From owner-freebsd-smp Sun Aug 29 6:36:39 1999 Delivered-To: freebsd-smp@freebsd.org Received: from thehousleys.net (frenchknot.ne.mediaone.net [24.218.96.75]) by hub.freebsd.org (Postfix) with ESMTP id DEDD414BE6 for ; Sun, 29 Aug 1999 06:36:32 -0700 (PDT) (envelope-from jim@thehousleys.net) Received: from thehousleys.net (housley@localhost [127.0.0.1]) by thehousleys.net (8.9.3/8.9.3) with ESMTP id JAA86485; Sun, 29 Aug 1999 09:33:07 -0400 (EDT) (envelope-from jim@thehousleys.net) Message-ID: <37C93692.EB93AC98@thehousleys.net> Date: Sun, 29 Aug 1999 09:33:07 -0400 From: "James E. Housley" X-Mailer: Mozilla 4.51 [en] (X11; U; FreeBSD 3.2-STABLE i386) X-Accept-Language: en MIME-Version: 1.0 To: Suresh Rajagopalan Cc: freebsd-smp@FreeBSD.ORG Subject: Re: SMP freezes on 3.2-STABLE References: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Suresh Rajagopalan wrote: > ... > The machines are very busy web servers and simply freeze randomly after > 2-5 days of uptime. There is no response from the console at > the point of freeze. I cannot get it to dump core either. > > Machines are PIII-450 on ASUS P2B-D motherboards with 256Mb of RAM & 800M > swap. All machines are NFS clients mounting off a Solaris X86 server. > The output of mptable is attached below. I've tried MP 1.4 as well as > 1.1. > > I don't know if this is a problem with 3.x in general, and I have read > about larger sites like yahoo running smp-stable. If some of you are > running SMP in a busy web/nfs enviroment, I'd love to hear from you. > The problem is probably NFS. There has been lots of discussion on -hackers IIRC about heavy NFS traffic over the last three months. You might try checking the archives. Jim -- James E. Housley PGP: 1024/03983B4D System Supply, Inc. 2C 3F 3A 0D A8 D8 C3 13 Pager: pagejim@notepage.com 7C F0 B5 BF 27 8B 92 FE "The box said 'Requires Windows 95, NT, or better,' so I installed FreeBSD" To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Aug 29 9:16:41 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 AB4D6156F6 for ; Sun, 29 Aug 1999 09:16:39 -0700 (PDT) (envelope-from dillon@apollo.backplane.com) Received: (from dillon@localhost) by apollo.backplane.com (8.9.3/8.9.1) id JAA06121; Sun, 29 Aug 1999 09:14:01 -0700 (PDT) (envelope-from dillon) Date: Sun, 29 Aug 1999 09:14:01 -0700 (PDT) From: Matthew Dillon Message-Id: <199908291614.JAA06121@apollo.backplane.com> To: Suresh Rajagopalan Cc: freebsd-smp@FreeBSD.ORG Subject: Re: SMP freezes on 3.2-STABLE References: Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org :I've been trying to track down continual freezes on 3 servers running :3.2-STABLE. : :The machines are very busy web servers and simply freeze randomly after :2-5 days of uptime. There is no response from the console at :the point of freeze. I cannot get it to dump core either. : :Machines are PIII-450 on ASUS P2B-D motherboards with 256Mb of RAM & 800M :swap. All machines are NFS clients mounting off a Solaris X86 server. :The output of mptable is attached below. I've tried MP 1.4 as well as :1.1. : :I don't know if this is a problem with 3.x in general, and I have read :about larger sites like yahoo running smp-stable. If some of you are :running SMP in a busy web/nfs enviroment, I'd love to hear from you. : :Any help is appreciated. : :-Suresh Is this 3.x-STABLE or 3.2-RELEASE? Did you compile up a kernel with DDB support configured? (i.e. to be able to ctl-alt-esc into ddb from the console)? None of the currently known NFS related bugs freeze up the machine to the point where ctl-alt-esc fails to break into DDB. I would recommend recompiling the kernel for one of the machines to the absolute latest -STABLE (making sure DDB support has been configured), and the other one to the absolute latest -STABLE, with DDB, but without SMP, then see if none, one, or both machines crash. If when a machine freezes up you are able to escape into DDB, run the 'trace' command and the 'ps' command prio to 'panic'ing the machine. Also, post the 'dmesg' output from the boot (cat /var/run/dmesg.boot). -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 Sun Aug 29 11: 3:55 1999 Delivered-To: freebsd-smp@freebsd.org Received: from tahoe.cinenet.net (ns1.cinenet.net [198.147.76.65]) by hub.freebsd.org (Postfix) with ESMTP id 140EC14BE4 for ; Sun, 29 Aug 1999 11:03:53 -0700 (PDT) (envelope-from sraja@cinenet.net) Received: from hermosa.cinenet.net (hermosa.cinenet.net [198.147.76.90]) by tahoe.cinenet.net (8.8.6/8.8.6) with SMTP id LAA18122; Sun, 29 Aug 1999 11:03:50 -0700 (PDT) Date: Sun, 29 Aug 1999 11:03:48 -0700 (PDT) From: Suresh Rajagopalan To: Matthew Dillon Cc: freebsd-smp@FreeBSD.ORG Subject: Re: SMP freezes on 3.2-STABLE In-Reply-To: <199908291614.JAA06121@apollo.backplane.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 > Is this 3.x-STABLE or 3.2-RELEASE? It is 3.2-STABLE. > > Did you compile up a kernel with DDB support configured? (i.e. to be > able to ctl-alt-esc into ddb from the console)? Yes, I did compile with DDB support. But I tried the break key, not the ctl-alt-esc sequence. I will attempt that. I am running one of the machines w/o SMP support. It seems to be fine. Thanks Suresh To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Sun Aug 29 15:56: 6 1999 Delivered-To: freebsd-smp@freebsd.org Received: from office.omc.net (office.omc.net [195.185.142.22]) by hub.freebsd.org (Postfix) with ESMTP id 147D91507B for ; Sun, 29 Aug 1999 15:55:51 -0700 (PDT) (envelope-from LutzRab@omc.net) Received: from lutz (lutz.omc.net [195.185.142.3]) by office.omc.net (8.9.3/8.9.3) with ESMTP id AAA09980 for ; Mon, 30 Aug 1999 00:55:48 +0200 (CEST) Message-Id: <199908292255.AAA09980@office.omc.net> From: "Lutz Rabing" Organization: OMCnet IS GmbH To: freebsd-smp@FreeBSD.ORG Date: Mon, 30 Aug 1999 00:55:39 +0200 MIME-Version: 1.0 Content-type: text/plain; charset=US-ASCII Content-transfer-encoding: 7BIT Subject: SMP broken in 3.2-STABLE? Reply-To: LutzRab@omc.net X-mailer: Pegasus Mail for Win32 (v3.12a) Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I just CVSUPed to the latest 3.2-STABLE, "make world" and recompiled the kernel. After rebooting the system (ASUS P2B-DS, 2* PIII-450) I get core dumps from csh, sendmail, ntpdate, cron ... Most of the tools (sh, top, gcc ...) work. When I rebuild the kernel without SMP everything works fine. The system worked fine in SMP-mode with 3.2-STABLE (Aug.21). Any ideas? Thanks, Lutz Rabing -OMCnet- Mit freundlichen Gruessen, Lutz Rabing -OMCnet- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Aug 30 0: 6: 4 1999 Delivered-To: freebsd-smp@freebsd.org Received: from dingo.cdrom.com (castles523.castles.com [208.214.165.87]) by hub.freebsd.org (Postfix) with ESMTP id 9BB0B14EDC for ; Mon, 30 Aug 1999 00:05:58 -0700 (PDT) (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 XAA17927; Sun, 29 Aug 1999 23:59:11 -0700 (PDT) (envelope-from mike@dingo.cdrom.com) Message-Id: <199908300659.XAA17927@dingo.cdrom.com> X-Mailer: exmh version 2.0.2 2/24/98 To: Suresh Rajagopalan Cc: freebsd-smp@freebsd.org Subject: Re: SMP freezes on 3.2-STABLE In-reply-to: Your message of "Sat, 28 Aug 1999 23:55:37 PDT." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 29 Aug 1999 23:59:11 -0700 From: Mike Smith Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > I've been trying to track down continual freezes on 3 servers running > 3.2-STABLE. > > The machines are very busy web servers and simply freeze randomly after > 2-5 days of uptime. There is no response from the console at > the point of freeze. I cannot get it to dump core either. This makes it very hard to debug. You might want to try generating a hardware NMI by shorting the relevant pins on the ISA bus (look for postings from Rod Grimes about using a paperclip for this, or use a POST card). > I don't know if this is a problem with 3.x in general, and I have read > about larger sites like yahoo running smp-stable. If some of you are > running SMP in a busy web/nfs enviroment, I'd love to hear from you. This is not a typical symptom for problems, but there are a few ways that your systems could become so tied up they won't respond to the keyboard to pull them into the debugger. Typically these issues are related to hardware interaction, and it's usually hardware doing something unexpected that's the problem. We'll need more input before we can offer you any real help, I'm sorry. -- \\ The mind's the standard \\ Mike Smith \\ of the man. \\ msmith@freebsd.org \\ -- Joseph Merrick \\ 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 Aug 30 10:43:18 1999 Delivered-To: freebsd-smp@freebsd.org Received: from tahoe.cinenet.net (ns1.cinenet.net [198.147.76.65]) by hub.freebsd.org (Postfix) with ESMTP id A95AE1515F for ; Mon, 30 Aug 1999 10:43:13 -0700 (PDT) (envelope-from sraja@cinenet.net) Received: from hermosa.cinenet.net (hermosa.cinenet.net [198.147.76.90]) by tahoe.cinenet.net (8.8.6/8.8.6) with SMTP id KAA29948; Mon, 30 Aug 1999 10:43:05 -0700 (PDT) Date: Mon, 30 Aug 1999 10:43:03 -0700 (PDT) From: Suresh Rajagopalan To: Matthew Dillon Cc: freebsd-smp@FreeBSD.ORG Subject: Re: SMP freezes on 3.2-STABLE In-Reply-To: <199908291614.JAA06121@apollo.backplane.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 > If when a machine freezes up you are able to escape into DDB, run the > 'trace' command and the 'ps' command prio to 'panic'ing the machine. > Hi: a) This time I was able to break into DDB. Here's the output (BTW, I had to transcribe this, is there any way to dump the output to a file?) ----------------------- panic rslock cpu #0 addr 0xcc848cb0 lock=0x00000001 mp_lock=0000000x, cpuid=0, lapic.id=010000000 Debugger panic, stopped at debugger 0x37, movl, $0_in debugger ---- b) trace shows the following: panic (c0241d25,0,cc848cb0,1,c01769d4) at panic 0xa4 bsl1 (cc848c40, 2, cbbe6860, cbe3043f, cc848d00) at bs1 nfs_lookup (cbf3de30, cbafce00, cbf3dedc, cbf3deb80) at nfs_lookup 0x22f lookup(cbf3deb8,cbbe6860,cbbe6860,cbf3df94,cbf3de74) at lookup 0x2c1 namei(cbf3deb8,cbbe6860,c0292740,0,8162ce4) at namei 0x133 stat(cbbe6860,cbf3df94,13,5,bfbfdd50) at stat 0x44 syscall(27,bfbf0027,bfbfdd50,5,bfbfdb28) at syscall 0x107 Xint0x80_syscall() at Xint)x80_syscall + 0x4c --- c) ps shows a few http and perl processes, some in nfsrecv state. The output was too long to copy by hand. Hope this helps. -- d) On one of the machines, I now also see this message in the syslog: xl0: command never completed (at random intervals) e) Also here is the output of /var/run/dmesg.boot Copyright (c) 1992-1999 FreeBSD Inc. Copyright (c) 1982, 1986, 1989, 1991, 1993 The Regents of the University of California. All rights reserved. FreeBSD 3.2-STABLE #0: Thu Aug 26 10:40:37 PDT 1999 root@web2.mediasc.com:/usr/src/sys/compile/STABLE1 Timecounter "i8254" frequency 1193182 Hz CPU: Pentium III (686-class CPU) Origin = "GenuineIntel" Id = 0x672 Stepping = 2 Features=0x387fbff,MMX,FXSR,> real memory = 268435456 (262144K bytes) avail memory = 257658880 (251620K 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 0xc0314000. Probing for devices on PCI bus 0: chip0: rev 0x03 on pci0.0.0 chip1: rev 0x03 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 xl0: <3Com 3c905B-TX Fast Etherlink XL> rev 0x30 int a irq 18 on pci0.10.0 xl0: Ethernet address: 00:50:04:7d:08:1b xl0: autoneg complete, link status good (half-duplex, 10Mbps) Probing for devices on PCI bus 1: vga0: rev 0x7a 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 sio1 at 0x2f8-0x2ff irq 3 on isa sio1: 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: 8011MB (16408224 sectors), 16278 cyls, 16 heads, 63 S/T, 512 B/S wdc0: unit 1 (atapi): , removable, intr, dma, iordis acd0: drive speed 0 - 5500KB/sec, 128KB cache acd0: supported read types: CD-R, CD-RW, CD-DA, packet track acd0: Audio: play, 255 volume levels acd0: Mechanism: ejectable tray acd0: Medium: no/blank disc inside, unlocked wdc1 not found at 0x170 wt0 not found at 0x300 mcd0 not found at 0x300 scd0 not found at 0x230 ppc0 at 0x378 irq 7 on isa ppc0: Generic chipset (NIBBLE-only) in COMPATIBLE mode ppi0: on ppbus 0 plip0: on ppbus 0 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 changing root device to wd0s1a SMP: AP CPU #1 Launched! WARNING: / was not properly dismounted Thanks for your help. Suresh To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Aug 30 10:58:55 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 4F8D4152D2 for ; Mon, 30 Aug 1999 10:58:53 -0700 (PDT) (envelope-from dillon@apollo.backplane.com) Received: (from dillon@localhost) by apollo.backplane.com (8.9.3/8.9.1) id KAA16067; Mon, 30 Aug 1999 10:58:48 -0700 (PDT) (envelope-from dillon) Date: Mon, 30 Aug 1999 10:58:48 -0700 (PDT) From: Matthew Dillon Message-Id: <199908301758.KAA16067@apollo.backplane.com> To: Suresh Rajagopalan Cc: freebsd-smp@FreeBSD.ORG Subject: Re: SMP freezes on 3.2-STABLE References: Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org :a) : :This time I was able to break into DDB. Here's the output (BTW, I had to :transcribe this, is there any way to dump the output to a file?) No, you have to transcribe it or set the console up on a serial port (which can be annoying) and log it. :b) :trace shows the following: : :panic (c0241d25,0,cc848cb0,1,c01769d4) at panic 0xa4 :bsl1 (cc848c40, 2, cbbe6860, cbe3043f, cc848d00) at bs1 :nfs_lookup (cbf3de30, cbafce00, cbf3dedc, cbf3deb80) at nfs_lookup 0x22f :lookup(cbf3deb8,cbbe6860,cbbe6860,cbf3df94,cbf3de74) at lookup 0x2c1 :namei(cbf3deb8,cbbe6860,c0292740,0,8162ce4) at namei 0x133 :stat(cbbe6860,cbf3df94,13,5,bfbfdd50) at stat 0x44 :syscall(27,bfbf0027,bfbfdd50,5,bfbfdb28) at syscall 0x107 :Xint0x80_syscall() at Xint)x80_syscall + 0x4c : :--- :c) :ps shows a few http and perl processes, some in nfsrecv state. The :output was too long to copy by hand. Hope this helps. : :-- :d) : :On one of the machines, I now also see this message in the syslog: : :xl0: command never completed : :(at random intervals) :... :Thanks for your help. : :Suresh Hmm. This is very odd. It seems unlikely to be a bug in NFS if turning off SMP fixes the problem. If the machine didn't crash and ctl-alt-esc breaking it shows an active stack frame from a running process, it could be that the process is getting stuck in an endless loop somewhere or somehow. I'm at a loss at the moment. It sounds like an SMP problem of some sort. I'm hoping one of the SMP guys will have a brainfart :-) If it cannot be resolved soon I would back both machines down to a single-cpu and wait... having production machines crash is not fun. Current ought to work better for SMP then stable, but I hesitate to suggest it for a production machine. -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 Mon Aug 30 11: 5:49 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 2E996158F0 for ; Mon, 30 Aug 1999 11:05:32 -0700 (PDT) (envelope-from julian@whistle.com) Received: from current1.whistle.com (current1.whistle.com [207.76.205.22]) by alpo.whistle.com (8.9.1a/8.9.1) with SMTP id LAA63615; Mon, 30 Aug 1999 11:04:38 -0700 (PDT) Date: Mon, 30 Aug 1999 11:05:20 -0700 (PDT) From: Julian Elischer To: Matthew Dillon Cc: Suresh Rajagopalan , freebsd-smp@FreeBSD.ORG Subject: Re: SMP freezes on 3.2-STABLE In-Reply-To: <199908301758.KAA16067@apollo.backplane.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 Mon, 30 Aug 1999, Matthew Dillon wrote: > :a) > : > :This time I was able to break into DDB. Here's the output (BTW, I had to > :transcribe this, is there any way to dump the output to a file?) > > No, you have to transcribe it or set the console up on a serial port > (which can be annoying) and log it. The best way is to set flags 0x20 on your sio0, and then hook it to sio1 on another machine. (and do same from the other machine.) then use script to log the activity on it when you 'tip' into it. julian > > :b) > :trace shows the following: > : > :panic (c0241d25,0,cc848cb0,1,c01769d4) at panic 0xa4 > :bsl1 (cc848c40, 2, cbbe6860, cbe3043f, cc848d00) at bs1 > :nfs_lookup (cbf3de30, cbafce00, cbf3dedc, cbf3deb80) at nfs_lookup 0x22f > :lookup(cbf3deb8,cbbe6860,cbbe6860,cbf3df94,cbf3de74) at lookup 0x2c1 > :namei(cbf3deb8,cbbe6860,c0292740,0,8162ce4) at namei 0x133 > :stat(cbbe6860,cbf3df94,13,5,bfbfdd50) at stat 0x44 > :syscall(27,bfbf0027,bfbfdd50,5,bfbfdb28) at syscall 0x107 > :Xint0x80_syscall() at Xint)x80_syscall + 0x4c > : > :--- > :c) > :ps shows a few http and perl processes, some in nfsrecv state. The > :output was too long to copy by hand. Hope this helps. > : > :-- > :d) > : > :On one of the machines, I now also see this message in the syslog: > : > :xl0: command never completed > : > :(at random intervals) > :... > :Thanks for your help. > : > :Suresh > > Hmm. This is very odd. It seems unlikely to be a bug in NFS if > turning off SMP fixes the problem. > > If the machine didn't crash and ctl-alt-esc breaking it shows an > active stack frame from a running process, it could be that the > process is getting stuck in an endless loop somewhere or somehow. > > I'm at a loss at the moment. It sounds like an SMP problem of some > sort. I'm hoping one of the SMP guys will have a brainfart :-) > > If it cannot be resolved soon I would back both machines down to > a single-cpu and wait... having production machines crash is not fun. > > Current ought to work better for SMP then stable, but I hesitate to > suggest it for a production machine. > > -Matt > Matthew Dillon > > > > 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 Mon Aug 30 11: 6: 7 1999 Delivered-To: freebsd-smp@freebsd.org Received: from isbalham.ist.co.uk (isbalham.ist.co.uk [192.31.26.1]) by hub.freebsd.org (Postfix) with ESMTP id 3681415951; Mon, 30 Aug 1999 11:05:48 -0700 (PDT) (envelope-from rb@gid.co.uk) Received: from gid.co.uk (uucp@localhost) by isbalham.ist.co.uk (8.9.2/8.8.7) with UUCP id TAA56037; Mon, 30 Aug 1999 19:05:46 +0100 (BST) (envelope-from rb@gid.co.uk) Received: from [194.32.164.2] by seagoon.gid.co.uk; Mon, 30 Aug 1999 19:02:40 +0100 (BST) X-Sender: rb@194.32.164.1 Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Mon, 30 Aug 1999 19:02:38 +0000 To: current@freebsd.org From: Bob Bishop Subject: NFS-related panic on SMP Cc: smp@freebsd.org Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, With a kernel built from -current cvsup'd at Sun Aug 22 04:03:19 BST 1999, I got the following panic doing make -j8 world with /usr/src via NFS and /usr/obj local: Panic: free vnode isn't mp_lock = 00000001; cpuid = 0; lapic.id = 00000000 backtrace (hand-transcribed): panic getnewvnode() + 0x157 nfs_nget() + 0x107 nfs_lookup()+0xe8d lookup() + 0x2ab namei() + 0x137 stat() + 0x44 syscall() + 0x186 Xint0x80_syscall() + 0x31 Any ideas? -- Bob Bishop (0118) 977 4017 international code +44 118 rb@gid.co.uk fax (0118) 989 4254 between 0800 and 1800 UK To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Aug 30 11:28:22 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 418AF14CE1; Mon, 30 Aug 1999 11:28:19 -0700 (PDT) (envelope-from dillon@apollo.backplane.com) Received: (from dillon@localhost) by apollo.backplane.com (8.9.3/8.9.1) id LAA16324; Mon, 30 Aug 1999 11:28:06 -0700 (PDT) (envelope-from dillon) Date: Mon, 30 Aug 1999 11:28:06 -0700 (PDT) From: Matthew Dillon Message-Id: <199908301828.LAA16324@apollo.backplane.com> To: Bob Bishop Cc: current@FreeBSD.ORG, smp@FreeBSD.ORG Subject: Re: NFS-related panic on SMP References: Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org :Hi, : :With a kernel built from -current cvsup'd at Sun Aug 22 04:03:19 BST 1999, :I got the following panic doing make -j8 world with /usr/src via NFS and :/usr/obj local: : :Panic: free vnode isn't :mp_lock = 00000001; cpuid = 0; lapic.id = 00000000 : :backtrace (hand-transcribed): : :panic :getnewvnode() + 0x157 :nfs_nget() + 0x107 :nfs_lookup()+0xe8d :lookup() + 0x2ab :namei() + 0x137 :stat() + 0x44 :syscall() + 0x186 :Xint0x80_syscall() + 0x31 : :Any ideas? :-- :Bob Bishop (0118) 977 4017 international code +44 118 :rb@gid.co.uk fax (0118) 989 4254 between 0800 and 1800 UK This is the second person with a crash in nfs_lookup. No major changes have been made to NFS yet -- my patch set is still uncommitted. I can only conclude that something else someone has done has broken something. With my patchset and the latest current, I am able to do buildworlds in a loop forever on an SMP box using precisely that setup: /usr/src via NFS, /usr/obj local (and also /usr/obj/ via VN). At this instant my patch set will not patch cleanly due to some inappropriate commits made last night which I am trying to resolve. (However, these commits are probably not responsible for this particular problem since you indicate an early sunday morning image). I recommend waiting a few days, hopefully this mess will be straightened out by then and we can put everyone on the same page again. I can't update my patch until the mess that was committed last night has been backed out. -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 Mon Aug 30 15:39:28 1999 Delivered-To: freebsd-smp@freebsd.org Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (Postfix) with ESMTP id 456CC14E32 for ; Mon, 30 Aug 1999 15:39:23 -0700 (PDT) (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost.cdrom.com [127.0.0.1]) by dingo.cdrom.com (8.9.3/8.8.8) with ESMTP id PAA00751 for ; Mon, 30 Aug 1999 15:30:41 -0700 (PDT) (envelope-from mike@dingo.cdrom.com) Message-Id: <199908302230.PAA00751@dingo.cdrom.com> X-Mailer: exmh version 2.0.2 2/24/98 To: smp@freebsd.org Subject: BIOS calls on SMP systems Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 30 Aug 1999 15:30:41 -0700 From: Mike Smith Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Can someone familiar with the Intel MP spec clarify for me whether it's permissible for an AP to call the BIOS, or whether only the BSP may do this? Thanks. -- \\ The mind's the standard \\ Mike Smith \\ of the man. \\ msmith@freebsd.org \\ -- Joseph Merrick \\ 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 Aug 30 18:53:44 1999 Delivered-To: freebsd-smp@freebsd.org Received: from mail.rdc1.sfba.home.com (ha1.rdc1.sfba.home.com [24.0.0.66]) by hub.freebsd.org (Postfix) with ESMTP id B8E9E152A6 for ; Mon, 30 Aug 1999 18:53:43 -0700 (PDT) (envelope-from adsharma@c62443-a.frmt1.sfba.home.com) Received: from c62443-a.frmt1.sfba.home.com ([24.0.69.165]) by mail.rdc1.sfba.home.com (InterMail v4.01.01.00 201-229-111) with ESMTP id <19990831015148.DXRO29524.mail.rdc1.sfba.home.com@c62443-a.frmt1.sfba.home.com>; Mon, 30 Aug 1999 18:51:48 -0700 Received: (from adsharma@localhost) by c62443-a.frmt1.sfba.home.com (8.9.3/8.9.3) id UAA31591; Mon, 30 Aug 1999 20:51:43 -0700 Date: Mon, 30 Aug 1999 20:51:43 -0700 From: Arun Sharma To: Mike Smith Cc: smp@FreeBSD.ORG Subject: Re: BIOS calls on SMP systems Message-ID: <19990830205143.A31493@home.com> References: <199908302230.PAA00751@dingo.cdrom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.6i In-Reply-To: <199908302230.PAA00751@dingo.cdrom.com>; from Mike Smith on Mon, Aug 30, 1999 at 03:30:41PM -0700 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, Aug 30, 1999 at 03:30:41PM -0700, Mike Smith wrote: > > Can someone familiar with the Intel MP spec clarify for me whether it's > permissible for an AP to call the BIOS, or whether only the BSP may do > this? MP Spec doesn't say that APs can't call the BIOS. So I think you're okay if you make BIOS calls from an AP. However, the MP spec does mention that most BIOSes are not multithreaded and hence the caller must make sure that only one AP is executing the BIOS at anytime. [ Appendix A - System BIOS programming guidelines ] -Arun To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Aug 30 21:53:38 1999 Delivered-To: freebsd-smp@freebsd.org Received: from synthcom.com (beacon.synthcom.com [198.145.98.253]) by hub.freebsd.org (Postfix) with ESMTP id 5DA1C153B8 for ; Mon, 30 Aug 1999 21:53:32 -0700 (PDT) (envelope-from neil@synthcom.com) Received: from RETROCADE (unix.synthcom.com [198.145.98.7]) by synthcom.com (8.9.3/8.9.3) with SMTP id WAA04210 for ; Mon, 30 Aug 1999 22:03:49 -0700 (PDT) (envelope-from neil@synthcom.com) Message-ID: <009c01bef36d$394ee850$076291c6@Engineering.retrocade.com> From: "Neil Bradley" To: References: <199908302230.PAA00751@dingo.cdrom.com> Subject: Re: BIOS calls on SMP systems Date: Mon, 30 Aug 1999 21:56:44 -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 5.00.2918.2701 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2918.2701 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Can someone familiar with the Intel MP spec clarify for me whether it's > permissible for an AP to call the BIOS, or whether only the BSP may do > this? It depends upon what services of the BIOS you're looking to call. If it's the standard INT services, only the BSP will be able to do it reliably. If you're calling the PCI functions, then you're most likely safe. But it is safe for the BSP to call BIOS even when the AP's aren't. -->Neil To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Aug 30 22:48:56 1999 Delivered-To: freebsd-smp@freebsd.org Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (Postfix) with ESMTP id 4B60C159FB for ; Mon, 30 Aug 1999 22:48:53 -0700 (PDT) (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost.cdrom.com [127.0.0.1]) by dingo.cdrom.com (8.9.3/8.8.8) with ESMTP id WAA02756; Mon, 30 Aug 1999 22:42:31 -0700 (PDT) (envelope-from mike@dingo.cdrom.com) Message-Id: <199908310542.WAA02756@dingo.cdrom.com> X-Mailer: exmh version 2.0.2 2/24/98 To: "Neil Bradley" Cc: smp@freebsd.org Subject: Re: BIOS calls on SMP systems In-reply-to: Your message of "Mon, 30 Aug 1999 21:56:44 PDT." <009c01bef36d$394ee850$076291c6@Engineering.retrocade.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 30 Aug 1999 22:42:31 -0700 From: Mike Smith Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > Can someone familiar with the Intel MP spec clarify for me whether it's > > permissible for an AP to call the BIOS, or whether only the BSP may do > > this? > > It depends upon what services of the BIOS you're looking to call. If it's > the standard INT services, only the BSP will be able to do it reliably. If > you're calling the PCI functions, then you're most likely safe. But it is > safe for the BSP to call BIOS even when the AP's aren't. Ok, so the baseline answer is "only the BSP can reliably call the BIOS". Correct there? If so, I have some interesting tunneling work to do... -- \\ The mind's the standard \\ Mike Smith \\ of the man. \\ msmith@freebsd.org \\ -- Joseph Merrick \\ 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 Aug 30 23: 2: 0 1999 Delivered-To: freebsd-smp@freebsd.org Received: from synthcom.com (beacon.synthcom.com [198.145.98.253]) by hub.freebsd.org (Postfix) with ESMTP id 7E6C1153E8 for ; Mon, 30 Aug 1999 23:01:45 -0700 (PDT) (envelope-from neil@synthcom.com) Received: from RETROCADE (unix.synthcom.com [198.145.98.7]) by synthcom.com (8.9.3/8.9.3) with SMTP id XAA04495 for ; Mon, 30 Aug 1999 23:09:52 -0700 (PDT) (envelope-from neil@synthcom.com) Message-ID: <02d601bef376$74434a10$076291c6@Engineering.retrocade.com> From: "Neil Bradley" To: References: <199908310542.WAA02756@dingo.cdrom.com> Subject: Re: BIOS calls on SMP systems Date: Mon, 30 Aug 1999 23:02:47 -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 5.00.2918.2701 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2918.2701 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > > Can someone familiar with the Intel MP spec clarify for me whether it's > > > permissible for an AP to call the BIOS, or whether only the BSP may do > > > this? > > It depends upon what services of the BIOS you're looking to call. If it's > > the standard INT services, only the BSP will be able to do it reliably. If > > you're calling the PCI functions, then you're most likely safe. But it is > > safe for the BSP to call BIOS even when the AP's aren't. > Ok, so the baseline answer is "only the BSP can reliably call the BIOS". Correct - but with a "depends" attached to it. What interrupt services are you calling? > Correct there? If so, I have some interesting tunneling work to do... Yes, but it depends upon what services you want to call in the BIOS. ON second thought, even doing 32 bit PCI function calls won't be safe because of the access to 0xcf8/0xcfc for configuration access. In other words, I wouldn't use the AP's to call BIOS services. -->Neil To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Mon Aug 30 23: 9:43 1999 Delivered-To: freebsd-smp@freebsd.org Received: from dingo.cdrom.com (dingo.cdrom.com [204.216.28.145]) by hub.freebsd.org (Postfix) with ESMTP id 0565F15108 for ; Mon, 30 Aug 1999 23:09:41 -0700 (PDT) (envelope-from mike@dingo.cdrom.com) Received: from dingo.cdrom.com (localhost.cdrom.com [127.0.0.1]) by dingo.cdrom.com (8.9.3/8.8.8) with ESMTP id XAA02907; Mon, 30 Aug 1999 23:03:14 -0700 (PDT) (envelope-from mike@dingo.cdrom.com) Message-Id: <199908310603.XAA02907@dingo.cdrom.com> X-Mailer: exmh version 2.0.2 2/24/98 To: "Neil Bradley" Cc: smp@freebsd.org Subject: Re: BIOS calls on SMP systems In-reply-to: Your message of "Mon, 30 Aug 1999 23:02:47 PDT." <02d601bef376$74434a10$076291c6@Engineering.retrocade.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 30 Aug 1999 23:03:13 -0700 From: Mike Smith Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > Ok, so the baseline answer is "only the BSP can reliably call the BIOS". > > Correct - but with a "depends" attached to it. What interrupt services are > you calling? At the moment, the PnP, APM and SMB BIOS APIs. We will probably end up calling the PCI BIOS. > > Correct there? If so, I have some interesting tunneling work to do... > > Yes, but it depends upon what services you want to call in the BIOS. ON > second thought, even doing 32 bit PCI function calls won't be safe because > of the access to 0xcf8/0xcfc for configuration access. > > In other words, I wouldn't use the AP's to call BIOS services. Ok, thanks. More fun. 8) -- \\ The mind's the standard \\ Mike Smith \\ of the man. \\ msmith@freebsd.org \\ -- Joseph Merrick \\ 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 Wed Sep 1 22:36:37 1999 Delivered-To: freebsd-smp@freebsd.org Received: from mail.rdc1.sfba.home.com (ha1.rdc1.sfba.home.com [24.0.0.66]) by hub.freebsd.org (Postfix) with ESMTP id 06A8914D0A for ; Wed, 1 Sep 1999 22:36:32 -0700 (PDT) (envelope-from adsharma@c62443-a.frmt1.sfba.home.com) Received: from c62443-a.frmt1.sfba.home.com ([24.0.69.165]) by mail.rdc1.sfba.home.com (InterMail v4.01.01.00 201-229-111) with ESMTP id <19990902053607.RKWO29524.mail.rdc1.sfba.home.com@c62443-a.frmt1.sfba.home.com> for ; Wed, 1 Sep 1999 22:36:07 -0700 Received: (from adsharma@localhost) by c62443-a.frmt1.sfba.home.com (8.9.3/8.9.3) id WAA04623 for freebsd-smp@freebsd.org; Wed, 1 Sep 1999 22:36:07 -0700 Date: Wed, 1 Sep 1999 22:36:07 -0700 From: Arun Sharma To: freebsd-smp@freebsd.org Subject: Kernel + xosview patch for cp_time Message-ID: <19990901223607.A4616@home.com> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary=EeQfGwPcQSOJBaQU X-Mailer: Mutt 0.95.6i Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org --EeQfGwPcQSOJBaQU Content-Type: text/plain; charset=us-ascii Here's a minor patch to make the kernel keep track of per CPU times and another patch to help xosview take advantage of that. -Arun --EeQfGwPcQSOJBaQU Content-Type: text/plain; charset=us-ascii Content-Disposition: attachment; filename=xosview-bsd-patch diff -u /usr/ports/sysutils/xosview/work/xosview-1.7.1/bsd/MeterMaker.cc bsd/MeterMaker.cc --- /usr/ports/sysutils/xosview/work/xosview-1.7.1/bsd/MeterMaker.cc Mon Jan 25 12:15:26 1999 +++ bsd/MeterMaker.cc Wed Sep 1 21:33:15 1999 @@ -44,8 +44,12 @@ push(new LoadMeter(_xos)); // Standard meters (usually added, but users could turn them off) - if (_xos->isResourceTrue("cpu")) - push(new CPUMeter(_xos)); + if (_xos->isResourceTrue("cpu")) { + int cpuCount = CPUMeter::countCPUs(); + int start = (cpuCount == 0) ? 0 : 1; + for (int i = start ; i <= cpuCount ; i++) + push(new CPUMeter(_xos, CPUMeter::cpuStr(i), i)); + } if (_xos->isResourceTrue("mem")) push(new MemMeter(_xos)); diff -u /usr/ports/sysutils/xosview/work/xosview-1.7.1/bsd/cpumeter.cc bsd/cpumeter.cc --- /usr/ports/sysutils/xosview/work/xosview-1.7.1/bsd/cpumeter.cc Sun Jan 31 12:18:49 1999 +++ bsd/cpumeter.cc Wed Sep 1 22:27:33 1999 @@ -16,6 +16,9 @@ // #include // For CPUSTATES #define. BCG #include // For use of atoi BCG +#include +#include +#include #include "general.h" #include "cpumeter.h" #include "kernel.h" // For NetBSD-specific icky kvm_ code. BCG @@ -23,18 +26,20 @@ CVSID("$Id: cpumeter.cc,v 1.16 1999/01/31 20:18:49 bgrayson Exp $"); CVSID_DOT_H(CPUMETER_H_CVSID); -CPUMeter::CPUMeter( XOSView *parent ) +CPUMeter::CPUMeter( XOSView *parent, const char *cpuID, int cpuindex) #ifdef XOSVIEW_FREEBSD -: FieldMeterGraph( parent, 5, "CPU", "USR/NICE/SYS/INT/FREE" ){ +: FieldMeterGraph( parent, 5, cpuID, "USR/NICE/SYS/INT/FREE" ){ #define FREE_INDEX 4 #else -: FieldMeterGraph( parent, 4, "CPU", "USR/NICE/SYS/FREE" ){ +: FieldMeterGraph( parent, 4, cpuID, "USR/NICE/SYS/FREE" ){ #define FREE_INDEX 3 #endif for ( int i = 0 ; i < 2 ; i++ ) for ( int j = 0 ; j < 4 ; j++ ) cputime_[i][j] = 0; + cpuindex_ = 0; + index = cpuindex - 1; // The setting of the priority will be done in checkResources(). BCG dodecay_ = 0; @@ -74,7 +79,7 @@ // Begin NetBSD-specific code... BCG long tempCPU[CPUSTATES]; - BSDGetCPUTimes (tempCPU); + BSDGetCPUTimes (tempCPU, index); cputime_[cpuindex_][0] = tempCPU[0]; cputime_[cpuindex_][1] = tempCPU[1]; @@ -124,3 +129,31 @@ cpuindex_ = (cpuindex_ + 1) % 2; } + + +int CPUMeter::countCPUs(void) +{ + int mib[2], ncpu; + size_t len; + + mib[0] = CTL_HW; + mib[1] = HW_NCPU; + len = sizeof(ncpu); + sysctl(mib, 2, &ncpu, &len, NULL, 0); + + return ncpu; +} + +const char *CPUMeter::cpuStr(int num) +{ + static char buffer[32]; + ostrstream str(buffer, 32); + + str << "cpu"; + if (num != 0) + str << (num - 1); + str << ends; + + return buffer; +} + diff -u /usr/ports/sysutils/xosview/work/xosview-1.7.1/bsd/cpumeter.h bsd/cpumeter.h --- /usr/ports/sysutils/xosview/work/xosview-1.7.1/bsd/cpumeter.h Tue Oct 20 12:37:33 1998 +++ bsd/cpumeter.h Wed Sep 1 22:09:32 1999 @@ -23,16 +23,19 @@ class CPUMeter : public FieldMeterGraph { public: - CPUMeter( XOSView *parent ); + CPUMeter( XOSView *parent, const char *cpuID = "cpu", int cpuindex = 0); ~CPUMeter( void ); const char *name( void ) const { return "CPUMeter"; } void checkevent( void ); + static int countCPUs(void); + static const char *cpuStr(int num); void checkResources( void ); protected: long cputime_[2][5]; int cpuindex_; + int index; void getcputime( void ); private: diff -u /usr/ports/sysutils/xosview/work/xosview-1.7.1/bsd/kernel.cc bsd/kernel.cc --- /usr/ports/sysutils/xosview/work/xosview-1.7.1/bsd/kernel.cc Wed Sep 1 22:17:57 1999 +++ bsd/kernel.cc Wed Sep 1 21:52:36 1999 @@ -289,11 +289,13 @@ } void -BSDGetCPUTimes (long* timeArray) { +BSDGetCPUTimes (long* timeArray, int index) { + int size = sizeof (long) * CPUSTATES; if (!timeArray) errx (-1, "BSDGetCPUTimes(): passed pointer was null!\n"); if (CPUSTATES != 5) errx (-1, "Error: xosview for *BSD expects 5 cpu states!\n"); - safe_kvm_read_symbol (CP_TIME_SYM_INDEX, timeArray, sizeof (long) * CPUSTATES); + safe_kvm_read(nlst[CP_TIME_SYM_INDEX].n_value + (index * size), + timeArray, size); } diff -u /usr/ports/sysutils/xosview/work/xosview-1.7.1/bsd/kernel.h bsd/kernel.h --- /usr/ports/sysutils/xosview/work/xosview-1.7.1/bsd/kernel.h Fri May 29 14:21:36 1998 +++ bsd/kernel.h Wed Sep 1 21:28:43 1999 @@ -39,7 +39,7 @@ BSDCPUInit(); void -BSDGetCPUTimes(long* timesArray); +BSDGetCPUTimes(long* timesArray, int index); void BSDNetInit(); --EeQfGwPcQSOJBaQU Content-Type: text/plain; charset=us-ascii Content-Disposition: attachment; filename="mp_cp_time.patch" Index: kern//kern_clock.c =================================================================== RCS file: /home/adsharma/cvs_root/freebsd-sys/kern/kern_clock.c,v retrieving revision 1.1.1.2 diff -u -r1.1.1.2 kern_clock.c --- kern_clock.c 1999/08/31 05:12:42 1.1.1.2 +++ kern_clock.c 1999/09/02 04:13:35 @@ -89,7 +89,7 @@ /* Some of these don't belong here, but it's easiest to concentrate them. */ #if defined(SMP) && defined(BETTER_CLOCK) -long cp_time[CPUSTATES]; +long cp_time[NCPUS][CPUSTATES]; #else static long cp_time[CPUSTATES]; #endif @@ -389,9 +389,17 @@ */ p->p_uticks++; if (p->p_nice > NZERO) +#ifdef SMP + cp_time[cpuid][CP_NICE]++; +#else cp_time[CP_NICE]++; +#endif else +#ifdef SMP + cp_time[cpuid][CP_USER]++; +#else cp_time[CP_USER]++; +#endif } else { #ifdef GPROF /* @@ -428,12 +436,26 @@ if (CLKF_INTR(frame)) { if (p != NULL) p->p_iticks++; +#ifdef SMP + cp_time[cpuid][CP_INTR]++; +#else cp_time[CP_INTR]++; +#endif + } else if (p != NULL) { p->p_sticks++; +#ifdef SMP + cp_time[cpuid][CP_SYS]++; +#else cp_time[CP_SYS]++; +#endif } else +#ifdef SMP + cp_time[cpuid][CP_IDLE]++; +#else cp_time[CP_IDLE]++; +#endif + } pscnt = psdiv; Index: i386//i386/mp_machdep.c =================================================================== RCS file: /home/adsharma/cvs_root/freebsd-sys/i386/i386/mp_machdep.c,v retrieving revision 1.1.1.2 diff -u -r1.1.1.2 mp_machdep.c --- mp_machdep.c 1999/08/31 05:12:10 1.1.1.2 +++ mp_machdep.c 1999/09/02 04:15:26 @@ -2227,8 +2227,9 @@ int checkstate_cpustate[NCPU]; u_long checkstate_pc[NCPU]; -extern long cp_time[CPUSTATES]; +extern long cp_time[NCPUS][CPUSTATES]; + #define PC_TO_INDEX(pc, prof) \ ((int)(((u_quad_t)((pc) - (prof)->pr_off) * \ (u_quad_t)((prof)->pr_scale)) >> 16) & ~1) @@ -2278,9 +2279,9 @@ return; p->p_uticks++; if (p->p_nice > NZERO) - cp_time[CP_NICE]++; + cp_time[cpuid][CP_NICE]++; else - cp_time[CP_USER]++; + cp_time[cpuid][CP_USER]++; break; case CHECKSTATE_SYS: #ifdef GPROF @@ -2300,10 +2301,10 @@ return; if (!p) - cp_time[CP_IDLE]++; + cp_time[cpuid][CP_IDLE]++; else { p->p_sticks++; - cp_time[CP_SYS]++; + cp_time[cpuid][CP_SYS]++; } break; case CHECKSTATE_INTR: @@ -2325,7 +2326,7 @@ return; if (p) p->p_iticks++; - cp_time[CP_INTR]++; + cp_time[cpuid][CP_INTR]++; } if (p != NULL) { p->p_cpticks++; --EeQfGwPcQSOJBaQU-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Thu Sep 2 7:26: 4 1999 Delivered-To: freebsd-smp@freebsd.org Received: from server3.yawl.com.br (server3.yawl.com.br [200.252.114.3]) by hub.freebsd.org (Postfix) with SMTP id E5B5215647 for ; Thu, 2 Sep 1999 07:25:57 -0700 (PDT) (envelope-from einstein@yawl.com.br) Received: (qmail 95781 invoked from network); 2 Sep 1999 14:25:55 -0000 Received: from unknown (HELO yawl.com.br) (200.252.114.253) by server3.yawl.com.br with SMTP; 2 Sep 1999 14:25:55 -0000 Message-ID: <37CE88F2.58F9C930@yawl.com.br> Date: Thu, 02 Sep 1999 11:25:54 -0300 From: Einstein Oliveira X-Mailer: Mozilla 4.61 [en] (WinNT; I) X-Accept-Language: en MIME-Version: 1.0 To: freebsd-smp@freebsd.org Subject: subscribe Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: 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 2 17:57:40 1999 Delivered-To: freebsd-smp@freebsd.org Received: from pathlink.net (linda.pathlink.com [165.113.238.197]) by hub.freebsd.org (Postfix) with ESMTP id 603B314DC3 for ; Thu, 2 Sep 1999 17:57:35 -0700 (PDT) (envelope-from kachun@pathlink.com) Received: from dvl-1 (dvl-1.pathlink.com [207.211.168.211]) by pathlink.net (8.9.3/8.8.7) with SMTP id RAA88215 for ; Thu, 2 Sep 1999 17:57:54 -0700 (PDT) Message-Id: <199909030057.RAA88215@pathlink.net> To: freebsd-smp@freebsd.org Subject: Re: SMP broken in 3.2-STABLE? From: kachun@pathlink.com (Kachun Lee) Date: Thu, 02 Sep 1999 17:57:20 -0700 X-Mailer: WinVN 0.99.8 (x86 32bit) In-Reply-To: <199908292255.AAA09980@office.omc.net> References: <199908292255.AAA09980@office.omc.net> MIME-Version: 1.0 Content-Type: Text/Plain; charset=US-ASCII Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Samething here... CVSUP on 8/31/99... ASUS P2B-DS 2*PII-450. In article <199908292255.AAA09980@office.omc.net>, you say... > > >I just CVSUPed to the latest 3.2-STABLE, "make world" and >recompiled the kernel. > >After rebooting the system (ASUS P2B-DS, 2* PIII-450) I get >core dumps from csh, sendmail, ntpdate, cron ... > >Most of the tools (sh, top, gcc ...) work. When I rebuild the >kernel without SMP everything works fine. > >The system worked fine in SMP-mode with 3.2-STABLE (Aug.21). > >Any ideas? > > >Thanks, >Lutz Rabing -OMCnet- > >Mit freundlichen Gruessen, >Lutz Rabing -OMCnet- > > > >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 3 8:13:40 1999 Delivered-To: freebsd-smp@freebsd.org Received: from smtp1.xs4all.nl (smtp1.xs4all.nl [194.109.127.48]) by hub.freebsd.org (Postfix) with ESMTP id 8103F14FD5; Fri, 3 Sep 1999 08:13:27 -0700 (PDT) (envelope-from arnout@tomcat.xs4all.nl) Received: from tomcat.xs4all.nl (tomcat.xs4all.nl [194.109.15.187]) by smtp1.xs4all.nl (8.9.3/8.9.3) with ESMTP id RAA06397; Fri, 3 Sep 1999 17:13:13 +0200 (CEST) Received: (from arnout@localhost) by tomcat.xs4all.nl (8.9.3/8.9.1) id RAA10998; Fri, 3 Sep 1999 17:16:27 +0200 (CEST) Date: Fri, 3 Sep 1999 17:16:27 +0200 From: Arnout Boer To: freebsd-emulation@freebsd.org, freebsd-stable@freebsd.org Subject: SO 5.1 and SMP under stable!? Message-ID: <19990903171627.B10976@tomcat.xs4all.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.6i X-Operating-System: FreeBSD 3.3-RC i386 Sender: owner-freebsd-smp@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I like to play with an office Suite under FreeBSD but SO doens't work onder an SMP stable kernel... StarOffice is ok.. It does crash frequently while playing with 'new things' but I like it.... A Free Office package on BSD is a plus! But can anybody tell me why it doesn't work ok a SMP kernel... I would like to play with it on my dual P200 instead of only one proc... Is there any change SMP changes will be ported to stable that make this work... Probably threags stuff.... I Think it's about problems with PosixThread vs SMP!? Btw.. I'm running STABLE 3.3 RC Single proc now because of SO :-( Greetz, Arnout p.s. If patches will be ported to stable I'll love to test them! To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 3 10:27:59 1999 Delivered-To: freebsd-smp@freebsd.org Received: from alpha.netaccess.on.ca (alpha.netaccess.on.ca [199.243.225.10]) by hub.freebsd.org (Postfix) with ESMTP id 369E315230 for ; Fri, 3 Sep 1999 10:27:35 -0700 (PDT) (envelope-from rob@ControlQ.com) Received: from fatlady.controlq.com (dial220.nas.net [199.243.225.220]) by alpha.netaccess.on.ca (8.9.0/8.9.0) with SMTP id NAA25856 for ; Fri, 3 Sep 1999 13:26:27 -0400 (EDT) Date: Fri, 3 Sep 1999 13:28:45 -0400 (EDT) From: "Robert S. Sciuk" To: freebsd-smp@freebsd.org Subject: 3.2 cvsup/kernel build now fixed ... 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 TWIMC, Thanks for the source fixes ... 3.x smp is now compiling properly ... I've managed to cvsup and build a 3.2-STABLE smp kernel on a dual PII 266 QDI board ... the uname -a reports a 3.3-RC identifier ... but it does seem to work just fine ... I'm now in the process of updating a dual 333mhz PII QDI board based system from 3.1-RELEASE to the same level ... Cheers, Rob. -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Robert S. Sciuk 1032 Howard Rd. PO Box 6A Ph:905 632-2466 Control-Q Research Burlington, Ont. Canada Fx:905 632-7417 rob@ControlQ.com L7R 3X5 http://www.ControlQ.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message From owner-freebsd-smp Fri Sep 3 12:20:59 1999 Delivered-To: freebsd-smp@freebsd.org Received: from ady.warpnet.ro (ady.warpnet.ro [194.102.224.1]) by hub.freebsd.org (Postfix) with ESMTP id E50BE1565A for ; Fri, 3 Sep 1999 12:20:44 -0700 (PDT) (envelope-from ady@freebsd.ady.ro) Received: from localhost (ady@localhost) by ady.warpnet.ro (8.9.3/8.9.3) with ESMTP id WAA61159; Fri, 3 Sep 1999 22:21:28 +0300 (EEST) (envelope-from ady@freebsd.ady.ro) Date: Fri, 3 Sep 1999 22:21:28 +0300 (EEST) From: Adrian Penisoara X-Sender: ady@ady.warpnet.ro To: "Robert S. Sciuk" Cc: freebsd-smp@FreeBSD.ORG Subject: Re: 3.2 cvsup/kernel build now fixed ... 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 Hi, On Fri, 3 Sep 1999, Robert S. Sciuk wrote: > > TWIMC, > > Thanks for the source fixes ... 3.x smp is now compiling properly ... Was there a breakage in RELENG_3 for SMP ?? It seems that I missed the fact... > > I've managed to cvsup and build a 3.2-STABLE smp kernel on a dual PII 266 > QDI board ... the uname -a reports a 3.3-RC identifier ... but it does > seem to work just fine ... I'm now in the process of updating a dual > 333mhz PII QDI board based system from 3.1-RELEASE to the same level ... There's nothing to worry about the 3.3-RC string, jkh just entered RELENG_3 into 3.3-Release Candidate stage, 3.3-REL will probably go out on mid-September, as planned. > > Cheers, > Rob. > > -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- > Robert S. Sciuk 1032 Howard Rd. PO Box 6A Ph:905 632-2466 > Control-Q Research Burlington, Ont. Canada Fx:905 632-7417 > rob@ControlQ.com L7R 3X5 http://www.ControlQ.com > Ady (@freebsd.ady.ro) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message