From owner-freebsd-stable Sun Jan 14 03:04:15 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id DAA27518 for stable-outgoing; Sun, 14 Jan 1996 03:04:15 -0800 (PST) Received: from DeepCore.dk (dial114.cybercity.dk [194.16.56.114]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id DAA27484 Sun, 14 Jan 1996 03:03:36 -0800 (PST) From: sos@DeepCore.dk Received: (from sos@localhost) by DeepCore.dk (8.7.3/8.7.3) id LAA00835; Sun, 14 Jan 1996 11:58:54 +0100 (MET) Message-Id: <199601141058.LAA00835@DeepCore.dk> Subject: Re: abuse and the -stable linux emu (mknod & mkfifo) To: bde@zeta.org.au (Bruce Evans) Date: Sun, 14 Jan 1996 11:58:54 +0100 (MET) Cc: ache@astral.msk.su, gpalmer@cdrom.com, hasty@rah.star-gate.com, hackers@FreeBSD.ORG, stable@FreeBSD.ORG In-Reply-To: <199601132301.KAA06594@godzilla.zeta.org.au> from "Bruce Evans" at Jan 14, 96 10:01:32 am Reply-to: sos@FreeBSD.ORG X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-stable@FreeBSD.ORG Precedence: bulk In reply to Bruce Evans who wrote: > > >I remember recent discussion for adding mkfifo capabilities > >to system mknod (instead of this hack). > >Could anybody tell me the result, I miss it... > > The Linux emulator should convert Linux mknod()'s to mkfifo()s > if necessary. The BSD mknod() shouldn't be changed. Note that > it is reasonable for the Linux library to call mknod() directly > although Linux apps shouldn't. I've allready commit the prober fix for this in -current... -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Soren Schmidt (sos@FreeBSD.org) FreeBSD Core Team Even more code to hack -- will it ever end .. From owner-freebsd-stable Sun Jan 14 04:14:51 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id EAA02585 for stable-outgoing; Sun, 14 Jan 1996 04:14:51 -0800 (PST) Received: from silvia.HIP.Berkeley.EDU (silvia.HIP.Berkeley.EDU [136.152.64.181]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id EAA02573 Sun, 14 Jan 1996 04:14:42 -0800 (PST) Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.7.3/8.6.9) id EAA01042; Sun, 14 Jan 1996 04:14:40 -0800 (PST) Date: Sun, 14 Jan 1996 04:14:40 -0800 (PST) Message-Id: <199601141214.EAA01042@silvia.HIP.Berkeley.EDU> To: hackers@freebsd.org CC: stable@freebsd.org, ccd@forgery.cs.berkeley.edu Subject: ccd driver or 2.1R available Reply-to: stable@silvia.HIP.Berkeley.EDU From: asami@cs.berkeley.edu (Satoshi Asami) Sender: owner-stable@freebsd.org Precedence: bulk (Note crossposting -- followups to -stable only please, this is 2.1R material.) I put up the initial snapshot of the ccd driver on ftp. The README is included below. Satoshi ------- ********************************************************** * The FreeBSD ccd driver (pre-alpha) * * for 2.1.0-RELEASE users * * by Satoshi Asami * * and Nisha Talagala * * version of 96/01/14 * ********************************************************** (1) Introduction This is a port of NetBSD's ccd (concatenated disk) driver. It is not a complete work in any way, but seems to be working fairly ok here, so we decided to make it available so that people can test it and even fix some bugs. :) (2) Warning As the title above says, this is pre-alpha software and is therefore VERY GREEN. You should not expect this to work. It may eat your system disk for lunch. It may even fry your microwave in the process. Make sure you have backed up all your data and locked your children in the basement before you attempt to try this. (3) What it does In case you don't know what it is, ccd is a disk array driver. You can combine several disk partitions into one "virtual disk". Then you can partition it or use the whole thing or add some pepper and salt or whatever you want. (4) What it does not There is no parity support yet. That's why its name doesn't resemble RAID in any form. (5) How to compile Note this package is for FreeBSD-2.1.0R. It probably won't work right away for -stable or -current. We are planning to upgrade our machines soon, and will release snapshots for -current as well. Ok, first you need to unpack the distribution. It contains the following files: ccd.README (this file) ccd.PLIST ccd.patch sys/ccd/ccd.4 sys/ccd/ccd.c sys/sys/ccdvar.h usr.sbin/ccdcontrol/Makefile usr.sbin/ccdcontrol/ccdcontrol.8 usr.sbin/ccdcontrol/ccdcontrol.c usr.sbin/ccdcontrol/pathnames.h You can unpack them in your /usr/src directory. After that, apply the ccd.patch. It makes several changes to the header files and such, including one totally ridiculous change to sys/disk.h that was done by a certain FreeBSD hacker who owns a hamster with one black eye and one red eye. (Someone please tell me how to fix it properly.) You can use the following command to apply the patch: patch -p < ccd.patch (On a separate line for your triple-clicking pleasure.) Then add the following to your kernel configuration file: device ccd0 at isa? device ccd1 at isa? device ccd2 at isa? device ccd3 at isa? (You can have as many of them as you want, or fewer than four, of course.) We recommend you to add "options DDB" too. This will make the kernel go into a debugger in case of a panic. That will make it easier for you to send us a complete bug report. (Note this will disable auto-rebooting after a panic, so don't do this on a machine that has to run unattended.) (6) How to use ccd Wait, you need to compile ccdcontrol too! Go in there, do "make depend all install". If install complains about missing directories and such, make sure you have the correct Makefile.inc in the parent (/usr/src/usr.sbin) directory. If you don't have one (like, you don't have the usr.sbin source tree), just editing the ccdcontrol Makefile and adding "BINDIR=/usr/sbin" by hand should do. Also, you will need to create the device files. There is a patch to MAKEDEV included in ccd.patch, so go into etc/etc.i386 and install the new MAKEDEV into /dev. Then you can, say, "cd /dev; sh MAKEDEV ccd0" to create the appropriate device files for your first ccd devices, etc. Now, go and read the man pages. ccdcontrol.8 should be installed by now; ccd.4 is still sitting in the source directory so go read it there. (Try "nroff -man ccd.4 | less -s".) Assuming you've read them, here is an example, if you have four partitions (sd1g, sd2g, sd3g, sd4g) you want to combine into one: echo "ccd0 16 none /dev/sd1g /dev/sd2g /dev/sd3g /dev/sd4g" > /etc/ccd.conf ccdcontrol -Cv (to configure; you'll see a message here) disklabel ccd0 (just to make sure there is a valid disklabel) newfs /dev/rccd0c mount /dev/ccd0c /mnt (play with /mnt) umount /mnt ccdcontrol -U (to unconfigure) (7) What's the second field in /etc/ccd.conf? That's the "interleave size". Basically, the ccd driver will write this many sectors (usually 512 bytes) to one disk before it moves to the next disk. As a special case, a zero here means no interleave, i.e., to concatenate disks serially. We have found that in FFS, a value of 16 usually optimizes read performance, while the write peaks with a much larger value (like 512). This probably has to do with cluster_write() thinking it's writing to a single disk when it's actually not. This is one of the things we are planning to fix. (8) Are there any caveats? Oh sure. One of them is "don't use a partition that starts at a beginning of the slice". So please leave some space at the beginning of the slice in the partitions you are combining (sd[1-4]g in the above example). Of course, if someone can figure out why and fix it, that will be great. (9) My disk is totally hosed. It's all your fault! See (2). (10) Anything I can help? Well, any bug fix is welcome. In addition to the stuff mentioned above, we are aware of at least the following: a. Since I didn't know how to properly register a pseudo-device, this baby thinks it's a ISA device (look at kern_devconf and isa_driver and such in ccd.c). It should be specified like "pseudo-device ccd 4" in the kernel configuration file (like the manual says). b. I'm not exactly sure what the "geometry" of this pseudo-disk means. If it doesn't matter, it's ok; if newfs (for instance) cares, we should give it a better set of default values. (11) Where should I send bug reports/fixes? Please send them to ccd@forgery.cs.berkeley.edu. This will reach the two developers (Satoshi and Nisha). If you want to be added to this list, please send mail to Satoshi (asami@cs.berkeley.edu). (12) Where do I get new versions? They will be made available as ftp://forgery.cs.berkeley.edu/pub/ccd-.tar.gz so check this site regularly. (13) I'm tired Yeah, I'm tired too. Well go to sleep now then, and try it tomorrow! Good night! Satoshi From owner-freebsd-stable Mon Jan 15 19:18:46 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id TAA05139 for stable-outgoing; Mon, 15 Jan 1996 19:18:46 -0800 (PST) Received: from haven.uniserve.com (haven.uniserve.com [198.53.215.121]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id TAA05132 for ; Mon, 15 Jan 1996 19:18:38 -0800 (PST) Received: by haven.uniserve.com id <30954-1>; Mon, 15 Jan 1996 19:21:08 -0000 Date: Mon, 15 Jan 1996 19:21:06 -0800 (PST) From: Tom Samplonius To: freebsd-stable@freebsd.org Subject: ccd driver on -stable Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@freebsd.org Precedence: bulk Has anyone tried the ccd driver released yesterday on a -stable system? Does it work? Tom From owner-freebsd-stable Tue Jan 16 08:24:15 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id IAA23473 for stable-outgoing; Tue, 16 Jan 1996 08:24:15 -0800 (PST) Received: from who.cdrom.com (who.cdrom.com [192.216.222.3]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id IAA23419 for ; Tue, 16 Jan 1996 08:24:00 -0800 (PST) Received: from sivka.carrier.kiev.ua (root@sivka.carrier.kiev.ua [193.125.68.130]) by who.cdrom.com (8.6.12/8.6.11) with ESMTP id FAA25884 for ; Tue, 16 Jan 1996 05:23:53 -0800 Received: from elvisti.kiev.ua (uucp@localhost) by sivka.carrier.kiev.ua (Sendmail 8.who.cares/5) with UUCP id PAA01060 for stable@freebsd.org; Tue, 16 Jan 1996 15:24:01 +0200 Received: from office.elvisti.kiev.ua (office.elvisti.kiev.ua [193.125.28.33]) by spider2.elvisti.kiev.ua (8.6.12/8.ElVisti) with ESMTP id PAA09863 for ; Tue, 16 Jan 1996 15:17:19 +0200 Received: (from stesin@localhost) by office.elvisti.kiev.ua (8.6.12/8.ElVisti) id PAA14528 for stable@freebsd.org; Tue, 16 Jan 1996 15:17:18 +0200 From: "Andrew V. Stesin" Message-Id: <199601161317.PAA14528@office.elvisti.kiev.ua> Subject: 2.1 config and IObase constants strangeness To: stable@FreeBSD.org Date: Tue, 16 Jan 1996 15:17:17 +0200 (EET) X-Mailer: ELM [version 2.4 PL24alpha5] Content-Type: text Sender: owner-stable@FreeBSD.org Precedence: bulk Hello, while I'm busy with equipping a new box, I discovered an unexpected config (8) behavior. As the hardware configuration is comparatively complex, IObase space is filled in an unusual manner, I'm afraid of overlaps and I never bothered remembering all "magic" default IObase values for different devices (like IO_FD1 == 0x3f0 for fdc0), I changed the "magic" constants in the definitions of fdc0 and sio[01] to their hex values, (i.e. "IO_FD1" to 0x3f0 for fdc0 line) for fdc0, sio0, sio1 devices. Surprizingly, config (8) didn't tell me about any errors, but resulting ioconf.c has become broken: config (8) inserted extra (erroneous) declarations for siointr (two times) and fdintr. The system is 2.1-very-late-beta (951026-SNAP). Is this a correct behaviour? (Or maybe fixed in -current, as always? :) Aren't magic IObase constants considered harmful? -- With best regards -- Andrew Stesin. +380 (44) 2760188 +380 (44) 2713457 +380 (44) 2713560 An undocumented feature is a coding error. From owner-freebsd-stable Tue Jan 16 10:45:38 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id KAA02653 for stable-outgoing; Tue, 16 Jan 1996 10:45:38 -0800 (PST) Received: from asstdc.scgt.oz.au (root@asstdc.scgt.oz.au [202.14.234.65]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id KAA02648 for ; Tue, 16 Jan 1996 10:45:30 -0800 (PST) Received: (from imb@localhost) by asstdc.scgt.oz.au (8.6.12/BSD4.4) id FAA02175; Wed, 17 Jan 1996 05:45:17 +1100 From: michael butler Message-Id: <199601161845.FAA02175@asstdc.scgt.oz.au> Subject: Re: Adaptec 2842 driver To: gibbs@freefall.freebsd.org (Justin T. Gibbs) Date: Wed, 17 Jan 1996 05:45:13 +1100 (EST) Cc: freebsd-stable@freebsd.org In-Reply-To: <199601161607.IAA22696@freefall.freebsd.org> from "Justin T. Gibbs" at Jan 16, 96 08:07:30 am X-Mailer: ELM [version 2.4 PL24beta] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org Precedence: bulk Justin T. Gibbs writes: > >I have a couple of problems with this $#@& machine running a 2842. > >On start-up, it complains bitterly about too many IRQ 7s .. > That has nothing to do with the 2842. It means that you have some > card configured to the wrong IRQ or a card that is generating interrupts > but doesn't have a driver assigned to it. I only happened to notice it with -stable, prior to that the machine had been running 2.1-release without tagged-queueing. It's been running the same hardware since I received the 2.0.5 CD. However, I dismantled it this evening and the "stray irq 7" and spontaneous reboot problems went away as soon as I replaced the EtherLink III with a cruddy 8-bit WD8003-EBT. Given that the 3Com card was working fine until just recently (within the last week or so), I suspect that some software change has invited the instability :-( michael From owner-freebsd-stable Fri Jan 19 11:26:35 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id LAA06401 for stable-outgoing; Fri, 19 Jan 1996 11:26:35 -0800 (PST) Received: from brasil.moneng.mei.com (brasil.moneng.mei.com [151.186.109.160]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id LAA06395 Fri, 19 Jan 1996 11:26:31 -0800 (PST) Received: (from jgreco@localhost) by brasil.moneng.mei.com (8.7.Beta.1/8.7.Beta.1) id NAA24673; Fri, 19 Jan 1996 13:24:41 -0600 From: Joe Greco Message-Id: <199601191924.NAA24673@brasil.moneng.mei.com> Subject: Re: ccd driver or 2.1R available To: stable@silvia.HIP.Berkeley.EDU Date: Fri, 19 Jan 1996 13:24:40 -0600 (CST) Cc: hackers@FreeBSD.ORG, stable@FreeBSD.ORG, ccd@forgery.cs.berkeley.edu In-Reply-To: <199601141214.EAA01042@silvia.HIP.Berkeley.EDU> from "Satoshi Asami" at Jan 14, 96 04:14:40 am X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-stable@FreeBSD.ORG Precedence: bulk > Oh sure. One of them is "don't use a partition that starts at a > beginning of the slice". So please leave some space at the beginning > of the slice in the partitions you are combining (sd[1-4]g in the > above example). Of course, if someone can figure out why and fix it, > that will be great. I may be wrong, but isn't this because the standard disklabel leaves some space for the BIOS partition table, just in case it's the first slice on the disk? ... Joe ------------------------------------------------------------------------------- Joe Greco - Systems Administrator jgreco@ns.sol.net Solaria Public Access UNIX - Milwaukee, WI 414/342-4847 From owner-freebsd-stable Fri Jan 19 14:23:02 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id OAA18947 for stable-outgoing; Fri, 19 Jan 1996 14:23:02 -0800 (PST) Received: from haven.uniserve.com (haven.uniserve.com [198.53.215.121]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id OAA18919 Fri, 19 Jan 1996 14:22:39 -0800 (PST) Received: by haven.uniserve.com id <30753-4>; Fri, 19 Jan 1996 14:25:11 -0000 Date: Fri, 19 Jan 1996 14:25:01 -0800 (PST) From: Tom Samplonius To: Joe Greco cc: hackers@FreeBSD.ORG, stable@FreeBSD.ORG Subject: Re: ccd driver or 2.1R available In-Reply-To: <199601191924.NAA24673@brasil.moneng.mei.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@FreeBSD.ORG Precedence: bulk On Fri, 19 Jan 1996, Joe Greco wrote: > > Oh sure. One of them is "don't use a partition that starts at a > > beginning of the slice". So please leave some space at the beginning > > of the slice in the partitions you are combining (sd[1-4]g in the > > above example). Of course, if someone can figure out why and fix it, > > that will be great. > > I may be wrong, but isn't this because the standard disklabel leaves some > space for the BIOS partition table, just in case it's the first slice on the > disk? As far as I can tell from the manpage, the reason is that the ccd disklabel could be mistaken by the system as the disk's label, because they are in the same spot. The first bit of every partition has space reserved for the label, but ccd will not reserve space. I believe swap will not work as the first partition on a disk for the same reason, except that swap will overwrite the label! Tom From owner-freebsd-stable Fri Jan 19 19:29:09 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id TAA26408 for stable-outgoing; Fri, 19 Jan 1996 19:29:09 -0800 (PST) Received: from asstdc.scgt.oz.au (root@asstdc.scgt.oz.au [202.14.234.65]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id TAA26384 for ; Fri, 19 Jan 1996 19:29:00 -0800 (PST) Received: (from imb@localhost) by asstdc.scgt.oz.au (8.6.12/BSD4.4) id OAA11176; Sat, 20 Jan 1996 14:28:36 +1100 From: michael butler Message-Id: <199601200328.OAA11176@asstdc.scgt.oz.au> Subject: spontaneous reboot with -stable To: gibbs@freefall.freebsd.org (Justin T. Gibbs) Date: Sat, 20 Jan 1996 14:28:31 +1100 (EST) Cc: stable@freebsd.org In-Reply-To: <199601091544.HAA16694@freefall.freebsd.org> from "Justin T. Gibbs" at Jan 9, 96 07:44:32 am X-Mailer: ELM [version 2.4 PL24beta] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org Precedence: bulk I finally caught the error message from my 2842-fitted -stable machine that gives me so much trouble (reboots more than once a day :-(). It says .. panic: getnewbuf: inconsistent LRU queue, qindex=0 syncing ... Any suggestions ? michael From owner-freebsd-stable Fri Jan 19 20:35:02 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id UAA04370 for stable-outgoing; Fri, 19 Jan 1996 20:35:02 -0800 (PST) Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id UAA04348 Fri, 19 Jan 1996 20:34:57 -0800 (PST) Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.6.12/8.6.12) id UAA10983; Fri, 19 Jan 1996 20:33:25 -0800 From: "Rodney W. Grimes" Message-Id: <199601200433.UAA10983@GndRsh.aac.dev.com> Subject: Re: ccd driver or 2.1R available To: tom@uniserve.com (Tom Samplonius) Date: Fri, 19 Jan 1996 20:33:25 -0800 (PST) Cc: jgreco@brasil.moneng.mei.com, hackers@freebsd.org, stable@freebsd.org In-Reply-To: from "Tom Samplonius" at Jan 19, 96 02:25:01 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org Precedence: bulk > > > On Fri, 19 Jan 1996, Joe Greco wrote: > > > > Oh sure. One of them is "don't use a partition that starts at a > > > beginning of the slice". So please leave some space at the beginning > > > of the slice in the partitions you are combining (sd[1-4]g in the > > > above example). Of course, if someone can figure out why and fix it, > > > that will be great. > > > > I may be wrong, but isn't this because the standard disklabel leaves some > > space for the BIOS partition table, just in case it's the first slice on the > > disk? > > As far as I can tell from the manpage, the reason is that the ccd > disklabel could be mistaken by the system as the disk's label, because > they are in the same spot. The first bit of every partition has space > reserved for the label, but ccd will not reserve space. > > I believe swap will not work as the first partition on a disk for the > same reason, except that swap will overwrite the label! The swap code does not use the first 64 blocks or some such for that very reason. You can have swap at the front of a disk, though this has not always been true (I think it was David Greenman who finally fixed it). > Tom > -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Reliable computers for FreeBSD From owner-freebsd-stable Fri Jan 19 20:47:52 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id UAA05636 for stable-outgoing; Fri, 19 Jan 1996 20:47:52 -0800 (PST) Received: from rocky.sri.MT.net (rocky.sri.MT.net [204.182.243.10]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id UAA05625 Fri, 19 Jan 1996 20:47:48 -0800 (PST) Received: (from nate@localhost) by rocky.sri.MT.net (8.6.12/8.6.12) id VAA17390; Fri, 19 Jan 1996 21:50:27 -0700 Date: Fri, 19 Jan 1996 21:50:27 -0700 From: Nate Williams Message-Id: <199601200450.VAA17390@rocky.sri.MT.net> To: stable@FreeBSD.org, current@FreeBSD.sri.MT.net, hackers@FreeBSD.org Subject: Possible solution to the Proxy-ARP bug enclosed Sender: owner-stable@FreeBSD.org Precedence: bulk Sorry for cross-posting this to so many lists, but I wanted to get as much coverage as possible for this patch as possible. For those who are wondering what this is all about, there is a problem that affects all 4.4Lite derived systems with regards to ARP handling. (This includes FreeBSD, NetBSD, and BSDi.) The problem manifested itself most often when remote hosts couldn't re-establish broken PPP connections when using proxy-arp. Bill Fenner has graciously spent some time looking at the problem, and has sent me a patch which appears to have fixed my problem. Until it's tested further and others have had time to check it out it's not going in the tree, but for those folks (like me) who are being annoyed by this bug here is the patch Bill supplied me. I've tested this patch as best I can, but I'd like to get other folks to test this as well. Nate ---- [Standard disclaimers applies] *** in_rmx.c.orig Sun Jul 23 01:44:59 1995 --- in_rmx.c Fri Jan 19 14:33:57 1996 *************** *** 52,57 **** --- 52,58 ---- #include #include + #include #include #include #include *************** *** 77,82 **** --- 78,84 ---- { struct rtentry *rt = (struct rtentry *)treenodes; struct sockaddr_in *sin = (struct sockaddr_in *)rt_key(rt); + struct radix_node *ret; /* * For IP, all unicast non-host routes are automatically cloning. *************** *** 115,121 **** ? rt->rt_ifp->if_mtu : tcp_mssdflt + sizeof(struct tcpiphdr)); ! return rn_addroute(v_arg, n_arg, head, treenodes); } /* --- 117,148 ---- ? rt->rt_ifp->if_mtu : tcp_mssdflt + sizeof(struct tcpiphdr)); ! ret = rn_addroute(v_arg, n_arg, head, treenodes); ! if (ret == NULL && rt->rt_flags & RTF_HOST) { ! struct rtentry *rt2; ! /* ! * We are trying to add a host route, but can't. ! * Find out if it is because of an unresolved ! * ARP request and delete it if so. ! */ ! rt2 = rtalloc1((struct sockaddr *)sin, 0, ! RTF_CLONING | RTF_PRCLONING); ! if (rt2 && rt2->rt_flags & RTF_LLINFO && ! rt2->rt_flags & RTF_HOST && ! rt2->rt_gateway && ! rt2->rt_gateway->sa_family == AF_LINK && ! #define SDL(s) ((struct sockaddr_dl *)s) ! SDL(rt2->rt_gateway)->sdl_alen == 0) { ! #undef SDL ! rtrequest(RTM_DELETE, ! (struct sockaddr *)rt_key(rt2), ! rt2->rt_gateway, ! rt_mask(rt2), rt2->rt_flags, 0); ! RTFREE(rt2); ! ret = rn_addroute(v_arg, n_arg, head, treenodes); ! } ! } ! return ret; } /* From owner-freebsd-stable Fri Jan 19 21:17:01 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id VAA07255 for stable-outgoing; Fri, 19 Jan 1996 21:17:01 -0800 (PST) Received: from widget.xmission.com (root@widget.xmission.com [198.60.22.228]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id VAA07237 Fri, 19 Jan 1996 21:16:49 -0800 (PST) Received: (from rlenk@localhost) by widget.xmission.com (8.6.12/8.6.12) id WAA02402; Fri, 19 Jan 1996 22:16:24 -0700 From: Ron Lenk Message-Id: <199601200516.WAA02402@widget.xmission.com> Subject: Re: spontaneous reboot with -stable To: imb@scgt.oz.au (michael butler) Date: Fri, 19 Jan 1996 22:16:23 -0700 (MST) Cc: stable@freebsd.org, hackers@freebsd.org In-Reply-To: <199601200328.OAA11176@asstdc.scgt.oz.au> from "michael butler" at Jan 20, 96 02:28:31 pm X-Mailer: ELM [version 2.4 PL24 ME8a] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org Precedence: bulk > I finally caught the error message from my 2842-fitted -stable machine that > gives me so much trouble (reboots more than once a day :-(). It says .. > > panic: getnewbuf: inconsistent LRU queue, qindex=0 > syncing ... > > Any suggestions ? Yeah, don't run -stable. :-D I started running -stable again after Justin Gibbs commited the new ahc driver, and I have been unable to get a kernel to run for more than 5 minutes. I too have a 2842, and I have seen both total hang conditions ( ddb shows the processes sleeping on "newbuf" ), and panics about the various queues, both "inconsistent EMPTY queue", and "inconsistent LRU queue". A kernel from 2.1-RELEASE sources runs fine. I have exchanged mail with Justin Gibbs, and David Greenman about this, and haven't seen/heard anything in about 2 weeks. Obviously, I'm not the only person seeing this, and, in fact, one other person with an Adaptec 1742 under -stable was/is having the same problems as well. ( could this be something in the eisaconf code? ) Ron BTW: I apologize for sending a cc to hackers, but I've been unable to get any kind of response from anyone about this after the initial mail with Justin and David. ( A little pat on the head, along with "we're working on it" might be nice :) ) -- Ron Lenk -- rlenk@xmission.com From owner-freebsd-stable Fri Jan 19 22:50:05 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id WAA11020 for stable-outgoing; Fri, 19 Jan 1996 22:50:05 -0800 (PST) Received: from localhost.cdrom.com (localhost.cdrom.com [127.0.0.1]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id WAA11014 Fri, 19 Jan 1996 22:50:02 -0800 (PST) Message-Id: <199601200650.WAA11014@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: Host localhost.cdrom.com [127.0.0.1] didn't use HELO protocol To: Ron Lenk cc: imb@scgt.oz.au (michael butler), stable@freebsd.org, hackers@freebsd.org Subject: Re: spontaneous reboot with -stable In-reply-to: Your message of "Fri, 19 Jan 1996 22:16:23 MST." <199601200516.WAA02402@widget.xmission.com> Date: Fri, 19 Jan 1996 22:50:01 -0800 From: "Justin T. Gibbs" Sender: owner-stable@freebsd.org Precedence: bulk >I too have a 2842, and I have seen both total hang conditions ( ddb shows >the processes sleeping on "newbuf" ), and panics about the various queues, >both "inconsistent EMPTY queue", and "inconsistent LRU queue". A kernel >from 2.1-RELEASE sources runs fine. Did you upgrade your entire system (lkms too) to 2.1-STABLE, or did you only compile a kernel? >I have exchanged mail with Justin Gibbs, and David Greenman about this, and >haven't seen/heard anything in about 2 weeks. I have a day job. >Obviously, I'm not the only person seeing this, and, in fact, one other >person with an Adaptec 1742 under -stable was/is having the same problems >as well. ( could this be something in the eisaconf code? ) I don't know how the eisaconf code could cause this kind of problem. My guess is that its a side effect of the initialization order of devices. We now do: EISA PCI ISA and used to do: ISA EISA PCI You could try moving the call to eisa_configure() in i386/i386/autoconf.c after the initialization of ISA devices and see if that heps. >Ron > >BTW: I apologize for sending a cc to hackers, but I've been unable to get >any kind of response from anyone about this after the initial mail with >Justin and David. ( A little pat on the head, along with "we're working >on it" might be nice :) ) I'm in the process of creating a 2.1-STABLE bootable partition to help debug these issues, but I'm not positive that its just a -STABLE issue. Did you also say you had a 3c509 in the system? >-- >Ron Lenk -- rlenk@xmission.com -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-stable Fri Jan 19 23:22:05 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id XAA14342 for stable-outgoing; Fri, 19 Jan 1996 23:22:05 -0800 (PST) Received: from asstdc.scgt.oz.au (root@asstdc.scgt.oz.au [202.14.234.65]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id XAA14321 Fri, 19 Jan 1996 23:21:56 -0800 (PST) Received: (from imb@localhost) by asstdc.scgt.oz.au (8.6.12/BSD4.4) id SAA18912; Sat, 20 Jan 1996 18:21:20 +1100 From: michael butler Message-Id: <199601200721.SAA18912@asstdc.scgt.oz.au> Subject: Re: spontaneous reboot with -stable To: gibbs@freefall.freebsd.org (Justin T. Gibbs) Date: Sat, 20 Jan 1996 18:21:15 +1100 (EST) Cc: rlenk@widget.xmission.com, stable@freebsd.org, hackers@freebsd.org In-Reply-To: <199601200650.WAA11014@freefall.freebsd.org> from "Justin T. Gibbs" at Jan 19, 96 10:50:01 pm X-Mailer: ELM [version 2.4 PL24beta] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-stable@freebsd.org Precedence: bulk Justin T. Gibbs writes: > Did you upgrade your entire system (lkms too) to 2.1-STABLE, or did you > only compile a kernel? "make world" then made the kernel and installed the lot in one pass. I've tried all three combinations, with and without TAG_ENABLE and full queueing. > I'm in the process of creating a 2.1-STABLE bootable partition to help > debug these issues, but I'm not positive that its just a -STABLE issue. > Did you also say you had a 3c509 in the system? I had a 3c509 in the system, however, this panic occurs with or without it. It's currently got a (yuk! 8-bit) WD8003-EBT. That cured my other problem of "stray irq 7" but I haven't yet tried this with the -RELEASE kernel I've just installed to get the box going again. I'll do this after I come back from "Miss Saigon" this evening to see if that's -stable-specific as well, michael From owner-freebsd-stable Sat Jan 20 01:31:23 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id BAA27921 for stable-outgoing; Sat, 20 Jan 1996 01:31:23 -0800 (PST) Received: from Root.COM (implode.Root.COM [198.145.90.17]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id BAA27891 Sat, 20 Jan 1996 01:31:13 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by Root.COM (8.6.12/8.6.5) with SMTP id BAA10142; Sat, 20 Jan 1996 01:30:28 -0800 Message-Id: <199601200930.BAA10142@Root.COM> X-Authentication-Warning: implode.Root.COM: Host localhost didn't use HELO protocol To: Ron Lenk cc: imb@scgt.oz.au (michael butler), stable@freebsd.org, hackers@freebsd.org Subject: Re: spontaneous reboot with -stable In-reply-to: Your message of "Fri, 19 Jan 1996 22:16:23 MST." <199601200516.WAA02402@widget.xmission.com> From: David Greenman Reply-To: davidg@Root.COM Date: Sat, 20 Jan 1996 01:30:28 -0800 Sender: owner-stable@freebsd.org Precedence: bulk > panic: getnewbuf: inconsistent LRU queue, qindex=0 ... >BTW: I apologize for sending a cc to hackers, but I've been unable to get >any kind of response from anyone about this after the initial mail with >Justin and David. ( A little pat on the head, along with "we're working >on it" might be nice :) ) I know you're frustrated. I've been very busy with preparing for and doing an upgrade of wcarchive. I've been in the Bay area the last two days installing the new equipment and I just got back. What little time I've had while working on this the last 3 weeks has been spent on other bugs that were critical to wcarchive's stability...and I'm still working on this. If Justin Gibbs or John Dyson haven't fixed the problem for you after I've finished what I'm working on now, then I'll look into this again and try to fix the problem for you. -DG David Greenman Core Team/Principal Architect, The FreeBSD Project From owner-freebsd-stable Sat Jan 20 03:22:02 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id DAA17000 for stable-outgoing; Sat, 20 Jan 1996 03:22:02 -0800 (PST) Received: from iaehv.IAEhv.nl (root@iaehv.IAEhv.nl [192.87.208.2]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id DAA16991 for ; Sat, 20 Jan 1996 03:21:53 -0800 (PST) Received: from oasis.IAEhv.nl by iaehv.IAEhv.nl (8.6.12/1.63) id MAA26924; Sat, 20 Jan 1996 12:21:44 +0100 X-Disclaimer: iaehv.nl is a public access UNIX system and cannot be held responsible for the opinions of its individual users. Received: by oasis.IAEhv.nl (8.6.12/1.63) id MAA00915; Sat, 20 Jan 1996 12:20:40 +0100 From: volf@oasis.IAEhv.nl (Frank Volf) Message-Id: <199601201120.MAA00915@oasis.IAEhv.nl> Subject: How to keep up with -STABLE To: stable@freebsd.org Date: Sat, 20 Jan 1996 12:20:40 +0100 (MET) X-Mailer: ELM [version 2.4 PL25] Content-Type: text Sender: owner-stable@freebsd.org Precedence: bulk Hi, I installed the FreeBSD 2.1 release from the Walnut Creek CDROM. Now I would like to stay up to date with the FreeBSD stable tree. I know I can use ctm or sup to this. However, I can't find the appropriate configuration files to do so: on ftp.freebsd.org the FreeBSD-stable/ctm directory is emptye and the FreeBSD-stable/sup* files are not world readable. Does this mean that you can't ctm or sup them or should I look for the config files in another directory? Best regards, Frank ---------------------------------------------------------------------------- Frank Volf - Internet Access Eindhoven - Digitale Stad Eindhoven ---------------------------------------------------------------------------- || volf@oasis.IAEhv.nl - use for personal mail || || volf@IAEhv.nl - use for Internet Access Eindhoven related mail || || volf@dse.dse.nl - use for Digital City of Eindhoven related mail || ---------------------------------------------------------------------------- IAE Public Access Unix System - Dial +31.40.2439436 and login as new. ---------------------------------------------------------------------------- From owner-freebsd-stable Sat Jan 20 06:52:10 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id GAA28349 for stable-outgoing; Sat, 20 Jan 1996 06:52:10 -0800 (PST) Received: from irz301.inf.tu-dresden.de (irz301.inf.tu-dresden.de [141.76.1.11]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id GAA28328 Sat, 20 Jan 1996 06:52:03 -0800 (PST) Received: from sax.sax.de by irz301.inf.tu-dresden.de (8.6.12/8.6.12-s1) with ESMTP id PAA11387; Sat, 20 Jan 1996 15:51:52 +0100 Received: by sax.sax.de (8.6.11/8.6.12-s1) with UUCP id PAA00324; Sat, 20 Jan 1996 15:51:46 +0100 Received: (from j@localhost) by uriah.heep.sax.de (8.7.3/8.6.9) id KAA06747; Sat, 20 Jan 1996 10:51:57 +0100 (MET) From: J Wunsch Message-Id: <199601200951.KAA06747@uriah.heep.sax.de> Subject: Re: ccd driver or 2.1R available To: jgreco@brasil.moneng.mei.com (Joe Greco) Date: Sat, 20 Jan 1996 10:51:57 +0100 (MET) Cc: stable@silvia.HIP.Berkeley.EDU, hackers@freebsd.org, stable@freebsd.org, ccd@forgery.cs.berkeley.edu Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199601191924.NAA24673@brasil.moneng.mei.com> from "Joe Greco" at Jan 19, 96 01:24:40 pm X-Phone: +49-351-2012 669 X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-stable@freebsd.org Precedence: bulk As Joe Greco wrote: > > > Oh sure. One of them is "don't use a partition that starts at a > > beginning of the slice". > I may be wrong, but isn't this because the standard disklabel leaves some > space for the BIOS partition table, just in case it's the first slice on the > disk? No, the standard disklabel leaves some space for the disklabel. :) There used to be a problem with swapping all over the disklabel in earlier BSDs (including its derivatives like Illtrix), but this has been solved since. The only problem by now is (dunno whether it's already fixed) that if you configure your kernel with ``dumps on xxx0'', it will dump over the disklabel if the swap partition starts at offset 0. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-stable Sat Jan 20 09:21:01 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id JAA06404 for stable-outgoing; Sat, 20 Jan 1996 09:21:01 -0800 (PST) Received: from haven.uniserve.com (haven.uniserve.com [198.53.215.121]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id JAA06399 for ; Sat, 20 Jan 1996 09:20:58 -0800 (PST) Received: by haven.uniserve.com id <30753-3>; Sat, 20 Jan 1996 09:23:38 -0000 Date: Sat, 20 Jan 1996 09:23:29 -0800 (PST) From: Tom Samplonius To: Frank Volf cc: stable@freeBSD.org Subject: Re: How to keep up with -STABLE In-Reply-To: <199601201120.MAA00915@oasis.IAEhv.nl> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@freeBSD.org Precedence: bulk On Sat, 20 Jan 1996, Frank Volf wrote: > > Hi, > > I installed the FreeBSD 2.1 release from the Walnut Creek CDROM. Now I > would like to stay up to date with the FreeBSD stable tree. I know I can > use ctm or sup to this. However, I can't find the appropriate configuration > files to do so: on ftp.freebsd.org the FreeBSD-stable/ctm directory is > emptye and the FreeBSD-stable/sup* files are not world readable. Does > this mean that you can't ctm or sup them or should I look for the config > files in another directory? Yes. On YOUR systems look in /usr/share/examples/sup Tom From owner-freebsd-stable Sat Jan 20 09:37:12 1996 Return-Path: owner-stable Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id JAA08255 for stable-outgoing; Sat, 20 Jan 1996 09:37:12 -0800 (PST) Received: from haven.uniserve.com (haven.uniserve.com [198.53.215.121]) by freefall.freebsd.org (8.7.3/8.7.3) with ESMTP id JAA08249 for ; Sat, 20 Jan 1996 09:37:07 -0800 (PST) Received: by haven.uniserve.com id <30752-3>; Sat, 20 Jan 1996 09:39:53 -0000 Date: Sat, 20 Jan 1996 09:39:45 -0800 (PST) From: Tom Samplonius To: Ron Lenk cc: michael butler , stable@freeBSD.org Subject: Re: spontaneous reboot with -stable In-Reply-To: <199601200516.WAA02402@widget.xmission.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-stable@freeBSD.org Precedence: bulk On Fri, 19 Jan 1996, Ron Lenk wrote: > > I finally caught the error message from my 2842-fitted -stable machine that > > gives me so much trouble (reboots more than once a day :-(). It says .. > > > > panic: getnewbuf: inconsistent LRU queue, qindex=0 > > syncing ... > > > > Any suggestions ? > > Yeah, don't run -stable. :-D > > I started running -stable again after Justin Gibbs commited the new ahc > driver, and I have been unable to get a kernel to run for more than 5 > minutes. Not if you have a 2940. The new ahc driver is really, really nice in comparison to the old one. I'm now seeing record uptime's on a -stable system. Tom