From owner-freebsd-bugs Sun Apr 9 02:24:56 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id CAA13865 for bugs-outgoing; Sun, 9 Apr 1995 02:24:56 -0700 Received: (from joerg@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA13850 ; Sun, 9 Apr 1995 11:24:55 +0200 Date: Sun, 9 Apr 1995 11:24:55 +0200 From: Joerg Wunsch Message-Id: <199504090924.LAA13850@freefall.cdrom.com> To: lowell@epilogue.com, joerg, freebsd-bugs Subject: Changed information for PR gnu/210 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: tar support for rsh doesn't work State-Changed-From-To: open-closed State-Changed-By: joerg State-Changed-When: Sun Apr 9 11:21:29 MET DST 1995 State-Changed-Why: Among many other fixes (e.g. the /etc/rmt symlink), the latest change to the rmt utility to create new files with specific permission bits instead of random garbage seems to have killed the last show-stopper for this. From owner-freebsd-bugs Sun Apr 9 05:25:02 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id FAA19304 for bugs-outgoing; Sun, 9 Apr 1995 05:25:02 -0700 Received: from zap.zap.qc.ca (zap.zap.qc.ca [198.168.127.8]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id FAA19294 for ; Sun, 9 Apr 1995 05:24:56 -0700 Received: (from fortin@localhost) by zap.zap.qc.ca (8.6.9/8.6.6) id IAA01727; Sun, 9 Apr 1995 08:24:41 -0400 From: Denis Fortin Message-Id: <199504091224.IAA01727@zap.zap.qc.ca> Subject: Re: bin/308: Do-it-yourself FTP needs to prompt for what to download FDIV001 To: uhclem%nemesis@fw.ast.com Date: Sun, 9 Apr 1995 08:24:40 -0400 (EDT) Cc: freebsd-bugs@freefall.cdrom.com In-Reply-To: <199504090550.WAA03910@freefall.cdrom.com> from "uhclem%nemesis@fw.ast.com" at Apr 8, 95 10:50:01 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 304 Sender: bugs-owner@FreeBSD.org Precedence: bulk > Longer term, I also suggest putting a prefix of some sort on files with > common names, such as CKSUMS (src.CKSUMS) That *would* be nice! -- Denis Fortin fortin@acm.org DMR Group Inc, (514) 877-3301 These opinions are my own From owner-freebsd-bugs Sun Apr 9 09:10:02 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id JAA25770 for bugs-outgoing; Sun, 9 Apr 1995 09:10:02 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id JAA25763 ; Sun, 9 Apr 1995 09:10:01 -0700 Date: Sun, 9 Apr 1995 09:10:01 -0700 Message-Id: <199504091610.JAA25763@freefall.cdrom.com> From: pritc003@maroon.tc.umn.edu Reply-To: pritc003@maroon.tc.umn.edu To: freebsd-bugs Subject: kern/312: link system call on msdos filesystem causes panic In-Reply-To: Your message of Sun, 9 Apr 1995 11:01:57 -0500 <199504091601.LAA00372@mpp.com> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 312 >Category: kern >Synopsis: link system call on msdos filesystem causes panic >Confidential: no >Severity: serious >Priority: high >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 09:10:01 1995 >Originator: Mike Pritchard >Organization: >Release: FreeBSD 2.1.0-Development i386 >Environment: >Description: If a program does a link("file1", "file2") system call followed by an unlink("file1") system call on a msdos file system, the system will panic with the following message: panic: msdosfs_lock: locking against myself This is because the msdosfs link code does not vput the vnode like it should before returning. The link code should also return an error to the user so that the calling program doesn't accidently destroy the file it was attempting to link, thinking that the link() call actually worked. >How-To-Repeat: Compile and run the following program on a msdos file system: #include #include main() { int fd; if ((fd = open("testfile", O_RDWR | O_CREAT, 0666)) < 0) { perror("open"); exit(1); } if (link("testfile", "newfile") < 0) { perror("link"); exit(1); } if (unlink("testfile") < 0) { perror("unlink"); exit(1); } close(fd); } >Fix: The attached fix corrects the msdosfs_link() routine to correctly vput the vnode before returning, and return an error of EMLINK (too many links) to the user. Feel free to choose another errno if you think something else is more appropriate. ENODEV would be my next choice. *** msdosfs/orig2/msdosfs_vnops.c Sun Apr 9 10:29:49 1995 --- msdosfs/msdosfs_vnops.c Sun Apr 9 10:34:30 1995 *************** *** 907,913 **** struct componentname *a_cnp; } */ *ap; { ! return VOP_ABORTOP(ap->a_vp, ap->a_cnp); } /* --- 907,915 ---- struct componentname *a_cnp; } */ *ap; { ! VOP_ABORTOP(ap->a_vp, ap->a_cnp); ! vput(ap->a_vp); ! return EMLINK; } /* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 10:59:41 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id KAA28151 for bugs-outgoing; Sun, 9 Apr 1995 10:59:41 -0700 Received: from goof.com (goof.com [198.82.204.15]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id KAA28145 ; Sun, 9 Apr 1995 10:59:39 -0700 Received: (from mmead@localhost) by goof.com (8.6.11/8.6.9) id NAA00394; Sun, 9 Apr 1995 13:59:31 -0400 From: "matthew c. mead" Message-Id: <199504091759.NAA00394@goof.com> Subject: kernel supped 4/10 1pm est To: current@FreeBSD.org Date: Sun, 9 Apr 1995 13:59:31 -0400 (EDT) Cc: bugs@FreeBSD.org X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 408 Sender: bugs-owner@FreeBSD.org Precedence: bulk The latest kernel breaks pidentd (it hangs). It appears the port needs to be fixed, too... -matt -- Matthew C. Mead -> Virginia Tech Center for Transportation Research - -> Multiple Platform System and Network Administration Work Related -> mmead@ctr.vt.edu | mmead@goof.com <- All Other ---- ------- WWW -> http://www.goof.com/~mmead --- ----- From owner-freebsd-bugs Sun Apr 9 13:20:17 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01322 for bugs-outgoing; Sun, 9 Apr 1995 13:20:17 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01315 ; Sun, 9 Apr 1995 13:20:15 -0700 Date: Sun, 9 Apr 1995 13:20:15 -0700 Message-Id: <199504092020.NAA01315@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: bin/316: SNAP950322 less stable on IDE than earlier releases FDIV009 In-Reply-To: Your message of Sun, 9 Apr 95 13:00 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 316 >Category: bin >Synopsis: SNAP950322 less stable on IDE than earlier releases FDIV009 >Confidential: no >Severity: critical >Priority: high >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:20:14 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: system is a GRiD/DEC 486DX-33 EISA system with: one WD Caviar 528Meg IDE drive, OAK SVGA, SMC 8013 network adapter, Soundblaster 16 with two Matsushita CD-ROM drives. (can be removed - no effect) 8 Meg of RAM. The hard disk was completely wiped during multiple installations. Kernel can be the stock kernel, a "make world" kernel, or a custom kernel. >Description: [FDIV009] SNAP-032295 is less stable than the previous SNAP. I get system lockups during kernel builds, ls -alR | more, even grep commnds. This same system ran SNAP-021095 for a month without problems. The system also ran 2.0 and 1.1.5.1 without problems. When I rebuild the kernel, typically three to six drivers get recompiled. During this process, the system will lockup an average of one time per build. I have had to restart the same build three times to make it all the way through, and then ten minutes later a recompile of the same files goes through without incident. The lockup is just that - no panic. The system simply goes dead, no activity, Numlock and Capslock no longer toggle when you push keys. Pings are not recognized. I have removed and reseated RAM, processor, all cards, and finally replaced all the RAM and the problem doesn't go away. I finally moved the hard disk to a similar system (almost identical configuration that has been running SNAP-021095 since it was available with no problems) and now the other system randomly lockups. I assume it is either a memory management problem or a problem with the wd disk driver. It will not lockup if the system isn't doing anything. It has to be crunching to cause the problem to show itself. In fact, I wrote this report by telneting from a SNAP-032295 system to a 1.1.5.1 system that was otherwise idle. While typing this note, cron ran /etc/daily on the 032295 system and locked the system up. Nothing else was running and there are no other users. >How-To-Repeat: Here it is easy. Give the system something disk-intensive to do. >Fix: *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 13:20:10 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01283 for bugs-outgoing; Sun, 9 Apr 1995 13:20:10 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01273 ; Sun, 9 Apr 1995 13:20:07 -0700 Date: Sun, 9 Apr 1995 13:20:07 -0700 Message-Id: <199504092020.NAA01273@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: bin/313: BACKSPACE key not producing 0x08 breaks many things FDIV007 In-Reply-To: Your message of Sun, 9 Apr 95 12:41 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 313 >Category: bin >Synopsis: BACKSPACE key not producing 0x08 breaks many things FDIV007 >Confidential: no >Severity: serious >Priority: medium >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:20:05 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: Appeared in SNAP-032295 >Description: [FDIV007] The character generated by the BACKSPACE key was changed in this SNAP despite a lengthy discussion that I thought said it would not be changed. Anyway. During installation, neither BACKSPACE nor DELETE produces an ERASE operation. As soon as you touch a shell for the first time, the BACKSPACE is now 0x7f, not 0x08. This makes connecting to other systems via telnet or cu a major hassle since most other systems accept BACKSPACE as 0x08. And if you pop into the command layer of either, now the BACKSPACE key is backwards from what was working on the session. Locally, even if you stty erase ^?, vi still only accepts the BACKSPACE key as a ERASE operation in Insert mode. The BACKSPACE key no longer can be used for cursor motions, something that has worked for over a decade. CTRL-BACKSPACE produces 0x08 which is accepted a cursor motion, but this is a poor substitute, BUT CTRL-BACKSPACE isn't accepted in INSERT mode. Other applications (like prompts in tin/trn) have the same dual-personality problem that they did not have before. I also can't get BACKSPACE to be accepted anywhere in X no matter what I set stty to. Oh and boot single user. Even the stty "what you have set" message has trouble dealing with BACKSPACE <> 0x08. Instead of ^H, or ^? you get a little stick-house symbol. >How-To-Repeat: For VI Login stty erase ^H vi /.profile Ainsert test note backspacing works now backspacing doesn't work. You now have to use CTRL-BACKSPACE to do cursor motions. For TELNET/CU, make connection, connect to SCO XENIX, SCO UNIX, FreeBSD 1.1.5.1, SUNos, and probably others. Try to get backspace recognized by changing stty settings. Now ~ in cu a command or ^] and prompt in telnet. For X try typing text and backspaces into anything. For tin, try to subscribe to something and try to correct typing errors. Single user shell is easy - kernel -s. Select "sh". >Fix: Make the BACKSPACE key produce the 0x08 character like it was before. This change just breaks too many applications and threatens our ability to run binaries from other platforms. On PCs, the BACKSPACE key is supposed to be 0x08 and some programs have that notion hard-wired in. OK, it's bad programming practice, but making SCO and other executables malfunction on this user-obvious point when we are trying to get them to work is not a good way to win converts. If we are serious about allowing other i386 executables run on FreeBSD, we should try not to deviate from the established norms for that platform. *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 13:20:14 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01309 for bugs-outgoing; Sun, 9 Apr 1995 13:20:14 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01301 ; Sun, 9 Apr 1995 13:20:13 -0700 Date: Sun, 9 Apr 1995 13:20:13 -0700 Message-Id: <199504092020.NAA01301@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: bin/315: Kernel and boot block don't agree on kernel name FDIV010 In-Reply-To: Your message of Sun, 9 Apr 95 13:07 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 315 >Category: bin >Synopsis: Kernel and boot block don't agree on kernel name FDIV010 >Confidential: yes >Severity: serious >Priority: high >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:20:11 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: SNAP-950322 >Description: [FDIV010] The boot block looks for a kernel named 386bsd (or 386bsd.old) rather than kernel*. At first I thought I had downloaded the wrong floppy images, but a second download of the files in SNAP-032295/floppies resulted in the same problems. SNAP-021095 did not have this problem and searched for "kernel". >How-To-Repeat: >Fix: Rename kernel or link to 386bsd will get around the problem of autobooting. Hopefully we will switch back to the boot blocks we had in SNAP-950210 which thought the kernel was caller "kernel". *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 13:20:11 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01295 for bugs-outgoing; Sun, 9 Apr 1995 13:20:11 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01287 ; Sun, 9 Apr 1995 13:20:10 -0700 Date: Sun, 9 Apr 1995 13:20:10 -0700 Message-Id: <199504092020.NAA01287@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: bin/314: CTRL-Z during installation crashes install FDIV008 In-Reply-To: Your message of Sun, 9 Apr 95 12:50 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 314 >Category: bin >Synopsis: CTRL-Z during installation crashes install FDIV008 >Confidential: no >Severity: serious >Priority: low >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:20:08 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: SNAP-032295 >Description: [FDIV008] During installation, if you press CTRL-Z, the installation aborts, reports something about abnormal child termination and immediately reboots the system. (The message doesn't stay on the screen long enough to read.) >How-To-Repeat: In my case I selected do-it-yourself FTP and while at the FTP prompt I hit CTRL-Z. I also believe I did it at one of the installation prompts, such as "Do you want to delete /usr/tmp?" >Fix: This signal should either deliver a shell (when possible) or be trapped and ignored. *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 13:20:28 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01353 for bugs-outgoing; Sun, 9 Apr 1995 13:20:28 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01345 ; Sun, 9 Apr 1995 13:20:25 -0700 Date: Sun, 9 Apr 1995 13:20:25 -0700 Message-Id: <199504092020.NAA01345@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: bin/318: matcd driver not on boot or fixit floppy FDIV012 In-Reply-To: Your message of Sun, 9 Apr 95 13:15 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 318 >Category: bin >Synopsis: matcd driver not on boot or fixit floppy FDIV012 >Confidential: no >Severity: serious >Priority: medium >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: change-request >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:20:21 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: FreeBSD 2.0.0-SNAP950322 i386 >Description: [FDIV012] The kernel on the boot floppy does not have the matcd driver present. I have been told (but have not personally checked) that the fixit disk also doesn't have matcd. This should be fixed on at least the fixit disk so that files can be recovered from the distribution CD-ROM if needed using the fixit disk. >How-To-Repeat: >Fix: *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 13:20:24 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01344 for bugs-outgoing; Sun, 9 Apr 1995 13:20:24 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01331 ; Sun, 9 Apr 1995 13:20:19 -0700 Date: Sun, 9 Apr 1995 13:20:19 -0700 Message-Id: <199504092020.NAA01331@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: docs/317: Install README doesn't list matcd as install choice FDIV011 In-Reply-To: Your message of Sun, 9 Apr 95 13:12 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 317 >Category: docs >Synopsis: Install README doesn't list matcd as install choice FDIV011 >Confidential: no >Severity: non-critical >Priority: medium >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: doc-bug >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:20:17 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: FreeBSD 2.0.0-SNAP950322 i386 >Description: [FDIV011] The installation README says that the Creative CD-ROM cannot be used for installation, but this should be possible in SNAP-032295 and later. >How-To-Repeat: >Fix: *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 13:20:33 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01367 for bugs-outgoing; Sun, 9 Apr 1995 13:20:33 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01359 ; Sun, 9 Apr 1995 13:20:29 -0700 Date: Sun, 9 Apr 1995 13:20:29 -0700 Message-Id: <199504092020.NAA01359@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: bin/319: Install doesn't strip sys name in prompt for domain FDIV013 In-Reply-To: Your message of Sun, 9 Apr 95 13:18 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 319 >Category: bin >Synopsis: Install doesn't strip sys name in prompt for domain FDIV013 >Confidential: no >Severity: non-critical >Priority: low >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:20:27 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: FreeBSD 2.0.0-SNAP950322 i386 >Description: [FDIV013] When doing FTP download, it asks for the machines fully-qualified domain name. I typed: dalek.lonestar.org The next question asked what my domain name was, and it filled the box in with: dalek.lonestar.org| with the cursor on the right. Why didn't the install at least delete the left-most field off the fully-qualified domain name and offer that as a prompt? >How-To-Repeat: Install system and select that you want to do a FTP install. >Fix: *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 13:20:37 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01387 for bugs-outgoing; Sun, 9 Apr 1995 13:20:37 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01373 ; Sun, 9 Apr 1995 13:20:34 -0700 Date: Sun, 9 Apr 1995 13:20:34 -0700 Message-Id: <199504092020.NAA01373@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: bin/320: TCP/IP settings entered b4 install are not kept FDIV014 In-Reply-To: Your message of Sun, 9 Apr 95 13:24 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 320 >Category: bin >Synopsis: TCP/IP settings entered b4 install are not kept FDIV014 >Confidential: no >Severity: non-critical >Priority: medium >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:20:33 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: FreeBSD 2.0.0-SNAP950322 i386 >Description: [FDIV014] When doing FTP download, the installation procedire asks for the machines domain name, system name, IP number and other information prior to downloading modules. After downloads are complete, the installation procedure offers to finish configuring TCP/IP settings. If you select this, it asks the same questions again and has discarded the answers from the previous questions. (There is a related problem [FDIV015] where many of these values are lost anyway when the system reboots normally at the end of the install.) >How-To-Repeat: Install the system and select FTP download as the method. After installation is complete and you are offered the chance to finish configuring TCP/IP, select that choice. Note that it has forgotten the information you gave it earlier. >Fix: I suggest that the second set of questions not even be offered as a choice if the information was provided prior to downloads. Alternatively, all the boxes in the second pass should have the answers from the first configuration session as the defaults. *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 13:20:41 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01395 for bugs-outgoing; Sun, 9 Apr 1995 13:20:41 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01385 ; Sun, 9 Apr 1995 13:20:36 -0700 Date: Sun, 9 Apr 1995 13:20:36 -0700 Message-Id: <199504092020.NAA01385@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: bin/321: TCP/IP settings given during install are not kept FDIV015 In-Reply-To: Your message of Sun, 9 Apr 95 13:29 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 321 >Category: bin >Synopsis: TCP/IP settings given during install are not kept FDIV015 >Confidential: no >Severity: serious >Priority: medium >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:20:35 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: FreeBSD 2.0.0-SNAP950322 i386 >Description: [FDIV015] When doing FTP download, the installation procedure asks for the machines domain name, system name, IP number and other information prior to downloading modules. After extracting modules you may also specify details of your domain, system name and IP information a second time (See FDIV014). But when you reboot, your system name is still "myname.mydomain" instead of being set based on the information given during the installation. >How-To-Repeat: Install the system and select the FTP download option. Enter all domain, system name, IP number, gateway, DNS and other information both times. Then let the system reboot normally. As the system boots, note the system name is myname.mydomain and that other IP-related files do not contain the settings you provided and these must now be entered a second (or third) time. >Fix: The hosts and sysconfig files should be updated with the information provided during the installation. *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 13:20:45 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01411 for bugs-outgoing; Sun, 9 Apr 1995 13:20:45 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01401 ; Sun, 9 Apr 1995 13:20:42 -0700 Date: Sun, 9 Apr 1995 13:20:42 -0700 Message-Id: <199504092020.NAA01401@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: bin/322: No way to restart a failed FTP download FDIV016 In-Reply-To: Your message of Sun, 9 Apr 95 13:37 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 322 >Category: bin >Synopsis: No way to restart a failed FTP download FDIV016 >Confidential: no >Severity: serious >Priority: high >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:20:40 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: FreeBSD 2.0.0-SNAP950322 i386 >Description: [FDIV016] During the installation process, if you select FTP downloads and the transfer of a files for one reason or another, there is no way to recover other than completely starting the install process over. In my case, I was down to one of the last of the files that makes up the bin distribution (although this could happen during any distribution) and I the remote system apparently crashed. I got: 421 Service not available, remote server has close connection 98304 bytes received in 8.4e+02 seconds (0.11 kbytes/sec) Continue with mget? Of course you can't answer the question. If you press CTRL-C, you get: User interrupted Aborting installation and the system immediately reboots. If you answer "n" the system takes off and tries to extract with what it has (which should fail). >How-To-Repeat: Umm, ok, start an install, select FTP download, and during the transfers unplug your network cable. That ought to produce errors similar to the problem I encountered. >Fix: There needs to be a way to restart an entire download if something goes wrong without having to restart the entire installation process. *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 13:20:49 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01425 for bugs-outgoing; Sun, 9 Apr 1995 13:20:49 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01415 ; Sun, 9 Apr 1995 13:20:45 -0700 Date: Sun, 9 Apr 1995 13:20:45 -0700 Message-Id: <199504092020.NAA01415@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: bin/323: Creating lost+found causes fsck to stop fsck -y FDIV017 In-Reply-To: Your message of Sun, 9 Apr 95 13:44 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 323 >Category: bin >Synopsis: Creating lost+found causes fsck to stop fsck -y FDIV017 >Confidential: no >Severity: serious >Priority: medium >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:20:43 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: FreeBSD 2.0.0-SNAP950322 i386 >Description: [FDIV017] The first time FSCK runs after an installation and finds an unreferenced file or directory, it creates lost+found. This is OK. However, fsck then reports: LINK COUNT INCREASING (duh) UNEXPECTED INCONSISTENCY; RUN FSCK MANUALLY >How-To-Repeat: Cause a non-zero-length file or directory to be unreferences and run fsck on a partition that doesn't have a lost+found. >Fix: I suggest that fsck should consider the inode (and associated links) it used to create lost+found to not be an inconsistency. *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 13:20:52 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01439 for bugs-outgoing; Sun, 9 Apr 1995 13:20:52 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01429 ; Sun, 9 Apr 1995 13:20:49 -0700 Date: Sun, 9 Apr 1995 13:20:49 -0700 Message-Id: <199504092020.NAA01429@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: bin/324: /usr/include/sys -> /usr/sys/sys -> /usr/sys/sys FDIV018 In-Reply-To: Your message of Sun, 9 Apr 95 13:54 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 324 >Category: bin >Synopsis: /usr/include/sys -> /usr/sys/sys -> /usr/sys/sys FDIV018 >Confidential: no >Severity: serious >Priority: high >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:20:47 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: FreeBSD 2.0.0-SNAP950322 i386 >Description: [FDIV018] Hopefully everybody has noticed this by now. After install, /usr/include/sys ends up as a link to /usr/sys/sys, which is a link to /usr/sys/sys. The files that used to be in /usr/include/sys (or its symlink) appear to be erased. Make world cannot be done with the links like this. If you correct this problem, "make world" will try to change it near the start of the make "world process" back to the incorrect linking and wlll change it again a second time at the end of the "make world" process. Make sure you fix both attempts to scramble links in the "make world" process. >How-To-Repeat: Install complete system. Try to do a "make world". After twenty minutes or so it will fail and stop. When that fails, rm /usr/include/sys and re-extract that directory manually from the distrbution (if you kept it around). Now make tar of /usr/include/sys and keep it safe. You will need it. Start "make world" again. About fifteen minutes into the process before it actually starts compiling anything, verify that /usr/include/sys is still intact. Keep checking every few minutes until /usr/include/sys turns back into the mangled links. Now ^Z the make. Extract the tar you made earlier into /usr/include/sys and let the make world resume. "Make world" will now complete correctly with no fatal errors. After "make world" finishes, look at /usr/include/sys and you will find the files are gone again and the crazy links are back. Re-extact the tar you made of /usr/include/sys if you feel energetic. >Fix: *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 13:20:55 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01452 for bugs-outgoing; Sun, 9 Apr 1995 13:20:55 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01443 ; Sun, 9 Apr 1995 13:20:53 -0700 Date: Sun, 9 Apr 1995 13:20:53 -0700 Message-Id: <199504092020.NAA01443@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: bin/325: boot -s SH is confused by BACKSPACE <> 0x08 FDIV019 In-Reply-To: Your message of Sun, 9 Apr 95 13:58 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 325 >Category: bin >Synopsis: boot -s SH is confused by BACKSPACE <> 0x08 FDIV019 >Confidential: no >Severity: non-critical >Priority: low >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:20:51 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: FreeBSD 2.0.0-SNAP950322 i386 >Description: [FDIV019] When you tell the system to boot -s, it comes up and says: Enter pathname of shell or RETURN for sh: If you press RETURN, you get: erase *, kill ^U, intr ^C where * is a symbol that looks like a stick drawing of a house. This is yet-another symptom of erase not being set to ^H and the BACKSPACE key not generating ^H and programs not coping with it. >How-To-Repeat: >Fix: Return erase character to being 0x08 and make the BACKSPACE key produce a 0x08. *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 13:21:00 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01470 for bugs-outgoing; Sun, 9 Apr 1995 13:21:00 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01458 ; Sun, 9 Apr 1995 13:20:57 -0700 Date: Sun, 9 Apr 1995 13:20:57 -0700 Message-Id: <199504092020.NAA01458@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: bin/326: Weekly cron generates some usage and error messages FDIV021 In-Reply-To: Your message of Sun, 9 Apr 95 14:13 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 326 >Category: bin >Synopsis: Weekly cron generates some usage and error messages FDIV021 >Confidential: no >Severity: non-critical >Priority: medium >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:20:55 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: FreeBSD 2.0.0-SNAP950322 i386 All distributions except X present. >Description: [FDIV021] When weekly cron runs on a system that has everything except X11 installed, you get mail that says: weekly run output Rotating messages: *** cat: /var/run/syslog.pid: no such file or directory *** usage: kill [-l] [-sig] pid... Rotating cron log: *** /usr/X11R6/man/whatis.tmp no such file or directory >How-To-Repeat: Install system. I downloaded all but X, but this may not be a requirement. Allow cron to run /etc/weekly. >Fix: *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 13:21:04 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01490 for bugs-outgoing; Sun, 9 Apr 1995 13:21:04 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01474 ; Sun, 9 Apr 1995 13:21:01 -0700 Date: Sun, 9 Apr 1995 13:21:01 -0700 Message-Id: <199504092021.NAA01474@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: bin/327: Clock management punishes you if CMOS != GMT FDIV020 In-Reply-To: Your message of Sun, 9 Apr 95 14:10 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 327 >Category: bin >Synopsis: Clock management punishes you if CMOS != GMT FDIV020 >Confidential: no >Severity: serious >Priority: high >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:20:59 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: FreeBSD 2.0.0-SNAP950322 i386 >Description: [FDIV020] The installation procedure seems to be written assuming you are going to set the CMOS clock to GMT, which is the least likely time choice for PC users. This is because people who use the system for MS-DOS, Windows and other things on other partitions (or from floppy) must keep the clock correct to local time or those systems screw up. If they run certain networking software on those systems, all they have to boot and the CMOS is instantly changed to the correct local time, which can screw-up things under FreeBSD later. If you elect to keep the CMOS at local time, the installation will offer you cities (not many choices in CST by the way), then show you a time with the CST suffix, WITH THE GMT OFFSET ADDED even if you don't want it. It then asks "Is this what you wanted?". The normal user isn't going to realize that all the installation process is asking about at that point is the CST suffix and will answer the question NO because the time is wrong. The user has to answer that question yes and then select 98 (CMOS isn't GMT) later to end up with a correct time in FreeBSD multi-user mode.* This part of the installation is pretty user-unfriendly and needs improvement or at least better instructions. * Also note that if you have CMOS set to LOCAL, and boot the system in maint mode, the date shown is wrong (behind by several hours). If you change it to be correct, and then boot multi-user, the date is now ahead of where it should be by several hours. This is pretty confusing. >How-To-Repeat: Using CMOS setup, set CMOS to correct local time (or something non-GMTish if you happen to live in GMT), save, install FreeBSD, select LOCAL time. Now boot multi-user. Verify time is what you set CMOS to. Now halt and reboot single-user. Do date command and note that the time displayed is hours off (slow.) So, set time using date command to be correct. Sync. Now re-boot in multi-user mode and note that now time in multi-user mode is several hours fast of correct time and many more hours ahead of time just set in single-user mode. Now halt and run CMOS setup and see what the clock is set to compared to what you originally set. >Fix: Improve clock management for people who set CMOS to non-GMT, for people who can't afford an 100% exclusive FreeBSD system as they are likely to set the CMOS to the local time. *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 13:21:07 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01504 for bugs-outgoing; Sun, 9 Apr 1995 13:21:07 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01491 ; Sun, 9 Apr 1995 13:21:04 -0700 Date: Sun, 9 Apr 1995 13:21:04 -0700 Message-Id: <199504092021.NAA01491@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: bin/328: Checking for -c kernel changes randomly hangs FDIV023 In-Reply-To: Your message of Sun, 9 Apr 95 14:22 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 328 >Category: bin >Synopsis: Checking for -c kernel changes randomly hangs FDIV023 >Confidential: no >Severity: serious >Priority: high >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:21:02 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: FreeBSD 2.0.0-SNAP950322 i386 Stock kernel, "make world" kernel, or custom kernel. >Description: [FDIV023] >From time to time during the normal boot process, it will hang immediately after the message: check for kernel -c changes After waiting a minute or so to make sure it is hung, if you press CTRL-C, the system will proceed with: clearing /tmp and continue to boot normally. This may be related to booting a kernel named something other than /kernel, but it seemed more random than that. As mentioned above, I tried several different kernels (including the distribution "stocl" kernel) and tried naming them different things and it didn't seem to matter. There didn't seem to be a pattern to the failures. >How-To-Repeat: Let the system boot normally, although you have to specify a kernel name (due to boot block bug). If it doesn't hang, shut down and try again. It would hang probably four or five times out of ten boots. >Fix: *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 13:21:09 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01517 for bugs-outgoing; Sun, 9 Apr 1995 13:21:09 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA01508 ; Sun, 9 Apr 1995 13:21:08 -0700 Date: Sun, 9 Apr 1995 13:21:08 -0700 Message-Id: <199504092021.NAA01508@freefall.cdrom.com> From: uhclem%nemesis@fw.ast.com Reply-To: uhclem%nemesis@fw.ast.com To: freebsd-bugs Subject: bin/329: FTP transfers above 99K shown in scientific notation FDIV022 In-Reply-To: Your message of Sun, 9 Apr 95 14:17 CDT Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 329 >Category: bin >Synopsis: FTP transfers above 99K shown in scientific notation FDIV022 >Confidential: no >Severity: non-critical >Priority: low >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: change-request >Submitter-Id: current-users >Arrival-Date: Sun Apr 9 13:21:06 1995 >Originator: Frank Durda IV >Organization: >Release: FreeBSD 2.0.0-SNAP950322 i386 >Environment: FreeBSD 2.0.0-SNAP950322 i386 >Description: [FDIV022] When FTP completes a transfer that runs faster the 99K/sec, it displays the results in scientific notation. Although people who have taken (and remember) chemistry and related scientific courses, the average user has not. I got a call from someone I support who thought the system was broken because it was displaying letters in numbers (ie "e"). Can't we fix this so that it doesn't switch to scientific notation that average people don't understand? I note that SCO and NCSA telnet packages have done this. >How-To-Repeat: FTP something from a system that you can achieve a >99K/sec transfer rate on. >Fix: Change a printf? *END* >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sun Apr 9 14:51:11 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id OAA17479 for bugs-outgoing; Sun, 9 Apr 1995 14:51:11 -0700 Received: from irz301.inf.tu-dresden.de (irz301.inf.tu-dresden.de [141.76.1.11]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id OAA17464 for ; Sun, 9 Apr 1995 14:50:59 -0700 Received: from sax.sax.de by irz301.inf.tu-dresden.de with SMTP (5.67b+/DEC-Ultrix/4.3) id AA12090; Sun, 9 Apr 1995 23:50:15 +0200 Received: by sax.sax.de (8.6.9/8.6.9-s1) with UUCP id XAA19371; Sun, 9 Apr 1995 23:50:13 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.6.11/8.6.9) id XAA08977; Sun, 9 Apr 1995 23:47:47 +0200 From: J Wunsch Message-Id: <199504092147.XAA08977@uriah.heep.sax.de> Subject: Re: bin/313: BACKSPACE key not producing 0x08 breaks many things FDIV007 To: uhclem%nemesis@fw.ast.com Date: Sun, 9 Apr 1995 23:47:47 +0200 (MET DST) Cc: freebsd-bugs@freefall.cdrom.com In-Reply-To: <199504092020.NAA01273@freefall.cdrom.com> from "uhclem%nemesis@fw.ast.com" at Apr 9, 95 01:20:07 pm Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) 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 Content-Length: 651 Sender: bugs-owner@FreeBSD.org Precedence: bulk As uhclem%nemesis@fw.ast.com wrote: > > > >Number: 313 > >Category: bin > >Synopsis: BACKSPACE key not producing 0x08 breaks many things FDIV007 I understand that you've been annoyed by the new behaviour (i've been annoyed by the previous behaviour :-], but that doesn't matter). However, could you please file only ONE bug report for one problem? Søren has already reverted the defaults to the same like 1.1.5.1 (and 2.0) were, so i assume it's safe to close both your reports now? -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ Never trust an operating system you don't have sources for. ;-) From owner-freebsd-bugs Sun Apr 9 15:12:35 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id PAA18576 for bugs-outgoing; Sun, 9 Apr 1995 15:12:35 -0700 Received: from tom.compulink.co.uk (tom.compulink.co.uk [192.188.69.2]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id PAA18568 for ; Sun, 9 Apr 1995 15:12:30 -0700 Received: from gonzales.compulink.co.uk (gonzales.compulink.co.uk [192.188.69.4]) by tom.compulink.co.uk (8.6.9/8.6.9) id XAA14591 for bugs@freebsd.org; Sun, 9 Apr 1995 23:10:39 +0100 Date: Sun, 9 Apr 95 23:09 BST-1 From: geg@cix.compulink.co.uk (Gordon Glasgow) Subject: CD install fail To: bugs@FreeBSD.org Reply-To: geg@cix.compulink.co.uk Message-Id: Sender: bugs-owner@FreeBSD.org Precedence: bulk My install of FreeBSD 2.0 failed. I followed the install instructions on the CD and created the 2 Diskettes. The diskette boot worked ok. I set up slices as below wd0 a: 50Mb / b: 16Mb /swap HPFS 91 MB for OS/2 g: 45MB /usr wd1 HPFS 300 MB 104 mb reserved for FreeBSD After (W) to save to I entered ( P) to proceed. The errors are as below wd0 interrupt time out wd;0 status 59 error 10 wd0a: hard error writting fsbn 102399 wd0 bn 515259 cn 1226 tn 9 sn 24 Status 51 error 10 I followed the troubleshooting hints on the install diskette with no success. Free Bsd reports the same Disk geometry as OS/2, ie 989 Cyls 12 trks 35 sectors. Machine Details: 486DX2 66 VL bus Taiwanese Motherboard / 256K Cache. 8 Mb RAM SB16 sound card IDE controller with wd0 = Western digital 202 MB Caviar Disk wd1 =Western Digital 404 MB Caviar Disk ISA bus 16550 serial/parallel port. Genoa 8900VL 2mb (ET400W32i) video card. Mitsumi FX001D CDROM via SB16. irq 11 H'340' 3.5 Diskette Clone Mouse. OS/2 FDISK sees the '5a' partitions and the Boot works, ie I get the enter F1 etc messages on Booting. The HPFS partitions have not been affected! I hope you can help me Gordon.......... geg@cix.compulink.co.uk From owner-freebsd-bugs Sun Apr 9 15:40:11 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id PAA19359 for bugs-outgoing; Sun, 9 Apr 1995 15:40:11 -0700 Received: from unlisys.unlisys.NET (unlisys.unlisys.net [194.64.15.1]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id PAA19350 for ; Sun, 9 Apr 1995 15:40:04 -0700 Received: by unlisys.unlisys.NET from bolzen.in-berlin.de with bsmtp (Smail3.1.28.1 [@@]) id m0ry5e6-000HzMC; Mon, 10 Apr 95 00:39 MET DST Received: from luthien.in-berlin.de by bolzen.in-berlin.de with uucp (Smail3.1.29.1 #3) id m0ry5BH-0000ZXC; Sun, 9 Apr 95 23:10 GMT+0100 Received: by luthien.in-berlin.de (8.6.8/1.34) id OAA00187; Sun, 9 Apr 1995 14:31:57 +0200 From: bolzen.in-berlin.de!luthien.in-berlin.de!wagner@unlisys.NET (Olaf Wagner) Message-Id: <199504091231.OAA00187@luthien.in-berlin.de> Subject: kernel panic To: bugs@FreeBSD.org Date: Sun, 9 Apr 1995 14:31:57 +1567541 (MET DST) X-Mailer: ELM [version 2.4 PL21] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 5468 Sender: bugs-owner@FreeBSD.org Precedence: bulk Trying to integrate the language support for Modula-3 into the FreeBSD version of gdb, I was surprised to crash the whole system with the first start of m3gdb. The syslog messages state the following: Apr 9 13:25:21 luthien /kernel: pid 14905: m3gdb: uid 15: exited on signal 11 Apr 9 13:46:03 luthien /kernel: pid 15008: gdb: uid 15: exited on signal 11 Apr 9 13:47:45 luthien /kernel: pid 15018: gdb: uid 15: exited on signal 11 Apr 9 14:09:43 luthien /kernel: pid 15103: gdb: uid 15: exited on signal 11 Apr 9 14:23:11 luthien /kernel: kernel page directory invalid pdir=0x994063, va=0x21000 Apr 9 14:23:11 luthien /kernel: panic: invalid kernel page directory Apr 9 14:23:11 luthien /kernel: Apr 9 14:23:11 luthien /kernel: syncing disks... 10 10 8 4 done Apr 9 14:23:11 luthien /kernel: Automatic reboot in 15 seconds - press a key on the console to abort Apr 9 14:23:12 luthien /kernel: --> Press a key on the console to reboot <-- Apr 9 14:23:12 luthien /kernel: Rebooting... Apr 9 14:23:12 luthien /kernel: FreeBSD 2.1.0-Development #0: Mon Mar 20 12:18:37 MET 1995 Apr 9 14:23:12 luthien /kernel: root@luthien.in-berlin.de:/usr2/src/sys/compile/LUTHIEN Apr 9 14:23:12 luthien /kernel: CPU: i486DX (486-class CPU) Id = 0x435 Origin = "GenuineIntel" Apr 9 14:23:12 luthien /kernel: real memory = 33161216 (8096 pages) Apr 9 14:23:12 luthien /kernel: avail memory = 31461376 (7681 pages) Apr 9 14:23:12 luthien /kernel: Probing for devices on the ISA bus: Apr 9 14:23:12 luthien /kernel: sc0 at 0x60-0x6f irq 1 on motherboard Apr 9 14:23:12 luthien /kernel: sc0: VGA color <16 virtual consoles, flags=0x0> Apr 9 14:23:13 luthien /kernel: ed0 at 0x300-0x31f irq 10 maddr 0xcc000 msize 16384 on isa Apr 9 14:23:13 luthien /kernel: ed0: address 00:00:c0:46:85:a2, type SMC8416C/SMC8416BT (16 bit) Apr 9 14:23:13 luthien /kernel: sio0 not found at 0x2a0 Apr 9 14:23:13 luthien /kernel: sio2 not found at 0x2b0 Apr 9 14:23:13 luthien /kernel: sio3 not found at 0x2b8 Apr 9 14:23:13 luthien /kernel: sio4 at 0x3f8-0x3ff irq 4 on isa Apr 9 14:23:13 luthien /kernel: sio4: type 16550A Apr 9 14:23:13 luthien /kernel: sio5 at 0x2f8-0x2ff irq 3 on isa Apr 9 14:23:13 luthien /kernel: sio5: type 16550A Apr 9 14:23:14 luthien /kernel: lpt0 at 0x3bc-0x3c3 irq 7 on isa Apr 9 14:23:14 luthien /kernel: lpt0: Interrupt-driven port Apr 9 14:23:14 luthien /kernel: lp0: TCP/IP capable interface Apr 9 14:23:14 luthien /kernel: fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa Apr 9 14:23:14 luthien /kernel: fdc0: (NEC 72065B) [0: fd0: 1.2MB 5.25in] [1: fd1: 1.44MB 3.5in] Apr 9 14:23:14 luthien /kernel: aha0 not found at 0x330 Apr 9 14:23:14 luthien /kernel: npx0 on motherboard Apr 9 14:23:14 luthien /kernel: npx0: INT 16 interface Apr 9 14:23:14 luthien /kernel: pci0: scanning device 0..15, mechanism=2. Apr 9 14:23:14 luthien /kernel: chip0 on pci0:0 Apr 9 14:23:14 luthien /kernel: ncr0 int a irq 9 on pci0:1 Apr 9 14:23:14 luthien /kernel: reg20: virtual=0xf409b000 physical=0xc0000000 Apr 9 14:23:14 luthien /kernel: ncr0: restart (scsi reset). Apr 9 14:23:14 luthien /kernel: ncr0 scanning for targets 0..6 (V2 pl10 95/02/06) Apr 9 14:23:15 luthien /kernel: probe0(ncr0:0:0): 200ns (5 Mb/sec) offset 8. Apr 9 14:23:15 luthien /kernel: ncr0 targ 0 lun 0: type 0(direct) fixed SCSI2 Apr 9 14:23:15 luthien /kernel: ncr0 targ 0 lun 0: Apr 9 14:23:15 luthien /kernel: sd0(ncr0:0:0): FAST SCSI-2 100ns (10 Mb/sec) offset 8. Apr 9 14:23:15 luthien /kernel: sd0: 496MB (1015812 total sec), 1429 cyl, 11 head, 64 sec, bytes/sec 512 Apr 9 14:23:15 luthien /kernel: probe0(ncr0:1:0): 200ns (5 Mb/sec) offset 8. Apr 9 14:23:15 luthien /kernel: ncr0 targ 1 lun 0: type 0(direct) fixed SCSI2 Apr 9 14:23:15 luthien /kernel: ncr0 targ 1 lun 0: Apr 9 14:23:16 luthien /kernel: sd1(ncr0:1:0): FAST SCSI-2 100ns (10 Mb/sec) offset 8. Apr 9 14:23:16 luthien /kernel: sd1: 1033MB (2117025 total sec), 1819 cyl, 15 head, 77 sec, bytes/sec 512 Apr 9 14:23:16 luthien /kernel: ncr0 targ 2 lun 0: type 1(sequential) removable SCSI1 Apr 9 14:23:16 luthien /kernel: ncr0 targ 2 lun 0: Apr 9 14:23:16 luthien /kernel: st0: Tandberg tdc3600 is a known rogue Apr 9 14:23:16 luthien /kernel: st0: density code 0x10, drive empty Apr 9 14:23:16 luthien /kernel: chip1 on pci0:2 Apr 9 14:23:16 luthien /kernel: vga0 int a irq 5 on pci0:6 Apr 9 14:23:16 luthien /kernel: pci uses physical addresses from 0xc0000000 to 0xc0001000 Apr 9 14:23:16 luthien /kernel: WARNING: / was not properly dismounted Apr 9 14:23:17 luthien /kernel: sd1(ncr0:1:0): Apr 9 14:23:13 luthien lpd[83]: restarted Apr 9 14:23:13 luthien lpd[86]: /var/log/lpd-errs: No such file or directory Apr 9 14:23:34 luthien login: login on ttyv1 as wagner This is the SNAPSHOT dated 10.02.1995, and it was the first kernel panic I got for months... Olaf Wagner -- /\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ || Olaf Wagner | wagner@luthien.in-berlin.de (private) | || Sanderstrasse 13 | olaf@logware.de (work) | || 12047 Berlin | phone: 49 30 623 36 35 | || Germany / Deutschland | please don't call before 9 o'clock | \///////////////////////////////////////////////////////////////// From owner-freebsd-bugs Sun Apr 9 16:31:29 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id QAA20339 for bugs-outgoing; Sun, 9 Apr 1995 16:31:29 -0700 Received: from todonix.ping.de (root@todonix.ping.de [193.100.14.43]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id QAA20333 for ; Sun, 9 Apr 1995 16:31:15 -0700 Received: by todonix.ping.de (Smail3.1.28.1 #1) id m0ppmUx-000GnLC; Sun, 10 Apr 94 01:31 MET DST Message-Id: From: jan@todonix.ping.de (Jan Wedekind) Subject: Cyclades driver bug report To: bugs@FreeBSD.org Date: Sun, 10 Apr 1994 01:31:34 +0200 (MET DST) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 2760 Sender: bugs-owner@FreeBSD.org Precedence: bulk Hello, we bought and installed 2 Cyclades Cyclom 8Yo cards, using FreeBSD-2.0 SNAP 950322. we found the following problems: a) the driver only works for one installed card because of global variables for the IO-Memory and the internal structure (is there anybody else working on a further version ?) b) the driver itself (or some other parts of the kernel) are buggy because of the following reason: configuration: (with options for KTRACE, DDB and DIAGNOSTIC) device cy0 at isa? tty iomem 0xd4000 iosiz 0x2000 vector cyintr (so the cyclom_base has been set to 0xf00d4000) the kernel found the card, but when trying to use it, to configure a modem with kermit, we got a kernel trap, reason 'page not found': virtaddr 0x4 reason / type supervisor read, page not present at: 0x8:f01a8dd9 base 0, limit 0xffff, type 0x1b DPL 0, ... type 12 trap at _cytimeout+0x2d movl 0x4(%ecx), (%esi) some of the 'show reg' output: cs mmclose+0x4 ds isadmarangecheck+0x50 cax cy_tty+0x78 we testet in two computer's, a PCI 486DX4-100 with an ASUS SP3G board, and a 486DX2-66 with VL-Bus architecture. Both showed the same results. Any suggestions ? Thanks, Jan PS: the error occurs at the following code (see /sys/i386/isa/cy.c) /* upper level character processing routine */ void cytimeout(void *ptr) { int unit; timeout_scheduled = 0; #ifdef CyDebug cy_timeouts++; #endif /* check each port in turn */ for (unit = 0; unit < NCY*PORTS_PER_CYCLOM; unit++) { struct cy *ip = info[unit]; #ifndef TxBuffer struct tty *tp = ip->tty; #endif **** i guess here the trap occurs (see assembler input below) this would would mean, that the info struct hasn't been initialized! *** /* ignore anything that is not open */ if (!ip->tty) continue; /* * any received chars to handle? (doesn't matter if intr routine * kicks in while we're testing this) */ if (ip->rx_buf->free != CY_RX_BUF_SIZE) service_upper_rx(unit); [continued] the assembler input: .globl _cytimeout .type _cytimeout,@function _cytimeout: pushl %ebp movl %esp,%ebp subl $24,%esp pushl %edi pushl %esi pushl %ebx movb $0,_timeout_scheduled incl _cy_timeouts movl $0,-8(%ebp) .align 2,0x90 L157: movl -8(%ebp),%ecx movl _info(,%ecx,4),%ecx movl %ecx,-12(%ebp) --> movl 4(%ecx),%esi <-- here the page not present trap occurs testl %esi,%esi je L156 movl 48(%ecx),%eax cmpl $256,4(%eax) je L159 movl %esi,-16(%ebp) movl %eax,-20(%ebp) #APP cli #NO_APP movl 8(%eax),%eax movl %eax,48(%ecx) #APP sti #NO_APP -- PING e.V. Jan Wedekind (Kassenwart) jan@ping.de Emil-Figge-Str. 85 Lindemannstr. 12 kasse@ping.de 44227 Dortmund 44137 Dortmund vorstand@ping.de Tel.: noch nicht bek. Tel.: +49 (231) 13 44 04 info@ping.de From owner-freebsd-bugs Sun Apr 9 17:58:12 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id RAA22682 for bugs-outgoing; Sun, 9 Apr 1995 17:58:12 -0700 Received: from localhost (localhost [127.0.0.1]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id RAA22675 ; Sun, 9 Apr 1995 17:58:11 -0700 X-Authentication-Warning: freefall.cdrom.com: Host localhost didn't use HELO protocol To: uhclem%nemesis@fw.ast.com cc: freebsd-bugs@freefall.cdrom.com Subject: Re: bin/319: Install doesn't strip sys name in prompt for domain FDIV013 In-reply-to: Your message of "Sun, 09 Apr 95 13:20:29 PDT." <199504092020.NAA01359@freefall.cdrom.com> Date: Sun, 09 Apr 1995 17:58:10 -0700 Message-ID: <22674.797475490@freefall.cdrom.com> From: "Jordan K. Hubbard" Sender: bugs-owner@FreeBSD.org Precedence: bulk This is a bug introduced into sed by our library ordering changes (static vs dynamic). It should be fixed in the latest snap (I won't try to close this until it's verified). Jordan > > >Number: 319 > >Category: bin > >Synopsis: Install doesn't strip sys name in prompt for domain FDIV013 > >Confidential: no > >Severity: non-critical > >Priority: low > >Responsible: freebsd-bugs (FreeBSD bugs mailing list) > >State: open > >Class: sw-bug > >Submitter-Id: current-users > >Arrival-Date: Sun Apr 9 13:20:27 1995 > >Originator: Frank Durda IV > >Organization: > >Release: FreeBSD 2.0.0-SNAP950322 i386 > >Environment: > > FreeBSD 2.0.0-SNAP950322 i386 > > > >Description: > > [FDIV013] > > When doing FTP download, it asks for the machines fully-qualified domain > name. I typed: > > dalek.lonestar.org > > The next question asked what my domain name was, and it filled the box in > with: > > dalek.lonestar.org| > > with the cursor on the right. > > Why didn't the install at least delete the left-most field off the > fully-qualified domain name and offer that as a prompt? > > > >How-To-Repeat: > > Install system and select that you want to do a FTP install. > > > >Fix: > > > *END* > >Audit-Trail: > >Unformatted: > > From owner-freebsd-bugs Mon Apr 10 00:40:01 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id AAA00493 for bugs-outgoing; Mon, 10 Apr 1995 00:40:01 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id AAA00486 ; Mon, 10 Apr 1995 00:40:01 -0700 Date: Mon, 10 Apr 1995 00:40:01 -0700 Message-Id: <199504100740.AAA00486@freefall.cdrom.com> From: edward@edcom.com Reply-To: edward@edcom.com To: freebsd-bugs Subject: bin/330: lastcomm output format error In-Reply-To: Your message of Mon, 10 Apr 1995 00:36:37 -0700 <199504100736.AAA01425@edcom.com> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 330 >Category: bin >Synopsis: lastcomm outputs garbage when command name is 10 chars long >Confidential: no >Severity: serious >Priority: medium >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Mon Apr 10 00:40:00 1995 >Originator: Edward Wang >Organization: unix guru inc. >Release: FreeBSD 2.1.0-Development i386 >Environment: 2.0-950322-SNAP release >Description: the printf format string for lastcomm is incorrect causing garbage output on lines with command names of 10 chars >How-To-Repeat: lastcomm | more and look for something long >Fix: *** lastcomm.c.dist Fri May 27 05:31:57 1994 --- lastcomm.c Mon Apr 10 00:29:57 1995 *************** *** 135,142 **** continue; t = expand(ab.ac_utime) + expand(ab.ac_stime); ! (void)printf("%-*s %-7s %-*s %-*s %6.2f secs %.16s\n", ! fldsiz(acct, ac_comm), ab.ac_comm, flagbits(ab.ac_flag), UT_NAMESIZE, user_from_uid(ab.ac_uid, 0), UT_LINESIZE, getdev(ab.ac_tty), t / (double)AHZ, ctime(&ab.ac_btime)); --- 135,143 ---- continue; t = expand(ab.ac_utime) + expand(ab.ac_stime); ! (void)printf("%-*.*s %-7s %-*s %-*s %6.2f secs %.16s\n", ! fldsiz(acct, ac_comm), fldsiz(acct, ac_comm), ! ab.ac_comm, flagbits(ab.ac_flag), UT_NAMESIZE, user_from_uid(ab.ac_uid, 0), UT_LINESIZE, getdev(ab.ac_tty), t / (double)AHZ, ctime(&ab.ac_btime)); >Audit-Trail: >Unformatted: From owner-freebsd-bugs Mon Apr 10 00:50:02 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id AAA00673 for bugs-outgoing; Mon, 10 Apr 1995 00:50:02 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id AAA00666 ; Mon, 10 Apr 1995 00:50:01 -0700 Date: Mon, 10 Apr 1995 00:50:01 -0700 Message-Id: <199504100750.AAA00666@freefall.cdrom.com> From: edward@edcom.com Reply-To: edward@edcom.com To: freebsd-bugs Subject: kern/331: SCSI tape short-read failure In-Reply-To: Your message of Mon, 10 Apr 1995 00:45:50 -0700 <199504100745.AAA01711@edcom.com> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 331 >Category: kern >Synopsis: a read that returns short on a SCSI tape fails >Confidential: no >Severity: serious >Priority: medium >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Mon Apr 10 00:50:00 1995 >Originator: Edward Wang >Organization: feh! >Release: FreeBSD 2.1.0-Development i386 >Environment: 2.0-950322-SNAP >Description: On a SCSI tape (and really other devices with similar characteristics), a read of a record smaller than the request size always fails. >How-To-Repeat: find a SCSI tape drive put a tape in do dd if=/dev/zero of=/dev/rst0 bs=1b count=1 dd if=/dev/rst0 of=/dev/null bs=1k count=1 the read fails when it should return 512 bytes >Fix: *** scsi_base.c.dist Tue Mar 21 03:21:01 1995 --- scsi_base.c Sun Apr 9 22:56:52 1995 *************** *** 593,604 **** xs->flags &= ~ITSDONE; goto retry; } } - retval = EIO; /* Too many retries */ if (bp) { bp->b_error = 0; - bp->b_resid = 0; if (retval) { bp->b_flags |= B_ERROR; bp->b_error = retval; --- 593,603 ---- xs->flags &= ~ITSDONE; goto retry; } + retval = EIO; /* Too many retries */ } if (bp) { bp->b_error = 0; if (retval) { bp->b_flags |= B_ERROR; bp->b_error = retval; The first change (position of retval = EIO) is the bug. The second change is to return the correct b_resid. The second problem seems to be in the 1.1.5.1 code as well. It's less used so possibly nobody noticed. I can't be sure it doesn't affect something else -- something that doesn't set b_resid correctly and relies on this to clear it. However, I'm running this on a machine with SCSI disks and tapes. They seem to work. >Audit-Trail: >Unformatted: From owner-freebsd-bugs Mon Apr 10 06:37:46 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id GAA00240 for bugs-outgoing; Mon, 10 Apr 1995 06:37:46 -0700 Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.34]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id GAA00231 for ; Mon, 10 Apr 1995 06:37:12 -0700 Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id XAA16593; Mon, 10 Apr 1995 23:23:41 +1000 Date: Mon, 10 Apr 1995 23:23:41 +1000 From: Bruce Evans Message-Id: <199504101323.XAA16593@godzilla.zeta.org.au> To: bugs@FreeBSD.org, jan@todonix.ping.de Subject: Re: Cyclades driver bug report Sender: bugs-owner@FreeBSD.org Precedence: bulk >we bought and installed 2 Cyclades Cyclom 8Yo cards, >using FreeBSD-2.0 SNAP 950322. >we found the following problems: >a) the driver only works for one installed card because > of global variables for the IO-Memory and the internal structure > (is there anybody else working on a further version ?) Yes. >... >the kernel found the card, but when trying to use it, to configure >a modem with kermit, we got a kernel trap, reason 'page not found': The driver has some hard-coded configuration that is suitable only for cy16's. The changes at the should allow it to work for ONE cy8. Another cyclades driver was posted to freebsd-current on 31 March. It is more complete but less efficient. Bruce *** cy.c~ Wed Mar 29 05:56:57 1995 --- cy.c Sun Apr 9 23:56:00 1995 *************** *** 60,68 **** #if NCY > 0 ! /* This disgusing hack because we actually have 16 units on one controller */ #if NCY < 2 #undef NCY #define NCY (16) #endif #include --- 63,73 ---- #if NCY > 0 ! #if 0 ! /* This disgusting hack because we actually have 16 units on one controller */ #if NCY < 2 #undef NCY #define NCY (16) #endif + #endif #include *************** *** 126,131 **** /* #define CD1400s_PER_CYCLOM 1 */ /* cyclom-4y */ ! /* #define CD1400s_PER_CYCLOM 2 */ /* cyclom-8y */ ! #define CD1400s_PER_CYCLOM 4 /* cyclom-16y */ /* FreeBSD's getty doesn't know option for setting RTS/CTS handshake. Its --- 131,136 ---- /* #define CD1400s_PER_CYCLOM 1 */ /* cyclom-4y */ ! #define CD1400s_PER_CYCLOM 2 /* cyclom-8y */ ! /* #define CD1400s_PER_CYCLOM 4 */ /* cyclom-16y */ /* FreeBSD's getty doesn't know option for setting RTS/CTS handshake. Its *************** *** 140,145 **** --- 145,152 ---- #define CD1400_MEMSIZE 0x100 /* 256 bytes per chip: cyclom-16y */ /* XXX or is it 0x400 like the rest? */ + #if 0 #define CYCLOM_16 1 /* This is a cyclom-16Y */ #endif + #endif #define PORTS_PER_CYCLOM (CD1400_NO_OF_CHANNELS * CD1400s_PER_CYCLOM) From owner-freebsd-bugs Mon Apr 10 08:07:53 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id IAA00151 for bugs-outgoing; Mon, 10 Apr 1995 08:07:53 -0700 Received: from mail.barrnet.net (mail.BARRNET.NET [131.119.246.7]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id IAA00136 for ; Mon, 10 Apr 1995 08:07:51 -0700 Received: from ns1.win.net (ns1.win.net [204.215.209.3]) by mail.barrnet.net (8.6.10/MAIL-RELAY-LEN) with ESMTP id IAA18143 for ; Mon, 10 Apr 1995 08:05:17 -0700 Received: (from bugs@localhost) by ns1.win.net (8.6.9/8.6.9) id LAA01975 for freebsd-bugs@freebsd.org; Mon, 10 Apr 1995 11:08:43 -0400 From: Mark Hittinger Message-Id: <199504101508.LAA01975@ns1.win.net> Subject: inet problems -current To: freebsd-bugs@FreeBSD.org Date: Mon, 10 Apr 1995 11:08:41 -0400 (EDT) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 663 Sender: bugs-owner@FreeBSD.org Precedence: bulk I've just realized that I have some sort of kernel problem with my internet traffic on the -current box. When using rcp or ftp to copy a large file it will hang shortly after the command. For example, copying a 3 meg file will get the first 400K chunk and then get stuck. This has happened after 16K as well, but most of the time I seem to get at least 350K before the stall. The inetd appears to be stuck in an input wait when you do a ps. Shifted back to the snap-0322 kernel and everything runs as expected. I seem to have the problem with my -current kernels going back to april 6. I have the 3c509 card btw. Regards, Mark Hittinger bugs@win.net From owner-freebsd-bugs Mon Apr 10 09:56:15 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id JAA10120 for bugs-outgoing; Mon, 10 Apr 1995 09:56:15 -0700 Received: from halloran-eldar.lcs.mit.edu (halloran-eldar.lcs.mit.edu [18.26.0.159]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id JAA10108 for ; Mon, 10 Apr 1995 09:56:12 -0700 Received: by halloran-eldar.lcs.mit.edu; id AA03250; Mon, 10 Apr 1995 12:56:02 -0400 Date: Mon, 10 Apr 1995 12:56:02 -0400 From: Garrett Wollman Message-Id: <9504101656.AA03250@halloran-eldar.lcs.mit.edu> To: uhclem%nemesis@fw.ast.com Cc: freebsd-bugs@freefall.cdrom.com Subject: bin/327: Clock management punishes you if CMOS != GMT FDIV020 In-Reply-To: <199504092021.NAA01474@freefall.cdrom.com> References: <199504092021.NAA01474@freefall.cdrom.com> Sender: bugs-owner@FreeBSD.org Precedence: bulk <> Number: 327 > If you elect to keep the CMOS at local time, the installation will offer > you cities (not many choices in CST by the way), then show you a time with > the CST suffix, WITH THE GMT OFFSET ADDED even if you don't want it. The first thing it asks you is: Is $4, $1 $2 $3 $6 the correct current date and time? (with appropriate values for the positional variables). If your CMOS clock is set to local time, then this will be the current time, and so you are supposed to answer `y'. It then tells you: Unless your local time is GMT, this probably means that your CMOS clock is on local time. In this case, the local times that you will be shown will be incorrect. You should either set your CMOS clock to GMT, or select option 98 from the main menu after selecting a timezone. Can't get much more explicit than that. Note that we VERY EXPLICITLY do not allow users to select something like `CST' as their timezone, because the notion is not well-defined. Instead, we ask: Please select a location from the following list which has the same legal time as your location: This is a very different question, because the time zone database keeps track of summer time arrangements and changes in zone boundaries. Thus, for most of the U.S. locations that observe ``Central time'', `Chicago' is the correct choice, because most locations in that area started observing both standard and summer time at the same time as Chicago did. Most parts of Indiana started out with Eastern, summer time, but in 1946, Indiana decided not to observe summer time, and so those locations use `Fort_Wayne'. Starke County, Indiana, used to be on Central time following the `Chicago' definition, but in October 1991 it decided to follow the rest of Indiana and be on Eastern time all the time, with no summer time; hence, `Knox_IN'. Michigan has always been in the Eastern zone, but didn't observe summer time from 1968 to 1973, hence `Detroit'. At some point in time, I'll re-do all this a la `dmenu'. (Or Jordan is welcome to...) It's all explained in gory detail in /usr/src/share/zoneinfo. > It then asks "Is this what you wanted?". The normal user isn't going to > realize that all the installation process is asking about at that point is > the CST suffix and will answer the question NO because the time is wrong. I agree that there is a problem here, but there's nothing we can do about it, since we can't correct the time without knowing what the user's timezone is. I probably should have asked ``Is this the time zone you wanted?'' > Also note that if you have CMOS set to LOCAL, and boot the system in maint > mode, the date shown is wrong (behind by several hours). Another good reason to use a UTC CMOS clock... There's no other way it can work the way you want it. -GAWollman -- Garrett A. Wollman | Shashish is simple, it's discreet, it's brief. ... wollman@lcs.mit.edu | Shashish is the bonding of hearts in spite of distance. Opinions not those of| It is a bond more powerful than absence. We like people MIT, LCS, ANA, or NSA| who like Shashish. - Claude McKenzie + Florent Vollant From owner-freebsd-bugs Mon Apr 10 11:59:02 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA18233 for bugs-outgoing; Mon, 10 Apr 1995 11:59:02 -0700 Received: (from wollman@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA18219 ; Mon, 10 Apr 1995 11:59:00 -0700 Date: Mon, 10 Apr 1995 11:59:00 -0700 From: "Garrett A. Wollman" Message-Id: <199504101859.LAA18219@freefall.cdrom.com> To: gary@palmer.demon.co.uk, wollman, freebsd-bugs Subject: Changed information for PR bin/182 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: Strange error from umount State-Changed-From-To: open-closed State-Changed-By: wollman State-Changed-When: Mon Apr 10 11:57:20 PDT 1995 State-Changed-Why: Fixed by correctly naming the type of MOUNT_CD9660, and correcting initialization bug in umount/umount.c. From owner-freebsd-bugs Mon Apr 10 12:17:57 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id GAA00240 for bugs-outgoing; Mon, 10 Apr 1995 06:37:46 -0700 Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.34]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id GAA00231 for ; Mon, 10 Apr 1995 06:37:12 -0700 Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id XAA16593; Mon, 10 Apr 1995 23:23:41 +1000 Date: Mon, 10 Apr 1995 23:23:41 +1000 From: Bruce Evans Message-Id: <199504101323.XAA16593@godzilla.zeta.org.au> To: bugs@freebsd.org, jan@todonix.ping.de Subject: Re: Cyclades driver bug report Sender: bugs-owner@freebsd.org Precedence: bulk >we bought and installed 2 Cyclades Cyclom 8Yo cards, >using FreeBSD-2.0 SNAP 950322. >we found the following problems: >a) the driver only works for one installed card because > of global variables for the IO-Memory and the internal structure > (is there anybody else working on a further version ?) Yes. >... >the kernel found the card, but when trying to use it, to configure >a modem with kermit, we got a kernel trap, reason 'page not found': The driver has some hard-coded configuration that is suitable only for cy16's. The changes at the should allow it to work for ONE cy8. Another cyclades driver was posted to freebsd-current on 31 March. It is more complete but less efficient. Bruce *** cy.c~ Wed Mar 29 05:56:57 1995 --- cy.c Sun Apr 9 23:56:00 1995 *************** *** 60,68 **** #if NCY > 0 ! /* This disgusing hack because we actually have 16 units on one controller */ #if NCY < 2 #undef NCY #define NCY (16) #endif #include --- 63,73 ---- #if NCY > 0 ! #if 0 ! /* This disgusting hack because we actually have 16 units on one controller */ #if NCY < 2 #undef NCY #define NCY (16) #endif + #endif #include *************** *** 126,131 **** /* #define CD1400s_PER_CYCLOM 1 */ /* cyclom-4y */ ! /* #define CD1400s_PER_CYCLOM 2 */ /* cyclom-8y */ ! #define CD1400s_PER_CYCLOM 4 /* cyclom-16y */ /* FreeBSD's getty doesn't know option for setting RTS/CTS handshake. Its --- 131,136 ---- /* #define CD1400s_PER_CYCLOM 1 */ /* cyclom-4y */ ! #define CD1400s_PER_CYCLOM 2 /* cyclom-8y */ ! /* #define CD1400s_PER_CYCLOM 4 */ /* cyclom-16y */ /* FreeBSD's getty doesn't know option for setting RTS/CTS handshake. Its *************** *** 140,145 **** --- 145,152 ---- #define CD1400_MEMSIZE 0x100 /* 256 bytes per chip: cyclom-16y */ /* XXX or is it 0x400 like the rest? */ + #if 0 #define CYCLOM_16 1 /* This is a cyclom-16Y */ #endif + #endif #define PORTS_PER_CYCLOM (CD1400_NO_OF_CHANNELS * CD1400s_PER_CYCLOM) From owner-freebsd-bugs Mon Apr 10 13:37:39 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA29250 for bugs-outgoing; Mon, 10 Apr 1995 13:37:39 -0700 Received: (from wollman@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA29233 ; Mon, 10 Apr 1995 13:37:34 -0700 Date: Mon, 10 Apr 1995 13:37:34 -0700 From: "Garrett A. Wollman" Message-Id: <199504102037.NAA29233@freefall.cdrom.com> To: wollman@khavrinen.lcs.mit.edu, wollman, freebsd-bugs Subject: Changed information for PR kern/306 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: if_tun LKM does not set up cdevsw entry State-Changed-From-To: open-closed State-Changed-By: wollman State-Changed-When: Mon Apr 10 13:36:18 PDT 1995 State-Changed-Why: tunnel driver now sets up its cdevsw entry on its own, with the help of some goop in conf.c. Rev. 1.5 of if_tun.c. From owner-freebsd-bugs Mon Apr 10 13:43:34 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA29993 for bugs-outgoing; Mon, 10 Apr 1995 13:43:34 -0700 Received: from hda.com (hda.com [199.232.40.182]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id NAA29973 for ; Mon, 10 Apr 1995 13:43:27 -0700 Received: (dufault@localhost) by hda.com (8.6.9/8.3) id QAA12359; Mon, 10 Apr 1995 16:43:03 -0400 From: Peter Dufault Message-Id: <199504102043.QAA12359@hda.com> Subject: Re: kern/331: SCSI tape short-read failure To: edward@edcom.com Date: Mon, 10 Apr 1995 16:43:03 -0400 (EDT) Cc: freebsd-bugs@freefall.cdrom.com In-Reply-To: <199504100750.AAA00666@freefall.cdrom.com> from "edward@edcom.com" at Apr 10, 95 00:50:01 am X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 474 Sender: bugs-owner@FreeBSD.org Precedence: bulk Thanks. I have a whole slew of fixes to the "st" resid code, but I can't commit them until I test them myself. The last time I just committed some code Julian gave me I managed to drop all the changes to one of the files. Look for the commit in the next day or so when I can get a few hours for test. -- Peter Dufault Real Time Machine Control and Simulation HD Associates, Inc. Voice: 508 433 6936 dufault@hda.com Fax: 508 433 5267 From owner-freebsd-bugs Mon Apr 10 17:22:18 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id RAA17959 for bugs-outgoing; Mon, 10 Apr 1995 17:22:18 -0700 Received: from mail.barrnet.net (mail.BARRNET.NET [131.119.246.7]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id RAA17951 for ; Mon, 10 Apr 1995 17:22:16 -0700 Received: from trinity.radio-do.de (gate.Radio-do.de [193.101.164.225]) by mail.barrnet.net (8.6.10/MAIL-RELAY-LEN) with ESMTP id RAA23412 for ; Mon, 10 Apr 1995 17:19:38 -0700 Date: Tue, 11 Apr 1995 02:22:53 +0200 From: Frank Nobis Message-Id: <199504110022.CAA03381@trinity.radio-do.de> Received: by trinity.radio-do.de (8.6.9/TRINITY-1.2.0-a) via EUnet id CAA03381; Tue, 11 Apr 1995 02:22:53 +0200 To: freebsd-bugs@FreeBSD.org Subject: Patch to ports/mail/popper Sender: bugs-owner@FreeBSD.org Precedence: bulk I just tried to compile popper. Through a typo the Makefile didn't get the tar file. Then I found out, that there is a newer version of popper than the one in the Makfile. so here is the simple patch. BTW. the new version compiles without any erros. even the patches get applied correctly. *** Makefile Sun Apr 2 05:28:51 1995 --- Makefile.fn Tue Apr 11 02:01:04 1995 *************** *** 6,12 **** # $Id: Makefile,v 1.6 1995/04/02 01:13:42 pst Exp $ # ! DISTNAME= qpop2.1.4-R1 MASTER_SITES= ftp://ftp.qualcomm.com/quest/unix/servers/popper/ EXTRACT_SUFX= .tar.Z WRKSRC= ${WRKDIR}/qpopper2.1.4 --- 6,12 ---- # $Id: Makefile,v 1.6 1995/04/02 01:13:42 pst Exp $ # ! DISTNAME= qpop2.1.4-r2 MASTER_SITES= ftp://ftp.qualcomm.com/quest/unix/servers/popper/ EXTRACT_SUFX= .tar.Z WRKSRC= ${WRKDIR}/qpopper2.1.4 Virtually fn -- Frank Nobis Email: fn@Radio-do.de Landgrafenstr. 130 fn@Germany.EU.net 44139 Dortmund Fax: +49 231 7213816 From owner-freebsd-bugs Mon Apr 10 18:43:27 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id SAA28701 for bugs-outgoing; Mon, 10 Apr 1995 18:43:27 -0700 Received: (from rgrimes@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id SAA28635 ; Mon, 10 Apr 1995 18:43:21 -0700 Date: Mon, 10 Apr 1995 18:43:21 -0700 From: "Rodney W. Grimes" Message-Id: <199504110143.SAA28635@freefall.cdrom.com> To: smp@clem.systemsix.com, rgrimes, freebsd-bugs Subject: Changed information for PR misc/249 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: $hostname variable in /etc/netstart not always set State-Changed-From-To: open-closed State-Changed-By: rgrimes State-Changed-When: Mon Apr 10 18:41:51 PDT 1995 State-Changed-Why: iThe variable hostname is now always set in sysconfig, maeing the sited conditions imposible. [Arg, this sucks, I can't edit :-(] From owner-freebsd-bugs Tue Apr 11 00:58:15 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id AAA15763 for bugs-outgoing; Tue, 11 Apr 1995 00:58:15 -0700 Received: from mail.eskimo.com (root@mail.eskimo.com [204.122.16.4]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id AAA15757 for ; Tue, 11 Apr 1995 00:58:13 -0700 Received: from eskimo.com by mail.eskimo.com (5.65c/1.35) id AA08124; Tue, 11 Apr 1995 00:57:53 -0700 Date: Tue, 11 Apr 1995 00:57:53 -0700 (PDT) From: Dave Haas To: bugs@FreeBSD.org Subject: matcd0: error during 2.0-950322-SNAP installation Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: bugs-owner@FreeBSD.org Precedence: bulk Hi there. I'm Dave Haas of Bainbridge Island, Washington. I've been using the 2/10 snapshot release of FreeBSD since around March 5th on another system and love it! When I wanted to install FreeBSD onto another system, I decided to try the 3/22 snapshot as it was said to include support for Intel EtherExpress 16 cards and improved sound card functionality. I ftp'd everything except src from ftp.cdrom.com:pub/FreeBSD/2.0-950322-SNAP, rawrote boot.flp and cpio.flp onto freshly formatted floppies, ran through the fdisk and disklabel procedures, and successfully proceeded until the initial boot from the hard disk. Then, when wd(0,a)/kernel went probing devices, the device matcd0: would eventually come up for probing and/or attachment, and the kernel would declare emphatically: matcd0: get_stat: after reading status byte, buss didn't go idle This would go on forever and head off into infinity until a hard reset was issued. So, after resetting the computer and doing a wd(0,a)/kernel -c to check out this problematic device. config> ls reveals: matcd int0xffffffff -1 -1 0x0 0 0 0x0 yes I tried to disable the device in the following ways: config> disable matcd config> di matcd config> disable matcd0 config> di matcd0 and was always told... No such device: matcd0 Invalid command or syntax. I also tried disabling, probing and attaching matcd0: with various creative and unusual spellings, thinking the device might have some weird tweak to its spelling. But always the device didn't exist. This would explain a great deal about why the kernel falls into an infinite loop when it tries to probe/attach to matcd0: After the disabling fiasco, I took the SoundBlaster 16 out of this machine, thinking that matcd0: was meant to represent the Matsushita cd-rom which attaches to these sound cards. I went through the entire procedure described above, and once again the same problem appeared. I am confused and prepared to start eating eggplant if that's what it takes to figure this out. So, the system I'm working with here is an i486DX2 66Mhz system with 16mb RAM, a 540mb Western Digital HD, a 540mb Quantum SCSI HD and a Toshiba XM3501B cd-rom (so far unrecognizable to FreeBSD because it's on a QLogic VLB SCSI controller card), a SoundBlaster 16 (presumably this is what's supposed to be attached to matcd0: ???), a Diamond SpeedStar Pro w 1mb RAM, and an Intel EtherExpress 16, Please give me guidance here so I can avoid the unpleasantries of eggplant. And thanks for putting together such a great un*x system! I recommend it to everyone. Dave Haas daveh@eskimo,com (206) 842-0679 From owner-freebsd-bugs Tue Apr 11 01:50:12 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id BAA16862 for bugs-outgoing; Tue, 11 Apr 1995 01:50:12 -0700 Received: from localhost (localhost [127.0.0.1]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id BAA16853 ; Tue, 11 Apr 1995 01:50:09 -0700 X-Authentication-Warning: freefall.cdrom.com: Host localhost didn't use HELO protocol To: Dave Haas cc: bugs@FreeBSD.org Subject: Re: matcd0: error during 2.0-950322-SNAP installation In-reply-to: Your message of "Tue, 11 Apr 95 00:57:53 PDT." MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Tue, 11 Apr 1995 01:50:08 -0700 Message-ID: <16852.797590208@freefall.cdrom.com> From: "Jordan K. Hubbard" Sender: bugs-owner@FreeBSD.org Precedence: bulk > Hi there. I'm Dave Haas of Bainbridge Island, Washington. I've been using Hi! Nice place. Dad lives over on Fox.. :-) > Then, when wd(0,a)/kernel went probing devices, the device matcd0: would > eventually come up for probing and/or attachment, and the kernel would > declare emphatically: > > matcd0: get_stat: after reading status byte, buss didn't go idle Indeed. I'm sorry, but the matcd0 driver was tragically flawed in 0322 and flamed some people out of the sky like this. I'm working right now on another snap dist that solves these problems, but have been delayed by a number of other problems. Watch the announce@freebsd.org mailing list for it once I work the last of the kinks out. Jordan From owner-freebsd-bugs Tue Apr 11 02:50:01 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id CAA17675 for bugs-outgoing; Tue, 11 Apr 1995 02:50:01 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id CAA17668 ; Tue, 11 Apr 1995 02:50:00 -0700 Date: Tue, 11 Apr 1995 02:50:00 -0700 Message-Id: <199504110950.CAA17668@freefall.cdrom.com> From: pritc003@maroon.tc.umn.edu Reply-To: pritc003@maroon.tc.umn.edu To: freebsd-bugs Subject: kern/332: Incorrect routes can cause system to reset In-Reply-To: Your message of Tue, 11 Apr 1995 04:46:56 -0500 <199504110946.EAA00290@mpp.com> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 332 >Category: kern >Synopsis: Incorrect routes can cause system to reset >Confidential: no >Severity: critical >Priority: high >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Tue Apr 11 02:50:00 1995 >Originator: Mike Pritchard >Organization: >Release: FreeBSD 2.1.0-Development i386 >Environment: System configured with 1 SLIP interface and 1 PPP interface. IPFIREWALL and MROUTING are configured. No other network devices are configured or present. >Description: Under certain conditions, an incorrect routing setup can cause the system to reset (no panic, a reset just like you hit the reset button). Any attempt to communicate with the newtwork (ping, netstat -r, telnet, whatever) when this route setup exists will cause the machine to reset. I also saw this when running with the 2.0R system when I was initially setting up my machine, but forgot about it until just now when I screwed something up in a config file tonight and caused it to happen again. >How-To-Repeat: (the SLIP interface is NOT connected, and does not need to be to duplicate the problem) # ifconfig sl0 sl0: flags=c011 mtu 552 inet 172.16.0.1 --> 128.101.118.21 netmask 0xffffff00 # netstat -rn Routing tables Internet: Destination Gateway Flags Refs Use Netif Expire default 128.101.118.21 UGSc 0 0 sl0 127.0.0.1 127.0.0.1 UH 0 0 lo0 172.16.0.1 127.0.0.1 UH 0 0 lo0 224 172.16.0.1 US 0 0 sl0 # ping freebsd.org [machine resets here] Right after a fresh reboot, the routing table looks like: # netstat -rn Routing tables Internet: Destination Gateway Flags Refs Use Netif Expire 127.0.0.1 127.0.0.1 UH 0 0 lo0 128.101.118.21 172.16.0.1 UH 0 15 sl0 172.16.0.1 127.0.0.1 UH 0 0 lo0 224 172.16.0.1 US 0 0 sl0 I did the following to change the above routes into the routes that cause the problem: route delete 128.101.118.21 172.16.0.1 route add default 128.101.118.21 >Fix: >Audit-Trail: >Unformatted: From owner-freebsd-bugs Tue Apr 11 03:50:02 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id DAA18152 for bugs-outgoing; Tue, 11 Apr 1995 03:50:02 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id DAA18145 ; Tue, 11 Apr 1995 03:50:01 -0700 Date: Tue, 11 Apr 1995 03:50:01 -0700 Message-Id: <199504111050.DAA18145@freefall.cdrom.com> From: Paul Traina Reply-To: Paul Traina To: freebsd-bugs Subject: bin/333: cvs isn't being built In-Reply-To: Your message of Tue, 11 Apr 1995 03:41:11 -0700 <199504111041.DAA04230@precipice.shockwave.com> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 333 >Category: bin >Synopsis: CVS isn't being built >Confidential: no >Severity: non-critical >Priority: low >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Tue Apr 11 03:50:00 1995 >Originator: Paul Traina >Organization: Shockwave Engineering >Release: FreeBSD 2.1.0-Development i386 >Environment: FreeBSD-current as of today (April 10, 1995) >Description: CVS isn't being built >How-To-Repeat: make world ; ls -l /usr/bin/cvs >Fix: fix /usr/src/gnu/usr.sbin/Makefile I seem to recall that cvs was removed for a reason (it was broken), but since Nate has been beating on it, I assume it's now fixed? >Audit-Trail: >Unformatted: From owner-freebsd-bugs Tue Apr 11 06:30:01 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id GAA19723 for bugs-outgoing; Tue, 11 Apr 1995 06:30:01 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id GAA19716 ; Tue, 11 Apr 1995 06:30:01 -0700 Date: Tue, 11 Apr 1995 06:30:01 -0700 Message-Id: <199504111330.GAA19716@freefall.cdrom.com> From: miff@spam.apana.org.au Reply-To: miff@spam.apana.org.au To: freebsd-bugs Subject: bin/334: -n option to strings does not work In-Reply-To: Your message of Tue, 11 Apr 1995 23:40:57 GMT <199504112340.XAA02102@spam.apana.org.au> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 334 >Category: bin >Synopsis: -n option to strings does not work >Confidential: no >Severity: non-critical >Priority: low >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Tue Apr 11 06:30:00 1995 >Originator: michael smith >Organization: Genesis Software >Release: FreeBSD 2.0-RELEASE i386 >Environment: Any. >Description: The -n number option to /usr/bin/strings is not implemented correctly. Any attempt to specify a number generates an error. >How-To-Repeat: specify the -n option to strings : # strings -n 10 /kernel |more >Fix: In /usr/src/usr.bin/strings/strings.c, add the marked lines : + if (minlen < 1) + { if (minlen == -1) minlen = DEF_LEN; else { (void)fprintf(stderr, "strings: length less than 1\n"!); exit(1); } + } >Audit-Trail: >Unformatted: From owner-freebsd-bugs Tue Apr 11 07:29:32 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id HAA20685 for bugs-outgoing; Tue, 11 Apr 1995 07:29:32 -0700 Received: from trout.sri.MT.net (trout.sri.MT.net [204.182.243.12]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id HAA20679 for ; Tue, 11 Apr 1995 07:29:26 -0700 Received: (from nate@localhost) by trout.sri.MT.net (8.6.11/8.6.10) id IAA08857; Tue, 11 Apr 1995 08:33:25 -0600 Date: Tue, 11 Apr 1995 08:33:25 -0600 Message-Id: <199504111433.IAA08857@trout.sri.MT.net> To: Paul Traina Cc: freebsd-bugs@freefall.cdrom.com Subject: Re: bin/333: cvs isn't being built In-Reply-To: <199504111050.DAA18145@freefall.cdrom.com> References: <199504111041.DAA04230@precipice.shockwave.com> <199504111050.DAA18145@freefall.cdrom.com> Reply-To: nate@sneezy.sri.com (Nate Williams) From: nate@sneezy.sri.com (Nate Williams) Sender: bugs-owner@FreeBSD.org Precedence: bulk Paul Traina writes: > >Synopsis: CVS isn't being built ... > >Fix: > > fix /usr/src/gnu/usr.sbin/Makefile > > I seem to recall that cvs was removed for a reason (it was broken), but > since Nate has been beating on it, I assume it's now fixed? It wasn't broken, but it has changed alot and Rod wanted to have it tested more before enabling it. If he doesn't mind, I'd like to re-enable it since I think the one minor boo-boo he saw wasn't critical AND it hasn't trashed anyone's Repository. Nate From owner-freebsd-bugs Tue Apr 11 09:59:39 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id JAA23137 for bugs-outgoing; Tue, 11 Apr 1995 09:59:39 -0700 Received: from halloran-eldar.lcs.mit.edu (halloran-eldar.lcs.mit.edu [18.26.0.159]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id JAA23131 for ; Tue, 11 Apr 1995 09:59:37 -0700 Received: by halloran-eldar.lcs.mit.edu; id AA05366; Tue, 11 Apr 1995 12:59:19 -0400 Date: Tue, 11 Apr 1995 12:59:19 -0400 From: Garrett Wollman Message-Id: <9504111659.AA05366@halloran-eldar.lcs.mit.edu> To: pritc003@maroon.tc.umn.edu Cc: freebsd-bugs@freefall.cdrom.com Subject: kern/332: Incorrect routes can cause system to reset In-Reply-To: <199504110950.CAA17668@freefall.cdrom.com> References: <199504110946.EAA00290@mpp.com> <199504110950.CAA17668@freefall.cdrom.com> Sender: bugs-owner@FreeBSD.org Precedence: bulk < Destination Gateway Flags Refs Use Netif Expire > default 128.101.118.21 UGSc 0 0 sl0 > 127.0.0.1 127.0.0.1 UH 0 0 lo0 > 172.16.0.1 127.0.0.1 UH 0 0 lo0 > 224 172.16.0.1 US 0 0 sl0 > # ping freebsd.org > [machine resets here] Your problem here is very simple: you have specified a gateway (128.101.118.21) for which there is no route. When something attempts to use the default route, it must first find a route for the gateway. The only route that matches is the default route! This results in bottomless mutual recursion. The answer is to NEVER perform this step: > route delete 128.101.118.21 172.16.0.1 I am also working on code which should be able to detect this situation and avoid the problem entirely. -GAWollman -- Garrett A. Wollman | Shashish is simple, it's discreet, it's brief. ... wollman@lcs.mit.edu | Shashish is the bonding of hearts in spite of distance. Opinions not those of| It is a bond more powerful than absence. We like people MIT, LCS, ANA, or NSA| who like Shashish. - Claude McKenzie + Florent Vollant From owner-freebsd-bugs Tue Apr 11 10:19:16 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id KAA23626 for bugs-outgoing; Tue, 11 Apr 1995 10:19:16 -0700 Received: (from bde@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id KAA23612 ; Tue, 11 Apr 1995 10:19:14 -0700 Date: Tue, 11 Apr 1995 10:19:14 -0700 From: Bruce Evans Message-Id: <199504111719.KAA23612@freefall.cdrom.com> To: pritc003@maroon.tc.umn.edu, bde, freebsd-bugs Subject: Changed information for PR kern/303 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: msdosfs: moving a file into another directory causes panic State-Changed-From-To: open-closed State-Changed-By: bde State-Changed-When: Tue Apr 11 10:16:40 PDT 1995 State-Changed-Why: Patch applied in revision 1.11 1995/04/11 of msdosfs_denode.c. From owner-freebsd-bugs Tue Apr 11 10:24:48 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id KAA23735 for bugs-outgoing; Tue, 11 Apr 1995 10:24:48 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id KAA23727 for ; Tue, 11 Apr 1995 10:24:44 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.8/8.6.6) id KAA04039; Tue, 11 Apr 1995 10:24:28 -0700 From: "Rodney W. Grimes" Message-Id: <199504111724.KAA04039@gndrsh.aac.dev.com> Subject: Re: bin/333: cvs isn't being built To: nate@sneezy.sri.com Date: Tue, 11 Apr 1995 10:24:27 -0700 (PDT) Cc: pst@Shockwave.COM, freebsd-bugs@freefall.cdrom.com In-Reply-To: <199504111433.IAA08857@trout.sri.MT.net> from "Nate Williams" at Apr 11, 95 08:33:25 am X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 796 Sender: bugs-owner@FreeBSD.org Precedence: bulk > > Paul Traina writes: > > >Synopsis: CVS isn't being built > ... > > >Fix: > > > > fix /usr/src/gnu/usr.sbin/Makefile > > > > I seem to recall that cvs was removed for a reason (it was broken), but > > since Nate has been beating on it, I assume it's now fixed? > > It wasn't broken, but it has changed alot and Rod wanted to have it > tested more before enabling it. If he doesn't mind, I'd like to > re-enable it since I think the one minor boo-boo he saw wasn't critical > AND it hasn't trashed anyone's Repository. Agreed, enable it, as I can't duplicate the one problem I have seen and have seen nothing else go wrong. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Custom computers for FreeBSD From owner-freebsd-bugs Tue Apr 11 10:32:27 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id KAA24178 for bugs-outgoing; Tue, 11 Apr 1995 10:32:27 -0700 Received: from trout.sri.MT.net (trout.sri.MT.net [204.182.243.12]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id KAA24159 for ; Tue, 11 Apr 1995 10:32:20 -0700 Received: (from nate@localhost) by trout.sri.MT.net (8.6.11/8.6.10) id LAA09357; Tue, 11 Apr 1995 11:34:56 -0600 Date: Tue, 11 Apr 1995 11:34:56 -0600 From: Nate Williams Message-Id: <199504111734.LAA09357@trout.sri.MT.net> In-Reply-To: "Rodney W. Grimes" "Re: bin/333: cvs isn't being built" (Apr 11, 10:24am) X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: "Rodney W. Grimes" Subject: Re: bin/333: cvs isn't being built Cc: pst@Shockwave.COM, freebsd-bugs@freefall.cdrom.com Sender: bugs-owner@FreeBSD.org Precedence: bulk > Agreed, enable it, as I can't duplicate the one problem I have seen > and have seen nothing else go wrong. Okay, can I have you build a new CVS binary and install it on freefall CVS-meister? The speedups alone are worth the possibility of the *very* minor bug you ran into. (Not the spdup version, but the stock version) Nate From owner-freebsd-bugs Tue Apr 11 10:43:14 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id KAA24519 for bugs-outgoing; Tue, 11 Apr 1995 10:43:14 -0700 Received: (from davidg@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id KAA24499 ; Tue, 11 Apr 1995 10:43:12 -0700 Date: Tue, 11 Apr 1995 10:43:12 -0700 From: David Greenman Message-Id: <199504111743.KAA24499@freefall.cdrom.com> To: miff@spam.apana.org.au, davidg, freebsd-bugs Subject: Changed information for PR bin/334 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: -n option to strings does not work State-Changed-From-To: open-closed State-Changed-By: davidg State-Changed-When: Tue Apr 11 10:42:26 PDT 1995 State-Changed-Why: This bug was fixed on 2-2-95. From owner-freebsd-bugs Tue Apr 11 11:06:20 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA25195 for bugs-outgoing; Tue, 11 Apr 1995 11:06:20 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id LAA25189 for ; Tue, 11 Apr 1995 11:06:18 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.8/8.6.6) id LAA04255; Tue, 11 Apr 1995 11:05:56 -0700 From: "Rodney W. Grimes" Message-Id: <199504111805.LAA04255@gndrsh.aac.dev.com> Subject: Re: bin/333: cvs isn't being built To: nate@trout.sri.MT.net (Nate Williams) Date: Tue, 11 Apr 1995 11:05:56 -0700 (PDT) Cc: pst@Shockwave.COM, freebsd-bugs@freefall.cdrom.com In-Reply-To: <199504111734.LAA09357@trout.sri.MT.net> from "Nate Williams" at Apr 11, 95 11:34:56 am X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 698 Sender: bugs-owner@FreeBSD.org Precedence: bulk > > > Agreed, enable it, as I can't duplicate the one problem I have seen > > and have seen nothing else go wrong. > > Okay, can I have you build a new CVS binary and install it on freefall > CVS-meister? The speedups alone are worth the possibility of the *very* > minor bug you ran into. (Not the spdup version, but the stock version) I was actually going to try and do that.... give me until late tonigh, it is going to be a very busy day (I can tell just by the number of email messages I have and by the ringing of my phone :-)). -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Custom computers for FreeBSD From owner-freebsd-bugs Tue Apr 11 11:12:46 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA25506 for bugs-outgoing; Tue, 11 Apr 1995 11:12:46 -0700 Received: (from nate@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA25492 ; Tue, 11 Apr 1995 11:12:45 -0700 Date: Tue, 11 Apr 1995 11:12:45 -0700 From: Nate Williams Message-Id: <199504111812.LAA25492@freefall.cdrom.com> To: pst@Shockwave.COM, nate, freebsd-bugs Subject: Changed information for PR bin/333 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: CVS isn't being built State-Changed-From-To: open-closed State-Changed-By: nate State-Changed-When: Tue Apr 11 11:11:18 PDT 1995 State-Changed-Why: CVS is now being built since it has been adequately tested by the CVS Repository guru and is declared safe for general consumption From owner-freebsd-bugs Tue Apr 11 11:17:11 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA25728 for bugs-outgoing; Tue, 11 Apr 1995 11:17:11 -0700 Received: from trout.sri.MT.net (trout.sri.MT.net [204.182.243.12]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id LAA25718 for ; Tue, 11 Apr 1995 11:17:06 -0700 Received: (from nate@localhost) by trout.sri.MT.net (8.6.11/8.6.10) id MAA09598; Tue, 11 Apr 1995 12:21:08 -0600 Date: Tue, 11 Apr 1995 12:21:08 -0600 From: Nate Williams Message-Id: <199504111821.MAA09598@trout.sri.MT.net> In-Reply-To: "Rodney W. Grimes" "Re: bin/333: cvs isn't being built" (Apr 11, 11:05am) X-Mailer: Mail User's Shell (7.2.5 10/14/92) To: "Rodney W. Grimes" Subject: Re: bin/333: cvs isn't being built Cc: freebsd-bugs@freefall.cdrom.com Sender: bugs-owner@FreeBSD.org Precedence: bulk [ New CVS binary on freefall ] Rod writes: > I was actually going to try and do that.... give me until late tonigh, Thanks alot! Nate From owner-freebsd-bugs Tue Apr 11 11:33:06 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA26686 for bugs-outgoing; Tue, 11 Apr 1995 11:33:06 -0700 Received: (from ache@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA26672 ; Tue, 11 Apr 1995 11:33:03 -0700 Date: Tue, 11 Apr 1995 11:33:03 -0700 From: "Andrey A. Chernov" Message-Id: <199504111833.LAA26672@freefall.cdrom.com> To: pritc003@maroon.tc.umn.edu, ache, freebsd-bugs Subject: Changed information for PR kern/312 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: link system call on msdos filesystem causes panic State-Changed-From-To: open-closed State-Changed-By: ache State-Changed-When: Tue Apr 11 11:32:30 PDT 1995 State-Changed-Why: Fix applied. From owner-freebsd-bugs Tue Apr 11 11:51:45 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA28455 for bugs-outgoing; Tue, 11 Apr 1995 11:51:45 -0700 Received: (from ache@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA28434 ; Tue, 11 Apr 1995 11:51:43 -0700 Date: Tue, 11 Apr 1995 11:51:43 -0700 From: "Andrey A. Chernov" Message-Id: <199504111851.LAA28434@freefall.cdrom.com> To: edward@edcom.com, ache, freebsd-bugs Subject: Changed information for PR bin/330 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: lastcomm outputs garbage when command name is 10 chars long State-Changed-From-To: open-closed State-Changed-By: ache State-Changed-When: Tue Apr 11 11:51:04 PDT 1995 State-Changed-Why: Fix applied. From owner-freebsd-bugs Tue Apr 11 12:10:38 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id MAA29659 for bugs-outgoing; Tue, 11 Apr 1995 12:10:38 -0700 Received: (from ache@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id MAA29642 ; Tue, 11 Apr 1995 12:10:37 -0700 Date: Tue, 11 Apr 1995 12:10:37 -0700 From: "Andrey A. Chernov" Message-Id: <199504111910.MAA29642@freefall.cdrom.com> To: edward@edcom.com, ache, freebsd-bugs Subject: Changed information for PR kern/331 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: a read that returns short on a SCSI tape fails State-Changed-From-To: open-closed State-Changed-By: ache State-Changed-When: Tue Apr 11 12:09:47 PDT 1995 State-Changed-Why: Fix applied. From owner-freebsd-bugs Tue Apr 11 12:30:06 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id MAA00614 for bugs-outgoing; Tue, 11 Apr 1995 12:30:06 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id MAA00607 ; Tue, 11 Apr 1995 12:30:04 -0700 Date: Tue, 11 Apr 1995 12:30:04 -0700 Message-Id: <199504111930.MAA00607@freefall.cdrom.com> From: edward@edcom.com Reply-To: edward@edcom.com To: freebsd-bugs Subject: bin/335: ls -k with BLOCKSIZE=K uses 2k blocks In-Reply-To: Your message of Tue, 11 Apr 1995 12:22:01 -0700 <199504111922.MAA02366@edcom.com> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 335 >Category: bin >Synopsis: ls -k with BLOCKSIZE=K uses 2k blocks >Confidential: no >Severity: non-critical >Priority: medium >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Tue Apr 11 12:30:03 1995 >Originator: Edward Wang >Organization: UC Berkeley >Release: FreeBSD 2.1.0-Development i386 >Environment: 2.0-950322-SNAP BLOCKSIZE=K What does environment mean anyway? >Description: see below >How-To-Repeat: setenv BLOCKSIZE K dd if=/dev/zero of=x bs=1k count=10 /bin/ls -kls x says "5 -rw-r--r-- 1 edward staff 10240 Apr 11 12:12 x" >Fix: Actually, it's only the binary that's screwed up. The source looks fine, and worked when I compiled and installed it. So. Why am I reporting this? >Audit-Trail: >Unformatted: From owner-freebsd-bugs Tue Apr 11 12:41:29 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id MAA01296 for bugs-outgoing; Tue, 11 Apr 1995 12:41:29 -0700 Received: from irz301.inf.tu-dresden.de (irz301.inf.tu-dresden.de [141.76.1.11]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id MAA01280 for ; Tue, 11 Apr 1995 12:41:16 -0700 Received: from sax.sax.de by irz301.inf.tu-dresden.de with SMTP (5.67b+/DEC-Ultrix/4.3) id AA28789; Tue, 11 Apr 1995 21:40:45 +0200 Received: by sax.sax.de (8.6.9/8.6.9-s1) with UUCP id VAA04554; Tue, 11 Apr 1995 21:40:44 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.6.11/8.6.9) id UAA05218; Tue, 11 Apr 1995 20:21:44 +0200 From: J Wunsch Message-Id: <199504111821.UAA05218@uriah.heep.sax.de> Subject: Re: bin/334: -n option to strings does not work To: miff@spam.apana.org.au Date: Tue, 11 Apr 1995 20:21:44 +0200 (MET DST) Cc: freebsd-bugs@freefall.cdrom.com In-Reply-To: <199504111330.GAA19716@freefall.cdrom.com> from "miff@spam.apana.org.au" at Apr 11, 95 06:30:01 am Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) 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 Content-Length: 700 Sender: bugs-owner@FreeBSD.org Precedence: bulk As miff@spam.apana.org.au wrote: > > The -n number option to /usr/bin/strings is not implemented correctly. Any > attempt to specify a number generates an error. This has already been fixed before: revision 1.2 date: 1995/02/15 02:47:47; author: davidg; state: Exp; lines: +1 -1 >Description: The "strings" program chokes if you try using "-n". >How-To-Repeat: Try "strings -n SOMENUMBER SOMEFILE". >Fix: Here's a "diff -u" patch that corrects the problem. Submitted by: Lon Willett via NetBSD -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ Never trust an operating system you don't have sources for. ;-) From owner-freebsd-bugs Tue Apr 11 13:00:12 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA02796 for bugs-outgoing; Tue, 11 Apr 1995 13:00:12 -0700 Received: from relay1.UU.NET (relay1.UU.NET [192.48.96.5]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id NAA02780 for ; Tue, 11 Apr 1995 13:00:06 -0700 Received: from ast.com by relay1.UU.NET with SMTP id QQylat01567; Tue, 11 Apr 1995 15:59:44 -0400 Received: from trsvax.fw.ast.com (fw.ast.com) by ast.com with SMTP id AA01336 (5.67b/IDA-1.5 for uunet!pilchuck!rwing!eskimo!daveh); Tue, 11 Apr 1995 13:03:48 -0700 Received: by trsvax.fw.ast.com (/\=-/\ Smail3.1.18.1 #18.1) id ; Tue, 11 Apr 95 14:59 CDT Received: by nemesis.lonestar.org (Smail3.1.27.1 #18) id m0rym1x-0004vrC; Tue, 11 Apr 95 14:55 CDT Message-Id: Date: Tue, 11 Apr 95 14:55 CDT To: daveh@eskimo.com, bugs@FreeBSD.org From: uhclem@nemesis.lonestar.org (Frank Durda IV) Sent: Tue Apr 11 1995, 14:55:20 CDT Subject: matcd0: error during 2.0-950322-SNAP installation Cc: uhclem@nemesis.lonestar.org Sender: bugs-owner@FreeBSD.org Precedence: bulk [0]Then, when wd(0,a)/kernel went probing devices, the device matcd0: would [0]eventually come up for probing and/or attachment, and the kernel would [0]declare emphatically: [0] [0] matcd0: get_stat: after reading status byte, buss didn't go idle The problem is that at one of the addresses the driver is probing (which are 0x230, 0x240, 0x250, 0x260) there is some other device and that device is confusing the probe code. Removing the SoundBlaster card WILL NOT HELP, but adding a drive to it will. This has already been fixed, but until the new snap is available, your choices are: 1. Remove the offending adapter card at the above addresses or relocate them to other I/O ports until you can rebuild the kernel or get the new version. 2. Download the new version of matcd 1(16) from the -current tree. It now assumes that if the device doesn't honor a reset, it must not be the device it was looking for. The new SNAP (0408 or later) should have the newer version built-in. 3. Send me mail and I will send the new driver to you or a bootable kernel (0322-based). 4. Rebuild the kernel with a hard port specified (I know this one is tough if you don't already have a working system.) [0]re: kernel -c [0] No such device: matcd0 [0] Invalid command or syntax. This is another bug that the newer version fixes. The string that kernel -c is looking for is "matcd int", but I don't think there is any way to get it to accept a name with a space in it. The newer version makes "matcd" work in kernel -c. [0]After the disabling fiasco, I took the SoundBlaster 16 out of this [0]machine, thinking that matcd0: was meant to represent the Matsushita [0]cd-rom which attaches to these sound cards. I went through the entire [0]procedure described above, and once again the same problem appeared. The problem is that it is spotting some other device at one of the above addresses LOWER than the address that your SoundBlaster card is at *OR* you don't have a Matsushita drive attached at all, so the driver will look at ALL of the addresses listed above. You have to remove the conflicting devices (0322 RELEASE ONLY) to get around this problem. [0]So, the system I'm working with here is an i486DX2 66Mhz system with 16mb [0]RAM, a 540mb Western Digital HD, a 540mb Quantum SCSI HD and a Toshiba [0]XM3501B cd-rom (so far unrecognizable to FreeBSD because it's on a QLogic [0]VLB SCSI controller card), a SoundBlaster 16 (presumably this is what's [0]supposed to be attached to matcd0: ???), a Diamond SpeedStar Pro w 1mb [0]RAM, and an Intel EtherExpress 16, Try removing the network card or moving its address higher. The SCSI board may also be the source of the conflict. Don't give up - the next version is solves these problems. Frank Durda IV |"The Knights who say "LETNi" or uhclem%nemesis@trsvax.ast.com (Internet)| demand... A SEGMENT REGISTER!!!" ...letni!rwsys!nemesis!uhclem |"A what?" ...decvax!trsvax.fw.ast.com!nemesis!uhclem |"LETNi! LETNi! LETNi!" - 1983 From owner-freebsd-bugs Tue Apr 11 13:33:56 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA05068 for bugs-outgoing; Tue, 11 Apr 1995 13:33:56 -0700 Received: from ns1.win.net (ns1.win.net [204.215.209.3]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id NAA05060 for ; Tue, 11 Apr 1995 13:33:48 -0700 Received: (from bugs@localhost) by ns1.win.net (8.6.9/8.6.9) id QAA09904 for freebsd-bugs@freebsd.org; Tue, 11 Apr 1995 16:35:18 -0400 From: Mark Hittinger Message-Id: <199504112035.QAA09904@ns1.win.net> Subject: tcp back under control (thanks davidg) To: freebsd-bugs@FreeBSD.org Date: Tue, 11 Apr 1995 16:35:17 -0400 (EDT) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 205 Sender: bugs-owner@FreeBSD.org Precedence: bulk The problem I reported earlier about ftp/rcp hanging after ~400K and inetd stuck in an input wait is gone (in today's -current). It even shuts down cleanly now :-). Thanks Mark Hittinger bugs@win.net From owner-freebsd-bugs Tue Apr 11 13:40:06 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA05511 for bugs-outgoing; Tue, 11 Apr 1995 13:40:06 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA05504 ; Tue, 11 Apr 1995 13:40:05 -0700 Date: Tue, 11 Apr 1995 13:40:05 -0700 Message-Id: <199504112040.NAA05504@freefall.cdrom.com> From: Michael Paepcke Reply-To: Michael Paepcke To: freebsd-bugs Subject: kern/336: In-Reply-To: Your message of Tue, 11 Apr 1995 21:14:20 +0200 <199504111914.VAA01319@arcway.snafu.de> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 336 >Category: kern >Synopsis: >Confidential: no >Severity: serious >Priority: high >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Tue Apr 11 13:40:03 1995 >Originator: Michael Paepcke >Organization: arcway >Release: FreeBSD 2.1.0-Development i386 >Environment: SCSI Controller: ncr53c810 Harddisk : Quantum Empire 1400S (src-tree on Quantum, obj-tree on second SCSI Bus) (don't know >Description: (/usr/src) make world fails after 3-4 hours work every try on different places, the critical parts of the source tree compile fine last messages before the system stops: ncr0 targ 0? ERROR (81:10) (0-21-52) @ (c3c:0). reg: da 10 c0 13 47 8 0 1f 3 0 80 21 80 0 81 0. ncr0: have to clear fifos. ncr0: resart (fatal error). sd0(ncr0:0:0): COMMAND FAILED (9 ff) @f084a000. ncr0: reset by timeout. sd0(ncr0:0:0): FAST SCSI-2 100ns (10MB/sec) offset 8. [some vnode_pager / vm_fault error messages] >How-To-Repeat: ncr810 + quantum empire1400 + make world (or any other long & heavy disk access periods ???) ^^^^ >Fix: ? >Audit-Trail: >Unformatted: From owner-freebsd-bugs Tue Apr 11 16:01:37 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id QAA12739 for bugs-outgoing; Tue, 11 Apr 1995 16:01:37 -0700 Received: from isl.cf.ac.uk (isl-gate.elsy.cf.ac.uk [131.251.22.1]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id QAA12731 ; Tue, 11 Apr 1995 16:01:35 -0700 Received: (from paul@localhost) by isl.cf.ac.uk (8.6.9/8.6.9) id AAA05291; Wed, 12 Apr 1995 00:01:26 +0100 From: Paul Richards Message-Id: <199504112301.AAA05291@isl.cf.ac.uk> Subject: Re: Changed information for PR kern/331 To: ache@freefall.cdrom.com (Andrey A. Chernov) Date: Wed, 12 Apr 1995 00:01:26 +0100 (BST) Cc: edward@edcom.com, ache@freefall.cdrom.com, freebsd-bugs@freefall.cdrom.com In-Reply-To: <199504111910.MAA29642@freefall.cdrom.com> from "Andrey A. Chernov" at Apr 11, 95 12:10:37 pm X-Mailer: ELM [version 2.4 PL23] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 610 Sender: bugs-owner@FreeBSD.org Precedence: bulk In reply to Andrey A. Chernov who said > > Synopsis: a read that returns short on a SCSI tape fails > > State-Changed-From-To: open-closed > State-Changed-By: ache > State-Changed-When: Tue Apr 11 12:09:47 PDT 1995 > State-Changed-Why: > Fix applied. > Umm, please adopt the "standard" we're aiming for of referencing the version of the file you just changed. -- Paul Richards, FreeBSD core team member. Internet: paul@FreeBSD.org, URL: http://isl.cf.ac.uk/~paul/ Phone: +44 1222 874000 x6646 (work), +44 1222 457651 (home) Dept. Mechanical Engineering, University of Wales, College Cardiff. From owner-freebsd-bugs Tue Apr 11 21:40:06 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id VAA26988 for bugs-outgoing; Tue, 11 Apr 1995 21:40:06 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id VAA26981 ; Tue, 11 Apr 1995 21:40:04 -0700 Date: Tue, 11 Apr 1995 21:40:04 -0700 Message-Id: <199504120440.VAA26981@freefall.cdrom.com> From: Stephen Hocking Reply-To: Stephen Hocking To: freebsd-bugs Subject: i386/337: Seagate code hangs on bootup. In-Reply-To: Your message of Wed, 12 Apr 1995 14:29:44 +1000 <199504120429.EAA12627@netfl15a.devetir.qld.gov.au> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 337 >Category: i386 >Synopsis: seagate controller code hangs on boot. >Confidential: no >Severity: critical >Priority: medium >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Tue Apr 11 21:40:02 1995 >Originator: Stephen Hocking >Organization: DEVETIR >Release: FreeBSD 2.1.0-Development i386 >Environment: FreeBSD-2.1-current src-cur.0500 Hardware - 25MHz 386sx/386sx, 8Mb, S3911, TMC 885 SCSI controller, WREN 6 harddrive "CDC 94191-15 0136" >Description: On boot up, the machine hangs after identifying the controller, while attempting to attach devices. >How-To-Repeat: Get hold of appropriate hardware and boot. >Fix: I had some success by making a bunch of variables volatile. However this was still unstable and would usually start printing out timeouts shortly after boot, if not during. This often had sad consequences for the integrity of the filesystems. The final fix was to drag the source off the 2.0 CDROM and undefine SEA_BLINDTRANSFER and SEA_ASSEMBLER (at line 107 and 108 of /sys/i386/isa/seagate.c). It is slow, but at the moment it is partway through a "make world" without complaining. This controller is at the moment eating upto 50% of the CPU on interrupts, the rest mostly going on system CPU. Why am I running this pile of canine ordure? My beloved Buslogic 542B finally died after many years of service and I was desperate & cashless. The Buslogic card would get by on 3-5% interrupt CPU, some system time (mostly fs code) leaving plenty of idle CPU for other things. It could drive the disk at platter speed, the seagate only manages about 30% (if that). I do not speak for the Worker's Compensation Board of Queensland - They don't pay me enough for that! >Audit-Trail: >Unformatted: From owner-freebsd-bugs Wed Apr 12 00:30:02 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id AAA00440 for bugs-outgoing; Wed, 12 Apr 1995 00:30:02 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id AAA00432 ; Wed, 12 Apr 1995 00:30:01 -0700 Date: Wed, 12 Apr 1995 00:30:01 -0700 Message-Id: <199504120730.AAA00432@freefall.cdrom.com> From: pritc003@maroon.tc.umn.edu Reply-To: pritc003@maroon.tc.umn.edu To: freebsd-bugs Subject: bin/338: Cron allows users to obtain root access In-Reply-To: Your message of Wed, 12 Apr 1995 02:23:32 -0500 <199504120723.CAA04652@mpp.com> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 338 >Category: bin >Synopsis: Cron allows users to obtain root access >Confidential: no >Severity: serious >Priority: high >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Wed Apr 12 00:30:01 1995 >Originator: Mike Pritchard >Organization: >Release: FreeBSD 2.1.0-Development i386 >Environment: >Description: Cron can allow users to obtain root access the same way that the atrun command allowed. >How-To-Repeat: See all of the relavent "atrun" mail for a detailed description. Basically, just modify the MAILTO variable in a users crontab file in the same way that the atrun mail information was changed to spoof sendmail. >Fix: Cron was changed to only accept MAILTO variables that point to user names that actually exist in the password file. This disallows users from passing arguments to sendmail and spoofing it into running as root and giving them root access. If cron detected that a user name did not exist, and it is able to find a user name to send the mail to, it will add a: X-Cron-Err: User specified in MAILTO variable does not exist header line to the mail message to give the user some idea of why their mail isn't going where they wanted. Here is the patch to /usr/src/usr.sbin/cron/cron/do_command.c to fix the problem: *** orig/do_command.c Wed Apr 12 01:31:00 1995 --- do_command.c Wed Apr 12 01:58:28 1995 *************** *** 28,33 **** --- 28,34 ---- #if defined(SYSLOG) # include #endif + #include static void child_process __P((entry *, user *)), *************** *** 76,81 **** --- 77,83 ---- int stdin_pipe[2], stdout_pipe[2]; register char *input_data; char *usernm, *mailto; + int badmailto = 0; int children = 0; Debug(DPROC, ("[%d] child_process('%s')\n", getpid(), e->cmd)) *************** *** 358,366 **** --- 360,385 ---- /* if we are supposed to be mailing, MAILTO will * be non-NULL. only in this case should we set * up the mail command and subjects and stuff... + * + * Verify that the username we are being asked to + * mail to actually exists. This is to prevent + * spoofing the mail delivery program. */ if (mailto) { + struct passwd *pwd; + + if ((pwd = getpwnam(mailto)) == NULL) { + Debug(DPROC, ("[%d] MAILTO user (%s) does not exist\n", + getpid(), mailto)) + badmailto = 1; + if ((pwd = getpwnam(usernm)) != NULL) + mailto = usernm; + else + mailto = NULL; + } + } + if (mailto) { register char **env; auto char mailcmd[MAX_COMMAND]; auto char hostname[MAXHOSTNAMELEN]; *************** *** 384,389 **** --- 403,415 ---- for (env = e->envp; *env; env++) fprintf(mail, "X-Cron-Env: <%s>\n", *env); + /* + * Give the user some idea of why their mail + * didn't go where they expected if we + * rejected their MAILTO variable. + */ + if (badmailto) + fprintf(mail, "X-Cron-Err: User specified in MAILTO variable does not exist\n"); fprintf(mail, "\n"); /* this was the first char from the pipe >Audit-Trail: >Unformatted: From owner-freebsd-bugs Wed Apr 12 04:01:13 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id EAA09409 for bugs-outgoing; Wed, 12 Apr 1995 04:01:13 -0700 Received: from regen.dkrz.de (regen.dkrz.de [136.172.60.241]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id EAA09403 for ; Wed, 12 Apr 1995 04:01:05 -0700 X400-Received: by /PRMD=dkrz/ADMD=d400/C=de/; Relayed; 12 Apr 95 12:55:20+0200 Date: 12 Apr 95 12:55:20+0200 From: "Georg-W. Koltermann" Message-ID: <9504121055.AA08106@racer.dkrz.de> To: bde@zeta.org.au cc: bugs@FreeBSD.org, nate@sneezy.sri.com, smace@metal-mail.neosoft.com References: , <8 Apr 1995 22:56:47 +1000> Subject: Re: 80387 hangs system at divide by zero Reply-To: gwk@cray.com Sender: bugs-owner@FreeBSD.org Precedence: bulk Hi Bruce, thanks for explaining in long what's wrong with my '387 setup. Also thanks to the others telling me ULSI '387s are broke (especially to Nate who is from Montana... :) . Well, I did some more testing last night. Bruce Evans wrote: > Some buggy i387 (in)compatibles and/or motherboards are reported to lock > up the whole system (interrupts stop working; I think this is an i387 > bug); others are reported to only stop IRQ13 working (I think this is > a motherboard problem). FreeBSD makes no attempt to handle either > problem. Linux handles the second problem using a timeout, so an i387 > error only wastes an average of 1/2 a clock tick. If I generate an overflow or divide by zero on MS-DOG(tm), the result prints out as "infinity". I tried with Symantec C++ 6.11 as well as gcc 2.6.3 (the EMX port). Both environments mask off all exceptions by default. If I enable exceptions, the program will hang just as with FreeBSD. In case of gcc/MS-DOG I can ctrl-break out of the hang, but the system gets somewhat instable--the next invocation of the gcc compiler will hang irreversibly. With Linux [Slackware 2.1 binaries with a 1.1.78 kernel, gcc 2.5.8] both overflow and divide by zero are reported as floating point exception. Linux can handle it! When I mask off exceptions, results are printed as infinity just as with MS-DOG. On FreeBSD overflow and divide by zero hang my system irreversibly. Can't even switch consoles, or CTRL-ALT-DEL, only reset will help. > Actually it should be possible to use the chip in IEEE mode (with all > exceptions masked). This should be the default mode, and probably will > be in FreeBSD-2.1. Yup, when I mask off exceptions on FreeBSD my test program works and prints the results as infinity. I only get a warning when the program terminates stating that it terminated with pending exceptions (or something to that effect). Is there a way that I can mask off exceptions globally for every program on my system? Or would you think there is a chance that FreeBSD gets as good as coping with my hardware as Linux? I have already tried to compare FreeBSD and Linux sources, the parts that do exception handling. Not much luck though. As the German saying goes: "I couln't find the forest because there were so many trees". Georg-W. Koltermann, gwk@cray.com P.S.: For the curious I will append my test program here. Yes, it's a quick and dirty one... ---------------------------------------------------------------------- #include #if linux # include #endif #if __FreeBSD__ # include #endif static void divide (double divisor) { double dividend = 42.0; printf ("%g / %g = %g\n", dividend, divisor, dividend / divisor); } int main (int argc, char *argv []) { #if linux unsigned control = 0; #define fnstcw(addr) __asm("fnstcw %0" : : "m" (*(addr))) fnstcw (&control); printf ("fpu_control = 0x%x\n", control); __setfpucw (control & ~0xd); sleep (1); fnstcw (&control); printf ("fpu_control now 0x%x\n", control); #elif __FreeBSD__ #define fldcw(addr) __asm("fldcw %0" : : "m" (*(addr))) /* previous #define stolen from npx.c--why is it missing from floatingpoint.h ?? */ unsigned control = 0; setvbuf (stdout, NULL, _IONBF, 0); __fnstcw (&control); printf ("x387 control word: 0x%x\n", control); control &= ~0xd; fldcw (&control); __fnstcw (&control); printf ("x387 control word now: 0x%x\n", control); #else /* MS-DOG */ printf ("x387 control word: 0x%x\n", _control87 (0x0, 0xd)); printf ("x387 control word now: 0x%x\n", _control87 (0, 0)); #endif divide (7.0); divide (3e-308); divide (0.0); return 0; } From owner-freebsd-bugs Wed Apr 12 06:46:29 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id GAA13471 for bugs-outgoing; Wed, 12 Apr 1995 06:46:29 -0700 Received: from munich.gcomm.com (munich.gcomm.com [199.227.15.15]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id GAA13464 for ; Wed, 12 Apr 1995 06:46:28 -0700 Received: (from craig@localhost) by munich.gcomm.com (8.6.8.1/8.6.6) id JAA27823 for bugs@freebsd.org; Wed, 12 Apr 1995 09:48:27 -0400 From: Craig Yap Message-Id: <199504121348.JAA27823@munich.gcomm.com> Subject: seekdir bug To: bugs@FreeBSD.org Date: Wed, 12 Apr 1995 09:48:27 -0400 (EDT) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 1599 Sender: bugs-owner@FreeBSD.org Precedence: bulk There is a bug when seeking through a directory. telldir() seems to return a position in the directory that doesnt match the name of the file. I've included a program that will exploit the bug. I ran it as followed: First made some test files: touch 1 2 3 4 then ran the program: ./a "?" I will continually show the name=a eventhough the position is incremented Thanks for your time, Craig /* * test the viability of seekdir() after closedir() * * requires: fnmatch() and dirent routines */ #include #include static long position=-1; char * findfile(char *spec) { struct dirent *dent; DIR *dirp; if ((dirp=opendir(".")) == NULL) { perror("opendir"); exit(2); } if (position != -1) { seekdir(dirp,position); } while((dent=readdir(dirp)) != NULL) { if (fnmatch(spec,dent->d_name,0) == 0 && strcmp(dent->d_name,".") != 0 && strcmp(dent->d_name,"..") != 0) { position=telldir(dirp); closedir(dirp); return(dent->d_name); } } closedir(dirp); return(NULL); } main( int argc, char *argv[]) { char *fname; if (argc != 2) { printf("usage: testdirent \"\"\n"); return(1); } while ((fname=findfile(argv[1])) != NULL) { printf("name=%s position=%ld\n",fname,position); } return(0); } -- ------------------------------------------------------------------------------ Craig Yap craig@munich.gcomm.com Software Engineer Unix Technical Division Galacticomm, Inc From owner-freebsd-bugs Wed Apr 12 11:20:08 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA19592 for bugs-outgoing; Wed, 12 Apr 1995 11:20:08 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA19585 ; Wed, 12 Apr 1995 11:20:07 -0700 Date: Wed, 12 Apr 1995 11:20:07 -0700 Message-Id: <199504121820.LAA19585@freefall.cdrom.com> From: Paul Traina Reply-To: Paul Traina To: freebsd-bugs Subject: bin/339: kerberos violates s/key interaction rules In-Reply-To: Your message of Wed, 12 Apr 1995 11:14:08 -0700 <199504121814.LAA24399@precipice.shockwave.com> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 339 >Category: bin >Synopsis: users may enter kerberos password at login prompt >Confidential: no >Severity: serious >Priority: medium >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Wed Apr 12 11:20:04 1995 >Originator: Paul Traina >Organization: Shockwave Engineering >Release: FreeBSD 2.1.0-Development i386 >Environment: FreeBSD with eBones made and installed, s/key enabled for a user, kerberos tickets available for a user. >Description: There's a disconnect between kerberos and s/key access rules. If I restrict password logins using /etc/skey.access in order to force users to use one-time passwords (or a kerberos ticket), if Kerberos is enabled, a user may enter their kerberos password at the login prompt to gain access to the system. The whole point of /etc/skey.access is to stop people from entering passwords over the net, so the /etc/skey.access system should apply to locally entered kerberos tickets at the login prompt as well. >How-To-Repeat: pst@precipice$ rlogin -K remote-host s/key 98 qu08742 (s/key required) Password: Last login: Wed Apr 12 10:54:44 from precipice Copyright (c) 1980, 1983, 1986, 1988, 1990, 1991, 1993, 1994 The Regents of the University of California. All rights reserved. FreeBSD 2.1.0-Development (QUEMADURA) #0: Tue Apr 11 11:54:26 PDT 1995 Welcome to FreeBSD! >Fix: This isn't totally trivial, because you want to allow kerberos athentication to occur if a remote kerberos ticket has been validated. >Audit-Trail: >Unformatted: From owner-freebsd-bugs Wed Apr 12 13:55:22 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA24082 for bugs-outgoing; Wed, 12 Apr 1995 13:55:22 -0700 Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.34]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id NAA24055 for ; Wed, 12 Apr 1995 13:54:42 -0700 Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id GAA25937; Thu, 13 Apr 1995 06:52:28 +1000 Date: Thu, 13 Apr 1995 06:52:28 +1000 From: Bruce Evans Message-Id: <199504122052.GAA25937@godzilla.zeta.org.au> To: bugs@FreeBSD.org, craig@munich.gcomm.com Subject: Re: seekdir bug Sender: bugs-owner@FreeBSD.org Precedence: bulk >There is a bug when seeking through a directory. telldir() seems to return >a position in the directory that doesnt match the name of the file. I've >included a program that will exploit the bug. I ran it as followed: The man page says that "Values returned by telldir() are good only for the lifetime of the DIR pointer, dirp, from which they are derived." However, it also says that "It is safe to use a previous telldir() value immediately after a call to opendir() and before any calls to readdir()." The second statement is now wrong because telldir() cookies are reclaimed in closedir(). telldir() cookies are malloc()ed and (I think) the memory for them used to be lost in closedir(). Bruce From owner-freebsd-bugs Wed Apr 12 14:34:10 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id OAA26038 for bugs-outgoing; Wed, 12 Apr 1995 14:34:10 -0700 Received: from cs.weber.edu (cs.weber.edu [137.190.16.16]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id OAA26019 for ; Wed, 12 Apr 1995 14:34:06 -0700 Received: by cs.weber.edu (4.1/SMI-4.1.1) id AA19075; Wed, 12 Apr 95 15:27:13 MDT From: terry@cs.weber.edu (Terry Lambert) Message-Id: <9504122127.AA19075@cs.weber.edu> Subject: Re: seekdir bug To: craig@munich.gcomm.com (Craig Yap) Date: Wed, 12 Apr 95 15:27:12 MDT Cc: bugs@FreeBSD.org In-Reply-To: <199504121348.JAA27823@munich.gcomm.com> from "Craig Yap" at Apr 12, 95 09:48:27 am X-Mailer: ELM [version 2.4dev PL52] Sender: bugs-owner@FreeBSD.org Precedence: bulk > There is a bug when seeking through a directory. telldir() seems to return > a position in the directory that doesnt match the name of the file. I've > included a program that will exploit the bug. I ran it as followed: [ ... seekdir after close then reopen of dir ... ] If there is a bug, it's in the man page. Here's what the Sun man page has to say: seekdir() sets the position of the next readdir() operation on the directory stream. The new position reverts to the one associated with the directory stream when the telldir() operation was performed. Values returned by telldir() are good only for the lifetime of the DIR pointer from which they are derived. If the directory is closed and then reopened, the telldir() value may be invalidated due to undetected directory compaction. It is safe to use a previ- ous telldir() value immediately after a call to opendir() and before any calls to readdir. Admittedly, no compation is taking place in your use of the code, at least not from the code itself. The main problem is that telldir/seekdir imply the ability to save state across opens; in reality, they are abominations, and their use is to be frowned upon. The way directory entries are returned in the opendir is through the use of the internal system call getdents() which is supposed to return an atomic block of directory entries (an atomic block being defined as a block of data wherin changes to the data will not span a block boundry). In other words, getdents() takes a snapshot that is guranteed to be consistant for an external observer. Then the readdir/telldir operations operate on the user space copy of the block. The telldir() returns an offset which is not supposed to be a seek offset in the file itself, but a block/offset_into_block combination. A lot of the current code assumes that it is an actual file offset -- this is bogus, since POSIX does not guarantee that directories are themselves represented via a file abstraction in the first place (it just so happens that for the majority of UNIX file systems that they are). So the first problem is that you are violating the usage protocol, and the second problem is that you are doing so with an interface that doesn't guarantee atomicity for lookup and directory modification operations. Terry Lambert terry@cs.weber.edu --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-bugs Wed Apr 12 14:44:17 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id OAA26677 for bugs-outgoing; Wed, 12 Apr 1995 14:44:17 -0700 Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.34]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id OAA26664 for ; Wed, 12 Apr 1995 14:43:47 -0700 Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id HAA26990; Thu, 13 Apr 1995 07:39:50 +1000 Date: Thu, 13 Apr 1995 07:39:50 +1000 From: Bruce Evans Message-Id: <199504122139.HAA26990@godzilla.zeta.org.au> To: gwk@cray.com Subject: Re: 80387 hangs system at divide by zero Cc: bugs@FreeBSD.org, nate@sneezy.sri.com, smace@metal-mail.neosoft.com Sender: bugs-owner@FreeBSD.org Precedence: bulk >If I generate an overflow or divide by zero on MS-DOG(tm), the result >prints out as "infinity". I tried with Symantec C++ 6.11 as well as >gcc 2.6.3 (the EMX port). Both environments mask off all exceptions >by default. If I enable exceptions, the program will hang just as Interesting. An old version of Turbo C++ (1.0) unmasks some exceptions by default. Turbo Debugger mishandles the resulting exceptions. >with FreeBSD. In case of gcc/MS-DOG I can ctrl-break out of the hang, >but the system gets somewhat instable--the next invocation of the gcc >compiler will hang irreversibly. This might be because the IRQ13 for the exception never occurred, so the FPU busy latch didn't get cleared. The next FPU instruction will then hang (I'm not sure if it hangs irreversibly). >With Linux [Slackware 2.1 binaries with a 1.1.78 kernel, gcc 2.5.8] >both overflow and divide by zero are reported as floating point >exception. Linux can handle it! When I mask off exceptions, results >are printed as infinity just as with MS-DOG. I think it clears the busy latch after an FPU hang. You could try this under DOS: break out of a hang, then use DEBUG to clear the hang "o f0 0" command, then see if gcc still hangs. >Is there a way that I can mask off exceptions globally for every >program on my system? Or would you think there is a chance that You can change the #define of __INITIAL_NPXCW__ in from __BDE_NPXCW__ to __BETTER_BDE_NPXCW__, or arrange to set the control word in crt0 like Linux does. There is no way to stop an application from changing it to something harmful. >P.S.: For the curious I will append my test program here. Yes, it's a > quick and dirty one... >... >#elif __FreeBSD__ >#define fldcw(addr) __asm("fldcw %0" : : "m" (*(addr))) > /* previous #define stolen from npx.c--why is it missing from > floatingpoint.h ?? */ floatingpoint.h uses `fnstenv(...); modify_env(); fldenv();' instead. This is more general but slower. Here is a better test program. It uses direct asm() statements. Unfortunately this is more portable than the C interfaces to the special FPU registers. You won't be able to run this under FreeBSD until the hangs are fixed; under Linux it should be a good test for the workarounds for the hangs. --- /* fpetest.c - test x87/486 floating point error handling */ #ifndef __FreeBSD__ #include #endif #include #include #include #define CW_ZM (1 << 2) /* divide by zero mask */ #define SW_BUSY (1 << 15) /* FPU busy */ #define SW_ES (1 << 7) /* exception summary */ #define SW_ZE (1 << 2) /* divide by zero (pending) exception seen */ #define NR_TESTS 15 #define TEST(n, x, y) ( setup(), ({ asm(x); 0; }), check(n, x, y) ) static double double_in_mem; static unsigned short fp_cw; static unsigned fp_env[7]; static unsigned fp_state[7 + 8 * 10 / sizeof(unsigned)]; static unsigned short fp_sw; static unsigned long max_failures; static unsigned long max_tests; static unsigned short msw; static unsigned char silent; static volatile sig_atomic_t sigfpe_handled; static volatile sig_atomic_t sigint_handled; static void delay(void); static int check(int test_nr, char *insn, int sigfpe_expected); static void report_control_words(char *when); static void setup(void); static void sigfpe_handler(int sig_num); static void sigint_handler(int sig_num); static int check(int test_nr, char *insn, int sigfpe_expected) { static char *insns[NR_TESTS]; static unsigned long failures[NR_TESTS]; static int prev_passed[NR_TESTS]; static unsigned long tests[NR_TESTS]; static unsigned long total_failures; static unsigned long total_tests; int passed; if (sigfpe_handled) { signal(SIGFPE, sigfpe_handler); sigfpe_handled = 0; passed = sigfpe_expected; } else passed = !sigfpe_expected; insns[test_nr] = insn; ++tests[test_nr]; ++total_tests; if (!passed) { ++failures[test_nr]; ++total_failures; if (!silent && (failures[test_nr] != 1 || passed == prev_passed[test_nr])) fprintf(stderr, "T %lu F %lu: %s SIGFPE for `%s'\n", total_tests, total_failures, sigfpe_expected ? "no" : " ", insn); } if (total_failures >= max_failures || total_tests >= max_tests || sigint_handled) { fprintf(stderr, " Tests Failures Test\n"); for (test_nr = 0; test_nr < NR_TESTS; ++test_nr) if (insns[test_nr] != NULL) fprintf(stderr, "%7lu %7lu `%s'\n", tests[test_nr], failures[test_nr], insns[test_nr]); fprintf(stderr, "%7lu %7lu totals\n", total_tests, total_failures); report_control_words("Final"); exit(total_failures ? 1 : 0); } prev_passed[test_nr] = passed; return passed; } static void delay(void) { volatile unsigned countdown; for (countdown = 100; countdown != 0; --countdown) ; } int main(argc, argv) int argc; char **argv; { unsigned char skip3; unsigned char skip4; int opt; report_control_words("Initial"); max_failures = 2 * NR_TESTS; max_tests = 2 * NR_TESTS; skip3 = 0; skip4 = 0; silent = 0; while ((opt = getopt(argc, argv, "34f:st:")) != EOF) switch(opt) { case '3': skip3 = 1; break; case '4': skip3 = 1; skip4 = 1; break; case 'f': max_failures = strtoul(optarg, (char **) NULL, 0); break; case 's': silent = 1; break; case 't': max_tests = strtoul(optarg, (char **) NULL, 0); break; default: fprintf(stderr, "usage: %s [-34s] [-f max-failures] [-t max-tests]\n", argv[0]); fprintf(stderr, " -3 = skip tests that often fail on 386's\n"); fprintf(stderr, " -4 = skip tests that often fail on [3-4]86's\n"); fprintf(stderr, " -s = silent\n"); exit(2); } signal(SIGINT, sigint_handler); while (1) { /* * Cause a divide by zero error. This should not trigger an exception. * The next no-wait FP instruction should trigger the exception. */ if (TEST(0, "fldz; fld1; fdiv %st,%st(1)", 0)) { /* * The fwait instruction should always trigger a pending exception. */ TEST(1, "fldz; fld1; fdiv %st,%st(1); fwait", 1); /* * One way for this to fail is if the kernel uses CR0_EM instead * of CR0_TS | CR0_MP to handle FP context switching. This fails * to trap fwaits immediately after an FP context switch. It is * especially bad when FP is being emulated. Then all fwaits are * ignored! */ TEST(2, "fldz; fld1; fdiv %st,%st(1); nop; fwait", 1); TEST(3, "fldz; fld1; fdiv %st,%st(1); call _delay; fwait", 1); /* * No-wait instructions should never trigger a pending exception. * * On my 486 system, they are all broken when the IRQ13 FP * exception reporting method is used. On at least one 386 system, * fnclex usually works but some of the others are broken, and * fnclex fails after a context switch, presumably because frstor * fails in the kernel. */ if (!skip3) { TEST(4, "fldz; fld1; fdiv %st,%st(1); fninit", 0); TEST(5, "fldz; fld1; fdiv %st,%st(1); fnstcw _fp_cw", 0); TEST(6, "fldz; fld1; fdiv %st,%st(1); fnstsw _fp_sw", 0); } if (!skip4) TEST(7, "fldz; fld1; fdiv %st,%st(1); fnclex", 0); if (!skip3) { TEST(8, "fldz; fld1; fdiv %st,%st(1); fnstenv _fp_env", 0); TEST(9, "fldz; fld1; fdiv %st,%st(1); fnsave _fp_state", 0); } } /* * fldenv and frstor of an error state should not trigger an exception, * and they should not lose the pending exception. Fake the pending * exception so that these tests can be done even if the tests for * fnstenv and fnsave of the pending exception failed. */ if (!skip3) { setup(); asm("fnstenv _fp_env"); /* an almost clean env */ fp_env[1] |= SW_BUSY | SW_ES | SW_ZE; /* fake excepttion */ asm("fldenv _fp_env"); if (check(10, "fldenv of pending exception", 0)) { delay(); asm("fwait"); check(11, "fwait after fldenv of pending exception", 1); } setup(); asm("fnsave _fp_state"); /* an almost clean state */ fp_state[1] |= SW_BUSY | SW_ES | SW_ZE; /* fake excepttion */ asm("frstor _fp_state"); if (check(12, "frstor of pending exception", 0)) { delay(); asm("fwait"); check(13, "fwait after frstor of pending exception", 1); } } /* * fstpl to memory when the FP stack is empty sometimes causes an * IRQ13 a little after the intstruction. When the fstpl is traced, * the exception appears to come from the trace trap handler! frstor * of a pending error may also cause an IRQ13 after the instruction. * * XXX - this test has never been observed to fail. */ TEST(14, "fstpl _double_in_mem", 0); } } static void report_control_words(char *when) { asm("smsw _msw"); asm("fnstcw _fp_cw"); fprintf(stderr, "%7s machine status word = %04x\n", when, msw); fprintf(stderr, "%7s FP control word = %04x\n", when, fp_cw); } static void setup(void) { asm("fninit; fnstcw _fp_cw"); fp_cw &= ~CW_ZM; asm("fldcw _fp_cw"); signal(SIGFPE, sigfpe_handler); sigfpe_handled = 0; } static void sigfpe_handler(int sig_num) { sigfpe_handled = 1; } static void sigint_handler(int sig_num) { sigint_handled = 1; } --- Bruce From owner-freebsd-bugs Wed Apr 12 15:37:50 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id PAA28768 for bugs-outgoing; Wed, 12 Apr 1995 15:37:50 -0700 Received: from cs.weber.edu (cs.weber.edu [137.190.16.16]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id PAA28756 for ; Wed, 12 Apr 1995 15:37:45 -0700 Received: by cs.weber.edu (4.1/SMI-4.1.1) id AA19462; Wed, 12 Apr 95 16:30:39 MDT From: terry@cs.weber.edu (Terry Lambert) Message-Id: <9504122230.AA19462@cs.weber.edu> Subject: Re: seekdir bug To: bde@zeta.org.au (Bruce Evans) Date: Wed, 12 Apr 95 16:30:38 MDT Cc: bugs@FreeBSD.org, craig@munich.gcomm.com In-Reply-To: <199504122052.GAA25937@godzilla.zeta.org.au> from "Bruce Evans" at Apr 13, 95 06:52:28 am X-Mailer: ELM [version 2.4dev PL52] Sender: bugs-owner@FreeBSD.org Precedence: bulk > The man page says that "Values returned by telldir() are good only for > the lifetime of the DIR pointer, dirp, from which they are derived." > However, it also says that "It is safe to use a previous telldir() value > immediately after a call to opendir() and before any calls to > readdir()." The second statement is now wrong because telldir() cookies > are reclaimed in closedir(). telldir() cookies are malloc()ed and (I > think) the memory for them used to be lost in closedir(). It's safe to use, but it may result in seekdir() succeeding and a subsequent readdir() returning NULL for a read at the now invalid position. Oh, non-linear iteration of directory entries should be banned, yes indeedy. 8-(. Terry Lambert terry@cs.weber.edu --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-bugs Wed Apr 12 15:42:55 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id PAA28917 for bugs-outgoing; Wed, 12 Apr 1995 15:42:55 -0700 Received: from localhost (localhost [127.0.0.1]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id PAA28908 ; Wed, 12 Apr 1995 15:42:49 -0700 X-Authentication-Warning: freefall.cdrom.com: Host localhost didn't use HELO protocol To: gwk@cray.com cc: bde@zeta.org.au, bugs@FreeBSD.org, nate@sneezy.sri.com, smace@metal-mail.neosoft.com, phk@FreeBSD.org Subject: Re: 80387 hangs system at divide by zero In-reply-to: Your message of "12 Apr 95 12:55:20 +0200." <9504121055.AA08106@racer.dkrz.de> Date: Wed, 12 Apr 1995 15:42:47 -0700 Message-ID: <28907.797726567@freefall.cdrom.com> From: "Jordan K. Hubbard" Sender: bugs-owner@FreeBSD.org Precedence: bulk > Hi Bruce, > > thanks for explaining in long what's wrong with my '387 setup. Also > thanks to the others telling me ULSI '387s are broke (especially to > Nate who is from Montana... :) . This is just one of a number of examples that have been cited lately where Linux/DOS runs on hardware that FreeBSD won't touch with a pole. DOS running I can live with since we all know it doesn't use the machine anywhere nearly comprehensively enough to trigger problems, but when Linux copes and we don't, that stings. People don't care WHY Linux works, or that their hardware is evil and braindead and should be replaced, not papered over, so when this situation is encountered we get a: "Why does Linux work with my computer when FreeBSD just dies?" In the newsgroups and it makes us look bad. I would be much happier if one of the core people who actually know about this kind of stuff (and there are a few of you who do - no use trying to hide! ;) were to take it on as a personal mission to bring Linux's compatability hacks to FreeBSD, perhaps placing them in a lead-lined directory in our kernel source tree with a sign saying "EVIL WORK-AROUND HACKS! DON'T LOOK!" placed in a prominent position. I'm not talking about hacking our core technology to support evilness, that would just come back to bite us later, but can we somehow segregate it under the COMPAT_HACK label and pull even with Linux on these silly boxes again? Thanks! Jordan From owner-freebsd-bugs Wed Apr 12 16:25:42 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id QAA00339 for bugs-outgoing; Wed, 12 Apr 1995 16:25:42 -0700 Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.34]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id QAA00298 ; Wed, 12 Apr 1995 16:24:57 -0700 Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id JAA30027; Thu, 13 Apr 1995 09:22:39 +1000 Date: Thu, 13 Apr 1995 09:22:39 +1000 From: Bruce Evans Message-Id: <199504122322.JAA30027@godzilla.zeta.org.au> To: gwk@cray.com, jkh@freefall.cdrom.com Subject: Re: 80387 hangs system at divide by zero Cc: bde@zeta.org.au, bugs@FreeBSD.org, nate@sneezy.sri.com, phk@FreeBSD.org, smace@metal-mail.neosoft.com Sender: bugs-owner@FreeBSD.org Precedence: bulk >In the newsgroups and it makes us look bad. I would be much happier >if one of the core people who actually know about this kind of stuff >(and there are a few of you who do - no use trying to hide! ;) were to >take it on as a personal mission to bring Linux's compatability hacks >to FreeBSD, perhaps placing them in a lead-lined directory in our I need interrupts enabled at probe time to use the Linux hack. I've been waiting for this for 2.5 years. npx.c already has about 5 times as much code as Linux to handle i387 hardware bugs. It needs to have more because of the missing interrupts and other complications in our environment (such as paging the FPU state instead of keeping it in the process table). It has more than necessary because minimal fixes are too hard to test. Bruce From owner-freebsd-bugs Wed Apr 12 19:16:16 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id TAA07037 for bugs-outgoing; Wed, 12 Apr 1995 19:16:16 -0700 Received: from localhost (localhost [127.0.0.1]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id TAA07030 ; Wed, 12 Apr 1995 19:16:11 -0700 X-Authentication-Warning: freefall.cdrom.com: Host localhost didn't use HELO protocol To: Bruce Evans cc: gwk@cray.com, bugs@FreeBSD.org, nate@sneezy.sri.com, phk@FreeBSD.org, smace@metal-mail.neosoft.com Subject: Re: 80387 hangs system at divide by zero In-reply-to: Your message of "Thu, 13 Apr 95 09:22:39 +1000." <199504122322.JAA30027@godzilla.zeta.org.au> Date: Wed, 12 Apr 1995 19:16:06 -0700 Message-ID: <7028.797739366@freefall.cdrom.com> From: "Jordan K. Hubbard" Sender: bugs-owner@FreeBSD.org Precedence: bulk > I need interrupts enabled at probe time to use the Linux hack. I've > been waiting for this for 2.5 years. npx.c already has about 5 times > as much code as Linux to handle i387 hardware bugs. It needs to have So what do we need to get this long-delayed functionality? Anyone you think could do it and should be prodded into doing so? We're about to seriously go head-to-head with Linux for "mindshare" in 2.1 (trust me, folks, this is imminent :-) as we start playing dueling PR campaigns, and I'd really rather not have magazines coming back at me with "Linux worked in our testing lab - FreeBSD fell over" as I'm working overtime to get them to publish articles about us. Thanks! Jordan From owner-freebsd-bugs Wed Apr 12 21:14:16 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id VAA09775 for bugs-outgoing; Wed, 12 Apr 1995 21:14:16 -0700 Received: from main.statsci.com (main.statsci.com [198.145.125.110]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id VAA09769 for ; Wed, 12 Apr 1995 21:14:14 -0700 Received: by main.statsci.com (Smail3.1.28.1 #9) id m0rzGIA-000r3vC; Wed, 12 Apr 95 21:14 PDT Message-Id: To: "Jordan K. Hubbard" cc: bugs@FreeBSD.org Subject: Re: 80387 hangs system at divide by zero In-reply-to: Your message of "Wed, 12 Apr 1995 15:42:47 -0700." <28907.797726567@freefall.cdrom.com> Reply-to: scott@statsci.com MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Wed, 12 Apr 1995 21:12:36 -0700 From: Scott Blachowicz Sender: bugs-owner@FreeBSD.org Precedence: bulk "Jordan K. Hubbard" wrote: > "Why does Linux work with my computer when FreeBSD just dies?" Does that mean I should mention that my brand new Pentium-90 system boots from a Linux boot disc just fine, but just sits and spins trying to boot a rawriten boot.flp from either the 0210 or 0322 SNAPs? :-( The only suggestion I've gotten on trying to figure out the problem is to start disabling things in my BIOS setup to try to dumb things down. I haven't gotten that far (mostly because I don't know that much about the various BIOS setup configuration items). I've tried various combinations of rawrite, rawrite3.zip (v1.3) and 'dd' from Linux to produce the disc. I think I'm nearly convinced I've the boot disc right (hmmm...maybe I should try booting a different computer with one). So, it seems like it's gotta be something in my BIOS setup or particular flavor of some piece of hardware (or as was suggested - not having summoned the right daemons to haunt my system). Still out of it in Seattle... Scott Blachowicz Ph: 206/283-8802x240 StatSci, a div of MathSoft, Inc. 1700 Westlake Ave N #500 scott@statsci.com Seattle, WA USA 98109 Scott.Blachowicz@seaslug.org From owner-freebsd-bugs Thu Apr 13 00:13:50 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id AAA05439 for bugs-outgoing; Thu, 13 Apr 1995 00:13:50 -0700 Received: from skynet.ctr.columbia.edu (skynet.ctr.columbia.edu [128.59.64.70]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id AAA05433 for ; Thu, 13 Apr 1995 00:13:46 -0700 Received: (from wpaul@localhost) by skynet.ctr.columbia.edu (8.6.8/8.6.6) id CAA04290; Thu, 13 Apr 1995 02:10:43 -0400 From: "House of Debuggin'" Message-Id: <199504130610.CAA04290@skynet.ctr.columbia.edu> Subject: Re: 80387 hangs system at divide by zero To: scott@statsci.com Date: Thu, 13 Apr 1995 02:10:40 -0400 (EDT) Cc: bugs@FreeBSD.org In-Reply-To: from "Scott Blachowicz" at Apr 12, 95 09:12:36 pm X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 3959 Sender: bugs-owner@FreeBSD.org Precedence: bulk They say this Scott Blachowicz person was kidding when he wrote: > > "Jordan K. Hubbard" wrote: > > > "Why does Linux work with my computer when FreeBSD just dies?" > > Does that mean I should mention that my brand new Pentium-90 system boots > from a Linux boot disc just fine, but just sits and spins trying to boot a > rawriten boot.flp from either the 0210 or 0322 SNAPs? :-( Argh... No, it actually means you should tell us exactly what sort of Pentium-90 system you have and then provide more details. :) > The only suggestion I've gotten on trying to figure out the problem is to > start disabling things in my BIOS setup to try to dumb things down. Well, you have to be a bit more specific than just 'sits and spins.' At what point does it start to 'sit and spin?' Does it at least present you with a 'Boot:' prompt? If not, can you tell if there's any floppy activity at all? I'm inclined to think that the BIOS settings won't affect much until the kernel is loaded, but I could be wrong. If the kernel does manage to load, how far does it get? > I > haven't gotten that far (mostly because I don't know that much about the > various BIOS setup configuration items). I've tried various combinations of > rawrite, rawrite3.zip (v1.3) and 'dd' from Linux to produce the disc. Have you also tried FTPing new disk images? In binary mode? (Yes, I know: stupid questions, but we have to cover all the bases.) > I > think I'm nearly convinced I've the boot disc right (hmmm...maybe I should > try booting a different computer with one). That would help, yes. If you have another machine handy, please try it. > So, it seems like it's gotta be > something in my BIOS setup or particular flavor of some piece of hardware (or > as was suggested - not having summoned the right daemons to haunt my system). > > Still out of it in Seattle... > > Scott Blachowicz Ph: 206/283-8802x240 StatSci, a div of MathSoft, Inc. One thing that could be causing a problem (*shudder*) is the keyboard probe code in the boot block. The new boot blocks in 2.1 check to see if there's a keyboard available, and if not they use serial port COM1 as the console device. One way to see if this is a problem is to keep a sharp eye on your keyboard LEDs. The keyboard probe actually tries to reset the keyboard, and when it does this it generally causes the keyboard LEDs to flash. If the LEDs flash, but nothing else happens, then it's possible that the keyboard probe is hanging the machine (how, I don't know, since it's virtually the same probe used in syscons, which seems to work fine). If the LEDs flash, then things pause momentatily, and then you see/hear more floppy activity, then it's possible that the probe incorrectly decided that you had no keyboard attached and defaulted over into serial console mode. The kernel will still be loaded if this happens, which is why you might see more activity after a brief pause. If absolutely *nothing* happens (no keyboard LED flashes, no floppy activity, no nuthin', except for the floppy drive motor spinning) then the bootblock isn't even being loaded correctly. In this case, I would begin to suspect the boot disk. If you do in fact manage to get past the 'Boot:' prompt, then this is largely irrelevant (and I can sleep better since this whole 'automatic serial console' thing was my idea), but you still need to provide more details. -Bill -- ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~T~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -Bill Paul (212) 854-6020 | System Manager Work: wpaul@ctr.columbia.edu | Center for Telecommunications Research Home: wpaul@skynet.ctr.columbia.edu | Columbia University, New York City ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ The Møøse Illuminati: ignore it and be confused, or join it and be confusing! ~~~~~~~~ FreeBSD 2.1.0-Development #0: Tue Mar 14 11:11:25 EST 1995 ~~~~~~~~~ From owner-freebsd-bugs Thu Apr 13 02:27:01 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id CAA08187 for bugs-outgoing; Thu, 13 Apr 1995 02:27:01 -0700 Received: (from ache@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id CAA08173 ; Thu, 13 Apr 1995 02:27:00 -0700 Date: Thu, 13 Apr 1995 02:27:00 -0700 From: "Andrey A. Chernov" Message-Id: <199504130927.CAA08173@freefall.cdrom.com> To: pritc003@maroon.tc.umn.edu, ache, freebsd-bugs Subject: Changed information for PR bin/338 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: Cron allows users to obtain root access State-Changed-From-To: open-closed State-Changed-By: ache State-Changed-When: Thu Apr 13 02:26:21 PDT 1995 State-Changed-Why: Fixed in do_command.c 1.2 From owner-freebsd-bugs Thu Apr 13 06:29:10 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id GAA13605 for bugs-outgoing; Thu, 13 Apr 1995 06:29:10 -0700 Received: from cix.cict.fr (cix.cict.fr [192.70.79.11]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id GAA13596 for ; Thu, 13 Apr 1995 06:29:08 -0700 Received: from corail.cict.fr by cix.cict.fr; Thu, 13 Apr 95 15:30:58 +0200 From: cnama23@corail.cict.fr (Boudet Thierry(cnama23)) Received: by corail.cict.fr; Thu, 13 Apr 95 15:31:44 +0200 Date: Thu, 13 Apr 95 15:31:44 +0200 Message-Id: <9504131331.AA03962@corail.cict.fr> Apparently-To: bugs@FreeBSD.org Sender: bugs-owner@FreeBSD.org Precedence: bulk hello. i have a big problem with my FreeBsd 2.0 from the Walnut Creek cd-rom suddenly, after a random time (from 2 mins to 3 days) when i use the ALT-Fn key to swith console, the computer don't react... no key is operating... but all the other parts of the system continue to work. example : i am logged on ttyv0 and ttyv1 on ttyv1 i run a long-time-run command : find / -print | grep f2c > f2c.list (it's take long time because i have a cd mounted) ALR-F1 for going back to ttyv0 : DAMNDED, NO ANSWER after 1 or 2 minutes, the find is done, the shell prompt come on screen, but my keyboard is still frozzzzzen... i have this problem *only* when i switch consoles, never found with another keystroke. i don't run X. ;---( my configuration is : computer Intel Champion 486sx33 with 8 megs of ram Buslogic BT545C scsi controler hard-disk HP C3323-300 1 giga cd-rom Sony CDU-541 please, help me ! Thierry From owner-freebsd-bugs Thu Apr 13 08:21:08 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id IAA17752 for bugs-outgoing; Thu, 13 Apr 1995 08:21:08 -0700 Received: from goof.com (root@goof.com [198.82.204.15]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id IAA17744 for ; Thu, 13 Apr 1995 08:21:05 -0700 Received: (from mmead@localhost) by goof.com (8.6.11/8.6.9) id LAA12090 for bugs@freebsd.org; Thu, 13 Apr 1995 11:21:32 -0400 From: "matthew c. mead" Message-Id: <199504131521.LAA12090@goof.com> Subject: /var/mail To: bugs@FreeBSD.org Date: Thu, 13 Apr 1995 11:21:32 -0400 (EDT) X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 441 Sender: bugs-owner@FreeBSD.org Precedence: bulk /var/mail should be owner bin, group bin, permissions 775, so that packages like elm can write dot-lock files in /var/mail. -matt -- Matthew C. Mead -> Virginia Tech Center for Transportation Research - -> Multiple Platform System and Network Administration Work Related -> mmead@ctr.vt.edu | mmead@goof.com <- All Other ---- ------- WWW -> http://www.goof.com/~mmead --- ----- From owner-freebsd-bugs Thu Apr 13 08:22:29 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id IAA17925 for bugs-outgoing; Thu, 13 Apr 1995 08:22:29 -0700 Received: (from wollman@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id IAA17906 ; Thu, 13 Apr 1995 08:22:27 -0700 Date: Thu, 13 Apr 1995 08:22:27 -0700 From: "Garrett A. Wollman" Message-Id: <199504131522.IAA17906@freefall.cdrom.com> To: pst@Shockwave.COM, wollman, freebsd-bugs Subject: Changed information for PR bin/339 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: users may enter kerberos password at login prompt State-Changed-From-To: open-closed State-Changed-By: wollman State-Changed-When: Thu Apr 13 08:20:52 PDT 1995 State-Changed-Why: Paul's suggested patch appears to do the right thing, applied in login.c 1.8. From owner-freebsd-bugs Thu Apr 13 08:39:36 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id IAA18842 for bugs-outgoing; Thu, 13 Apr 1995 08:39:36 -0700 Received: from main.statsci.com (main.statsci.com [198.145.125.110]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id IAA18831 for ; Thu, 13 Apr 1995 08:39:32 -0700 Received: by main.statsci.com (Smail3.1.28.1 #9) id m0rzQzQ-000r3vC; Thu, 13 Apr 95 08:39 PDT Message-Id: To: "House of Debuggin'" cc: bugs@FreeBSD.org Subject: Re: 80387 hangs system at divide by zero In-reply-to: Your message of "Thu, 13 Apr 1995 02:10:40 -0400." <199504130610.CAA04290@skynet.ctr.columbia.edu> Reply-to: scott@statsci.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 13 Apr 1995 08:39:26 -0700 From: Scott Blachowicz Sender: bugs-owner@FreeBSD.org Precedence: bulk "House of Debuggin'" wrote: > Argh... No, it actually means you should tell us exactly what sort of > Pentium-90 system you have and then provide more details. :) I've done that to the freebsd-{questions,hackers,current} lists. Guess I'll try 'bugs', too. > > The only suggestion I've gotten on trying to figure out the problem is to > > start disabling things in my BIOS setup to try to dumb things down. > > Well, you have to be a bit more specific than just 'sits and spins.' I'll append details from previous postings to this message. > what point does it start to 'sit and spin?' At the point where I would expect it to be reading the kernel or something from the diskette. The drive light goes on & stays on and I can hear the diskette spinning if I listen closely enough. > Have you also tried FTPing new disk images? In binary mode? (Yes, I know: > stupid questions, but we have to cover all the bases.) Yup. Also, the first one I did was a boot.flp.gz - I would expect gunzip to complain about the .gz file if it weren't done in binary mode. I've used the perl mirror-2.3 script (on a Internet-connected SunOS system) and/or dired/efs under GNU emacs v19 to do the downloads. They should both be doing binary mode automatically. > > I think I'm nearly convinced I've the boot disc right (hmmm...maybe I > > should try booting a different computer with one). > > That would help, yes. If you have another machine handy, please try it. I just recreated a 0322-SNAP boot disc with 'rawrite' on a work PC (a 486/66 w/32Mb RAM). I got to the "Boot:" prompt which timed out, sucks in the kernel off of the floppy, then drops me into the "Welcome to FreeBSD!" screen. Now I'll have to go try that at home with the exact same disc just to be sure. I suppose it's possible that there was some problem writing the boot disc at home, but I find that hard to believe. > One way to see if this is a problem is to keep a sharp eye on your > keyboard LEDs. The keyboard probe actually tries to reset the keyboard, I'll check for that. Thanx!! Scott Blachowicz Ph: 206/283-8802x240 StatSci, a div of MathSoft, Inc. 1700 Westlake Ave N #500 scott@statsci.com Seattle, WA USA 98109 Scott.Blachowicz@seaslug.org ---------- Start of forwarded message I just pulled down the 950322-SNAP tree to use as my first attempt at installing FreeBSD on my home system. That system is: Pentium-90; 16 Mb RAM; 256K cache CAF Tech motherboard / SiS chipset / Award 4.50G BIOS PCI IDE - boot disk NCR 8150S PCI SCSI - Toshiba 3501 CDROM, Exabyte EXB-8200 8mm drive ATI Mach64 (2Mb)/17" display Sound Blaster AWE32 floppy/2S/1P/game card one 3.5" floppy drive; no 5.25" floppy drive I've tried using both RAWRITE & RAWRITE3 from the tools/dos-tools directory to put a gunzip'd boot.flp.gz onto a 3.5" diskette, then boot from that. My first attempt (with RAWRITE) ended up having my boot process completely ignore the inserted diskette. My 2nd & 3rd attempts using RAWRITE3 ended up just sitting there spinning the diskette drive for a while with no output to my screen before I came back and tried CTRL-ALT-DEL with no results and the RESET button. I then pulled the boot floppy out. The reboot sequence got to the point where I think it normally scans the system & spits out the Award BIOS banner and just hung. I power cycled and got the same hang. I then turned the system off for a while and turned it back on - it came up fine. After all that, I repeated the same business with a the 950210-SNAP boot.flp and got the same results. Possible explanations? 1) Do I really need a 1.2Mb 5.25" drive in order to use that boot.flp image? If so, what are the chances of getting a 1.4M 3.5" drive boot.flp image? It seems many systems are shipped without 5.25" drives these days. 2) Is there something about my machine configuration that is confusing the boot code? So, does anyone have any idea 1) what I'm doing wrong? 2) how I can do it right? 3) what the heck is happening here? 4) has anyone successfully booted the SNAPs using the supplied boot.flp on a 1.4M 3.5" drive? help? [whimper] ---------- End of forwarded message From owner-freebsd-bugs Thu Apr 13 09:07:16 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id JAA19926 for bugs-outgoing; Thu, 13 Apr 1995 09:07:16 -0700 Received: from halloran-eldar.lcs.mit.edu (halloran-eldar.lcs.mit.edu [18.26.0.159]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id JAA19917 for ; Thu, 13 Apr 1995 09:07:14 -0700 Received: by halloran-eldar.lcs.mit.edu; id AA08853; Thu, 13 Apr 1995 12:07:01 -0400 Date: Thu, 13 Apr 1995 12:07:01 -0400 From: Garrett Wollman Message-Id: <9504131607.AA08853@halloran-eldar.lcs.mit.edu> To: "matthew c. mead" Cc: bugs@FreeBSD.org Subject: /var/mail In-Reply-To: <199504131521.LAA12090@goof.com> References: <199504131521.LAA12090@goof.com> Sender: bugs-owner@FreeBSD.org Precedence: bulk < said: > /var/mail should be owner bin, group bin, permissions 775, so that packages > like elm can write dot-lock files in /var/mail. NO!!!!!!! If you make /var/mail group writable, it should be writable by group `mail', NOT by group `bin'. -GAWollman -- Garrett A. Wollman | Shashish is simple, it's discreet, it's brief. ... wollman@lcs.mit.edu | Shashish is the bonding of hearts in spite of distance. Opinions not those of| It is a bond more powerful than absence. We like people MIT, LCS, ANA, or NSA| who like Shashish. - Claude McKenzie + Florent Vollant From owner-freebsd-bugs Thu Apr 13 09:44:50 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id JAA20702 for bugs-outgoing; Thu, 13 Apr 1995 09:44:50 -0700 Received: from cs.utah.edu (cs.utah.edu [128.110.4.21]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id JAA20696 for ; Thu, 13 Apr 1995 09:44:44 -0700 From: moore@ced.utah.edu Received: from canyonlands.ced.utah.edu by cs.utah.edu (8.6.12/utah-2.21-cs) id KAA23049; Thu, 13 Apr 1995 10:44:41 -0600 Received: from ruin.ced.utah.edu by canyonlands.ced.utah.edu (4.1/SMI-4.1) id AA11279; Thu, 13 Apr 95 10:44:10 MDT Received: by ruin.ced.utah.edu (940816.SGI.8.6.9//ident-1.0) id KAA14834; Thu, 13 Apr 1995 10:44:10 -0600 Date: Thu, 13 Apr 1995 10:44:10 -0600 Message-Id: <199504131644.KAA14834@ruin.ced.utah.edu> To: bugs@FreeBSD.org Subject: integer divide fault with newer boot floppies on a Micron P90 Sender: bugs-owner@FreeBSD.org Precedence: bulk I've been having problems installing FreeBSD on a Micron P90 PCI system with a Micronics motherboard. On advice from Terry Lambert I went in search of newer boot floppies than 2.0-Release, and downloaded the images from the 0322-Snapshot in floppies/newer. These get a little further, but always die after I select Install from the menu: Fatal Trap 18: integer divide fault while in kernel mode instruction pointer = 0x8:0xf015b1ec code segment = base 0x0, limit 0xfffff,type 0x1b = DPL0, prs 1 def 321 gran 1 processor eflags = interrupts enabled, resume, IOPL=0 current process = 1 (sysinstall) interrupt mask = panic: integer divide fault So, what's up? I'd like to help solve this, obviously :) -- Tim Moore Sarcos Research Corp. moore@ced.utah.edu "Wind in my hair - Shifting and drifting - Mechanical music - Adrenaline surge" - Rush From owner-freebsd-bugs Thu Apr 13 10:10:44 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id KAA21174 for bugs-outgoing; Thu, 13 Apr 1995 10:10:44 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id KAA21132 ; Thu, 13 Apr 1995 10:10:38 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.8/8.6.6) id KAA10573; Thu, 13 Apr 1995 10:10:20 -0700 From: "Rodney W. Grimes" Message-Id: <199504131710.KAA10573@gndrsh.aac.dev.com> Subject: Re: /var/mail To: mmead@goof.com (matthew c. mead) Date: Thu, 13 Apr 1995 10:10:20 -0700 (PDT) Cc: bugs@FreeBSD.org, FAQ@FreeBSD.org In-Reply-To: <199504131521.LAA12090@goof.com> from "matthew c. mead" at Apr 13, 95 11:21:32 am X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 909 Sender: bugs-owner@FreeBSD.org Precedence: bulk > > /var/mail should be owner bin, group bin, permissions 775, so that packages > like elm can write dot-lock files in /var/mail. This has been gone over several times. Use of dot-lock files for elm and such mailers or for having site wide NFS mounted /var/mail is a site decision and requires the relaxation of the permission on /var/mail. Since having /var/mail bin:bin 775 presents a possible security risk FreeBSD as shipped keeps uses bin:bin 755. You also must install elm sgid bin if you want the mode 775 stuff to work, yet another potential security whole (and yes there have been versions of elm that allowed you to ! out of them and become bin and then go happely trash the whole system :-(. Can some one please add this to a FAQ!!! -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Custom computers for FreeBSD From owner-freebsd-bugs Thu Apr 13 13:25:05 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA25789 for bugs-outgoing; Thu, 13 Apr 1995 13:25:05 -0700 Received: from irz301.inf.tu-dresden.de (irz301.inf.tu-dresden.de [141.76.1.11]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id NAA25768 for ; Thu, 13 Apr 1995 13:24:15 -0700 Received: from sax.sax.de by irz301.inf.tu-dresden.de with SMTP (5.67b+/DEC-Ultrix/4.3) id AA22154; Thu, 13 Apr 1995 22:22:41 +0200 Received: by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id WAA19883; Thu, 13 Apr 1995 22:22:41 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.6.11/8.6.9) id WAA14843; Thu, 13 Apr 1995 22:20:03 +0200 From: J Wunsch Message-Id: <199504132020.WAA14843@uriah.heep.sax.de> Subject: Re: 80387 hangs system at divide by zero To: scott@statsci.com Date: Thu, 13 Apr 1995 22:20:03 +0200 (MET DST) Cc: wpaul@skynet.ctr.columbia.edu, bugs@FreeBSD.org In-Reply-To: from "Scott Blachowicz" at Apr 13, 95 08:39:26 am Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) 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 Content-Length: 602 Sender: bugs-owner@FreeBSD.org Precedence: bulk As Scott Blachowicz wrote: > > 1) Do I really need a 1.2Mb 5.25" drive in order to use that boot.flp > image? If so, what are the chances of getting a 1.4M 3.5" drive > boot.flp image? It seems many systems are shipped without 5.25" drives > these days. The image is kept small enough to fit on a 1.2 MB disk, but it can be placed as well on a 1.44 MB disk -- this way, we satisfy both camps. Sorry, i can't help you with your weird problem. :( -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ Never trust an operating system you don't have sources for. ;-) From owner-freebsd-bugs Thu Apr 13 13:33:19 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id NAA26124 for bugs-outgoing; Thu, 13 Apr 1995 13:33:19 -0700 Received: from main.statsci.com (main.statsci.com [198.145.125.110]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id NAA26114 for ; Thu, 13 Apr 1995 13:33:16 -0700 Received: by main.statsci.com (Smail3.1.28.1 #9) id m0rzVZd-000r3wC; Thu, 13 Apr 95 13:33 PDT Message-Id: To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) cc: wpaul@skynet.ctr.columbia.edu, bugs@FreeBSD.org Subject: Re: 80387 hangs system at divide by zero In-reply-to: Your message of "Thu, 13 Apr 1995 22:20:03 +0200." <199504132020.WAA14843@uriah.heep.sax.de> Reply-to: scott@statsci.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 13 Apr 1995 13:34:17 -0700 From: Scott Blachowicz Sender: bugs-owner@FreeBSD.org Precedence: bulk J Wunsch wrote: > The image is kept small enough to fit on a 1.2 MB disk, but it can be > placed as well on a 1.44 MB disk -- this way, we satisfy both camps. That's what I figured - I know I've heard that said of both FreeBSD & Linux boot disks. Just wondering if it is possible for the boot disc to be created in such a way as to make it not _always_ work on 1.44Mb drives. I don't know the structure of the bytes on the disc, so I even tried this (from Linux): dd if=/dev/zero of=/dev/fd0 before rawrite-ing the disc to make sure there was no stray garbage laying around (a naive-i-don't-know-how-it-works kind of thing to do). No difference in bootability of the disc. Scott Blachowicz Ph: 206/283-8802x240 StatSci, a div of MathSoft, Inc. 1700 Westlake Ave N #500 scott@statsci.com Seattle, WA USA 98109 Scott.Blachowicz@seaslug.org From owner-freebsd-bugs Thu Apr 13 14:09:04 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id OAA27632 for bugs-outgoing; Thu, 13 Apr 1995 14:09:04 -0700 Received: from pluto.ops.NeoSoft.com (root@pluto.ops.NeoSoft.COM [198.64.212.23]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id OAA27619 for ; Thu, 13 Apr 1995 14:08:59 -0700 Received: from metal.ops.neosoft.com (glenn-slip43.nmt.edu [129.138.5.143]) by pluto.ops.NeoSoft.com (8.6.10/8.6.10) with ESMTP id QAA03299; Thu, 13 Apr 1995 16:07:54 -0500 Received: (from smace@localhost) by metal.ops.neosoft.com (8.6.11/8.6.10) id PAA00503; Thu, 13 Apr 1995 15:07:51 -0600 From: Scott Mace Message-Id: <199504132107.PAA00503@metal.ops.neosoft.com> Subject: Re: /var/mail To: wollman@halloran-eldar.lcs.mit.edu (Garrett Wollman) Date: Thu, 13 Apr 1995 15:07:51 -0600 (MDT) Cc: mmead@goof.com, bugs@FreeBSD.org In-Reply-To: <9504131607.AA08853@halloran-eldar.lcs.mit.edu> from "Garrett Wollman" at Apr 13, 95 12:07:01 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 393 Sender: bugs-owner@FreeBSD.org Precedence: bulk > > < said: > > > /var/mail should be owner bin, group bin, permissions 775, so that packages > > like elm can write dot-lock files in /var/mail. > > NO!!!!!!! > > If you make /var/mail group writable, it should be writable by group > `mail', NOT by group `bin'. or not use .dot-locking at all...> Scott From owner-freebsd-bugs Thu Apr 13 15:19:40 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id PAA29684 for bugs-outgoing; Thu, 13 Apr 1995 15:19:40 -0700 Received: (from joerg@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id AAA29669 ; Fri, 14 Apr 1995 00:19:37 +0200 Date: Fri, 14 Apr 1995 00:19:37 +0200 From: Joerg Wunsch Message-Id: <199504132219.AAA29669@freefall.cdrom.com> To: edward@edcom.com, joerg, freebsd-bugs Subject: Changed information for PR bin/335 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: ls -k with BLOCKSIZE=K uses 2k blocks State-Changed-From-To: open-closed State-Changed-By: joerg State-Changed-When: Fri Apr 14 00:15:13 MET DST 1995 State-Changed-Why: The problem had already been fixed with revision 1.6 of ls.c by phk; this was just one day after the 950322 NSAP however. From owner-freebsd-bugs Thu Apr 13 15:43:17 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id PAA00276 for bugs-outgoing; Thu, 13 Apr 1995 15:43:17 -0700 Received: from dub-img-4.compuserve.com (dub-img-4.compuserve.com [198.4.9.4]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id PAA00268 for ; Thu, 13 Apr 1995 15:43:15 -0700 Received: by dub-img-4.compuserve.com (8.6.10/5.941228sam) id SAA06781; Thu, 13 Apr 1995 18:42:36 -0400 Date: 13 Apr 95 18:41:07 EDT From: "Regner A. Cunanan" <75452.2021@compuserve.com> To: FreeBSD Subject: Problem encountered at CompuServe/Internet mail gate Message-ID: <950413224107_75452.2021_FHQ71-1@CompuServe.COM> Sender: bugs-owner@FreeBSD.org Precedence: bulk Message number: 950413051953 75452.2021 FHQ33-1 Sent at: 01:19 EDT 13-Apr-95 Subject: Installing FreeBSD 2 on 2.9 Gigabytes ST43400N I am trying to install FreeBSD on this 2.9 Gigabytes SCSI drive. MSDOS occupies the first 650 Megabytes, OS2 the next 200 Megs and the rest assigned to BSD. The root is assigned next as 50 Megs, followed by SWAP (128Megs) and the rest for /usr. I keep getting FATAL error code 5888. When I change the geometry from the mapped geometry to DISK specs, the load continues but when I am supposed to reset and boot again to complete the installation, the boot manager just loops whether I chose BSD, MSDOS or OS2. Plus when I boot from floppy and use MSDOS fdisk, it can not read the fdisk table. I think the problem is that the mapping to handle > 1 Gigabyte disk in SCSI is not being handle correctly. Would appreciate all help I can get. PS: I tried using a smaller SCSI drive and it installed correctly. But I sure would like to just use the large disk I just bought for this purpose. How do I join your group to work on this problem if no one else is interested in it? Thanks..Regner From owner-freebsd-bugs Thu Apr 13 15:59:43 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id PAA00630 for bugs-outgoing; Thu, 13 Apr 1995 15:59:43 -0700 Received: from cs.weber.edu (cs.weber.edu [137.190.16.16]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id PAA00621 for ; Thu, 13 Apr 1995 15:59:38 -0700 Received: by cs.weber.edu (4.1/SMI-4.1.1) id AA25381; Thu, 13 Apr 95 16:53:11 MDT From: terry@cs.weber.edu (Terry Lambert) Message-Id: <9504132253.AA25381@cs.weber.edu> Subject: Re: Problem encountered at CompuServe/Internet mail gate To: 75452.2021@compuserve.com (Regner A. Cunanan) Date: Thu, 13 Apr 95 16:53:10 MDT Cc: bugs@FreeBSD.org In-Reply-To: <950413224107_75452.2021_FHQ71-1@CompuServe.COM> from "Regner A. Cunanan" at Apr 13, 95 06:41:07 pm X-Mailer: ELM [version 2.4dev PL52] Sender: bugs-owner@FreeBSD.org Precedence: bulk > Plus when I boot from floppy and > use MSDOS fdisk, it can not read the fdisk table. Please download pfdisk.ese from freebsd.cdrom.com. Put it on your floppy. Boot from floppy. Tell us what partition information and ID's it sees. Terry Lambert terry@cs.weber.edu --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-bugs Thu Apr 13 16:34:53 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id QAA01617 for bugs-outgoing; Thu, 13 Apr 1995 16:34:53 -0700 Received: from localhost (localhost [127.0.0.1]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id QAA01610 ; Thu, 13 Apr 1995 16:34:51 -0700 X-Authentication-Warning: freefall.cdrom.com: Host localhost didn't use HELO protocol To: "Regner A. Cunanan" <75452.2021@compuserve.com> cc: FreeBSD Subject: Re: Problem encountered at CompuServe/Internet mail gate In-reply-to: Your message of "13 Apr 95 18:41:07 EDT." <950413224107_75452.2021_FHQ71-1@CompuServe.COM> Date: Thu, 13 Apr 1995 16:34:51 -0700 Message-ID: <1609.797816091@freefall.cdrom.com> From: "Jordan K. Hubbard" Sender: bugs-owner@FreeBSD.org Precedence: bulk This won't work in 2.0. You need to get the latest snapshot, I'm afraid! ftp://ftp.freebsd.org/pub/FreeBSD/2.0-950412-SNAP Jordan > > Message number: 950413051953 75452.2021 FHQ33-1 > Sent at: 01:19 EDT 13-Apr-95 > Subject: Installing FreeBSD 2 on 2.9 Gigabytes ST43400N > > I am trying to install FreeBSD on this 2.9 Gigabytes SCSI drive. MSDOS occup ies > the first 650 Megabytes, OS2 the next 200 Megs and the rest assigned to BSD. > The root is assigned next as 50 Megs, followed by SWAP (128Megs) and the rest > for /usr. I keep getting FATAL error code 5888. When I change the geometry > from the mapped geometry to DISK specs, the load continues but when I am > supposed to reset and boot again to complete the installation, the boot manag er > just loops whether I chose BSD, MSDOS or OS2. Plus when I boot from floppy a nd > use MSDOS fdisk, it can not read the fdisk table. > > I think the problem is that the mapping to handle > 1 Gigabyte disk in SCSI i s > not being handle correctly. Would appreciate all help I can get. > > PS: I tried using a smaller SCSI drive and it installed correctly. But I sur e > would like to just use the large disk I just bought for this purpose. > > How do I join your group to work on this problem if no one else is interested in > it? > > Thanks..Regner > > From owner-freebsd-bugs Thu Apr 13 17:58:37 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id RAA03508 for bugs-outgoing; Thu, 13 Apr 1995 17:58:37 -0700 Received: from fgwmail.fujitsu.co.jp (fgwmail.fujitsu.co.jp [164.71.1.133]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id RAA03501 for ; Thu, 13 Apr 1995 17:58:33 -0700 Received: from fdmmail.fujitsu.co.jp by fgwmail.fujitsu.co.jp (8.6.12+2.4W/3.3W5-MX941209-Fujitsu Mail Gateway) id JAA00207; Fri, 14 Apr 1995 09:58:28 +0900 Received: from fdm.fujitsu.co.jp by fdmmail.fujitsu.co.jp (8.6.12+2.4W/3.3W5-MX950127-Fujitsu Domain Mail Master) id JAA25261; Fri, 14 Apr 1995 09:57:57 +0900 Received: from sysrap by fdm.fujitsu.co.jp (5.65/6.4J.6) id AA18211; Fri, 14 Apr 95 09:57:57 +0900 Received: from seki.sysrap.cs.fujitsu.co.jp by spad.sysrap.cs.fujitsu.co.jp with smtp (Smail3.1.28.1 #1) id m0rzZpN-0001yJC; Fri, 14 Apr 95 10:05 JST Date: Fri, 14 Apr 95 09:53:37 JST From: Masahiro SEKIGUCHI Message-Id: <9504140053.AA10643@seki.sysrap.cs.fujitsu.co.jp> To: FreeBSD-BUGs@FreeBSD.org, FreeBSD-current@FreeBSD.org Subject: bug on /bin/ps permission? Sender: bugs-owner@FreeBSD.org Precedence: bulk I'm not sure this is a bug. In 950322 SNAP bin, /bin/ps has the following permission/owner/group: -r-sr-xr-x 1 root bin 143360 Mar 24 09:33 /bin/ps I guess the SUID bit and root owning is to access /dev/kmem. Then, should it be: -r-xr-sr-x 1 bin kmem 143360 Mar 24 09:33 /bin/ps to give minimum privilege? From owner-freebsd-bugs Fri Apr 14 01:00:04 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id BAA10074 for bugs-outgoing; Fri, 14 Apr 1995 01:00:04 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id BAA10062 ; Fri, 14 Apr 1995 01:00:03 -0700 Date: Fri, 14 Apr 1995 01:00:03 -0700 Message-Id: <199504140800.BAA10062@freefall.cdrom.com> From: Mark Murray Reply-To: Mark Murray To: freebsd-bugs Subject: ports/340: USE_GMAKE does not work with USE_IMAKE in ports In-Reply-To: Your message of Fri, 14 Apr 1995 09:58:19 +0200 <199504140758.JAA18422@grunt.grondar.za> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 340 >Category: ports >Synopsis: USE_GMAKE does not work with USE_IMAKE in ports >Confidential: no >Severity: serious >Priority: medium >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Fri Apr 14 01:00:01 1995 >Originator: Mark Murray >Organization: GTA >Release: FreeBSD 2.1.0-Development i386 >Environment: Current - ctm#510 and Ports - ctm#333 >Description: If a port has both USE_IMAKE and USE_GMAKE, the ${GMAKE} is not used at the 'make makefiles' stage; ${MAKE} is. >How-To-Repeat: cd ports/print/xdvi make >Fix: Dunno >Audit-Trail: >Unformatted: Mark Murray From owner-freebsd-bugs Fri Apr 14 02:08:55 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id CAA11905 for bugs-outgoing; Fri, 14 Apr 1995 02:08:55 -0700 Received: from irz301.inf.tu-dresden.de (irz301.inf.tu-dresden.de [141.76.1.11]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id CAA11889 for ; Fri, 14 Apr 1995 02:08:45 -0700 Received: from sax.sax.de by irz301.inf.tu-dresden.de with SMTP (5.67b+/DEC-Ultrix/4.3) id AA02406; Fri, 14 Apr 1995 11:08:31 +0200 Received: by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id LAA23732; Fri, 14 Apr 1995 11:08:29 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.6.11/8.6.9) id KAA17434; Fri, 14 Apr 1995 10:49:12 +0200 From: J Wunsch Message-Id: <199504140849.KAA17434@uriah.heep.sax.de> Subject: Re: bug on /bin/ps permission? To: seki@sysrap.cs.fujitsu.co.jp (Masahiro SEKIGUCHI) Date: Fri, 14 Apr 1995 10:49:11 +0200 (MET DST) Cc: FreeBSD-BUGs@FreeBSD.org In-Reply-To: <9504140053.AA10643@seki.sysrap.cs.fujitsu.co.jp> from "Masahiro SEKIGUCHI" at Apr 14, 95 09:53:37 am Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) 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 Content-Length: 1186 Sender: bugs-owner@FreeBSD.org Precedence: bulk As Masahiro SEKIGUCHI wrote: > > I'm not sure this is a bug. > > In 950322 SNAP bin, /bin/ps has the following permission/owner/group: > > -r-sr-xr-x 1 root bin 143360 Mar 24 09:33 /bin/ps This is the change that made it: revision 1.2 date: 1994/08/11 20:06:55; author: wollman; state: Exp; lines: +2 -2 Make `ps' set-uid root so that it can read the same set of arguments that old `ps'es did. I'm not too thrilled about this, but I'm not enough of an FS person to hack procfs so that /proc/xxx/mem is readable by members of group `kmem'. If this is done, then `ps' can go back to being set-gid kmem. Unfortunately, it doesn't seem to make a good job, since the remainder of ps is not aware of its new privileges: /* * Discard setgid privileges if not the running kernel so that bad * guys can't print interesting stuff from kernel memory. */ if (nlistf != NULL || memf != NULL || swapf != NULL) setgid(getgid()); This has been turned into a no-op by the suidness. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ Never trust an operating system you don't have sources for. ;-) From owner-freebsd-bugs Fri Apr 14 02:10:07 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id CAA11967 for bugs-outgoing; Fri, 14 Apr 1995 02:10:07 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id CAA11960 ; Fri, 14 Apr 1995 02:10:06 -0700 Date: Fri, 14 Apr 1995 02:10:06 -0700 Message-Id: <199504140910.CAA11960@freefall.cdrom.com> From: J Wunsch Reply-To: J Wunsch To: freebsd-bugs Subject: docs/341: vnconfig(8) disagrees with its man page In-Reply-To: Your message of Fri, 14 Apr 1995 10:16:22 +0200 <199504140816.KAA17118@uriah.heep.sax.de> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 341 >Category: docs >Synopsis: vnconfig(8) disagrees with its man page >Confidential: no >Severity: non-critical >Priority: low >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: doc-bug >Submitter-Id: current-users >Arrival-Date: Fri Apr 14 02:10:03 1995 >Originator: J Wunsch >Organization: >Release: FreeBSD 2.1.0-Development i386 >Environment: >Description: -current, as of CTM #cvs-cur 553 >How-To-Repeat: j@uriah 96% /usr/sbin/vnconfig usage: /usr/sbin/vnconfig [-acdefguv] [-s option] [-r option] \ [special-device file] options: labels, follow, debug, io, all, none j@uriah 97% man vnconfig VNCONFIG(8) UNIX System Manager's Manual VNCONFIG(8) NAME vnconfig - configure and enable vnode disks SYNOPSIS vnconfig [-cdeuv] special_file regular_file [feature] vnconfig -a [-cdeuv] [-f config_file] The option set do not match. Option ``-s labels'' seems to be important e.g. to create a diskette image that can be disklabelled, so it should be documented. (What's ``-r''?) >Fix: >Audit-Trail: >Unformatted: From owner-freebsd-bugs Fri Apr 14 03:36:07 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id DAA14049 for bugs-outgoing; Fri, 14 Apr 1995 03:36:07 -0700 Received: from irz301.inf.tu-dresden.de (irz301.inf.tu-dresden.de [141.76.1.11]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id DAA14037 for ; Fri, 14 Apr 1995 03:35:52 -0700 Received: from sax.sax.de by irz301.inf.tu-dresden.de with SMTP (5.67b+/DEC-Ultrix/4.3) id AA03349; Fri, 14 Apr 1995 12:34:28 +0200 Received: by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id MAA24587 for freebsd-bugs@freefall.cdrom.com; Fri, 14 Apr 1995 12:34:28 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.6.11/8.6.9) id MAA18792 for freebsd-bugs@freefall.cdrom.com; Fri, 14 Apr 1995 12:32:32 +0200 From: J Wunsch Message-Id: <199504141032.MAA18792@uriah.heep.sax.de> Subject: Re: ports/340: USE_GMAKE does not work with USE_IMAKE in ports To: freebsd-bugs@freefall.cdrom.com Date: Fri, 14 Apr 1995 12:32:31 +0200 (MET DST) Reply-To: freebsd-bugs@freefall.cdrom.com In-Reply-To: <199504140800.BAA10062@freefall.cdrom.com> from "Mark Murray" at Apr 14, 95 01:00:03 am Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) 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 Content-Length: 387 Sender: bugs-owner@FreeBSD.org Precedence: bulk As Mark Murray wrote: > > If a port has both USE_IMAKE and USE_GMAKE, the ${GMAKE} is not used at > the 'make makefiles' stage; ${MAKE} is. imake-ports should never require gmake. imake-created Makefiles are IMHO safe to be run with bmake. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ Never trust an operating system you don't have sources for. ;-) From owner-freebsd-bugs Fri Apr 14 06:46:16 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id GAA16894 for bugs-outgoing; Fri, 14 Apr 1995 06:46:16 -0700 Received: from grunt.grondar.za (grunt.grondar.za [196.7.18.129]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id GAA16883 for ; Fri, 14 Apr 1995 06:45:56 -0700 Received: from localhost (localhost [127.0.0.1]) by grunt.grondar.za (8.6.11/8.6.9) with SMTP id PAA10885; Fri, 14 Apr 1995 15:45:36 +0200 Message-Id: <199504141345.PAA10885@grunt.grondar.za> X-Authentication-Warning: grunt.grondar.za: Host localhost didn't use HELO protocol To: freebsd-bugs@freefall.cdrom.com, joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) Subject: Re: ports/340: USE_GMAKE does not work with USE_IMAKE in ports Date: Fri, 14 Apr 1995 15:45:35 +0200 From: Mark Murray Sender: bugs-owner@FreeBSD.org Precedence: bulk > As Mark Murray wrote: > > > > If a port has both USE_IMAKE and USE_GMAKE, the ${GMAKE} is not used at > > the 'make makefiles' stage; ${MAKE} is. > > imake-ports should never require gmake. imake-created Makefiles are > IMHO safe to be run with bmake. Please try to compile ports/xdvi with and without gmake. You will have to do the gmake bit by hand. M -- Mark Murray 46 Harvey Rd, Claremont, Cape Town 7700, South Africa +27 21 61-3768 GMT+0200 From owner-freebsd-bugs Fri Apr 14 07:29:16 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id HAA17662 for bugs-outgoing; Fri, 14 Apr 1995 07:29:16 -0700 Received: from cabri.obs-besancon.fr (cabri.obs-besancon.fr [193.52.184.3]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id HAA17654 for ; Fri, 14 Apr 1995 07:29:07 -0700 Received: by cabri.obs-besancon.fr (5.57/Ultrix3.0-C) id AA15625; Fri, 14 Apr 95 16:27:28 +0100 Date: Fri, 14 Apr 95 16:27:28 +0100 From: jmz@cabri.obs-besancon.fr (Jean-Marc Zucconi) Message-Id: <9504141527.AA15625@cabri.obs-besancon.fr> To: mark@grondar.za Cc: freebsd-bugs@freefall.cdrom.com, joerg_wunsch@uriah.heep.sax.de In-Reply-To: <199504141345.PAA10885@grunt.grondar.za> (message from Mark Murray on Fri, 14 Apr 1995 15:45:35 +0200) Subject: Re: ports/340: USE_GMAKE does not work with USE_IMAKE in ports X-Mailer: Emacs Sender: bugs-owner@FreeBSD.org Precedence: bulk >>>>> "Mark" == Mark Murray writes: >> As Mark Murray wrote: >> > >> > If a port has both USE_IMAKE and USE_GMAKE, the ${GMAKE} is not used at >> > the 'make makefiles' stage; ${MAKE} is. >> >> imake-ports should never require gmake. imake-created Makefiles are >> IMHO safe to be run with bmake. > Please try to compile ports/xdvi with and without gmake. You will have to > do the gmake bit by hand. Are you using a recent bsd.port.mk? I solved this problem a long time ago. From the cvs log: revision 1.67 date: 1994/11/16 23:14:22; author: jmz; state: Exp; lines: +5 -1 If USE_GMAKE is defined, use ${GMAKE} in ${XMKMF} && ${MAKE} Makefiles Jean-Marc. > M > -- > Mark Murray > 46 Harvey Rd, Claremont, Cape Town 7700, South Africa > +27 21 61-3768 GMT+0200 ~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~-~ Jean-Marc Zucconi | jmz@cabri.obs-besancon.fr Observatoire de Besancon | F 25010 Besancon cedex | PGP Key: finger jmz@cabri.obs-besancon.fr ========================================================================= From owner-freebsd-bugs Fri Apr 14 08:10:19 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id IAA19041 for bugs-outgoing; Fri, 14 Apr 1995 08:10:19 -0700 Received: from grunt.grondar.za (grunt.grondar.za [196.7.18.129]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id IAA19018 for ; Fri, 14 Apr 1995 08:10:07 -0700 Received: from localhost (localhost [127.0.0.1]) by grunt.grondar.za (8.6.11/8.6.9) with SMTP id RAA12179; Fri, 14 Apr 1995 17:09:23 +0200 Message-Id: <199504141509.RAA12179@grunt.grondar.za> X-Authentication-Warning: grunt.grondar.za: Host localhost didn't use HELO protocol To: jmz@cabri.obs-besancon.fr (Jean-Marc Zucconi) cc: freebsd-bugs@freefall.cdrom.com, joerg_wunsch@uriah.heep.sax.de Subject: Re: ports/340: USE_GMAKE does not work with USE_IMAKE in ports Date: Fri, 14 Apr 1995 17:09:23 +0200 From: Mark Murray Sender: bugs-owner@FreeBSD.org Precedence: bulk > > Please try to compile ports/xdvi with and without gmake. You will have to > > do the gmake bit by hand. > > Are you using a recent bsd.port.mk? I solved this problem a long time > ago. From the cvs log: > > revision 1.67 > date: 1994/11/16 23:14:22; author: jmz; state: Exp; lines: +5 -1 > If USE_GMAKE is defined, use ${GMAKE} in ${XMKMF} && ${MAKE} Makefiles I get updates via ctm (I am at ctm 510 right now), and I had just done a 'make world' when this happened. I looked in /usr/share/mk/bsd.port.mk, and saw the line that _should_ have done what you say above. It does not seem to work. Either the make logic is not working, or GMAKE at that stage contains 'make'. (This is happening at the configure stage when the make system tries to 'make makefiles'. M -- Mark Murray 46 Harvey Rd, Claremont, Cape Town 7700, South Africa +27 21 61-3768 GMT+0200 From owner-freebsd-bugs Fri Apr 14 08:30:02 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id IAA20690 for bugs-outgoing; Fri, 14 Apr 1995 08:30:02 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id IAA20683 ; Fri, 14 Apr 1995 08:30:01 -0700 Date: Fri, 14 Apr 1995 08:30:01 -0700 Message-Id: <199504141530.IAA20683@freefall.cdrom.com> From: pritc003@maroon.tc.umn.edu Reply-To: pritc003@maroon.tc.umn.edu To: freebsd-bugs Subject: bin/342: lpd can allow users access to all of root's groups In-Reply-To: Your message of Fri, 14 Apr 1995 10:27:58 -0500 <199504141527.KAA02469@mpp.com> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 342 >Category: bin >Synopsis: lpd can allow users access to all of root's groups >Confidential: no >Severity: serious >Priority: high >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Fri Apr 14 08:30:00 1995 >Originator: Mike Pritchard >Organization: >Release: FreeBSD 2.1.0-Development i386 >Environment: >Description: If a /etc/printcap entry is setup to use a filtering program that allows the user some control over it (e.g. apsfilter), then the user can run commands as user daemon with all of root's groups. >How-To-Repeat: The latest version of apsfilter allows users to create a file called ~/.apsfilterrc which is then executed by apsfilter anytime it runs. During execution, the process is running as daemon with all of root's groups. Verify this by adding a "id > /tmp/id.aps" as the first line of the .apsfilterrc file. >Fix: The following patch fixes lpd to call initgroups() and setgid() to restrict its group list anytime it forks off a command. This still allows users to run as daemon.daemon, which really isn't desirable either, but changing it to run the filtering programs as the user would probably break any filters that depend on being daemon to write accounting files. At least this is a start. Apply this patch to /usr/src/usr.sbin/lpr/lpd/printjob.c *** orig/printjob.c Fri Apr 14 08:27:42 1995 --- printjob.c Fri Apr 14 09:22:21 1995 *************** *** 1065,1070 **** --- 1065,1071 ---- int action; { register int i, pid; + struct passwd *pwd; for (i = 0; i < 20; i++) { if ((pid = fork()) < 0) { *************** *** 1074,1081 **** /* * Child should run as daemon instead of root */ ! if (pid == 0) setuid(DU); return(pid); } syslog(LOG_ERR, "can't fork"); --- 1075,1089 ---- /* * Child should run as daemon instead of root */ ! if (pid == 0) { ! if ((pwd = getpwuid(DU)) == NULL) { ! syslog(LOG_ERR, "Can't lookup default uid in password file"); ! break; ! } ! initgroups(pwd->pw_name, pwd->pw_gid); ! setgid(pwd->pw_gid); setuid(DU); + } return(pid); } syslog(LOG_ERR, "can't fork"); >Audit-Trail: >Unformatted: From owner-freebsd-bugs Fri Apr 14 08:59:55 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id IAA22512 for bugs-outgoing; Fri, 14 Apr 1995 08:59:55 -0700 Received: from irz301.inf.tu-dresden.de (irz301.inf.tu-dresden.de [141.76.1.11]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id IAA22498 for ; Fri, 14 Apr 1995 08:59:45 -0700 Received: from sax.sax.de by irz301.inf.tu-dresden.de with SMTP (5.67b+/DEC-Ultrix/4.3) id AA07704; Fri, 14 Apr 1995 17:58:11 +0200 Received: by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id RAA26135; Fri, 14 Apr 1995 17:58:11 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.6.11/8.6.9) id RAA20464; Fri, 14 Apr 1995 17:46:28 +0200 From: J Wunsch Message-Id: <199504141546.RAA20464@uriah.heep.sax.de> Subject: Re: ports/340: USE_GMAKE does not work with USE_IMAKE in ports To: mark@grondar.za (Mark Murray) Date: Fri, 14 Apr 1995 17:46:28 +0200 (MET DST) Cc: freebsd-bugs@freefall.cdrom.com, joerg_wunsch@uriah.heep.sax.de In-Reply-To: <199504141345.PAA10885@grunt.grondar.za> from "Mark Murray" at Apr 14, 95 03:45:35 pm Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) 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 Content-Length: 406 Sender: bugs-owner@FreeBSD.org Precedence: bulk As Mark Murray wrote: > > > imake-ports should never require gmake. imake-created Makefiles are > > IMHO safe to be run with bmake. > > Please try to compile ports/xdvi with and without gmake. You will have to > do the gmake bit by hand. Doing it right now. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ Never trust an operating system you don't have sources for. ;-) From owner-freebsd-bugs Fri Apr 14 09:20:02 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id JAA23298 for bugs-outgoing; Fri, 14 Apr 1995 09:20:02 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id JAA23291 ; Fri, 14 Apr 1995 09:20:02 -0700 Date: Fri, 14 Apr 1995 09:20:02 -0700 Message-Id: <199504141620.JAA23291@freefall.cdrom.com> From: jc@irbs.com Reply-To: jc@irbs.com To: freebsd-bugs Subject: kern/343: tcpdump with filter fails on tun and lo devices In-Reply-To: Your message of Fri, 14 Apr 1995 12:12:21 -0400 <199504141612.MAA01541@irbs.com> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 343 >Category: kern >Synopsis: tcpdump with filter fails on tun and lo devices >Confidential: no >Severity: non-critical >Priority: low >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Fri Apr 14 09:20:01 1995 >Originator: John Capo >Organization: IRBS Engineering >Release: FreeBSD 2.1.0-Development i386 >Environment: FreeBSD-current >Description: The packet filter fails to capture any packets when tcpdump is used with a filter expression on the tunnel or loopback device. >How-To-Repeat: tcpdump -i lo0 icmp on one terminal ping localhost on another terminal >Fix: I am not sure if the problem is in the drivers, the kernel packet filter, or in tcpdump. The loopback and tunnel devices attach to the bpf with a device type of DLT_NULL and dummy up a link level header that contains the adddress family for the packet, always AF_INET. Using a dummy PPP header in the drivers and attaching to the bpf with DLT_PPP works. This patch is probably not the correct fix but it will provide the missing functionality. A bpf guru needs to fix this the right way. *** sys/net/if_tun.c.orig Mon Apr 10 22:48:34 1995 --- sys/net/if_tun.c Fri Apr 14 09:44:20 1995 *************** *** 117,123 **** ifp->if_opackets = 0; if_attach(ifp); #if NBPFILTER > 0 ! bpfattach(&tunctl[i].tun_bpf, ifp, DLT_NULL, sizeof(u_int)); #endif } } --- 117,123 ---- ifp->if_opackets = 0; if_attach(ifp); #if NBPFILTER > 0 ! bpfattach(&tunctl[i].tun_bpf, ifp, DLT_PPP, sizeof(u_int)); #endif } } *************** *** 286,299 **** #if NBPFILTER > 0 if (tp->tun_bpf) { /* ! * We need to prepend the address family as ! * a four byte field. Cons up a dummy header ! * to pacify bpf. This is safe because bpf ! * will only read from the mbuf (i.e., it won't ! * try to free it or keep a pointer to it). ! */ struct mbuf m; ! u_int af = dst->sa_family; m.m_next = m0; m.m_len = 4; --- 286,298 ---- #if NBPFILTER > 0 if (tp->tun_bpf) { /* ! * Dummy up a PPP link level header. ! * This is safe because bpf will only read from the ! * mbuf (i.e., it won't try to free it or keep a ! * pointer to it). ! */ struct mbuf m; ! u_int af = htonl(0x0021); m.m_next = m0; m.m_len = 4; *************** *** 510,523 **** #if NBPFILTER > 0 if (tunctl[unit].tun_bpf) { /* ! * We need to prepend the address family as ! * a four byte field. Cons up a dummy header ! * to pacify bpf. This is safe because bpf ! * will only read from the mbuf (i.e., it won't ! * try to free it or keep a pointer to it). ! */ struct mbuf m; ! u_int af = AF_INET; m.m_next = top; m.m_len = 4; --- 509,521 ---- #if NBPFILTER > 0 if (tunctl[unit].tun_bpf) { /* ! * Dummy up a PPP link level header. ! * This is safe because bpf will only read from the ! * mbuf (i.e., it won't try to free it or keep a ! * pointer to it). ! */ struct mbuf m; ! u_int af = (0x0021); m.m_next = top; m.m_len = 4; *** sys/net/if_loop.c.orig Mon Mar 20 16:26:39 1995 --- sys/net/if_loop.c Fri Apr 14 09:44:31 1995 *************** *** 97,103 **** ifp->if_addrlen = 0; if_attach(ifp); #if NBPFILTER > 0 ! bpfattach(&ifp->if_bpf, ifp, DLT_NULL, sizeof(u_int)); #endif } --- 97,103 ---- ifp->if_addrlen = 0; if_attach(ifp); #if NBPFILTER > 0 ! bpfattach(&ifp->if_bpf, ifp, DLT_PPP, sizeof(u_int)); #endif } *************** *** 119,132 **** #if NBPFILTER > 0 if (loif.if_bpf) { /* ! * We need to prepend the address family as ! * a four byte field. Cons up a dummy header ! * to pacify bpf. This is safe because bpf ! * will only read from the mbuf (i.e., it won't ! * try to free it or keep a pointer a to it). ! */ struct mbuf m0; ! u_int af = dst->sa_family; m0.m_next = m; m0.m_len = 4; --- 119,131 ---- #if NBPFILTER > 0 if (loif.if_bpf) { /* ! * Dummy up a PPP link level header. ! * This is safe because bpf will only read from the ! * mbuf (i.e., it won't try to free it or keep a ! * pointer to it). ! */ struct mbuf m0; ! u_int af = htonl(0x0021); m0.m_next = m; m0.m_len = 4; >Audit-Trail: >Unformatted: From owner-freebsd-bugs Fri Apr 14 11:20:02 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA28258 for bugs-outgoing; Fri, 14 Apr 1995 11:20:02 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA28251 ; Fri, 14 Apr 1995 11:20:02 -0700 Date: Fri, 14 Apr 1995 11:20:02 -0700 Message-Id: <199504141820.LAA28251@freefall.cdrom.com> From: dkelly@tomcat1.tbe.com (David Kelly) Reply-To: dkelly@tomcat1.tbe.com (David Kelly) To: freebsd-bugs Subject: misc/344: Signal 11 when running a user program In-Reply-To: Your message of Fri, 14 Apr 95 13:14:20 -0500 <9504141814.AA03294@tomcat1.tbe.com> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 344 >Category: misc >Synopsis: Signal 11 when running a user program >Confidential: no >Severity: non-critical >Priority: medium >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Fri Apr 14 11:20:01 1995 >Originator: David Kelly >Organization: just little old me >Release: FreeBSD 2.1.0-Development i386 >Environment: plain old 486DX33, 8M RAM, Maxtor 850AV IDE hard drive, 950322 snap. >Description: Signal 11 on execution of user program. >How-To-Repeat: Download ftp://cochlea.hut.fi/pub/som_pak/som_pak-3.1.tar.Z uncompress un-tar cp makefile.unix Makefile remove the tab on line 95 of Makefile make all make example ./randinit bombs with Signal 11 (segmentation violation) >Fix: Found 2 ways: - comment out LDFLAGS=-s in Makefile and remake. No problem if binary has debugging symbols. - copy the binaries prior to use of the original. No problem if the entire binary has recently been read. Still don't have my FreeBSD system properly configured for mailing to the outside world, so this file was plucked from /var/spool/mqueue. Hope that's ok. Hope I did it right. Reply-To: dkelly@nebula.tbe.com (this works for sure) david.kelly@pobox.tbe.com (also works) dkelly@iquest.com (also works) dkelly@tomcat1.tbe.com (heck, I don't know, it sends ok) This stuff was in /var/spool/mqueue/q*, looked important: P121064 T797881489 DdfMAA01363 MDeferred: No route to host $_dkelly@localhost Sdkelly RFreeBSD-gnats-submit@freebsd.org H?P?Return-Path: dkelly HReceived: (from dkelly@localhost) by PeeCee.tbe.com (8.6.11/8.6.9) id MAA01363; Fri, 14 Apr 1995 12:44:49 -0500 H?D?Date: Fri, 14 Apr 1995 12:44:49 -0500 H?F?From: David Kelly H?x?Full-Name: David Kelly H?M?Message-Id: <199504141744.MAA01363@PeeCee.tbe.com> HTo: FreeBSD-gnats-submit@freebsd.org HReply-To: dkelly@nebula.tbe.com HX-send-pr-version: 3.2 -- David Kelly N4HHE, n4hhe@amsat.org, dkelly@nebula.tbe.com ---------------------------------------------------------------------- The human mind ordinarily operates at only ten percent of its capacity -- the rest is overhead for the operating system. >Audit-Trail: >Unformatted: From owner-freebsd-bugs Fri Apr 14 12:45:30 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id MAA02952 for bugs-outgoing; Fri, 14 Apr 1995 12:45:30 -0700 Received: from barkley.ME.Berkeley.EDU (barkley.ME.Berkeley.EDU [128.32.125.131]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id MAA02946 for ; Fri, 14 Apr 1995 12:45:29 -0700 Received: by barkley.ME.Berkeley.EDU (4.1/1.42) id AA18992; Fri, 14 Apr 95 12:45:28 PDT Message-Id: <9504141945.AA18992@barkley.ME.Berkeley.EDU> From: Allyn Hardyck To: freebsd-bugs@FreeBSD.org Subject: Typo in /etc/security? Organization: Avalanche Systems Inc., NYC Date: Fri, 14 Apr 95 12:45:27 -0700 Sender: bugs-owner@FreeBSD.org Precedence: bulk So far I'm pretty psyched by the way FreeBSD 2.0 is set up and performs (although I'd still like some ideas on how to set up CAP with Native Ethertalk). When I discovered that it's configured to print a list of setuid files every day and compare it to yesterday's list, I looked for ps, just as an example. I was surprised not to find it. After some experimentation I found that the command in /etc/security: find $mount -xdev -perm -u+s -or -perm -g+s seems only to return the files that have the gid bit set and not those that have the uid or both uid and gid bits set. This seems to give the union: find $mount -xdev \( -perm -u+s -or -perm -g+s \) Why are the number of returned filenames limited to 20? From owner-freebsd-bugs Fri Apr 14 23:24:54 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id XAA29664 for bugs-outgoing; Fri, 14 Apr 1995 23:24:54 -0700 Received: (from gpalmer@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id XAA29647 ; Fri, 14 Apr 1995 23:24:52 -0700 Date: Fri, 14 Apr 1995 23:24:52 -0700 From: Gary Palmer (FreeBSD/ARM Team) Message-Id: <199504150624.XAA29647@freefall.cdrom.com> To: root@rz-wb.fh-sw.de, gpalmer, freebsd-bugs Subject: Changed information for PR bin/201 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: PPP does not find config-files State-Changed-From-To: open-closed State-Changed-By: gpalmer State-Changed-When: Fri Apr 14 23:23:11 PDT 1995 State-Changed-Why: Fixed in rev 1.2 of systems.c. From owner-freebsd-bugs Fri Apr 14 23:58:33 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id XAA01880 for bugs-outgoing; Fri, 14 Apr 1995 23:58:33 -0700 Received: from irz301.inf.tu-dresden.de (irz301.inf.tu-dresden.de [141.76.1.11]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id XAA01865 for ; Fri, 14 Apr 1995 23:58:27 -0700 Received: from sax.sax.de by irz301.inf.tu-dresden.de with SMTP (5.67b+/DEC-Ultrix/4.3) id AA24898; Sat, 15 Apr 1995 08:58:22 +0200 Received: by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id IAA02748; Sat, 15 Apr 1995 08:58:22 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.6.11/8.6.9) id HAA01352; Sat, 15 Apr 1995 07:48:33 +0200 From: J Wunsch Message-Id: <199504150548.HAA01352@uriah.heep.sax.de> Subject: Re: bin/342: lpd can allow users access to all of root's groups To: pritc003@maroon.tc.umn.edu Date: Sat, 15 Apr 1995 07:48:32 +0200 (MET DST) Cc: freebsd-bugs@freefall.cdrom.com In-Reply-To: <199504141530.IAA20683@freefall.cdrom.com> from "pritc003@maroon.tc.umn.edu" at Apr 14, 95 08:30:01 am Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) 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 Content-Length: 719 Sender: bugs-owner@FreeBSD.org Precedence: bulk As pritc003@maroon.tc.umn.edu wrote: > > If a /etc/printcap entry is setup to use a filtering program that > allows the user some control over it (e.g. apsfilter), then the user > can run commands as user daemon with all of root's groups. It's arguable that this is a bug in apsfilter as well. (And actually, just *this* is the reason why Andreas Klemm didn't yet offer a port of apsfilter to FreeBSD - he knows about this bug/problem.) OTOH, it seems to be a good idea to switch UIDs regardless of this problem, since lpr filters tend to be shell scripts in some cases. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ Never trust an operating system you don't have sources for. ;-) From owner-freebsd-bugs Sat Apr 15 00:50:05 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id AAA03097 for bugs-outgoing; Sat, 15 Apr 1995 00:50:05 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id AAA03090 ; Sat, 15 Apr 1995 00:50:03 -0700 Date: Sat, 15 Apr 1995 00:50:03 -0700 Message-Id: <199504150750.AAA03090@freefall.cdrom.com> From: Heikki Suonsivu Reply-To: Heikki Suonsivu To: freebsd-bugs Subject: kern/345: Panic with 14th April supped kernel (EET, late evening 13th US time) In-Reply-To: Your message of Sat, 15 Apr 1995 10:46:45 +0300 <199504150746.KAA00893@katiska.clinet.fi> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 345 >Category: kern >Synopsis: panic "biodone: page busy < 0" >Confidential: no >Severity: serious >Priority: high >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sat Apr 15 00:50:02 1995 >Originator: Heikki Suonsivu >Organization: Helsinki University of Technology, Finland >Release: FreeBSD 2.1.0-Development i386 >Environment: P60 MSI motherboard, BT PCI SCSI (a month old), 32M, 512kB cache, 1G IBM 0662 and Seagate Barracuda 2G. FreeBSD 2.1.0-Development #0: Fri Apr 14 07:12:38 EET DST 1995 hsu@katiska.clinet.fi:/usr/current/src/sys/compile/CLINETSERVER CPU: 56-MHz Pentium 510\60 or 567\66 (Pentium-class CPU) Origin = "GenuineIntel" Id = 0x515 Stepping=5 Features=0x1bf real memory = 33161216 (8096 pages) avail memory = 30236672 (7382 pages) Probing for devices on the ISA bus: sc0 at 0x60-0x6f irq 1 on motherboard sc0: VGA color <16 virtual consoles, flags=0x0> ed0 at 0x280-0x29f irq 5 on isa ed0: address 00:4f:56:00:93:c4, type NE2000 (16 bit) bpf: ed0 attached lpt0 not found at 0x3bc sio0 at 0x3f8-0x3ff irq 4 on isa sio0: type 16450 sio1 not found at 0x2f8 sio2 not found at 0x3e8 sio3 not found at 0x2e8 pca0 on isa pca0: PC speaker audio driver bt0: Bt946C/ 0-PCI/EISA/VLB(32bit) bus bt0: reading board settings, busmastering, int=11 bt0: version 4.23, sync, parity, 32 mbxs, 32 ccbs bt0: targ 0 sync rate=10.00MB/s(100ns), offset=15 bt0: targ 1 sync rate=10.00MB/s(100ns), offset=15 bt0: Enabling Round robin scheme bt0 at 0x330 irq 11 on isa bt0 waiting for scsi devices to settle (bt0:0:0): "IBM 0662S12 !O 2 23" is a type 0 fixed SCSI 2 sd is configured at 0 sd0(bt0:0:0): Direct-Access 1003MB (2055035 S), 4119 C 5 H 99 S/T 512 B/S (bt0:1:0): "SEAGATE ST32550N 0012" is a type 0 fixed SCSI 2 sd is configured at 1 sd1(bt0:1:0): Direct-Access 2047MB (4194058 S), 3511 C 11 H 108 S/T 512 B/S aha0 not probed due to I/O address conflict with bt0 at 0x330 wdc0 not found at 0x1f0 wdc1 not found at 0x170 fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: NEC 765 fd0: 1.44MB 3.5in mcd0: timeout getting status mcd0 not found at 0x300 le0: no board found at 0x300 le0 not found at 0x300 npx0 on motherboard npx0: INT 16 interface matcd0 not found at 0xffffffff matcd1 not found at 0xffffffff matcd2 not found at 0xffffffff matcd3 not found at 0xffffffff bio_imask c0000840 tty_imask c0030032 net_imask c0030032 Probing for devices on the pci0 bus: configuration mode 2 allows 16 devices. pci0:0: vendor=0x10b9, device=0x1451, class=bridge [not supported] pci0:2: vendor=0x10b9, device=0x1449, class=old [not supported] pci0:3: vendor=0x104b, device=0x1040, class=storage [not supported] map(10): io(ffe4) changing root device to sd0a sd0s1: start 0, end = 2055034, size 2055035: OK sd1s1: start 0, end = 4194057, size 4194058: OK sd0s1: start 0, end = 2055034, size 2055035: OK bpf: ds0 attached bpf: lo0 attached bpf: ppp0 attached bpf: ppp1 attached bpf: ppp2 attached bpf: ppp3 attached bpf: ppp4 attached bpf: ppp5 attached bpf: ppp6 attached bpf: ppp7 attached bpf: ppp8 attached bpf: ppp9 attached bpf: ppp10 attached bpf: ppp11 attached bpf: ppp12 attached bpf: ppp13 attached bpf: ppp14 attached bpf: ppp15 attached bpf: ppp16 attached bpf: ppp17 attached bpf: ppp18 attached bpf: ppp19 attached bpf: ppp20 attached bpf: ppp21 attached bpf: ppp22 attached bpf: ppp23 attached bpf: ppp24 attached bpf: ppp25 attached bpf: ppp26 attached bpf: ppp27 attached bpf: ppp28 attached bpf: ppp29 attached bpf: ppp30 attached bpf: ppp31 attached bpf: sl0 attached bpf: sl1 attached bpf: sl2 attached bpf: sl3 attached bpf: sl4 attached bpf: sl5 attached bpf: sl6 attached bpf: sl7 attached bpf: sl8 attached bpf: sl9 attached bpf: sl10 attached bpf: sl11 attached bpf: sl12 attached bpf: sl13 attached bpf: sl14 attached bpf: sl15 attached bpf: tun0 attached sd0s1: start 0, end = 2055034, size 2055035: OK sd1s1: start 0, end = 4194057, size 4194058: OK sd1s1: start 0, end = 4194057, size 4194058: OK biodone: page busy < 0, off: 196608, foff: 196608, resid: 4096, index: 0 iosize: 8192, lblkno: 24 valid: 0xff, dirty: 0x0, mapped: 0 panic: biodone: page busy < 0 syncing disks... 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 giving up dumping to dev 401, offset 344064 dump 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 >Description: Current directory is /var/crash/ Reading symbol data from /var/crash/kernel.15...done. (kgdb) core vmcore.15 IdlePTD 246000 panic: biodone: page busy < 0 current pcb at 1f74c4 Reading in symbols for ../../i386/i386/machdep.c...done. (kgdb) up Reading in symbols for ../../kern/subr_prf.c...done. #1 0xf0114ab3 in panic (fmt=(char *) 0xf0126f68 "biodone: page busy < 0\n") (../../kern/subr_prf.c line 128) 128 (../../kern/subr_prf.c) (kgdb) directory /usr/src/sys/compile/CLINETSERVER Source directories searched: /m/katiska/news/crash:/usr/src/sys/compile/CLINETSERVER (kgdb) up Reading in symbols for ../../kern/vfs_bio.c...done. #2 0xf0127187 in biodone (bp=(struct buf *) 0xf27043b8) (../../kern/vfs_bio.c line 1134) (kgdb) print *m $1 = {pageq = {tqe_next = 0x0, tqe_prev = 0xf02c12c4}, hashq = {tqe_next = 0x0, tqe_prev = 0xf0268580}, listq = {tqe_next = 0xf02bb2e4, tqe_prev = 0xf02bec08}, object = 0xf0e2a280, offset = 0x00030000, phys_addr = 0x018dc000, wire_count = 0x0000, flags = 0x0042, hold_count = 0, act_count = 0x0000, bmapped = 0x0000, busy = 0x0000, valid = 0x00ff, dirty = 0x0000} (kgdb) bt #0 boot (arghowto=256) (../../i386/i386/machdep.c line 850) #1 0xf0114ab3 in panic (fmt=(char *) 0xf0126f68 "biodone: page busy < 0\n") (../../kern/subr_prf.c line 128) #2 0xf0127187 in biodone (bp=(struct buf *) 0xf27043b8) (../../kern/vfs_bio.c line 1134) #3 0xf015cf2b in nfs_doio (...) #4 0xf015bcba in nfs_bioread (...) #5 0xf0174022 in nfs_read (...) #6 0xf012f012 in vn_read (...) #7 0xf0115c9b in read (...) #8 0xf01b9a8d in syscall (...) (kgdb) up Reading in symbols for ../../nfs/nfs_bio.c...done. #3 0xf015cf2b in nfs_doio (bp=(struct buf *) 0xf27043b8, cr=(struct ucred *) 0xf0db7f00, p=(struct proc *) 0xf0d4a900) (../../nfs/nfs_bio.c line 867) (kgdb) up #4 0xf015bcba in nfs_bioread (vp=(struct vnode *) 0xf0ab9800, uio=(struct uio *) 0xefbfff28, ioflag=0, cred=(struct ucred *) 0xf0db7f00) (../../nfs/nfs_bio.c line 249) (kgdb) down #3 0xf015cf2b in nfs_doio (bp=(struct buf *) 0xf27043b8, cr=(struct ucred *) 0xf0db7f00, p=(struct proc *) 0xf0d4a900) (../../nfs/nfs_bio.c line 867) (kgdb) down #2 0xf0127187 in biodone (bp=(struct buf *) 0xf27043b8) (../../kern/vfs_bio.c line 1134) (kgdb) down #1 0xf0114ab3 in panic (fmt=(char *) 0xf0126f68 "biodone: page busy < 0\n") (../../kern/subr_prf.c line 128) (kgdb) up #2 0xf0127187 in biodone (bp=(struct buf *) 0xf27043b8) (../../kern/vfs_bio.c line 1134) (kgdb) up #3 0xf015cf2b in nfs_doio (bp=(struct buf *) 0xf27043b8, cr=(struct ucred *) 0xf0db7f00, p=(struct proc *) 0xf0d4a900) (../../nfs/nfs_bio.c line 867) (kgdb) print bp $2 = (struct buf *) 0xf27043b8 (kgdb) print *bp $3 = {b_hash = {le_next = 0xf26e8408, le_prev = 0xf0228c24}, b_vnbufs = {le_next = 0xf26e9928, le_prev = 0xf0ab982c}, b_freelist = {tqe_next = 0x0, tqe_prev = 0xf01f7648}, b_actf = 0x0, b_actb = 0x0, b_proc = 0x0, b_flags = 537920016, b_qindex = 0, b_error = 0, b_bufsize = 8192, b_bcount = 8192, b_resid = 0, b_dev = 0xffffffff, b_un = {b_addr = 0xf3438000 "\366t\005\270D\024\004"}, b_saveaddr = 0x0, b_lblkno = 24, b_blkno = 384, b_iodone = 0x0, b_iodone_chain = 0x0, b_vp = 0xf0ab9800, b_pfcent = 0, b_dirtyoff = 0, b_dirtyend = 0, b_rcred = 0xffffffff, b_wcred = 0xffffffff, b_validoff = 0, b_validend = 8192, b_pblkno = 233688, b_savekva = 0x0, b_driver1 = 0x0, b_driver2 = 0x0, b_spc = 0x0, b_pages = {0xf02c9c7c, 0xf0283488, 0x0 }, b_npages = 2} (kgdb) up #4 0xf015bcba in nfs_bioread (vp=(struct vnode *) 0xf0ab9800, uio=(struct uio *) 0xefbfff28, ioflag=0, cred=(struct ucred *) 0xf0db7f00) (../../nfs/nfs_bio.c line 249) (kgdb) print cred $4 = (struct ucred *) 0xf0db7f00 (kgdb) print *cred $5 = {cr_ref = 0x002d, cr_uid = 0x000005b5, cr_ngroups = 2, cr_groups = {0x000003e8, 0x000003e8, 0x00000000 }} (kgdb) print p $6 = (struct proc *) 0xf0d4a900 (kgdb) print *proc No symbol "proc" in current context. (kgdb) print *p $7 = {p_forw = 0xf0228efc, p_back = 0x0, p_next = 0xf0bdad00, p_prev = 0xf0e65c08, p_cred = 0xf0cbfa20, p_fd = 0xf0ce0180, p_stats = 0xf4694288, p_limit = 0xf022fdf4, p_vmspace = 0xf0b8d100, p_sigacts = 0xf469415c, p_flag = 16390, p_stat = 2, p_pad1 = {"\276\255\336"}, p_pid = 6952, p_hash = 0x0, p_pgrpnxt = 0x0, p_pptr = 0xf0bdad00, p_osptr = 0x0, p_ysptr = 0x0, p_cptr = 0x0, p_oppid = 0, p_dupfd = 0, p_estcpu = 0x000003f8, p_cpticks = 994, p_pctcpu = 0x0000001e, p_wchan = 0x0, p_wmesg = 0xf01e41d0 "netio", p_swtime = 0x00000090, p_slptime = 0x00000000, p_realtimer = {it_interval = {tv_sec = 0, tv_usec = 0}, it_value = {tv_sec = 0, tv_usec = 0}}, p_rtime = {tv_sec = 5, tv_usec = 29118}, p_uticks = 0x18e, p_sticks = 0x48d, p_iticks = 0x56, p_traceflag = 0, p_tracep = 0x0, p_siglist = 0, p_textvp = 0xf0ab9800, p_lock = 0, p_pad2 = {"\000\000\000"}, p_spare = {0, 0}, p_sigmask = 0x00000000, p_sigignore = 0x10489000, p_sigcatch = 0x28044003, p_priority = 0x18, p_usrpri = 0x7f, ! p_nice = 0, p_comm = {"irc1\000n", '\000' }, p_pgrp = 0xf0e85e40, p_sysent = 0xf01e3620, p_rtprio = {type = 0x0001, prio = 0x0000}, p_thread = -559038737, p_addr = 0xf4694000, p_md = {md_flags = -559038737, md_regs = 0xefbfffbc}, p_xstat = 0xbeef, p_acflag = 0x0000, p_ru = 0xdeadbeef} (kgdb) list 244 got_buf = 1; 245 if ((bp->b_flags & (B_DONE | B_DELWRI)) == 0) { 246 bp->b_flags |= B_READ; 247 not_readin = 0; 248 vfs_busy_pages(bp, 0); 249 error = nfs_doio(bp, cred, p); 250 if (error) { 251 brelse(bp); 252 return (error); 253 } (kgdb) up Reading in symbols for ../../nfs/nfs_vnops.c...done. #5 0xf0174022 in nfs_read (ap=(struct vop_read_args *) 0xefbffedc) (../../nfs/nfs_vnops.c line 780) (kgdb) list 775 { 776 register struct vnode *vp = ap->a_vp; 777 778 if (vp->v_type != VREG) 779 return (EPERM); 780 return (nfs_bioread(vp, ap->a_uio, ap->a_ioflag, ap->a_cred)); 781 } 782 783 /* 784 * nfs readlink call (kgdb) print *vp Cannot read memory: address 0x0 out of bounds. (kgdb) print *ap Cannot read memory: address 0x0 out of bounds. (kgdb) print ap $8 = (struct vop_read_args *) 0x0 (kgdb) print vp $9 = (struct vnode *) 0x0 (kgdb) print (struct vop_read_args *) 0xefbffedc Undefined command: "". (kgdb) print *(struct vop_read_args *) 0xefbffedc $10 = (struct vop_read_args *) 0xefbffedc (kgdb) print *(struct vop_read_args *) 0xefbffedc $11 = {a_desc = 0xf01e223c, a_vp = 0xf0ab9800, a_uio = 0xefbfff28, a_ioflag = 0, a_cred = 0xf0db7f00} (kgdb) up Reading in symbols for ../../kern/vfs_vnops.c...done. #6 0xf012f012 in vn_read (fp=(struct file *) 0xf0ec9fc0, uio=(struct uio *) 0xefbfff28, cred=(struct ucred *) 0xf0db7f00) (./vnode_if.h line 211) Source file is more recent than executable. (kgdb) print *vp $12 = {v_flag = 0x00002002, v_usecount = 4, v_writecount = 0, v_holdcnt = 25, v_lastr = 23, v_id = 0x0001c621, v_mount = 0xf0afb600, v_op = 0xf0a18f00, v_freelist = {tqe_next = 0xf0c02a80, tqe_prev = 0xf0adec9c}, v_mntvnodes = {le_next = 0xf0ba4c80, le_prev = 0xf0b800a4}, v_cleanblkhd = {lh_first = 0xf27043b8}, v_dirtyblkhd = {lh_first = 0x0}, v_numoutput = 0, v_type = VREG, v_un = {vu_mountedhere = 0x0, vu_socket = 0x0, vu_specinfo = 0x0, vu_fifoinfo = 0x0}, v_lease = 0x0, v_lastw = 0, v_cstart = 0, v_lasta = 0, v_clen = 0, v_ralen = 0, v_maxra = 0, v_vmdata = 0xf0e2a280 "\364\324,\360\364\262+\360", v_tag = VT_NFS, v_data = 0xf0df5900} (kgdb) print *uio $13 = {uio_iov = 0xefbfff20, uio_iovcnt = 1, uio_offset = 196608, uio_resid = 2048, uio_segflg = UIO_USERSPACE, uio_rw = UIO_READ, uio_procp = 0xf0d4a900} (kgdb) print *ioflag No symbol "ioflag" in current context. (kgdb) print *cred $14 = {cr_ref = 0x002d, cr_uid = 0x000005b5, cr_ngroups = 2, cr_groups = {0x000003e8, 0x000003e8, 0x00000000 }} (kgdb) list 206 a.a_desc = VDESC(vop_read); 207 a.a_vp = vp; 208 a.a_uio = uio; 209 a.a_ioflag = ioflag; 210 a.a_cred = cred; 211 return (VCALL(vp, VOFFSET(vop_read), &a)); 212 } 213 struct vop_write_args { 214 struct vnodeop_desc *a_desc; 215 struct vnode *a_vp; (kgdb) up Reading in symbols for ../../kern/sys_generic.c...done. #7 0xf0115c9b in read (p=(struct proc *) 0xf0d4a900, uap=(struct read_args *) 0xefbfff94, retval=(int *) 0xefbfff8c) (../../kern/sys_generic.c line 112) (kgdb) print *fp Cannot read memory: address 0x0 out of bounds. (kgdb) print fp $15 = (struct file *) 0x0 (kgdb) print *(struct read_args *) 0xefbfff94 $16 = {fd = 7, buf = 0xefbfbfb8 , nbyte = 0x00000800} (kgdb) up Reading in symbols for ../../i386/i386/trap.c...done. #8 0xf01b9a8d in syscall (frame={tf_es = 39, tf_ds = 39, tf_edi = 617216, tf_esi = -272646216, tf_ebp = -272644164, tf_isp = -272629788, tf_ebx = 617216, tf_edx = 292332, tf_ecx = 8, tf_eax = 3, tf_trapno = 530, tf_err = 530, tf_eip = 134625685, tf_cs = 31, tf_eflags = 530, tf_esp = -272646244, tf_ss = 39}) (../../i386/i386/trap.c line 828) (kgdb) print args $17 = {7, -272646216, 2048, 0, 0, 0, 1073741823, -272644164} (kgdb) print rval $18 = {0, 292332} (kgdb) up Initial frame selected; you cannot go up. (kgdb) list 823 ktrsyscall(p->p_tracep, code, callp->sy_narg, args); 824 #endif 825 rval[0] = 0; 826 rval[1] = frame.tf_edx; 827 828 error = (*callp->sy_call)(p, args, rval); 829 830 switch (error) { 831 832 case 0: (kgdb) down #7 0xf0115c9b in read (p=(struct proc *) 0xf0d4a900, uap=(struct read_args *) 0xefbfff94, retval=(int *) 0xefbfff8c) (../../kern/sys_generic.c line 112) (kgdb) list 107 */ 108 if (KTRPOINT(p, KTR_GENIO)) 109 ktriov = aiov; 110 #endif 111 cnt = uap->nbyte; 112 if ((error = (*fp->f_ops->fo_read)(fp, &auio, fp->f_cred))) 113 if (auio.uio_resid != cnt && (error == ERESTART || 114 error == EINTR || error == EWOULDBLOCK)) 115 error = 0; 116 cnt -= auio.uio_resid; (kgdb) down #6 0xf012f012 in vn_read (fp=(struct file *) 0xf0ec9fc0, uio=(struct uio *) 0xefbfff28, cred=(struct ucred *) 0xf0db7f00) (./vnode_if.h line 211) (kgdb) list 206 a.a_desc = VDESC(vop_read); 207 a.a_vp = vp; 208 a.a_uio = uio; 209 a.a_ioflag = ioflag; 210 a.a_cred = cred; 211 return (VCALL(vp, VOFFSET(vop_read), &a)); 212 } 213 struct vop_write_args { 214 struct vnodeop_desc *a_desc; 215 struct vnode *a_vp; (kgdb) down #5 0xf0174022 in nfs_read (ap=(struct vop_read_args *) 0xefbffedc) (../../nfs/nfs_vnops.c line 780) (kgdb) list 775 { 776 register struct vnode *vp = ap->a_vp; 777 778 if (vp->v_type != VREG) 779 return (EPERM); 780 return (nfs_bioread(vp, ap->a_uio, ap->a_ioflag, ap->a_cred)); 781 } 782 783 /* 784 * nfs readlink call (kgdb) down #4 0xf015bcba in nfs_bioread (vp=(struct vnode *) 0xf0ab9800, uio=(struct uio *) 0xefbfff28, ioflag=0, cred=(struct ucred *) 0xf0db7f00) (../../nfs/nfs_bio.c line 249) (kgdb) list 244 got_buf = 1; 245 if ((bp->b_flags & (B_DONE | B_DELWRI)) == 0) { 246 bp->b_flags |= B_READ; 247 not_readin = 0; 248 vfs_busy_pages(bp, 0); 249 error = nfs_doio(bp, cred, p); 250 if (error) { 251 brelse(bp); 252 return (error); 253 } (kgdb) down #3 0xf015cf2b in nfs_doio (bp=(struct buf *) 0xf27043b8, cr=(struct ucred *) 0xf0db7f00, p=(struct proc *) 0xf0d4a900) (../../nfs/nfs_bio.c line 867) (kgdb) list 862 } 863 bp->b_dirtyoff = bp->b_dirtyend = 0; 864 } 865 } 866 bp->b_resid = uiop->uio_resid; 867 biodone(bp); 868 return (error); 869 } (kgdb) down #2 0xf0127187 in biodone (bp=(struct buf *) 0xf27043b8) (../../kern/vfs_bio.c line 1134) (kgdb) list 1129 m->offset, foff, resid, i); 1130 printf(" iosize: %d, lblkno: %d\n", 1131 bp->b_vp->v_mount->mnt_stat.f_iosize, bp->b_lblkno); 1132 printf(" valid: 0x%x, dirty: 0x%x, mapped: %d\n", 1133 m->valid, m->dirty, m->bmapped); 1134 panic("biodone: page busy < 0\n"); 1135 } 1136 --m->busy; 1137 if( (m->busy == 0) && (m->flags & PG_WANTED)) 1138 wakeup((caddr_t) m); (kgdb) list 1122 1117 if (!bogusflag && resid > 0) { 1118 vm_page_set_valid(m, foff & (PAGE_SIZE-1), resid); 1119 vm_page_set_clean(m, foff & (PAGE_SIZE-1), resid); 1120 } 1121 1122 /* 1123 * when debugging new filesystems or buffer I/O methods, this 1124 * is the most common error that pops up. if you see this, you 1125 * have not set the page busy flag correctly!!! 1126 */ (kgdb) list 1127 if (m->busy == 0) { 1128 printf("biodone: page busy < 0, off: %d, foff: %d, resid: %d, index: %d\n", 1129 m->offset, foff, resid, i); 1130 printf(" iosize: %d, lblkno: %d\n", 1131 bp->b_vp->v_mount->mnt_stat.f_iosize, bp->b_lblkno); 1132 printf(" valid: 0x%x, dirty: 0x%x, mapped: %d\n", 1133 m->valid, m->dirty, m->bmapped); 1134 panic("biodone: page busy < 0\n"); 1135 } 1136 --m->busy; (kgdb) list 1137 if( (m->busy == 0) && (m->flags & PG_WANTED)) 1138 wakeup((caddr_t) m); 1139 --obj->paging_in_progress; 1140 foff += resid; 1141 iosize -= resid; 1142 } 1143 if (obj && obj->paging_in_progress == 0 && 1144 (obj->flags & OBJ_PIPWNT)) { 1145 obj->flags &= ~OBJ_PIPWNT; 1146 wakeup((caddr_t) obj); (kgdb) list 1147 } 1148 } 1149 /* 1150 * For asynchronous completions, release the buffer now. The brelse 1151 * checks for B_WANTED and will do the wakeup there if necessary - so 1152 * no need to do a wakeup here in the async case. 1153 */ 1154 1155 if (bp->b_flags & B_ASYNC) { 1156 brelse(bp); (kgdb) >How-To-Repeat: Hard to say. The system tries to run as an nntp server but there are often other users in the system as well. With -current it seems to panic about twice a day, sometimes corrupting a couple of files or a directory. I'll try to send more dumps and update kernel every couple of days or so. >Fix: Dunno, -- Heikki Suonsivu, T{ysikuu 10 C 83/02210 Espoo/FINLAND, hsu@cs.hut.fi home +358-0-8031121 work -4513377 fax -4555276 riippu SN >Audit-Trail: >Unformatted: From owner-freebsd-bugs Sat Apr 15 04:30:09 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id EAA11978 for bugs-outgoing; Sat, 15 Apr 1995 04:30:09 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id EAA11971 for freebsd-bugs; Sat, 15 Apr 1995 04:30:08 -0700 Date: Sat, 15 Apr 1995 04:30:08 -0700 From: GNU GNATS Message-Id: <199504151130.EAA11971@freefall.cdrom.com> To: freebsd-bugs Subject: List of open Problem Reports Sender: bugs-owner@FreeBSD.org Precedence: bulk This is the list of currently open problem reports [1994/09/14] kern/1 Toshiba MK156FB scsi drive does not work with 2.0 ke [1994/10/07] kern/6 System hangs while using slip or ppp connection [1994/11/13] i386/14 No MBR put on prestine IDE disk drive [1994/11/14] bin/17 "mount -t foo /dev/wd1h /mnt" gives incorrect error [1994/11/15] gnu/19 cvs commit core dumps on multiple files [1994/11/18] kern/21 kernel panic. [1994/11/18] conf/22 Cannot use links to share kernel objects [1994/11/23] kern/25 mounting a photocd panics the kernel [1994/11/23] kern/27 Loopback mounts confuse unmount-on-reboot code [1994/11/30] kern/34 nullfs and union mounts can result in wild pointer r [1994/12/03] kern/39 Recent config changes hosed ft config [1994/12/04] kern/40 Floppy tape probe hosed (FIX) [1994/12/06] kern/42 PC-NFS on -current can't access disks [1994/12/07] docs/43 manpage for keysu is missing. reference from keyinit [1994/12/07] kern/44 ep driver won't work with anything but bnc [1994/12/08] kern/46 ed driver don't know the link0 flag (IFF_LINK0) [1994/12/13] misc/48 /etc/rc does not recover recovery files [1994/12/14] kern/51 Panic accessing NFS mounted file system [1994/12/14] gnu/54 No manual entry for uucpd [1994/12/18] misc/67 It took 20 tries to get free-bsd installed [1994/12/18] misc/68 bininst leaves system kinda hung [1994/12/21] misc/75 Netboot ignores gateway [1994/12/20] conf/76 fdisk can't be trusted. [1994/12/21] i386/79 Reboots sometimes don't complete. [1994/12/23] misc/83 System crash after abrupt end of slip session [1995/01/03] kern/95 Configuring a SLIP device with source = destination [1995/01/09] kern/102 PANIC after running Merged Buffer Cache kernel for a [1995/01/10] bin/104 pax -rwl may corrupt filesystem [1995/01/11] i386/105 Distributed libm (msun) has non-standard error handl [1995/01/13] gnu/107 kernel build produces internal compiler error [1995/01/14] bin/109 at command doesn't parse dates time and doesn't prom [1995/01/14] misc/112 ^C cause panic during install [1995/01/14] bin/113 /bin/csh and /usr/local/bin/tcsh choke on my nasty c [1995/01/14] bin/115 systat iostat display doesn't scale high enough [1995/01/14] bin/116 disk transfer rates reported by systat :iostat are t [1995/01/14] misc/118 use of 386bsd partition tag makes multi-os boot hard [1995/01/14] misc/119 you cannot tune your newfs parameters at intall time [1995/01/14] misc/120 first Initial security report is bogus becuase of ls [1995/01/14] bin/124 traceroute doesn't support -g flag. [1995/01/14] docs/127 locate man page does not have any clues to how the d [1995/01/14] misc/128 default owners of system directories is a security h [1995/01/14] bin/129 fsck cannot take a mount point as an argument [1995/01/14] docs/131 there is no man page for uucpd [1995/01/14] docs/132 gettytab documentation is split between /etc/gettyta [1995/01/14] bin/133 mail program doesn't have editheaders option [1995/01/14] misc/134 ispell (in ports) thinks the dictionary is in /usr/d [1995/01/14] bin/135 not enough ptys; virtual console names conflict with [1995/01/14] docs/137 domainname man page references getdomainname(2) whic [1995/01/14] kern/140 cdplay will not work with NEC210 [1995/01/15] bin/146 version of compress is kinda old and slow [1995/01/19] bin/151 One of the pkg_add set /tmp mod to 755 [1995/01/19] kern/152 Kernel config doesn't handle "irq 2" as irq 9 [1995/01/20] bin/153 mount -u improvement for diskless systems [1995/01/21] kern/157 Quota system does not work [1995/01/21] docs/159 uucp manual files have wrong file locations. [1995/01/21] docs/161 Not MAKEDEV(8) manual page. [1995/01/21] bin/163 telneting sometimes doesn't yield a "login:" prompt [1995/01/21] docs/164 no ptrace(2) manual page. [1995/01/21] docs/165 No documentation on kernel boot flags. [1995/01/21] misc/166 /usr/include/machine/asmacros.h includes uninstalled [1995/01/21] bin/167 makewhatis ignores symbolic links. [1995/01/21] kern/169 Errors from mkdir & mv when directories paths end wi [1995/01/21] bin/170 ar does not print an error message with readonly lib [1995/01/21] docs/171 No documentation on .forward files. [1995/01/21] bin/172 cp -f does not work [1995/01/21] bin/173 rc trys to mount modload fs before ld is available. [1995/01/21] bin/174 Poor error message from stty [1995/01/22] kern/175 Syscons does not recover X graphics mode [1995/01/22] kern/176 EIDRM not defined in errno.h [1995/01/22] docs/177 man pages missing for SYSV IPC funtions [1995/01/23] kern/180 Fiddling with process stack and using scanf with inv [1995/01/24] gnu/183 can't resolve "operator <<" overload [1995/01/24] bin/184 send-pr says "Aborting ..." and happily removes the [1995/01/25] kern/185 kernel stability problems - can't sustain uptimes > [1995/01/25] conf/187 Files referenced in Makefiles do not exist [1995/01/25] bin/188 disklabel uses c partition by default, kernel says d [1995/01/30] gnu/196 size of bss in *.o's reported wrong by size [1995/01/30] bin/198 1.1.5.1 pine binary loops; top shows fancy values [1995/02/01] bin/199 quiz(6) reading database bug, pattern matching bug [1995/02/01] misc/202 System hang or reboot without clues, tcp ppp named r [1995/02/02] bin/204 Installing failes [1995/02/08] kern/211 mv returns "Invalid argument" unreasonably [1995/02/14] kern/214 Using wt driver crashes system [1995/02/14] kern/216 /kernel: panic: ffs_alloccg: map corrupted [1995/02/14] docs/218 dbm references from hash(3) [1995/02/16] kern/219 Performance on local net drops too much when SNDBUF [1995/02/16] misc/220 Cancel still doesn't work during install [1995/02/16] misc/221 lots of error messages during unpacking via ftp inst [1995/02/16] i386/222 boot prompt doesn't always work [1995/02/16] misc/223 default domain shouldn't be hostname [1995/02/18] kern/224 ppp net serial [1995/02/18] kern/225 Panic: invalid wire count when GDB kills stopped inf [1995/03/02] misc/229 acos() core dump [1995/03/03] bin/230 C++ contructors not being called [1995/03/05] gnu/231 send-pr initializes Organization with ~/.signature [1995/03/05] docs/232 The mandoc .St macro doesn't work or misses IEEE754 [1995/03/10] misc/236 2.0-950210-SNAP/src/extract.sh references "release" [1995/03/11] kern/238 failed assertion in ncr.c --> no more scsi disk acce [1995/03/13] bin/241 strange behavior of /bin/sh [1995/03/17] kern/247 Berkeley Packet Filter fix [1995/03/17] kern/248 scbus attach/probe information is inconsistent [1995/03/18] i386/251 disklabel(8) refuses to install new bootstrap [1995/03/18] kern/253 Tagged queuing with an Adaptec 2842 controller doesn [1995/03/19] misc/254 fsck -p misbehaves for SIGQUIT [1995/03/19] gnu/256 nasty gdb bug back in 031595 system [1995/03/20] docs/263 There is absolutely no documentation for the dialog [1995/03/21] docs/265 mtrace command lacks man pages [1995/03/21] kern/266 ifconfig panics kernel built without option NFS [1995/03/27] bin/274 touch -t with century doesn't work [1995/03/28] kern/275 qic-02 streamer won't work [1995/03/28] kern/277 netstat -r reports entries with negative expire time [1995/03/28] kern/280 the new slice code is bitching about my old slices [1995/03/28] kern/281 Messages printed when checking CD ROM device too ver [1995/03/28] kern/282 buslogic adapter information WAY too verbose [1995/03/29] kern/284 Network gives permission denied [1995/03/30] gnu/285 ld.so problem with -lg++ [1995/03/30] conf/286 'make depend' on GENERIC kernel stops [1995/03/30] kern/287 Several header files are not consistent [1995/03/31] kern/288 panic: free: multiple frees [1995/04/01] gnu/289 info(1) cannot handle arrow keys on a VT100 [1995/04/01] kern/291 PCI devices still probe/attach after being disabled [1995/04/01] misc/292 rfc1323 and rfc1644 support can confuse terminal ser [1995/04/01] kern/293 wd0: interrupt timeout [1995/04/02] kern/294 Bootstrap has problems with -g compiled kernels [1995/04/04] i386/297 matcd does not play audio CDs from cdplay [1995/04/04] kern/299 SCSI probes stop at ID 7 even for WIDE busses [1995/04/05] bin/300 __fdnlist fails on kernels created by cc -g and stri [1995/04/06] conf/301 Log rotation wastes _much_ to much space [1995/04/06] kern/302 Changing st tapes after extract crashes system [1995/04/07] kern/304 root rm fails, if user set uchg/uappnd flag [1995/04/08] pending/307 request for a customer id [1995/04/08] bin/308 Do-it-yourself FTP needs to prompt for what to downl [1995/04/08] bin/309 Do-it-yourself FTP deletes /usr/tmp dir rather than [1995/04/08] bin/310 ^C or ^Z in do-it-yourself FTP installation crashes [1995/04/08] bin/311 Do-it-yourself FTP deletes /usr/tmp dir rather than [1995/04/09] bin/313 BACKSPACE key not producing 0x08 breaks many things [1995/04/09] bin/314 CTRL-Z during installation crashes install FDIV008 [1995/04/09] bin/315 Kernel and boot block don't agree on kernel name FD [1995/04/09] bin/316 SNAP950322 less stable on IDE than earlier releases [1995/04/09] docs/317 Install README doesn't list matcd as install choice [1995/04/09] bin/318 matcd driver not on boot or fixit floppy FDIV012 [1995/04/09] bin/319 Install doesn't strip sys name in prompt for domain [1995/04/09] bin/320 TCP/IP settings entered b4 install are not kept FDIV [1995/04/09] bin/321 TCP/IP settings given during install are not kept FD [1995/04/09] bin/322 No way to restart a failed FTP download FDIV016 [1995/04/09] bin/323 Creating lost+found causes fsck to stop fsck -y FDIV [1995/04/09] bin/324 /usr/include/sys -> /usr/sys/sys -> /usr/sys/sys FD [1995/04/09] bin/325 boot -s SH is confused by BACKSPACE <> 0x08 FDIV019 [1995/04/09] bin/326 Weekly cron generates some usage and error messages [1995/04/09] bin/327 Clock management punishes you if CMOS != GMT FDIV02 [1995/04/09] bin/328 Checking for -c kernel changes randomly hangs FDIV0 [1995/04/09] bin/329 FTP transfers above 99K shown in scientific notation [1995/04/11] kern/332 Incorrect routes can cause system to reset [1995/04/11] kern/336 [1995/04/11] i386/337 seagate controller code hangs on boot. [1995/04/14] ports/340 USE_GMAKE does not work with USE_IMAKE in ports [1995/04/14] docs/341 vnconfig(8) disagrees with its man page [1995/04/14] bin/342 lpd can allow users access to all of root's groups [1995/04/14] kern/343 tcpdump with filter fails on tun and lo devices [1995/04/14] misc/344 Signal 11 when running a user program [1995/04/15] kern/345 panic "biodone: page busy < 0" From owner-freebsd-bugs Sat Apr 15 04:30:51 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id EAA12005 for bugs-outgoing; Sat, 15 Apr 1995 04:30:51 -0700 Received: from grunt.grondar.za (grunt.grondar.za [196.7.18.129]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id EAA11982 ; Sat, 15 Apr 1995 04:30:13 -0700 Received: from localhost (localhost [127.0.0.1]) by grunt.grondar.za (8.6.11/8.6.9) with SMTP id NAA09007; Sat, 15 Apr 1995 13:29:56 +0200 Message-Id: <199504151129.NAA09007@grunt.grondar.za> X-Authentication-Warning: grunt.grondar.za: Host localhost didn't use HELO protocol To: bugs@FreeBSD.org, ports@FreeBSD.org Subject: xdvi is hosed. Date: Sat, 15 Apr 1995 13:29:56 +0200 From: Mark Murray Sender: bugs-owner@FreeBSD.org Precedence: bulk Hi Subject line says it all. If I go in (after make configure) and manually edit the xdvi: target to include '-lXt -lXaw -lX11' it looks a whole lot better, except that two of its own symbols are not found at link time. I am no X hacker. Help! M -- Mark Murray 46 Harvey Rd, Claremont, Cape Town 7700, South Africa +27 21 61-3768 GMT+0200 From owner-freebsd-bugs Sat Apr 15 07:42:42 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id HAA14115 for bugs-outgoing; Sat, 15 Apr 1995 07:42:42 -0700 Received: from irz301.inf.tu-dresden.de (irz301.inf.tu-dresden.de [141.76.1.11]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id HAA14108 ; Sat, 15 Apr 1995 07:42:32 -0700 Received: from sax.sax.de by irz301.inf.tu-dresden.de with SMTP (5.67b+/DEC-Ultrix/4.3) id AA05870; Sat, 15 Apr 1995 16:41:55 +0200 Received: by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id QAA05407; Sat, 15 Apr 1995 16:41:53 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.6.11/8.6.9) id QAA04740; Sat, 15 Apr 1995 16:33:52 +0200 From: J Wunsch Message-Id: <199504151433.QAA04740@uriah.heep.sax.de> Subject: Re: xdvi is hosed. To: mark@grondar.za (Mark Murray) Date: Sat, 15 Apr 1995 16:33:50 +0200 (MET DST) Cc: bugs@FreeBSD.org, ports@FreeBSD.org In-Reply-To: <199504151129.NAA09007@grunt.grondar.za> from "Mark Murray" at Apr 15, 95 01:29:56 pm Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) 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 Content-Length: 967 Sender: bugs-owner@FreeBSD.org Precedence: bulk As Mark Murray wrote: > > Hi > > Subject line says it all. If I go in (after make configure) and manually > edit the xdvi: target to include '-lXt -lXaw -lX11' it looks a whole lot > better, except that two of its own symbols are not found at link time. I'm the last one who's put his fingers on this port. :-/ Mysterious. When i've tried it here before commiting (and also on thud after commiting) it was compilable. I could swear it. :) Ok. This whole Makefile/Imakefile crud there is too weird. The most terrible Imakefile i've ever seen. :-( They do only rely on gmake since gmake continues if it cannot find an include file (while bmake would bail out), and the Makefile is used to generate an include file for the Makefile. Strange abuse. Anyway, i give up and apologize. Gonna back out my changes... :-( -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ Never trust an operating system you don't have sources for. ;-) From owner-freebsd-bugs Sat Apr 15 07:56:16 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id HAA14672 for bugs-outgoing; Sat, 15 Apr 1995 07:56:16 -0700 Received: (from joerg@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id QAA14658 ; Sat, 15 Apr 1995 16:56:15 +0200 Date: Sat, 15 Apr 1995 16:56:15 +0200 From: Joerg Wunsch Message-Id: <199504151456.QAA14658@freefall.cdrom.com> To: uhclem%nemesis@fw.ast.com, joerg, freebsd-bugs Subject: Changed information for PR bin/313 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: BACKSPACE key not producing 0x08 breaks many things FDIV007 State-Changed-From-To: open-closed State-Changed-By: joerg State-Changed-When: Sat Apr 15 16:53:10 MET DST 1995 State-Changed-Why: This was already obsoleted at time of submission, kbdtables.h rev 1.28 reverted back to the prevoius behaviour: revision 1.28 date: 1995/03/27 19:41:11; author: sos; state: Exp; lines: +6 -6 Give backspace to the people (again) From owner-freebsd-bugs Sat Apr 15 08:00:23 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id IAA14824 for bugs-outgoing; Sat, 15 Apr 1995 08:00:23 -0700 Received: (from joerg@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id RAA14810 ; Sat, 15 Apr 1995 17:00:22 +0200 Date: Sat, 15 Apr 1995 17:00:22 +0200 From: Joerg Wunsch Message-Id: <199504151500.RAA14810@freefall.cdrom.com> To: uhclem%nemesis@fw.ast.com, joerg, freebsd-bugs Subject: Changed information for PR bin/325 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: boot -s SH is confused by BACKSPACE <> 0x08 FDIV019 State-Changed-From-To: open-closed State-Changed-By: joerg State-Changed-When: Sat Apr 15 16:59:19 MET DST 1995 State-Changed-Why: This was already obsoleted at time of submission, kbdtables.h rev 1.28 reverted back to the prevoius behaviour: revision 1.28 date: 1995/03/27 19:41:11; author: sos; state: Exp; lines: +6 -6 Give backspace to the people (again) From owner-freebsd-bugs Sat Apr 15 08:04:32 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id IAA14948 for bugs-outgoing; Sat, 15 Apr 1995 08:04:32 -0700 Received: (from joerg@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id RAA14934 ; Sat, 15 Apr 1995 17:04:32 +0200 Date: Sat, 15 Apr 1995 17:04:32 +0200 From: Joerg Wunsch Message-Id: <199504151504.RAA14934@freefall.cdrom.com> To: ache@astral.msk.su, joerg, freebsd-bugs Subject: Changed information for PR kern/304 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: root rm fails, if user set uchg/uappnd flag State-Changed-From-To: open-closed State-Changed-By: joerg State-Changed-When: Sat Apr 15 17:01:50 MET DST 1995 State-Changed-Why: Andrey has fixed this himself: Working file: rm.c revision 1.6 date: 1995/04/13 13:35:09; author: ache; state: Exp; lines: +64 -29 Allow root to delete uchg/uappnd files From owner-freebsd-bugs Sat Apr 15 08:49:59 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id IAA15335 for bugs-outgoing; Sat, 15 Apr 1995 08:49:59 -0700 Received: from peri.swt.edu (peri.swt.edu [147.26.10.13]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id IAA15328 for ; Sat, 15 Apr 1995 08:49:51 -0700 Received: from swt.edu by swt.edu (PMDF V4.3-10 #6249) id <01HPD5N3R99SDMWRGH@swt.edu>; Sat, 15 Apr 1995 10:49:43 -0600 (CST) Date: Sat, 15 Apr 1995 10:49:43 -0600 (CST) From: "James C. Cook" Subject: Bugs with 950412-SNAP To: bugs@FreeBSD.org Message-id: <01HPD5N3RC3MDMWRGH@swt.edu> Organization: Southwest Texas State University X-VMS-To: IN%"bugs@freebsd.org" MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII Content-transfer-encoding: 7BIT Sender: bugs-owner@FreeBSD.org Precedence: bulk Hi, I am trying to install FreeBSD via floppies on my PC. Alright, it went through the initial installation fine. I got through setting up the partitions, and labeling the sectors, etc. Alright, I am trying to install the bin files, and well.. I am getting problems. I direct it to look at a DOS partition I have set up, and well, it goes and looks at it, and says Checksum error.. Not at a particular file, as it seems it would want to say with do_cksum.sh, but it just says CheckSum error.. Also, at the same time at the bottom of the screen, it says: do_cksum.sh 62: Syntax Error: end of file unexpected (expecting "fi") I added fi to the end of the file, and it didn't do anything for me, and instead said the error was on the next line. Could you help me? Jim --------------------------------------------------------------------------- Jim Cook Email Addresses: jc03436@academia.swt.edu jc03436@dali.math.swt.edu From owner-freebsd-bugs Sat Apr 15 08:55:16 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id IAA15427 for bugs-outgoing; Sat, 15 Apr 1995 08:55:16 -0700 Received: from mail.eskimo.com (root@mail.eskimo.com [204.122.16.4]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id IAA15421 ; Sat, 15 Apr 1995 08:55:15 -0700 Received: from eskimo.com by mail.eskimo.com (5.65c/1.35) id AA27955; Sat, 15 Apr 1995 08:55:12 -0700 Date: Sat, 15 Apr 1995 08:55:10 -0700 (PDT) From: Dave Haas To: freebsd-hardware@FreeBSD.org Cc: freebsd-bugs@FreeBSD.org Subject: Three button emulation in XFree with MS Ergonomic Mouse Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: bugs-owner@FreeBSD.org Precedence: bulk Hi there. I've experienced a weird problem with three-button emulation using a Microsoft Mouse (one of the two button, ergonomic deals) in XFree86 3.1.1. Starting from the configuration of XFree, I have selected three-button emulation. And I have subsequently and frequently verified in /etc/XF86Config that indeeed three-button emulation is specified. However, regardless of how I start XFree, three-button emulation doesn't really work. Or, I should say that, on occasion, I've gotten it to work after doing [Ctrl]-[Alt]-[F1]'s followed by [Alt]-[F4]'s. But this is largely ineffectual and seems to be luck, as it's not reproducible. As an example of how the mouse doesn't work, take Xman. When I start Xman and click on the [Manual Page] button, it correctly brings up the initial Xman help page. From the Options and Sections pulldowns menus, however, I am unable to select any options. The cursor simply floats over any of these menu items, without highlighting them. Similarly, in editres, viewres, xfontsel, etc., the same problem exists. Other than fundamental philosophical differences, can anyone suggest what may be causing this disagreement between the mouse and Xfree. Dave Haas daveh@eskimo.com From owner-freebsd-bugs Sat Apr 15 09:16:46 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id JAA15901 for bugs-outgoing; Sat, 15 Apr 1995 09:16:46 -0700 Received: from mpp.com (dialup-4-61.gw.umn.edu [128.101.96.61]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id JAA15894 for ; Sat, 15 Apr 1995 09:16:25 -0700 Received: (from mpp@localhost) by mpp.com (8.6.11/8.6.9) id LAA07985 for freebsd-bugs@freebsd.org; Sat, 15 Apr 1995 11:15:37 -0500 From: Mike Pritchard Message-Id: <199504151615.LAA07985@mpp.com> Subject: at (was Re: List of open Problem Reports) To: freebsd-bugs@FreeBSD.org Date: Sat, 15 Apr 1995 11:15:36 -0500 (CDT) In-Reply-To: <199504151130.EAA11971@freefall.cdrom.com> from "GNU GNATS" at Apr 15, 95 04:30:08 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 972 Sender: bugs-owner@FreeBSD.org Precedence: bulk > [1995/01/14] bin/109 at command doesn't parse dates time and doesn't prom Nearly everything mentioned in this pr was fixed by the new at that was installed about a week ago, except that the new at still doesn't give any type of prompt message asking you for the commands that you want to run. Speaking of the new at, it no longer supports abbreviations for minute/hour/day/week like the old at did. E.g. I can't do: "at now +1d", instead I have to do "at now +1day" instead. The old abbreviations were: min/m/hr/h/d/w. It might be nice to get them back. The new at also is unable to remove at jobs from the queue. E.g. atrm doesn't work, and interrupting at while it is reading stdin for commands fails to remove the /var/at/jobs/XXX file. This is because the REDUCDE_PRIVS macro in at winds up setting both uid/euid to the uid of the caller. -- Mike Pritchard pritc003@maroon.tc.umn.edu "Go that way. Really fast. If something gets in your way, turn" From owner-freebsd-bugs Sat Apr 15 11:11:50 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA19125 for bugs-outgoing; Sat, 15 Apr 1995 11:11:50 -0700 Received: from star-gate.com (hasty.vip.best.com [204.156.141.143]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id LAA19098 ; Sat, 15 Apr 1995 11:11:41 -0700 Received: from localhost (localhost [127.0.0.1]) by star-gate.com (8.6.11/8.6.9) with SMTP id LAA02794; Sat, 15 Apr 1995 11:04:05 GMT Message-Id: <199504151104.LAA02794@star-gate.com> X-Authentication-Warning: star-gate.com: Host localhost didn't use HELO protocol X-Mailer: exmh version 1.6delta 4/7/95 To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) cc: mark@grondar.za (Mark Murray), bugs@FreeBSD.org, ports@FreeBSD.org Subject: Re: xdvi is hosed. In-reply-to: Your message of "Sat, 15 Apr 1995 16:33:50 +0200." <199504151433.QAA04740@uriah.heep.sax.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 15 Apr 1995 11:04:03 +0000 From: Amancio Hasty Sender: bugs-owner@FreeBSD.org Precedence: bulk >>> J Wunsch said: > As Mark Murray wrote: > > > > Hi > > > > Subject line says it all. If I go in (after make configure) and manually > > edit the xdvi: target to include '-lXt -lXaw -lX11' it looks a whole lot > > better, except that two of its own symbols are not found at link time. > Try including the following X libs : -lICE -lSM -lXt -lXaw -lX11 if that does not work, try to post the undefined symbols. Amancio From owner-freebsd-bugs Sat Apr 15 12:40:24 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id MAA21517 for bugs-outgoing; Sat, 15 Apr 1995 12:40:24 -0700 Received: from irz301.inf.tu-dresden.de (irz301.inf.tu-dresden.de [141.76.1.11]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id MAA21488 for ; Sat, 15 Apr 1995 12:39:12 -0700 Received: from sax.sax.de by irz301.inf.tu-dresden.de with SMTP (5.67b+/DEC-Ultrix/4.3) id AA10351; Sat, 15 Apr 1995 21:16:07 +0200 Received: by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id VAA07152; Sat, 15 Apr 1995 21:16:06 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.6.11/8.6.9) id VAA24292; Sat, 15 Apr 1995 21:07:20 +0200 From: J Wunsch Message-Id: <199504151907.VAA24292@uriah.heep.sax.de> Subject: Re: Three button emulation in XFree with MS Ergonomic Mouse To: daveh@eskimo.com (Dave Haas) Date: Sat, 15 Apr 1995 21:07:19 +0200 (MET DST) Cc: freebsd-bugs@FreeBSD.org In-Reply-To: from "Dave Haas" at Apr 15, 95 08:55:10 am Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) 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 Content-Length: 776 Sender: bugs-owner@FreeBSD.org Precedence: bulk As Dave Haas wrote: > > As an example of how the mouse doesn't work, take Xman. When I start Xman > and click on the [Manual Page] button, it correctly brings up the initial > Xman help page. From the Options and Sections pulldowns menus, however, I am > unable to select any options. >From your description, this ain't a problem with 3-button emulation. It's rather one of *the* FAQs these days. Try Section "Keyboard" ... ServerNumLock ... EndSection in your {/etc,/usr/X11R6/lib/X11}/XF86Config file. It seems that one of the recent changes in XFree86 has caused servers without ServerNumLock to break. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ Never trust an operating system you don't have sources for. ;-) From owner-freebsd-bugs Sat Apr 15 12:57:49 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id MAA21900 for bugs-outgoing; Sat, 15 Apr 1995 12:57:49 -0700 Received: from silver.sms.fi (silver.sms.fi [193.64.137.1]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id MAA21894 for ; Sat, 15 Apr 1995 12:57:47 -0700 Received: (from pete@localhost) by silver.sms.fi (8.6.11/8.6.9) id WAA20830; Sat, 15 Apr 1995 22:57:43 +0300 Date: Sat, 15 Apr 1995 22:57:43 +0300 Message-Id: <199504151957.WAA20830@silver.sms.fi> From: Petri Helenius To: freebsd-bugs@FreeBSD.org Subject: vat audio Sender: bugs-owner@FreeBSD.org Precedence: bulk Is there anything that could be done to get the vat audio working with PAS/16 ? Pete From owner-freebsd-bugs Sat Apr 15 15:32:25 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id PAA05131 for bugs-outgoing; Sat, 15 Apr 1995 15:32:25 -0700 Received: (from jmz@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id PAA05113 ; Sat, 15 Apr 1995 15:32:21 -0700 Date: Sat, 15 Apr 1995 15:32:21 -0700 From: Jean-Marc Zucconi Message-Id: <199504152232.PAA05113@freefall.cdrom.com> To: mark@grondar.za, jmz, freebsd-bugs Subject: Changed information for PR ports/340 Sender: bugs-owner@FreeBSD.org Precedence: bulk Synopsis: USE_GMAKE does not work with USE_IMAKE in ports State-Changed-From-To: open-closed State-Changed-By: jmz State-Changed-When: Sat Apr 15 15:23:12 PDT 1995 State-Changed-Why: Fixed From owner-freebsd-bugs Sat Apr 15 16:03:20 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id QAA06254 for bugs-outgoing; Sat, 15 Apr 1995 16:03:20 -0700 Received: from sequent.kiae.su (sequent.kiae.su [144.206.136.6]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id QAA06248 for ; Sat, 15 Apr 1995 16:03:17 -0700 Received: by sequent.kiae.su id AA29968 (5.65.kiae-2 ); Sun, 16 Apr 1995 02:46:26 +0400 Received: by sequent.KIAE.su (UUMAIL/2.0); Sun, 16 Apr 95 02:46:25 +0400 Received: (from ache@localhost) by astral.msk.su (8.6.8/8.6.6) id CAA00689; Sun, 16 Apr 1995 02:08:09 +0400 To: freebsd-bugs@FreeBSD.org, Mike Pritchard References: <199504151615.LAA07985@mpp.com> In-Reply-To: <199504151615.LAA07985@mpp.com>; from Mike Pritchard at Sat, 15 Apr 1995 11:15:36 -0500 (CDT) Message-Id: Organization: Olahm Ha-Yetzirah Date: Sun, 16 Apr 1995 02:08:09 +0400 X-Mailer: Mail/@ [v2.32 FreeBSD] From: "Andrey A. Chernov, Black Mage" X-Class: Fast Subject: Re: at (was Re: List of open Problem Reports) Lines: 24 Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Length: 1028 Sender: bugs-owner@FreeBSD.org Precedence: bulk In message <199504151615.LAA07985@mpp.com> Mike Pritchard writes: >Speaking of the new at, it no longer supports abbreviations for >minute/hour/day/week like the old at did. E.g. I can't do: >"at now +1d", instead I have to do "at now +1day" instead. >The old abbreviations were: min/m/hr/h/d/w. It might be nice >to get them back. It is original author intention, please contact with him on this subject. >The new at also is unable to remove at jobs from the queue. E.g. atrm >doesn't work, and interrupting at while it is reading stdin for >commands fails to remove the /var/at/jobs/XXX file. This is because >the REDUCDE_PRIVS macro in at winds up setting both uid/euid to >the uid of the caller. I just commit the fix. -- Andrey A. Chernov : And I rest so composedly, /Now, in my bed, ache@astral.msk.su : That any beholder /Might fancy me dead - FidoNet: 2:5020/230.3 : Might start at beholding me, /Thinking me dead. RELCOM Team,FreeBSD Team : E.A.Poe From "For Annie" 1849 From owner-freebsd-bugs Sat Apr 15 21:20:24 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id VAA17874 for bugs-outgoing; Sat, 15 Apr 1995 21:20:24 -0700 Received: from skynet.ctr.columbia.edu (skynet.ctr.columbia.edu [128.59.64.70]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id VAA17868 for ; Sat, 15 Apr 1995 21:20:21 -0700 Received: (from wpaul@localhost) by skynet.ctr.columbia.edu (8.6.8/8.6.6) id XAA00363 for bugs@freefall.cdrom.com; Sat, 15 Apr 1995 23:17:42 -0400 Date: Sat, 15 Apr 1995 23:17:42 -0400 From: "House of Debuggin'" Message-Id: <199504160317.XAA00363@skynet.ctr.columbia.edu> To: bugs@freefall.cdrom.com Subject: if_ppp, if_sl loadable module problems Sender: bugs-owner@FreeBSD.org Precedence: bulk Tonight I discovered that if you load either the if_ppp_mod or if_sl_mod loadable kernel modules, then try to do a 'modstat,' the system panics inside lkmexists() in kern_lkm.c. (I think. This is a close approximation.) I think this has to do with the fact that it's trying to do a strcmp() on the module name, but there may not a name associated with either of these modules because they aren't put together like any of the others. I might just be installing them wrong. Unfortunately, I have no way to tell for sure because these bloody modules appear to be completely undocumented, so there's no way to know what the right way is. I've been doing it like this: # modload -e pppattach if_ppp_mod.o This works, and I can ifconfig the ppp0 and ppp1 interfaces with no trouble. Unfortunately, modstat causes a panic and modunload -i 0 hangs the machine hard (unfortunately, the machine is at work, and I'm at home, so I can't do any more testing until I go in tomorrow and reset it :( ). I have a feeling this is also caused by the same lkmexists() lossage. I'm going to do more hacking on this tomorrow, though if someone can spot the bug before I do, I'd love to hear from you. -Bill ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~T~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -Bill Paul (212) 854-6020 | System Manager Work: wpaul@ctr.columbia.edu | Center for Telecommunications Research Home: wpaul@skynet.ctr.columbia.edu | Columbia University, New York City ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ The Møøse Illuminati: ignore it and be confused, or join it and be confusing! ~~~~~~~~ FreeBSD 2.1.0-Development #0: Tue Mar 14 11:11:25 EST 1995 ~~~~~~~~~ From owner-freebsd-bugs Sat Apr 15 22:30:06 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id WAA19179 for bugs-outgoing; Sat, 15 Apr 1995 22:30:06 -0700 Received: (from gnats@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id WAA19170 ; Sat, 15 Apr 1995 22:30:04 -0700 Date: Sat, 15 Apr 1995 22:30:04 -0700 Message-Id: <199504160530.WAA19170@freefall.cdrom.com> From: mcgovern@spoon.beta.com Reply-To: mcgovern@spoon.beta.com To: freebsd-bugs Subject: kern/347: ft0 not detected at boot In-Reply-To: Your message of Sun, 16 Apr 1995 00:59:29 -0400 <199504160459.AAA13111@spoon.beta.com> Sender: bugs-owner@FreeBSD.org Precedence: bulk >Number: 347 >Category: kern >Synopsis: ft0 fails detection on boot >Confidential: no >Severity: non-critical >Priority: medium >Responsible: freebsd-bugs (FreeBSD bugs mailing list) >State: open >Class: sw-bug >Submitter-Id: current-users >Arrival-Date: Sat Apr 15 22:30:02 1995 >Originator: Brian J. McGovern >Organization: B.E.T.A. Mountain >Release: FreeBSD 2.1.0-Development i386 >Environment: Intel 486/33, Adaptec 1542CF SCSI Adapter, Jumbo 250 Tape drive and/or Irwin 40MB Tape drive Description: During boot process, kernel accurately detects existance of 5.25" 1.2MB Floppy disk. However, it fails to detect the existance of a QIC-40 or QIC-80 tape device, although ft0 has been properly configured in to the kernel (config lines follow). Older revisions (up to and including 2.0-RELEASE) see drives properly. Config lines used - controller fdc0 at isa? port "IO_FD1" bio irq 6 drq 2 vector fdintr disk fd0 at fdc0 drive 0 disk fd1 at fdc0 drive 1 tape ft0 at fdc0 drive 2 How-To-Repeat: Repeatable simply by rebooting system. Fix: >Description: >How-To-Repeat: >Fix: >Audit-Trail: >Unformatted: