From owner-freebsd-smp Mon Mar 8 0:51:46 1999 Delivered-To: freebsd-smp@freebsd.org Received: from orbit.flnet.com (orbit.flnet.com [205.240.232.32]) by hub.freebsd.org (Postfix) with ESMTP id 40DCF14C25; Mon, 8 Mar 1999 00:51:43 -0800 (PST) (envelope-from henrich@orbit.flnet.com) Received: (from henrich@localhost) by orbit.flnet.com (8.8.5/8.8.4) id DAA02696; Mon, 8 Mar 1999 03:51:25 -0500 (EST) Message-ID: <19990308005125.10555@orbit.flnet.com> Date: Mon, 8 Mar 1999 00:51:25 -0800 From: Charles Henrich To: freebsd-hackers@freebsd.org, freebsd-smp@freebsd.org Subject: Erorr: isa_dmainit(2, 1024) failed Mail-Followup-To: freebsd-hackers@freebsd.org, freebsd-smp@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89i X-Operating-System: FreeBSD 2.2-BETA_A X-PGP-Fingerprint: 1024/F7 FD C7 3A F5 6A 23 BF 76 C4 B8 C9 6E 41 A4 4F Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org During startup on my FreeBSD box (a dual proc 1GB system) I see: isa_dmainit(2, 1024) failed And (perhaps) coincidentally any access to the ISA sound card causes the machine to panic with an isa page map missing. Does anyone out there have any ideas whats causing this, and what a solution might be? Thanks! -Crh Charles Henrich Manex Visual Effects henrich@flnet.com http://orbit.flnet.com/~henrich To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Mar 8 1:17:35 1999 Delivered-To: freebsd-smp@freebsd.org Received: from dingo.cdrom.com (castles178.castles.com [208.214.165.178]) by hub.freebsd.org (Postfix) with ESMTP id C9B6514E7F; Mon, 8 Mar 1999 01:17:02 -0800 (PST) (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (LOCALHOST [127.0.0.1]) by dingo.cdrom.com (8.9.3/8.8.8) with ESMTP id BAA05661; Mon, 8 Mar 1999 01:11:30 -0800 (PST) (envelope-from mike@dingo.cdrom.com) Message-Id: <199903080911.BAA05661@dingo.cdrom.com> X-Mailer: exmh version 2.0.2 2/24/98 To: Charles Henrich Cc: freebsd-hackers@freebsd.org, freebsd-smp@freebsd.org Subject: Re: Erorr: isa_dmainit(2, 1024) failed In-reply-to: Your message of "Mon, 08 Mar 1999 00:51:25 PST." <19990308005125.10555@orbit.flnet.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 08 Mar 1999 01:11:30 -0800 From: Mike Smith Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > During startup on my FreeBSD box (a dual proc 1GB system) I see: > > isa_dmainit(2, 1024) failed > > And (perhaps) coincidentally any access to the ISA sound card causes the > machine to panic with an isa page map missing. Does anyone out there have any > ideas whats causing this, and what a solution might be? Thanks! You have too much memory. 8) Seriously, by the time the ISA code gets a chance to allocate memory, all the physically dma-able-to memory is gone for other uses. This is basically a bug in the way kernel memory is handed out. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Mar 8 9:41:42 1999 Delivered-To: freebsd-smp@freebsd.org Received: from orbit.flnet.com (orbit.flnet.com [205.240.232.32]) by hub.freebsd.org (Postfix) with ESMTP id 25B7914EEE; Mon, 8 Mar 1999 09:41:33 -0800 (PST) (envelope-from henrich@orbit.flnet.com) Received: (from henrich@localhost) by orbit.flnet.com (8.8.5/8.8.4) id MAA09237; Mon, 8 Mar 1999 12:41:01 -0500 (EST) Message-ID: <19990308094100.14836@orbit.flnet.com> Date: Mon, 8 Mar 1999 09:41:00 -0800 From: Charles Henrich To: Mike Smith Cc: freebsd-hackers@freebsd.org, freebsd-smp@freebsd.org Subject: Re: Erorr: isa_dmainit(2, 1024) failed Mail-Followup-To: Mike Smith , freebsd-hackers@freebsd.org, freebsd-smp@freebsd.org References: <19990308005125.10555@orbit.flnet.com> <199903080911.BAA05661@dingo.cdrom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89i In-Reply-To: <199903080911.BAA05661@dingo.cdrom.com>; from Mike Smith on Mon, Mar 08, 1999 at 01:11:30AM -0800 X-Operating-System: FreeBSD 2.2-BETA_A X-PGP-Fingerprint: 1024/F7 FD C7 3A F5 6A 23 BF 76 C4 B8 C9 6E 41 A4 4F Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On the subject of Re: Erorr: isa_dmainit(2, 1024) failed, Mike Smith stated: > > During startup on my FreeBSD box (a dual proc 1GB system) I see: > > > > isa_dmainit(2, 1024) failed > > > > And (perhaps) coincidentally any access to the ISA sound card causes the > > machine to panic with an isa page map missing. Does anyone out there have > > any ideas whats causing this, and what a solution might be? Thanks! > > You have too much memory. 8) > > Seriously, by the time the ISA code gets a chance to allocate memory, all > the physically dma-able-to memory is gone for other uses. > > This is basically a bug in the way kernel memory is handed out. How does one fix this? :) -Crh Charles Henrich Manex Visual Effects henrich@flnet.com http://orbit.flnet.com/~henrich To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Mar 8 9:46:17 1999 Delivered-To: freebsd-smp@freebsd.org Received: from dingo.cdrom.com (castles302.castles.com [208.214.167.2]) by hub.freebsd.org (Postfix) with ESMTP id CABBD14EF8; Mon, 8 Mar 1999 09:46:10 -0800 (PST) (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (LOCALHOST [127.0.0.1]) by dingo.cdrom.com (8.9.3/8.8.8) with ESMTP id JAA06967; Mon, 8 Mar 1999 09:40:28 -0800 (PST) (envelope-from mike@dingo.cdrom.com) Message-Id: <199903081740.JAA06967@dingo.cdrom.com> X-Mailer: exmh version 2.0.2 2/24/98 To: Charles Henrich Cc: Mike Smith , freebsd-hackers@freebsd.org, freebsd-smp@freebsd.org Subject: Re: Erorr: isa_dmainit(2, 1024) failed In-reply-to: Your message of "Mon, 08 Mar 1999 09:41:00 PST." <19990308094100.14836@orbit.flnet.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 08 Mar 1999 09:40:28 -0800 From: Mike Smith Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > On the subject of Re: Erorr: isa_dmainit(2, 1024) failed, Mike Smith stated: > > > > During startup on my FreeBSD box (a dual proc 1GB system) I see: > > > > > > isa_dmainit(2, 1024) failed > > > > > > And (perhaps) coincidentally any access to the ISA sound card causes the > > > machine to panic with an isa page map missing. Does anyone out there have > > > any ideas whats causing this, and what a solution might be? Thanks! > > > > You have too much memory. 8) > > > > Seriously, by the time the ISA code gets a chance to allocate memory, all > > the physically dma-able-to memory is gone for other uses. > > > > This is basically a bug in the way kernel memory is handed out. > > How does one fix this? :) With a text editor. Find the code that does it wrong, change it and recompile. Send us the diffs when you're done. Actually, fixing it "right" will be very difficult. I suspect that the ISA DMA code will need a statically-allocated buffer to overcome this. -- \\ Sometimes you're ahead, \\ Mike Smith \\ sometimes you're behind. \\ mike@smith.net.au \\ The race is long, and in the \\ msmith@freebsd.org \\ end it's only with yourself. \\ msmith@cdrom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Mar 8 10:11:30 1999 Delivered-To: freebsd-smp@freebsd.org Received: from orbit.flnet.com (orbit.flnet.com [205.240.232.32]) by hub.freebsd.org (Postfix) with ESMTP id 5715D14BE0; Mon, 8 Mar 1999 10:11:23 -0800 (PST) (envelope-from henrich@orbit.flnet.com) Received: (from henrich@localhost) by orbit.flnet.com (8.8.5/8.8.4) id NAA09977; Mon, 8 Mar 1999 13:10:59 -0500 (EST) Message-ID: <19990308101059.50251@orbit.flnet.com> Date: Mon, 8 Mar 1999 10:10:59 -0800 From: Charles Henrich To: Mike Smith Cc: freebsd-hackers@freebsd.org, freebsd-smp@freebsd.org Subject: Re: Erorr: isa_dmainit(2, 1024) failed Mail-Followup-To: Mike Smith , freebsd-hackers@freebsd.org, freebsd-smp@freebsd.org References: <19990308094100.14836@orbit.flnet.com> <199903081740.JAA06967@dingo.cdrom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89i In-Reply-To: <199903081740.JAA06967@dingo.cdrom.com>; from Mike Smith on Mon, Mar 08, 1999 at 09:40:28AM -0800 X-Operating-System: FreeBSD 2.2-BETA_A X-PGP-Fingerprint: 1024/F7 FD C7 3A F5 6A 23 BF 76 C4 B8 C9 6E 41 A4 4F Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On the subject of Re: Erorr: isa_dmainit(2, 1024) failed, Mike Smith stated: > With a text editor. Find the code that does it wrong, change it and > recompile. Send us the diffs when you're done. Smartass :) > Actually, fixing it "right" will be very difficult. I suspect that the ISA > DMA code will need a statically-allocated buffer to overcome this. Ah, that makes sense. I shall look into this. -Crh Charles Henrich Manex Visual Effects henrich@flnet.com http://orbit.flnet.com/~henrich To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Mar 8 10:28:58 1999 Delivered-To: freebsd-smp@freebsd.org Received: from midten.fast.no (midten.fast.no [195.139.251.11]) by hub.freebsd.org (Postfix) with ESMTP id D260814F26; Mon, 8 Mar 1999 10:28:45 -0800 (PST) (envelope-from tegge@fast.no) Received: from fast.no (IDENT:tegge@midten.fast.no [195.139.251.11]) by midten.fast.no (8.9.1/8.9.1) with ESMTP id TAA70476; Mon, 8 Mar 1999 19:27:59 +0100 (CET) Message-Id: <199903081827.TAA70476@midten.fast.no> To: mike@smith.net.au Cc: henrich@flnet.com, freebsd-hackers@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: Erorr: isa_dmainit(2, 1024) failed From: Tor.Egge@fast.no In-Reply-To: Your message of "Mon, 08 Mar 1999 09:40:28 -0800" References: <199903081740.JAA06967@dingo.cdrom.com> X-Mailer: Mew version 1.70 on Emacs 19.34.1 Mime-Version: 1.0 Content-Type: Multipart/Mixed; boundary="--Next_Part(Mon_Mar__8_19:22:22_1999)--" Content-Transfer-Encoding: 7bit Date: Mon, 08 Mar 1999 19:27:59 +0100 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org ----Next_Part(Mon_Mar__8_19:22:22_1999)-- Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit > > How does one fix this? :) > > With a text editor. Find the code that does it wrong, change it and > recompile. Send us the diffs when you're done. > > Actually, fixing it "right" will be very difficult. I suspect that the > ISA DMA code will need a statically-allocated buffer to overcome this. I've been using the appended patch to use SoundBlaster AWE64 Gold on a machine with 512 MB memory. - Tor Egge ----Next_Part(Mon_Mar__8_19:22:22_1999)-- Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Index: sys/i386/include/pmap.h =================================================================== RCS file: /home/ncvs/src/sys/i386/include/pmap.h,v retrieving revision 1.58 diff -u -r1.58 pmap.h --- pmap.h 1999/03/02 16:20:39 1.58 +++ pmap.h 1999/03/03 15:24:59 @@ -265,6 +285,7 @@ void putfmtrr __P((void)); void pmap_setdevram __P((unsigned long long, unsigned)); void pmap_setvidram __P((void)); +int pmap_defer_avail __P((vm_offset_t)); #endif /* KERNEL */ Index: sys/i386/i386/pmap.c =================================================================== RCS file: /home/ncvs/src/sys/i386/i386/pmap.c,v retrieving revision 1.223 diff -u -r1.223 pmap.c --- pmap.c 1999/02/19 14:25:33 1.223 +++ pmap.c 1999/02/26 04:25:22 @@ -71,6 +71,7 @@ #include "opt_disable_pse.h" #include "opt_pmap.h" #include "opt_msgbuf.h" +#include "isa.h" #include #include @@ -3411,6 +3623,16 @@ return val; } +int +pmap_defer_avail(vm_offset_t pa) +{ +#if NISA > 0 + if (pa >= 0x0 && pa < 0x1000000) + return 1; +#endif + return 0; +} + void pmap_activate(struct proc *p) { Index: sys/i386/i386/machdep.c =================================================================== RCS file: /home/ncvs/src/sys/i386/i386/machdep.c,v retrieving revision 1.326 diff -u -r1.326 machdep.c --- machdep.c 1999/02/13 17:45:15 1.326 +++ machdep.c 1999/02/18 15:20:53 @@ -1544,7 +1545,9 @@ * so that we keep going. The first bad page * will terminate the loop. */ - if (phys_avail[pa_indx] == target_page) { + if (phys_avail[pa_indx] == target_page && + pmap_defer_avail(target_page) == + pmap_defer_avail(target_page - PAGE_SIZE)) { phys_avail[pa_indx] += PAGE_SIZE; if (speculative_mprobe == TRUE && phys_avail[pa_indx] >= (64*1024*1024)) Index: sys/vm/vm_page.c =================================================================== RCS file: /home/ncvs/src/sys/vm/vm_page.c,v retrieving revision 1.127 diff -u -r1.127 vm_page.c --- vm_page.c 1999/02/24 21:26:26 1.127 +++ vm_page.c 1999/02/26 04:10:32 @@ -201,11 +203,13 @@ register vm_page_t m; register struct vm_page **bucket; vm_size_t npages, page_range; + vm_size_t tmpnpages; register vm_offset_t new_start; int i; vm_offset_t pa; int nblocks; vm_offset_t first_managed_page; + int pass; /* the biggest memory array is the second group of pages */ vm_offset_t start; @@ -324,22 +328,30 @@ cnt.v_page_count = 0; cnt.v_free_count = 0; - for (i = 0; phys_avail[i + 1] && npages > 0; i += 2) { - if (i == biggestone) - pa = ptoa(first_managed_page); - else - pa = phys_avail[i]; - while (pa < phys_avail[i + 1] && npages-- > 0) { - ++cnt.v_page_count; - ++cnt.v_free_count; - m = PHYS_TO_VM_PAGE(pa); - m->phys_addr = pa; - m->flags = 0; - m->pc = (pa >> PAGE_SHIFT) & PQ_L2_MASK; - m->queue = m->pc + PQ_FREE; - TAILQ_INSERT_TAIL(vm_page_queues[m->queue].pl, m, pageq); - ++(*vm_page_queues[m->queue].lcnt); - pa += PAGE_SIZE; + for (pass = 0; pass < 2; pass++) { + tmpnpages = npages; + for (i = 0; phys_avail[i + 1] && npages > 0; i += 2) { + if (i == biggestone) + pa = ptoa(first_managed_page); + else + pa = phys_avail[i]; + while (pa < phys_avail[i + 1] && tmpnpages-- > 0) { + if (pass != pmap_defer_avail(pa)) { + pa += PAGE_SIZE; + continue; + } + ++cnt.v_page_count; + ++cnt.v_free_count; + m = PHYS_TO_VM_PAGE(pa); + m->phys_addr = pa; + m->flags = 0; + m->pc = (pa >> PAGE_SHIFT) & PQ_L2_MASK; + m->queue = m->pc + PQ_FREE; + TAILQ_INSERT_TAIL(vm_page_queues[m->queue].pl, + m, pageq); + ++(*vm_page_queues[m->queue].lcnt); + pa += PAGE_SIZE; + } } } return (mapped); ----Next_Part(Mon_Mar__8_19:22:22_1999)---- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Mar 8 12:30:24 1999 Delivered-To: freebsd-smp@freebsd.org Received: from cs.rice.edu (cs.rice.edu [128.42.1.30]) by hub.freebsd.org (Postfix) with ESMTP id 9BAE815259; Mon, 8 Mar 1999 12:30:21 -0800 (PST) (envelope-from alc@cs.rice.edu) Received: from nonpc.cs.rice.edu (nonpc.cs.rice.edu [128.42.1.219]) by cs.rice.edu (8.9.0/8.9.0) with ESMTP id OAA23842; Mon, 8 Mar 1999 14:29:48 -0600 (CST) Received: (from alc@localhost) by nonpc.cs.rice.edu (8.9.2/8.7.3) id OAA00544; Mon, 8 Mar 1999 14:29:47 -0600 (CST) Date: Mon, 8 Mar 1999 14:29:47 -0600 From: Alan Cox To: Charles Henrich Cc: Mike Smith , freebsd-hackers@freebsd.org, freebsd-smp@freebsd.org Subject: Re: Erorr: isa_dmainit(2, 1024) failed Message-ID: <19990308142947.A471@nonpc.cs.rice.edu> References: <19990308094100.14836@orbit.flnet.com> <199903081740.JAA06967@dingo.cdrom.com> <19990308101059.50251@orbit.flnet.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.1i In-Reply-To: <19990308101059.50251@orbit.flnet.com>; from Charles Henrich on Mon, Mar 08, 1999 at 10:10:59AM -0800 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Charles, Please try the following simpler patch. It should accomplish the same thing as Tor's patch. Replace the TAILQ_INSERT_TAIL in vm_page_startup (in vm_page.c) with a TAILQ_INSERT_HEAD. Alan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Mar 8 14: 0: 3 1999 Delivered-To: freebsd-smp@freebsd.org Received: from orbit.flnet.com (orbit.flnet.com [205.240.232.32]) by hub.freebsd.org (Postfix) with ESMTP id A0C4F159F3; Mon, 8 Mar 1999 14:00:00 -0800 (PST) (envelope-from henrich@orbit.flnet.com) Received: (from henrich@localhost) by orbit.flnet.com (8.8.5/8.8.4) id QAA15790; Mon, 8 Mar 1999 16:59:40 -0500 (EST) Message-ID: <19990308135939.46895@orbit.flnet.com> Date: Mon, 8 Mar 1999 13:59:39 -0800 From: Charles Henrich To: Alan Cox Cc: Mike Smith , freebsd-hackers@freebsd.org, freebsd-smp@freebsd.org Subject: Re: Erorr: isa_dmainit(2, 1024) failed Mail-Followup-To: Alan Cox , Mike Smith , freebsd-hackers@freebsd.org, freebsd-smp@freebsd.org References: <19990308094100.14836@orbit.flnet.com> <199903081740.JAA06967@dingo.cdrom.com> <19990308101059.50251@orbit.flnet.com> <19990308142947.A471@nonpc.cs.rice.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89i In-Reply-To: <19990308142947.A471@nonpc.cs.rice.edu>; from Alan Cox on Mon, Mar 08, 1999 at 02:29:47PM -0600 X-Operating-System: FreeBSD 2.2-BETA_A X-PGP-Fingerprint: 1024/F7 FD C7 3A F5 6A 23 BF 76 C4 B8 C9 6E 41 A4 4F Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On the subject of Re: Erorr: isa_dmainit(2, 1024) failed, Alan Cox stated: > Charles, > > Please try the following simpler patch. It should accomplish the same thing > as Tor's patch. > > Replace the TAILQ_INSERT_TAIL in vm_page_startup (in vm_page.c) with a > TAILQ_INSERT_HEAD. It works almost perfectly! Whenever I first open the audio device there is all sorts of wacky siezing of the seizing until the audio starts to play, then everything goes back to normal. In any case, thanks (!) -Crh Charles Henrich Manex Visual Effects henrich@flnet.com http://orbit.flnet.com/~henrich To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Mar 9 1:28: 7 1999 Delivered-To: freebsd-smp@freebsd.org Received: from peach.ocn.ne.jp (peach.ocn.ne.jp [210.145.254.87]) by hub.freebsd.org (Postfix) with ESMTP id 398B515046; Tue, 9 Mar 1999 01:27:42 -0800 (PST) (envelope-from dcs@newsguy.com) Received: from newsguy.com by peach.ocn.ne.jp (8.9.1a/OCN) id SAA05986; Tue, 9 Mar 1999 18:26:57 +0900 (JST) Message-ID: <36E4E25D.DE57F27B@newsguy.com> Date: Tue, 09 Mar 1999 17:57:01 +0900 From: "Daniel C. Sobral" X-Mailer: Mozilla 4.5 [en] (Win98; I) X-Accept-Language: pt-BR,ja MIME-Version: 1.0 To: Mike Smith Cc: Charles Henrich , freebsd-hackers@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: Erorr: isa_dmainit(2, 1024) failed References: <199903081740.JAA06967@dingo.cdrom.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Mike Smith wrote: > > > How does one fix this? :) > > With a text editor. Find the code that does it wrong, change it and > recompile. Send us the diffs when you're done. A text editor... :-) The ultimate bug-fixing tool! :-) I loved it... :-) -- Daniel C. Sobral (8-DCS) dcs@newsguy.com dcs@freebsd.org "FreeBSD is Yoda, Linux is Luke Skywalker." To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Mar 9 1:28: 7 1999 Delivered-To: freebsd-smp@freebsd.org Received: from peach.ocn.ne.jp (peach.ocn.ne.jp [210.145.254.87]) by hub.freebsd.org (Postfix) with ESMTP id 2AFB81515C; Tue, 9 Mar 1999 01:28:04 -0800 (PST) (envelope-from dcs@newsguy.com) Received: from newsguy.com by peach.ocn.ne.jp (8.9.1a/OCN) id SAA06086; Tue, 9 Mar 1999 18:27:32 +0900 (JST) Message-ID: <36E4E31B.9CA2AA70@newsguy.com> Date: Tue, 09 Mar 1999 18:00:11 +0900 From: "Daniel C. Sobral" X-Mailer: Mozilla 4.5 [en] (Win98; I) X-Accept-Language: pt-BR,ja MIME-Version: 1.0 To: Tor.Egge@fast.no Cc: mike@smith.net.au, henrich@flnet.com, freebsd-hackers@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG Subject: Re: Erorr: isa_dmainit(2, 1024) failed References: <199903081740.JAA06967@dingo.cdrom.com> <199903081827.TAA70476@midten.fast.no> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Tor.Egge@fast.no wrote: > > Index: sys/i386/i386/pmap.c > =================================================================== > RCS file: /home/ncvs/src/sys/i386/i386/pmap.c,v > retrieving revision 1.223 > diff -u -r1.223 pmap.c > --- pmap.c 1999/02/19 14:25:33 1.223 > +++ pmap.c 1999/02/26 04:25:22 > @@ -71,6 +71,7 @@ > #include "opt_disable_pse.h" > #include "opt_pmap.h" > #include "opt_msgbuf.h" > +#include "isa.h" ^^^^^ I have a slight suspicion that Bruce would not be very please with this... :-) -- Daniel C. Sobral (8-DCS) dcs@newsguy.com dcs@freebsd.org "FreeBSD is Yoda, Linux is Luke Skywalker." To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Mar 9 6:15:18 1999 Delivered-To: freebsd-smp@freebsd.org Received: from luke.pmr.com (luke.pmr.com [207.170.114.132]) by hub.freebsd.org (Postfix) with ESMTP id E787615341 for ; Tue, 9 Mar 1999 06:15:08 -0800 (PST) (envelope-from bob@luke.pmr.com) Received: (from bob@localhost) by luke.pmr.com (8.9.2/8.9.2) id IAA00684 for freebsd-smp@freebsd.org; Tue, 9 Mar 1999 08:14:52 -0600 (CST) (envelope-from bob) Date: Tue, 9 Mar 1999 08:14:52 -0600 From: Bob Willcox To: SMP list Subject: rtc0 stopping under load Message-ID: <19990309081452.A563@luke.pmr.com> Reply-To: Bob Willcox Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.1i Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I have a dual Xeon system here on which the rtc0 clock stops as soon as I put a load on it (its running right after boot). Once the interrupts for rtc0 stop they stay that way till next boot. When running a UP kernel rtc0 never stops. My hardware is: ASUS XG-DLS motherboard 2 400MHz Xeons/512KB cache 512MB ECC SDRAM Also, my son has a dual Celeron system on an ASUS P2B-D2 motherboard that has the exact same problem. Does anyone have any suggestions on what might be causing this or what I can do to further isolate the problem? Thanks, Bob -- Bob Willcox The man who follows the crowd will usually get no bob@luke.pmr.com further than the crowd. The man who walks alone is Austin, TX likely to find himself in places no one has ever been. -- Alan Ashley-Pitt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Tue Mar 9 23:57: 0 1999 Delivered-To: freebsd-smp@freebsd.org Received: from venus.dbft.daimlerbenz.com (venus.DaimlerBenz.com [141.113.7.196]) by hub.freebsd.org (Postfix) with ESMTP id 851EE14EF7 for ; Tue, 9 Mar 1999 23:56:54 -0800 (PST) (envelope-from andreas.strobel@dbag.ulm.daimlerbenz.com) Received: (from uucp@localhost) by venus.dbft.daimlerbenz.com (8.7.1/8.7.1) id JAA03714 for ; Wed, 10 Mar 1999 09:07:35 +0100 (MET) Received: from axmail.dbag.ulm.daimlerbenz.com(53.16.7.62) by venus.dbft.daimlerbenz.com via smap (3.2) id xma003569; Wed, 10 Mar 99 09:07:09 +0100 Received: from dagobert by axmail.dbag.ulm.DaimlerBenz.COM (5.65v4.0/1.1.8.2/9-Juli-1996-a) id AA02244; Wed, 10 Mar 1999 08:56:07 +0100 Received: from balu by dagobert.dbag.ulm.DaimlerBenz.COM (4.1/SMI-4.1-18.9.1995-gm) id AA04301; Wed, 10 Mar 99 08:56:16 +0100 Message-Id: <36E6260F.43AF@dbag.ulm.daimlerbenz.com> Date: Wed, 10 Mar 1999 08:58:07 +0100 From: Andreas Strobel Organization: Daimler Benz AG X-Mailer: Mozilla 3.0Gold (X11; I; HP-UX B.10.20 9000/735) Mime-Version: 1.0 To: freebsd-smp@FreeBSD.ORG Subject: isa error Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, using the sound with the luigi sound driver causes the following error message: timeout flushing dbuf_out chan 7 cnt 0xff00 flags 0x00000141 i hear only a second music than the speakers are silent. I have FreeBSD 3.0 and two PIIs with 128MB of ram. Does anyone out there have any ideas whats causing this, and what a solution might be? Thanks! - Andreas To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Mar 10 5: 9:13 1999 Delivered-To: freebsd-smp@freebsd.org Received: from excalibur.lps.ens.fr (excalibur.lps.ens.fr [129.199.120.3]) by hub.freebsd.org (Postfix) with ESMTP id 25AEE14C92 for ; Wed, 10 Mar 1999 05:09:04 -0800 (PST) (envelope-from Thierry.Besancon@lps.ens.fr) Received: (from besancon@localhost) by excalibur.lps.ens.fr (8.8.5/8.8.6) id NAA24194; Wed, 10 Mar 1999 13:08:41 GMT Message-Id: <199903101308.NAA24194@excalibur.lps.ens.fr> From: Thierry.Besancon@lps.ens.fr (Thierry Besancon) Date: Wed, 10 Mar 1999 14:08:40 +0000 X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: freebsd-smp@freebsd.org, dwhite@gdi.uoregon.edu Subject: lockmgr() panic Cc: besancon@lps.ens.fr, Pierre.David@prism.uvsq.fr, jt@ratp.fr Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hello I have in production a ASUS P2B machine with dual 350 MHz PII (with Intel 440BX chip ?) and 512 Mb of ram. It runs FreeBSD 3.1 It paniced last week so I compiled in DDB in order to have more information. Today it paniced again, with the following messages : Fatal trap 12: page fault while in kernel mode mp_lock = 010000002; cpuid = 1; lapic.id = 00000000 fault virtual address = 0xb8 fault code = supervisor read, page not present instruction pointer = 0x8:0xf01430a7 stack pointer = 0x10:0xff804eb4 frame pointer = 0x10:0xff804ec8 code segment = base 0x0, limit 0xfffff, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags = interrupt enabled, resume, IOPL = 0 current process = Idle interrupt mask = net tty bio cam <- SMP : XXX kernel: type 12 trap, code=0 Stopped at tsleep+0x1b: movl 0xb8(%ebx),%eax The "trace" command gave : db> trace tsleep(f027bab4,4,f023b455,0,f027bab4) at tsleep+0x1b acquire(f027bab4,1000000,600,f027bab4,c6040000) at acquire+0x91 lockmgr(f027bab4,2,0,0,1) at lockmgr+0x260 kmem_malloc(f027bab4,1000,1,f113f300,ff804f88) at kmem_malloc+0x51 m_clalloc(1,1) ar m_clalloc+0x2e fxp_add_rfabuf(f1762e00,f11e9980) at fxp_add_rfabuf+0xa5 fxp_intr(f1762e00,0,f0250010,10,0) at fxp_intr+0xe1 Xresume18() at Xresume18+0x51 --- interrupt, eip = 0xf020dfe0, esp = 0xff804ff0, ebp = 0 --- default_halt() at default_halt (I rebooted with a new kernel with Alan Cox's path about "update_needed" in src/sys/i386/i386/pmap.c,v just in case it might help but i don't have any idea whether this patch relates to my bug) My machine doesn't panic when running this piece of C code found in the mailing list : #include #include #include #include #define SIZE (32 * 1024 * 1024) int main(void) { void *p; char vec[SIZE / PAGE_SIZE]; p = malloc(SIZE); if (mincore(p, SIZE, vec) != 0) err(1, "mincore"); return (0); } While typing this mail, my machine paniced again with a more explicit message ? panic: lockmgr: locking against myself mp_lock = 010000003 ; cpuid = 1; lapic.id = 00000000 Debugger("panic") Stopped at Debugger+0x37: movl $0,in_Debugger db> trace Debugger(f0230f52) at Debugger+0x37 panic() at panic+0xa4 lockmgr() at lockmgr+0x228 vm_map_growstack() at vm_map_growstack+0x29 grow_stack() at grow_stack+0xe trap_pfault() at trap+0x33e calltap() at calltrap+0x3c --- trap 0xc, eip = 0xf020c3a3, esp = 0xfa7f6d6c, ebp = 0xfa7f6d84 --- pmap_enter() at pmap_enter+0xa7 vm_fault() at vm_fault+0x83a trap_pfault() at trap_pfault+0xc4 trap() at trap+0x33e calltrap() at calltrap+0x3c --- trap 0xc, eip = ..., esp = ..., ebp = ... --- zalloci() at zalloci+0x33 vm_map_entry_create() at vm_map_entry_create+0x27 _vm_map_clip_end() at _vm_map_clip_end+0x42 vm_map_madvise() at vm_map_madvise+0xc9 madvise() at madvise+0x47 syscall(27,27,2c,2811c000,efbfb9ec) at syscall+0x187 Xint0x80_syscall() at Xint0x80_syscall+0x4c So, does anybody have an idea about why my problem occurs ? Any fix ? Thanks in advance. A special thanks for all the people that made FreeBSD runnable on SMP. Thierry ******************************************************************************** ******************************************************************************** ps 1 : here's my "mptable" output =============================================================================== MPTable, version 2.0.15 ------------------------------------------------------------------------------- MP Floating Pointer Structure: location: BIOS physical address: 0x000f6df0 signature: '_MP_' length: 16 bytes version: 1.1 checksum: 0x3f mode: Virtual Wire ------------------------------------------------------------------------------- MP Config Table Header: physical address: 0x000f69ec signature: 'PCMP' base table length: 252 version: 1.1 checksum: 0x5c OEM ID: 'OEM00000' Product ID: 'PROD00000000' OEM table pointer: 0x00000000 OEM table size: 0 entry count: 23 local APIC address: 0xfee00000 extended table length: 0 extended table checksum: 0 ------------------------------------------------------------------------------- MP Config Base Table Entries: -- Processors: APIC ID Version State Family Model Step Flags 1 0x11 BSP, usable 6 5 2 0x183fbff 0 0x11 AP, usable 6 5 2 0x183fbff -- Bus: Bus ID Type 0 PCI 1 PCI 2 ISA -- I/O APICs: APIC ID Version State Address 2 0x11 usable 0xfec00000 -- I/O Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT conforms conforms 2 0 2 0 INT conforms conforms 2 1 2 1 INT conforms conforms 2 0 2 2 INT conforms conforms 2 3 2 3 INT conforms conforms 2 4 2 4 INT conforms conforms 2 5 2 5 INT conforms conforms 2 6 2 6 INT conforms conforms 2 7 2 7 INT conforms conforms 2 8 2 8 INT conforms conforms 2 9 2 9 INT conforms conforms 2 14 2 14 INT conforms conforms 2 15 2 15 INT active-lo level 2 11 2 16 INT active-lo level 2 10 2 18 INT active-lo level 2 12 2 19 -- Local Ints: Type Polarity Trigger Bus ID IRQ APIC ID PIN# ExtINT active-hi edge 2 0 255 0 NMI active-hi edge 2 0 255 1 ------------------------------------------------------------------------------- # SMP kernel config file options: # Required: options SMP # Symmetric MultiProcessor Kernel options APIC_IO # Symmetric (APIC) I/O # Optional (built-in defaults will work in most cases): #options NCPU=2 # number of CPUs #options NBUS=3 # number of busses #options NAPIC=1 # number of IO APICs #options NINTR=24 # number of INTs =============================================================================== ******************************************************************************** ******************************************************************************** ps 2 : here's my dmesg output : Copyright (c) 1992-1999 FreeBSD Inc. Copyright (c) 1982, 1986, 1989, 1991, 1993 The Regents of the University of California. All rights reserved. FreeBSD 3.1-RELEASE #5: Wed Mar 10 13:12:39 CET 1999 besancon@tournesol.lps.ens.fr:/usr/src/sys/compile/TOURNESOL Timecounter "i8254" frequency 1193182 Hz CPU: Pentium II/Xeon/Celeron (686-class CPU) Origin = "GenuineIntel" Id = 0x652 Stepping=2 Features=0x183fbff> real memory = 536870912 (524288K bytes) avail memory = 519774208 (507592K bytes) Programming 24 pins in IOAPIC #0 FreeBSD/SMP: Multiprocessor motherboard cpu0 (BSP): apic id: 1, version: 0x00040011, at 0xfee00000 cpu1 (AP): apic id: 0, version: 0x00040011, at 0xfee00000 io0 (APIC): apic id: 2, version: 0x00170011, at 0xfec00000 Preloaded elf kernel "kernel" at 0xf02d7000. Probing for devices on PCI bus 0: chip0: rev 0x02 on pci0.0.0 chip1: rev 0x02 on pci0.1.0 chip2: rev 0x02 on pci0.4.0 ide_pci0: rev 0x01 on pci0.4.1 chip3: rev 0x02 on pci0.4.3 fxp0: rev 0x02 int a irq 18 on pci0.10.0 fxp0: Ethernet address 00:a0:c9:43:9d:99 ncr0: rev 0x26 int a irq 16 on pci0.12.0 Probing for devices on PCI bus 1: vga0: rev 0x5c int a irq 16 on pci1.0.0 Probing for devices on the ISA bus: sc0 on isa sc0: VGA color <16 virtual consoles, flags=0x0> atkbdc0 at 0x60-0x6f on motherboard atkbd0 irq 1 on isa psm0 not found sio0 at 0x3f8-0x3ff irq 4 flags 0x10 on isa sio0: type 16550A fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: FIFO enabled, 8 bytes threshold fd0: 1.44MB 3.5in wdc0 at 0x1f0-0x1f7 irq 14 on isa wdc0: unit 0 (wd0): wd0: 2014MB (4124736 sectors), 4092 cyls, 16 heads, 63 S/T, 512 B/S wdc1 at 0x170-0x177 irq 15 on isa wdc1: unit 0 (atapi): , removable, accel, dma, iordis acd0: drive speed 2412KB/sec, 128KB cache acd0: supported read types: CD-R, CD-RW, CD-DA, packet track acd0: Audio: play, 256 volume levels acd0: Mechanism: ejectable tray acd0: Medium: no/blank disc inside, unlocked scd0 not found at 0x230 vga0 at 0x3b0-0x3df maddr 0xa0000 msize 131072 on isa npx0 on motherboard npx0: INT 16 interface APIC_IO: Testing 8254 interrupt delivery APIC_IO: routing 8254 via pin 2 SMP: AP CPU #1 Launched! changing root device to da0s1a da0 at ncr0 bus 0 target 0 lun 0 da0: Fixed Direct Access SCSI-2 device da0: 40.0MB/s transfers (20.0MHz, offset 16, 16bit), Tagged Queueing Enabled da0: 4350MB (8910423 512 byte sectors: 255H 63S/T 554C) ******************************************************************************** ******************************************************************************** ps 3 : here's my kernel config file : machine "i386" cpu "I686_CPU" ident TOURNESOL-SMP maxusers 128 options MATH_EMULATE #Support for x87 emulation options INET #InterNETworking options FFS #Berkeley Fast Filesystem options FFS_ROOT #FFS usable as root device [keep this!] options MFS #Memory Filesystem options MFS_ROOT #MFS usable as root device, "MFS" req'ed options NFS #Network Filesystem options NFS_ROOT #NFS usable as root device, "NFS" req'ed options MSDOSFS #MSDOS Filesystem options "CD9660" #ISO 9660 Filesystem options "CD9660_ROOT" #CD-ROM usable as root. "CD9660" req'ed options PROCFS #Process filesystem options "COMPAT_43" #Compatible with BSD 4.3 [KEEP THIS!] options SCSI_DELAY=1500 #Be pessimistic about Joe SCSI device options UCONSOLE #Allow users to grab the console options FAILSAFE #Be conservative options USERCONFIG #boot -c editor options VISUAL_USERCONFIG #visual boot -c editor config kernel root on wd0 # To make an SMP kernel, the next two are needed options SMP # Symmetric MultiProcessor Kernel options APIC_IO # Symmetric (APIC) I/O controller isa0 controller eisa0 controller pci0 controller fdc0 at isa? port "IO_FD1" bio irq 6 drq 2 disk fd0 at fdc0 drive 0 options "CMD640" # work around CMD640 chip deficiency controller wdc0 at isa? port "IO_WD1" bio irq 14 disk wd0 at wdc0 drive 0 disk wd1 at wdc0 drive 1 controller wdc1 at isa? port "IO_WD2" bio irq 15 disk wd2 at wdc1 drive 0 disk wd3 at wdc1 drive 1 # # ATAPI enables the support for ATAPI-compatible IDE devices # options ATAPI #Enable ATAPI support for IDE bus options ATAPI_STATIC #Don't do it as an LKM # # This option allow you to override the default probe time for IDE # devices, to get a faster probe. Setting this below 10000 violate # the IDE specs, but may still work for you (it will work for most # people). # options IDE_DELAY=8000 # Be optimistic about Joe IDE device # IDE CD-ROM & CD-R/RW driver - requires wdc controller and ATAPI option device acd0 device da0 #SCSI direct access devices (aka disks) device sa0 #SCSI tapes device pass0 #CAM passthrough driver device cd0 #SCSI CD-ROMs device ch0 #SCSI media changers device scd0 at isa? port 0x230 bio # atkbdc0 controlls both the keyboard and the PS/2 mouse controller atkbdc0 at isa? port IO_KBD tty device atkbd0 at isa? tty irq 1 device psm0 at isa? tty irq 12 device vga0 at isa? port ? conflicts # splash screen/screen saver pseudo-device splash # syscons is the default console driver, resembling an SCO console device sc0 at isa? tty # Enable this and PCVT_FREEBSD for pcvt vt220 compatible console driver #device vt0 at isa? tty #options XSERVER # support for X server #options FAT_CURSOR # start with block cursor # If you have a ThinkPAD, uncomment this along with the rest of the PCVT lines #options PCVT_SCANSET=2 # IBM keyboards are non-std # # The Numeric Processing eXtension driver. This should be configured if # your machine has a math co-processor, unless the coprocessor is very # buggy. If it is not configured then you *must* configure math emulation # (see above). If both npx0 and emulation are configured, then only npx0 # is used (provided it works). device npx0 at isa? port IO_NPX irq 13 device sio0 at isa? port "IO_COM1" flags 0x10 tty irq 4 #device sio1 at isa? port "IO_COM2" tty irq 3 ## Parallel port #device ppc0 at isa? port? net irq 7 #controller ppbus0 #device nlpt0 at ppbus? #device plip0 at ppbus? #device ppi0 at ppbus? ##controller vpo0 at ppbus? # Order is important here due to intrusive probes, do *not* alphabetize # this list of network interfaces until the probes have been fixed. # Right now it appears that the ie0 must be probed before ep0. See # revision 1.20 of this file. device de0 device fxp0 pseudo-device loop pseudo-device ether pseudo-device pty 256 # Enable the kernel debugger. options DDB #options DDB_UNATTENDED # KTRACE enables the system-call tracing facility ktrace(2). # This adds 4 KB bloat to your kernel, and slightly increases # the costs of each syscall. options KTRACE #kernel tracing # This provides support for System V shared memory and message queues. # options SYSVSHM options SYSVMSG options SYSVSEM # The `bpfilter' pseudo-device enables the Berkeley Packet Filter. Be # aware of the legal and administrative consequences of enabling this # option. The number of devices determines the maximum number of # simultaneous BPF clients programs runnable. #pseudo-device bpfilter 4 #Berkeley packet filter ##################################################################### options NETATALK #Appletalk communications protocols controller ncr0 controller scbus0 at ncr0 disk da0 at scbus0 target 0 disk da1 at scbus0 target 1 disk da2 at scbus0 target 2 disk da3 at scbus0 target 3 disk da4 at scbus0 target 4 disk da5 at scbus0 target 5 disk da6 at scbus0 target 6 options SCSI_REPORT_GEOMETRY options "MD5" options "ICMP_BANDLIM" # # This allows you to actually store this configuration file into # the kernel binary itself, where it may be later read by saying: # strings -aout -n 3 /kernel | grep ^___ | sed -e 's/^___//' > MYKERNEL # options INCLUDE_CONFIG_FILE # Include this file in kernel # # If you want to disable loadable kernel modules (LKM), you # might want to use this option. # #options NO_LKM # # Disable swapping. This option removes all code which actually performs # swapping, so it's not possible to turn it back on at run-time. # # This is sometimes usable for systems which don't have any swap space # (see also sysctls "vm.defer_swapspace_pageouts" and # "vm.disable_swapspace_pageouts") # #options NO_SWAPPING #pseudo-device vn #Vnode driver (turns a file into a device) pseudo-device snp 3 #Snoop device - to look at pty/vty/etc.. #pseudo-device ppp 1 #pseudo-device tun 1 #pseudo-device gzip # Exec gzipped a.out's To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Mar 10 6: 3:45 1999 Delivered-To: freebsd-smp@freebsd.org Received: from excalibur.lps.ens.fr (excalibur.lps.ens.fr [129.199.120.3]) by hub.freebsd.org (Postfix) with ESMTP id 66D21150EB for ; Wed, 10 Mar 1999 06:03:39 -0800 (PST) (envelope-from Thierry.Besancon@lps.ens.fr) Received: (from besancon@localhost) by excalibur.lps.ens.fr (8.8.5/8.8.6) id OAA28667; Wed, 10 Mar 1999 14:02:40 GMT To: "Richard Seaman Jr." , Thierry Besancon Subject: Re: lockmgr() panic References: <199903101308.NAA24194@excalibur.lps.ens.fr> <19990310075417.D4440@tar.com> Cc: Pierre.David@prism.uvsq.fr, jt@ratp.fr, freebsd-smp@freebsd.org, besancon@lps.ens.fr From: Thierry.Besancon@lps.ens.fr Date: 10 Mar 1999 15:02:40 +0100 In-Reply-To: "Richard Seaman, Jr."'s message of Wed, 10 Mar 1999 07:54:17 -0600 Message-ID: Lines: 26 X-Mailer: Gnus v5.3/Emacs 19.34 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Dixit "Richard Seaman, Jr." (le Wed, 10 Mar 1999 07:54:17 -0600) : >> >> On Wed, Mar 10, 1999 at 02:08:40PM +0000, Thierry Besancon wrote: >> > So, does anybody have an idea about why my problem occurs ? >> > >> > Any fix ? >> >> How up to date is your source tree? I believe Alan Cox committed >> changes to vm_grow_stack and to mincore that were designed to >> fix this problem. I forget the exact date of the commits, but >> the latest is probably not more than a week old. It's 3.1-RELEASE, plain stock. Given it is a production machine, I generally don't keep in sync with current source tree. I freeze a configuration and I keep it over time up to a next major release. Of course, I'm willing to solve my problem so I can update the kernel source tree if it gives a working and stable kernel, something that might be in production without any new trouble during weeks... :-) Thierry ps: I gonna subscribe to commit mailing list... To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Mar 10 6:37:56 1999 Delivered-To: freebsd-smp@freebsd.org Received: from lor.watermarkgroup.com (lor.watermarkgroup.com [207.202.73.33]) by hub.freebsd.org (Postfix) with ESMTP id CD5AE15132 for ; Wed, 10 Mar 1999 06:37:53 -0800 (PST) (envelope-from luoqi@watermarkgroup.com) Received: (from luoqi@localhost) by lor.watermarkgroup.com (8.8.8/8.8.8) id JAA11204; Wed, 10 Mar 1999 09:35:39 -0500 (EST) (envelope-from luoqi) Date: Wed, 10 Mar 1999 09:35:39 -0500 (EST) From: Luoqi Chen Message-Id: <199903101435.JAA11204@lor.watermarkgroup.com> To: Thierry.Besancon@lps.ens.fr, dwhite@gdi.uoregon.edu, freebsd-smp@FreeBSD.ORG Subject: Re: lockmgr() panic Cc: Pierre.David@prism.uvsq.fr, alc@cs.rice.edu, besancon@lps.ens.fr, dillon@apollo.backplane.com, jt@ratp.fr Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I have in production a ASUS P2B machine with dual 350 MHz PII > (with Intel 440BX chip ?) and 512 Mb of ram. > It runs FreeBSD 3.1 > > It paniced last week so I compiled in DDB in order to have more > information. Today it paniced again, with the following messages : > > Fatal trap 12: page fault while in kernel mode > mp_lock = 010000002; cpuid = 1; lapic.id = 00000000 > fault virtual address = 0xb8 > fault code = supervisor read, page not present > instruction pointer = 0x8:0xf01430a7 > stack pointer = 0x10:0xff804eb4 > frame pointer = 0x10:0xff804ec8 > code segment = base 0x0, limit 0xfffff, type 0x1b > = DPL 0, pres 1, def32 1, gran 1 > processor eflags = interrupt enabled, resume, IOPL = 0 > current process = Idle > interrupt mask = net tty bio cam <- SMP : XXX > kernel: type 12 trap, code=0 > Stopped at tsleep+0x1b: movl 0xb8(%ebx),%eax > > The "trace" command gave : > > db> trace > tsleep(f027bab4,4,f023b455,0,f027bab4) at tsleep+0x1b > acquire(f027bab4,1000000,600,f027bab4,c6040000) at acquire+0x91 > lockmgr(f027bab4,2,0,0,1) at lockmgr+0x260 > kmem_malloc(f027bab4,1000,1,f113f300,ff804f88) at kmem_malloc+0x51 > m_clalloc(1,1) ar m_clalloc+0x2e > fxp_add_rfabuf(f1762e00,f11e9980) at fxp_add_rfabuf+0xa5 > fxp_intr(f1762e00,0,f0250010,10,0) at fxp_intr+0xe1 > Xresume18() at Xresume18+0x51 > --- interrupt, eip = 0xf020dfe0, esp = 0xff804ff0, ebp = 0 --- > default_halt() at default_halt > kmem_malloc() blocked when trying to lock the mbuf map, in an interrupt context, even though M_NOWAIT was set. We need a non-blocking version of vm_map_lock() [vm_map_lock_try()?] in this situation. I'm forwarding this to Alan Cox and Matt Dillon... > While typing this mail, my machine paniced again with a more explicit message ? > > panic: lockmgr: locking against myself > mp_lock = 010000003 ; cpuid = 1; lapic.id = 00000000 > Debugger("panic") > Stopped at Debugger+0x37: movl $0,in_Debugger > > db> trace > Debugger(f0230f52) at Debugger+0x37 > panic() at panic+0xa4 > lockmgr() at lockmgr+0x228 > vm_map_growstack() at vm_map_growstack+0x29 > grow_stack() at grow_stack+0xe > trap_pfault() at trap+0x33e > calltap() at calltrap+0x3c > --- trap 0xc, eip = 0xf020c3a3, esp = 0xfa7f6d6c, ebp = 0xfa7f6d84 --- > pmap_enter() at pmap_enter+0xa7 > vm_fault() at vm_fault+0x83a > trap_pfault() at trap_pfault+0xc4 > trap() at trap+0x33e > calltrap() at calltrap+0x3c > --- trap 0xc, eip = ..., esp = ..., ebp = ... --- > zalloci() at zalloci+0x33 > vm_map_entry_create() at vm_map_entry_create+0x27 > _vm_map_clip_end() at _vm_map_clip_end+0x42 > vm_map_madvise() at vm_map_madvise+0xc9 > madvise() at madvise+0x47 > syscall(27,27,2c,2811c000,efbfb9ec) at syscall+0x187 > Xint0x80_syscall() at Xint0x80_syscall+0x4c > This is a completely different problem. It looks like a page fault on a kernel page table page, which should have been wired down. Could you provide more detail, e.g. fault address for both traps? > > So, does anybody have an idea about why my problem occurs ? > > Any fix ? > > Thanks in advance. > A special thanks for all the people that made FreeBSD runnable on SMP. > > Thierry -lq To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Mar 10 6:44:31 1999 Delivered-To: freebsd-smp@freebsd.org Received: from mel.alcatel.fr (mel.alcatel.fr [212.208.74.132]) by hub.freebsd.org (Postfix) with ESMTP id C4364150F6 for ; Wed, 10 Mar 1999 06:44:18 -0800 (PST) (envelope-from Thierry.Herbelot@alcatel.fr) Received: from aifhs2.alcatel.fr (mailhub.alcatel.fr [155.132.180.80]) by mel.alcatel.fr (ALCANET/SMTP) with ESMTP id PAA16341; Wed, 10 Mar 1999 15:42:32 +0100 Received: from lune.telspace.alcatel.fr (lune.telspace.alcatel.fr [155.132.144.65]) by aifhs2.alcatel.fr (ALCANET/SMTP2) with ESMTP id PAA13619; Wed, 10 Mar 1999 15:40:07 +0100 (MET) Received: from telss1.telspace.alcatel.fr (telss1.telspace.alcatel.fr [155.132.51.4]) by lune.telspace.alcatel.fr (8.9.1a/8.9.1) with SMTP id PAA20304; Wed, 10 Mar 1999 15:21:07 +0100 (MET) Received: from telspace.alcatel.fr (nairobi.telspace.alcatel.fr) by telss1.telspace.alcatel.fr (4.1/SMI-4.1) id AA02027; Wed, 10 Mar 99 15:30:53 +0100 Message-Id: <36E68355.40CC8A44@telspace.alcatel.fr> Date: Wed, 10 Mar 1999 15:36:05 +0100 From: HERBELOT Thierry Organization: Alcatel Telspace X-Mailer: Mozilla 4.5 [en] (X11; I; SunOS 5.5.1 sun4m) X-Accept-Language: en Mime-Version: 1.0 To: Thierry.Besancon@lps.ens.fr Cc: "Richard Seaman Jr." , Thierry Besancon , Pierre.David@prism.uvsq.fr, jt@ratp.fr, freebsd-smp@FreeBSD.ORG, besancon@lps.ens.fr Subject: Re: lockmgr() panic References: <199903101308.NAA24194@excalibur.lps.ens.fr> <19990310075417.D4440@tar.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Thierry.Besancon@lps.ens.fr wrote: > [SNIP] > > It's 3.1-RELEASE, plain stock. > > Given it is a production machine, I generally don't keep in sync with > current source tree. I freeze a configuration and I keep it over time ^^^^^^^ BzzzT !!! 3.1 is the new -Stable Branch. Even on a production machine, you should follow it (for example every other week : a script was given in Greg Lehey's Complete FreeBSD book just for this purpose : cvsup of the sources, followed by a make world and reconfig of kernel + reboot - as it is donne automatically, it can be done at a time when there are no users) TfH > up to a next major release. > Of course, I'm willing to solve my problem so I can update the kernel > source tree if it gives a working and stable kernel, something that > might be in production without any new trouble during weeks... :-) > > Thierry > > ps: I gonna subscribe to commit mailing list... > > 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 Wed Mar 10 6:48:28 1999 Delivered-To: freebsd-smp@freebsd.org Received: from excalibur.lps.ens.fr (excalibur.lps.ens.fr [129.199.120.3]) by hub.freebsd.org (Postfix) with ESMTP id 34D1A1505D for ; Wed, 10 Mar 1999 06:47:52 -0800 (PST) (envelope-from Thierry.Besancon@lps.ens.fr) Received: (from besancon@localhost) by excalibur.lps.ens.fr (8.8.5/8.8.6) id OAA01137; Wed, 10 Mar 1999 14:45:25 GMT To: Luoqi Chen Cc: Thierry.Besancon@tournesol.lps.ens.fr, dwhite@gdi.uoregon.edu, freebsd-smp@FreeBSD.ORG, Pierre.David@prism.uvsq.fr, alc@cs.rice.edu, dillon@apollo.backplane.com, jt@ratp.fr Subject: Re: lockmgr() panic References: <199903101435.JAA11204@lor.watermarkgroup.com> From: Thierry.Besancon@lps.ens.fr Date: 10 Mar 1999 15:45:24 +0100 In-Reply-To: Luoqi Chen's message of Wed, 10 Mar 1999 09:35:39 -0500 (EST) Message-ID: Lines: 58 X-Mailer: Gnus v5.3/Emacs 19.34 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Dixit Luoqi Chen (le Wed, 10 Mar 1999 09:35:39 -0500 (EST)) : >> > While typing this mail, my machine paniced again with a more explicit message ? >> > >> > panic: lockmgr: locking against myself >> > mp_lock = 010000003 ; cpuid = 1; lapic.id = 00000000 >> > Debugger("panic") >> > Stopped at Debugger+0x37: movl $0,in_Debugger >> > >> > db> trace >> > Debugger(f0230f52) at Debugger+0x37 >> > panic() at panic+0xa4 >> > lockmgr() at lockmgr+0x228 >> > vm_map_growstack() at vm_map_growstack+0x29 >> > grow_stack() at grow_stack+0xe >> > trap_pfault() at trap+0x33e >> > calltap() at calltrap+0x3c >> > --- trap 0xc, eip = 0xf020c3a3, esp = 0xfa7f6d6c, ebp = 0xfa7f6d84 --- >> > pmap_enter() at pmap_enter+0xa7 >> > vm_fault() at vm_fault+0x83a >> > trap_pfault() at trap_pfault+0xc4 >> > trap() at trap+0x33e >> > calltrap() at calltrap+0x3c >> > --- trap 0xc, eip = ..., esp = ..., ebp = ... --- >> > zalloci() at zalloci+0x33 >> > vm_map_entry_create() at vm_map_entry_create+0x27 >> > _vm_map_clip_end() at _vm_map_clip_end+0x42 >> > vm_map_madvise() at vm_map_madvise+0xc9 >> > madvise() at madvise+0x47 >> > syscall(27,27,2c,2811c000,efbfb9ec) at syscall+0x187 >> > Xint0x80_syscall() at Xint0x80_syscall+0x4c >> > >> This is a completely different problem. Life sucks today. Reading src/sys/kern/kern_lock.c, I saw this piece of code that might be relevant with this problem : #if !defined(MAX_PERF) if ((extflags & LK_CANRECURSE) == 0) panic("lockmgr: locking against myself"); #endif Before defining the flag, does anybody know what MAX_PERF implies other than not panicing the kernel ? What does it means ? >> It looks like a page fault on >> a kernel page table page, which should have been wired down. Could you >> provide more detail, e.g. fault address for both traps? I'm sorry, I can't. There were too many adresses and I thought they were irrelevant so I didn't note them... I'll note them next time (i hope not to have to do it though ;-) Thierry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Mar 10 10:17:21 1999 Delivered-To: freebsd-smp@freebsd.org Received: from cs.rice.edu (cs.rice.edu [128.42.1.30]) by hub.freebsd.org (Postfix) with ESMTP id 84CFB14BDE for ; Wed, 10 Mar 1999 10:17:15 -0800 (PST) (envelope-from alc@cs.rice.edu) Received: from nonpc.cs.rice.edu (nonpc.cs.rice.edu [128.42.1.219]) by cs.rice.edu (8.9.0/8.9.0) with ESMTP id MAA13713; Wed, 10 Mar 1999 12:12:57 -0600 (CST) Received: (from alc@localhost) by nonpc.cs.rice.edu (8.9.2/8.7.3) id MAA03237; Wed, 10 Mar 1999 12:12:57 -0600 (CST) Date: Wed, 10 Mar 1999 12:12:56 -0600 From: Alan Cox To: Luoqi Chen Cc: Thierry.Besancon@lps.ens.fr, dwhite@gdi.uoregon.edu, freebsd-smp@freebsd.org, Pierre.David@prism.uvsq.fr, besancon@lps.ens.fr, dillon@apollo.backplane.com, jt@ratp.fr Subject: Re: lockmgr() panic Message-ID: <19990310121256.Q362@nonpc.cs.rice.edu> References: <199903101435.JAA11204@lor.watermarkgroup.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.1i In-Reply-To: <199903101435.JAA11204@lor.watermarkgroup.com>; from Luoqi Chen on Wed, Mar 10, 1999 at 09:35:39AM -0500 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I agree with Luoqi. It appears that you are running into two different problems. Can you try applying the following two changes? rev 1.38 of vm_map.h rev 1.154 of vm_map.c They may address one of the problems. Alan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Mar 10 10:53:50 1999 Delivered-To: freebsd-smp@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [209.157.86.2]) by hub.freebsd.org (Postfix) with ESMTP id 569BD15197 for ; Wed, 10 Mar 1999 10:53:45 -0800 (PST) (envelope-from dillon@apollo.backplane.com) Received: (from dillon@localhost) by apollo.backplane.com (8.9.3/8.9.1) id KAA56970; Wed, 10 Mar 1999 10:50:36 -0800 (PST) (envelope-from dillon) Date: Wed, 10 Mar 1999 10:50:36 -0800 (PST) From: Matthew Dillon Message-Id: <199903101850.KAA56970@apollo.backplane.com> To: Thierry.Besancon@lps.ens.fr Cc: Luoqi Chen , Thierry.Besancon@tournesol.lps.ens.fr, dwhite@gdi.uoregon.edu, freebsd-smp@FreeBSD.ORG, Pierre.David@prism.uvsq.fr, alc@cs.rice.edu, jt@ratp.fr Subject: Re: lockmgr() panic References: <199903101435.JAA11204@lor.watermarkgroup.com> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org :>> This is a completely different problem. : :Life sucks today. : :Reading src/sys/kern/kern_lock.c, I saw this piece of code that might :be relevant with this problem : : :#if !defined(MAX_PERF) : if ((extflags & LK_CANRECURSE) == 0) : panic("lockmgr: locking against myself"); :#endif : :Before defining the flag, does anybody know what MAX_PERF implies :other than not panicing the kernel ? What does it means ? MAX_PERF means 'maximum performance'. Note that the conditional is an '#if !defined ...' conditional, not an '#if defined ...' conditional. This means that by default, when MAX_PERF is *not* defined, the above code *will* be executed. If MAX_PERF is defined, the system removes many sanity checks from the kernel that would otherwise be compiled in. I would recommend that you *NOT* define it. -Matt : Thierry : Matthew Dillon To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Mar 10 10:55:11 1999 Delivered-To: freebsd-smp@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [209.157.86.2]) by hub.freebsd.org (Postfix) with ESMTP id 8E5E314DD0 for ; Wed, 10 Mar 1999 10:55:10 -0800 (PST) (envelope-from dillon@apollo.backplane.com) Received: (from dillon@localhost) by apollo.backplane.com (8.9.3/8.9.1) id KAA57013; Wed, 10 Mar 1999 10:54:38 -0800 (PST) (envelope-from dillon) Date: Wed, 10 Mar 1999 10:54:38 -0800 (PST) From: Matthew Dillon Message-Id: <199903101854.KAA57013@apollo.backplane.com> To: Alan Cox Cc: Luoqi Chen , Thierry.Besancon@lps.ens.fr, dwhite@gdi.uoregon.edu, freebsd-smp@freebsd.org, Pierre.David@prism.uvsq.fr, besancon@lps.ens.fr, jt@ratp.fr Subject: Re: lockmgr() panic References: <199903101435.JAA11204@lor.watermarkgroup.com> <19990310121256.Q362@nonpc.cs.rice.edu> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org :I agree with Luoqi. It appears that you are running into two different :problems. : :Can you try applying the following two changes? : : rev 1.38 of vm_map.h : rev 1.154 of vm_map.c : :They may address one of the problems. : :Alan Nice to see the growstack stuff has been fixed! -Matt Matthew Dillon To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Mar 10 11:12: 1 1999 Delivered-To: freebsd-smp@freebsd.org Received: from panzer.plutotech.com (panzer.plutotech.com [206.168.67.125]) by hub.freebsd.org (Postfix) with ESMTP id 0B03E15407 for ; Wed, 10 Mar 1999 11:11:55 -0800 (PST) (envelope-from ken@panzer.plutotech.com) Received: (from ken@localhost) by panzer.plutotech.com (8.9.3/8.8.5) id MAA09630; Wed, 10 Mar 1999 12:10:45 -0700 (MST) From: "Kenneth D. Merry" Message-Id: <199903101910.MAA09630@panzer.plutotech.com> Subject: Re: lockmgr() panic In-Reply-To: <199903101850.KAA56970@apollo.backplane.com> from Matthew Dillon at "Mar 10, 1999 10:50:36 am" To: dillon@apollo.backplane.com (Matthew Dillon) Date: Wed, 10 Mar 1999 12:10:45 -0700 (MST) Cc: Thierry.Besancon@lps.ens.fr, luoqi@watermarkgroup.com, Thierry.Besancon@tournesol.lps.ens.fr, dwhite@gdi.uoregon.edu, freebsd-smp@FreeBSD.ORG, Pierre.David@prism.uvsq.fr, alc@cs.rice.edu, jt@ratp.fr X-Mailer: ELM [version 2.4ME+ PL43 (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 Matthew Dillon wrote... > > :>> This is a completely different problem. > : > :Life sucks today. > : > :Reading src/sys/kern/kern_lock.c, I saw this piece of code that might > :be relevant with this problem : > : > :#if !defined(MAX_PERF) > : if ((extflags & LK_CANRECURSE) == 0) > : panic("lockmgr: locking against myself"); > :#endif > : > :Before defining the flag, does anybody know what MAX_PERF implies > :other than not panicing the kernel ? What does it means ? > > MAX_PERF means 'maximum performance'. Note that the conditional > is an '#if !defined ...' conditional, not an '#if defined ...' > conditional. > > This means that by default, when MAX_PERF is *not* defined, the above > code *will* be executed. > > If MAX_PERF is defined, the system removes many sanity checks from the > kernel that would otherwise be compiled in. I would recommend that you > *NOT* define it. > > -Matt I'll second this. One of the few times I've really seriously hosed my filesystem came one time a couple of years ago (Summer '97) when I defined MAX_PERF. Even during the early CAM development, I never managed to hose my filesystem, but defining MAX_PERF was a really bad idea. I think that's one of those "John-Dyson-only" options. Ken -- Kenneth Merry ken@plutotech.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Mar 10 11:34:35 1999 Delivered-To: freebsd-smp@freebsd.org Received: from palrel1.hp.com (palrel1.hp.com [156.153.255.242]) by hub.freebsd.org (Postfix) with ESMTP id 5C38014C58 for ; Wed, 10 Mar 1999 11:34:32 -0800 (PST) (envelope-from darrylo@sr.hp.com) Received: from srmail.sr.hp.com (srmail.sr.hp.com [15.4.45.14]) by palrel1.hp.com (8.8.6/8.8.5tis) with ESMTP id LAA13428 for ; Wed, 10 Mar 1999 11:34:14 -0800 (PST) Received: from mina.sr.hp.com by srmail.sr.hp.com with ESMTP (1.37.109.16/15.5+ECS 3.3) id AA124544453; Wed, 10 Mar 1999 11:34:13 -0800 Received: from localhost (darrylo@mina.sr.hp.com [15.4.42.247]) by mina.sr.hp.com with ESMTP (8.8.6 (PHNE_14041)/8.7.3 TIS 5.0) id LAA13789 for ; Wed, 10 Mar 1999 11:34:12 -0800 (PST) Message-Id: <199903101934.LAA13789@mina.sr.hp.com> To: freebsd-smp@freebsd.org Subject: SMP benchmarks? Reply-To: Darryl Okahata Mime-Version: 1.0 (generated by tm-edit 1.1.1.1) Content-Type: text/plain; charset=US-ASCII Date: Wed, 10 Mar 1999 11:34:12 -0800 From: Darryl Okahata Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Does anyone know of any SMP benchmarks? In particular, I've been wondering how much of an effect a 66MHz FSB has, compared to a 100MHz one, and so I'm looking for something that runs on all CPUs and thrashes the caches, so that I can get an idea of the effects of memory contention and FSB speed (does it scale linearly with bus speed and number of CPUs?). (It might also be interesting to run, once processor affinity is implemented.) Thanks. -- Darryl Okahata darrylo@sr.hp.com DISCLAIMER: this message is the author's personal opinion and does not constitute the support, opinion, or policy of Hewlett-Packard, or of the little green men that have been following him all day. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Mar 10 12: 0:36 1999 Delivered-To: freebsd-smp@freebsd.org Received: from smtp04.primenet.com (smtp04.primenet.com [206.165.6.134]) by hub.freebsd.org (Postfix) with ESMTP id DE87A15157 for ; Wed, 10 Mar 1999 12:00:33 -0800 (PST) (envelope-from tlambert@usr04.primenet.com) Received: (from daemon@localhost) by smtp04.primenet.com (8.8.8/8.8.8) id NAA28717; Wed, 10 Mar 1999 13:41:04 -0700 (MST) Received: from usr04.primenet.com(206.165.6.204) via SMTP by smtp04.primenet.com, id smtpd028659; Wed Mar 10 13:40:58 1999 Received: (from tlambert@localhost) by usr04.primenet.com (8.8.5/8.8.5) id MAA08386; Wed, 10 Mar 1999 12:59:28 -0700 (MST) From: Terry Lambert Message-Id: <199903101959.MAA08386@usr04.primenet.com> Subject: Re: lockmgr() panic To: Thierry.Besancon@lps.ens.fr Date: Wed, 10 Mar 1999 19:59:28 +0000 (GMT) Cc: luoqi@watermarkgroup.com, Thierry.Besancon@tournesol.lps.ens.fr, dwhite@gdi.uoregon.edu, freebsd-smp@FreeBSD.ORG, Pierre.David@prism.uvsq.fr, alc@cs.rice.edu, dillon@apollo.backplane.com, jt@ratp.fr In-Reply-To: from "Thierry.Besancon@lps.ens.fr" at Mar 10, 99 03:45:24 pm X-Mailer: ELM [version 2.4 PL25] 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 > Life sucks today. > > Reading src/sys/kern/kern_lock.c, I saw this piece of code that might > be relevant with this problem : > > #if !defined(MAX_PERF) > if ((extflags & LK_CANRECURSE) == 0) > panic("lockmgr: locking against myself"); > #endif > > Before defining the flag, does anybody know what MAX_PERF implies > other than not panicing the kernel ? What does it means ? "Maximum Performance" -- Don't do tests that will never fail because we don't have any bugs. The result of defining MAX_PERF in this case will be a cascade failure later, at some point in the future, since there is, in fact, a bug. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Mar 10 22: 0:27 1999 Delivered-To: freebsd-smp@freebsd.org Received: from acetylene.vapornet.net (acetylene.vapornet.net [209.100.218.11]) by hub.freebsd.org (Postfix) with ESMTP id 9676D150F9 for ; Wed, 10 Mar 1999 22:00:17 -0800 (PST) (envelope-from john@vapornet.net) Received: from datapit.home.vapornet.net (vapornet.xnet.com. [205.243.141.107]) by acetylene.vapornet.net (8.9.1/8.9.1/VaporServer 1.4) with ESMTP id AAA01564 for ; Thu, 11 Mar 1999 00:00:10 -0600 (CST) (envelope from: john@vapornet.net) Received: from habanero.chili-pepper.net (habanero.chili-pepper.net [192.168.0.11]) by datapit.home.vapornet.net (8.9.2/8.9.1/VaporServer v1.3.1) with ESMTP id AAA26354 for ; Thu, 11 Mar 1999 00:00:11 -0600 (CST) (envelope from: john@vapornet.net) Received: (from john@localhost) by habanero.chili-pepper.net (8.9.2/8.9.1/VaporClient v3.1) id XAA77815; Wed, 10 Mar 1999 23:59:53 -0600 (CST) (envelope from: john@vapornet.net) From: John Preisler MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Date: Wed, 10 Mar 1999 23:59:53 -0600 (CST) To: freebsd-smp@freebsd.org Subject: celeron smp hack X-Mailer: VM 6.43 under 20.4 "Emerald" XEmacs Lucid Message-ID: <14055.23441.864135.866871@habanero.chili-pepper.net> Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org http://www.bxboards.com/dual.htm anyone brave enough to try such a hack? is it stable? -j To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Wed Mar 10 23:54:39 1999 Delivered-To: freebsd-smp@freebsd.org Received: from atlrel1.hp.com (atlrel1.hp.com [156.153.255.210]) by hub.freebsd.org (Postfix) with ESMTP id DC726152A0 for ; Wed, 10 Mar 1999 23:54:37 -0800 (PST) (envelope-from darrylo@sr.hp.com) Received: from srmail.sr.hp.com (srmail.sr.hp.com [15.4.45.14]) by atlrel1.hp.com (8.8.6 (PHNE_14041)/8.8.5tis) with ESMTP id CAA29279; Thu, 11 Mar 1999 02:54:09 -0500 (EST) Received: from mina.sr.hp.com by srmail.sr.hp.com with ESMTP (1.37.109.16/15.5+ECS 3.3) id AA287958748; Wed, 10 Mar 1999 23:52:28 -0800 Received: from localhost (darrylo@mina.sr.hp.com [15.4.42.247]) by mina.sr.hp.com with ESMTP (8.8.6 (PHNE_14041)/8.7.3 TIS 5.0) id XAA01513; Wed, 10 Mar 1999 23:52:28 -0800 (PST) Message-Id: <199903110752.XAA01513@mina.sr.hp.com> To: John Preisler Cc: freebsd-smp@FreeBSD.ORG Subject: Re: celeron smp hack Reply-To: Darryl Okahata In-Reply-To: Your message of "Wed, 10 Mar 1999 23:59:53 CST." <14055.23441.864135.866871@habanero.chili-pepper.net> Mime-Version: 1.0 (generated by tm-edit 1.1.1.1) Content-Type: text/plain; charset=US-ASCII Date: Wed, 10 Mar 1999 23:52:27 -0800 From: Darryl Okahata Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > http://www.bxboards.com/dual.htm > > anyone brave enough to try such a hack? is it stable? That's the hard way. For the easier way, see the various dual celeron articles at: http://www.cpu-central.com/ -- Darryl Okahata darrylo@sr.hp.com DISCLAIMER: this message is the author's personal opinion and does not constitute the support, opinion, or policy of Hewlett-Packard, or of the little green men that have been following him all day. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Mar 11 7: 7:40 1999 Delivered-To: freebsd-smp@freebsd.org Received: from world.virtual-earth.de (world.virtual-earth.de [194.231.209.38]) by hub.freebsd.org (Postfix) with ESMTP id 4A3EA153AA for ; Thu, 11 Mar 1999 07:07:33 -0800 (PST) (envelope-from mathiasp@newton.virtual-earth.de) Received: from newton.virtual-earth.de (newton.virtual-earth.de [194.231.209.37]) by world.virtual-earth.de (8.9.2/8.8.5) with ESMTP id QAA04568 for ; Thu, 11 Mar 1999 16:07:06 +0100 (MET) Message-Id: <199903111507.QAA04568@world.virtual-earth.de> Date: Thu, 11 Mar 1999 16:06:05 +0000 (GMT) From: Mathias Picker Subject: oas fail under smp, work without To: freebsd-smp@freebsd.org MIME-Version: 1.0 Content-Type: MULTIPART/signed; micalg=pgp-md5; protocol="application/pgp-signature"; BOUNDARY="0-1804289383-921168375=:3240" Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org --0-1804289383-921168375=:3240 Content-Type: TEXT/plain; CHARSET=US-ASCII I just installed Oracle/Linux and the oracle web application server on a brand new dual PII running -stable from 1999-03-09 on an asus p2b-ds motherboard with 512MB. With the help of Marcel Moolenaar and his infos and linux-base and -devel pkg at http://www.wi.leidenuniv.nl/~mhmoolen/ this was relatively easy. One problem remained for both of us: the server would not execute the oracle plsql cartridge, only giving back "not found" errors. Then Marcel tried running it without smp-support in the kernel and voila - it worked. The oracle web application server uses _lots_ of processes, and I guess some inter-process communication won't work in an smp enviroment. How can I start to debug this? Where to look? I'm not a kernel hacker, this is my first smp system, but I really want to try. Everything else seems to work in an smp-enviroment, connection to the database, remote connection via ip, the webserver as webserver, configuring the webserver via it's html administration interface, starting and stopping web listeners etc, only calling cartridges did not work. Thanks, Mathias -- Mathias Picker Consultant Information Architecture Mathias.Picker@virtual-earth.de +49 172 / 89 19 381 --0-1804289383-921168375=:3240 Content-Type: APPLICATION/pgp-signature -----BEGIN PGP MESSAGE----- Version: 2.6.3ia iQCVAwUBNufp8azvoSjVwVSdAQFAEQQAxWZXRYJGwhbXdFKPDpQIjN8+QuC8DTrt jb095RnWwGCs4o3/OzzPlTZkAtizr9ZKrPClsynflhZlK5CI7kRPAyv8NWnu5oB2 C9vRR6Ip4eFK1kcJ2E5mk+hCv+N34JNlMf3ex1Oh4UbQlpiyShRrU5ubYou93Ngo ghauwf4L2ZM= =ir7S -----END PGP MESSAGE----- --0-1804289383-921168375=:3240-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Mar 11 10:47:26 1999 Delivered-To: freebsd-smp@freebsd.org Received: from cs.rice.edu (cs.rice.edu [128.42.1.30]) by hub.freebsd.org (Postfix) with ESMTP id 546E915056 for ; Thu, 11 Mar 1999 10:47:07 -0800 (PST) (envelope-from alc@cs.rice.edu) Received: from nonpc.cs.rice.edu (nonpc.cs.rice.edu [128.42.1.219]) by cs.rice.edu (8.9.0/8.9.0) with ESMTP id MAA12863; Thu, 11 Mar 1999 12:46:49 -0600 (CST) Received: (from alc@localhost) by nonpc.cs.rice.edu (8.9.2/8.7.3) id MAA01635; Thu, 11 Mar 1999 12:46:48 -0600 (CST) Date: Thu, 11 Mar 1999 12:46:48 -0600 From: Alan Cox To: Mathias Picker Cc: freebsd-smp@freebsd.org Subject: Re: oas fail under smp, work without Message-ID: <19990311124648.D892@nonpc.cs.rice.edu> References: <199903111507.QAA04568@world.virtual-earth.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.1i In-Reply-To: <199903111507.QAA04568@world.virtual-earth.de>; from Mathias Picker on Thu, Mar 11, 1999 at 04:06:05PM +0000 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Try applying the patches under the very first bullet at http://www.freebsd.org/~luoqi/ If it's using linux threads, that could fix the problem. Alan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Mar 11 11:28:59 1999 Delivered-To: freebsd-smp@freebsd.org Received: from exchange-server.modacad.com (fw.modacad.com [207.199.66.93]) by hub.freebsd.org (Postfix) with ESMTP id 4791914BF7 for ; Thu, 11 Mar 1999 11:28:57 -0800 (PST) (envelope-from MattL@ModaCAD.com) Message-ID: From: Matt Liu To: "'smp@freebsd.org'" Subject: why maxusers may cause page-fault panic? Date: Thu, 11 Mar 1999 11:28:01 -0800 MIME-Version: 1.0 Content-Type: text/plain Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In older version of fbsd, you can set maxuser large but the newer version, I don't knwo from which version-3.0?, it's different. What is changed. A freely settable maxuser may make users happy-:). UNIX/NT ADM @ModaCad Inc. A Berkeley follower To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Mar 12 7:33: 7 1999 Delivered-To: freebsd-smp@freebsd.org Received: from gaia.euronet.nl (gaia.euronet.nl [194.134.0.10]) by hub.freebsd.org (Postfix) with ESMTP id 4B48B153DC for ; Fri, 12 Mar 1999 07:32:22 -0800 (PST) (envelope-from freebsd-smp@scc.nl) Received: from scones.sup.scc.nl (i130.ztm.euronet.nl [194.134.112.91]) by gaia.euronet.nl (8.8.8/8.8.8) with ESMTP id QAA04711 from for ; Fri, 12 Mar 1999 16:32:03 +0100 (MET) Received: (from daemon@localhost) by scones.sup.scc.nl (8.9.3/8.9.1) id QAA85976 for smp@FreeBSD.ORG; Fri, 12 Mar 1999 16:29:09 +0100 (CET) (envelope-from freebsd-smp@scc.nl) Received: from GATEWAY by scones.sup.scc.nl with netnews for smp@FreeBSD.ORG (smp@FreeBSD.ORG) To: smp@FreeBSD.ORG Date: Fri, 12 Mar 1999 16:29:05 +0100 From: Marcel Moolenaar Message-ID: <36E932C1.C6CB9DEE@scc.nl> Organization: SCC vof Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit References: <199903111507.QAA04568@world.virtual-earth.de>, <19990311124648.D892@nonpc.cs.rice.edu> Subject: Re: oas fail under smp, work without Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Alan Cox wrote: > > Try applying the patches under the very first bullet at > > http://www.freebsd.org/~luoqi/ > > If it's using linux threads, that could fix the problem. The exact problem is in the fact that OAS forks with some shared memory allocated. The message you get is: shared address space fork attempted: pid: %d (in file: /sys/kern/kern_fork.c) If someone is working on this particular problem and wants support/help or testers, speak up! I'm currently not in the position to attack this myself, but that doesn't mean I can't help :-) marcel To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Mar 12 8:56: 1 1999 Delivered-To: freebsd-smp@freebsd.org Received: from lor.watermarkgroup.com (lor.watermarkgroup.com [207.202.73.33]) by hub.freebsd.org (Postfix) with ESMTP id 2338C14D28 for ; Fri, 12 Mar 1999 08:55:59 -0800 (PST) (envelope-from luoqi@watermarkgroup.com) Received: (from luoqi@localhost) by lor.watermarkgroup.com (8.8.8/8.8.8) id LAA29937; Fri, 12 Mar 1999 11:55:08 -0500 (EST) (envelope-from luoqi) Date: Fri, 12 Mar 1999 11:55:08 -0500 (EST) From: Luoqi Chen Message-Id: <199903121655.LAA29937@lor.watermarkgroup.com> To: marcel@scc.nl, smp@FreeBSD.ORG Subject: Re: oas fail under smp, work without Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Alan Cox wrote: > > > > Try applying the patches under the very first bullet at > > > > http://www.freebsd.org/~luoqi/ > > > > If it's using linux threads, that could fix the problem. > > The exact problem is in the fact that OAS forks with some shared memory > allocated. The message you get is: > shared address space fork attempted: pid: %d > (in file: /sys/kern/kern_fork.c) > No, this message means the process tries to fork a child that shares its own address space (a linux clone() syscall), which is not possible without the aforementioned patch under SMP. It has nothing to do with (SYSV) shared memory. -lq To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Mar 12 9:59:17 1999 Delivered-To: freebsd-smp@freebsd.org Received: from alpo.whistle.com (alpo.whistle.com [207.76.204.38]) by hub.freebsd.org (Postfix) with ESMTP id F18C614C82 for ; Fri, 12 Mar 1999 09:58:59 -0800 (PST) (envelope-from julian@whistle.com) Received: (from daemon@localhost) by alpo.whistle.com (8.8.5/8.8.5) id JAA06573; Fri, 12 Mar 1999 09:55:58 -0800 (PST) Received: from current1.whistle.com(207.76.205.22) via SMTP by alpo.whistle.com, id smtpdEa6568; Fri Mar 12 17:55:53 1999 Date: Fri, 12 Mar 1999 09:55:48 -0800 (PST) From: Julian Elischer To: Luoqi Chen Cc: marcel@scc.nl, smp@FreeBSD.org Subject: Re: oas fail under smp, work without In-Reply-To: <199903121655.LAA29937@lor.watermarkgroup.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 Hi luoqi.. i think it's time to check it in.. :-) On Fri, 12 Mar 1999, Luoqi Chen wrote: > > Alan Cox wrote: > > > > > > Try applying the patches under the very first bullet at > > > > > > http://www.freebsd.org/~luoqi/ > > > > > > If it's using linux threads, that could fix the problem. > > > > The exact problem is in the fact that OAS forks with some shared memory > > allocated. The message you get is: > > shared address space fork attempted: pid: %d > > (in file: /sys/kern/kern_fork.c) > > > No, this message means the process tries to fork a child that shares its > own address space (a linux clone() syscall), which is not possible without > the aforementioned patch under SMP. It has nothing to do with (SYSV) shared > memory. > > -lq > > > 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 Mar 12 10: 4:41 1999 Delivered-To: freebsd-smp@freebsd.org Received: from gaia.euronet.nl (gaia.euronet.nl [194.134.0.10]) by hub.freebsd.org (Postfix) with ESMTP id 08FA5151F9 for ; Fri, 12 Mar 1999 10:04:37 -0800 (PST) (envelope-from marcel@scc.nl) Received: from scones.sup.scc.nl (i319.ztm.euronet.nl [194.134.67.80]) by gaia.euronet.nl (8.8.8/8.8.8) with ESMTP id TAA14197; Fri, 12 Mar 1999 19:04:18 +0100 (MET) Received: from scc.nl (scones.sup.scc.nl [192.168.2.4]) by scones.sup.scc.nl (8.9.3/8.9.1) with ESMTP id TAA25971; Fri, 12 Mar 1999 19:04:15 +0100 (CET) (envelope-from marcel@scc.nl) Message-ID: <36E9571F.BE5A5DE7@scc.nl> Date: Fri, 12 Mar 1999 19:04:15 +0100 From: Marcel Moolenaar Organization: SCC vof X-Mailer: Mozilla 4.5 [en] (X11; I; FreeBSD 4.0-CURRENT i386) X-Accept-Language: en MIME-Version: 1.0 To: Luoqi Chen Cc: smp@FreeBSD.ORG Subject: Re: oas fail under smp, work without References: <199903121655.LAA29937@lor.watermarkgroup.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Luoqi Chen wrote: > > The exact problem is in the fact that OAS forks with some shared memory > > allocated. The message you get is: > > shared address space fork attempted: pid: %d > > (in file: /sys/kern/kern_fork.c) > > > No, this message means the process tries to fork a child that shares its > own address space (a linux clone() syscall), which is not possible without > the aforementioned patch under SMP. It has nothing to do with (SYSV) shared > memory. Ah... that explains... thanks, marcel To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Mar 12 11: 4:50 1999 Delivered-To: freebsd-smp@freebsd.org Received: from gaia.euronet.nl (gaia.euronet.nl [194.134.0.10]) by hub.freebsd.org (Postfix) with ESMTP id E84D014CFE for ; Fri, 12 Mar 1999 11:04:45 -0800 (PST) (envelope-from marcel@scc.nl) Received: from scones.sup.scc.nl (i253.ztm.euronet.nl [194.134.67.54]) by gaia.euronet.nl (8.8.8/8.8.8) with ESMTP id UAA15263; Fri, 12 Mar 1999 20:04:26 +0100 (MET) Received: from scc.nl (scones.sup.scc.nl [192.168.2.4]) by scones.sup.scc.nl (8.9.3/8.9.1) with ESMTP id UAA02801; Fri, 12 Mar 1999 20:04:20 +0100 (CET) (envelope-from marcel@scc.nl) Message-ID: <36E96534.D5068F8E@scc.nl> Date: Fri, 12 Mar 1999 20:04:20 +0100 From: Marcel Moolenaar Organization: SCC vof X-Mailer: Mozilla 4.5 [en] (X11; I; FreeBSD 4.0-CURRENT i386) X-Accept-Language: en MIME-Version: 1.0 To: Julian Elischer Cc: Luoqi Chen , smp@FreeBSD.org Subject: Re: oas fail under smp, work without References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Julian Elischer wrote: > > Hi luoqi.. > i think it's time to check it in.. :-) After applying luoqi's patches on a -current machine, I get the following: scones# /compat/linux/bin/bash bash: Out of virtual memory! I'll get back to you when I know more; in the mean time, do not commit :-) marcel To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Mar 12 12:36:26 1999 Delivered-To: freebsd-smp@freebsd.org Received: from gaia.euronet.nl (gaia.euronet.nl [194.134.0.10]) by hub.freebsd.org (Postfix) with ESMTP id 313AC15316 for ; Fri, 12 Mar 1999 12:35:55 -0800 (PST) (envelope-from marcel@scc.nl) Received: from scones.sup.scc.nl (i119.ztm.euronet.nl [194.134.112.80]) by gaia.euronet.nl (8.8.8/8.8.8) with ESMTP id VAA01604; Fri, 12 Mar 1999 21:35:20 +0100 (MET) Received: from scc.nl (scones.sup.scc.nl [192.168.2.4]) by scones.sup.scc.nl (8.9.3/8.9.1) with ESMTP id VAA17362; Fri, 12 Mar 1999 21:35:12 +0100 (CET) (envelope-from marcel@scc.nl) Message-ID: <36E97A80.8170F280@scc.nl> Date: Fri, 12 Mar 1999 21:35:12 +0100 From: Marcel Moolenaar Organization: SCC vof X-Mailer: Mozilla 4.5 [en] (X11; I; FreeBSD 4.0-CURRENT i386) X-Accept-Language: en MIME-Version: 1.0 To: Luoqi Chen Cc: smp@FreeBSD.ORG Subject: SMP vmspace sharing [was: Re: oas fail under smp, work without] References: <199903121655.LAA29937@lor.watermarkgroup.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, Applying luoqi's patches results in a different behavior for the linux_brk syscall. This breaks Linux bash. a -current system without the patches produces the following ktrace: ... 555 bash CALL getegid 555 bash RET getegid 1000/0x3e8 555 bash CALL linux_brk(0) 555 bash RET linux_brk 134873088/0x80a0000 555 bash CALL linux_brk(0x80a0030) 555 bash RET linux_brk 134873136/0x80a0030 555 bash CALL linux_brk(0x80a1000) 555 bash RET linux_brk 134877184/0x80a1000 555 bash CALL linux_time(0) 555 bash RET linux_time 921267035/0x36e96b5b ... After applying the patches, the following ktrace is produced: ... 2993 bash CALL getegid 2993 bash RET getegid 0 2993 bash CALL linux_brk(0) 2993 bash RET linux_brk -256558252/0xf0b53b54 2993 bash CALL linux_brk(0xf0b53b84) 2993 bash RET linux_brk -256558252/0xf0b53b54 2993 bash CALL write(0x2,0xefbfb1ac,0x6) 2993 bash GIO fd 2 wrote 6 bytes "bash: " 2993 bash RET write 6 2993 bash CALL write(0x2,0xefbfb1c4,0x16) 2993 bash GIO fd 2 wrote 22 bytes "Out of virtual memory!" 2993 bash RET write 22/0x16 ... marcel To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Mar 12 14:48:53 1999 Delivered-To: freebsd-smp@freebsd.org Received: from lor.watermarkgroup.com (lor.watermarkgroup.com [207.202.73.33]) by hub.freebsd.org (Postfix) with ESMTP id 67AD814D75 for ; Fri, 12 Mar 1999 14:48:50 -0800 (PST) (envelope-from luoqi@watermarkgroup.com) Received: (from luoqi@localhost) by lor.watermarkgroup.com (8.8.8/8.8.8) id RAA02336; Fri, 12 Mar 1999 17:47:59 -0500 (EST) (envelope-from luoqi) Date: Fri, 12 Mar 1999 17:47:59 -0500 (EST) From: Luoqi Chen Message-Id: <199903122247.RAA02336@lor.watermarkgroup.com> To: marcel@scc.nl Subject: Re: SMP vmspace sharing [was: Re: oas fail under smp, work without] Cc: smp@FreeBSD.ORG Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Did you recompile linux module? brk can't be at 0xf0b53b54, that's in kernel address space. I can run linux bash just fine, but truss output didn't show any brk(0) call, so I may be using a different version of bash (the one comes with linux-lib 2.6 port). -lq > > Applying luoqi's patches results in a different behavior for the linux_brk > syscall. This breaks Linux bash. > > a -current system without the patches produces the following ktrace: > > ... > 555 bash CALL getegid > 555 bash RET getegid 1000/0x3e8 > 555 bash CALL linux_brk(0) > 555 bash RET linux_brk 134873088/0x80a0000 > 555 bash CALL linux_brk(0x80a0030) > 555 bash RET linux_brk 134873136/0x80a0030 > 555 bash CALL linux_brk(0x80a1000) > 555 bash RET linux_brk 134877184/0x80a1000 > 555 bash CALL linux_time(0) > 555 bash RET linux_time 921267035/0x36e96b5b > ... > > After applying the patches, the following ktrace is produced: > > ... > 2993 bash CALL getegid > 2993 bash RET getegid 0 > 2993 bash CALL linux_brk(0) > 2993 bash RET linux_brk -256558252/0xf0b53b54 > 2993 bash CALL linux_brk(0xf0b53b84) > 2993 bash RET linux_brk -256558252/0xf0b53b54 > 2993 bash CALL write(0x2,0xefbfb1ac,0x6) > 2993 bash GIO fd 2 wrote 6 bytes > "bash: " > 2993 bash RET write 6 > 2993 bash CALL write(0x2,0xefbfb1c4,0x16) > 2993 bash GIO fd 2 wrote 22 bytes > "Out of virtual memory!" > 2993 bash RET write 22/0x16 > ... > > marcel > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Mar 13 1:35:13 1999 Delivered-To: freebsd-smp@freebsd.org Received: from iquest3.iquest.net (iquest3.iquest.net [209.43.20.203]) by hub.freebsd.org (Postfix) with SMTP id 940D014E62 for ; Sat, 13 Mar 1999 01:35:11 -0800 (PST) (envelope-from toor@y.dyson.net) Received: (qmail 7274 invoked from network); 13 Mar 1999 09:34:51 -0000 Received: from dyson.iquest.net (HELO y.dyson.net) (198.70.144.127) by iquest3.iquest.net with SMTP; 13 Mar 1999 09:34:51 -0000 Received: (from toor@localhost) by y.dyson.net (8.9.3/8.9.1) id EAA01281; Sat, 13 Mar 1999 04:34:45 -0500 (EST) Message-Id: <199903130934.EAA01281@y.dyson.net> Subject: Re: lockmgr() panic In-Reply-To: <199903101910.MAA09630@panzer.plutotech.com> from "Kenneth D. Merry" at "Mar 10, 99 12:10:45 pm" To: ken@plutotech.com (Kenneth D. Merry) Date: Sat, 13 Mar 1999 04:34:45 -0500 (EST) Cc: dillon@apollo.backplane.com, Thierry.Besancon@lps.ens.fr, luoqi@watermarkgroup.com, Thierry.Besancon@tournesol.lps.ens.fr, dwhite@gdi.uoregon.edu, freebsd-smp@FreeBSD.ORG, Pierre.David@prism.uvsq.fr, alc@cs.rice.edu, jt@ratp.fr From: "John S. Dyson" Reply-To: dyson@iquest.net X-Mailer: ELM [version 2.4ME+ PL38 (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 Kenneth D. Merry said: > > I think that's one of those "John-Dyson-only" options. > It was mostly for investigating the cost of consistancy checks. If the cost is insignificant, then remove the #ifdef, and enable the code all of the time. Note that adding lots of cruft with wordy asserts adds to bloat. On X86 machines, registers are expensive, and subroutine calls really mess things up. IMO, you cannot check everything every time. Sometimes, I added asserts that were silly, and they should be carefully placed, and not put willy-nilly all over the code. Lots of asserts make code almost as unreadable as lots of '#ifdefs' and silly macros. Not only that, lots of asserts unconditionally enabled that don't diagnose a carefully considered error syndrome just take up memory unnecessarily, and often don't really help debugging. Being able to remove asserts and debugging stuff (like DIAGNOSTIC) increases flexibility. Like any option, making silly decisions makes silly results. -- John | Never try to teach a pig to sing, dyson@iquest.net | it makes one look stupid jdyson@nc.com | and it irritates the pig. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Mar 13 7:48:21 1999 Delivered-To: freebsd-smp@freebsd.org Received: from gaia.euronet.nl (gaia.euronet.nl [194.134.0.10]) by hub.freebsd.org (Postfix) with ESMTP id 5686A14D5A for ; Sat, 13 Mar 1999 07:48:18 -0800 (PST) (envelope-from marcel@scc.nl) Received: from scones.sup.scc.nl (i249.ztm.euronet.nl [194.134.67.50]) by gaia.euronet.nl (8.8.8/8.8.8) with ESMTP id QAA18790; Sat, 13 Mar 1999 16:47:59 +0100 (MET) Received: from scc.nl (scones.sup.scc.nl [192.168.2.4]) by scones.sup.scc.nl (8.9.3/8.9.1) with ESMTP id QAA19948; Sat, 13 Mar 1999 16:47:54 +0100 (CET) (envelope-from marcel@scc.nl) Message-ID: <36EA88AA.4892D69D@scc.nl> Date: Sat, 13 Mar 1999 16:47:54 +0100 From: Marcel Moolenaar Organization: SCC vof X-Mailer: Mozilla 4.5 [en] (X11; I; FreeBSD 4.0-CURRENT i386) X-Accept-Language: en MIME-Version: 1.0 To: Luoqi Chen Cc: smp@FreeBSD.ORG Subject: Re: SMP vmspace sharing [was: Re: oas fail under smp, work without] References: <199903122247.RAA02336@lor.watermarkgroup.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Luoqi Chen wrote: > Did you recompile linux module? ...and how I wish you didn't asked that question... this is embarrassing (and also because a pointy-hat looks stupid :-) Yes, recompiling the linux module solved it. Thanks and sorry, marcel To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sat Mar 13 14:51:42 1999 Delivered-To: freebsd-smp@freebsd.org Received: from luke.pmr.com (luke.pmr.com [207.170.114.132]) by hub.freebsd.org (Postfix) with ESMTP id A00C014D4A for ; Sat, 13 Mar 1999 14:51:37 -0800 (PST) (envelope-from bob@luke.pmr.com) Received: (from bob@localhost) by luke.pmr.com (8.9.2/8.9.2) id QAA41505; Sat, 13 Mar 1999 16:49:51 -0600 (CST) (envelope-from bob) Date: Sat, 13 Mar 1999 16:49:51 -0600 From: Bob Willcox To: Darryl Okahata Cc: John Preisler , freebsd-smp@FreeBSD.org Subject: Re: celeron smp hack Message-ID: <19990313164951.A41414@luke.pmr.com> Reply-To: Bob Willcox References: <14055.23441.864135.866871@habanero.chili-pepper.net> <199903110752.XAA01513@mina.sr.hp.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.1i In-Reply-To: <199903110752.XAA01513@mina.sr.hp.com>; from Darryl Okahata on Wed, Mar 10, 1999 at 11:52:27PM -0800 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Wed, Mar 10, 1999 at 11:52:27PM -0800, Darryl Okahata wrote: > > http://www.bxboards.com/dual.htm > > > > anyone brave enough to try such a hack? is it stable? My son and I did this with one of his systems. We spent about 6 hours on the first processor and 2 on the second making the modifications (it really helped to know what we were doing). So far, his system has been stable at 375MHz but not at 450MHz. > > That's the hard way. For the easier way, see the various dual > celeron articles at: > > http://www.cpu-central.com/ Our preference would have been to do it the easier way (with the MSI converter boards), but he was unable to locate any within the timeframe that we would have needed them so we modified the slot 1 processor boards. I've also heard that there are converter boards available with jumpers to effect the change, though I've never seen them or anyplace claiming to have them so they may not exist. Doing the mods to the processor boards wasn't too bad but it is fairly tedious and requires pretty good soldering skills. Also, had I not had a stereo microscope I doubt that I would have been successful (the wires and land patterns on the boards are pretty small). Bob > > -- > Darryl Okahata > darrylo@sr.hp.com > > DISCLAIMER: this message is the author's personal opinion and does not > constitute the support, opinion, or policy of Hewlett-Packard, or of the > little green men that have been following him all day. > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-smp" in the body of the message -- Bob Willcox The man who follows the crowd will usually get no bob@luke.pmr.com further than the crowd. The man who walks alone is Austin, TX likely to find himself in places no one has ever been. -- Alan Ashley-Pitt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message