works with either 2.2.x or 2.4.test.x. When I have scsi support enabled in the kernel the machine only gives me the error message, it will not totally boot. I don't ever get a terminal, it just hangs. Thanks, Wes "William S. Duncanson" wrote: > > Hm, I have a Tyan Tiger 100 with dual processors running in SMP mode with a > 2940UW. I would suggest making sure that the BIOS for the motherboard and > the card are up to date. Does any other MP capable OS work? What are the > outputs of dmesg and mptable? > > At 00:46 9/9/2000, Wes Warner wrote: > > >Hi, I have a abit bp6 and an adaptec 2940 uw scsi card. If I compile a > >SMP > >kernel with cupport for my scsi card I get this error... > > > >Waiting 15 seconds for SCSI devices to settle > >Timeout SCB handled by another timeout > >(probe1:ahc0:0:1:0)SCB 0xd timedout > >ahc0 Issued channel A bus reset 14 SCBs Aborted > > > >I can run SMP fine without the card or run with the card and no SMP, but > >I can't do them both together. Is there a fix for this? > > > > Thanks, > > Wes To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Sep 9 11:50:37 2000 Delivered-To: freebsd-smp@freebsd.org Received: from misery.sdf.com (misery.sdf.com [204.244.213.49]) by hub.freebsd.org (Postfix) with ESMTP id 103D837B424; Sat, 9 Sep 2000 11:50:33 -0700 (PDT) Received: from tom (helo=localhost) by misery.sdf.com with local-esmtp (Exim 2.12 #1) id 13XpM3-0003gy-00; Sat, 9 Sep 2000 11:27:55 -0700 Date: Sat, 9 Sep 2000 11:27:48 -0700 (PDT) From: Tom Samplonius To: Wes Warner Cc: freebsd-scsi@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: adaptec 2940uw with smp kernel on a bp6 In-Reply-To: <39BA6A83.8B800FD9@earthlink.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, 9 Sep 2000, Wes Warner wrote: > I"m running 4.1. > > Wes 4.1-RELEASE or 4.1-STABLE? I believe an up-to-date -STABLE has some ahc fixes. Tom To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Sep 9 12:10: 8 2000 Delivered-To: freebsd-smp@freebsd.org Received: from critter.freebsd.dk (flutter.freebsd.dk [212.242.40.147]) by hub.freebsd.org (Postfix) with ESMTP id 436D537B422; Sat, 9 Sep 2000 12:10:03 -0700 (PDT) Received: from critter (localhost [127.0.0.1]) by critter.freebsd.dk (8.11.0/8.9.3) with ESMTP id e89JA1N89356; Sat, 9 Sep 2000 21:10:02 +0200 (CEST) (envelope-from phk@critter.freebsd.dk) To: arch@freebsd.org, smp@freebsd.org Subject: Device probing... From: Poul-Henning Kamp Date: Sat, 09 Sep 2000 21:10:01 +0200 Message-ID: <89354.968526601@critter> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I would really like to se us move all the device probe/attach code to run in "normal context", ie enable interrupts before we probe everything. The main argument for this is the drivers can then use the full complement of kernel infrastructure and interrupts during probe attach. Drivers needs to be able to function in this environment anyway if they support removeable devices (pccard, usb, etc). Are there reasons to not do this I have overlooked ? -- Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 phk@FreeBSD.ORG | TCP/IP since RFC 956 FreeBSD coreteam member | BSD since 4.3-tahoe Never attribute to malice what can adequately be explained by incompetence. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Sep 9 12:41:22 2000 Delivered-To: freebsd-smp@freebsd.org Received: from mail.m.iinet.net.au (opera3.iinet.net.au [203.59.24.51]) by hub.freebsd.org (Postfix) with SMTP id 4F22837B424 for ; Sat, 9 Sep 2000 12:41:17 -0700 (PDT) Received: (qmail 10013 invoked by uid 666); 9 Sep 2000 19:41:13 -0000 Received: from reggae-12-250.nv.iinet.net.au (HELO jules.elischer.org) (203.59.92.250) by mail.m.iinet.net.au with SMTP; 9 Sep 2000 19:41:13 -0000 Message-ID: <39BA9252.794BDF32@elischer.org> Date: Sat, 09 Sep 2000 12:41:06 -0700 From: Julian Elischer X-Mailer: Mozilla 3.04Gold (X11; I; FreeBSD 5.0-CURRENT i386) MIME-Version: 1.0 To: Poul-Henning Kamp Cc: arch@freebsd.org, smp@freebsd.org Subject: Re: Device probing... References: <89354.968526601@critter> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Poul-Henning Kamp wrote: > > I would really like to se us move all the device probe/attach code > to run in "normal context", ie enable interrupts before we probe > everything. > > The main argument for this is the drivers can then use the full > complement of kernel infrastructure and interrupts during probe > attach. > > Drivers needs to be able to function in this environment anyway > if they support removeable devices (pccard, usb, etc). > > Are there reasons to not do this I have overlooked ? I know of no reason not to, and in fact the old BSD code used to do this in order to try work out what interrupt a device was on. (though with shared interrupts this gets harder) > > -- > Poul-Henning Kamp | UNIX since Zilog Zeus 3.20 > phk@FreeBSD.ORG | TCP/IP since RFC 956 > FreeBSD coreteam member | BSD since 4.3-tahoe > Never attribute to malice what can adequately be explained by incompetence. > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-smp" in the body of the message -- __--_|\ Julian Elischer / \ julian@elischer.org ( OZ ) World tour 2000 ---> X_.---._/ presently in: Perth v To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Sep 9 17:33:57 2000 Delivered-To: freebsd-smp@freebsd.org Received: from wantadilla.lemis.com (wantadilla.lemis.com [192.109.197.80]) by hub.freebsd.org (Postfix) with ESMTP id C66FB37B422; Sat, 9 Sep 2000 17:33:49 -0700 (PDT) Received: (from grog@localhost) by wantadilla.lemis.com (8.11.0/8.9.3) id e8A0XYG15727; Sun, 10 Sep 2000 10:03:34 +0930 (CST) (envelope-from grog) Date: Sun, 10 Sep 2000 10:03:34 +0930 From: Greg Lehey To: Wes Warner Cc: FreeBSD Questions Subject: Re: adaptec 2940uw with smp kernel on a bp6 Message-ID: <20000910100334.A15703@wantadilla.lemis.com> References: <39B9CEAC.E2F5AF0A@earthlink.net> <39BA6B77.CA1D2C5F@earthlink.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0i In-Reply-To: <39BA6B77.CA1D2C5F@earthlink.net>; from ufodziner@earthlink.net on Sat, Sep 09, 2000 at 09:55:19AM -0700 Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia Phone: +61-8-8388-8286 Fax: +61-8-8388-8725 Mobile: +61-418-838-708 WWW-Home-Page: http://www.lemis.com/~grog X-PGP-Fingerprint: 6B 7B C3 8C 61 CD 54 AF 13 24 52 F8 6D A4 95 EF Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org [Format recovered--see http://www.lemis.com/email/email-format.html] [Following up to -questions] On Saturday, 9 September 2000 at 9:55:19 -0700, Wes Warner wrote: > "William S. Duncanson" wrote: >> At 00:46 9/9/2000, Wes Warner wrote: >> >>> Hi, I have a abit bp6 and an adaptec 2940 uw scsi card. If I >>> compile a SMP kernel with cupport for my scsi card I get this >>> error... >>> >>> Waiting 15 seconds for SCSI devices to settle >>> Timeout SCB handled by another timeout >>> (probe1:ahc0:0:1:0)SCB 0xd timedout >>> ahc0 Issued channel A bus reset 14 SCBs Aborted >>> >>> I can run SMP fine without the card or run with the card and no >>> SMP, but I can't do them both together. Is there a fix for this? >> >> Hm, I have a Tyan Tiger 100 with dual processors running in SMP >> mode with a 2940UW. I would suggest making sure that the BIOS for >> the motherboard and the card are up to date. Does any other MP >> capable OS work? What are the outputs of dmesg and mptable? > > I updated both the scsi card bios and my mobo bios to see if that > helped already. I didn't get any improvement. I currently am > running slackware and it runs fine both smp and the scsi card. It > works with either 2.2.x or 2.4.test.x. When I have scsi support > enabled in the kernel the machine only gives me the error message, > it will not totally boot. I don't ever get a terminal, it just > hangs. Well, the simple answer is that this is not a general bug. Plenty of people, myself included, are running this combination of motherboard and host adapter. What is the rest of your hardware? dmesg output would be a good idea. It's also a good idea to try a GENERIC kernel modified for SMP, rather than a kernel with other modifications, just in case that's the problem. Greg -- When replying to this message, please copy the original recipients. For more information, see http://www.lemis.com/questions.html When replying to this message, please take care not to mutilate the original text. For more information, see http://www.lemis.com/email.html Finger grog@lemis.com for PGP public key See complete headers for address and phone numbers To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Sep 9 17:49: 7 2000 Delivered-To: freebsd-smp@freebsd.org Received: from hawk.prod.itd.earthlink.net (hawk.prod.itd.earthlink.net [207.217.120.22]) by hub.freebsd.org (Postfix) with ESMTP id ADABA37B422; Sat, 9 Sep 2000 17:49:03 -0700 (PDT) Received: from earthlink.net (dialup-63.212.128.135.LosAngeles1.Level3.net [63.212.128.135]) by hawk.prod.itd.earthlink.net (8.9.3-EL_1_3/8.9.3) with ESMTP id RAA08596; Sat, 9 Sep 2000 17:49:01 -0700 (PDT) Message-ID: <39BADA7D.F68366E8@earthlink.net> Date: Sat, 09 Sep 2000 17:49:01 -0700 From: Wes Warner X-Mailer: Mozilla 4.73 [en] (X11; U; Linux 2.4.0-test8 i686) X-Accept-Language: en MIME-Version: 1.0 To: freebsd-scsi@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: adaptec 2940uw with smp kernel on a bp6 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I'm not sure to be honest. I downloaded it when it wasn't an rc anymore. I can tr to re-download it and see if htat works. Thanks, Wes Tom Samplonius wrote: > > On Sat, 9 Sep 2000, Wes Warner wrote: > > > I"m running 4.1. > > > > Wes > > 4.1-RELEASE or 4.1-STABLE? I believe an up-to-date -STABLE has some ahc > fixes. > > Tom To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Sep 9 20: 9:50 2000 Delivered-To: freebsd-smp@freebsd.org Received: from gidora.zeta.org.au (gidora.zeta.org.au [203.26.10.25]) by hub.freebsd.org (Postfix) with SMTP id BB7CF37B42C for ; Sat, 9 Sep 2000 20:09:45 -0700 (PDT) Received: (qmail 17553 invoked from network); 10 Sep 2000 03:09:41 -0000 Received: from unknown (HELO bde.zeta.org.au) (203.2.228.102) by gidora.zeta.org.au with SMTP; 10 Sep 2000 03:09:41 -0000 Date: Sun, 10 Sep 2000 14:09:40 +1100 (EST) From: Bruce Evans X-Sender: bde@besplex.bde.org To: Poul-Henning Kamp Cc: arch@FreeBSD.ORG, smp@FreeBSD.ORG Subject: Re: Device probing... In-Reply-To: <89354.968526601@critter> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, 9 Sep 2000, Poul-Henning Kamp wrote: > I would really like to se us move all the device probe/attach code > to run in "normal context", ie enable interrupts before we probe > everything. > > The main argument for this is the drivers can then use the full > complement of kernel infrastructure and interrupts during probe > attach. > > Drivers needs to be able to function in this environment anyway > if they support removeable devices (pccard, usb, etc). > > Are there reasons to not do this I have overlooked ? Active level-triggered interrupts will repeat endlessly if they have no driver. Active edge-triggered interrupts will prevent further interrupts on the same irq if they have no driver. Active interrupts with no driver hopefully arren't a problem with removable devices (after all drivers have had a chance to run). Bruce To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Sep 9 21: 1:42 2000 Delivered-To: freebsd-smp@freebsd.org Received: from swan.prod.itd.earthlink.net (swan.prod.itd.earthlink.net [207.217.120.123]) by hub.freebsd.org (Postfix) with ESMTP id 1795737B424; Sat, 9 Sep 2000 21:01:39 -0700 (PDT) Received: from earthlink.net (dialup-63.212.142.182.LosAngeles1.Level3.net [63.212.142.182]) by swan.prod.itd.earthlink.net (8.9.3-EL_1_3/8.9.3) with ESMTP id VAA09925; Sat, 9 Sep 2000 21:01:37 -0700 (PDT) Message-ID: <39BB07A2.D04EF9E7@earthlink.net> Date: Sat, 09 Sep 2000 21:01:39 -0700 From: Wes Warner X-Mailer: Mozilla 4.73 [en] (X11; U; Linux 2.4.0-test8 i686) X-Accept-Language: en MIME-Version: 1.0 To: freebsd-scsi@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: adaptec 2940uw with smp kernel on a bp6 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org What I am using is a generic kernel with SMP enabled. The kernel I use with the scsi working is just the regular generic. I'm trying to do this as tactful as possible. The rest of my hardware is... SBLive, Linksys 10/100 nic, ATI All-in-wonder 32meg, Colorado Floppy tape (with card), 3com 56k external modem, 3ea. WD 50 pin SCSI drives, Maxtor 20 gig ata66 drive (OS is on this drive), Quantum 5 gig, creative 52x cdrom, Philips 2x2x6 burner, ls-120. I think that is about it. If it matters, my processors are 433's and I have 256megs of RAM. I have tried it both overclocked and not with no luck either way. I can't get a dmesg output with both smp and scsi because it doesn't boot. Thanks, Wes > Well, the simple answer is that this is not a general bug. Plenty of > people, myself included, are running this combination of motherboard > and host adapter. What is the rest of your hardware? dmesg output > would be a good idea. It's also a good idea to try a GENERIC kernel > modified for SMP, rather than a kernel with other modifications, just > in case that's the problem. > > Greg > -- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Sep 9 23:28:19 2000 Delivered-To: freebsd-smp@freebsd.org Received: from femail2.sdc1.sfba.home.com (femail2.sdc1.sfba.home.com [24.0.95.82]) by hub.freebsd.org (Postfix) with ESMTP id 93D2237B422; Sat, 9 Sep 2000 23:28:16 -0700 (PDT) Received: from cx443070b ([24.0.36.170]) by femail2.sdc1.sfba.home.com (InterMail vM.4.01.03.00 201-229-121) with SMTP id <20000910062809.OYIP27630.femail2.sdc1.sfba.home.com@cx443070b>; Sat, 9 Sep 2000 23:28:09 -0700 Message-ID: <000701c01af0$72e24ac0$aa240018@cx443070b> From: "Jeremiah Gowdy" To: "Wes Warner" , , References: <39BB07A2.D04EF9E7@earthlink.net> Subject: Re: adaptec 2940uw with smp kernel on a bp6 Date: Sat, 9 Sep 2000 23:29:15 -0700 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > What I am using is a generic kernel with SMP enabled. The kernel I use > with the scsi working is just the regular generic. I'm trying to do > this as tactful as possible. The rest of my hardware is... > > SBLive, Linksys 10/100 nic, ATI All-in-wonder 32meg, Colorado Floppy > tape (with card), 3com 56k external modem, 3ea. WD 50 pin SCSI drives, > Maxtor 20 gig ata66 drive (OS is on this drive), Quantum 5 gig, creative > 52x cdrom, Philips 2x2x6 burner, ls-120. > > I think that is about it. If it matters, my processors are 433's and I > have 256megs of RAM. I have tried it both overclocked and not with no > luck either way. I can't get a dmesg output with both smp and scsi > because it doesn't boot. > > Thanks, > Wes I wouldn't think a GENERIC kernel with SMP enabled would be better/cleaner/safer/less likely to crash than a cleaned kernel containing only the drivers you need. Seems like less likely of a chance of a conflict. I suggest, clean a GENERIC, enable SMP and go with that To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message