From owner-freebsd-hackers Sun Mar 30 00:10:36 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA14919 for hackers-outgoing; Sun, 30 Mar 1997 00:10:36 -0800 (PST) Received: from nlsystems.com (nlsys.demon.co.uk [158.152.125.33]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA14908 for ; Sun, 30 Mar 1997 00:10:28 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by nlsystems.com (8.8.5/8.8.5) with SMTP id JAA01792; Sun, 30 Mar 1997 09:08:17 +0100 (BST) Date: Sun, 30 Mar 1997 09:08:17 +0100 (BST) From: Doug Rabson To: "David E. Cross" cc: freebsd-hackers@FreeBSD.ORG, support@4front-tech.com Subject: Re: OSS/FreeBSD and Ensoniq Soundscape In-Reply-To: <199703300516.AAA04782@phoenix.its.rpi.edu> Message-ID: MIME-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="0-1998856922-859709297=:271" Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. Send mail to mime@docserver.cac.washington.edu for more info. --0-1998856922-859709297=:271 Content-Type: TEXT/PLAIN; charset=US-ASCII [Added support@4front-tech.com to Cc list] On Sun, 30 Mar 1997, David E. Cross wrote: > I got the OSS/FreeBSD drivers earlier this week, and have my SoundScape card > "mostly" working. The MIDI works flawlessly, but the wave support is verry > unstable, it acts as if there is a DMA/IRQ conflict, but there is not... > (I have tried both my "tuned" kernel and GENERIC, both have the same > failure mode). If anyone else has a SoundScape card, and would be willing > to work with me on this, please let me know. I have an Ensoniq Soundscape S2000 in my machine. I installed OSS last night and playback seems to work pretty well. I have a lot of problems with recording though. Even if I select a high data rate, sound recording is extremely noisy. Also, if I try to record 8 bit samples, it complains about DMA problems a few times and them completely locks up the system. The attached program shows this when used to record from the Line-1 input (CD player). -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 181 951 1891 --0-1998856922-859709297=:271 Content-Type: APPLICATION/octet-stream; name="recordplay.c" Content-Transfer-Encoding: BASE64 Content-ID: Content-Description: Simple record/playback I2luY2x1ZGUgPHN0ZGlvLmg+CiNpbmNsdWRlIDxmY250bC5oPgojaW5jbHVk ZSA8bWFjaGluZS9zb3VuZGNhcmQuaD4KCmludCBtYWluKGludCBhcmdjLCBj aGFyKiogYXJndikKewogICAgaW50IGZkOwogICAgaW50IHQ7CgogICAgZmQg PSBvcGVuKCIvZGV2L2RzcCIsIE9fUkRXUik7CiAgICBpZiAoZmQgPCAwKSB7 CglwZXJyb3IoImRzcCIpOwoJZXhpdCgxKTsKICAgIH0KICAgIHQgPSBBRk1U X1M4OwogICAgaWYgKGlvY3RsKGZkLCBTTkRDVExfRFNQX1NFVEZNVCwgJnQp IDwgMCkgewoJcGVycm9yKCJmb3JtYXQiKTsKCWV4aXQoMSk7CiAgICB9CiAg ICBpZiAodCAhPSBBRk1UX1M4KSB7CglmcHJpbnRmKHN0ZGVyciwgIndlaXJk IGZvcm1hdCAleFxuIiwgdCk7CglleGl0KDEpOwogICAgfQogICAgdCA9IDA7 CiAgICBpZiAoaW9jdGwoZmQsIFNORENUTF9EU1BfU1RFUkVPLCAmdCkgPCAw KSB7CglwZXJyb3IoInN0ZXJlbyIpOwoJZXhpdCgxKTsKICAgIH0KICAgIGlm ICh0ICE9IDApIHsKCWZwcmludGYoc3RkZXJyLCAibm8gc3RlcmVvXG4iKTsK CWV4aXQoMSk7CiAgICB9CiAgICB0ID0gMzIwMDA7CiAgICBpZiAoaW9jdGwo ZmQsIFNORENUTF9EU1BfU1BFRUQsICZ0KSA8IDApIHsKCXBlcnJvcigic3Bl ZWQiKTsKCWV4aXQoMSk7CiAgICB9CiAgICBpZiAodCAhPSAzMjAwMCkgewoJ ZnByaW50ZihzdGRlcnIsICJ3ZWlyZCBzcGVlZCAlZFxuIiwgdCk7CglleGl0 KDEpOwogICAgfQoKICAgIGlmIChhcmdjICE9IDIgfHwgKHN0cmNtcChhcmd2 WzFdLCAiLXIiKSAmJiBzdHJjbXAoYXJndlsxXSwgIi13IikpKSB7CglmcHJp bnRmKHN0ZGVyciwgInVzYWdlOiByZWNvcmRwbGF5IFstcnwtd11cbiIpOwoJ ZXhpdCgxKTsKICAgIH0KICAgIAogICAgaWYgKCFzdHJjbXAoYXJndlsxXSwg Ii1yIikpCglmb3IgKDs7KSB7CgkgICAgY2hhciBidWZbNDA5Nl07CgkgICAg aW50IG47CgoJICAgIG4gPSByZWFkKGZkLCBidWYsIHNpemVvZiBidWYpOwoJ ICAgIGlmIChuIDwgMCkgewoJCXBlcnJvcigicmVhZCIpOwoJCWV4aXQoMSk7 CgkgICAgfQoJICAgIGlmIChuID09IDApCgkJYnJlYWs7CgkgICAgd3JpdGUo MSwgYnVmLCBuKTsKCX0KICAgIGVsc2UKCWZvciAoOzspIHsKCSAgICBjaGFy IGJ1Zls0MDk2XTsKCSAgICBpbnQgbjsKCgkgICAgbiA9IHJlYWQoMCwgYnVm LCBzaXplb2YgYnVmKTsKCSAgICBpZiAobiA8IDApIHsKCQlwZXJyb3IoInJl YWQiKTsKCQlleGl0KDEpOwoJICAgIH0KCSAgICBpZiAobiA9PSAwKQoJCWJy ZWFrOwoJICAgIHdyaXRlKGZkLCBidWYsIG4pOwoJfQp9CgAAAAA= --0-1998856922-859709297=:271-- From owner-freebsd-hackers Sun Mar 30 01:16:15 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA16768 for hackers-outgoing; Sun, 30 Mar 1997 01:16:15 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA16763 for ; Sun, 30 Mar 1997 01:16:13 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id BAA08407; Sun, 30 Mar 1997 01:16:11 -0800 (PST) To: Brian Tao cc: FREEBSD-HACKERS-L Subject: Re: StrongARM based NC with NetBSD In-reply-to: Your message of "Sun, 30 Mar 1997 02:03:18 EST." Date: Sun, 30 Mar 1997 01:16:10 -0800 Message-ID: <8403.859713370@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > On Thu, 27 Mar 1997, Jordan K. Hubbard wrote: > > > > FreeBSD is also involved in this project - it runs on the machines > > (NC servers) which feed these little diskless NC boxes, > > These NC servers are presumably on Intel hardware then? Yes. Though an ALPHA port of FreeBSD is definitely wanted so that they can run on *all* of DEC's server hardware. :) [And DEC actually makes some very nice x86 PC hardware]. Anyone interested in consulting on an ALPHA port? Chris D. is sort of interested, but he'd prefer to work on just the generic bits of the ALPHA port and NOT get involved in re-doing all the things in FreeBSD that he had to do in NetBSD - he just can't contemplate doing all that over. So it'd be nice if we could maybe get a couple of people involved - one or two to play ALPHA portmeisters on the FreeBSD side and maybe some of Chris's time as a consultant or something. And I already asked the NetBSD core group but there were no takers. :) Jordan From owner-freebsd-hackers Sun Mar 30 02:16:36 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id CAA18455 for hackers-outgoing; Sun, 30 Mar 1997 02:16:36 -0800 (PST) Received: from rah.star-gate.com (rah.star-gate.com [204.188.121.18]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id CAA18450 for ; Sun, 30 Mar 1997 02:16:31 -0800 (PST) Received: from rah.star-gate.com (localhost.star-gate.com [127.0.0.1]) by rah.star-gate.com (8.8.5/8.7.3) with ESMTP id CAA15207; Sun, 30 Mar 1997 02:16:23 -0800 (PST) Message-Id: <199703301016.CAA15207@rah.star-gate.com> X-Mailer: exmh version 1.6.9 8/22/96 To: Doug Rabson cc: "David E. Cross" , freebsd-hackers@freebsd.org Subject: Re: OSS/FreeBSD and Ensoniq Soundscape In-reply-to: Your message of "Sun, 30 Mar 1997 09:08:17 +0100." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 30 Mar 1997 02:16:23 -0800 From: Amancio Hasty Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, Try to post multimedia questions to the multimedia mailing list;specially, due to the high volume posting in this list. Amancio From owner-freebsd-hackers Sun Mar 30 06:49:40 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA26846 for hackers-outgoing; Sun, 30 Mar 1997 06:49:40 -0800 (PST) Received: from suntan.tandem.com (suntan.tandem.com [192.216.221.8]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id GAA26827 for ; Sun, 30 Mar 1997 06:49:23 -0800 (PST) Received: from adm.loc201.tandem.com by suntan.tandem.com (8.6.12/suntan5.970212) for id GAA20581; Sun, 30 Mar 1997 06:49:18 -0800 Received: from idc.tandem.com (adm.idc.tandem.com) by adm.loc201.tandem.com (4.1/6main.940209) id AA15236; Sun, 30 Mar 97 06:49:11 PST Received: from Arkavathi. idc.tandem.com ([192.168.14.16]) by idc.tandem.com (4.1/SMI-4.1) id AA14624; Sun, 30 Mar 97 20:18:32+050 Received: from localhost by Arkavathi. idc.tandem.com (4.1/SMI-4.1) id AA25122; Sun, 30 Mar 97 20:19:34+050 Date: Sun, 30 Mar 1997 20:19:33 +0530 (GMT+0530) From: Chirayu Patel To: hackers@freebsd.org Subject: More info reqd. Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, I am currently using a Pentium 75Mhz system with Linux and Win95 installed on it. But i never use Win95..it's just there. I recently came across some postings on FreeBSD ..which got me interested in it. Is it posible to install FreeBSD in the Win95 partition on my machine..if yes..the how?? Also i would like to know...how fast is the FreeBSd kernel code..as compared to the Linux kernel code...and can some someone give me some pointers to sites where i can find more info abt the kerel code. Thanx in advance Chirayu Chirayu Patel -/- email: chirayu@wipinfo.soft.net ,chirayu@idc.tandem.com ------------------------------------------------------------------------- Linux is addictive, I am hooked! From owner-freebsd-hackers Sun Mar 30 06:54:35 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA27145 for hackers-outgoing; Sun, 30 Mar 1997 06:54:35 -0800 (PST) Received: from plum.cyber.com.au (plum.cyber.com.au [203.7.155.24]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id GAA27134 for ; Sun, 30 Mar 1997 06:54:27 -0800 (PST) Received: (from darrenr@localhost) by plum.cyber.com.au (8.6.12/8.6.6) id AAA28455 for hackers@freebsd.org; Mon, 31 Mar 1997 00:54:11 +1000 From: Darren Reed Message-Id: <199703301454.AAA28455@plum.cyber.com.au> Subject: dump & restore for MS-DOS filesystems. To: hackers@freebsd.org Date: Mon, 31 Mar 1997 00:54:11 +1000 (EST) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Well, a seemingly hard task has been completed...how "complete" it is, I won't venture to say, except that I've hacked on dump & restore enough to make them both work on MS-DOS format file systems. I'm pretty confident that dump is working correctly: for each file, it dumps the entire cluster of sectors that are allocated to it and directories are the same. restore works...but is very naive about long names, etc. I'm not quite sure where any translation between modes should be done, as well as the extra info. that gets stored in the directory blocks. For now, those entries are just ignored (even though they are on the backup `tape'). When restoring, it uses the mode bits as they are for the MS-DOS file. I'd really like to hear back if anyone has any useful ideas on what to do with file permission bits. I've been working, compiling and testing on FreeBSD-2.1.6. The source is available at: ftp://ftp.cyber.com.au/pub/unix/dump_msdos.tgz ftp://ftp.cyber.com.au/pub/unix/restore_msdos.tgz (and retains the BSD copyrights...) Oh, one last question, is it ok to add these to the -current tree, under usr.sbin for FreeBSD ? (At this point they've really just been hacked until they work in an `expected' manner). Darren sample output: % ll /dos/d/SDD total 1208 -rwxrwxr-x 1 root wheel 334 Dec 10 11:14 file_id.diz* -rwxrwxr-x 1 root wheel 561132 Jan 25 00:52 install.dat* -rwxrwxr-x 1 root wheel 62122 Jan 24 22:07 install.exe* -rwxrwxr-x 1 root wheel 5211 Dec 12 17:01 readme.txt* -rwxrwxr-x 1 root wheel 595133 Mar 17 00:53 sdd53a-d.zip* % ./dump_msdos 0f - /dos/d > wd1s0.dump DUMP: Date of this level 0 dump: Sun Mar 30 23:37:19 1997 DUMP: Date of last level 0 dump: the epoch DUMP: Dumping /dev/rwd2s1 (/dos/d) to standard output DUMP: mapping (Pass I) [regular files, directories] DUMP: estimated 17359 tape blocks. DUMP: dumping (Pass II) [directories] DUMP: dumping (Pass III) [regular files] DUMP: DUMP: 17408 tape blocks DUMP: DUMP IS DONE % ./restore_msdos ivf ../dump_msdos/wd1s0.dump Verify tape and initialize maps Tape block size is 32 Dump date: Sun Mar 30 23:37:19 1997 Dumped from: the epoch Level 0 dump of /dos/d on freebsd:/dev/wd2s1 Label: NO NAME FAT16 Extract directories from tape Volume: [MS-DOS_6 ] Initialize symbol table. restore > ls .: 2 *./ 3 NETSCAPE/ 2629 S16DW3UP/ 489 SBBASIC/ 2810 WIN32S/ 2811 CTCM31BB/ 1283 RECYCLED 1671 S2/ 1672 SDD/ restore > add SDD Make node ./SDD restore > restore > extract Extract requested files You have not read any tapes yet. Unless you know which volume your file(s) are on you should start with the last volume and work towards towards the first. Specify next volume #: 1 extract file ./SDD/FILE_ID.DIZ extract file ./SDD/INSTALL.EXE extract file ./SDD/README.TXT extract file ./SDD/INSTALL.DAT extract file ./SDD/SDD53A-D.ZIP Add links Set directory mode, owner, and times. set owner/mode for '.'? [yn] n restore > % ll -d SDD d----w---- 2 darrenr cyber 512 Oct 17 1972 SDD/ % chmod 755 SDD % ll SDD total 1220 ----r----- 1 darrenr cyber 334 Sep 21 1972 FILE_ID.DIZ ----r----- 1 darrenr cyber 561132 Oct 11 1972 INSTALL.DAT ----r----- 1 darrenr cyber 62122 Oct 11 1972 INSTALL.EXE ----r----- 1 darrenr cyber 5211 Sep 21 1972 README.TXT ----r----- 1 darrenr cyber 595133 Oct 18 1972 SDD53A-D.ZIP % chmod u+r SDD/* % cd SDD % foreach i (*) foreach? cmp $i /dos/d/sdd/$i foreach? end % From owner-freebsd-hackers Sun Mar 30 06:56:13 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA27226 for hackers-outgoing; Sun, 30 Mar 1997 06:56:13 -0800 (PST) Received: from fallout.campusview.indiana.edu (fallout.campusview.indiana.edu [149.159.1.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA27221 for ; Sun, 30 Mar 1997 06:56:10 -0800 (PST) Received: from localhost (jfieber@localhost) by fallout.campusview.indiana.edu (8.8.5/8.8.5) with SMTP id JAA02969; Sun, 30 Mar 1997 09:55:59 -0500 (EST) Date: Sun, 30 Mar 1997 09:55:58 -0500 (EST) From: John Fieber Reply-To: John Fieber To: "David E. Cross" cc: freebsd-hackers@FreeBSD.org, support@4front-tech.com Subject: Re: OSS/FreeBSD and Ensoniq Soundscape In-Reply-To: <199703300516.AAA04782@phoenix.its.rpi.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk On Sun, 30 Mar 1997, David E. Cross wrote: > I got the OSS/FreeBSD drivers earlier this week, and have my SoundScape card > "mostly" working. The MIDI works flawlessly, but the wave support is verry > unstable, it acts as if there is a DMA/IRQ conflict, but there is not... Check the IRQs used by rndcontrol(8) at boot time. I had my soundblaster in the kernel at IRQ 10. When I tried out OSS, it reported a conflict getting IRQ 10 which stumped me for a long time, then I realized that rndcontrol was grabbing IRQ 10 for use by /dev/random before the sound driver could get it. My problem now is anything that tries to write to /dev/dsp becomes unkillable, and (according to top) the kernel continuously pages out at about 700K/s until I reboot. Top also reports the cache size to grow to about 15 megabytes (32 megabyte machine). Bizarre. Anyway, I don't really have the spare time right now to fiddle so I've gone back to the stock sound driver. -john From owner-freebsd-hackers Sun Mar 30 09:53:28 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA03616 for hackers-outgoing; Sun, 30 Mar 1997 09:53:28 -0800 (PST) Received: from news.IAEhv.nl (root@news.IAEhv.nl [194.151.64.4]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id JAA03611 for ; Sun, 30 Mar 1997 09:53:16 -0800 (PST) Received: from LOCAL (uucp@localhost) by news.IAEhv.nl (8.6.13/1.63) with IAEhv.nl; pid 19026 on Sun, 30 Mar 1997 19:53:02 +0200; id TAA19026 efrom: peter@grendel.IAEhv.nl; eto: UNKNOWN Received: (from peter@localhost) by grendel.IAEhv.nl (8.8.5/8.8.4) id EAA03026; Sat, 29 Mar 1997 04:10:03 +0100 (MET) Message-ID: <19970329040952.25223@grendel.IAEhv.nl> Date: Sat, 29 Mar 1997 04:09:52 +0100 From: Peter Korsten To: "Gregory D. Moncreaff" Cc: "hackers@freebsd.org" Subject: Re: C++ Code in Kernel References: <01BC3BA3.6FFD23A0@dominus.ultranet.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.65e_p2,4-5,7,11,15,18,21-22 In-Reply-To: <01BC3BA3.6FFD23A0@dominus.ultranet.com>; from Gregory D. Moncreaff on Fri, Mar 28, 1997 at 06:10:51PM -0500 Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Gregory D. Moncreaff shared with us: > maybe its just that I'm taking a c++ class now, but I'm > cusious as to the pro's and con's on writing kernel source > in c++. I haven't seen nor heard on this happening on FreeBSD > [but then my travels haven't been that wide] I like C++ very much as a language. It hands me constructs which enable me to make abstractions of programming problems that are all solveable in C, be it with much more effort. With the right classes, my programs become very simple and thus less error-prone. I no longer care about buffer overruns, they simply don't occur. On the other hand, Unix is not really an object-oriented OS, so the added value of C++ (primarily data abstraction and object orientation) has little meaning. Also, performance is a consideration. C++ code tends to have more overhead. Finally, it hides a lot from you. I don't know if that's desirable in kernel code. - Peter -- Peter Korsten | peter@grendel.IAEhv.nl (UUCP) | peterk@IAEhv.nl C/C++/Perl/Java hacker From owner-freebsd-hackers Sun Mar 30 11:42:01 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA08267 for hackers-outgoing; Sun, 30 Mar 1997 11:42:01 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA08262 for ; Sun, 30 Mar 1997 11:41:58 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id MAA08284; Sun, 30 Mar 1997 12:25:35 -0700 From: Terry Lambert Message-Id: <199703301925.MAA08284@phaeton.artisoft.com> Subject: Re: Switching to FreeBSD To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Sun, 30 Mar 1997 12:25:35 -0700 (MST) Cc: michaelh@cet.co.jp, irq@stepahead.net, hackers@freebsd.org In-Reply-To: <5841.859702394@time.cdrom.com> from "Jordan K. Hubbard" at Mar 29, 97 10:13:14 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > FreeBSD passwords aren't in /etc/passwd, they're in a db. You need to > > Well, to be more accurate - they're shadowed in /etc/passwd but > actually in /etc/master.passwd (from which the db is built). > > If you have DES style passwords on your Linux box, just make sure > that the DES distribution is installed on your FreeBSD box then > copy the entries into /etc/master.passwd using vipw(1). When > you're done adjusting the entries to taste (home directory, GCOS > field changes, etc), vipw will rebuild the databases automatically. Even more accurately, there is a shell script to do this conversion, including fixing the field differences. The shell script is in the -questions and -hackers list archives. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Sun Mar 30 11:43:49 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA08342 for hackers-outgoing; Sun, 30 Mar 1997 11:43:49 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA08337 for ; Sun, 30 Mar 1997 11:43:47 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id MAA08301; Sun, 30 Mar 1997 12:27:25 -0700 From: Terry Lambert Message-Id: <199703301927.MAA08301@phaeton.artisoft.com> Subject: Re: StrongARM based NC with NetBSD To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Sun, 30 Mar 1997 12:27:24 -0700 (MST) Cc: taob@risc.org, freebsd-hackers@freebsd.org In-Reply-To: <8403.859713370@time.cdrom.com> from "Jordan K. Hubbard" at Mar 30, 97 01:16:10 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Anyone interested in consulting on an ALPHA port? Chris D. is sort of > interested, but he'd prefer to work on just the generic bits of the > ALPHA port and NOT get involved in re-doing all the things in FreeBSD > that he had to do in NetBSD - he just can't contemplate doing all that > over. So it'd be nice if we could maybe get a couple of people > involved - one or two to play ALPHA portmeisters on the FreeBSD side > and maybe some of Chris's time as a consultant or something. > > And I already asked the NetBSD core group but there were no takers. :) Well, considering that Jeffrey Hsu and I both have a start on the code, and I believe we both have the hardware for the port... Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Sun Mar 30 11:50:31 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA08748 for hackers-outgoing; Sun, 30 Mar 1997 11:50:31 -0800 (PST) Received: from stepahead.net (stepahead.net [205.161.119.233]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA08741 for ; Sun, 30 Mar 1997 11:50:26 -0800 (PST) Received: from localhost (irq@localhost) by stepahead.net (8.8.5/8.8.4) with SMTP id OAA08974; Sun, 30 Mar 1997 14:49:40 -0500 Date: Sun, 30 Mar 1997 14:49:34 -0500 (EST) From: interrupt request To: Terry Lambert cc: "Jordan K. Hubbard" , michaelh@cet.co.jp, hackers@freebsd.org Subject: Re: Switching to FreeBSD In-Reply-To: <199703301925.MAA08284@phaeton.artisoft.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Where can I access these archives? ----------------------------------------- On Sun, 30 Mar 1997, Terry Lambert wrote: > > > FreeBSD passwords aren't in /etc/passwd, they're in a db. You need to > > > > Well, to be more accurate - they're shadowed in /etc/passwd but > > actually in /etc/master.passwd (from which the db is built). > > > > If you have DES style passwords on your Linux box, just make sure > > that the DES distribution is installed on your FreeBSD box then > > copy the entries into /etc/master.passwd using vipw(1). When > > you're done adjusting the entries to taste (home directory, GCOS > > field changes, etc), vipw will rebuild the databases automatically. > > Even more accurately, there is a shell script to do this conversion, > including fixing the field differences. The shell script is in the > -questions and -hackers list archives. > > > Regards, > Terry Lambert > terry@lambert.org > --- > Any opinions in this posting are my own and not those of my present > or previous employers. > From owner-freebsd-hackers Sun Mar 30 12:14:53 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA09921 for hackers-outgoing; Sun, 30 Mar 1997 12:14:53 -0800 (PST) Received: from ki1.chemie.fu-berlin.de (ki1.chemie.fu-berlin.de [160.45.24.21]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA09916 for ; Sun, 30 Mar 1997 12:14:51 -0800 (PST) Received: by ki1.chemie.fu-berlin.de (Smail3.1.28.1) from mail.hanse.de (193.174.9.9) with smtp id ; Sun, 30 Mar 97 22:14 MEST Received: from wavehh.UUCP by mail.hanse.de with UUCP for moncrg@dominus.ma.ultranet.COM id ; Sun, 30 Mar 97 22:14 MET DST Received: by wavehh.hanse.de (4.1/SMI-4.1) id AA06383; Sun, 30 Mar 97 13:48:32 +0200 Date: Sun, 30 Mar 97 13:48:32 +0200 From: cracauer@wavehh.hanse.de (Martin Cracauer) Message-Id: <9703301148.AA06383@wavehh.hanse.de> To: moncrg@dominus.ma.ultranet.COM Cc: freebsd-hackers@freebsd.org Subject: Re: C++ Code in Kernel References: <01BC3BA3.6FFD23A0@dominus.ultranet.com> Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >maybe its just that I'm taking a c++ class now, but I'm >cusious as to the pro's and con's on writing kernel source >in c++. I haven't seen nor heard on this happening on FreeBSD >[but then my travels haven't been that wide] My current list of argument against C++: - The compilers in general and g++ in special are not as mature as the C equivalents. We have enough problems with gcc itself that show up in deep system space (like floating-point - handling). - The performance characteristic of C++ in general is less predictable than for pure C. It is very easy to use constructs that end up being compiled to code with hidden extra work. And even worse in practice, it is quite easy to write wrong code when trying to avoid these hidden costs. - The performance characteristic of g++ is surprisingly often different from that of gcc for the same code. g++ disables some optimizations and aligns data often quite different than gcc does. I have done some little benchmarks, let me know if you want them. It is not that g++ is really slower than gcc for C, the disabled optimizations are not too effective for my code anyway and the other effects of g++ cause faster code in some place, too. But our kernel hackers can't reuse their compiler experience that easy. - C++ is a very complicated language and using it (read: using some real C++ features) would make it fundamentally more difficult to let new developers hack on the code. The most dramatic effect (IMHO) is that error messages in C++ are very often a quite esoteric abstraction from the real problem. - You can reuse C code in other languages as you want. This is several numbers of magnitude more difficult in C++ because of name mangling, constructors and destructors of static data, initialization of virtual function tables etc etc etc. - C++ code is more diffucult to browse, to search for definitions and declarations, to present to the user in a debugger and such. Many if not all FreeBSD developers use self-constructed browse tools and couldn't reuse them for C++. Don't misunderstand me: C++ is a nice tool for a lot of things and I sheduled it for use in some projects of mine. Especially the STL, despite some bloating problems, enables quite some code compression for me. I use languages like Common Lisp and C++ when I have to start coding before I really understand the problem I have to write a program for. That is usually the case in almost every project I get payed for, noone can expect customers to give straigt outline of a project in a domain they do their first project in, which in turn applies to 70% of my work. But when I have the strong will to do something really really right, I use C and nothing else than C. For example, when I need a collection class code for several data types, I don't want templates to expand them. I think over it where casting or macros or code duplication is applicable and use the right one in every place. And build normal libraries so that the code isn't constructed everytime I compile the program. I know I didn't do it right when compile time prohibits developing on my old Sparc 1 (no joke). And last but not leat, I can use C libraries in Common Lisp and C++. I clearly want FreeBSD's kernel code to be of the way outlined in the last paragraph. Martin P.S. When some OO features are really useful, maybe for constructing device driver or network protocol families with inheritance and runtime lookup of functionality, I would think Objectve-C is a better bet. Objective-C is just a C library for message calls and some syntactic shugar to constructs calls to messages in source code. You can even leave out all the syntactic stuff and do everything ObjC and its runtime can do from plain C code. You just link in the runtime which is a plain C library and a brave citizen inside other code of every kind. -- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Martin_Cracauer@wavehh.hanse.de http://cracauer.cons.org Fax.: +4940 5228536 "As far as I'm concerned, if something is so complicated that you can't ex- plain it in 10 seconds, then it's probably not worth knowing anyway"- Calvin From owner-freebsd-hackers Sun Mar 30 12:28:48 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA10581 for hackers-outgoing; Sun, 30 Mar 1997 12:28:48 -0800 (PST) Received: from news1.gtn.com (news1.gtn.com [194.77.0.15]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA10572 for ; Sun, 30 Mar 1997 12:28:43 -0800 (PST) Received: (from uucp@localhost) by news1.gtn.com (8.7.2/8.7.2) with UUCP id WAA21856; Sun, 30 Mar 1997 22:15:54 +0200 (MET DST) Received: (from andreas@localhost) by klemm.gtn.com (8.8.5/8.8.2) id WAA26743; Sun, 30 Mar 1997 22:10:56 +0200 (MET DST) Message-ID: <19970330221055.59085@klemm.gtn.com> Date: Sun, 30 Mar 1997 22:10:55 +0200 From: Andreas Klemm To: bsdi-probleme@genua.de Cc: hackers@FreeBSD.ORG Subject: Can I mount FreeBSD ufs fs's with BSDI and vice versa ? Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.67 X-Disclaimer: A free society is one where it is safe to be unpopular X-Operating-System: FreeBSD 3.0-CURRENT Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk [ My BSDI host ID is: 407004174, demo version ] Hi ! I had an installation problem with BSDI 3.0, see below ... but first of all my ideas, how I plan to configure my system, is that possible to do it the following way .... I'm planning to reconfigure my machine to support 4 operating systems using 3 x 2.1 GB SCSI harddisks and need some advice. I'd love to get the following configuration up and running, so that it's possible, to share my "data" filesystems on the 3rd disk between FreeBSD-current and BSDI. So I could try and test both operating systems without loosing user data or mails ... the data partition with home dirs and news/www proxy fs would be accessable from both OS's. Sounds like a dream, eh ?! ;-) Well, hope this dream comes true. ;-) Should I disable sector translation so that BSDI installation is happier ? When the DOS/NT partition C: and the extended DOS partition D: for the shared swap area lives within 1024 cylinders, then it should be nor problem or ? Or do I have to begin the BSDI partition within the 1024 cylinders as well ???? Or is there another way, to keep sector translation enabled on the SCSI host adapter and to get BSDI peacefully installed, without BSDI telling me, that it has to create 32k/4k ufs filesystems, which fails then, see below... ??? | |shared | | | DOS/Win95/NT | swap | BSDI 3.0 32 MB root | sd0 |---------------------------|-------|--------------------------| |C:\ 900MB FAT DOS/WIN/NT |NT/BSDI| 96 MB var, ~900 MB /usr | | | | | | FreeBSD-current | sd1 |--------------------------------------------------------------| | 32 MB root, 64 MB swap, 196 MB var, ~1.7 GB usr | | FreeBSD-current / BSDI 3.0 | sd2 |--------------------------------------------------------------| | 300 MB proxy-cache/news, 64 MB swap, ~1.7 GB user data | Another question ... yesterday I had an BSDI installation problem when I tried to install BSDI on sd2. During installation BSDI install program was unable to create filesystems. Here the steps up to the error message ... - custom setup - custom software setup - local cd-rom - configure sd0 - no - configure sd1 - no - configure sd2 - yes "Disksetup has detected an fdisk label on your disk.." "Use a fdisk table (required for sharing with other OS) [YES] YES" CMOS BIOS FDISK BSD/OS Heads : 0 0 255 0 sectors/track : 0 0 63 0 cylinders : 0 0 263 0 Total sectors : 0 0 4225095 42226725 - [T]able used by fdisk - change geometry [NO] "Disksetup has detected an FDISK table. You may either start with this fdisk table. If you create a new fdisk table you will loose all information on this disk. Use the current FDISK label [YES]" YES "A standard BSD partition table includes room for the root fs .... Would you like to use a standard BSD partition table ? NO" a 4.2 / 64260 ( 31.4 4.0 0 64259 b swap swap 257040 ( 125.5 16.0 64260 321299 c ---- ---- 4225095 (2063.0 263.0 0 4225094 d ---- ---- e 4.2 /bench 706860 ( 345.1 44.0 321300 1028159 f 4.2 /var 385560 ( 188.3 24.0 1028160 1413719 g 4.2 /var/news 401625 ( 196.1 25.0 1413720 1815344 h 4.2 /usr 2409750 (1176.6 150.0 1815345 4225094 - Install a new MBR ? YES - bootany.sys ? YES - Install the new BSD/OS boot blocks ? YES - "bios" "current FDISK partition of type (0xa5) will be destroyed" - ok to write new MBR ? YES "Is partition of disk C: bootable..." YES - Write out new MBR [YES] ? NO - Ok to write new disk label and BSD/OS boot blocks [YES] ? YES .... Which file systems would you like to initialize ? [all] all After that the error message ... " Custom installation failed to create a filesystem. Custom installation will try again, though it will most likely fail. You should investigate the problem and try custom installation after you have resolved the problem. With 16065 sectors per cylinder, minimum cylinders per group is 64. This requires the block size to be changed from 8192 to 32768 and the fragmented size to be changed from 1024 to 4096" The configuration I wished to setup temporarily: | | | | | DOS/Win95/NT | DOS DATA | FreeBSD-current | sd0 |---------------|---------------|------------------------------| | | /, /www, /var, /usr | | FreeBSD-current | sd1 |--------------------------------------------------------------| | 200 MB news, 1.8 GB /local | | BSDI 3.0 | sd2 |--------------------------------------------------------------| | /, /var, /usr, /local | I configured sector translation in my AHA 2940 SCSI controller because of the coexistence with DOS and the 1024 cylinder boundary. Although (if I re-think) DOS resides within the first 1024 cylinders... This leads to the following disk geometry ... Heads : 255 sectors/track : 63 cylinders : 263 sectors/cylinder: 16065 Total sectors : 4225095 BSDI install process seems to be unable to handle this accordingly. Since I have removeable SCSI storage chassis with a key that turns the drive on/off I simply installed BSDI on the 3rd disk under SCSI ID 0 and have to enable/disable the right disks now. Not very fine but ok ... The Question is ... if I re-install everything with the goal, that I share filesystems between the BSD's... Should I better disable Sector translation since BSDI doesn't grok filesystems with only a few cylinders because of sector translation ? Or is there a workaround how I can tell BSDI during installation the real Drive geometry ? But without loosing fdisk compatibility ... Since I still want to use the FreeBSD boot manager to switch between the different OS's ... Any advice would be great ! Andreas /// -- andreas@klemm.gtn.com /\/\___ Wiechers & Partner Datentechnik GmbH Andreas Klemm ___/\/\/ Support Unix -- andreas.klemm@wup.de pgp p-key http://www-swiss.ai.mit.edu/~bal/pks-toplev.html >>> powered by <<< ftp://sunsite.unc.edu/pub/Linux/system/Printing/aps-491.tgz >>> FreeBSD <<< From owner-freebsd-hackers Sun Mar 30 12:37:49 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA11135 for hackers-outgoing; Sun, 30 Mar 1997 12:37:49 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA11128 for ; Sun, 30 Mar 1997 12:37:47 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id MAA20321; Sun, 30 Mar 1997 12:36:24 -0800 (PST) To: Terry Lambert cc: taob@risc.org, freebsd-hackers@freebsd.org Subject: Re: StrongARM based NC with NetBSD In-reply-to: Your message of "Sun, 30 Mar 1997 12:27:24 MST." <199703301927.MAA08301@phaeton.artisoft.com> Date: Sun, 30 Mar 1997 12:36:23 -0800 Message-ID: <20317.859754183@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Well, considering that Jeffrey Hsu and I both have a start on the code, > and I believe we both have the hardware for the port... I asked Jeff, but he says there are more ALPHA-knowledgeable people out there than he and he'd rather they do it so he can focus on trying to get someone to subsidize his JDK 1.1 port, which sounds like what he'd much rather be doing. :) Jordan From owner-freebsd-hackers Sun Mar 30 12:41:03 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA11266 for hackers-outgoing; Sun, 30 Mar 1997 12:41:03 -0800 (PST) Received: from homeport.org (lighthouse.homeport.org [205.136.65.198]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA11259 for ; Sun, 30 Mar 1997 12:40:58 -0800 (PST) Received: (adam@localhost) by homeport.org (8.8.5/8.6.9) id PAA03573 for freebsd-hackers@FreeBSD.ORG; Sun, 30 Mar 1997 15:38:15 -0500 (EST) From: Adam Shostack Message-Id: <199703302038.PAA03573@homeport.org> Subject: Forcing IP packets through a virtual interface? To: freebsd-hackers@FreeBSD.ORG Date: Sun, 30 Mar 1997 15:38:15 -0500 (EST) X-Mailer: ELM [version 2.4ME+ PL27 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk I'm having a problem NFS mounting disks where the server is a FreeBSD (2.1.5) machine with several virtual interfaces and the client is an OpenBSD 2.0 machine. The freebsd machine has two web servers in two domains (www.homeport.org, www.chips4less.com). Having looked at a lot of tcpdumps, I believe that the problem is that when Openbsd is trying to mount from machine A, it is unwilling to accept NFS packets from machine B. Is there a way to force Freebsd to respond on the interface to which a packet was addressed? I've examined the release notes for versions between 2.1.5 and the current -RELEASE, and didn't see anything that seemed applicable. (I'm considering moving the machine to 2.2.1, to get the NFS 3 code, and see if that works any better, but don't want to move without knowing that it will/is likely to fix the problem.) Adam PS: If this is the wrong list, let me know. "General technical discussion" sounded closer than "user questions." -- "It is seldom that liberty of any kind is lost all at once." -Hume From owner-freebsd-hackers Sun Mar 30 13:03:25 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA12351 for hackers-outgoing; Sun, 30 Mar 1997 13:03:25 -0800 (PST) Received: from Ilsa.StevesCafe.com (sc-gw.StevesCafe.com [205.168.119.191]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA12339 for ; Sun, 30 Mar 1997 13:03:15 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.7.5/8.6.12) with SMTP id OAA01705 for ; Sun, 30 Mar 1997 14:03:11 -0700 (MST) Message-Id: <199703302103.OAA01705@Ilsa.StevesCafe.com> X-Authentication-Warning: Ilsa.StevesCafe.com: Host localhost [127.0.0.1] didn't use HELO protocol X-Mailer: exmh version 1.6.5 12/11/95 From: Steve Passe To: hackers@freefall.freebsd.org Subject: sendmail and 2.2.1R not happy Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sun, 30 Mar 1997 14:03:11 -0700 Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, I've just installed 2.2.1-RELEASE obtained 970328 from freefall on a P5-90, pretty much standard hardware. Uneventful, except that I cannot send mail to anyone on the local system, mailing to other systems is no problem. /var/log/maillog: Mar 29 19:19:39 Sam sendmail[540]: /etc/pwd.db: Permission denied Mar 29 19:19:39 Sam sendmail[540]: LAA04887: SYSERR(UID1): Cannot exec /usr/libexec/mail.local: Permission denied # lt `which sendmail` -r-sr-xr-x 5 root bin 274432 Mar 25 07:50 /usr/sbin/sendmail* # lt /etc/pwd.db -rw-r--r-- 1 root wheel 40960 Mar 30 03:31 /etc/pwd.db # lt /usr/libexec/mail.local -r-sr-xr-x 1 root bin 12288 Mar 25 07:42 /usr/libexec/mail.local* Anyone have any ideas??? tia, -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-hackers Sun Mar 30 13:05:34 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA12474 for hackers-outgoing; Sun, 30 Mar 1997 13:05:34 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id NAA12469 for ; Sun, 30 Mar 1997 13:05:31 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id NAA08454; Sun, 30 Mar 1997 13:49:36 -0700 From: Terry Lambert Message-Id: <199703302049.NAA08454@phaeton.artisoft.com> Subject: Re: Switching to FreeBSD To: irq@stepahead.net (interrupt request) Date: Sun, 30 Mar 1997 13:49:36 -0700 (MST) Cc: terry@lambert.org, jkh@time.cdrom.com, michaelh@cet.co.jp, hackers@freebsd.org In-Reply-To: from "interrupt request" at Mar 30, 97 02:49:34 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Where can I access these archives? > ----------------------------------------- > > Even more accurately, there is a shell script to do this conversion, > > including fixing the field differences. The shell script is in the > > -questions and -hackers list archives. www.freebsd.org. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Sun Mar 30 20:03:41 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA11862 for hackers-outgoing; Sun, 30 Mar 1997 20:03:41 -0800 (PST) Received: from thelab.hub.org (hal-ns3-01.netcom.ca [207.181.94.129]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA11845 for ; Sun, 30 Mar 1997 20:03:36 -0800 (PST) Received: from thelab.hub.org (localhost [127.0.0.1]) by thelab.hub.org (8.8.5/8.8.2) with SMTP id AAA22658 for ; Mon, 31 Mar 1997 00:03:27 -0400 (AST) Date: Mon, 31 Mar 1997 00:03:27 -0400 (AST) From: The Hermit Hacker To: hackers@freebsd.org Subject: ftruncate("directory")... Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi... Has anyone written (is it possible?) a utility that can 'truncate' a directory? Essentially, what I'm looking at is something that would open a directory, re-org the entries in it so that only the 'good' entries are at the head of the file, and then truncate it... Not sure if this is actually possible, but am going to play with it here...but if its already been done, all the better... :) Thanks... From owner-freebsd-hackers Sun Mar 30 20:08:14 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA12111 for hackers-outgoing; Sun, 30 Mar 1997 20:08:14 -0800 (PST) Received: from vinyl.quickweb.com (vinyl.quickweb.com [206.222.77.8]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA12061 for ; Sun, 30 Mar 1997 20:07:49 -0800 (PST) Received: from localhost (mark@localhost) by vinyl.quickweb.com (8.8.5/8.6.12) with SMTP id WAA29369; Sun, 30 Mar 1997 22:59:58 -0500 (EST) Date: Sun, 30 Mar 1997 22:59:56 -0500 (EST) From: Mark Mayo To: "Jordan K. Hubbard" cc: Terry Lambert , taob@risc.org, freebsd-hackers@freebsd.org Subject: Re: StrongARM based NC with NetBSD In-Reply-To: <20317.859754183@time.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Sun, 30 Mar 1997, Jordan K. Hubbard wrote: > > Well, considering that Jeffrey Hsu and I both have a start on the code, > > and I believe we both have the hardware for the port... > > I asked Jeff, but he says there are more ALPHA-knowledgeable people > out there than he and he'd rather they do it so he can focus on trying > to get someone to subsidize his JDK 1.1 port, which sounds like what > he'd much rather be doing. :) 1. What do you mean by "subsidize his JDK 1.1 port" exactly? As in find someone to pay him salary for the hours he works on it?? I would *love* to see a good JDK 1.1 port on FreeBSD (especially with the thread stuff from libc_r/3.0). Also, I believe Nate had applied for a porting license from SunSoft but never heard back.. Perhaps the JDK effort should be coordinated? 2. FreeBSD-Alpha. Is this beast going to be a reality? Is the source code tree cleaned up now to allow another hardware platform? (Sorry, I haven't been following -current.. I'll catch up) I'm a techincally incompetent individual when it comes to this type of porting (I just haven't had enough time to look at Unix/FreeBSD system internals to feel confident enough to dig into the source..), but I do have experience porting large FORTRAN apps from NEC SX3/4 supercomputers to DEC alphas runnning Digital Unix. I have an aplphastation-255 (300MHz) that is "free for beating on" - i.e. I can test code!! Finally, I need to buy a new PC, and I wouldn't mind picking up one of the shiny new 21164A-PC based alphas! For $2500 you can get a screamin' 533MHz beastie these days... Wow. With DEC now sleeping with Microsoft, and dedicated to NT, and with Microsoft promising a 64-bit NT for the Alpha with the NT 5.0 release, I'm finally confident that the Alpha may just make it into mainstream "PC" world. It would really rock to have FreeBSD on the alpha. Just my $.02. -Mark > > Jordan > ---------------------------------------------------------------------------- Mark Mayo mark@quickweb.com RingZero Comp. http://vinyl.quickweb.com/mark finger mark@quickweb.com for my PGP key and GCS code ---------------------------------------------------------------------------- "Typically, I don't use JAVA -- I think that strong typing is for weak minds (and lazy compiler/interpreter writers)." -- Terry Lambert From owner-freebsd-hackers Mon Mar 31 00:36:23 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA27242 for hackers-outgoing; Mon, 31 Mar 1997 00:36:23 -0800 (PST) Received: from nlsystems.com (nlsys.demon.co.uk [158.152.125.33]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA27213 for ; Mon, 31 Mar 1997 00:36:06 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by nlsystems.com (8.8.5/8.8.5) with SMTP id JAA27568; Mon, 31 Mar 1997 09:35:28 +0100 (BST) Date: Mon, 31 Mar 1997 09:35:28 +0100 (BST) From: Doug Rabson To: Adam Shostack cc: freebsd-hackers@FreeBSD.ORG Subject: Re: Forcing IP packets through a virtual interface? In-Reply-To: <199703302038.PAA03573@homeport.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sun, 30 Mar 1997, Adam Shostack wrote: > I'm having a problem NFS mounting disks where the server is a > FreeBSD (2.1.5) machine with several virtual interfaces and the client > is an OpenBSD 2.0 machine. The freebsd machine has two web servers in > two domains (www.homeport.org, www.chips4less.com). > > Having looked at a lot of tcpdumps, I believe that the problem > is that when Openbsd is trying to mount from machine A, it is > unwilling to accept NFS packets from machine B. > > Is there a way to force Freebsd to respond on the interface to > which a packet was addressed? I've examined the release notes for > versions between 2.1.5 and the current -RELEASE, and didn't see > anything that seemed applicable. (I'm considering moving the machine > to 2.2.1, to get the NFS 3 code, and see if that works any better, but > don't want to move without knowing that it will/is likely to fix the > problem.) This is on my list of things to fix. I would expect that a fix for this should be in the 2.2.5 release. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 181 951 1891 From owner-freebsd-hackers Mon Mar 31 00:37:31 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA27446 for hackers-outgoing; Mon, 31 Mar 1997 00:37:31 -0800 (PST) Received: from nlsystems.com (nlsys.demon.co.uk [158.152.125.33]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA27439 for ; Mon, 31 Mar 1997 00:37:26 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by nlsystems.com (8.8.5/8.8.5) with SMTP id JAA27578; Mon, 31 Mar 1997 09:36:53 +0100 (BST) Date: Mon, 31 Mar 1997 09:36:53 +0100 (BST) From: Doug Rabson To: Adam Shostack cc: freebsd-hackers@FreeBSD.ORG Subject: Re: Forcing IP packets through a virtual interface? In-Reply-To: <199703302038.PAA03573@homeport.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sun, 30 Mar 1997, Adam Shostack wrote: > I'm having a problem NFS mounting disks where the server is a > FreeBSD (2.1.5) machine with several virtual interfaces and the client > is an OpenBSD 2.0 machine. The freebsd machine has two web servers in > two domains (www.homeport.org, www.chips4less.com). > > Having looked at a lot of tcpdumps, I believe that the problem > is that when Openbsd is trying to mount from machine A, it is > unwilling to accept NFS packets from machine B. > > Is there a way to force Freebsd to respond on the interface to > which a packet was addressed? I've examined the release notes for > versions between 2.1.5 and the current -RELEASE, and didn't see > anything that seemed applicable. (I'm considering moving the machine > to 2.2.1, to get the NFS 3 code, and see if that works any better, but > don't want to move without knowing that it will/is likely to fix the > problem.) In the meantime, you might try using the -c flag to mount_nfs on the OpenBSD box. I am not sure if this will help, but it might. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 181 951 1891 From owner-freebsd-hackers Mon Mar 31 01:04:35 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA28872 for hackers-outgoing; Mon, 31 Mar 1997 01:04:35 -0800 (PST) Received: from frig.mt.cs.keio.ac.jp (frig.mt.cs.keio.ac.jp [131.113.32.7]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id BAA28853 for ; Mon, 31 Mar 1997 01:04:30 -0800 (PST) Received: (from hosokawa@localhost) by frig.mt.cs.keio.ac.jp (8.6.12+2.4W/3.4Wbeta3) id SAA13630; Mon, 31 Mar 1997 18:04:28 +0900 Date: Mon, 31 Mar 1997 18:04:28 +0900 Message-Id: <199703310904.SAA13630@frig.mt.cs.keio.ac.jp> To: hackers@freebsd.org Subject: Re: I18N Japanese/English boot.flp for 2.2.1-RELEASE In-Reply-To: Your message of "Sat, 29 Mar 1997 01:37:18 +0900". <199703281637.BAA17778@frig.mt.cs.keio.ac.jp> From: hosokawa@mt.cs.keio.ac.jp (HOSOKAWA Tatsumi) X-Mailer: mnews [version 1.20] 1996-12/08(Sun) Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I updated I18N floppy today. In article <199703281637.BAA17778@frig.mt.cs.keio.ac.jp> hosokawa@mt.cs.keio.ac.jp writes: >> ftp://jaz.jp.freebsd.org/pub/FreeBSD-jp/I18N-flp/2.2.1-RELEASE/boot.flp >> ftp://jaz.jp.freebsd.org/pub/FreeBSD-jp/I18N-flp/2.2.1-RELEASE/bootpao.flp New checksums: MD5 (boot.flp) = 6ef7a84de3ba580524a8c01a61ffb98d MD5 (bootpao.flp) = 917d384e83d7ebd849306610f0c2e211 >> o No English messages and help files are removed. If you select >> English mode, it performs just as the original boot.flp. Sorry, I removed English *.hlp files by my mistake. This version includes full English *.hlp files (and of course all Japanese *.hlp files). This version is basically based on today's RELENG_2_2 src/release. I put the source at the same directory, but currently it's totally undocumented, and I think nobody can't compile it even s/he gets this source (sorry!). ftp://jaz.jp.freebsd.org/pub/FreeBSD-jp/I18N-flp/2.2.1-RELEASE/release.I18N.PAO.src-970331.tar.gz MD5 (release.I18N.PAO.src-970331.tar.gz) = 5cca427336e33f74e22ab05a42d0d0de -- HOSOKAWA, Tatsumi E-mail: hosokawa@mt.cs.keio.ac.jp WWW homepage: http://www.mt.cs.keio.ac.jp/person/hosokawa.html Department of Computer Science, Keio University, Yokohama, Japan From owner-freebsd-hackers Mon Mar 31 01:10:25 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA29378 for hackers-outgoing; Mon, 31 Mar 1997 01:10:25 -0800 (PST) Received: from trinity.radio-do.de (trinity.Radio-do.de [193.101.164.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA29373 for ; Mon, 31 Mar 1997 01:10:16 -0800 (PST) From: Frank Nobis Message-Id: <199703310909.LAA00694@trinity.radio-do.de> Received: by trinity.radio-do.de (8.8.5/CLIENT-1.2.7-i) via EUnet id LAA00694; Mon, 31 Mar 1997 11:09:36 +0200 (CEST) Subject: SCSI subsystem died To: freebsd-hackers@freebsd.org Date: Mon, 31 Mar 1997 11:09:36 +0200 (CEST) X-Mailer: ELM [version 2.4 PL24 ME8a] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk My system died this night after upgrading with source to 2.2.1-RELEASE. Previous system was 2.2-RELEASE. The messages on the console: sd0(ahc0:0:0): SCB 0x1 - timed out in dataout phase, SCSISIGI == 0x4 SEQADDR == 0x125 sd0(ahc0:0:0): SCB 0x0 - timed out while recovery phase in progress sd0(ahc0:0:0): SCB 0x2 - timed out while recovery phase in progress sd0(ahc0:0:0): SCB 0x3 - timed out while recovery phase in progress sd0(ahc0:0:0): SCB 0x1 - timed out while recovery phase in progress sd0(ahc0:0:0): SCB 0x1 - timed out in dataout phase, SCSISIGI == 0x4 SEQADDR == 0x126 sd0(ahc0:0:0): abort message in message buffer Here is a verbose boot message from the 2.2.1 kernel. After 3 hours uptime the maschine was not reponding on the console,but yet pingable. Any sugestions, what I can do? Gru"s Frank Mar 31 00:26:17 trinity /kernel: Copyright (c) 1992-1996 FreeBSD Inc. Mar 31 00:26:17 trinity /kernel: Copyright (c) 1982, 1986, 1989, 1991, 1993 Mar 31 00:26:17 trinity /kernel: The Regents of the University of California. All rights reserved. Mar 31 00:26:17 trinity /kernel: Mar 31 00:26:17 trinity /kernel: FreeBSD 2.2.1-RELEASE #0: Mon Mar 31 02:16:34 CEST 1997 Mar 31 00:26:17 trinity /kernel: root@:/usr/src/sys/compile/TRINITY Mar 31 00:26:17 trinity /kernel: Calibrating clock(s) ... i586 clock: 132947974 Hz, i8254 clock: 1193125 Hz Mar 31 00:26:18 trinity /kernel: CPU: Pentium (132.95-MHz 586-class CPU) Mar 31 00:26:18 trinity /kernel: Origin = "GenuineIntel" Id = 0x52c Stepping=12 Mar 31 00:26:18 trinity /kernel: Features=0x1bf Mar 31 00:26:18 trinity /kernel: real memory = 67108864 (65536K bytes) Mar 31 00:26:18 trinity /kernel: avail memory = 62599168 (61132K bytes) Mar 31 00:26:18 trinity /kernel: pcibus_setup(1): mode 1 addr port (0x0cf8) is 0x8000005c Mar 31 00:26:18 trinity /kernel: pcibus_setup(1a): mode1res=0x80000000 (0x80000000) Mar 31 00:26:18 trinity /kernel: pcibus_check: device 0 is there (id=122d8086) Mar 31 00:26:18 trinity /kernel: Probing for devices on PCI bus 0: Mar 31 00:26:18 trinity /kernel: configuration mode 1 allows 32 devices. Mar 31 00:26:18 trinity /kernel: chip0 rev 2 on pci0:0 Mar 31 00:26:18 trinity /kernel: CPU Inactivity timer: clocks Mar 31 00:26:18 trinity /kernel: Peer Concurrency: enabled Mar 31 00:26:18 trinity /kernel: CPU-to-PCI Write Bursting: enabled Mar 31 00:26:18 trinity /kernel: PCI Streaming: enabled Mar 31 00:26:18 trinity /kernel: Bus Concurrency: enabled Mar 31 00:26:18 trinity /kernel: Cache: 256K pipelined-burst secondary; L1 enabled Mar 31 00:26:18 trinity /kernel: DRAM: no memory hole, 66 MHz refresh Mar 31 00:26:18 trinity /kernel: Read burst timing: x-2-2-2/x-3-3-3 Mar 31 00:26:18 trinity /kernel: Write burst timing: x-3-3-3 Mar 31 00:26:18 trinity /kernel: RAS-CAS delay: 3 clocks Mar 31 00:26:18 trinity /kernel: chip1 rev 2 on pci0:7:0 Mar 31 00:26:18 trinity /kernel: I/O Recovery Timing: 8-bit 3.5 clocks, 16-bit 3.5 clocks Mar 31 00:26:19 trinity /kernel: Extended BIOS: disabled Mar 31 00:26:19 trinity /kernel: Lower BIOS: enabled Mar 31 00:26:19 trinity /kernel: Coprocessor IRQ13: enabled Mar 31 00:26:19 trinity /kernel: Mouse IRQ12: disabled Mar 31 00:26:19 trinity /kernel: Interrupt Routing: A: IRQ11, B: IRQ12, C: IRQ9, D: disabled Mar 31 00:26:19 trinity /kernel: MB0: IRQ15, MB1: disabled Mar 31 00:26:19 trinity /kernel: chip2 rev 2 on pci0:7:1 Mar 31 00:26:19 trinity /kernel: mapreg[20] type=1 addr=0000e800 size=0010. Mar 31 00:26:19 trinity /kernel: Primary IDE: enabled Mar 31 00:26:19 trinity /kernel: Secondary IDE: enabled Mar 31 00:26:19 trinity /kernel: vga0 rev 1 int a irq 9 on pci0:10 Mar 31 00:26:19 trinity /kernel: mapreg[10] type=0 addr=fb76c000 size=4000. Mar 31 00:26:19 trinity /kernel: mapreg[14] type=0 addr=fb800000 size=800000. Mar 31 00:26:19 trinity /kernel: de0 rev 17 int a irq 12 on pci0:11 Mar 31 00:26:19 trinity /kernel: mapreg[10] type=1 addr=0000e400 size=0080. Mar 31 00:26:19 trinity /kernel: mapreg[14] type=0 addr=fb76b000 size=0080. Mar 31 00:26:19 trinity /kernel: reg16: ioaddr=0xe400 size=0x80 Mar 31 00:26:19 trinity /kernel: de0: 21041 [10Mb/s] pass 1.1 Mar 31 00:26:19 trinity /kernel: de0: address 00:00:c0:91:ca:d3 Mar 31 00:26:19 trinity /kernel: bpf: de0 attached Mar 31 00:26:19 trinity /kernel: ahc0 rev 3 int a irq 11 on pci0:12 Mar 31 00:26:19 trinity /kernel: mapreg[10] type=1 addr=0000e000 size=0100. Mar 31 00:26:20 trinity /kernel: mapreg[14] type=0 addr=fb76a000 size=1000. Mar 31 00:26:20 trinity /kernel: reg20: virtual=0xf68a0000 physical=0xfb76a000 size=0x1000 Mar 31 00:26:20 trinity /kernel: ahc0: Reading SEEPROM...done. Mar 31 00:26:20 trinity /kernel: low byte termination disabled, high byte termination enabled Mar 31 00:26:20 trinity /kernel: ahc0: aic7870 Single Channel, SCSI Id=7, 16 SCBs Mar 31 00:26:20 trinity /kernel: ahc0: Resetting Channel A Mar 31 00:26:20 trinity /kernel: ahc0: Downloading Sequencer Program...ahc0: 379 instructions downloaded Mar 31 00:26:20 trinity /kernel: Done Mar 31 00:26:20 trinity /kernel: ahc0: Probing channel A Mar 31 00:26:20 trinity /kernel: ahc0 waiting for scsi devices to settle Mar 31 00:26:20 trinity /kernel: ahc0: target 0 synchronous at 10.0MHz, offset = 0xf Mar 31 00:26:20 trinity /kernel: ahc0: target 0 Tagged Queuing Device Mar 31 00:26:20 trinity /kernel: (ahc0:0:0): "CONNER CFP2105S 2.14GB 2B4B" type 0 fixed SCSI 2 Mar 31 00:26:20 trinity /kernel: sd0(ahc0:0:0): Direct-Access 2048MB (4194304 512 byte sectors) Mar 31 00:26:20 trinity /kernel: sd0(ahc0:0:0): with 3940 cyls, 10 heads, and an average 106 sectors/track Mar 31 00:26:20 trinity /kernel: ahc0: target 1 synchronous at 10.0MHz, offset = 0xf Mar 31 00:26:20 trinity /kernel: ahc0: target 1 Tagged Queuing Device Mar 31 00:26:20 trinity /kernel: (ahc0:1:0): "QUANTUM FIREBALL_TM3200S 300N" type 0 fixed SCSI 2 Mar 31 00:26:20 trinity /kernel: sd1(ahc0:1:0): Direct-Access 3067MB (6281856 512 byte sectors) Mar 31 00:26:20 trinity /kernel: sd1(ahc0:1:0): with 6810 cyls, 5 heads, and an average 184 sectors/track Mar 31 00:26:20 trinity /kernel: (ahc0:3:0): "SANYO CRD-254S 1.05" type 5 removable SCSI 2 Mar 31 00:26:20 trinity /kernel: cd0(ahc0:3:0): CD-ROM cd present [244303 x 2048 byte records] Mar 31 00:26:21 trinity /kernel: ahc0:A:4: refuses synchronous negotiation. Using asynchronous transfers Mar 31 00:26:21 trinity /kernel: (ahc0:4:0): "TANDBERG TDC 3600 Z08:" type 1 removable SCSI 1 Mar 31 00:26:21 trinity /kernel: st0(ahc0:4:0): Sequential-Access density code 0x10, drive empty Mar 31 00:26:21 trinity /kernel: pci0: uses 8409216 bytes of memory from fb76a000 upto fbffffff. Mar 31 00:26:21 trinity /kernel: pci0: uses 400 bytes of I/O space from e000 upto e80f. Mar 31 00:26:21 trinity /kernel: Probing for devices on the ISA bus: Mar 31 00:26:21 trinity /kernel: sc0: the current keyboard controller command byte 0047 Mar 31 00:26:21 trinity /kernel: kbdio: RESET_KBD return code:00fa Mar 31 00:26:21 trinity /kernel: kbdio: RESET_KBD status:00aa Mar 31 00:26:21 trinity /kernel: sc0 at 0x60-0x6f irq 1 on motherboard Mar 31 00:26:21 trinity /kernel: sc0: VGA color <16 virtual consoles, flags=0x0> Mar 31 00:26:21 trinity /kernel: ed0 at 0x280-0x28f irq 5 maddr 0xd8000 msize 8192 on isa Mar 31 00:26:21 trinity /kernel: ed0: address 02:60:8c:9c:1f:2b, type 3c503 (8 bit) Mar 31 00:26:21 trinity /kernel: bpf: ed0 attached Mar 31 00:26:21 trinity /kernel: sio0 at 0x3f8-0x3ff irq 4 on isa Mar 31 00:26:21 trinity /kernel: sio0: type 16550A Mar 31 00:26:21 trinity /kernel: sio1 at 0x2f8-0x2ff irq 3 on isa Mar 31 00:26:21 trinity /kernel: sio1: type 16550A Mar 31 00:26:21 trinity /kernel: lpt0 at 0x378-0x37f on isa Mar 31 00:26:21 trinity /kernel: lpt1 not found at 0xffffffff Mar 31 00:26:21 trinity /kernel: fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa Mar 31 00:26:21 trinity /kernel: fdc0: NEC 72065B Mar 31 00:26:22 trinity /kernel: fd0: 1.44MB 3.5in Mar 31 00:26:22 trinity /kernel: wdc0 not found at 0x1f0 Mar 31 00:26:22 trinity /kernel: npx0 on motherboard Mar 31 00:26:22 trinity /kernel: npx0: INT 16 interface Mar 31 00:26:22 trinity /kernel: joy0 at 0x201 on isa Mar 31 00:26:22 trinity /kernel: joy0: joystick Mar 31 00:26:22 trinity /kernel: sb0 at 0x220 irq 10 drq 1 on isa Mar 31 00:26:22 trinity /kernel: sb0: Mar 31 00:26:22 trinity /kernel: sbxvi0 at 0x0 drq 5 on isa Mar 31 00:26:22 trinity /kernel: sbxvi0: Mar 31 00:26:22 trinity /kernel: sbmidi0 at 0x300 on isa Mar 31 00:26:22 trinity /kernel: Mar 31 00:26:22 trinity /kernel: opl0 at 0x388 on isa Mar 31 00:26:22 trinity /kernel: opl0: Mar 31 00:26:22 trinity /kernel: imasks: bio c0000840, tty c003001a, net c0021020 Mar 31 00:26:22 trinity /kernel: sd0s1: type 0x6, start 32, end = 204799, size 204768 : OK Mar 31 00:26:22 trinity /kernel: sd0s2: type 0xa5, start 204800, end = 4194303, size 3989504 : OK Mar 31 00:26:22 trinity /kernel: BIOS Geometries: Mar 31 00:26:22 trinity /kernel: 0:03fe3f20 0..1022=1023 cylinders, 0..63=64 heads, 1..32=32 sectors Mar 31 00:26:22 trinity /kernel: 1:03fe3f20 0..1022=1023 cylinders, 0..63=64 heads, 1..32=32 sectors Mar 31 00:26:22 trinity /kernel: 0 accounted for Mar 31 00:26:22 trinity /kernel: Device configuration finished. Mar 31 00:26:22 trinity /kernel: Considering FFS root f/s. Mar 31 00:26:23 trinity /kernel: configure() finished. Mar 31 00:26:23 trinity /kernel: bpf: lo0 attached Mar 31 00:26:23 trinity /kernel: IP packet filtering initialized, divert enabled, unlimited logging Mar 31 00:26:23 trinity /kernel: sd0s1: type 0x6, start 32, end = 204799, size 204768 : OK From owner-freebsd-hackers Mon Mar 31 01:23:27 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA00435 for hackers-outgoing; Mon, 31 Mar 1997 01:23:27 -0800 (PST) Received: from frig.mt.cs.keio.ac.jp (frig.mt.cs.keio.ac.jp [131.113.32.7]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id BAA00430 for ; Mon, 31 Mar 1997 01:23:23 -0800 (PST) Received: (from hosokawa@localhost) by frig.mt.cs.keio.ac.jp (8.6.12+2.4W/3.4Wbeta3) id SAA13956; Mon, 31 Mar 1997 18:23:21 +0900 Date: Mon, 31 Mar 1997 18:23:21 +0900 Message-Id: <199703310923.SAA13956@frig.mt.cs.keio.ac.jp> To: hackers@freebsd.org Subject: Re: I18N Japanese/English boot.flp for 2.2.1-RELEASE In-Reply-To: Your message of "Mon, 31 Mar 1997 18:04:28 +0900". <199703310904.SAA13630@frig.mt.cs.keio.ac.jp> From: hosokawa@mt.cs.keio.ac.jp (HOSOKAWA Tatsumi) X-Mailer: mnews [version 1.20] 1996-12/08(Sun) Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Sorry, these are last version's checksums. >> New checksums: >> MD5 (boot.flp) = 6ef7a84de3ba580524a8c01a61ffb98d >> MD5 (bootpao.flp) = 917d384e83d7ebd849306610f0c2e211 Correct New checksums: MD5 (boot.flp) = f79286fff51a14b589823acc98d847e2 MD5 (bootpao.flp) = e5c40690ba02bed962bda1b6c55b85e6 -- HOSOKAWA, Tatsumi E-mail: hosokawa@mt.cs.keio.ac.jp WWW homepage: http://www.mt.cs.keio.ac.jp/person/hosokawa.html Department of Computer Science, Keio University, Yokohama, Japan From owner-freebsd-hackers Mon Mar 31 01:45:58 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA02188 for hackers-outgoing; Mon, 31 Mar 1997 01:45:58 -0800 (PST) Received: (from hsu@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA02169; Mon, 31 Mar 1997 01:45:54 -0800 (PST) Date: Mon, 31 Mar 1997 01:45:54 -0800 (PST) From: Jeffrey Hsu Message-Id: <199703310945.BAA02169@freefall.freebsd.org> To: jkh Subject: Re: StrongARM based NC with NetBSD Cc: hackers Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > I asked Jeff, but he says there are more ALPHA-knowledgeable people > out there than he Actually, what I said was that we already have other people on the FreeBSD team who can do a kernel port. > and he'd rather they do it so he can focus on trying > to get someone to subsidize his JDK 1.1 port, which sounds like what > he'd much rather be doing. :) True. From owner-freebsd-hackers Mon Mar 31 01:59:38 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA03065 for hackers-outgoing; Mon, 31 Mar 1997 01:59:38 -0800 (PST) Received: from f4.hotmail.com (F4.hotmail.com [207.82.250.15]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA03060 for ; Mon, 31 Mar 1997 01:59:34 -0800 (PST) Received: (from root@localhost) by f4.hotmail.com (8.7.5/8.7.3) id BAA12465; Mon, 31 Mar 1997 01:58:48 -0800 (PST) Date: Mon, 31 Mar 1997 01:58:48 -0800 (PST) Message-Id: <199703310958.BAA12465@f4.hotmail.com> Received: from 137.229.17.253 by www.hotmail.com with HTTP; Mon, 31 Mar 1997 01:58:48 PST X-Originating-IP: [137.229.17.253] From: " steve howe" To: hackers@freebsd.org Subject: float Content-Type: text/plain Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk can anyone tell me if FreeBSD will support IEEE DOUBLE FLOATS anytime soon? i am losing alot of precision in porting software that needs that precision in mathematical computations ... float to string to float conversion only give me 6 bytes of accuracy with floats, but i get 8 bytes under DOS with Double Floats. (i know Doubles are 40 bits, but i get 64 bits of accuracy in conversion with Doubles ...) From owner-freebsd-hackers Mon Mar 31 02:50:10 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id CAA06945 for hackers-outgoing; Mon, 31 Mar 1997 02:50:10 -0800 (PST) Received: from gdi.uoregon.edu (gdi.uoregon.edu [128.223.170.30]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id CAA06909 for ; Mon, 31 Mar 1997 02:50:00 -0800 (PST) Received: from localhost (dwhite@localhost) by gdi.uoregon.edu (8.8.5/8.6.12) with SMTP id CAA00905 for ; Mon, 31 Mar 1997 02:50:00 -0800 (PST) Date: Mon, 31 Mar 1997 02:50:00 -0800 (PST) From: Doug White X-Sender: dwhite@localhost Reply-To: Doug White To: hackers@freebsd.org Subject: Telling 2.2 from 3.0 (fwd) Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk This is your guys's area... Doug White | University of Oregon Internet: dwhite@resnet.uoregon.edu | Residence Networking Assistant http://gladstone.uoregon.edu/~dwhite | Computer Science Major ---------- Forwarded message ---------- Date: Sun, 30 Mar 1997 11:07:49 -0800 (PST) From: Archie Cobbs To: freebsd-questions@freebsd.org Subject: Telling 2.2 from 3.0 What is the "correct" way to tell which version of FreeBSD you're building on using #ifdef's in your C source code? Specifically, I have the problem that a file requires #include when building under 3.0, but not when building under 2.2 (the file doesn't exist under 2.2). Thanks, -Archie ___________________________________________________________________________ Archie Cobbs * Whistle Communications, Inc. * http://www.whistle.com From owner-freebsd-hackers Mon Mar 31 03:54:30 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id DAA12323 for hackers-outgoing; Mon, 31 Mar 1997 03:54:30 -0800 (PST) Received: from minor.stranger.com (stranger.vip.best.com [204.156.129.250]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id DAA12318 for ; Mon, 31 Mar 1997 03:54:26 -0800 (PST) Received: from dog.farm.org (dog.farm.org [207.111.140.47]) by minor.stranger.com (8.6.12/8.6.12) with ESMTP id DAA26710; Mon, 31 Mar 1997 03:53:11 -0800 Received: (from dk@localhost) by dog.farm.org (8.7.5/dk#3) id DAA06808; Mon, 31 Mar 1997 03:46:39 -0800 (PST) Date: Mon, 31 Mar 1997 03:46:39 -0800 (PST) From: Dmitry Kohmanyuk Message-Id: <199703311146.DAA06808@dog.farm.org> To: adrian@obiwan.aceonline.com.au (Adrian Chadd) Cc: freebsd-hackers@freebsd.org Subject: Re: Small Disk Xterminal Newsgroups: cs-monolit.gated.lists.freebsd.hackers Organization: FARM Computing Association Reply-To: dk+@ua.net X-Newsreader: TIN [version 1.2 PL2] Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In article you wrote: > On Thu, 27 Mar 1997, Joe Greco wrote: > > I have a not-very-polished set of tools that allows me to build a floppy > > disk (1.44MB) that contains a minimal (MINIMAL) FreeBSD configuration on > > an MFS filesystem. This is potentially very handy for things like small > > routers, terminal servers, etc. It is ALSO useful to build a minimal > > Xterminal (although the Xterminal requires an -ro NFS mount from > > someplace). > > > Why? > What other programs would the straight X server need? Could you fit them > compressed onto a 1.44mb floppy? I don't think that even a compressed server itself can fit into 1.44 floppy. see: dog:/usr/X11/bin# gzip -9 < XF86_S3 | wc 3697 22451 1050627 and we still have to put kernel there... (not saying about sh, init, etc.) > I was thinking of booting off a floppy disk, and "netbooting" (kinda), by > getting what was needed, placing it into a minimal ramdisk on the machine, > and then running totally diskless. or, get some DOSish tftp client which can unzip dosboot into RAM disk using a packet driver... seems to be not so hard? It should be even possible to write/find a standalone RAM disk and tftp programs for real mode, without using that M$-written non-reentrant interrupt handler mistaken for the operating system... there are emx and djgpp to compile the code with. I already have a public domain bootsector for FAT filesystem ;-) Oh well, this only have to be a single program, using packet driver to tftp the kernel and then load it properly (using netboot as a reference implementation, but maybe doing it more cleanly). Even better, it can be used to tftp the standard secondary-stage bootblock which would then load the kernel itself... hmm. Of course, Terry's idea of modular kernel with fallback drivers using vm86 is cleaner. > Yes, RAM is so relatively cheap you wouldn't really have problems with it. > I know of xterms with 4 and 8mb RAM that run just fine when netbooted. you probably mean `hardware' xterms (running their own OS)? btw, I have NCD 88k, and it's boot image is 2.5M, including TCP networking (BSD-derived, according to copyright), local WM, and telnet, nifty setup, and NFS- or tftp-based font access. Oh well, and it even does LAT and VMS pathnames ;-) and Xremote and who-knows-what else... All that on a RISC CPU... -- What do you mean you've never been to Alpha Centauri? - Prostetnic Vogon Jeltz (THHGTTG) From owner-freebsd-hackers Mon Mar 31 04:13:12 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id EAA13501 for hackers-outgoing; Mon, 31 Mar 1997 04:13:12 -0800 (PST) Received: from news1.gtn.com (news1.gtn.com [192.109.159.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id EAA13495 for ; Mon, 31 Mar 1997 04:13:05 -0800 (PST) Received: (from uucp@localhost) by news1.gtn.com (8.7.2/8.7.2) with UUCP id OAA13540; Mon, 31 Mar 1997 14:00:49 +0200 (MET DST) Received: (from andreas@localhost) by klemm.gtn.com (8.8.5/8.8.2) id MAA02192; Mon, 31 Mar 1997 12:37:26 +0200 (MET DST) Message-ID: <19970331123726.14611@klemm.gtn.com> Date: Mon, 31 Mar 1997 12:37:26 +0200 From: Andreas Klemm To: Steve Passe Cc: hackers@freefall.freebsd.org Subject: Re: sendmail and 2.2.1R not happy References: <199703302103.OAA01705@Ilsa.StevesCafe.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.67 In-Reply-To: <199703302103.OAA01705@Ilsa.StevesCafe.com>; from Steve Passe on Sun, Mar 30, 1997 at 02:03:11PM -0700 X-Disclaimer: A free society is one where it is safe to be unpopular X-Operating-System: FreeBSD 3.0-CURRENT Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sun, Mar 30, 1997 at 02:03:11PM -0700, Steve Passe wrote: I compared the permissions of your installation with my brand new -current installation, no difference. > Mar 29 19:19:39 Sam sendmail[540]: /etc/pwd.db: Permission denied > Mar 29 19:19:39 Sam sendmail[540]: LAA04887: SYSERR(UID1): Cannot exec > /usr/libexec/mail.local: Permission denied > > # lt `which sendmail` > -r-sr-xr-x 5 root bin 274432 Mar 25 07:50 /usr/sbin/sendmail* > > # lt /etc/pwd.db > -rw-r--r-- 1 root wheel 40960 Mar 30 03:31 /etc/pwd.db > > # lt /usr/libexec/mail.local > -r-sr-xr-x 1 root bin 12288 Mar 25 07:42 /usr/libexec/mail.local* -- andreas@klemm.gtn.com /\/\___ Wiechers & Partner Datentechnik GmbH Andreas Klemm ___/\/\/ Support Unix -- andreas.klemm@wup.de pgp p-key http://www-swiss.ai.mit.edu/~bal/pks-toplev.html >>> powered by <<< ftp://sunsite.unc.edu/pub/Linux/system/Printing/aps-491.tgz >>> FreeBSD <<< From owner-freebsd-hackers Mon Mar 31 04:13:20 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id EAA13522 for hackers-outgoing; Mon, 31 Mar 1997 04:13:20 -0800 (PST) Received: from news1.gtn.com (news1.gtn.com [192.109.159.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id EAA13505 for ; Mon, 31 Mar 1997 04:13:13 -0800 (PST) Received: (from uucp@localhost) by news1.gtn.com (8.7.2/8.7.2) with UUCP id OAA13532; Mon, 31 Mar 1997 14:00:42 +0200 (MET DST) Received: (from andreas@localhost) by klemm.gtn.com (8.8.5/8.8.2) id MAA02166; Mon, 31 Mar 1997 12:34:10 +0200 (MET DST) Message-ID: <19970331123409.52993@klemm.gtn.com> Date: Mon, 31 Mar 1997 12:34:09 +0200 From: Andreas Klemm To: bsdi-probleme@genua.de Cc: hackers@FreeBSD.ORG Subject: Re: Can I mount FreeBSD ufs fs's with BSDI and vice versa ? References: <19970330221055.59085@klemm.gtn.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.67 In-Reply-To: <19970330221055.59085@klemm.gtn.com>; from Andreas Klemm on Sun, Mar 30, 1997 at 10:10:55PM +0200 X-Disclaimer: A free society is one where it is safe to be unpopular X-Operating-System: FreeBSD 3.0-CURRENT Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sun, Mar 30, 1997 at 10:10:55PM +0200, Andreas Klemm wrote: > [ My BSDI host ID is: 407004174, demo version ] Well, what I found out yesterday evening is, that you can't mount a BSDI 3.0 filesystem on a FreeBSD-current system. I think the basic problem is, that the disklabel format of both OS's are different. When running disklabel on the BSDI 3.0 disk sd2, then disklabel tells you so. Bad luck. I'll try the other way around, but I think it's also not possible ... Would it be a very big effort, to create something like a compatibility mode, that enables you to mount filesystems from other flavours of BSD ? Andreas /// -- andreas@klemm.gtn.com /\/\___ Wiechers & Partner Datentechnik GmbH Andreas Klemm ___/\/\/ Support Unix -- andreas.klemm@wup.de pgp p-key http://www-swiss.ai.mit.edu/~bal/pks-toplev.html >>> powered by <<< ftp://sunsite.unc.edu/pub/Linux/system/Printing/aps-491.tgz >>> FreeBSD <<< From owner-freebsd-hackers Mon Mar 31 04:14:12 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id EAA13578 for hackers-outgoing; Mon, 31 Mar 1997 04:14:12 -0800 (PST) Received: from news1.gtn.com (news1.gtn.com [192.109.159.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id EAA13570 for ; Mon, 31 Mar 1997 04:14:07 -0800 (PST) Received: (from uucp@localhost) by news1.gtn.com (8.7.2/8.7.2) with UUCP id OAA13611; Mon, 31 Mar 1997 14:01:08 +0200 (MET DST) Received: (from andreas@localhost) by klemm.gtn.com (8.8.5/8.8.2) id MAA02895; Mon, 31 Mar 1997 12:49:43 +0200 (MET DST) Message-ID: <19970331124943.01545@klemm.gtn.com> Date: Mon, 31 Mar 1997 12:49:43 +0200 From: Andreas Klemm To: Darren Reed Cc: hackers@FreeBSD.ORG Subject: Re: dump & restore for MS-DOS filesystems. References: <199703301454.AAA28455@plum.cyber.com.au> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.67 In-Reply-To: <199703301454.AAA28455@plum.cyber.com.au>; from Darren Reed on Mon, Mar 31, 1997 at 12:54:11AM +1000 X-Disclaimer: A free society is one where it is safe to be unpopular X-Operating-System: FreeBSD 3.0-CURRENT Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hey ! Cool work ! I'd suggest making a port of it first and add it to -current's source tree later, after many different people had a chance to try it and to add functionality that might be missing at the moment. As you say currently restore doesn't restore all informations ... I'm not sure, if that couldn't break a Win95 or NT system on C: if you really perform a full backup and restore cycle ... Did you ever proof that ? Complete backup and reestore ? That would be interesting to hear ... If it works so far I'd vote to bring it in into -current. Andreas /// On Mon, Mar 31, 1997 at 12:54:11AM +1000, Darren Reed wrote: > Well, a seemingly hard task has been completed...how "complete" it is, I > won't venture to say, except that I've hacked on dump & restore enough to > make them both work on MS-DOS format file systems. I'm pretty confident > that dump is working correctly: for each file, it dumps the entire cluster > of sectors that are allocated to it and directories are the same. restore > works...but is very naive about long names, etc. I'm not quite sure where > any translation between modes should be done, as well as the extra info. > that gets stored in the directory blocks. For now, those entries are just > ignored (even though they are on the backup `tape'). When restoring, it > uses the mode bits as they are for the MS-DOS file. I'd really like to > hear back if anyone has any useful ideas on what to do with file permission > bits. > > I've been working, compiling and testing on FreeBSD-2.1.6. > > The source is available at: > ftp://ftp.cyber.com.au/pub/unix/dump_msdos.tgz > ftp://ftp.cyber.com.au/pub/unix/restore_msdos.tgz > (and retains the BSD copyrights...) > > Oh, one last question, is it ok to add these to the -current tree, under > usr.sbin for FreeBSD ? (At this point they've really just been hacked until > they work in an `expected' manner). > > Darren > > sample output: > % ll /dos/d/SDD > total 1208 > -rwxrwxr-x 1 root wheel 334 Dec 10 11:14 file_id.diz* > -rwxrwxr-x 1 root wheel 561132 Jan 25 00:52 install.dat* > -rwxrwxr-x 1 root wheel 62122 Jan 24 22:07 install.exe* > -rwxrwxr-x 1 root wheel 5211 Dec 12 17:01 readme.txt* > -rwxrwxr-x 1 root wheel 595133 Mar 17 00:53 sdd53a-d.zip* > % ./dump_msdos 0f - /dos/d > wd1s0.dump > DUMP: Date of this level 0 dump: Sun Mar 30 23:37:19 1997 > DUMP: Date of last level 0 dump: the epoch > DUMP: Dumping /dev/rwd2s1 (/dos/d) to standard output > DUMP: mapping (Pass I) [regular files, directories] > DUMP: estimated 17359 tape blocks. > DUMP: dumping (Pass II) [directories] > DUMP: dumping (Pass III) [regular files] > DUMP: DUMP: 17408 tape blocks > DUMP: DUMP IS DONE > > % ./restore_msdos ivf ../dump_msdos/wd1s0.dump > Verify tape and initialize maps > Tape block size is 32 > Dump date: Sun Mar 30 23:37:19 1997 > Dumped from: the epoch > Level 0 dump of /dos/d on freebsd:/dev/wd2s1 > Label: NO NAME FAT16 > Extract directories from tape > Volume: [MS-DOS_6 ] > Initialize symbol table. > restore > ls > .: > 2 *./ 3 NETSCAPE/ 2629 S16DW3UP/ 489 SBBASIC/ 2810 WIN32S/ > 2811 CTCM31BB/ 1283 RECYCLED 1671 S2/ 1672 SDD/ > > restore > add SDD > Make node ./SDD > restore > > restore > extract > Extract requested files > You have not read any tapes yet. > Unless you know which volume your file(s) are on you should start > with the last volume and work towards towards the first. > Specify next volume #: 1 > extract file ./SDD/FILE_ID.DIZ > extract file ./SDD/INSTALL.EXE > extract file ./SDD/README.TXT > extract file ./SDD/INSTALL.DAT > extract file ./SDD/SDD53A-D.ZIP > Add links > Set directory mode, owner, and times. > set owner/mode for '.'? [yn] n > restore > > % ll -d SDD > d----w---- 2 darrenr cyber 512 Oct 17 1972 SDD/ > % chmod 755 SDD > % ll SDD > total 1220 > ----r----- 1 darrenr cyber 334 Sep 21 1972 FILE_ID.DIZ > ----r----- 1 darrenr cyber 561132 Oct 11 1972 INSTALL.DAT > ----r----- 1 darrenr cyber 62122 Oct 11 1972 INSTALL.EXE > ----r----- 1 darrenr cyber 5211 Sep 21 1972 README.TXT > ----r----- 1 darrenr cyber 595133 Oct 18 1972 SDD53A-D.ZIP > % chmod u+r SDD/* > % cd SDD > % foreach i (*) > foreach? cmp $i /dos/d/sdd/$i > foreach? end > % -- andreas@klemm.gtn.com /\/\___ Wiechers & Partner Datentechnik GmbH Andreas Klemm ___/\/\/ Support Unix -- andreas.klemm@wup.de pgp p-key http://www-swiss.ai.mit.edu/~bal/pks-toplev.html >>> powered by <<< ftp://sunsite.unc.edu/pub/Linux/system/Printing/aps-491.tgz >>> FreeBSD <<< From owner-freebsd-hackers Mon Mar 31 04:52:35 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id EAA15907 for hackers-outgoing; Mon, 31 Mar 1997 04:52:35 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id EAA15902 for ; Mon, 31 Mar 1997 04:52:32 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id OAA10839; Mon, 31 Mar 1997 14:51:55 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id OAA10429; Mon, 31 Mar 1997 14:21:01 +0200 (MET DST) Message-ID: <19970331142101.YD01026@uriah.heep.sax.de> Date: Mon, 31 Mar 1997 14:21:01 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: un_x@hotmail.com (steve howe) Cc: hackers@freebsd.org Subject: Re: float References: <199703310958.BAA12465@f4.hotmail.com> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199703310958.BAA12465@f4.hotmail.com>; from steve howe on Mar 31, 1997 01:58:48 -0800 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As steve howe wrote: > can anyone tell me if FreeBSD will support IEEE DOUBLE FLOATS > anytime soon? Huh? We do. We don't support the complete IEEE exception vs. non-exception (NaN, infinity etc.) handling, but the default double is 64 bits. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Mon Mar 31 04:53:25 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id EAA16010 for hackers-outgoing; Mon, 31 Mar 1997 04:53:25 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id EAA16001 for ; Mon, 31 Mar 1997 04:53:22 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id OAA10851; Mon, 31 Mar 1997 14:52:36 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id OAA10440; Mon, 31 Mar 1997 14:22:39 +0200 (MET DST) Message-ID: <19970331142239.UY50152@uriah.heep.sax.de> Date: Mon, 31 Mar 1997 14:22:39 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freebsd.org Cc: archie@whistle.com (Archie Cobbs) Subject: Re: Telling 2.2 from 3.0 (fwd) References: X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: ; from Doug White on Mar 31, 1997 02:50:00 -0800 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Doug White wrote: > What is the "correct" way to tell which version of FreeBSD you're > building on using #ifdef's in your C source code? #include #if __FreeBSD_version >= 300000 #include #endif -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Mon Mar 31 05:37:50 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA17821 for hackers-outgoing; Mon, 31 Mar 1997 05:37:50 -0800 (PST) Received: from hellcat.umd.edu (hellcat.umd.edu [129.2.70.125]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA17816 for ; Mon, 31 Mar 1997 05:37:48 -0800 (PST) Received: from carrier.eng.umd.edu (carrier.eng.umd.edu [129.2.98.188]) by hellcat.umd.edu (8.8.5/8.8.5) with ESMTP id IAA11569 for ; Mon, 31 Mar 1997 08:37:44 -0500 (EST) Received: from localhost (chuckr@localhost) by carrier.eng.umd.edu (8.8.5/8.6.4) with SMTP id IAA20431 for ; Mon, 31 Mar 1997 08:37:43 -0500 (EST) X-Authentication-Warning: carrier.eng.umd.edu: chuckr owned process doing -bs Date: Mon, 31 Mar 1997 08:37:42 -0500 (EST) From: Chuck Robey X-Sender: chuckr@carrier.eng.umd.edu To: FreeBSD-Hackers Subject: Hard disk problems Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk I was wondering if someone can explain the error message I'm suddenly seeing on bootup of one of my systems. It hangs with the message: ncr0: SCSI phase error fixup: CCB already dequeued (0xf2467200) ncr0 timeout ccb=f2466600 skip ncr0 timeout ccb=f2466800 skip ncr0 timeout ccb=f2466a00 skip ncr0 timeout ccb=f2467200 skip (hangs here) I need to know what it means, and if there's any way to recover from it without having to lose the whole disk (my boot disk, darn it!) I get this error on the system after it does the full fsck (successfully), and the system has (in total) 2 - 2G disks (this is the first one), 1 - 200 meg disk, and one cdrom, all scsi, all hooked to one ncr 825 controller (Tyan). Thanks! ----------------------------+----------------------------------------------- Chuck Robey | Interests include any kind of voice or data chuckr@eng.umd.edu | communications topic, C programming, and Unix. 9120 Edmonston Ct #302 | Greenbelt, MD 20770 | I run Journey2 and picnic, both FreeBSD (301) 220-2114 | version 3.0 current -- and great FUN! ----------------------------+----------------------------------------------- From owner-freebsd-hackers Mon Mar 31 05:42:49 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA18057 for hackers-outgoing; Mon, 31 Mar 1997 05:42:49 -0800 (PST) Received: from cheops.anu.edu.au (avalon@cheops.anu.edu.au [150.203.76.24]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA18048 for ; Mon, 31 Mar 1997 05:42:44 -0800 (PST) Message-Id: <199703311342.FAA18048@freefall.freebsd.org> Received: by cheops.anu.edu.au (1.37.109.16/16.2) id AA204705452; Mon, 31 Mar 1997 23:37:32 +1000 From: Darren Reed Subject: Re: Telling 2.2 from 3.0 (fwd) To: joerg_wunsch@uriah.heep.sax.de Date: Mon, 31 Mar 1997 23:37:32 +1000 (EST) Cc: hackers@freebsd.org, archie@whistle.com In-Reply-To: <19970331142239.UY50152@uriah.heep.sax.de> from "J Wunsch" at Mar 31, 97 02:22:39 pm X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In some mail from J Wunsch, sie said: > > As Doug White wrote: > > > What is the "correct" way to tell which version of FreeBSD you're > > building on using #ifdef's in your C source code? > > #include > > #if __FreeBSD_version >= 300000 > #include > #endif but 2.1.6 defines __FreeBSD_version as 199612 has there been a change in how this is used ? From owner-freebsd-hackers Mon Mar 31 06:09:06 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA19027 for hackers-outgoing; Mon, 31 Mar 1997 06:09:06 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id GAA19019 for ; Mon, 31 Mar 1997 06:08:57 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id QAA11766 for freebsd-hackers@freebsd.org; Mon, 31 Mar 1997 16:08:54 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id PAA10746; Mon, 31 Mar 1997 15:51:45 +0200 (MET DST) Message-ID: <19970331155144.ZF08342@uriah.heep.sax.de> Date: Mon, 31 Mar 1997 15:51:44 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: freebsd-hackers@freebsd.org (FreeBSD hackers) Subject: Anybody with an HP ethernet card? X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk The subject says it all. Is there anybody around there with an HP Ethernet card? I've got an HP J2970A, and started to write a driver for it. Now, if there's anybody who'd be willing to test it, speak up please. Most preferred are people who could also test the 100VG AnyLan features... -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Mon Mar 31 06:09:20 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA19047 for hackers-outgoing; Mon, 31 Mar 1997 06:09:20 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id GAA19040 for ; Mon, 31 Mar 1997 06:09:12 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id QAA11768 for hackers@freebsd.org; Mon, 31 Mar 1997 16:09:04 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id PAA10758; Mon, 31 Mar 1997 15:55:09 +0200 (MET DST) Message-ID: <19970331155509.SF05502@uriah.heep.sax.de> Date: Mon, 31 Mar 1997 15:55:09 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freebsd.org Subject: Re: Telling 2.2 from 3.0 (fwd) References: <19970331142239.UY50152@uriah.heep.sax.de> <199703311342.PAA11401@sax.sax.de> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199703311342.PAA11401@sax.sax.de>; from Darren Reed on Mar 31, 1997 23:37:32 +1000 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Darren Reed wrote: > > #include > > > > #if __FreeBSD_version >= 300000 > > #include > > #endif > > but 2.1.6 defines __FreeBSD_version as 199612 Yes, but that's still < 300000, so the above logic is ok. > has there been a change in how this is used ? Yep, it has been changed with 2.2R (which uses 220000). We have had too many problems with the point releases before. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Mon Mar 31 07:03:35 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA21497 for hackers-outgoing; Mon, 31 Mar 1997 07:03:35 -0800 (PST) Received: from labinfo.iet.unipi.it (labinfo.iet.unipi.it [131.114.9.5]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id HAA21476 for ; Mon, 31 Mar 1997 07:03:25 -0800 (PST) Received: from localhost (luigi@localhost) by labinfo.iet.unipi.it (8.6.5/8.6.5) id QAA21003; Mon, 31 Mar 1997 16:10:23 +0200 From: Luigi Rizzo Message-Id: <199703311410.QAA21003@labinfo.iet.unipi.it> Subject: Re: Small Disk Xterminal To: dk+@ua.net Date: Mon, 31 Mar 1997 16:10:22 +0200 (MET DST) Cc: adrian@obiwan.aceonline.com.au, freebsd-hackers@freebsd.org In-Reply-To: <199703311146.DAA06808@dog.farm.org> from "Dmitry Kohmanyuk" at Mar 31, 97 03:46:20 am X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > Why? > > What other programs would the straight X server need? Could you fit them > > compressed onto a 1.44mb floppy? > > I don't think that even a compressed server itself can fit into 1.44 floppy. > see: > dog:/usr/X11/bin# gzip -9 < XF86_S3 | wc > 3697 22451 1050627 Actually if you recompile it with only the necessary options... prova# ls -l /usr/X11R6/bin/XF* -rwsr-xr-x 1 root wheel 1445888 Mar 28 1996 /usr/X11R6/bin/XF86_S3 -rwsr-xr-x 1 root 10 2668252 Mar 14 1996 /usr/X11R6/bin/XF86_S3.21R -rwsr-xr-x 1 root wheel 1458176 Mar 23 1996 /usr/X11R6/bin/XF86_SVGA -rwsr-xr-x 1 root 10 2828389 Nov 3 1995 /usr/X11R6/bin/XF86_SVGA.21R you save a lot of memory. This is a 2.1.R system with various utilities recompiled to fit my needs (including the use of phkmalloc). Luigi -----------------------------+-------------------------------------- Luigi Rizzo | Dip. di Ingegneria dell'Informazione email: luigi@iet.unipi.it | Universita' di Pisa tel: +39-50-568533 | via Diotisalvi 2, 56126 PISA (Italy) fax: +39-50-568522 | http://www.iet.unipi.it/~luigi/ _____________________________|______________________________________ From owner-freebsd-hackers Mon Mar 31 09:51:25 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA29587 for hackers-outgoing; Mon, 31 Mar 1997 09:51:25 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id JAA29542 for ; Mon, 31 Mar 1997 09:50:43 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA16780; Mon, 31 Mar 1997 12:50:06 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Mon, 31 Mar 1997 12:50 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id GAA10773; Mon, 31 Mar 1997 06:40:23 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id GAA15593; Mon, 31 Mar 1997 06:46:02 -0500 (EST) Date: Mon, 31 Mar 1997 06:46:02 -0500 (EST) From: Thomas David Rivers Message-Id: <199703311146.GAA15593@lakes.water.net> To: ponds!radio-do.de!fn, ponds!freebsd.org!freebsd-hackers Subject: Re: SCSI subsystem died Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > My system died this night after upgrading with source to 2.2.1-RELEASE. > Previous system was 2.2-RELEASE. > > The messages on the console: > > sd0(ahc0:0:0): SCB 0x1 - timed out in dataout phase, SCSISIGI == 0x4 > SEQADDR == 0x125 > sd0(ahc0:0:0): SCB 0x0 - timed out while recovery phase in progress > sd0(ahc0:0:0): SCB 0x2 - timed out while recovery phase in progress > sd0(ahc0:0:0): SCB 0x3 - timed out while recovery phase in progress > sd0(ahc0:0:0): SCB 0x1 - timed out while recovery phase in progress > sd0(ahc0:0:0): SCB 0x1 - timed out in dataout phase, SCSISIGI == 0x4 > SEQADDR == 0x126 > sd0(ahc0:0:0): abort message in message buffer This is very similar to my problem with my aha2940 and the SCSI tape drive I've reported. I'd bet when my particular (reproducible) problem is corrected; you're will be as well... - Dave Rivers - > > Here is a verbose boot message from the 2.2.1 kernel. After 3 hours > uptime the machine was not reponding on the console,but yet pingable. > > Any sugestions, what I can do? I believe someone is working on the problem... (I think it's Justin, but I'm not sure...) - Dave Rivers - > > Gru"s > Frank > > From owner-freebsd-hackers Mon Mar 31 10:48:39 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA02657 for hackers-outgoing; Mon, 31 Mar 1997 10:48:39 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA02652 for ; Mon, 31 Mar 1997 10:48:36 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id LAA09800; Mon, 31 Mar 1997 11:32:24 -0700 From: Terry Lambert Message-Id: <199703311832.LAA09800@phaeton.artisoft.com> Subject: Re: ftruncate("directory")... To: scrappy@hub.org (The Hermit Hacker) Date: Mon, 31 Mar 1997 11:32:24 -0700 (MST) Cc: hackers@freebsd.org In-Reply-To: from "The Hermit Hacker" at Mar 31, 97 00:03:27 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Has anyone written (is it possible?) a utility that can > 'truncate' a directory? > > Essentially, what I'm looking at is something that would > open a directory, re-org the entries in it so that only the 'good' > entries are at the head of the file, and then truncate it... > > Not sure if this is actually possible, but am going to play > with it here...but if its already been done, all the better... :) Directories are operated on in terms of "directory blocks". New entries are allocated from as early in the directory as it is possible to allocate them. If there are two entries with a gap between them, and there would be enough space in the block for the new entry if the gap wasn't there, then entries are moved. This is called "compaction", and only occurs on entry creation, and only inside a block, never spanning a block boundry. When a block is fully empty (the last entry in the block has just been deleted), then if it is not the last block, it's left there. If, however, it was the last block, then all contiguous empty blocks prior to the end of the directory are truncated back. It should be possible to modify the algorithm so that intermediate blocks which are fully empty are removed from the block list, either resulting in a sparse directory file, or (by reorganizing the block order) actually a smaller file by removing the placeholders. This is not recommended, since it would invlidate NFS cookies in such a way as to cause SunOS NFSv3 clients to fail unrecoverably in some situations where the clients expect traditional UFS behaviour and fail to deal gracefully with invalidated cookies (like they are supposed to, but do not). This is a bug in Sun's NFSv3 client code, and it is not likely that you will be able to get them to fix it. If you choose the sparse directory route, the ufs_dir.c operations must be made to recognize a zero "next" pointer for the first entry in a block as an indicator that the block contents are invalid and the next block should be consulted instead. It should be noted at this point that truncation on deletion is one of the reasons FreeBSD performs so poorly on the lmbench tests: the truncation affects FS metadata which is synchronously written unless you mount -async. There is considerable controversy as to whether or not lmbench tests on ext2fs on Linux (or any other -async mounted FS which fails to make POSIX guarantees, as -async mounted FS's are wont to do) can really be considered a "figure of merit". The general consensus among people with FS engineering backgrounds is that it can not. Inter-block compaction (as opposed to intra-block compaction) must include multiple non-atomic yet idempotent operations, since it will, by definition, span multiple blocks. It is highly unlikely that you will achive any significant block recovery by doing this, and the fact that you must implement the idempotence in muc the same way that "rename" implements it, means that this would be a very high-overhead operation: not something you would want to do during standard directory operations. For the amount of fragmentation necessary to achieve utility from this recovery, you would need to have set up a scenario terribly different from standard utilization patterns. Perhaps the lmbench directory create/delete test? Finally, it is unlikely that you will be able to successfully employ a "least fit" algorithm in allocating these entries without an O(N*(N-1)) traversal of the directory entries, noting that you must make specific exception for "." and "..", which *must* be the first entries in any directory. In any case, if you do pursue this, you should establish performance baselines for the operations which will be affected by the additional overhead, and one or more common scenarios where your changes will actually be beneficial instead of detrimental. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Mon Mar 31 10:49:43 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA02713 for hackers-outgoing; Mon, 31 Mar 1997 10:49:43 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA02708 for ; Mon, 31 Mar 1997 10:49:41 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id LAA09812; Mon, 31 Mar 1997 11:33:37 -0700 From: Terry Lambert Message-Id: <199703311833.LAA09812@phaeton.artisoft.com> Subject: Re: I18N Japanese/English boot.flp for 2.2.1-RELEASE To: hosokawa@mt.cs.keio.ac.jp (HOSOKAWA Tatsumi) Date: Mon, 31 Mar 1997 11:33:37 -0700 (MST) Cc: hackers@freebsd.org In-Reply-To: <199703310923.SAA13956@frig.mt.cs.keio.ac.jp> from "HOSOKAWA Tatsumi" at Mar 31, 97 06:23:21 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Sorry, these are last version's checksums. > > >> New checksums: > >> MD5 (boot.flp) = 6ef7a84de3ba580524a8c01a61ffb98d > >> MD5 (bootpao.flp) = 917d384e83d7ebd849306610f0c2e211 > > Correct New checksums: > MD5 (boot.flp) = f79286fff51a14b589823acc98d847e2 > MD5 (bootpao.flp) = e5c40690ba02bed962bda1b6c55b85e6 Heh. And how do we know you are you? 8-). Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Mon Mar 31 10:51:58 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA02852 for hackers-outgoing; Mon, 31 Mar 1997 10:51:58 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA02845 for ; Mon, 31 Mar 1997 10:51:53 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id LAA09830; Mon, 31 Mar 1997 11:36:04 -0700 From: Terry Lambert Message-Id: <199703311836.LAA09830@phaeton.artisoft.com> Subject: Re: float To: joerg_wunsch@uriah.heep.sax.de Date: Mon, 31 Mar 1997 11:36:04 -0700 (MST) Cc: un_x@hotmail.com, hackers@FreeBSD.ORG In-Reply-To: <19970331142101.YD01026@uriah.heep.sax.de> from "J Wunsch" at Mar 31, 97 02:21:01 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > can anyone tell me if FreeBSD will support IEEE DOUBLE FLOATS > > anytime soon? > > We do. Cool! > We don't support the complete IEEE exception vs. non-exception > (NaN, infinity etc.) handling, can anyone tell me if FreeBSD will support IEEE DOUBLE FLOATS anytime soon? 8-|. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Mon Mar 31 10:57:40 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA03094 for hackers-outgoing; Mon, 31 Mar 1997 10:57:40 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA03089 for ; Mon, 31 Mar 1997 10:57:37 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id LAA09842; Mon, 31 Mar 1997 11:41:28 -0700 From: Terry Lambert Message-Id: <199703311841.LAA09842@phaeton.artisoft.com> Subject: Re: Can I mount FreeBSD ufs fs's with BSDI and vice versa ? To: andreas@klemm.gtn.com (Andreas Klemm) Date: Mon, 31 Mar 1997 11:41:27 -0700 (MST) Cc: bsdi-probleme@genua.de, hackers@FreeBSD.ORG In-Reply-To: <19970331123409.52993@klemm.gtn.com> from "Andreas Klemm" at Mar 31, 97 12:34:09 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Would it be a very big effort, to create something like a compatibility > mode, that enables you to mount filesystems from other flavours of BSD ? The FS layout policy and directory entry management policy are not very seperable from gross hierarchical block management policy in the current VFS implementations. So it would be more than a little work if you attacked it directly, and didn't want to end up with a royal kludge. If you attacked it indirectly, by seperating policy implementation, it would probably be about half the work, and then you'd have to pay the other half of the work to get your changes checked in. If you need something fast, I'd say "kludge it and label it 'KLUDGE' in big, bold letters". Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Mon Mar 31 11:10:47 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA03917 for hackers-outgoing; Mon, 31 Mar 1997 11:10:47 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA03911 for ; Mon, 31 Mar 1997 11:10:42 -0800 (PST) Received: from ntws (ntws.etinc.com [204.141.95.142]) by etinc.com (8.8.3/8.6.9) with SMTP id OAA09152; Mon, 31 Mar 1997 14:14:21 -0500 (EST) Message-Id: <3.0.32.19970331141000.00ae5cc0@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Mon, 31 Mar 1997 14:10:03 -0500 To: Jeff Dickson From: dennis Subject: re: 2.2R real slow with 8 meg Cc: hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 09:51 AM 3/31/97 -0800, you wrote: >> 2.2 is noticably slower with only 8 meg of ram....is this to be expected, or >> is there some configuration option that needs to be tweeked? I'm pretty >> much using the GENERIC settings. >> > >Hi Dennis. You need more memory! I bet the machine is swapping like >crazy. Run top and see what the percentage of swap space used is. You >should also remake your kernel so it doesn't include a whole bunch of >stuff that isn't needed. That'll save alot of memory in itself and has- >ten the boot procedure. I have 24 MB on my system (2.1) and that's >barely enough! > I'll bet I do also...but the question is WHY? A machine with basic networking, a text console and virtually no services running shouldnt pig out with 8 meg. Or is FreeBSD now suffering from Windowsitis? Dennis From owner-freebsd-hackers Mon Mar 31 11:44:06 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA05807 for hackers-outgoing; Mon, 31 Mar 1997 11:44:06 -0800 (PST) Received: from pdx1.world.net (pdx1.world.net [192.243.32.18]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA05799 for ; Mon, 31 Mar 1997 11:44:02 -0800 (PST) From: proff@suburbia.net Received: from suburbia.net (suburbia.net [203.4.184.1]) by pdx1.world.net (8.7.5/8.7.3) with SMTP id LAA11189 for ; Mon, 31 Mar 1997 11:45:56 -0800 (PST) Received: (qmail 28586 invoked by uid 110); 31 Mar 1997 19:42:52 -0000 Message-ID: <19970331194252.28585.qmail@suburbia.net> Subject: Re: improvement of pcvt In-Reply-To: <199703311909.OAA00481@shandakor.tharsis.com> from George Robbins at "Mar 31, 97 02:09:45 pm" To: grr@shandakor.tharsis.com (George Robbins) Date: Tue, 1 Apr 1997 05:42:51 +1000 (EST) Cc: romanp@mutant.mybody.ryazan.ru, twpierce@mail.bsd.uchicago.edu, tech@openbsd.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > Subject: Re: improvement of pcvt > > From: Tim Pierce > > > > pcvt has the advantage of VT220 emulation, which is important for > > those of us who connect to remote systems from the console. > > Copying the pc3 termcap entry to all of the systems that I > > regularly use is very inconvenient. Maybe it would be valuable to > > make the pccons or syscons drivers emulate a VT100, for this > > reason. > > Or *anything* that's in the nearly universal termcap/termlib files subset > or generally known. vt52 or "ansi" would be fine as long as it really > worked with the termcap entries (and other things like bbs's) out there. > > George > Just run screen(1). Cheers, Julian. From owner-freebsd-hackers Mon Mar 31 12:39:23 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA08821 for hackers-outgoing; Mon, 31 Mar 1997 12:39:23 -0800 (PST) Received: from hod (hod.tera.com [206.215.142.67]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA08813 for ; Mon, 31 Mar 1997 12:39:21 -0800 (PST) Received: from athena.tera.com (athena.tera.com [206.215.142.62]) by hod (8.7.5/8.7.3) with ESMTP id MAA14865 for ; Mon, 31 Mar 1997 12:38:45 -0800 (PST) From: Gary Kline Received: (from kline@localhost) by athena.tera.com (8.7.5/8.7.3) id MAA17991 for hackers@freebsd.org; Mon, 31 Mar 1997 12:38:43 -0800 (PST) Date: Mon, 31 Mar 1997 12:38:43 -0800 (PST) Message-Id: <199703312038.MAA17991@athena.tera.com> To: hackers@freebsd.org Subject: porting from FBSD-X11 to Windoze.... Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Can anybody tell me how much is involved in porting an X Window program --- this also uses the SoundBlaster --- to DOS or WinNN? Please respond directly since I am not subscribed to the hackers list. Thanks. gary kline From owner-freebsd-hackers Mon Mar 31 12:49:12 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA09224 for hackers-outgoing; Mon, 31 Mar 1997 12:49:12 -0800 (PST) Received: from Ilsa.StevesCafe.com (sc-gw.StevesCafe.com [205.168.119.191]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA09207 for ; Mon, 31 Mar 1997 12:48:43 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by Ilsa.StevesCafe.com (8.7.5/8.6.12) with SMTP id NAA17433; Mon, 31 Mar 1997 13:48:26 -0700 (MST) Message-Id: <199703312048.NAA17433@Ilsa.StevesCafe.com> X-Authentication-Warning: Ilsa.StevesCafe.com: Host localhost [127.0.0.1] didn't use HELO protocol X-Mailer: exmh version 1.6.5 12/11/95 From: Steve Passe To: hackers@freefall.freebsd.org cc: John Polstra , Andreas Klemm Subject: Re: sendmail and 2.2.1R not happy Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 31 Mar 1997 13:48:26 -0700 Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, I previously wrote: > I've just installed 2.2.1-RELEASE obtained 970328 from freefall > ... > Mar 29 19:19:39 Sam sendmail[540]: /etc/pwd.db: Permission denied > Mar 29 19:19:39 Sam sendmail[540]: LAA04887: SYSERR(UID1): Cannot exec > /usr/libexec/mail.local: Permission denied The problem turned out to be that the permissions for '/' had somehow been set to 750 (don't ask me how, I'm clueless)! When I started adding users to the system it became evident that I had a more serious problem than sendmail failing. So all's well that ends well, I guess. -- Steve Passe | powered by smp@csn.net | Symmetric MultiProcessor FreeBSD From owner-freebsd-hackers Mon Mar 31 12:52:03 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA09458 for hackers-outgoing; Mon, 31 Mar 1997 12:52:03 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA09417 for ; Mon, 31 Mar 1997 12:51:54 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id WAA17395 for hackers@freebsd.org; Mon, 31 Mar 1997 22:51:43 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id WAA12125; Mon, 31 Mar 1997 22:27:45 +0200 (MET DST) Message-ID: <19970331222745.RV14852@uriah.heep.sax.de> Date: Mon, 31 Mar 1997 22:27:45 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freebsd.org Subject: Re: 2.2R real slow with 8 meg References: <3.0.32.19970331141000.00ae5cc0@etinc.com> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <3.0.32.19970331141000.00ae5cc0@etinc.com>; from dennis on Mar 31, 1997 14:10:03 -0500 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As dennis wrote: > >Hi Dennis. You need more memory! > I'll bet I do also...but the question is WHY? Your experience is surprising. A friend of mine recently changed to 3.0-some-time-ago-almost-2.2, from a 2.1R system. He reported me that the system now feels faster, this is with 8 MB RAM and X11. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Mon Mar 31 15:56:26 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA20114 for hackers-outgoing; Mon, 31 Mar 1997 15:56:26 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA20100 for ; Mon, 31 Mar 1997 15:56:19 -0800 (PST) Received: from ntws (ntws.etinc.com [204.141.95.142]) by etinc.com (8.8.3/8.6.9) with SMTP id TAA10670; Mon, 31 Mar 1997 19:00:20 -0500 (EST) Message-Id: <3.0.32.19970331185559.00b6b100@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Mon, 31 Mar 1997 18:56:02 -0500 To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch), hackers@freebsd.org From: dennis Subject: Re: 2.2R real slow with 8 meg Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 10:27 PM 3/31/97 +0200, J Wunsch wrote: >As dennis wrote: > >> >Hi Dennis. You need more memory! > >> I'll bet I do also...but the question is WHY? > >Your experience is surprising. A friend of mine recently changed to >3.0-some-time-ago-almost-2.2, from a 2.1R system. He reported me that >the system now feels faster, this is with 8 MB RAM and X11. > Hmmm...is more swap required, or maybe the generic kernel has more pithy settings than before...I haven't really examined it yet? Maybe its the IDE driver? Dennis From owner-freebsd-hackers Mon Mar 31 17:17:23 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA26087 for hackers-outgoing; Mon, 31 Mar 1997 17:17:23 -0800 (PST) Received: from root.com (implode.root.com [198.145.90.17]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA26082 for ; Mon, 31 Mar 1997 17:17:19 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by root.com (8.8.5/8.6.5) with SMTP id RAA13106; Mon, 31 Mar 1997 17:18:15 -0800 (PST) Message-Id: <199704010118.RAA13106@root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: dennis cc: Jeff Dickson , hackers@freebsd.org Subject: Re: 2.2R real slow with 8 meg In-reply-to: Your message of "Mon, 31 Mar 1997 14:10:03 EST." <3.0.32.19970331141000.00ae5cc0@etinc.com> From: David Greenman Reply-To: dg@root.com Date: Mon, 31 Mar 1997 17:18:15 -0800 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >At 09:51 AM 3/31/97 -0800, you wrote: >>> 2.2 is noticably slower with only 8 meg of ram....is this to be >expected, or >>> is there some configuration option that needs to be tweeked? I'm pretty >>> much using the GENERIC settings. >>> >> >>Hi Dennis. You need more memory! I bet the machine is swapping like >>crazy. Run top and see what the percentage of swap space used is. You >>should also remake your kernel so it doesn't include a whole bunch of >>stuff that isn't needed. That'll save alot of memory in itself and has- >>ten the boot procedure. I have 24 MB on my system (2.1) and that's >>barely enough! >> >I'll bet I do also...but the question is WHY? A machine with basic >networking, a text console and virtually no services running shouldnt >pig out with 8 meg. Or is FreeBSD now suffering from Windowsitis? You probably want a custom kernel. GENERIC has all sorts of crap in it that will cause considerable bloat; that's the price of adding new device driver support. Also, NFS has about doubled in size (to REALLY HUGE), so if you don't need NFS, take it out. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-hackers Mon Mar 31 17:50:37 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA27847 for hackers-outgoing; Mon, 31 Mar 1997 17:50:37 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id RAA27842 for ; Mon, 31 Mar 1997 17:50:34 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA00630; Mon, 31 Mar 1997 20:50:02 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Mon, 31 Mar 1997 20:50 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id UAA06227 for ; Mon, 31 Mar 1997 20:38:28 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id UAA00322 for freebsd-hackers@freefall.cdrom.com; Mon, 31 Mar 1997 20:44:07 -0500 (EST) Date: Mon, 31 Mar 1997 20:44:07 -0500 (EST) From: Thomas David Rivers Message-Id: <199704010144.UAA00322@lakes.water.net> To: ponds!freefall.cdrom.com!freebsd-hackers Subject: aha2940 problems on 2.1.7.1. Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Just to let everyone know what I've tried, in an attempt to diagnose the aha2940 problems in 2.1.7.1. I increased (to 20 minutes) all of the timeout parms in the scsi_scsi_cmd() calls in st.c; thinking that perhaps the write at the end of the time was timing out too soon; causing my problems. [Recall, my problem is that I can write to a Wangtek 5150ES (QIC-150) and fill the tape up; which locks down my 2.1.7.1 system completely - this is a 2.1.6.1 system with a 2.1.7.1 kernel.] However, this didn't help the problem, when the write that fills the tape up completes I get (on the console): sd0(ahc0:0:0): SCB 0x2 - timed out in command phase, SCSISIGI == 0x84 SEQADDR == 0x42 st0(ahc0:2:0): abort message in message buffer Note that ahc0:0:0 is my primary disk drive; apparently filling the tape up caused a SCSI bus reset which didn't do much for my primary disk drive. Also notice that I didn't get any I/O about the abort having completed... everything except ping'ing the machine has "gone south" at this point. Also, after a press-the-reset-button reboot (not a complete shutdown) I got as far as starting login and xdm when suddenly, I got: sd0(ahc0:0:0) SCB 0x0 - timed out in message out phasse, SCSISIGI == 0xa4 SEQADDR == 0x99 sd0(ahc0:0:0): abort message in message buffer sd0(ahc0:0:0): SCB 0 - Abort Completed. panic: Couldn't find next SCB A turn-off-the-machine/cold reboot seems to have gotten me working again... (phew) However; this means two things: 1) 2.1.7.1's AHA2940 support has a problem. 2) My idea about needing longer timeouts in st.c has nothing to do with the problem. I really need to back stuff up before an upgrade to 2.2.1 - does anyone have any suggestions? - Dave Rivers - p.s. Here's the pertinent dmesg from the last boot; to give everyone an idea about the devices I have: FreeBSD 2.1.7.1-RELEASE #1: Mon Mar 31 19:38:48 EST 1997 rivers@lakes.water.net:/usr/src/sys-2.1.7.1/compile/LAKES CPU: 133-MHz Pentium 735\\90 or 815\\100 (Pentium-class CPU) Origin = "GenuineIntel" Id = 0x52c Stepping=12 Features=0x1bf real memory = 33554432 (32768K bytes) avail memory = 30457856 (29744K bytes) Probing for devices on PCI bus 0: chip0 rev 1 on pci0:0 chip1 rev 1 on pci0:7:0 chip2 rev 0 on pci0:7:1 ahc0 rev 0 int a irq 15 on pci0:17 ahc0: aic7880 Wide Channel, SCSI Id=7, 16 SCBs ahc0 waiting for scsi devices to settle (ahc0:0:0): "HP C3323-300 4242" type 0 fixed SCSI 2 sd0(ahc0:0:0): Direct-Access 1003MB (2056008 512 byte sectors) (ahc0:1:0): "MICROP 1548-15MZ1077802 HZ2P" type 0 fixed SCSI 1 sd1(ahc0:1:0): Direct-Access 1635MB (3349512 512 byte sectors) (ahc0:2:0): "WANGTEK 5150ES SCSI FA23 08" type 1 removable SCSI 1 st0(ahc0:2:0): Sequential-Access drive offline (ahc0:3:0): "NEC CD-ROM DRIVE:400 1.0" type 5 removable SCSI 2 cd0(ahc0:3:0): CD-ROM cd present.[217422 x 2048 byte records] From owner-freebsd-hackers Mon Mar 31 20:30:39 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA08136 for hackers-outgoing; Mon, 31 Mar 1997 20:30:39 -0800 (PST) Received: from bacall.lodgenet.com (bacall.lodgenet.com [205.138.147.242]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id UAA08121 for ; Mon, 31 Mar 1997 20:30:33 -0800 (PST) Received: (from mail@localhost) by bacall.lodgenet.com (8.6.12/8.6.12) id WAA04269 for ; Mon, 31 Mar 1997 22:20:39 -0600 Received: from garbo.lodgenet.com(204.124.123.250) by bacall via smap (V1.3) id sma004262; Mon Mar 31 22:20:17 1997 Received: from knight.trosoft.com (jprince-home.lodgenet.com [10.0.9.53]) by garbo.lodgenet.com (8.6.12/8.6.9) with ESMTP id WAA31392 for ; Mon, 31 Mar 1997 22:20:25 -0600 Received: from lodgenet.com (localhost.trosoft.com [127.0.0.1]) by knight.trosoft.com (8.8.5/8.8.5) with ESMTP id WAA00706 for ; Mon, 31 Mar 1997 22:20:59 -0600 (CST) Message-Id: <199704010420.WAA00706@knight.trosoft.com> X-Mailer: exmh version 2.0gamma 1/27/96 To: hackers@freebsd.org Subject: DNS & Sendmail Reply-To: johnp@lodgenet.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 31 Mar 1997 22:20:59 -0600 From: John Prince Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I just updated to 2.2.1-RELEASE, and upon bootup sendmail hangs.. I am running a cache only name server on a local net, ppp on-demand for external stuff... When I kill named, I can start sendmail. I also noticed I can no longer 'telnet 0' I am using an unmodified named.root file. This seemed to work fine on 2.2-BETA. Examples were taken from the "PEDANTIC PRIMER" Tutorial Any one have similar problems or suggestions?? John Prince johnp@lodgenet.com jprince@iw.net --- START OF named.boot --- named.boot directory /etc/namedb ; type domain source host/file backup file cache . named.root primary trosoft.com. named.hosts ___ END OF named.boot ___ --- START OF named.host --- named.hosts ; named.hosts ; @ IN SOA trosoft.com. root.trosoft.com. ( 96120102 360000 3600 3600000 360000 ) IN NS knight.trosoft.com. 1 IN PTR localhost.trosoft.com. ; __________________________________________________________ knight.trosoft.com. IN A 192.168.200.1 twiggy.trosoft.com. IN A 192.168.200.100 simba.trosoft.com. IN A 192.168.200.101 fred.trosoft.com. IN A 192.168.200.102 localhost.trosoft.com. IN A 127.0.0.1 $ORIGIN 200.168.192.IN-ADDR.ARPA IN NS knight.trosoft.com. 1 IN PTR knight.trosoft.com. 100 IN PTR twiggy.trosoft.com. 101 IN PTR kids.trosoft.com. $ORIGIN 0.0.127.IN-ADDR.ARPA IN NS knight.trosoft.com. 1 IN PTR localhost.trosoft.com. ___ END OF named.host ___ From owner-freebsd-hackers Mon Mar 31 21:09:44 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id VAA10154 for hackers-outgoing; Mon, 31 Mar 1997 21:09:44 -0800 (PST) Received: from vinyl.quickweb.com (vinyl.quickweb.com [206.222.77.8]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id VAA10148 for ; Mon, 31 Mar 1997 21:09:39 -0800 (PST) Received: from localhost (mark@localhost) by vinyl.quickweb.com (8.8.5/8.6.12) with SMTP id AAA03825; Tue, 1 Apr 1997 00:03:21 -0500 (EST) Date: Tue, 1 Apr 1997 00:03:21 -0500 (EST) From: Mark Mayo To: David Greenman cc: dennis , Jeff Dickson , hackers@FreeBSD.org Subject: Re: 2.2R real slow with 8 meg In-Reply-To: <199704010118.RAA13106@root.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk On Mon, 31 Mar 1997, David Greenman wrote: > >At 09:51 AM 3/31/97 -0800, you wrote: > >>> 2.2 is noticably slower with only 8 meg of ram....is this to be > >expected, or > >>> is there some configuration option that needs to be tweeked? I'm pretty > >>> much using the GENERIC settings. [SNIP] > > You probably want a custom kernel. GENERIC has all sorts of crap in it > that will cause considerable bloat; that's the price of adding new device > driver support. Also, NFS has about doubled in size (to REALLY HUGE), so > if you don't need NFS, take it out. This worked for me (taking NFS out). Now, a *crappy* 486 laptop I'm using with FreeBSD works much faster under 2.2R than it did last week under 2.1R. At least it feels like it :-) -Mark > -DG > > David Greenman > Core-team/Principal Architect, The FreeBSD Project > ---------------------------------------------------------------------------- Mark Mayo mark@quickweb.com RingZero Comp. http://vinyl.quickweb.com/mark finger mark@quickweb.com for my PGP key and GCS code ---------------------------------------------------------------------------- "Typically, I don't use JAVA -- I think that strong typing is for weak minds (and lazy compiler/interpreter writers)." -- Terry Lambert From owner-freebsd-hackers Mon Mar 31 22:23:01 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id WAA13240 for hackers-outgoing; Mon, 31 Mar 1997 22:23:01 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id WAA13220 for ; Mon, 31 Mar 1997 22:22:56 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id IAA23236 for hackers@freebsd.org; Tue, 1 Apr 1997 08:21:39 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id IAA14576; Tue, 1 Apr 1997 08:17:33 +0200 (MET DST) Message-ID: <19970401081733.BQ30598@uriah.heep.sax.de> Date: Tue, 1 Apr 1997 08:17:33 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freebsd.org Subject: Re: 2.2R real slow with 8 meg References: <3.0.32.19970331185559.00b6b100@etinc.com> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <3.0.32.19970331185559.00b6b100@etinc.com>; from dennis on Mar 31, 1997 18:56:02 -0500 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As dennis wrote: > >Your experience is surprising. A friend of mine recently changed to > >3.0-some-time-ago-almost-2.2, from a 2.1R system. He reported me that > >the system now feels faster, this is with 8 MB RAM and X11. > Hmmm...is more swap required, or maybe the generic kernel has > more pithy settings than before...I haven't really examined it yet? > > Maybe its the IDE driver? Ah, that might be the culprit. No, not really the driver, but it's quite possible that 2.2 is more aggressively paging than 2.1 did. With a PIO-only disk subsystem, it's imaginable that this results in a loss of performance. With a decent (busmaster DMA) disk subsystem, it results in a gain of performance however since there's always RAM available when needed. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Mon Mar 31 22:48:14 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id WAA14503 for hackers-outgoing; Mon, 31 Mar 1997 22:48:14 -0800 (PST) Received: from mail.crl.com (mail.crl.com [165.113.1.22]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id WAA14498 for ; Mon, 31 Mar 1997 22:48:13 -0800 (PST) From: hitman.jack@djo.com Received: from djo.com by mail.crl.com with SMTP id AA27522 (5.65c/IDA-1.5 for ); Mon, 31 Mar 1997 22:46:29 -0800 Received: from MHS by djo.com with MHS id BGCOCECI ; Mon, 31 Mar 1997 22:47:04 -0500 Date: Mon, 31 Mar 1997 22:46:36 -0500 Message-Id: Subject: Hello again To: hackers@freebsd.org Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hello again, sorry to bother you, i sent mail a few days ago with problems with the display, if you remember. Well now i have tried 3 different video cards made by different companies and none of them work. Well, kind of. My old one doesnt work at all, the display gets screwed up, one of the new ones display gets messed up too, although it is a different kind of mess up, with colored blocks that flash, probably the same problem as before. And the last one, well, i thought it worked, because i booted a 2.1.7 kernel with 3.0 on disk and it worked fine, then later when some stuff happened and i was going to install 3.0 all over again with the 3.0 kernel and it froze up when it went into the setup, no graphics weirdness, it just froze. Sorry to keep bothering you peoople with my little troubles, i'll find someone else to pester sometime soon now. I would appreciate help though. thanks. --- Sent via: Disk Jockey Online - Portland's Premier Entertainment System Send e-mail to info@djo.com for information Sent by: Hitman Jack hitman.jack@djo.com From owner-freebsd-hackers Mon Mar 31 22:57:06 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id WAA14909 for hackers-outgoing; Mon, 31 Mar 1997 22:57:06 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id WAA14886 for ; Mon, 31 Mar 1997 22:56:51 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id IAA23492; Tue, 1 Apr 1997 08:56:48 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id IAA14763; Tue, 1 Apr 1997 08:49:21 +0200 (MET DST) Message-ID: <19970401084920.JH28821@uriah.heep.sax.de> Date: Tue, 1 Apr 1997 08:49:20 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freebsd.org Cc: johnp@lodgenet.com Subject: Re: DNS & Sendmail References: <199704010420.WAA00706@knight.trosoft.com> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199704010420.WAA00706@knight.trosoft.com>; from John Prince on Mar 31, 1997 22:20:59 -0600 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As John Prince wrote: > I just updated to 2.2.1-RELEASE, and upon bootup sendmail hangs.. > I am running a cache only name server on a local net, ppp on-demand for > external stuff... > When I kill named, I can start sendmail. > > I also noticed I can no longer 'telnet 0' I've also noticed this. > --- START OF named.boot --- > named.boot > directory /etc/namedb > > ; type domain source host/file backup file > > cache . named.root > primary trosoft.com. named.hosts > ___ END OF named.boot ___ If this is really your named.boot file, it's no surprise: it misses the reverse zone (although you've been posting the zone file for it below). If it still fails, start named with tracing enabled, and examine the trace file. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Mon Mar 31 23:45:30 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA17110 for hackers-outgoing; Mon, 31 Mar 1997 23:45:30 -0800 (PST) Received: from root.com (implode.root.com [198.145.90.17]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA17105 for ; Mon, 31 Mar 1997 23:45:26 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by root.com (8.8.5/8.6.5) with SMTP id XAA15537; Mon, 31 Mar 1997 23:46:50 -0800 (PST) Message-Id: <199704010746.XAA15537@root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: hitman.jack@djo.com cc: hackers@FreeBSD.ORG Subject: Re: Hello again In-reply-to: Your message of "Mon, 31 Mar 1997 22:46:36 EST." From: David Greenman Reply-To: dg@root.com Date: Mon, 31 Mar 1997 23:46:50 -0800 Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >Hello again, sorry to bother you, i sent mail a few days ago with >problems with the display, if you remember. Well now i have tried 3 > different video cards made by different companies and none of them >work. Well, kind of. My old one doesnt work at all, the display gets >screwed up, one of the new ones display gets messed up too, although it >is a different kind of mess up, with colored blocks that flash, probably >the same problem as before. And the last one, well, i thought it >worked, because i booted a 2.1.7 kernel with 3.0 on disk and it worked >fine, then later when some stuff happened and i was going to install 3.0 >all over again with the 3.0 kernel and it froze up when it went into the >setup, no graphics weirdness, it just froze. Sorry to keep bothering >you peoople with my little troubles, i'll find someone else to pester >sometime soon now. I would appreciate help though. thanks. You may wish to check the ISA bus speed in the motherboard BIOS configuration. It sounds like it might be set too fast (>8.5MHz). -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-hackers Mon Mar 31 23:57:44 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA17577 for hackers-outgoing; Mon, 31 Mar 1997 23:57:44 -0800 (PST) Received: from ravenock.cybercity.dk (ravenock.cybercity.dk [194.16.57.32]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA17570 for ; Mon, 31 Mar 1997 23:57:37 -0800 (PST) Received: (from sos@localhost) by ravenock.cybercity.dk (8.8.5/8.7.3) id JAA03263; Tue, 1 Apr 1997 09:57:39 +0200 (MET DST) From: Søren Schmidt Message-Id: <199704010757.JAA03263@ravenock.cybercity.dk> Subject: Re: Hello again In-Reply-To: <199704010746.XAA15537@root.com> from David Greenman at "Mar 31, 97 11:46:50 pm" To: dg@root.com Date: Tue, 1 Apr 1997 09:57:39 +0200 (MET DST) Cc: hitman.jack@djo.com, hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL30 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In reply to David Greenman who wrote: > >Hello again, sorry to bother you, i sent mail a few days ago with > >problems with the display, if you remember. Well now i have tried 3 > > different video cards made by different companies and none of them > >work. Well, kind of. My old one doesnt work at all, the display gets > >screwed up, one of the new ones display gets messed up too, although it > >is a different kind of mess up, with colored blocks that flash, probably > >the same problem as before. And the last one, well, i thought it > >worked, because i booted a 2.1.7 kernel with 3.0 on disk and it worked > >fine, then later when some stuff happened and i was going to install 3.0 > >all over again with the 3.0 kernel and it froze up when it went into the > >setup, no graphics weirdness, it just froze. Sorry to keep bothering > >you peoople with my little troubles, i'll find someone else to pester > >sometime soon now. I would appreciate help though. thanks. > > You may wish to check the ISA bus speed in the motherboard BIOS > configuration. It sounds like it might be set too fast (>8.5MHz). Erhm, do you have a pentium machine ?? Try disableing the pentium optimised bcopy and friends by giving npx a 0x7 flag setting in userconfig, if that helps you have HW thats broken in that respect... -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Søren Schmidt (sos@FreeBSD.org) FreeBSD Core Team Even more code to hack -- will it ever end .. From owner-freebsd-hackers Tue Apr 1 01:19:13 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA20537 for hackers-outgoing; Tue, 1 Apr 1997 01:19:13 -0800 (PST) Received: from pdx1.world.net (pdx1.world.net [192.243.32.18]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA20532 for ; Tue, 1 Apr 1997 01:19:09 -0800 (PST) From: proff@suburbia.net Received: from suburbia.net (suburbia.net [203.4.184.1]) by pdx1.world.net (8.7.5/8.7.3) with SMTP id BAA23918 for ; Tue, 1 Apr 1997 01:21:25 -0800 (PST) Received: (qmail 1648 invoked by uid 110); 1 Apr 1997 09:18:34 -0000 Message-ID: <19970401091834.1646.qmail@suburbia.net> Subject: file system layout In-Reply-To: from Doug Rabson at "Apr 1, 97 09:56:55 am" To: dfr@nlsystems.com (Doug Rabson) Date: Tue, 1 Apr 1997 19:18:33 +1000 (EST) Cc: hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Wouldn't it be useful to have /var/local as well as /usr/local. Machine > unique stuff like ssh keys etc. can go in /var/local and shared stuff can > go in /usr/local. > > -- > Doug Rabson Mail: dfr@nlsystems.com > Nonlinear Systems Ltd. Phone: +44 181 951 1891 The whole file system layout is not effective. Each package should have it's own tree. Configuration modifications should be done using unionfs. This allows you to store installed packages on write-protected media and union over them for any changes. The same goes for the entire base system. -- Prof. Julian Assange |If you want to build a ship, don't drum up people |together to collect wood and don't assign them tasks proff@iq.org |and work, but rather teach them to long for the endless proff@gnu.ai.mit.edu |immensity of the sea. -- Antoine de Saint Exupery From owner-freebsd-hackers Tue Apr 1 01:28:33 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA20805 for hackers-outgoing; Tue, 1 Apr 1997 01:28:33 -0800 (PST) Received: from fgate.flevel.co.uk (root@fgate.flevel.co.uk [194.6.101.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA20799; Tue, 1 Apr 1997 01:28:30 -0800 (PST) Received: from localhost (dev@localhost) by fgate.flevel.co.uk (8.8.3/8.6.9) with SMTP id KAA25766; Tue, 1 Apr 1997 10:28:28 +0100 (BST) Date: Tue, 1 Apr 1997 10:28:28 +0100 (BST) From: Developer To: freebsd-current@freebsd.org, freebsd-hackers@freebsd.org Subject: PPP Desperate for help! Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk We are really stuck trying to get the PPP user program (ppp) to connect to our Perle 833 dial in server. We can connect using Windows NT fine but on BSD ppp logs in and then data will travel in both directions (As I can see the modem lights flash at both ends for both send/recieve) but no data seems to actually get through as pings do not work. The packets from the server to the user, but seem to be lost on the way back. Ive had this problem for a week now and no solution?? Thanks in advance. Trefor S. From owner-freebsd-hackers Tue Apr 1 02:47:39 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id CAA23859 for hackers-outgoing; Tue, 1 Apr 1997 02:47:39 -0800 (PST) Received: from korin.warman.org.pl (korin.warman.org.pl [148.81.160.10]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id CAA23853; Tue, 1 Apr 1997 02:47:25 -0800 (PST) Received: from localhost (abial@localhost) by korin.warman.org.pl (8.8.3/8.7.3) with SMTP id MAA27130; Tue, 1 Apr 1997 12:23:56 +0200 (MET DST) Date: Tue, 1 Apr 1997 12:23:55 +0200 (MET DST) From: Andrzej Bialecki To: Developer cc: freebsd-current@freebsd.org, freebsd-hackers@freebsd.org Subject: Re: PPP Desperate for help! In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Tue, 1 Apr 1997, Developer wrote: > > We are really stuck trying to get the PPP user program (ppp) to connect to > our Perle 833 dial in server. We can connect using Windows NT fine but on > BSD ppp logs in and then data will travel in both directions (As I can see > the modem lights flash at both ends for both send/recieve) but no data > seems to actually get through as pings do not work. The packets from the > server to the user, but seem to be lost on the way back. Let me take a guess: maybe it's a known problem with tcp_extensions set to YES in /etc/sysconfig? Some broken TCP stacks don't like it, so the solution would be to turn it off. Andy +-------------------------------------------------------------------------+ Andrzej Bialecki _) _) _)_) _)_)_) _) _) --------------------------------------- _)_) _) _) _) _)_) _)_) Research and Academic Network in Poland _) _)_) _)_)_)_) _) _) _) Bartycka 18, 00-716 Warsaw, Poland _) _) _) _) _)_)_) _) _) +-------------------------------------------------------------------------+ From owner-freebsd-hackers Tue Apr 1 03:50:41 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id DAA26717 for hackers-outgoing; Tue, 1 Apr 1997 03:50:41 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id DAA26711 for ; Tue, 1 Apr 1997 03:50:39 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA11975; Tue, 1 Apr 1997 06:50:07 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Tue, 1 Apr 1997 06:50 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id GAA18970; Tue, 1 Apr 1997 06:12:47 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id GAA01239; Tue, 1 Apr 1997 06:18:26 -0500 (EST) Date: Tue, 1 Apr 1997 06:18:26 -0500 (EST) From: Thomas David Rivers Message-Id: <199704011118.GAA01239@lakes.water.net> To: ponds!yoda.fdt.net!frankd, ponds!lakes.water.net!rivers Subject: Re: aha2940 problems on 2.1.7.1. Cc: ponds!freefall.cdrom.com!freebsd-hackers Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Frank Seltzer writes: > On Mon, 31 Mar 1997, Thomas David Rivers wrote: > > > > > Just to let everyone know what I've tried, in an attempt to diagnose > > the aha2940 problems in 2.1.7.1. > > > > I increased (to 20 minutes) all of the timeout parms in the scsi_scsi_cmd() > > calls in st.c; thinking that perhaps the write at the end of the time > > was timing out too soon; causing my problems. [Recall, my problem is that > > I can write to a Wangtek 5150ES (QIC-150) and fill the tape up; which locks > > down my 2.1.7.1 system completely - this is a 2.1.6.1 system with a 2.1.7.1 > > kernel.] > > > > However, this didn't help the problem, when the write that fills the tape > > up completes I get (on the console): > > > > sd0(ahc0:0:0): SCB 0x2 - timed out in command phase, SCSISIGI == 0x84 > > SEQADDR == 0x42 > > st0(ahc0:2:0): abort message in message buffer > > I once got this same message while doing a dump to a DAT tape. I am > running 2.2 compiled from source. > > FreeBSD 2.2-RELEASE #0: Sun Mar 23 15:52:15 EST 1997 > root@Kryten.nina.org:/usr/src/sys/compile/Kryten > > ahc0 rev 0 int a irq 11 on pci0:12 > ahc0: aic7870 Single Channel, SCSI Id=7, 16 SCBs > (ahc0:0:0): "MICROP 4421-07 0329SJ 0329" type 0 fixed SCSI 2 > sd0(ahc0:0:0): Direct-Access 2047MB (4193360 512 byte sectors) > sd0(ahc0:0:0): with 4049 cyls, 7 heads, and an average 147 sectors/track > (ahc0:3:0): "HP HP35480A T503" type 1 removable SCSI 2 > st0(ahc0:3:0): Sequential-Access density code 0x13, variable blocks, > write-enabled > > It turned out that dump was corrupted somehow. I grabbed the binary from > another machine across the network that had 2.2 installed by ftp and it > cured the problem. > > I don't know if this is your problem but it worked for me. Nope - I'm using tar. I would be surprised if your problem simply wasn't time dependent; and the version of dump you were running wasn't actually the cause. Were you running 2.2 or 2.2.1 when you saw this? - Thanks - - Dave Rivers - From owner-freebsd-hackers Tue Apr 1 04:25:45 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id EAA28014 for hackers-outgoing; Tue, 1 Apr 1997 04:25:45 -0800 (PST) Received: from veda.is (ubiq.veda.is [193.4.230.60]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id EAA28007 for ; Tue, 1 Apr 1997 04:25:40 -0800 (PST) Received: (from adam@localhost) by veda.is (8.8.4/8.7.3) id MAA28673; Tue, 1 Apr 1997 12:47:36 GMT From: Adam David Message-Id: <199704011247.MAA28673@veda.is> Subject: Re: cvs commit: ports/security/ssh/pkg PLIST In-Reply-To: <199704010947.TAA08980@ogre.dtir.qld.gov.au> from Stephen McKay at "Apr 1, 97 07:47:42 pm" To: syssgm@dtir.qld.gov.au (Stephen McKay) Date: Tue, 1 Apr 1997 12:47:35 +0000 (GMT) Cc: syssgm@dtir.qld.gov.au, pst@shockwave.com, freebsd-hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Originally, it was all quite simple with /bin /etc /lib /usr (for user home > directories), but then /usr started to bloat with /usr/bin /usr/lib and so > on. Config files and executables were thrown in /usr/lib "because there was > some space" (or some equally inane reason). > > Sun stired this all up and symlinked like crazy to try to support their > "Network is the Computer" idea, which benefited from machine independant > shared areas, and machine local writable areas, etc. This gave us /var and > an almost-good-enough-but-really-a-pain shared /usr. > > I think 4.4BSD came up with /usr/share and /usr/libexec to try to clean up > which bits of /usr should be machine independant sharable, and which bits > should contain support executables, leaving /etc for just config files. Oh, > and I left out /sbin and /usr/sbin somewhere in that -- some misguided attempt > to stop ordinary users running system administration programs. > > This is all a complete mess, and (after a good hard look) complete crap. > > What do we need? I don't really have a good answer yet. Partially because > I'm sure nobody will follow me into the desert for purification, even if > I promise a lot of milk and honey at the final destination. At a minimum, > you'd need: > > /bin - executables ... overlaid with /bin for (dynamic) executables that are not needed during minimal singleuser maintenance mode. Most static executables can be overlaid with their dynamic counterparts to save execution VM space in normal operation. > /lib - libraries (dynamic and static) > /var - local logs, run time databases, and other writable stuff > /share - architecture independant, machine independant, shared data > /etc - config files (though "etc" is a crap name) /etc = local stuff that doesn't belong anywhere else. > /src - system sources > /home - user files > > All except /var and /etc could be shared to some degree. Even some of these can be shared, but perhaps it is easier to copy the identical files than to overlay the ones that differ? > Possibly you would have a /local with all this replicated in it, to replace > our /usr/local. Possibly, each of these would have a 'local' subdirectory. /local/* makes better sense than /*/local Also some kind of distinction needs to be made between "standard", "site-local" and "host-local". Traditionally this has been done with a shared or replicated /usr and /usr/local, and host-local stuff in /local. Of course, local stuff (both sitewide and within a single host) could be overlaid on the main directory structure, but for installation purposes it would still need to be referenced in separate directories according to actual storage location. > I've never liked libexec. Put it all in /bin and don't run support progs > directly. We don't protect people from other innocent accidents. Long > names, or prefixes will be enough. What's wrong with a directory prefix, for instance /bin/exec/ ? (perhaps a better name could also be found for it) > Things that contradict what I've just written: I've always favoured a small > root partition with just enough tools to do system repair work. That doesn't > easily fit this unless we have some sort of overlay mount mechanism. So use an overlay mount :) > PS To answer your original query, if I think /usr will be shared, I use > /etc/local to store config files for /usr/local/bin progs. But some of those progs need host-specific configs (in addition to generic configs). > PPS If people think total filesystem reorganisation will get any support, > feel free to move this to -hackers. Moved to -hackers, but I'm not making any statement about the idea gaining any support. -- Adam David From owner-freebsd-hackers Tue Apr 1 04:42:56 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id EAA29237 for hackers-outgoing; Tue, 1 Apr 1997 04:42:56 -0800 (PST) Received: from fgate.flevel.co.uk (root@fgate.flevel.co.uk [194.6.101.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id EAA29232; Tue, 1 Apr 1997 04:42:52 -0800 (PST) Received: from localhost (dev@localhost) by fgate.flevel.co.uk (8.8.3/8.6.9) with SMTP id NAA27341; Tue, 1 Apr 1997 13:42:45 +0100 (BST) Date: Tue, 1 Apr 1997 13:42:45 +0100 (BST) From: Developer cc: freebsd-current@freebsd.org, freebsd-hackers@freebsd.org Subject: Re: PPP Desperate for help! In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Tue, 1 Apr 1997, Andrzej Bialecki wrote: > On Tue, 1 Apr 1997, Developer wrote: > > > > > We are really stuck trying to get the PPP user program (ppp) to connect to > > our Perle 833 dial in server. We can connect using Windows NT fine but on > > BSD ppp logs in and then data will travel in both directions (As I can see > > the modem lights flash at both ends for both send/recieve) but no data > > seems to actually get through as pings do not work. The packets from the > > server to the user, but seem to be lost on the way back. > > Let me take a guess: maybe it's a known problem with tcp_extensions set to > YES in /etc/sysconfig? Some broken TCP stacks don't like it, so the > solution would be to turn it off. > > Andy We've tried this - no difference. Any more ideas please? From owner-freebsd-hackers Tue Apr 1 05:07:57 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA00138 for hackers-outgoing; Tue, 1 Apr 1997 05:07:57 -0800 (PST) Received: from milkyway.org (lta-r-1.usit.net [205.241.194.17]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA00133; Tue, 1 Apr 1997 05:07:53 -0800 (PST) Received: (from toby@localhost) by milkyway.org (8.8.3/8.8.3) id IAA00318; Tue, 4 Mar 1997 08:09:19 -0500 (EST) Date: Tue, 4 Mar 1997 08:09:19 -0500 (EST) From: "Toby J. Swanson" Message-Id: <199703041309.IAA00318@milkyway.org> To: dev@fgate.flevel.co.uk, freebsd-current@freebsd.org, freebsd-hackers@freebsd.org Subject: Re: PPP Desperate for help! Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Greetings, I am no expert, but I have sucessfully set up several PPP connections and am willing to help. Item first - Please describe the sequence of events fully. Such as, I enter the command "ppp destination", the machine responds "Log level is 09, Using i From owner-freebsd-hackers Tue Apr 1 05:12:09 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA00386 for hackers-outgoing; Tue, 1 Apr 1997 05:12:09 -0800 (PST) Received: from gateway.cybernet.com ([192.245.33.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA00381 for ; Tue, 1 Apr 1997 05:12:06 -0800 (PST) Received: from pcglenn (pcglenn.cybernet.com [192.245.33.35]) by gateway.cybernet.com (8.7.6/8.7.3) with SMTP id IAA24601 for ; Tue, 1 Apr 1997 08:14:25 -0500 (EST) Message-ID: <3340C326.6150@cybernet.com> Date: Tue, 01 Apr 1997 09:11:18 +0100 From: Glenn Beach Reply-To: gbeach@cybernet.com Organization: Cybernet Systems Corporation X-Mailer: Mozilla 3.01Gold (WinNT; I) MIME-Version: 1.0 To: hackers@freebsd.org Subject: Internal clock Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I have been searching for an answer to the following question. Doug White at questions@freebsd.org said that you may be able to help. So here it goes: How can I make time measurements in increments smaller than the clock tick? Specifically, I am running BSD on a Pentium PC which has 128 clock ticks per second. Is there any way to improve the resolution? Thanks, -- ----------------------------------------------------------------------- Cybernet Systems -- "Amplifying Human Performance through Advanced Technology" Glenn J. Beach Research Engineer Cybernet Systems gbeach@cybernet.com 727 Airport Blvd. http://www.cybernet.com Ann Arbor, MI 48108 PHONE (313) 668-2567 FAX (313) 668-8780 ----------------------------------------------------------------------- From owner-freebsd-hackers Tue Apr 1 05:26:07 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA00987 for hackers-outgoing; Tue, 1 Apr 1997 05:26:07 -0800 (PST) Received: from milkyway.org (lta-r-1.usit.net [205.241.194.17]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA00980; Tue, 1 Apr 1997 05:26:03 -0800 (PST) Received: (from toby@localhost) by milkyway.org (8.8.3/8.8.3) id IAA00337; Tue, 4 Mar 1997 08:27:35 -0500 (EST) Date: Tue, 4 Mar 1997 08:27:35 -0500 (EST) From: "Toby J. Swanson" Message-Id: <199703041327.IAA00337@milkyway.org> To: dev@fgate.flevel.co.uk, freebsd-current@freebsd.org, freebsd-hackers@freebsd.org Subject: Re: PPP Desperate for help! Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Greetings, I am no expert, but I have set up several PPP connections and am willing to help. Item first - Please describe fully the sequence of events as you execute the ppp command. Such as, I enter "ppp destination". The machine responds with "Log level is 09, Using interface: tun0, ...". Is the dial successful? Is the login successful. Item second - Can you show me the contents of your ppp.conf and ppp.linkup files, or at least the portions dealing with this connection? Any passwords or other sensitive data should be replaced by asterisks. Item last - Have you been able to connect in interactive mode using ppp's term command? Most problems I've encountered have been typos or bad settings in configuration files. If you got a partial message from me before this one, please ignore it. Hope I can be of help. Toby From owner-freebsd-hackers Tue Apr 1 05:38:13 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA01670 for hackers-outgoing; Tue, 1 Apr 1997 05:38:13 -0800 (PST) Received: from korin.warman.org.pl (korin.warman.org.pl [148.81.160.10]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA01665 for ; Tue, 1 Apr 1997 05:38:06 -0800 (PST) Received: from localhost (abial@localhost) by korin.warman.org.pl (8.8.3/8.7.3) with SMTP id PAA29765; Tue, 1 Apr 1997 15:37:35 +0200 (MET DST) Date: Tue, 1 Apr 1997 15:37:35 +0200 (MET DST) From: Andrzej Bialecki To: Developer cc: freebsd-hackers@FreeBSD.ORG Subject: Re: PPP Desperate for help! In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Tue, 1 Apr 1997, Developer wrote: > > > On Tue, 1 Apr 1997, Andrzej Bialecki wrote: > > > On Tue, 1 Apr 1997, Developer wrote: > > > > > > > > We are really stuck trying to get the PPP user program (ppp) to connect to > > > our Perle 833 dial in server. We can connect using Windows NT fine but on > > > BSD ppp logs in and then data will travel in both directions (As I can see > > > the modem lights flash at both ends for both send/recieve) but no data > > > seems to actually get through as pings do not work. The packets from the > > > server to the user, but seem to be lost on the way back. > > > > Let me take a guess: maybe it's a known problem with tcp_extensions set to > > YES in /etc/sysconfig? Some broken TCP stacks don't like it, so the > > solution would be to turn it off. > > > > Andy > > We've tried this - no difference. Any more ideas please? Did you try to hunt for problems in the output of: * ppp with debugging on * netstat -r * tcpdump -i tun0 The modem lights may flash, but it shows only that the two modems are waving hands at each other... :-) Andy, +-------------------------------------------------------------------------+ Andrzej Bialecki _) _) _)_) _)_)_) _) _) --------------------------------------- _)_) _) _) _) _)_) _)_) Research and Academic Network in Poland _) _)_) _)_)_)_) _) _) _) Bartycka 18, 00-716 Warsaw, Poland _) _) _) _) _)_)_) _) _) +-------------------------------------------------------------------------+ From owner-freebsd-hackers Tue Apr 1 05:47:00 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA01988 for hackers-outgoing; Tue, 1 Apr 1997 05:47:00 -0800 (PST) Received: from shadows.aeon.net (bsdhack@shadows.aeon.net [194.100.41.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA01976 for ; Tue, 1 Apr 1997 05:46:55 -0800 (PST) Received: (from bsdhack@localhost) by shadows.aeon.net (8.8.5/8.8.3) id QAA20476 for hackers@freebsd.org; Tue, 1 Apr 1997 16:47:25 +0300 (EET DST) From: mika ruohotie Message-Id: <199704011347.QAA20476@shadows.aeon.net> Subject: p166 vs. p166mmx To: hackers@freebsd.org Date: Tue, 1 Apr 1997 16:47:25 +0300 (EET DST) X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk well, i was curious about this, and wanted to find out the truth... this is exactly identical set of src, though today's 2.1-STABLE since my desktop at work runs stable, and i had to do it at work, couldnt carry the cpu to home so i could've done it on 3.0-CURRENT. machine was Gigabyte GA586HX512 board (with tag chip for full caching) using 64 megs of true parity micron ram, in two 32 meg simms. it has adaptec 2940 uw, only one quantum atlas-ii fast wide. i had /usr/src and /tmp mounted async (noatime doesnt work in stable, i tried), /usr/obj is goes to /tmp. oh, and ahc options scbpaging, tagenable and memio were used. /tmp was empty in the beginning for both. both created identical log from the compiling, i diffed it and only lines that differed included echoed dates. it would've compiled without the -k on both times. i think i should mention machine was on net and i ran afterstep pre6 and several xterms while it compiled. first normal intel pentium 166mhz: -- make -k world 7595.59s user 1843.94s system 87% cpu 3:00:48.28 total make depend 25.66s user 9.29s system 79% cpu 44.132 total text data bss dec hex 798720 53248 95220 947188 e73f4 make 237.70s user 27.34s system 84% cpu 5:12.39 total 960912 Apr 1 11:55 -- then, the intel pentium 166mhz with mmx -- make -k world 5982.25s user 1636.02s system 87% cpu 2:24:21.88 total make depend 21.54s user 8.26s system 78% cpu 37.911 total text data bss dec hex 798720 53248 95220 947188 e73f4 make 188.30s user 25.57s system 89% cpu 3:58.24 total 960912 Apr 1 15:31 -- i was somewhat impressed, and surprised. mickey From owner-freebsd-hackers Tue Apr 1 05:57:15 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA02460 for hackers-outgoing; Tue, 1 Apr 1997 05:57:15 -0800 (PST) Received: from bacall.lodgenet.com (bacall.lodgenet.com [205.138.147.242]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id FAA02455 for ; Tue, 1 Apr 1997 05:57:12 -0800 (PST) Received: (from mail@localhost) by bacall.lodgenet.com (8.6.12/8.6.12) id HAA11749; Tue, 1 Apr 1997 07:56:39 -0600 Received: from garbo.lodgenet.com(204.124.123.250) by bacall via smap (V1.3) id sma011733; Tue Apr 1 07:56:29 1997 Received: from milo.lodgenet.com (milo.lodgenet.com [10.0.11.142]) by garbo.lodgenet.com (8.6.12/8.6.9) with ESMTP id HAA03923; Tue, 1 Apr 1997 07:56:38 -0600 Received: from milo.lodgenet.com (localhost [127.0.0.1]) by milo.lodgenet.com (8.8.3/8.6.12) with ESMTP id HAA24649; Tue, 1 Apr 1997 07:56:54 -0600 (CST) Message-Id: <199704011356.HAA24649@milo.lodgenet.com> X-Mailer: exmh version 2.0gamma 1/27/96 To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) cc: hackers@freebsd.org, johnp@lodgenet.com Subject: Re: DNS & Sendmail In-reply-to: Your message of "Tue, 01 Apr 1997 08:49:20 +0200." <19970401084920.JH28821@uriah.heep.sax.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 01 Apr 1997 07:56:54 -0600 From: John Prince Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk J Wunsch writes: > As John Prince wrote: > > > I just updated to 2.2.1-RELEASE, and upon bootup sendmail hangs.. > > I am running a cache only name server on a local net, ppp on-demand for > > external stuff... > > When I kill named, I can start sendmail. > > > > I also noticed I can no longer 'telnet 0' > > I've also noticed this. > > If it still fails, start named with tracing enabled, and examine the > trace file. > It looks like named is trying to resolve 0.0.0.0 externaly. Starting PPP helps.. After a valid connection closes you can "telnet 0" and sendmail begins to work. From owner-freebsd-hackers Tue Apr 1 06:21:01 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA03455 for hackers-outgoing; Tue, 1 Apr 1997 06:21:01 -0800 (PST) Received: from florence.pavilion.net (florence.pavilion.net [194.242.128.25]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA03445 for ; Tue, 1 Apr 1997 06:20:55 -0800 (PST) Received: (from joe@localhost) by florence.pavilion.net (8.8.5/8.8.5) id PAA24384; Tue, 1 Apr 1997 15:19:47 +0100 (BST) Message-ID: <19970401151947.61221@pavilion.net> Date: Tue, 1 Apr 1997 15:19:47 +0100 From: Josef Karthauser To: gbeach@cybernet.com Cc: hackers@FreeBSD.ORG Subject: Re: Internal clock References: <3340C326.6150@cybernet.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.64 In-Reply-To: <3340C326.6150@cybernet.com>; from Glenn Beach on Tue, Apr 01, 1997 at 09:11:18AM +0100 Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Tue, Apr 01, 1997 at 09:11:18AM +0100, Glenn Beach wrote: > I have been searching for an answer to the following question. Doug > White at questions@freebsd.org said that you may be able to help. So > here it goes: > > How can I make time measurements in increments smaller than the clock > tick? Specifically, I am running BSD on a Pentium PC which has 128 > clock ticks per second. Is there any way to improve the resolution? > I'm using gettimeofday which returns: struct timeval { long tv_sec; /* seconds since Jan. 1, 1970 */ long tv_usec; /* and microseconds */ }; I've no idea of the actual system resolution, but I assume (without checking the source code) that the kernel will provide the best resolution available to the system. BTW a question to the hackers... does anyone have any plans for adding a real-time scheduling class, aka threads under solaris? I'm writing some midi software and could really do with a real-time class thread to handle the timing subsystem. Joe -- Josef Karthauser Technical Manager Email: joe@pavilion.net Pavilion Internet plc. [Tel: +44 1273 607072 Fax: +44 1273 607073] From owner-freebsd-hackers Tue Apr 1 06:22:25 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA03501 for hackers-outgoing; Tue, 1 Apr 1997 06:22:25 -0800 (PST) Received: from cheops.anu.edu.au (avalon@cheops.anu.edu.au [150.203.76.24]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA03481; Tue, 1 Apr 1997 06:22:14 -0800 (PST) Message-Id: <199704011422.GAA03481@freefall.freebsd.org> Received: by cheops.anu.edu.au (1.37.109.16/16.2) id AA253474249; Wed, 2 Apr 1997 00:17:29 +1000 From: Darren Reed Subject: root logins on secure tty's ? To: hackers@freebsd.org Date: Wed, 2 Apr 1997 00:17:28 +1000 (EST) Cc: security@freebsd.org Priority: urgent X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk for some reason, in 2.2.1 source, /bin/login root logins appear to be broken on secure tty's. line 271 of login.c (or thereabouts): } else if (pwd->pw_passwd[0] == '\0') { if (rootlogin && !rootok) { /* pretend password okay */ rval = 0; goto ttycheck; } } in my ttys, I enable ttyv1 as secure, rootok == 1 and I get prompted for a password. Were the tty insecure, I suspect this would work (I have a null password for root). btw, I only noticed this because it used to work on 2.1.6 and didn't after the upgrade... Is this (perhaps) a leftover from the breakin earlier in the year ? Darren From owner-freebsd-hackers Tue Apr 1 06:28:36 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA03807 for hackers-outgoing; Tue, 1 Apr 1997 06:28:36 -0800 (PST) Received: from nevis.oss.uswest.net (nevis.oss.uswest.net [204.147.85.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA03800 for ; Tue, 1 Apr 1997 06:28:33 -0800 (PST) Received: (from greg@localhost) by nevis.oss.uswest.net (8.8.2/8.8.2) id IAA08940; Tue, 1 Apr 1997 08:28:28 -0600 (CST) From: "Greg Rowe" Message-Id: <9704010828.ZM8938@nevis.oss.uswest.net> Date: Tue, 1 Apr 1997 08:28:28 -0600 In-Reply-To: Thomas David Rivers "aha2940 problems on 2.1.7.1." (Mar 31, 8:44pm) References: <199704010144.UAA00322@lakes.water.net> X-Mailer: Z-Mail (3.2.1 10oct95) To: Thomas David Rivers Subject: Re: aha2940 problems on 2.1.7.1. Cc: hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk You're not alone with this problem.... Although I'm seeing it when Amanda first goes out and does it's planner stuff with dump(8). The tape drive is located on another system, so it's not drive related. I've also increased the timeout values. My system lockup symptoms are also the same. Justin is working on the problem. Greg On Mar 31, 8:44pm, Thomas David Rivers wrote: > Subject: aha2940 problems on 2.1.7.1. > > Just to let everyone know what I've tried, in an attempt to diagnose > the aha2940 problems in 2.1.7.1. > > I increased (to 20 minutes) all of the timeout parms in the scsi_scsi_cmd() > calls in st.c; thinking that perhaps the write at the end of the time > was timing out too soon; causing my problems. [Recall, my problem is that > I can write to a Wangtek 5150ES (QIC-150) and fill the tape up; which locks > down my 2.1.7.1 system completely - this is a 2.1.6.1 system with a 2.1.7.1 > kernel.] > > However, this didn't help the problem, when the write that fills the tape > up completes I get (on the console): > > sd0(ahc0:0:0): SCB 0x2 - timed out in command phase, SCSISIGI == 0x84 > SEQADDR == 0x42 > st0(ahc0:2:0): abort message in message buffer > > Note that ahc0:0:0 is my primary disk drive; apparently filling the > tape up caused a SCSI bus reset which didn't do much for my primary disk > drive. Also notice that I didn't get any I/O about the abort having > completed... everything except ping'ing the machine has "gone south" > at this point. > > Also, after a press-the-reset-button reboot (not a complete shutdown) > I got as far as starting login and xdm when suddenly, I got: > > > sd0(ahc0:0:0) SCB 0x0 - timed out in message out phasse, SCSISIGI == 0xa4 > SEQADDR == 0x99 > sd0(ahc0:0:0): abort message in message buffer > sd0(ahc0:0:0): SCB 0 - Abort Completed. > panic: Couldn't find next SCB > > A turn-off-the-machine/cold reboot seems to have gotten me working > again... (phew) > > However; this means two things: > > 1) 2.1.7.1's AHA2940 support has a problem. > > 2) My idea about needing longer timeouts in st.c has nothing > to do with the problem. > > > I really need to back stuff up before an upgrade to 2.2.1 - does anyone > have any suggestions? > > > - Dave Rivers - > > p.s. Here's the pertinent dmesg from the last boot; to give everyone > an idea about the devices I have: > > FreeBSD 2.1.7.1-RELEASE #1: Mon Mar 31 19:38:48 EST 1997 > rivers@lakes.water.net:/usr/src/sys-2.1.7.1/compile/LAKES > CPU: 133-MHz Pentium 735\\90 or 815\\100 (Pentium-class CPU) > Origin = "GenuineIntel" Id = 0x52c Stepping=12 > Features=0x1bf > real memory = 33554432 (32768K bytes) > avail memory = 30457856 (29744K bytes) > Probing for devices on PCI bus 0: > chip0 rev 1 on pci0:0 > chip1 rev 1 on pci0:7:0 > chip2 rev 0 on pci0:7:1 > ahc0 rev 0 int a irq 15 on pci0:17 > ahc0: aic7880 Wide Channel, SCSI Id=7, 16 SCBs > ahc0 waiting for scsi devices to settle > (ahc0:0:0): "HP C3323-300 4242" type 0 fixed SCSI 2 > sd0(ahc0:0:0): Direct-Access 1003MB (2056008 512 byte sectors) > (ahc0:1:0): "MICROP 1548-15MZ1077802 HZ2P" type 0 fixed SCSI 1 > sd1(ahc0:1:0): Direct-Access 1635MB (3349512 512 byte sectors) > (ahc0:2:0): "WANGTEK 5150ES SCSI FA23 08" type 1 removable SCSI 1 > st0(ahc0:2:0): Sequential-Access drive offline > (ahc0:3:0): "NEC CD-ROM DRIVE:400 1.0" type 5 removable SCSI 2 > cd0(ahc0:3:0): CD-ROM cd present.[217422 x 2048 byte records] >-- End of excerpt from Thomas David Rivers -- Greg Rowe | U S West - Interact Services | INTERNET greg@uswest.net 111 Washington Ave. South | Fax: (612) 672-8537 Minneapolis, MN USA 55401 | Voice: (612) 672-8535 Never trust an operating system you don't have source for.... From owner-freebsd-hackers Tue Apr 1 06:28:56 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA03839 for hackers-outgoing; Tue, 1 Apr 1997 06:28:56 -0800 (PST) Received: from tyger.inna.net (root@tyger.inna.net [206.151.66.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA03834 for ; Tue, 1 Apr 1997 06:28:53 -0800 (PST) Received: from dolphin.inna.net (jamie@dolphin.inna.net [206.151.66.2]) by tyger.inna.net (8.8.3/8.7.3) with SMTP id JAA03660 for ; Tue, 1 Apr 1997 09:29:27 -0500 (EST) Date: Tue, 1 Apr 1997 09:35:45 -0500 (EST) From: Jamie Bowden To: hackers@FreeBSD.org Subject: Error Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk Hi all. My web server started complaining with the following errors last night, and I was wondering if someone could give me a clue on what I need to play with to make it stop. /kernel: file table is full /kernel: mb_map is full I know what the file table full error is, but the mb_map is a complete mystery. Jamie Bowden Network Administrator, TBI Ltd. From owner-freebsd-hackers Tue Apr 1 06:35:38 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA04256 for hackers-outgoing; Tue, 1 Apr 1997 06:35:38 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA04251; Tue, 1 Apr 1997 06:35:32 -0800 (PST) Received: from dbws.etinc.com (dbws.etinc.com [204.141.95.130]) by etinc.com (8.8.3/8.6.9) with SMTP id JAA15234; Tue, 1 Apr 1997 09:39:40 -0500 (EST) Message-Id: <3.0.32.19970401095010.006a2504@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Tue, 01 Apr 1997 09:50:12 -0500 To: Developer From: dennis Subject: Re: PPP Desperate for help! Cc: freebsd-current@FreeBSD.ORG, freebsd-hackers@FreeBSD.ORG Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk At 01:42 PM 4/1/97 +0100, Developer wrote: > > >On Tue, 1 Apr 1997, Andrzej Bialecki wrote: > >> On Tue, 1 Apr 1997, Developer wrote: >> >> > >> > We are really stuck trying to get the PPP user program (ppp) to connect to >> > our Perle 833 dial in server. We can connect using Windows NT fine but on >> > BSD ppp logs in and then data will travel in both directions (As I can see >> > the modem lights flash at both ends for both send/recieve) but no data >> > seems to actually get through as pings do not work. The packets from the >> > server to the user, but seem to be lost on the way back. >> >> Let me take a guess: maybe it's a known problem with tcp_extensions set to >> YES in /etc/sysconfig? Some broken TCP stacks don't like it, so the >> solution would be to turn it off. >> >> Andy > >We've tried this - no difference. Any more ideas please? Why not post a trace of the ppp negotiations? Perhaps there is a parameter conflict....are you getting to state IPCP_OPENED? If not, you cant pass traffic. Dennis From owner-freebsd-hackers Tue Apr 1 06:40:48 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA04469 for hackers-outgoing; Tue, 1 Apr 1997 06:40:48 -0800 (PST) Received: from life.eecs.umich.edu (pmchen@life.eecs.umich.edu [141.213.8.32]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA04464 for ; Tue, 1 Apr 1997 06:40:44 -0800 (PST) Received: (from pmchen@localhost) by life.eecs.umich.edu (8.8.5/8.8.0) id JAA26225 for freebsd-hackers@freebsd.org; Tue, 1 Apr 1997 09:39:42 -0500 (EST) Date: Tue, 1 Apr 1997 09:39:42 -0500 (EST) From: "Peter M. Chen" Message-Id: <199704011439.JAA26225@life.eecs.umich.edu> To: freebsd-hackers@freebsd.org Subject: question on buffer cache and VMIO Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, I'm starting to hack on FreeBSD and had some questions. Background: My goal is to make the file cache reliable without writing to disk under normal operation (see Rio paper in ASPLOS 1996). We protect against power loss by using a battery; we protect against kernel crashes by write-protecting the file cache. The main benefit is you get reliability equivalent to mounting the file system sync, yet with the performance of async (actually even better, since FreeBSD's async option still does substantial amounts of I/O). This works really well for mmap'ed files, which makes it possible to have VERY fast transactions. It should also extend the battery life for portables. See http://www.eecs.umich.edu/~pmchen/ for more info. Questions re: buffer cache and VMIO 1) What's the relationship between the buffer cache and VM cache (I'm not sure this is the proper terminology)? The buffer cache seems to hold file data and metadata, while the VM cache holds only file data. Do buffer headers point to the VM cache data? Can the VM cache hold file data that is not in the buffer cache? 2) The buffer cache seems small relative to the physical memory. E.g. on a 64 MB machine (51 MB available), maxbufspace defaults to only 6.3 MB. For working sets larger than this, there appears to be significant overhead in frequently moving data between the VM cache and buffer cache. Would it make sense to set NBUF larger (e.g. enough to have the buffer cache fill memory)? 3) Dirty data gets written to disk when it leaves the buffer cache, even if it is also in the VM cache. This makes sense normally (since Unix traditionally bawrote data to disk as soon as it filled a block), but this prevents my keeping lots of dirty file data around. 4) What happens to mmap'ed data? Does it reside in the VM cache? Are there buffer headers for mmap'ed data? 5) I came across a strange phenomenon when trying to get rid of all disk writes in ufs_remove. Even if i_nlink goes to 0, the file is still fsync'ed. The call graph is: ufs_remove -> vput -> vrele -> vm_object_deallocate -> vm_object_terminate -> vinvalbuf (with V_SAVE) -> VOP_FSYNC. The system fails when I have vm_object_terminate check for i_nlink and call vinvalbuf without V_SAVE. Can someone explain why a deleted file needs to get fsynced (note that this isn't the directory, but the actual file)? General kernel questions: 1) I'd like the ability to read and write kernel global variables (without going to ddb). I tried kvm, but that only works for variables in i386/i386/symbols.raw. kgdb only works for off-line core dumps. I finally used nm to get the symbol address and directly read and wrote to /dev/kmem. This works fine, but I was wondering if there's a better solution that exists already. By the way, I've been SUPER impressed with FreeBSD. The code is well-written, especially compared to other operating systems I've worked on; Compiling the kernel is VERY fast; the system is fast and small; the installation process is easy; the small number of packages and ports I've tried work right away; the boot manager understands the file system. It has the feel of a solid, well-put-together system. Thanks for your time, Pete Prof. Peter M. Chen EECS Department, 2225 EECS 1301 Beal Ave. University of Michigan Ann Arbor, MI 48109-2122 (313) 763-4472, fax: (313) 763-4617 pmchen@eecs.umich.edu http://www.eecs.umich.edu/~pmchen/ From owner-freebsd-hackers Tue Apr 1 07:17:13 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA06234 for hackers-outgoing; Tue, 1 Apr 1997 07:17:13 -0800 (PST) Received: from cheops.anu.edu.au (cheops.anu.edu.au [150.203.76.24]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA06197; Tue, 1 Apr 1997 07:16:54 -0800 (PST) Message-Id: <199704011516.HAA06197@freefall.freebsd.org> Received: by cheops.anu.edu.au (1.37.109.16/16.2) id AA264867486; Wed, 2 Apr 1997 01:11:26 +1000 From: Darren Reed Subject: Re: root logins on secure tty's ? To: avalon@coombs.anu.edu.au (Darren Reed) Date: Wed, 2 Apr 1997 01:11:26 +1000 (EST) Cc: hackers@freebsd.org, security@freebsd.org In-Reply-To: <199704011422.GAA03481@freefall.freebsd.org> from "Darren Reed" at Apr 2, 97 00:17:28 am X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Ignore this. Someone broke root logins, thats all. In some mail from Darren Reed, sie said: > > > for some reason, in 2.2.1 source, /bin/login root logins appear > to be broken on secure tty's. > > line 271 of login.c (or thereabouts): > } else if (pwd->pw_passwd[0] == '\0') { > if (rootlogin && !rootok) { > /* pretend password okay */ > rval = 0; > goto ttycheck; > } > } > > in my ttys, I enable ttyv1 as secure, rootok == 1 and I get prompted > for a password. Were the tty insecure, I suspect this would work (I > have a null password for root). btw, I only noticed this because it > used to work on 2.1.6 and didn't after the upgrade... > > Is this (perhaps) a leftover from the breakin earlier in the year ? From owner-freebsd-hackers Tue Apr 1 07:22:23 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA06498 for hackers-outgoing; Tue, 1 Apr 1997 07:22:23 -0800 (PST) Received: from cheops.anu.edu.au (avalon@cheops.anu.edu.au [150.203.76.24]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA06475 for ; Tue, 1 Apr 1997 07:21:40 -0800 (PST) Message-Id: <199704011521.HAA06475@freefall.freebsd.org> Received: by cheops.anu.edu.au (1.37.109.16/16.2) id AA265707777; Wed, 2 Apr 1997 01:16:17 +1000 From: Darren Reed Subject: 2.2.1 & 2940UW.... To: hackers@freebsd.org Date: Wed, 2 Apr 1997 01:16:17 +1000 (EST) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk so to test out the 2940 handling in 2.2.1, I boot up... ahc0 rev 0 int a irq 10 on pci0:12 ahc0: Using left over BIOS settings ahc0: aic7880 Wide Channel, SCSI Id=7, 16 SCBs ahc0 waiting for scsi devices to settle (ahc0:4:0): "WANGTEK 6130-HS 4G16" type 1 removable SCSI 2 st0(ahc0:4:0): Sequential-Access density code 0x13, drive empty probe0(ahc0:9:0): scsi_cmd probe0(ahc0:9:0): scsi_done (ahc0:9:0): command: 0,0,0,0,0,0-[0 bytes] probe0(ahc0:9:0): scsi_cmd probe0(ahc0:9:0): scsi_done (ahc0:9:0): command: 40,0,0,3,0,0,0,0,0,0-[0 bytes] probe0(ahc0:9:0): scsi_cmd probe0(ahc0:9:0): scsi_done (ahc0:9:0): command: 12,0,0,0,2c,0-[44 bytes] ------------------------------ 000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 016: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 032: 00 00 00 00 00 00 00 00 00 00 00 00 ------------------------------ ... then try to dump a filesystem (dump 0bdsf 64 108000 9000 /dev/nrst0 /usr). Soon after, the tape drive hangs...and I see this: st0(ahc0:4:0): SCB 0x0 - timed out in dataout phase, SCSISIGI == 0x6 SEQADDR == 0x125 st0(ahc0:4:0): abort message in message buffer st0(ahc0:4:0): SCB 0x0 - timed out in dataout phase, SCSISIGI == 0x16 SEQADDR == 0x125 st0(ahc0:4:0): no longer in timeout ahc0: Issued Channel A Bus Reset. 1 SCBs aborted Clearing bus reset Clearing 'in-reset' flag , I kill the remaining dump program and try "mt status" only to be greeted with the following: st0(ahc0:4:0): SCB 0x0 - timed out while idle, LASTPHASE == 0x1, SCSISIGI == 0x0 SEQADDR == 0x6 st0(ahc0:4:0): SCB 0: Immediate reset. Flags = 0x1 st0(ahc0:4:0): no longer in timeout ahc0: Issued Channel A Bus Reset. 1 SCBs aborted Clearing bus reset Clearing 'in-reset' flag which seems to be repeating in some loop. 2.2.1-RELEASE FreeBSD 2.2.1-RELEASE #0: Tue Apr 1 23:47:31 EST 1997 (this is no APril 1 joke) Is there anyone interested in even trying to fix this ? Darren From owner-freebsd-hackers Tue Apr 1 09:33:04 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA14283 for hackers-outgoing; Tue, 1 Apr 1997 09:33:04 -0800 (PST) Received: from akane.db-net.com (akane.db-net.com [206.103.247.225]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA14278 for ; Tue, 1 Apr 1997 09:33:02 -0800 (PST) Received: from catty (catty [206.103.247.226]) by akane.db-net.com (8.8.5/8.8.5) with ESMTP id MAA24036 for ; Tue, 1 Apr 1997 12:35:33 -0500 Message-ID: <33414845.51AD@db-net.com> Date: Tue, 01 Apr 1997 12:39:17 -0500 From: Wilson MacGyver Organization: DB-Net Inc. X-Mailer: Mozilla 4.0b2 (WinNT; I) MIME-Version: 1.0 To: hackers@freebsd.org Subject: Re: p166 vs. p166mmx X-Priority: 3 (Normal) References: <199704011347.QAA20476@shadows.aeon.net> Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=us-ascii Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk mika ruohotie wrote: > first normal intel pentium 166mhz: > -- > make -k world 7595.59s user 1843.94s system 87% cpu 3:00:48.28 total > > make depend 25.66s user 9.29s system 79% cpu 44.132 total > text data bss dec hex > 798720 53248 95220 947188 e73f4 > make 237.70s user 27.34s system 84% cpu 5:12.39 total > 960912 Apr 1 11:55 > -- > > then, the intel pentium 166mhz with mmx > -- > make -k world 5982.25s user 1636.02s system 87% cpu 2:24:21.88 total > > make depend 21.54s user 8.26s system 78% cpu 37.911 total > text data bss dec hex > 798720 53248 95220 947188 e73f4 > make 188.30s user 25.57s system 89% cpu 3:58.24 total > 960912 Apr 1 15:31 Wouldn't this be because of the larger L1 cache? If memory serves, L1 cache was increased from 16K to 32K... -- Wilson MacGyver macgyver@db-net.com -------------------------------------- Veni, Vidi, Concidi. From owner-freebsd-hackers Tue Apr 1 09:45:22 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA15230 for hackers-outgoing; Tue, 1 Apr 1997 09:45:22 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id JAA15223 for ; Tue, 1 Apr 1997 09:45:19 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id KAA11569; Tue, 1 Apr 1997 10:28:47 -0700 From: Terry Lambert Message-Id: <199704011728.KAA11569@phaeton.artisoft.com> Subject: Re: Internal clock To: gbeach@cybernet.com Date: Tue, 1 Apr 1997 10:28:47 -0700 (MST) Cc: hackers@FreeBSD.ORG In-Reply-To: <3340C326.6150@cybernet.com> from "Glenn Beach" at Apr 1, 97 09:11:18 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > I have been searching for an answer to the following question. Doug > White at questions@freebsd.org said that you may be able to help. So > here it goes: > > How can I make time measurements in increments smaller than the clock > tick? Specifically, I am running BSD on a Pentium PC which has 128 > clock ticks per second. Is there any way to improve the resolution? The clock can be set faster, and generally is for the kernel profiling code. What exactly do you need it for? If for profiling, then it's already coded up for you... if for timers and so on, the timer will fire at the expected time, but that won't necessarily context switch the process which was waiting for the event, only mark it ready to run. If your system is heavily loaded, it may take more than 10ms for your process to start running again. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Tue Apr 1 09:48:40 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA15417 for hackers-outgoing; Tue, 1 Apr 1997 09:48:40 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id JAA15412 for ; Tue, 1 Apr 1997 09:48:34 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id KAA11582; Tue, 1 Apr 1997 10:30:46 -0700 From: Terry Lambert Message-Id: <199704011730.KAA11582@phaeton.artisoft.com> Subject: Re: Internal clock To: joe@pavilion.net (Josef Karthauser) Date: Tue, 1 Apr 1997 10:30:46 -0700 (MST) Cc: gbeach@cybernet.com, hackers@FreeBSD.ORG In-Reply-To: <19970401151947.61221@pavilion.net> from "Josef Karthauser" at Apr 1, 97 03:19:47 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > BTW a question to the hackers... does anyone have any plans for > adding a real-time scheduling class, aka threads under solaris? > I'm writing some midi software and could really do with a real-time > class thread to handle the timing subsystem. You should contact the RT (Real Time) list (freebsd-realtime). I have seen patches on that list for nearly full POSIX RT scheduling, albiet for a slightly older than -current version of FreeBSD. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Tue Apr 1 09:50:32 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA15615 for hackers-outgoing; Tue, 1 Apr 1997 09:50:32 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id JAA15607 for ; Tue, 1 Apr 1997 09:50:29 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id KAA11592; Tue, 1 Apr 1997 10:33:35 -0700 From: Terry Lambert Message-Id: <199704011733.KAA11592@phaeton.artisoft.com> Subject: Re: file system layout To: proff@suburbia.net Date: Tue, 1 Apr 1997 10:33:35 -0700 (MST) Cc: dfr@nlsystems.com, hackers@FreeBSD.org In-Reply-To: <19970401091834.1646.qmail@suburbia.net> from "proff@suburbia.net" at Apr 1, 97 07:18:33 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > The whole file system layout is not effective. > > Each package should have it's own tree. > Configuration modifications should be done using unionfs. > > This allows you to store installed packages on write-protected media > and union over them for any changes. The same goes for the entire > base system. I agree totally. This is the DOS multiroot model being used correctly, as opposed to the way DOS boxes use it. I have long argued that the mount of an FS should be as its own root, and that the mapping of an FS root into the mapped FS hierarchical name space should be a seperate operation. As a nice side effect, all of the NFS export handling and root vs. non-root mount discrepancies, etc. all go away, and simplify the VFSOPS considerably. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Tue Apr 1 10:35:44 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA18947 for hackers-outgoing; Tue, 1 Apr 1997 10:35:44 -0800 (PST) Received: from pdx1.world.net (pdx1.world.net [192.243.32.18]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA18942 for ; Tue, 1 Apr 1997 10:35:42 -0800 (PST) From: proff@suburbia.net Received: from suburbia.net (suburbia.net [203.4.184.1]) by pdx1.world.net (8.7.5/8.7.3) with SMTP id KAA01745 for ; Tue, 1 Apr 1997 10:37:59 -0800 (PST) Received: (qmail 3990 invoked by uid 110); 1 Apr 1997 18:34:42 -0000 Message-ID: <19970401183442.3989.qmail@suburbia.net> Subject: Re: Internal clock In-Reply-To: <199704011730.KAA11582@phaeton.artisoft.com> from Terry Lambert at "Apr 1, 97 10:30:46 am" To: terry@lambert.org (Terry Lambert) Date: Wed, 2 Apr 1997 04:34:42 +1000 (EST) Cc: joe@pavilion.net, gbeach@cybernet.com, hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > BTW a question to the hackers... does anyone have any plans for > > adding a real-time scheduling class, aka threads under solaris? > > I'm writing some midi software and could really do with a real-time > > class thread to handle the timing subsystem. > > You should contact the RT (Real Time) list (freebsd-realtime). I have > seen patches on that list for nearly full POSIX RT scheduling, albiet > for a slightly older than -current version of FreeBSD. > The sad reality is if these things are not incorporated in -current then they fall by as the original authors move onto other projects. A lot of good ideas have died of because of this. When code makes it to current it guarantees a developer base that will take note of it, and pursue it beyond the attention span of the original author. Current should be a proving ground for new code, not merely -RELEASE x.x in the process of being debugged. If it is seen as unfit or unworthy then it can be #ifdef'd off by default, or eventually backed out. Cheers, Julian. From owner-freebsd-hackers Tue Apr 1 10:36:23 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA19046 for hackers-outgoing; Tue, 1 Apr 1997 10:36:23 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA19036 for ; Tue, 1 Apr 1997 10:36:20 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id LAA11677; Tue, 1 Apr 1997 11:19:28 -0700 From: Terry Lambert Message-Id: <199704011819.LAA11677@phaeton.artisoft.com> Subject: Re: question on buffer cache and VMIO To: pmchen@eecs.umich.edu (Peter M. Chen) Date: Tue, 1 Apr 1997 11:19:28 -0700 (MST) Cc: freebsd-hackers@freebsd.org In-Reply-To: <199704011439.JAA26225@life.eecs.umich.edu> from "Peter M. Chen" at Apr 1, 97 09:39:42 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Hi, > I'm starting to hack on FreeBSD and had some questions. > > Background: > My goal is to make the file cache reliable without writing to > disk under normal operation (see Rio paper in ASPLOS 1996). We > protect against power loss by using a battery; we protect against > kernel crashes by write-protecting the file cache. The main > benefit is you get reliability equivalent to mounting the file > system sync, yet with the performance of async (actually even > better, since FreeBSD's async option still does substantial > amounts of I/O). This works really well for mmap'ed files, which > makes it possible to have VERY fast transactions. It should also > extend the battery life for portables. > See http://www.eecs.umich.edu/~pmchen/ for more info. This is hardware... is it going to be generally available/purchasable? One big issue: I'm not sure how this will interact with the "soft updates" facility described in the Ganger/Patt paper. This facility is a planned future feature on nearly everyone's whiteboards. If you have read the paper, I can discuss the idea of a generic implementation that is not FS dependent (like the paper's Appendix A UFS example is) and the ramifications that would have on FS/cache interaction. > Questions re: buffer cache and VMIO > 1) What's the relationship between the buffer cache and VM cache (I'm > not sure this is the proper terminology)? The buffer cache seems to > hold file data and metadata, while the VM cache holds only file data. > Do buffer headers point to the VM cache data? Can the VM cache hold > file data that is not in the buffer cache? The VM and the buffer cache have been unified. File buffers are chains of pages hung off of the in core vnodes for the inodes backing them. RAM is for actualization of virtual pages, and for allocation of kernel resources. John Dyson and David Greenman are the people to contact about documenting this in detail. > 2) The buffer cache seems small relative to the physical memory. E.g. > on a 64 MB machine (51 MB available), maxbufspace defaults to only > 6.3 MB. For working sets larger than this, there appears to be > significant overhead in frequently moving data between the VM > cache and buffer cache. Would it make sense to set NBUF larger > (e.g. enough to have the buffer cache fill memory)? This is a different issue. This has to do with the ratio of in core pages backed by files to those not backed by files. > 3) Dirty data gets written to disk when it leaves the buffer cache, even > if it is also in the VM cache. This makes sense normally (since > Unix traditionally bawrote data to disk as soon as it filled a block), > but this prevents my keeping lots of dirty file data around. This is an issue which boils down to what happens to a vnode when it is about to be reused. Effectively, there can be "good" cache data in core, but no vnode which references it. When the vnode reference is being destroyed, dirty pages are forced out. The data will then need to be re-read. The main benefit to having a vnode/extent based caching mechanism over a device/extent caching mechanism, is that you effectively remove the device and FS size limitations inherent in the processor architecture limits on the VM system. The limit applies on a per vnode basis instead of applying on a per device basis. The main drawback is that pages in core which contain valid data, but without vnode references, must be reloaded from disk in order to be used. Part of this problem is that vnodes are referenced as entities that are seperate from the per FS on disk data which backs instantiation; that is, there is a global vnode pool, with a discrete data reference to the per FS data in the vnode, rather than the vnodes being allocated as part of the in core FS data object... ie: /* in core vnode*/ struct vnode { void *v_data; /* fs specific data*/ }; /* in core inode*/ struct inode { struct vnode *i_vnode; /* vnode for this inode*/ struct dinode i_din; /* on disk inode*/ }; Instead of the more desirable: /* in core inode*/ struct inode { struct vnode i_vnode; /* vnode instance for this inode*/ struct dinode i_din; /* on disk inode*/ }; Part of the reasoning behind this is explained by following the use of the "struct fileops" in the kernel, something which should probably go away as quickly as possible (the vnode v_un pointer union is an artifact of this same patchwork glue). > 4) What happens to mmap'ed data? Does it reside in the VM cache? Are > there buffer headers for mmap'ed data? All mmap'ed data is treated as file data; unlike System V shared memory, it does not need to remain in core at all times (the SYSV SHM has to because it is established in the kernel map, not in an independent file map; there is no concept of backing store other than file as backing store, and the SYSV SHM was not, last time I looked, swappable). > 5) I came across a strange phenomenon when trying to get rid of all > disk writes in ufs_remove. Even if i_nlink goes to 0, the file > is still fsync'ed. The call graph is: ufs_remove -> vput -> vrele -> > vm_object_deallocate -> vm_object_terminate -> vinvalbuf (with V_SAVE) > -> VOP_FSYNC. The system fails when I have vm_object_terminate > check for i_nlink and call vinvalbuf without V_SAVE. Can someone > explain why a deleted file needs to get fsynced (note that this > isn't the directory, but the actual file)? It probably does not. 8-(. I expect the problem is where the directory vs. open instance refrence counting occurs. > General kernel questions: > 1) I'd like the ability to read and write kernel global variables (without > going to ddb). I tried kvm, but that only works for variables in > i386/i386/symbols.raw. kgdb only works for off-line core dumps. > I finally used nm to get the symbol address and directly read and > wrote to /dev/kmem. This works fine, but I was wondering if there's > a better solution that exists already. Are these experimentation tunables? If they are, you should probably compile with debug and use SYSCTL_INT() to install debugging variables into the sysctl tree. There is a good example of this in the kernel source file /sys/ufs/ffs/ffs_alloc.c. Alternately, if these are permanent tunables, there is a good example of doing this for VFS in /sys/kern/vfs_init.c. To add your own hierarchy node, SYSCTL_NODE() in /sys/kern/subr_prof.c is a good example. If you are running a "cleaner" or similar process, you may need a process parameter. SYSCTL_PROC in /sys/kern/kern_proc.c is the example I'd use for that. > By the way, I've been SUPER impressed with FreeBSD. The code is > well-written, especially compared to other operating systems I've > worked on; Compiling the kernel is VERY fast; the system is fast and > small; the installation process is easy; the small number of packages > and ports I've tried work right away; the boot manager understands > the file system. It has the feel of a solid, well-put-together system. Heh. And with Rio added it should be 4-22 times faster, right? 8-) 8-). I really look forward to seeing how you do... and when you get done, you will probably find a ready market for the hardware you will be using. I'll help on any question I can, but it sure looks like you will need John Dyson's and David Greenman's help on some of these things which are not well documented... Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Tue Apr 1 11:21:02 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA22011 for hackers-outgoing; Tue, 1 Apr 1997 11:21:02 -0800 (PST) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA21994 for ; Tue, 1 Apr 1997 11:20:57 -0800 (PST) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id MAA04841; Tue, 1 Apr 1997 12:20:39 -0700 (MST) Date: Tue, 1 Apr 1997 12:20:39 -0700 (MST) Message-Id: <199704011920.MAA04841@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: proff@suburbia.net Cc: terry@lambert.org (Terry Lambert), hackers@freebsd.org Subject: Re: Internal clock In-Reply-To: <19970401183442.3989.qmail@suburbia.net> References: <199704011730.KAA11582@phaeton.artisoft.com> <19970401183442.3989.qmail@suburbia.net> X-Mailer: VM 6.22 under 19.15 XEmacs Lucid Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk proff@suburbia.net writes: > > > BTW a question to the hackers... does anyone have any plans for > > > adding a real-time scheduling class, aka threads under solaris? > > > I'm writing some midi software and could really do with a real-time > > > class thread to handle the timing subsystem. > > > > You should contact the RT (Real Time) list (freebsd-realtime). I have > > seen patches on that list for nearly full POSIX RT scheduling, albiet > > for a slightly older than -current version of FreeBSD. > > > > The sad reality is if these things are not incorporated in -current > then they fall by as the original authors move onto other projects. Then adding the code into the tree is a 'bad thing', since it becomes unsupported. If no-one is willing to incorporate/support the code, then it shouldn't be incorporated. Nate From owner-freebsd-hackers Tue Apr 1 11:29:54 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA23098 for hackers-outgoing; Tue, 1 Apr 1997 11:29:54 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA23076 for ; Tue, 1 Apr 1997 11:29:46 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id VAA04776; Tue, 1 Apr 1997 21:29:28 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id VAA16392; Tue, 1 Apr 1997 21:09:09 +0200 (MET DST) Message-ID: <19970401210909.DC36612@uriah.heep.sax.de> Date: Tue, 1 Apr 1997 21:09:09 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: freebsd-hackers@FreeBSD.ORG Cc: pmchen@eecs.umich.edu (Peter M. Chen) Subject: Re: question on buffer cache and VMIO References: <199704011439.JAA26225@life.eecs.umich.edu> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199704011439.JAA26225@life.eecs.umich.edu>; from Peter M. Chen on Apr 1, 1997 09:39:42 -0500 Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk As Peter M. Chen wrote: > Questions re: buffer cache and VMIO While i cannot answer these... > General kernel questions: > 1) I'd like the ability to read and write kernel global > variables (without going to ddb). I tried kvm, but that only works > for variables in i386/i386/symbols.raw. Are you sure about this? I've got my nose in dset(8) some time ago, and it really uses kvm for variables that are not listed in symbols.raw. > kgdb only works for > off-line core dumps. No. ``gdb -k /kernel /dev/mem'' debugs your currently running kernel. Add -w to the commandline, and you'll be allowed to write (unless you're running on a higher securelevel, but then you're at a loss anyway). -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Tue Apr 1 11:32:07 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA23339 for hackers-outgoing; Tue, 1 Apr 1997 11:32:07 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA23334 for ; Tue, 1 Apr 1997 11:32:03 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id VAA04827; Tue, 1 Apr 1997 21:31:53 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id UAA16300; Tue, 1 Apr 1997 20:34:29 +0200 (MET DST) Message-ID: <19970401203429.XR37988@uriah.heep.sax.de> Date: Tue, 1 Apr 1997 20:34:29 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freebsd.org Cc: joe@pavilion.net (Josef Karthauser), gbeach@cybernet.com Subject: Re: Internal clock References: <3340C326.6150@cybernet.com> <19970401151947.61221@pavilion.net> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <19970401151947.61221@pavilion.net>; from Josef Karthauser on Apr 1, 1997 15:19:47 +0100 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Josef Karthauser wrote: > I'm using gettimeofday which returns: > > struct timeval { > long tv_sec; /* seconds since Jan. 1, 1970 */ > long tv_usec; /* and microseconds */ > }; > > I've no idea of the actual system resolution, ... RTFM clocks(7). > BTW a question to the hackers... does anyone have any plans for > adding a real-time scheduling class, aka threads under solaris? There's already a pseudo-realtime scheduling class available, see rtprio(1). This has, of course, nothing to do with threads at all, nor will it be true realtime, since there's still the problem that scheduling doesn't happen while running in kernel context (i.e., the kernel cannot be preempted -- Solaris seems to have gone great lengths to create some preemption points for long-time running kernel functions). -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Tue Apr 1 11:32:38 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA23380 for hackers-outgoing; Tue, 1 Apr 1997 11:32:38 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA23374 for ; Tue, 1 Apr 1997 11:32:33 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id VAA04830 for hackers@freebsd.org; Tue, 1 Apr 1997 21:32:14 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id UAA16324; Tue, 1 Apr 1997 20:40:13 +0200 (MET DST) Message-ID: <19970401204013.QD16563@uriah.heep.sax.de> Date: Tue, 1 Apr 1997 20:40:13 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freebsd.org Subject: Re: root logins on secure tty's ? References: <199704011422.GAA03481@freefall.freebsd.org> <199704011516.HAA06197@freefall.freebsd.org> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199704011516.HAA06197@freefall.freebsd.org>; from Darren Reed on Apr 2, 1997 01:11:26 +1000 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Darren Reed wrote: > Ignore this. Someone broke root logins, thats all. This someone has also fixed them again in revs 1.22 and 1.12.2.3 of login.c. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Tue Apr 1 11:50:42 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA24642 for hackers-outgoing; Tue, 1 Apr 1997 11:50:42 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA24629 for ; Tue, 1 Apr 1997 11:50:34 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA12670; Tue, 1 Apr 1997 14:50:02 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Tue, 1 Apr 1997 14:50 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id NAA01192; Tue, 1 Apr 1997 13:48:54 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id NAA02002; Tue, 1 Apr 1997 13:54:37 -0500 (EST) Date: Tue, 1 Apr 1997 13:54:37 -0500 (EST) From: Thomas David Rivers Message-Id: <199704011854.NAA02002@lakes.water.net> To: ponds!coombs.anu.edu.au!avalon, ponds!freebsd.org!hackers Subject: Re: 2.2.1 & 2940UW.... Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > so to test out the 2940 handling in 2.2.1, I boot up... ... descriptions of aha2940 timeouts/resets/etc ... > > st0(ahc0:4:0): SCB 0x0 - timed out while idle, LASTPHASE == 0x1, SCSISIGI == 0x0 > SEQADDR == 0x6 > st0(ahc0:4:0): SCB 0: Immediate reset. Flags = 0x1 > st0(ahc0:4:0): no longer in timeout > ahc0: Issued Channel A Bus Reset. 1 SCBs aborted > Clearing bus reset > Clearing 'in-reset' flag ... > > Is there anyone interested in even trying to fix this ? Yes, there are several people interested in fixing this. I have it on good authority that Justin is looking into the problem. > > Darren > - Dave Rivers - From owner-freebsd-hackers Tue Apr 1 12:03:36 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA25885 for hackers-outgoing; Tue, 1 Apr 1997 12:03:36 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA25880 for ; Tue, 1 Apr 1997 12:03:33 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id MAA11874; Tue, 1 Apr 1997 12:46:06 -0700 From: Terry Lambert Message-Id: <199704011946.MAA11874@phaeton.artisoft.com> Subject: Re: Internal clock To: nate@mt.sri.com (Nate Williams) Date: Tue, 1 Apr 1997 12:46:06 -0700 (MST) Cc: proff@suburbia.net, terry@lambert.org, hackers@freebsd.org In-Reply-To: <199704011920.MAA04841@rocky.mt.sri.com> from "Nate Williams" at Apr 1, 97 12:20:39 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > The sad reality is if these things are not incorporated in -current > > then they fall by as the original authors move onto other projects. > > Then adding the code into the tree is a 'bad thing', since it becomes > unsupported. If no-one is willing to incorporate/support the code, then > it shouldn't be incorporated. Better that it be lost forever? Code shouldn't need a hell of a lot of maintenance, if the interfaces for plugging the code in are fairly static and well enough designed that they can remain that way. Seems to me that the issue is the fluidity of the kernel interfeaces, not the module code, that is at fault. Define a spanning set, and access it via macros, and the underlying implementation can change as much as you want without damaging the utlity of the unmaintained code. Network interfaces are a good example of one place where this should be happening. The RT scheduling question that started this particular thread is another. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Tue Apr 1 12:03:55 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA25923 for hackers-outgoing; Tue, 1 Apr 1997 12:03:55 -0800 (PST) Received: from pdx1.world.net (pdx1.world.net [192.243.32.18]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA25917 for ; Tue, 1 Apr 1997 12:03:53 -0800 (PST) From: proff@suburbia.net Received: from suburbia.net (suburbia.net [203.4.184.1]) by pdx1.world.net (8.7.5/8.7.3) with SMTP id MAA03128 for ; Tue, 1 Apr 1997 12:06:06 -0800 (PST) Received: (qmail 3723 invoked by uid 110); 1 Apr 1997 20:03:18 -0000 Message-ID: <19970401200318.3722.qmail@suburbia.net> Subject: Re: Internal clock In-Reply-To: <199704011920.MAA04841@rocky.mt.sri.com> from Nate Williams at "Apr 1, 97 12:20:39 pm" To: nate@mt.sri.com (Nate Williams) Date: Wed, 2 Apr 1997 06:03:18 +1000 (EST) Cc: terry@lambert.org, hackers@FreeBSD.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > proff@suburbia.net writes: > > then they fall by as the original authors move onto other projects. > > Then adding the code into the tree is a 'bad thing', since it becomes > unsupported. If no-one is willing to incorporate/support the code, then > it shouldn't be incorporated. > > > Nate > That philosophy guarentees one to failure. The code doesn't have a chance of being supported till it gets exposure and people start relying on it. If it starts bit-rotting then you let it fade (e.g x.25 code), or back it out. Don't shoot yourself before you are even on the track. Cheers, Julian. From owner-freebsd-hackers Tue Apr 1 12:08:50 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA26246 for hackers-outgoing; Tue, 1 Apr 1997 12:08:50 -0800 (PST) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA26239 for ; Tue, 1 Apr 1997 12:08:46 -0800 (PST) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id NAA05171; Tue, 1 Apr 1997 13:05:03 -0700 (MST) Date: Tue, 1 Apr 1997 13:05:03 -0700 (MST) Message-Id: <199704012005.NAA05171@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Terry Lambert Cc: nate@mt.sri.com (Nate Williams), proff@suburbia.net, hackers@freebsd.org Subject: Re: Internal clock In-Reply-To: <199704011946.MAA11874@phaeton.artisoft.com> References: <199704011920.MAA04841@rocky.mt.sri.com> <199704011946.MAA11874@phaeton.artisoft.com> X-Mailer: VM 6.22 under 19.14 XEmacs Lucid Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Terry Lambert writes: > > > The sad reality is if these things are not incorporated in -current > > > then they fall by as the original authors move onto other projects. > > > > Then adding the code into the tree is a 'bad thing', since it becomes > > unsupported. If no-one is willing to incorporate/support the code, then > > it shouldn't be incorporated. > > Better that it be lost forever? Yes. > Code shouldn't need a hell of a lot of maintenance, if the interfaces > for plugging the code in are fairly static and well enough designed > that they can remain that way. Yeah, right. If that were the case, you and I wouldn't be paid the big bucks to be software engineers, since any poor schmuck off the street could do our job. Nate From owner-freebsd-hackers Tue Apr 1 12:09:24 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA26293 for hackers-outgoing; Tue, 1 Apr 1997 12:09:24 -0800 (PST) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA26283 for ; Tue, 1 Apr 1997 12:09:20 -0800 (PST) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id NAA05197; Tue, 1 Apr 1997 13:08:42 -0700 (MST) Date: Tue, 1 Apr 1997 13:08:42 -0700 (MST) Message-Id: <199704012008.NAA05197@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: proff@suburbia.net Cc: nate@mt.sri.com (Nate Williams), hackers@freebsd.org Subject: Re: Internal clock In-Reply-To: <19970401200318.3722.qmail@suburbia.net> References: <199704011920.MAA04841@rocky.mt.sri.com> <19970401200318.3722.qmail@suburbia.net> X-Mailer: VM 6.22 under 19.14 XEmacs Lucid Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > then they fall by as the original authors move onto other projects. > > > > Then adding the code into the tree is a 'bad thing', since it becomes > > unsupported. If no-one is willing to incorporate/support the code, then > > it shouldn't be incorporated. > > > > That philosophy guarentees one to failure. The code doesn't have > a chance of being supported till it gets exposure and people start > relying on it. No, someone has to integrate it (hence support it). Nate From owner-freebsd-hackers Tue Apr 1 12:45:12 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA28937 for hackers-outgoing; Tue, 1 Apr 1997 12:45:12 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA28932 for ; Tue, 1 Apr 1997 12:45:03 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id NAA12015; Tue, 1 Apr 1997 13:27:47 -0700 From: Terry Lambert Message-Id: <199704012027.NAA12015@phaeton.artisoft.com> Subject: Re: Internal clock To: nate@mt.sri.com (Nate Williams) Date: Tue, 1 Apr 1997 13:27:47 -0700 (MST) Cc: terry@lambert.org, nate@mt.sri.com, proff@suburbia.net, hackers@freebsd.org In-Reply-To: <199704012005.NAA05171@rocky.mt.sri.com> from "Nate Williams" at Apr 1, 97 01:05:03 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > Better that it be lost forever? > > Yes. You can't be serious. You must be engaging in devil's advocacy... > > Code shouldn't need a hell of a lot of maintenance, if the interfaces > > for plugging the code in are fairly static and well enough designed > > that they can remain that way. > > Yeah, right. If that were the case, you and I wouldn't be paid the big > bucks to be software engineers, since any poor schmuck off the street > could do our job. And your point is that we are somehow superior to the average schmuck because we write code that needs a lot of maintenance? 8-). Really, the issue is one of designing good kernel interfaces, not the software that plugs into them. Are you saying that the general antipathy against well designed kernel interfaces is a result of engineers attempting to maintain job security? Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Tue Apr 1 12:53:55 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA29415 for hackers-outgoing; Tue, 1 Apr 1997 12:53:55 -0800 (PST) Received: from awfulhak.demon.co.uk (awfulhak.demon.co.uk [158.152.17.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA29405 for ; Tue, 1 Apr 1997 12:53:41 -0800 (PST) Received: from awfulhak.demon.co.uk (localhost.lan.awfulhak.org [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id VAA01643; Tue, 1 Apr 1997 21:20:18 +0100 (BST) Message-Id: <199704012020.VAA01643@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: johnp@lodgenet.com cc: hackers@FreeBSD.ORG Subject: Re: DNS & Sendmail In-reply-to: Your message of "Mon, 31 Mar 1997 22:20:59 MDT." <199704010420.WAA00706@knight.trosoft.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 01 Apr 1997 21:20:17 +0100 From: Brian Somers Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > I just updated to 2.2.1-RELEASE, and upon bootup sendmail hangs.. > I am running a cache only name server on a local net, ppp on-demand for > external stuff... > When I kill named, I can start sendmail. [.....] Does FEATURE(nodns) and FEATURE(nocanonify) help ? -- Brian , Don't _EVER_ lose your sense of humour.... From owner-freebsd-hackers Tue Apr 1 12:54:32 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA29451 for hackers-outgoing; Tue, 1 Apr 1997 12:54:32 -0800 (PST) Received: from awfulhak.demon.co.uk (awfulhak.demon.co.uk [158.152.17.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA29444 for ; Tue, 1 Apr 1997 12:54:26 -0800 (PST) Received: from awfulhak.demon.co.uk (localhost.lan.awfulhak.org [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id VAA01934; Tue, 1 Apr 1997 21:27:38 +0100 (BST) Message-Id: <199704012027.VAA01934@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: proff@suburbia.net cc: dfr@nlsystems.com (Doug Rabson), hackers@FreeBSD.org Subject: Re: file system layout In-reply-to: Your message of "Tue, 01 Apr 1997 19:18:33 +1000." <19970401091834.1646.qmail@suburbia.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 01 Apr 1997 21:27:38 +0100 From: Brian Somers Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk [.....] > Each package should have it's own tree. > Configuration modifications should be done using unionfs. > > This allows you to store installed packages on write-protected media > and union over them for any changes. The same goes for the entire > base system. > > -- > Prof. Julian Assange |If you want to build a ship, don't drum up people > |together to collect wood and don't assign them tasks > proff@iq.org |and work, but rather teach them to long for the endless > proff@gnu.ai.mit.edu |immensity of the sea. -- Antoine de Saint Exupery I know someone who attempted this on a linux box - seperating each package into a distinct tree of its own. Is unionfs functional in -current ? -- Brian , Don't _EVER_ lose your sense of humour.... From owner-freebsd-hackers Tue Apr 1 12:54:35 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA29469 for hackers-outgoing; Tue, 1 Apr 1997 12:54:35 -0800 (PST) Received: from awfulhak.demon.co.uk (awfulhak.demon.co.uk [158.152.17.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA29447 for ; Tue, 1 Apr 1997 12:54:30 -0800 (PST) Received: from awfulhak.demon.co.uk (localhost.lan.awfulhak.org [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id VAA01628; Tue, 1 Apr 1997 21:17:02 +0100 (BST) Message-Id: <199704012017.VAA01628@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: Greg Rowe cc: Thomas David Rivers , hackers@FreeBSD.ORG Subject: Re: aha2940 problems on 2.1.7.1. In-reply-to: Your message of "Tue, 01 Apr 1997 08:28:28 MDT." <9704010828.ZM8938@nevis.oss.uswest.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 01 Apr 1997 21:17:02 +0100 From: Brian Somers Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk FWIW, I'm getting similar problems in -current. I'm a bit naieve when it comes to this stuff, and I can't confirm that the hardware works 100%, all I can say is that it worked 'till October last year and then stopped. Being lazy, I just stopped doing backups (this is on my home machine), and hoped the problem would go away. Then, last month, I lost a 4Gb drive !!!! All my company documents etc - I was lucky that I'd backed up my company accounts just that day ! Anyway, I'm getting the following: sd0: SCB 0x6 - timed out in command phase, SCSISIGI == 0x84 SEQADDR == 0x42 st0: abort message in message buffer sd0: SCB 0x7 timedout while recovery in progress st0: SCB 2 - Abort Completed Panic: Couldn't find next SCB Debugger ("panic") The trace was in panic() in ahc_reset_device() in ahc_handle_scsiint() in ahc_intr() in Xresume10() This is reproducable by trying to write to a tape - even a brand new one. The hardware when that happened was a 2940W. Since then, I've got myself a new 1542C - when this drive worked before, it was on a 1542 (albeit not the same one). The tape drive (a 4/8 Gb DAT drive) is now alone on the 1542 and produces the same results (without the sd0 problem of course). No panics happen with the DAT alone on its own controller, but lots of nasty printfs from the kernel (as above) rear their ugly heads. To be more specific, I can read and write a variable amount of data from or to the drive using tar or dump/restore before this happens, I'm pretty sure it's not getting to the end of any tapes (all 120M 4mm). I don't think I've missed any details in my ramblings Now that I've got a scenario where I can test without panicing (I hate panicing when the sync's don't work, I have no backup), I'm willing to try things out if anyone can suggest anything. I can also arrange for an account on the machine if anyone needs to reproduce stuff on demand. Cheers. $ dmesg Copyright (c) 1992-1997 FreeBSD Inc. Copyright (c) 1982, 1986, 1989, 1991, 1993 The Regents of the University of California. All rights reserved. FreeBSD 3.0-CURRENT #0: Tue Apr 1 02:21:39 BST 1997 brian@awfulhak.demon.co.uk:/usr/src/sys/compile/AWFULHAK CPU: Pentium Pro (199.43-MHz 686-class CPU) Origin = "GenuineIntel" Id = 0x617 Stepping=7 Features=0xf9ff,MTRR,PGE,MCA,CMOV> real memory = 83886080 (81920K bytes) avail memory = 78352384 (76516K bytes) DEVFS: ready for devices bdevsw_add_generic: adding D_DISK flag for device 15 Probing for devices on PCI bus 0: chip0 rev 2 on pci0:0:0 chip1 rev 1 on pci0:7:0 chip2 rev 0 on pci0:7:1 vga0 rev 0 on pci0:11:0 ahc0 rev 0 int a irq 10 on pci0:15:0 ahc0: aic7870 Wide Channel, SCSI Id=7, 16 SCBs scbus1 at ahc0 bus 0 sd0 at scbus1 target 0 lun 0 sd0: type 0 fixed SCSI 2 sd0: Direct-Access 4153MB (8506782 512 byte sectors) sd0: with 3421 cyls, 18 heads, and an average 138 sectors/track sd1 at scbus1 target 1 lun 0 sd1: type 0 fixed SCSI 2 sd1: Direct-Access 507MB (1039329 512 byte sectors) sd1: with 2380 cyls, 6 heads, and an average 72 sectors/track sd2 at scbus1 target 2 lun 0 sd2: type 0 fixed SCSI 2 sd2: Direct-Access 2048MB (4194685 512 byte sectors) sd2: with 2621 cyls, 19 heads, and an average 84 sectors/track sd3 at scbus1 target 3 lun 0 sd3: type 0 fixed SCSI 2 sd3: Direct-Access 234MB (479350 512 byte sectors) sd3: with 1818 cyls, 4 heads, and an average 65 sectors/track sd4 at scbus1 target 4 lun 0 sd4: type 0 fixed SCSI 2 sd4: Direct-Access 1908MB (3907911 512 byte sectors) sd4: with 2621 cyls, 21 heads, and an average 71 sectors/track de0 rev 17 int a irq 9 on pci0:17:0 de0: 21041 [10Mb/s] pass 1.1 de0: address 00:00:c0:ff:e9:ce Probing for devices on the ISA bus: sc0 at 0x60-0x6f irq 1 on motherboard sc0: VGA color <12 virtual consoles, flags=0x0> sio0 at 0x3f8-0x3ff irq 4 on isa sio0: type 16550A sio1 at 0x2f8-0x2ff irq 3 on isa sio1: type 16550A sio2 at 0x3e8-0x3ef irq 5 on isa sio2: type 16550A sio3 not found at 0x2e8 <- de0 stole the IRQ instead of 15 :( lpt0 at 0x378-0x37f irq 7 on isa lpt0: Interrupt-driven port lp0: TCP/IP capable interface psm0 at 0x60-0x64 irq 12 on motherboard psm0: device ID 0, 2 buttons 100 nSEC ok, using 150 nSEC aha0 at 0x330-0x333 irq 11 drq 5 on isa scbus0 at aha0 bus 0 st0 at scbus0 target 5 lun 0 st0: type 1 removable SCSI 2 st0: Sequential-Access density code 0x24, variable blocks, write-enabled cd0 at scbus0 target 6 lun 0 cd0: type 5 removable SCSI 2 cd0: CD-ROM can't get the size fdc0 at 0x3f0-0x3f7 irq 6 drq 2 on isa fdc0: NEC 72065B fd0: 1.44MB 3.5in wdc0 at 0x1f0-0x1f7 irq 14 on isa wdc0: unit 0 (wd0): wd0: 1549MB (3173184 sectors), 3148 cyls, 16 heads, 63 S/T, 512 B/S npx0 on motherboard npx0: INT 16 interface changing root device to wd0a DEVFS: ready to run IP packet filtering initialized, divert enabled, logging disabled de0: enabling 10baseT port -- Brian , Don't _EVER_ lose your sense of humour.... From owner-freebsd-hackers Tue Apr 1 12:54:50 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA29500 for hackers-outgoing; Tue, 1 Apr 1997 12:54:50 -0800 (PST) Received: from awfulhak.demon.co.uk (awfulhak.demon.co.uk [158.152.17.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA29493; Tue, 1 Apr 1997 12:54:44 -0800 (PST) Received: from awfulhak.demon.co.uk (localhost.lan.awfulhak.org [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id UAA00720; Tue, 1 Apr 1997 20:49:07 +0100 (BST) Message-Id: <199704011949.UAA00720@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: dennis cc: Developer , freebsd-current@FreeBSD.ORG, freebsd-hackers@FreeBSD.ORG Subject: Re: PPP Desperate for help! In-reply-to: Your message of "Tue, 01 Apr 1997 09:50:12 CDT." <3.0.32.19970401095010.006a2504@etinc.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 01 Apr 1997 20:49:07 +0100 From: Brian Somers Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > At 01:42 PM 4/1/97 +0100, Developer wrote: > > > > > >On Tue, 1 Apr 1997, Andrzej Bialecki wrote: > > > >> On Tue, 1 Apr 1997, Developer wrote: > >> > >> > > >> > We are really stuck trying to get the PPP user program (ppp) to > connect to > >> > our Perle 833 dial in server. We can connect using Windows NT fine but on > >> > BSD ppp logs in and then data will travel in both directions (As I can > see > >> > the modem lights flash at both ends for both send/recieve) but no data > >> > seems to actually get through as pings do not work. The packets from the > >> > server to the user, but seem to be lost on the way back. > >> > >> Let me take a guess: maybe it's a known problem with tcp_extensions set to > >> YES in /etc/sysconfig? Some broken TCP stacks don't like it, so the > >> solution would be to turn it off. > >> > >> Andy > > > >We've tried this - no difference. Any more ideas please? > > Why not post a trace of the ppp negotiations? Perhaps there is a parameter > conflict....are you getting to state IPCP_OPENED? If not, you cant pass > traffic. > > Dennis I suspect you're way ahead of the problem. Has the original poster tried "set openmode active" ? Without this, a client ppp will wait for the server to initiate LCP. I think it may be frugal to make this the default for both client *and* server. A lot of server implementations wait for the client to start. Any comments ? -- Brian , Don't _EVER_ lose your sense of humour.... From owner-freebsd-hackers Tue Apr 1 12:55:39 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA29583 for hackers-outgoing; Tue, 1 Apr 1997 12:55:39 -0800 (PST) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA29572 for ; Tue, 1 Apr 1997 12:55:36 -0800 (PST) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id NAA05487; Tue, 1 Apr 1997 13:51:48 -0700 (MST) Date: Tue, 1 Apr 1997 13:51:48 -0700 (MST) Message-Id: <199704012051.NAA05487@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Terry Lambert Cc: nate@mt.sri.com (Nate Williams), proff@suburbia.net, hackers@freebsd.org Subject: Re: Internal clock In-Reply-To: <199704012027.NAA12015@phaeton.artisoft.com> References: <199704012005.NAA05171@rocky.mt.sri.com> <199704012027.NAA12015@phaeton.artisoft.com> X-Mailer: VM 6.22 under 19.15 XEmacs Lucid Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > Code shouldn't need a hell of a lot of maintenance, if the interfaces > > > for plugging the code in are fairly static and well enough designed > > > that they can remain that way. > > > > Yeah, right. If that were the case, you and I wouldn't be paid the big > > bucks to be software engineers, since any poor schmuck off the street > > could do our job. > > And your point is that we are somehow superior to the average schmuck > because we write code that needs a lot of maintenance? 8-). *laugh* Software 'engineering' is something I spent significant time studying, and no matter how good you are maintenance makes up 90% of the 'time' spent on code for most projects. One could argue that the entire FreeBSD project is doing 'maintenance' on the CSRG code tree. > Really, the issue is one of designing good kernel interfaces, not the > software that plugs into them. Really, the issue of putting a man on Mars is designing a good space ship, not actually building the darn ship. Nate From owner-freebsd-hackers Tue Apr 1 13:00:23 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA29973 for hackers-outgoing; Tue, 1 Apr 1997 13:00:23 -0800 (PST) Received: from matrix.42.org (sec@matrix.42.org [192.68.213.129]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id NAA29957 for ; Tue, 1 Apr 1997 13:00:16 -0800 (PST) Received: (from sec@localhost) by matrix.42.org (8.6.12/8.6.12) id XAA04589; Tue, 1 Apr 1997 23:00:08 +0200 To: freebsd-hackers@freebsd.org Path: sec From: sec@42.org (Stefan `Sec` Zehl) Newsgroups: muc.lists.freebsd.hackers Subject: Re: ftruncate("directory")... Date: 1 Apr 1997 23:00:06 +0200 Organization: Internet@home Lines: 39 Message-ID: References: <199703311832.LAA09800@phaeton.artisoft.com> X-Newsreader: slrn (0.9.3.0-2 BETA UNIX) Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I didn't get the original article, so i reply to this one instead :) > > Has anyone written (is it possible?) a utility that can > > 'truncate' a directory? > > > > Essentially, what I'm looking at is something that would > > open a directory, re-org the entries in it so that only the 'good' > > entries are at the head of the file, and then truncate it... > > > > Not sure if this is actually possible, but am going to play > > with it here...but if its already been done, all the better... :) Hmm, i always solved this problem by #!/bin/sh mkdir new_dir cd old_dir mv * ../new_dir cd .. rmdir old_dir mv new_dir old_dir okay, its not the nicest solution, but it works perfectly for me :) > For > the amount of fragmentation necessary to achieve utility from this > recovery, you would need to have set up a scenario terribly different > from standard utilization patterns. Perhaps the lmbench directory > create/delete test? I sometimes create a whole bunch of files (with a script) in a directory, and just want to keep an very few of them - so the directory is actually much bigger than it needs to be :) CU, Sec -- Fuer die Raupe ist es das Ende der Welt, Fuer den Rest der Welt ist es ein Schmetterling Error 0: No error From owner-freebsd-hackers Tue Apr 1 13:03:16 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA00283 for hackers-outgoing; Tue, 1 Apr 1997 13:03:16 -0800 (PST) Received: from matrix.42.org (sec@matrix.42.org [192.68.213.129]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id NAA00277 for ; Tue, 1 Apr 1997 13:03:11 -0800 (PST) Received: (from sec@localhost) by matrix.42.org (8.6.12/8.6.12) id XAA04676; Tue, 1 Apr 1997 23:03:07 +0200 To: freebsd-hackers@freebsd.org Path: sec From: sec@42.org (Stefan `Sec` Zehl) Newsgroups: muc.lists.freebsd.hackers Subject: Re: Internal clock Date: 1 Apr 1997 23:03:07 +0200 Organization: Internet@home Lines: 14 Message-ID: References: <3340C326.6150@cybernet.com> <19970401151947.61221@pavilion.net> X-Newsreader: slrn (0.9.3.0-2 BETA UNIX) Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In article <19970401151947.61221@pavilion.net>, Josef Karthauser wrote: > BTW a question to the hackers... does anyone have any plans for > adding a real-time scheduling class, aka threads under solaris? > I'm writing some midi software and could really do with a real-time > class thread to handle the timing subsystem. FreeBSD 2.1.0R has rtprio(2) ( /usr/sbin/rtprio ) - works great for me :) CU, Sec -- Fuer die Raupe ist es das Ende der Welt, Fuer den Rest der Welt ist es ein Schmetterling Error 0: No error From owner-freebsd-hackers Tue Apr 1 13:14:15 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA01119 for hackers-outgoing; Tue, 1 Apr 1997 13:14:15 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA01113 for ; Tue, 1 Apr 1997 13:14:14 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id NAA26919; Tue, 1 Apr 1997 13:14:02 -0800 (PST) To: Josef Karthauser cc: gbeach@cybernet.com, hackers@freebsd.org Subject: Re: Internal clock In-reply-to: Your message of "Tue, 01 Apr 1997 15:19:47 +0100." <19970401151947.61221@pavilion.net> Date: Tue, 01 Apr 1997 13:14:02 -0800 Message-ID: <26916.859929242@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > BTW a question to the hackers... does anyone have any plans for > adding a real-time scheduling class, aka threads under solaris? > I'm writing some midi software and could really do with a real-time > class thread to handle the timing subsystem. Hmmm. Have you looked at rtprio(1)? It might be enough for what you want, in a sort of brute-force way. Jordan From owner-freebsd-hackers Tue Apr 1 13:18:37 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA01335 for hackers-outgoing; Tue, 1 Apr 1997 13:18:37 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA01327 for ; Tue, 1 Apr 1997 13:18:34 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id NAA26970; Tue, 1 Apr 1997 13:18:26 -0800 (PST) To: Darren Reed cc: hackers@freebsd.org Subject: Re: 2.2.1 & 2940UW.... In-reply-to: Your message of "Wed, 02 Apr 1997 01:16:17 +1000." <199704011521.HAA06475@freefall.freebsd.org> Date: Tue, 01 Apr 1997 13:18:25 -0800 Message-ID: <26966.859929505@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > which seems to be repeating in some loop. > 2.2.1-RELEASE FreeBSD 2.2.1-RELEASE #0: Tue Apr 1 23:47:31 EST 1997 > > (this is no APril 1 joke) We know. :-) > Is there anyone interested in even trying to fix this ? Extremely. Justin is working on it now - it looks like there are still a few more problems, though he has nonetheless fixed a lot of the pathological cases we ran into before. Please work with Justin on this, he's Mr. Adaptec driver! Jordan From owner-freebsd-hackers Tue Apr 1 13:23:10 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA01640 for hackers-outgoing; Tue, 1 Apr 1997 13:23:10 -0800 (PST) Received: from hda.hda.com (hda-bicnet.bicnet.net [207.198.1.121]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA01633 for ; Tue, 1 Apr 1997 13:23:05 -0800 (PST) Received: (from dufault@localhost) by hda.hda.com (8.8.5/8.8.5) id MAA00473; Tue, 1 Apr 1997 12:45:59 -0500 (EST) From: Peter Dufault Message-Id: <199704011745.MAA00473@hda.hda.com> Subject: Re: Internal clock In-Reply-To: <199704012008.NAA05197@rocky.mt.sri.com> from Nate Williams at "Apr 1, 97 01:08:42 pm" To: nate@mt.sri.com (Nate Williams) Date: Tue, 1 Apr 1997 12:45:57 -0500 (EST) Cc: proff@suburbia.net, nate@mt.sri.com, hackers@FreeBSD.org X-Mailer: ELM [version 2.4ME+ PL25 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > > > > then they fall by as the original authors move onto other projects. > > > > > > Then adding the code into the tree is a 'bad thing', since it becomes > > > unsupported. If no-one is willing to incorporate/support the code, then > > > it shouldn't be incorporated. > > > > > > > That philosophy guarentees one to failure. The code doesn't have > > a chance of being supported till it gets exposure and people start > > relying on it. > > No, someone has to integrate it (hence support it). Nate's right. Jukka Ukonnen's patches were passed around by Jordan on -hackers, and I think sent to gnats around Christmas, so they should be easily found in the archives by searching for "ukonnen". I also have some local hacks that are similar but loaded via an LKM and run against a device for access. I'll put them up on Freefall by tomorrow AM if I can test them against -current. I haven't been eager to commit because I haven't had a -current system until recently, because it should be reexamined against the POSIX spec, have man pages written, peer reviewed, and because personally I want a skeleton for rt kernels in an SMP environment. Peter -- Peter Dufault (dufault@hda.com) Realtime Machine Control and Simulation HD Associates, Inc. Voice: 508 433 6936 From owner-freebsd-hackers Tue Apr 1 13:59:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA04777 for hackers-outgoing; Tue, 1 Apr 1997 13:59:11 -0800 (PST) Received: from nyx.pr.mcs.net (nyx.pr.mcs.net [204.95.55.81]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA04770 for ; Tue, 1 Apr 1997 13:59:07 -0800 (PST) Received: from nyx.pr.mcs.net (localhost [127.0.0.1]) by nyx.pr.mcs.net (8.8.5/8.8.5) with ESMTP id PAA00979; Tue, 1 Apr 1997 15:56:43 -0600 (CST) Message-Id: <199704012156.PAA00979@nyx.pr.mcs.net> X-Mailer: exmh version 1.6.9 8/22/96 To: Brian Somers cc: proff@suburbia.net, dfr@nlsystems.com (Doug Rabson), hackers@freebsd.org Subject: Re: file system layout In-reply-to: Your message of Tue, 01 Apr 1997 21:27:38 +0100. <199704012027.VAA01934@awfulhak.demon.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 01 Apr 1997 15:56:43 -0600 From: Chris Csanady Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk [snip] >I know someone who attempted this on a linux box - seperating each package >into a distinct tree of its own. > >Is unionfs functional in -current ? I read that it was, but I've had no luck. I think it spontaneously rebooted for me. :\ It's really too bad, this would be a wonderful thing to have working.. Chris >-- >Brian , > >Don't _EVER_ lose your sense of humour.... > > From owner-freebsd-hackers Tue Apr 1 14:01:16 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA05200 for hackers-outgoing; Tue, 1 Apr 1997 14:01:16 -0800 (PST) Received: from whizzo.transsys.com (whizzo.TransSys.COM [144.202.42.10]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA05188 for ; Tue, 1 Apr 1997 14:01:12 -0800 (PST) Received: from localhost.transsys.com (localhost.transsys.com [127.0.0.1]) by whizzo.transsys.com (8.8.5/8.7.3) with SMTP id RAA03711; Tue, 1 Apr 1997 17:00:19 -0500 (EST) Message-Id: <199704012200.RAA03711@whizzo.transsys.com> X-Mailer: exmh version 2.0gamma 1/27/96 To: gbeach@cybernet.com cc: hackers@freebsd.org From: "Louis A. Mamakos" Subject: Re: Internal clock References: <3340C326.6150@cybernet.com> In-reply-to: Your message of "Tue, 01 Apr 1997 09:11:18 +0100." <3340C326.6150@cybernet.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 01 Apr 1997 17:00:19 -0500 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > How can I make time measurements in increments smaller than the clock > tick? Specifically, I am running BSD on a Pentium PC which has 128 > clock ticks per second. Is there any way to improve the resolution? The gettimeofday() system call has resolution finer than the 128hz tick rate. If you look at the microtime subroutine in /sys/i386/i386/microtime.s, you'll see some pretty scary code which does this. On non-586/686 cpus, it looks at the counter which fires the 128hz interrupt to see how far along it is since the last interrupt, and uses that to compute the "current" time. The 586/686 case is somewhat more interesting.. The xntpd daemon uses this feature to synchronize the system clock to within a very few milliseconds by timestamping the arrival of packets delivered. As others have mentioned, arranging to wakeup at some arbitrary time in the future is a different thing all together. I think that the scheduling events are either based on the 128hz clock tick, or some other process blocking. louie From owner-freebsd-hackers Tue Apr 1 14:16:37 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA06473 for hackers-outgoing; Tue, 1 Apr 1997 14:16:37 -0800 (PST) Received: from spooky.rwwa.com (rwwa.com [198.115.177.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA06463 for ; Tue, 1 Apr 1997 14:16:30 -0800 (PST) Received: from spooky.rwwa.com (localhost.rwwa.com [127.0.0.1]) by spooky.rwwa.com (8.7.5/8.7.3) with ESMTP id RAA04747 for ; Tue, 1 Apr 1997 17:16:05 -0500 (EST) Message-Id: <199704012216.RAA04747@spooky.rwwa.com> X-Mailer: exmh version 2.0gamma 1/27/96 To: hackers@freebsd.org Subject: Can anyone here help these people? Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 01 Apr 1997 17:16:05 -0500 From: Robert Withrow Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk There is an order of magnitude difference in the performance of postgress between FreeBSD and Linux, with FreeBSD loosing. Here is a ktrace that shows the problem... ------- Forwarded Message Return-Path: scrappy@postgreSQL.org Received: from relay1.shore.net (root@relay1.shore.net [192.233.85.129]) by spooky.rwwa.com (8.7.5/8.7.3) with ESMTP id TAA04928 for ; Thu, 27 Mar 1997 19:59:52 -0500 (EST) Received: from hub.org (root@hub.org [207.107.138.200]) by relay1.shore.net (8.8.3/8.8.3) with ESMTP id SAA19485 for ; Thu, 27 Mar 1997 18:42:28 -0500 (EST) Received: from localhost (daemon@localhost) by hub.org (8.8.2/8.7.5) with SMTP id SAA02298; Thu, 27 Mar 1997 18:39:39 -0500 (EST) Received: by hub.org (bulk_mailer v1.5); Thu, 27 Mar 1997 18:39:04 -0500 Received: (from pgsql@localhost) by hub.org (8.8.2/8.7.5) id SAA02241 for hackers-outgoing; Thu, 27 Mar 1997 18:38:56 -0500 (EST) X-Authentication-Warning: hub.org: pgsql set sender to owner-pg95-dev@postgreSQL.org using -f Received: from phssthpok.illusionary.com (phssthpok.illusionary.com [206.174.15.210]) by hub.org (8.8.2/8.7.5) with ESMTP id SAA02226 for ; Thu, 27 Mar 1997 18:38:46 -0500 (EST) Received: from localhost (dglidden@localhost) by phssthpok.illusionary.com (8.8.5/8.8.3) with SMTP id SAA12047 for ; Thu, 27 Mar 1997 18:38:34 -0500 (EST) Date: Thu, 27 Mar 1997 18:38:33 -0500 (EST) From: Derek Glidden To: hackers@postgreSQL.org Subject: [HACKERS] FreeBSD execution traces Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-pg95-dev@hub.org Reply-To: hackers@hub.org, Derek Glidden Here's a sample of the ktrace from the backend showing one of the first calls of the INSERT query: 247 postgres 0.004788 RET read 8192/0x2000 247 postgres 0.000441 CALL write(0x5,0x158000,0xe) 247 postgres 0.000101 GIO fd 5 wrote 14 bytes "CINSERT 30560\0" 247 postgres 0.000365 RET write 14/0xe 247 postgres 0.000213 CALL read(0x4,0x14a000,0xe000) 247 postgres 0.018655 GIO fd 4 read 3 bytes "Q \0" 247 postgres 0.000082 RET read 3 247 postgres 0.000074 CALL write(0x5,0x158000,0x2) 247 postgres 0.000126 GIO fd 5 wrote 2 bytes "I\0" The number 0.0xxxx is the number of seconds since the previous system call. This is the final end of one of the first INSERT calls made by the client to the backend. Notice that the time between sending the "CINSERT" and receiving the next "Q " is very small. Here's a later call: 247 postgres 0.000182 CALL lseek(0xf,0,0,0,0x2) 247 postgres 0.000069 RET lseek 8192/0x2000 247 postgres 0.000702 CALL write(0x5,0x158000,0xe) 247 postgres 0.000101 GIO fd 5 wrote 14 bytes "CINSERT 30564\0" 247 postgres 0.000046 RET write 14/0xe 247 postgres 0.000190 CALL read(0x4,0x14a000,0xe000) 247 postgres 0.184748 GIO fd 4 read 3 bytes "Q \0" 247 postgres 0.000067 RET read 3 247 postgres 0.000068 CALL write(0x5,0x158000,0x2) 247 postgres 0.000126 GIO fd 5 wrote 2 bytes Notice how long it is before the backend receives the next "Q " after sending the "CINSERT." Now here's what it looks like on the client end: 809 posttest 0.000136 GIO fd 3 wrote 83 bytes "QINSERT INTO Torture VALUES (1, '03-27-1997', 'Data input at 16:00:27 \ 03-27-1997')\0" 809 posttest 0.003786 RET write 83/0x53 809 posttest 0.000124 CALL read(0x4,0x30000,0xe000) 809 posttest 0.000057 GIO fd 4 read 7 bytes "Pblank\0" 809 posttest 0.000034 RET read 7 809 posttest 0.000036 CALL read(0x4,0x30000,0xe000) 809 posttest 0.188054 GIO fd 4 read 14 bytes "CINSERT 42530\0" 809 posttest 0.000050 RET read 14/0xe 809 posttest 0.000056 CALL write(0x3,0x22000,0x3) 809 posttest 0.000126 GIO fd 3 wrote 3 bytes "Q \0" Here it's almost 1/5 of a second between getting the "Pblank" (I don't know what that signifies, maybe it's a clue to something.) after the insert and reading the "CINSERT." (The backend and client dumps are from different runs, so the numbers don't correspond to each other.) It looks like this might actually have something to do with the FreeBSD networking code, because it looks like the backend is writing the message to the socket, but the client isn't getting it until almost 1/5 of a second later. Why INSERT is the only operation that seems to be so slow, though, is beyond me. Perhaps it's just that each INSERT is more or less its own transaction while each SELECT is a single transaction no matter how many tuples it returns? If anyone wants the full trace/dump files, I've got 'em at: http://www.illusionary.com/~dglidden/postgres/ You'll need to either be running FreeBSD and use `kdump` to read the *.trace files, or get the *.dump files which have already been converted to "human-readable" text. - -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Derek Glidden http://www.illusionary.com Web development, databases, graphics and general plumbing Linux, FreeBSD, Apache, PostgreSQL, PHP (Say No to NT!) Unsolicited commercial EMail will be proofread for $500 ea. ------- End of Forwarded Message ----------------------------------------------------------------------------- Robert Withrow, Tel: +1 617 592 8935, Net: witr@rwwa.COM From owner-freebsd-hackers Tue Apr 1 14:17:56 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA06542 for hackers-outgoing; Tue, 1 Apr 1997 14:17:56 -0800 (PST) Received: from hda.hda.com (hda-bicnet.bicnet.net [207.198.1.121]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA06527 for ; Tue, 1 Apr 1997 14:17:46 -0800 (PST) Received: (from dufault@localhost) by hda.hda.com (8.8.5/8.8.5) id NAA00601; Tue, 1 Apr 1997 13:42:04 -0500 (EST) From: Peter Dufault Message-Id: <199704011842.NAA00601@hda.hda.com> Subject: Re: Internal clock In-Reply-To: <199704011745.MAA00473@hda.hda.com> from Peter Dufault at "Apr 1, 97 12:45:57 pm" To: dufault@hda.com (Peter Dufault) Date: Tue, 1 Apr 1997 13:42:03 -0500 (EST) Cc: hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL25 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Jukka Ukonnen's patches were passed around by Jordan on -hackers, > and I think sent to gnats around Christmas, so they should be Sorry - Jukka Ukkonen's patches will be much easily found by searching for Ukkonen than Ukonnen. Peter -- Peter Dufault (dufault@hda.com) Realtime Machine Control and Simulation HD Associates, Inc. Voice: 508 433 6936 From owner-freebsd-hackers Tue Apr 1 14:20:25 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA06872 for hackers-outgoing; Tue, 1 Apr 1997 14:20:25 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id OAA06841 for ; Tue, 1 Apr 1997 14:20:13 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id PAA12164; Tue, 1 Apr 1997 15:02:55 -0700 From: Terry Lambert Message-Id: <199704012202.PAA12164@phaeton.artisoft.com> Subject: Re: Internal clock To: nate@mt.sri.com (Nate Williams) Date: Tue, 1 Apr 1997 15:02:55 -0700 (MST) Cc: proff@suburbia.net, nate@mt.sri.com, hackers@freebsd.org In-Reply-To: <199704012008.NAA05197@rocky.mt.sri.com> from "Nate Williams" at Apr 1, 97 01:08:42 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > That philosophy guarentees one to failure. The code doesn't have > > a chance of being supported till it gets exposure and people start > > relying on it. > > No, someone has to integrate it (hence support it). If, for example, the character device driver to system interface were relatively static (which it could be if it were well designed and thought out in the first place, instead of being incrementally revised every Tuesday), then a character device driver written to that interface would *never* "go stale". It would keep it's utility indefinitely (code can be crufty, yet maintain utility) because, even though it was outdated, it would continue to function as it was intended, without additional maintenance. Interface changes are the predominant cause of maintenance hassles. Arguably, the person changing the interface is responsible for doing the associated maintenance for all consumers of the interface. At least that's what I'm told every time I want ot make VFS changes: that everything that currently works has to keep working. I don't see any reason to not apply this standard uniformly... so even if the interfaces keep mutating instead of settling down (like they should, if they were NP complete and functionally abstracted), there is no good reason to not include code which is not expected to be maintained. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Tue Apr 1 14:20:44 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA06935 for hackers-outgoing; Tue, 1 Apr 1997 14:20:44 -0800 (PST) Received: from yoda.fdt.net (root@yoda.fdt.net [205.229.48.17]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA06905 for ; Tue, 1 Apr 1997 14:20:34 -0800 (PST) Received: from Kryten.nina.org (port-50.ts1.gnv.fdt.net [205.229.51.50]) by yoda.fdt.net with SMTP id RAA21963; Tue, 1 Apr 1997 17:19:29 -0500 Date: Tue, 1 Apr 1997 17:18:54 -0500 (EST) From: Frank Seltzer X-Sender: frankd@Kryten.nina.org To: Thomas David Rivers cc: hackers@freebsd.org Subject: Re: aha2940 problems on 2.1.7.1. In-Reply-To: <199704011118.GAA01239@lakes.water.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Tue, 1 Apr 1997, Thomas David Rivers wrote: > Nope - I'm using tar. > > I would be surprised if your problem simply wasn't time dependent; > and the version of dump you were running wasn't actually the cause. > Were you running 2.2 or 2.2.1 when you saw this? This is with 2.2, I have not upgraded to 2.2.1 yet. > - Thanks - > - Dave Rivers - > > Frank -- Only in America can a homeless veteran sleep in a cardboard box while a draft dodger sleeps in the White House - anonymous From owner-freebsd-hackers Tue Apr 1 14:24:04 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA07337 for hackers-outgoing; Tue, 1 Apr 1997 14:24:04 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id OAA07306 for ; Tue, 1 Apr 1997 14:24:00 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id PAA12178; Tue, 1 Apr 1997 15:06:42 -0700 From: Terry Lambert Message-Id: <199704012206.PAA12178@phaeton.artisoft.com> Subject: Re: Internal clock To: nate@mt.sri.com (Nate Williams) Date: Tue, 1 Apr 1997 15:06:41 -0700 (MST) Cc: terry@lambert.org, nate@mt.sri.com, proff@suburbia.net, hackers@freebsd.org In-Reply-To: <199704012051.NAA05487@rocky.mt.sri.com> from "Nate Williams" at Apr 1, 97 01:51:48 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Software 'engineering' is something I spent significant time studying, > and no matter how good you are maintenance makes up 90% of the 'time' > spent on code for most projects. One could argue that the entire > FreeBSD project is doing 'maintenance' on the CSRG code tree. I'm not disagreeing, I'm just saying that maintenance time should be front-loaded by breaking the problem into cleanly divisable component areas, etc.. > > Really, the issue is one of designing good kernel interfaces, not the > > software that plugs into them. > > Really, the issue of putting a man on Mars is designing a good space > ship, not actually building the darn ship. And verifying the validity of the design... the simplest means being by putting a man on Mars. 8-). Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Tue Apr 1 14:24:09 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA07363 for hackers-outgoing; Tue, 1 Apr 1997 14:24:09 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA07342 for ; Tue, 1 Apr 1997 14:24:05 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id OAA27317; Tue, 1 Apr 1997 14:21:45 -0800 (PST) To: proff@suburbia.net cc: terry@lambert.org (Terry Lambert), joe@pavilion.net, gbeach@cybernet.com, hackers@freebsd.org Subject: Re: Internal clock In-reply-to: Your message of "Wed, 02 Apr 1997 04:34:42 +1000." <19970401183442.3989.qmail@suburbia.net> Date: Tue, 01 Apr 1997 14:21:45 -0800 Message-ID: <27313.859933305@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > The sad reality is if these things are not incorporated in -current > then they fall by as the original authors move onto other projects. The problem is that we need more committer/reviewers who are capable of bringing stuff in at this scale without also breaking everything. That's not as easy as it sounds, and even some of our best current people still break the tree now. :-) Breaking the tree is also a problem because of the amount of tech support it generates. It might not bother the end-user to have to stay away from -current for a few days, but as one of the developers who gets subjected to an endless stream (many messages a day!) of mail saying "current is broken!" "did you know current is broken?" "hey, I just thought I'd let you know that current is broken again. Get your act togther, guys!" every time current is broken, I mind it very very much. Break the tree for 3 days and you've just guaranteed about 200 messages in my inbox (sent to me *directly*, not even through the mailing lists - those account for another 300 or so). If people have begun to wonder at our conservatism, that's one very big reason for it. Jordan From owner-freebsd-hackers Tue Apr 1 14:25:14 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA07431 for hackers-outgoing; Tue, 1 Apr 1997 14:25:14 -0800 (PST) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA07422 for ; Tue, 1 Apr 1997 14:25:10 -0800 (PST) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id PAA09736; Tue, 1 Apr 1997 15:21:25 -0700 (MST) Date: Tue, 1 Apr 1997 15:21:25 -0700 (MST) Message-Id: <199704012221.PAA09736@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Terry Lambert Cc: nate@mt.sri.com (Nate Williams), proff@suburbia.net, hackers@freebsd.org Subject: Re: Internal clock In-Reply-To: <199704012202.PAA12164@phaeton.artisoft.com> References: <199704012008.NAA05197@rocky.mt.sri.com> <199704012202.PAA12164@phaeton.artisoft.com> X-Mailer: VM 6.22 under 19.15 XEmacs Lucid Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > If, for example, the character device driver to system interface > were relatively static (which it could be if it were well designed > and thought out in the first place, instead of being incrementally > revised every Tuesday),.... If 'ifs and butts were candy and nuts, we'd all have a Merry Christmas.' Nate From owner-freebsd-hackers Tue Apr 1 14:37:13 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA08278 for hackers-outgoing; Tue, 1 Apr 1997 14:37:13 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id OAA08264 for ; Tue, 1 Apr 1997 14:37:09 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id AAA06776 for hackers@freebsd.org; Wed, 2 Apr 1997 00:37:01 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id WAA17010; Tue, 1 Apr 1997 22:54:10 +0200 (MET DST) Message-ID: <19970401225410.UY37092@uriah.heep.sax.de> Date: Tue, 1 Apr 1997 22:54:10 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freebsd.org Subject: Re: Code maintenance References: <199704011920.MAA04841@rocky.mt.sri.com> <199704011946.MAA11874@phaeton.artisoft.com> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199704011946.MAA11874@phaeton.artisoft.com>; from Terry Lambert on Apr 1, 1997 12:46:06 -0700 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Terry Lambert wrote: > > Then adding the code into the tree is a 'bad thing', since it becomes > > unsupported. If no-one is willing to incorporate/support the code, then > > it shouldn't be incorporated. > > Better that it be lost forever? Yes, see the ft(4) and aic(4) drivers for two good examples. Both sorely lack a maintainer, and nobody seems to be interested in doing this job. p.s.: Please, avoid Cc'ing me in every followup. One message for the mailing list is enough. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Tue Apr 1 14:38:14 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA08338 for hackers-outgoing; Tue, 1 Apr 1997 14:38:14 -0800 (PST) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA08333 for ; Tue, 1 Apr 1997 14:38:07 -0800 (PST) Received: (from root@localhost) by dyson.iquest.net (8.8.4/8.6.9) id RAA03060; Tue, 1 Apr 1997 17:37:09 -0500 (EST) From: "John S. Dyson" Message-Id: <199704012237.RAA03060@dyson.iquest.net> Subject: Re: file system layout In-Reply-To: <199704012156.PAA00979@nyx.pr.mcs.net> from Chris Csanady at "Apr 1, 97 03:56:43 pm" To: ccsanady@nyx.pr.mcs.net (Chris Csanady) Date: Tue, 1 Apr 1997 17:37:09 -0500 (EST) Cc: brian@awfulhak.org, proff@suburbia.net, dfr@nlsystems.com, hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > [snip] > >I know someone who attempted this on a linux box - seperating each package > >into a distinct tree of its own. > > > >Is unionfs functional in -current ? > > I read that it was, but I've had no luck. I think it spontaneously rebooted > for me. :\ It's really too bad, this would be a wonderful thing to have > working.. > > Chris > > >-- > >Brian , > > > >Don't _EVER_ lose your sense of humour.... > > > > > > > > > From owner-freebsd-hackers Tue Apr 1 14:39:50 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA08414 for hackers-outgoing; Tue, 1 Apr 1997 14:39:50 -0800 (PST) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA08407 for ; Tue, 1 Apr 1997 14:39:45 -0800 (PST) Received: (from root@localhost) by dyson.iquest.net (8.8.4/8.6.9) id RAA03065; Tue, 1 Apr 1997 17:39:09 -0500 (EST) From: "John S. Dyson" Message-Id: <199704012239.RAA03065@dyson.iquest.net> Subject: Re: file system layout In-Reply-To: <199704012156.PAA00979@nyx.pr.mcs.net> from Chris Csanady at "Apr 1, 97 03:56:43 pm" To: ccsanady@nyx.pr.mcs.net (Chris Csanady) Date: Tue, 1 Apr 1997 17:39:09 -0500 (EST) Cc: brian@awfulhak.org, proff@suburbia.net, dfr@nlsystems.com, hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk (Sorry if I sent some email mistakenly -- here is my real reply :-)). > > [snip] > >I know someone who attempted this on a linux box - seperating each package > >into a distinct tree of its own. > > > >Is unionfs functional in -current ? > > I read that it was, but I've had no luck. I think it spontaneously rebooted > for me. :\ It's really too bad, this would be a wonderful thing to have > working.. > I had fully intended to get it working, but with the spectre of the Lite/2 merge, I didn't want to deal with it until afterwards. It is going to be tricky to get it working correctly. The whole issue of VM object sharing and coherency is difficult, and not addressed at all by other existing implementations. John From owner-freebsd-hackers Tue Apr 1 14:48:27 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA09057 for hackers-outgoing; Tue, 1 Apr 1997 14:48:27 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id OAA09052 for ; Tue, 1 Apr 1997 14:48:25 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id PAA12232; Tue, 1 Apr 1997 15:27:42 -0700 From: Terry Lambert Message-Id: <199704012227.PAA12232@phaeton.artisoft.com> Subject: Re: Internal clock To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Tue, 1 Apr 1997 15:27:42 -0700 (MST) Cc: proff@suburbia.net, terry@lambert.org, joe@pavilion.net, gbeach@cybernet.com, hackers@freebsd.org In-Reply-To: <27313.859933305@time.cdrom.com> from "Jordan K. Hubbard" at Apr 1, 97 02:21:45 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > The sad reality is if these things are not incorporated in -current > > then they fall by as the original authors move onto other projects. > > The problem is that we need more committer/reviewers who are capable > of bringing stuff in at this scale without also breaking everything. > That's not as easy as it sounds, and even some of our best current > people still break the tree now. :-) > > Breaking the tree is also a problem because of the amount of tech > support it generates. It might not bother the end-user to have to > stay away from -current for a few days, but as one of the developers > who gets subjected to an endless stream (many messages a day!) of mail > saying "current is broken!" "did you know current is broken?" "hey, I > just thought I'd let you know that current is broken again. Get your > act togther, guys!" every time current is broken, I mind it very very > much. Break the tree for 3 days and you've just guaranteed about 200 > messages in my inbox (sent to me *directly*, not even through the > mailing lists - those account for another 300 or so). > > If people have begun to wonder at our conservatism, that's one very > big reason for it. Do we need to go back to the discussion how a primitive transaction model can make this work and make all your detractors have nothing to detract about? Even at 10% overhead (to be overzealous on the down side) is worth it if you can safely accommodate 11.11% or more additional committers! STATUS: TREE IS BUILDABLE, SELF CONSISTANT BEGIN COMMIT PROCESS CHECK OUT TREE APPLY CHANGES TO CHECKED OUT TREE <---------. BUILD TREE | TREE SUCCESSFULLY BUILT? ------------ no ---' CHECK IN MODIFIED FILES END COMMIT PROCESS STATUS: TREE IS BUILDABLE, SELF CONSISTANT Are you worried about the CVSSup'able tree being buildable at all times? Or CTM delta's? Don't operate against the active tree, operate against a copy, and only make the copy when there isn't a commit process in progress (Implement type non-specific locking). Are you worried about the space for another copy of the tree? Implement mulitple reader/writer locking so you can use the same tree. Are you worried about two committers going at it at the same time? Implement single writer locking against the tree to keep them from stepping on each other's toes, and which the copy/sup participates in. Are you worried about all of these issues simultaneously? Are you worried about allowing concurrent readers? Implement multiple reader/single writer locking. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Tue Apr 1 14:51:25 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA09358 for hackers-outgoing; Tue, 1 Apr 1997 14:51:25 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id OAA09352 for ; Tue, 1 Apr 1997 14:51:21 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id AAA07060 for hackers@freebsd.org; Wed, 2 Apr 1997 00:51:19 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id AAA17424; Wed, 2 Apr 1997 00:49:57 +0200 (MET DST) Message-ID: <19970402004957.UW00921@uriah.heep.sax.de> Date: Wed, 2 Apr 1997 00:49:57 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freebsd.org Subject: Re: DNS & Sendmail References: <199704010420.WAA00706@knight.trosoft.com> <199704012020.VAA01643@awfulhak.demon.co.uk> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199704012020.VAA01643@awfulhak.demon.co.uk>; from Brian Somers on Apr 1, 1997 21:20:17 +0100 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Brian Somers wrote: > > When I kill named, I can start sendmail. > [.....] > > Does FEATURE(nodns) and FEATURE(nocanonify) help ? It shouldn't. Sendmail has no reason to try canonifying anything else than its own hostname on startup. The originator had a named set up for his own stuff, as it seems. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Tue Apr 1 15:03:40 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA10475 for hackers-outgoing; Tue, 1 Apr 1997 15:03:40 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA10457 for ; Tue, 1 Apr 1997 15:03:35 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id PAA27605; Tue, 1 Apr 1997 15:02:19 -0800 (PST) To: Terry Lambert cc: proff@suburbia.net, joe@pavilion.net, gbeach@cybernet.com, hackers@freebsd.org Subject: Re: Internal clock In-reply-to: Your message of "Tue, 01 Apr 1997 15:27:42 MST." <199704012227.PAA12232@phaeton.artisoft.com> Date: Tue, 01 Apr 1997 15:02:19 -0800 Message-ID: <27601.859935739@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Are you worried about all of these issues simultaneously? Are you > worried about allowing concurrent readers? Implement multiple > reader/single writer locking. You forgot about: Implement virtual time generator so that Jordan & co can actually get around to building the "frozen current" mechanism from hell, with duplicate CVS tree and all, along with the new build/relesae system, the setup tool, the fancy administration/help system, the new combined package/distribution scheme and and and... So can I have those virtual time generator plans by next week please? I'm sort of in a hurry. :-) Jordan From owner-freebsd-hackers Tue Apr 1 15:19:43 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA11856 for hackers-outgoing; Tue, 1 Apr 1997 15:19:43 -0800 (PST) Received: from pdx1.world.net (pdx1.world.net [192.243.32.18]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA11827 for ; Tue, 1 Apr 1997 15:19:38 -0800 (PST) Received: from suburbia.net (suburbia.net [203.4.184.1]) by pdx1.world.net (8.7.5/8.7.3) with SMTP id PAA05795 for ; Tue, 1 Apr 1997 15:21:32 -0800 (PST) Received: (qmail 7672 invoked by uid 110); 1 Apr 1997 23:18:13 -0000 MBOX-Line: From richard@a42.deep-thought.org Tue Apr 01 23:01:04 1997 remote from suburbia.net Delivered-To: proff@suburbia.net Received: (qmail 6931 invoked from network); 1 Apr 1997 23:01:03 -0000 Received: from a42.deep-thought.org (203.4.184.227) by suburbia.net with SMTP; 1 Apr 1997 23:01:03 -0000 Received: from a42.deep-thought.org ([127.0.0.1]) by a42.deep-thought.org with esmtp id m0wCCZO-0024vuC (Debian Smail-3.2 1996-Jul-4 #2); Wed, 2 Apr 1997 09:02:26 +1000 (EST) Message-Id: X-Mailer: exmh version 1.6.9 8/22/96 To: proff@suburbia.net Subject: interesting ehternet simulator Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 02 Apr 1997 09:02:25 +1000 From: Richard Jones Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk ------- Forwarded Message Return-Path: owner-linux-net-outgoing@vger.rutgers.edu Return-Path: Received: from nic.funet.fi ([128.214.248.6]) by a42.deep-thought.org with esmtp id m0wC6nY-0024xWa (Debian Smail-3.2 1996-Jul-4 #2); Wed, 2 Apr 1997 02:52:40 +1000 (EST) Received: from vger.rutgers.edu ([128.6.190.2]) by nic.funet.fi with ESMTP id <13932-7333>; Tue, 1 Apr 1997 19:39:16 +0300 Received: by vger.rutgers.edu id <970881-21088>; Tue, 1 Apr 1997 11:36:09 -0500 Date: Tue, 1 Apr 1997 18:41:01 +0200 Message-Id: <199704011641.SAA20432@ns.mad.servicom.es> X-Sender: a2810087@pop.mad.servicom.es X-Mailer: Windows Eudora Pro Version 2.1.2 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: linux-net@vger.rutgers.edu From: jose.nunez@mad.servicom.es Subject: Multi-Ethernet-simulator driver Sender: owner-linux-net@vger.rutgers.edu Precedence: bulk Hi! First of all, I'm not sure if this is the correct mailing list to post this message in, if it isn't just delete and I apologize. I've been working on a Ethernet-simulator driver. It 'creates' 8 ethernet devices which are connected two by two. Each of the devices of each pair can send and receive packets from the other one. It also 'loses' packets from time to time. Transmission and reception statistics are always updated and it also keeps track of the 'lost' packages. The idea behind this driver is to develop some routing and similar practices in university laboratories using only one machine; as not everyone can afford having 4 machines and a router for each student. If anyone wants to test this driver or thinks he might find it useful, drop me a line and I'll send you a copy. Please note that I'm just a student and both my networking & kernel knowledge are quite limited. JLN - -------------------------------------- Jose-Luis Nunez http://www.dat.etsit.upm.es/~jln - -------------------------------------- ------- End of Forwarded Message From owner-freebsd-hackers Tue Apr 1 15:47:26 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA14467 for hackers-outgoing; Tue, 1 Apr 1997 15:47:26 -0800 (PST) Received: from root.com (implode.root.com [198.145.90.17]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA14458 for ; Tue, 1 Apr 1997 15:47:23 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by root.com (8.8.5/8.6.5) with SMTP id PAA20241; Tue, 1 Apr 1997 15:48:34 -0800 (PST) Message-Id: <199704012348.PAA20241@root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: Robert Withrow cc: hackers@freebsd.org Subject: Re: Can anyone here help these people? In-reply-to: Your message of "Tue, 01 Apr 1997 17:16:05 EST." <199704012216.RAA04747@spooky.rwwa.com> From: David Greenman Reply-To: dg@root.com Date: Tue, 01 Apr 1997 15:48:34 -0800 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >There is an order of magnitude difference in the performance of >postgress between FreeBSD and Linux, with FreeBSD loosing. Here >is a ktrace that shows the problem... ... >Here it's almost 1/5 of a second between getting the "Pblank" (I don't >know what that signifies, maybe it's a clue to something.) after the The problem is caused by the Nagle Congestion Avoidence algorithm. See pages 269-270 of TCP/IP Illustrated, Volume 1. The problem can be eliminated by setting the TCP_NODELAY socket option. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-hackers Tue Apr 1 15:54:26 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA14890 for hackers-outgoing; Tue, 1 Apr 1997 15:54:26 -0800 (PST) Received: from softway95.softway.com (softway95.softway.com [206.80.1.38]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA14885; Tue, 1 Apr 1997 15:54:22 -0800 (PST) Received: (from renaud@localhost) by softway95.softway.com (8.8.5/8.6.12) id PAA15669; Tue, 1 Apr 1997 15:28:14 -0800 (PST) From: Renaud Waldura Message-Id: <199704012328.PAA15669@softway95.softway.com> Subject: Apache lingering_close error To: freebsd-isp@freebsd.org, freebsd-hackers@freebsd.org Date: Tue, 1 Apr 1997 15:28:14 -0800 (PST) Reply-To: Renaud Waldura Organization: Softway, Inc. (San Francisco) X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I'm using Apache 1.2b7. With the proxy module, I got lots of errors: [Fri Mar 28 22:42:11 1997] shutdown: Invalid argument [Fri Mar 28 22:42:11 1997] - lingering_close until I added EXTRA_CFLAGS=-DUSE_SO_LINGER -DNO_LINGCLOSE in the Configuration file. It works fine now, although the documentation http://www.apache.org/docs/misc/fin_wait_2.html makes me wonder if this was the Right Thing To Do. It says that FreeBSD 2.1 implements correctly the FIN_WAIT_2 timeout, but does this mean that SO_LINGER is ok too? -- -- Renaud Waldura -- -- Softway International, Inc -- -- rw@softway.com -- -- 185 Berry Street, Suite 5514 -- -- San Francisco, CA 94107 -- -- USA -- -- Tel (415) 896-0708 -- Fax (415) 896-0709 -- http://www.softway.com/ -- -- From owner-freebsd-hackers Tue Apr 1 15:56:25 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA15023 for hackers-outgoing; Tue, 1 Apr 1997 15:56:25 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id PAA14993 for ; Tue, 1 Apr 1997 15:56:16 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id QAA12377; Tue, 1 Apr 1997 16:29:08 -0700 From: Terry Lambert Message-Id: <199704012329.QAA12377@phaeton.artisoft.com> Subject: Re: Internal clock To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Tue, 1 Apr 1997 16:29:08 -0700 (MST) Cc: terry@lambert.org, proff@suburbia.net, joe@pavilion.net, gbeach@cybernet.com, hackers@freebsd.org In-Reply-To: <27601.859935739@time.cdrom.com> from "Jordan K. Hubbard" at Apr 1, 97 03:02:19 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > Are you worried about all of these issues simultaneously? Are you > > worried about allowing concurrent readers? Implement multiple > > reader/single writer locking. > > You forgot about: > > Implement virtual time generator so that Jordan & co can actually get > around to building the "frozen current" mechanism from hell, with > duplicate CVS tree and all, along with the new build/relesae system, > the setup tool, the fancy administration/help system, the new combined > package/distribution scheme and and and... > > So can I have those virtual time generator plans by next week please? > I'm sort of in a hurry. :-) You only need 'the "frozen current" mechanism from hell' if you refuse to consider multiple reader/single writer locking at all. Even then, you will need to implement on top of a similar framewrok (dropping the "writer-but-not-actively-committing" lock states from your graph). Snapshots of the tree are not necessary if you do not let in readers between the time the commit starts and the commit ends. A buildable tree is present any time a writer lock is not present, by definition, if you require that only buildable trees be checked in, and that a writer lock be held over the checkin itself. You only need to hold a writer lock across the whole transaction if you are trying to protect the tree from simultaneous writers. If all you are trying to do is protect readers from writers, you only need to hold the lock across the actuall commit itself. Single writer locking to protect writers from each other is held over the whole transaction. But you don't need to prevent readers between the time the transaction starts and the time the actual commit takes place, since the tree is unstable only *during* the actual commit. So it's: committers: LOCK: SOLE WRITER ... PROMOTE LOCK: SOLE COMMITTER UNLOCK: SOLE COMMITTER readers: LOCK: Nth READER UNLOCK: Nth READER Y wants ---------------------------------------- read write commit -------- -------- -------- X has | none | grant grant grant | read | grant grant block(1)(3) | write | grant block(4) block(1) | commit | block(2) block block(1) -------- -------- -------- (1) You would probably want to block additional "read" requests when there is a blocked commit pending to allow existing "read" requests to "drain" to allow the commit to go through. (2) A blocked read request may be treated as a "deny, try again later". (3) Depending on prioritization, an in-progress "read" might be preempted by a pending "commit". This should result in a "try again later", per note (2), but would require the cooperation of the user's CVSup and/or CTM mirroring. (4) Assumes inter-commiter contention is arbitrated by locking so that one modified tree is not committed on top of another, resulting in a conjoined ther that is unbuildable. If you want to implement 'the "frozen current" mechanism from hell', you can do so by applying the locking system to both trees, with the following algorithm: The "frozen" tree only gets "commit" locked (there is no "write" lock needed, since a "frozen" tree is only refreshed by one "writer", ever: the mirroring process). The "commit" tree only gets a "read" lock (or a "commit" lock if note (3) is implemented, so that a committer can have a write lock pending upgrade to "commit", and the "frozen" tree is not left in an indeterminate state) while the mirroring process is running. Readers and committers operate on their respective trees as if they were still operating on a single tree; the only difference is that the contention window is shortened to only occur during "mirroring". A frozen tree is probably not necessary if CVSup's from the commit tree are only done by CVSup sites doing mirroring rather than by individual users (a relatively low load means a relatively small contention window exists). --- I can write a simple program to do multiple reader/single writer locking like this using standard UNIX lock files and reader/writer counting, if you can't figure it out yourself... it's hardly a problem "from hell", however. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Tue Apr 1 16:02:58 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA15781 for hackers-outgoing; Tue, 1 Apr 1997 16:02:58 -0800 (PST) Received: from alpo.whistle.com (alpo.whistle.com [207.76.204.38]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA15770 for ; Tue, 1 Apr 1997 16:02:46 -0800 (PST) Received: from current1.whistle.com (current1.whistle.com [207.76.205.22]) by alpo.whistle.com (8.8.5/8.8.4) with SMTP id PAA20900 for ; Tue, 1 Apr 1997 15:51:04 -0800 (PST) Message-ID: <33419EB4.41C67EA6@whistle.com> Date: Tue, 01 Apr 1997 15:48:04 -0800 From: Julian Elischer Organization: Whistle Communications X-Mailer: Mozilla 3.0Gold (X11; I; FreeBSD 2.2-CURRENT i386) MIME-Version: 1.0 To: hackers@freebsd.org Subject: oops sorry (mail snafu) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I had a mail snafu when I HALF undid the redirection I had set up for my trip back to AUS. This led to a mail loop it's fixed now I believe sorry if you got bounced.. julian From owner-freebsd-hackers Tue Apr 1 16:07:55 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA16142 for hackers-outgoing; Tue, 1 Apr 1997 16:07:55 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id QAA16122 for ; Tue, 1 Apr 1997 16:07:49 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id QAA12448 for hackers@freebsd.org; Tue, 1 Apr 1997 16:51:39 -0700 From: Terry Lambert Message-Id: <199704012351.QAA12448@phaeton.artisoft.com> Subject: Re: Code maintenance To: hackers@freebsd.org Date: Tue, 1 Apr 1997 16:51:39 -0700 (MST) Reply-To: hackers@freebsd.org In-Reply-To: <19970401225410.UY37092@uriah.heep.sax.de> from "J Wunsch" at Apr 1, 97 10:54:10 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > Then adding the code into the tree is a 'bad thing', since it becomes > > > unsupported. If no-one is willing to incorporate/support the code, then > > > it shouldn't be incorporated. > > > > Better that it be lost forever? > > Yes, see the ft(4) and aic(4) drivers for two good examples. Both > sorely lack a maintainer, and nobody seems to be interested in doing > this job. Has this code suddenly mutated into unusability? Doesn't someone out there have a CDROM of the original, "unmutated" bits so that we can recover them to usability? Or is the problem that someone has changed the kernel interfaces used by them, without living up to the responsibility of changing the software that depends on those interfaces? One does not need to know the driver to change the driver from using one kernel interface (which one must have started with to get the new interface) to using another interface (which one defined). Or, to put it another way: The code for ft(4) and aic(4) did not change, the code used to work, if the code no longer works, it is not the code's fault. If you are going to pee in the public pool (by changing kernel interfaces), have the decency to add some chlorine (by changing the code which depends on those interfaces). > p.s.: Please, avoid Cc'ing me in every followup. One message for the > mailing list is enough. This would be easier if you didn't attach a "Reply-To:" header that specified you instead of the list. 8-(. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Tue Apr 1 16:12:02 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA16413 for hackers-outgoing; Tue, 1 Apr 1997 16:12:02 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA16335; Tue, 1 Apr 1997 16:10:38 -0800 (PST) Received: from dialup-usr11.etinc.com (dialup-usr11.etinc.com [204.141.95.132]) by etinc.com (8.8.3/8.6.9) with SMTP id TAA18745; Tue, 1 Apr 1997 19:14:36 -0500 (EST) Message-Id: <3.0.32.19970401190724.00b1d924@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Tue, 01 Apr 1997 19:07:32 -0500 To: Brian Somers From: dennis Subject: Re: PPP Desperate for help! Cc: Developer , freebsd-current@freebsd.org, freebsd-hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 08:49 PM 4/1/97 +0100, Brian Somers wrote: >> At 01:42 PM 4/1/97 +0100, Developer wrote: >> > >> > >> >On Tue, 1 Apr 1997, Andrzej Bialecki wrote: >> > >> >> On Tue, 1 Apr 1997, Developer wrote: >> >> >> >> > >> >> > We are really stuck trying to get the PPP user program (ppp) to >> connect to >> >> > our Perle 833 dial in server. We can connect using Windows NT fine but on >> >> > BSD ppp logs in and then data will travel in both directions (As I can >> see >> >> > the modem lights flash at both ends for both send/recieve) but no data >> >> > seems to actually get through as pings do not work. The packets from the >> >> > server to the user, but seem to be lost on the way back. >> >> >> >> Let me take a guess: maybe it's a known problem with tcp_extensions set to >> >> YES in /etc/sysconfig? Some broken TCP stacks don't like it, so the >> >> solution would be to turn it off. >> >> >> >> Andy >> > >> >We've tried this - no difference. Any more ideas please? >> >> Why not post a trace of the ppp negotiations? Perhaps there is a parameter >> conflict....are you getting to state IPCP_OPENED? If not, you cant pass >> traffic. >> >> Dennis > >I suspect you're way ahead of the problem. Has the original poster >tried "set openmode active" ? Without this, a client ppp will wait >for the server to initiate LCP. > >I think it may be frugal to make this the default for both client >*and* server. A lot of server implementations wait for the client >to start. Any comments ? It shouldnt make a difference, if implemented properly. There is no master-slave relationship in ppp...the ends are peers. Upon starting the interface it should send "n" configs, and then stop if no reply is received. When a config is received it starts again. The state machine allows for this nicely. One problem is when the electrical interface (UP) is not implemented (ie, receipt of DSR) properly. You should not be able to get into both ends being dormant if its done properly. Dennis From owner-freebsd-hackers Tue Apr 1 16:15:22 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA16740 for hackers-outgoing; Tue, 1 Apr 1997 16:15:22 -0800 (PST) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA16731 for ; Tue, 1 Apr 1997 16:15:17 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by who.cdrom.com (8.8.5/8.6.11) with SMTP id QAA15479 for ; Tue, 1 Apr 1997 16:15:15 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id QAA12469; Tue, 1 Apr 1997 16:55:20 -0700 From: Terry Lambert Message-Id: <199704012355.QAA12469@phaeton.artisoft.com> Subject: Re: Internal clock To: nate@mt.sri.com (Nate Williams) Date: Tue, 1 Apr 1997 16:55:20 -0700 (MST) Cc: terry@lambert.org, nate@mt.sri.com, proff@suburbia.net, hackers@freebsd.org In-Reply-To: <199704012221.PAA09736@rocky.mt.sri.com> from "Nate Williams" at Apr 1, 97 03:21:25 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > If, for example, the character device driver to system interface > > were relatively static (which it could be if it were well designed > > and thought out in the first place, instead of being incrementally > > revised every Tuesday),.... > > If 'ifs and butts were candy and nuts, we'd all have a Merry Christmas.' We *have* a well designed and thought out device interface: devfs. I have *no* idea what is preventing you from making it the default mechanism. (Do you think I would have engaged in this discussion if I did not already have answers to the questions which I would be asking?). Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Tue Apr 1 16:16:21 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA16859 for hackers-outgoing; Tue, 1 Apr 1997 16:16:21 -0800 (PST) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA16852 for ; Tue, 1 Apr 1997 16:16:18 -0800 (PST) Received: from agora.rdrop.com (root@agora.rdrop.com [199.2.210.241]) by who.cdrom.com (8.8.5/8.6.11) with SMTP id QAA15483 for ; Tue, 1 Apr 1997 16:16:17 -0800 (PST) Received: from seagull.rtd.com by agora.rdrop.com with smtp (Smail3.1.29.1 #17) id m0wCDim-0008zKC; Tue, 1 Apr 97 16:16 PST Received: (from dgy@localhost) by seagull.rtd.com (8.7.5/8.7.3) id RAA07641 for freebsd-hackers@freefall.cdrom.com; Tue, 1 Apr 1997 17:12:29 -0700 (MST) From: Don Yuniskis Message-Id: <199704020012.RAA07641@seagull.rtd.com> Subject: Re: Code maintenance To: freebsd-hackers@freefall.FreeBSD.org (FreeBSD hackers) Date: Tue, 1 Apr 1997 17:12:29 -0700 (MST) In-Reply-To: <19970401225410.UY37092@uriah.heep.sax.de> from "J Wunsch" at Apr 1, 97 10:54:10 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk It seems that J Wunsch said: > As Terry Lambert wrote: > > > > Then adding the code into the tree is a 'bad thing', since it becomes > > > unsupported. If no-one is willing to incorporate/support the code, then > > > it shouldn't be incorporated. > > > > Better that it be lost forever? > > Yes, see the ft(4) and aic(4) drivers for two good examples. Both > sorely lack a maintainer, and nobody seems to be interested in doing > this job. I had acquired all the docs to start on the ft driver coincident with my rewrite of my mt driver. However, I've still been unable to find an "unencumbered" source of information re: Pertec interfaces so have not invested any time in the mt driver, either :-( (since I suspect there are somewhat fewer than 0 folks interested in mt! :>) If I manage to find a good information source, I'll do my best to rewrite mine in a "compatible" way... > p.s.: Please, avoid Cc'ing me in every followup. One message for the > mailing list is enough. Consider it... Done! :> --don From owner-freebsd-hackers Tue Apr 1 16:54:07 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA19626 for hackers-outgoing; Tue, 1 Apr 1997 16:54:07 -0800 (PST) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA19620 for ; Tue, 1 Apr 1997 16:54:03 -0800 (PST) Received: from parkplace.cet.co.jp (parkplace.cet.co.jp [202.32.64.1]) by who.cdrom.com (8.8.5/8.6.11) with ESMTP id QAA15540 for ; Tue, 1 Apr 1997 16:53:23 -0800 (PST) Received: from localhost (michaelh@localhost) by parkplace.cet.co.jp (8.8.5/CET-v2.1) with SMTP id AAA22952 for ; Wed, 2 Apr 1997 00:50:19 GMT Date: Wed, 2 Apr 1997 09:50:19 +0900 (JST) From: Michael Hancock To: FreeBSD Hackers Subject: Bugs in libc_r (fwd) Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Some C precedence errors. ---------- Forwarded message ---------- Date: 01 Apr 1997 21:52:03 +0200 From: "Alex Fenyo (eowyn)" To: freebsd-current@freebsd.org Subject: Bugs in libc_r Hello, The same bug is repeated 4 times in libc_r. The problem is that the operator '&' has a lower precedence than '==' in C, and the author of the code forgot it at 4 places in the source : in /usr/src/lib/libc_r/uthread/uthread_read.c : ------------------------------------------------------------ if (_thread_fd_table[fd]->flags & O_NONBLOCK == 0 && (errno == EWOULDBLOCK || errno == EAGAIN)) { ------------------------------------------------------------ should be replaced by : ------------------------------------------------------------ if ((_thread_fd_table[fd]->flags & O_NONBLOCK) == 0 && (errno == EWOULDBLOCK || errno == EAGAIN)) { ------------------------------------------------------------ The same replacement should be done in : uthread_readv.c, uthread_write.c, uthread_writev.c Could someone apply the modifications ? Alexandre Fenyo From owner-freebsd-hackers Tue Apr 1 17:14:13 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA20751 for hackers-outgoing; Tue, 1 Apr 1997 17:14:13 -0800 (PST) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA20745 for ; Tue, 1 Apr 1997 17:14:10 -0800 (PST) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by mail.cdsnet.net (8.8.5/8.7.3) with SMTP id RAA07907 for ; Tue, 1 Apr 1997 17:14:09 -0800 (PST) Date: Tue, 1 Apr 1997 17:14:09 -0800 (PST) From: Jaye Mathisen To: hackers@freebsd.org Subject: de driver problems/ CRC errors with 2.2.1-RELEASE Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Digital HX6000, P6200/512, 32MB RAM, onboard Adaptec 7880, UW's, Toshiba 12X CDROM, 1 2GB seagate Barracude (wide), 1 SMC dual port 10/100 card. This box is hooked to a 3COM Superstack II Switch 3000, (A 100MB switch). Any heavy NFS access (like via iozone on NFS partition) generates kazillions of CRC align errors, at the rate of 30-40/second. (At least, according to the 3com switch, and the output of netstat). lighter access doesn't seem to do it. For example, running bonnie, the putc phase generates oodles of errors, but the rewrite phase generates maybe 1 or 2 per page of output. Any tips appreciated, I've been happy with the de driver, but this is not acceptable or usable. BTW, I took a box with a much older 2.2 kernel, from around October or so, and plugged it in, and zero problems. So it seems like a moderately recent development. From owner-freebsd-hackers Tue Apr 1 17:18:22 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA21147 for hackers-outgoing; Tue, 1 Apr 1997 17:18:22 -0800 (PST) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA21142 for ; Tue, 1 Apr 1997 17:18:19 -0800 (PST) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by mail.cdsnet.net (8.8.5/8.7.3) with SMTP id RAA08608 for ; Tue, 1 Apr 1997 17:18:18 -0800 (PST) Date: Tue, 1 Apr 1997 17:18:18 -0800 (PST) From: Jaye Mathisen To: hackers@freebsd.org Subject: Does de driver do 100MBIT Full Duplex? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I'm not sure exactly how the process works, but the SMC card (dual etherpower 10/100), seems unable to kick into full-duplex mode with my Linkswitch. Does the driver support it? ifconfig shows a "Simplex" flag, but if I remember correctly, that was something to do with if a card could read it's own output on coax or somesuch. From owner-freebsd-hackers Tue Apr 1 17:39:55 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA23108 for hackers-outgoing; Tue, 1 Apr 1997 17:39:55 -0800 (PST) Received: from scanner.worldgate.com (scanner.worldgate.com [198.161.84.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA23103; Tue, 1 Apr 1997 17:39:53 -0800 (PST) Received: from znep.com (uucp@localhost) by scanner.worldgate.com (8.8.5/8.7.3) with UUCP id SAA23555; Tue, 1 Apr 1997 18:39:50 -0700 (MST) Received: from localhost (marcs@localhost) by alive.znep.com (8.7.5/8.7.3) with SMTP id SAA09355; Tue, 1 Apr 1997 18:39:30 -0700 (MST) Date: Tue, 1 Apr 1997 18:39:30 -0700 (MST) From: Marc Slemko To: Renaud Waldura cc: freebsd-isp@freebsd.org, freebsd-hackers@freebsd.org Subject: Re: Apache lingering_close error In-Reply-To: <199704012328.PAA15669@softway95.softway.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Tue, 1 Apr 1997, Renaud Waldura wrote: > > I'm using Apache 1.2b7. With the proxy module, I got lots of errors: > > [Fri Mar 28 22:42:11 1997] shutdown: Invalid argument > [Fri Mar 28 22:42:11 1997] - lingering_close > > until I added > > EXTRA_CFLAGS=-DUSE_SO_LINGER -DNO_LINGCLOSE > > in the Configuration file. > > It works fine now, although the documentation > http://www.apache.org/docs/misc/fin_wait_2.html > makes me wonder if this was the Right Thing To Do. > > It says that FreeBSD 2.1 implements correctly the FIN_WAIT_2 timeout, > but does this mean that SO_LINGER is ok too? There is no reason to disable linging_close for FreeBSD[1]. The message is simply a combination between a warning and a not-yet-removed debugging message; most of the time it pops up simply due to a race condition in the code path; there are a couple of additional issues when using it with the proxy code that I haven't had time to follow, but nothing that hurts more than outputting that message sometimes. It should not hurt anything. We are having some trouble with the lingering_close function (which is why I wrote the above web page...) and are still playing around with a few things. I don't really want to comment on the SO_LINGER part other than to say that you should just use lingering_close in FreeBSD. [1] That is a lie. There are some performance issues that come up when using lingering_close (ie. child processes potentially hang around finishing up a request for longer), but SO_LINGER potentially has even more performance problems. What we really need is a better API, but lacking that lingering_close is the best solution[2]. It _IS_ necessary. [2] For now. This whole issue will go away in 2.0 because then the only overhead will be keeping the socket over, since 2.0 will threaded so the cleanup doesn't have to block a whole process. From owner-freebsd-hackers Tue Apr 1 17:41:34 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA23429 for hackers-outgoing; Tue, 1 Apr 1997 17:41:34 -0800 (PST) Received: from obiwan.aceonline.com.au (obiwan.aceonline.com.au [203.103.90.67]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA23388 for ; Tue, 1 Apr 1997 17:41:20 -0800 (PST) Received: from localhost (adrian@localhost) by obiwan.aceonline.com.au (8.8.5/8.8.5) with SMTP id JAA10850 for ; Wed, 2 Apr 1997 09:32:43 +0800 (WST) Date: Wed, 2 Apr 1997 09:32:43 +0800 (WST) From: Adrian Chadd To: freebsd-hackers@freebsd.org Subject: Re : select() ? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi. I've found the select() call in the kernel source, but I was wondering if there was a library call as a "front-end" to the kernel call ... I can't seem to find it. Also - does 2.2's select() support > 256 FDs out of the box? I've looked at the source briefly, set FD_SETSIZE to 1024, recompiled but it still doesn't work. Thanks, -- Adrian Chadd | UNIX, MS-DOS and Windows ... | (also known as the Good, the bad and the | ugly..) From owner-freebsd-hackers Tue Apr 1 18:56:37 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id SAA29591 for hackers-outgoing; Tue, 1 Apr 1997 18:56:37 -0800 (PST) Received: from root.com (implode.root.com [198.145.90.17]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA29583 for ; Tue, 1 Apr 1997 18:56:34 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by root.com (8.8.5/8.6.5) with SMTP id SAA21240; Tue, 1 Apr 1997 18:57:50 -0800 (PST) Message-Id: <199704020257.SAA21240@root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: Jaye Mathisen cc: hackers@freebsd.org Subject: Re: de driver problems/ CRC errors with 2.2.1-RELEASE In-reply-to: Your message of "Tue, 01 Apr 1997 17:14:09 PST." From: David Greenman Reply-To: dg@root.com Date: Tue, 01 Apr 1997 18:57:50 -0800 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >Digital HX6000, P6200/512, 32MB RAM, onboard Adaptec 7880, UW's, Toshiba >12X CDROM, 1 2GB seagate Barracude (wide), 1 SMC dual port 10/100 card. > >This box is hooked to a 3COM Superstack II Switch 3000, (A 100MB switch). > >Any heavy NFS access (like via iozone on NFS partition) generates >kazillions of CRC align errors, at the rate of 30-40/second. (At least, >according to the 3com switch, and the output of netstat). > >lighter access doesn't seem to do it. > >For example, running bonnie, the putc phase generates oodles of errors, >but the rewrite phase generates maybe 1 or 2 per page of output. > > >Any tips appreciated, I've been happy with the de driver, but this >is not acceptable or usable. > > >BTW, I took a box with a much older 2.2 kernel, from around October or so, >and plugged it in, and zero problems. So it seems like a moderately >recent development. Try forcing half duplex on the switch port. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-hackers Tue Apr 1 18:58:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id SAA29662 for hackers-outgoing; Tue, 1 Apr 1997 18:58:11 -0800 (PST) Received: from root.com (implode.root.com [198.145.90.17]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA29650 for ; Tue, 1 Apr 1997 18:58:03 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by root.com (8.8.5/8.6.5) with SMTP id SAA21261; Tue, 1 Apr 1997 18:59:08 -0800 (PST) Message-Id: <199704020259.SAA21261@root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: Jaye Mathisen cc: hackers@freebsd.org Subject: Re: Does de driver do 100MBIT Full Duplex? In-reply-to: Your message of "Tue, 01 Apr 1997 17:18:18 PST." From: David Greenman Reply-To: dg@root.com Date: Tue, 01 Apr 1997 18:59:08 -0800 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >I'm not sure exactly how the process works, but the SMC card (dual >etherpower 10/100), seems unable to kick into full-duplex mode with my >Linkswitch. > >Does the driver support it? ifconfig shows a "Simplex" flag, but if >I remember correctly, that was something to do with if a card could read >it's own output on coax or somesuch. No, the current de driver in FreeBSD does not support full duplex. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-hackers Tue Apr 1 19:15:56 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id TAA00744 for hackers-outgoing; Tue, 1 Apr 1997 19:15:56 -0800 (PST) Received: from life.eecs.umich.edu (pmchen@life.eecs.umich.edu [141.213.8.32]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id TAA00739 for ; Tue, 1 Apr 1997 19:15:53 -0800 (PST) Received: (from pmchen@localhost) by life.eecs.umich.edu (8.8.5/8.8.0) id WAA03318 for freebsd-hackers@freebsd.org; Tue, 1 Apr 1997 22:15:03 -0500 (EST) Date: Tue, 1 Apr 1997 22:15:03 -0500 (EST) From: "Peter M. Chen" Message-Id: <199704020315.WAA03318@life.eecs.umich.edu> To: freebsd-hackers@freebsd.org Subject: Re: question on buffer cache and VMIO Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >From terry@lambert.org Tue Apr 1 13:35:03 1997 > >This is hardware... is it going to be generally available/purchasable? The only hardware is a battery. The VM write protection is software. Those without batteries need to write to disk (or non-volatile memory); those with batteries (e.g. machine rooms, laptops) can keep data in memory if it survives crashes. >One big issue: I'm not sure how this will interact with the "soft >updates" facility described in the Ganger/Patt paper. This facility >is a planned future feature on nearly everyone's whiteboards. If you >have read the paper, I can discuss the idea of a generic implementation >that is not FS dependent (like the paper's Appendix A UFS example is) >and the ramifications that would have on FS/cache interaction. Soft updates allow you to delay metadata disk writes (just like data disk writes) without losing file system integrity. However, you will lose more metadata in a crash than before. E.g. if you mkdir then crash a few seconds later, soft updates will lose that directory (but in a way that doesn't compromise file system integrity). This is the same tradeoff that's made with data. Think of it as allowing more "data loss" without "integrity loss". Rio (my project on reliable file caches) gives you something different. It allows you to delay disk writes indefinitely, yet without losing any data in a crash. This minimizes both "data loss" and "integrity loss". Of course, Rio only works with a battery, while soft updates work without a battery. >John Dyson and David Greenman are the people to contact about documenting >this in detail. John has responded privately to me. I've been amazed at how helpful and responsive this mailing list has been to a detailed kernel hacking issue. You all deserve a raise. Pete Prof. Peter M. Chen EECS Department, 2225 EECS 1301 Beal Ave. University of Michigan Ann Arbor, MI 48109-2122 (313) 763-4472, fax: (313) 763-4617 pmchen@eecs.umich.edu http://www.eecs.umich.edu/~pmchen/ From owner-freebsd-hackers Tue Apr 1 19:21:49 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id TAA01110 for hackers-outgoing; Tue, 1 Apr 1997 19:21:49 -0800 (PST) Received: from mail13.digital.com (mail13.digital.com [192.208.46.30]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id TAA01099 for ; Tue, 1 Apr 1997 19:21:45 -0800 (PST) Received: from muggsy.lkg.dec.com by mail13.digital.com (8.7.5/UNX 1.5/1.0/WV) id WAA18715; Tue, 1 Apr 1997 22:14:02 -0500 (EST) Received: from tunsrv2-tunnel.imc.das.dec.com by muggsy.lkg.dec.com (5.65/DEC-Ultrix/4.3) with SMTP id AA26887; Tue, 1 Apr 1997 22:13:57 -0500 Message-Id: <3.0.1.32.19970401221054.00769af8@alphy.lkg.dec.com> X-Sender: popmatt@alphy.lkg.dec.com X-Mailer: Windows Eudora Pro Version 3.0.1 (32) Date: Tue, 01 Apr 1997 22:10:54 -0500 To: Jaye Mathisen , hackers@freebsd.org From: Matt Thomas Subject: Re: Does de driver do 100MBIT Full Duplex? In-Reply-To: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 05:18 PM 4/1/97 -0800, Jaye Mathisen wrote: > > >I'm not sure exactly how the process works, but the SMC card (dual >etherpower 10/100), seems unable to kick into full-duplex mode with my >Linkswitch. Is that the 9334bdt? What version of FreeBSD? >Does the driver support it? ifconfig shows a "Simplex" flag, but if >I remember correctly, that was something to do with if a card could read >it's own output on coax or somesuch. Correct. Note the driver in NetBSD is in much better shape than the current FreeBSD driver. (it isn't hard to get the NetBSD driver running under FreeBSD). -- Matt Thomas Internet: matt@3am-software.com 3am Software Foundry WWW URL: http://www.3am-software.com/bio/matt.html Westford, MA Disclaimer: I disavow all knowledge of this message From owner-freebsd-hackers Tue Apr 1 20:18:26 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA03907 for hackers-outgoing; Tue, 1 Apr 1997 20:18:26 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA03902 for ; Tue, 1 Apr 1997 20:18:24 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id UAA28840; Tue, 1 Apr 1997 20:17:05 -0800 (PST) To: Terry Lambert cc: nate@mt.sri.com (Nate Williams), proff@suburbia.net, hackers@FreeBSD.ORG Subject: Re: Internal clock In-reply-to: Your message of "Tue, 01 Apr 1997 16:55:20 MST." <199704012355.QAA12469@phaeton.artisoft.com> Date: Tue, 01 Apr 1997 20:17:05 -0800 Message-ID: <28836.859954625@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > We *have* a well designed and thought out device interface: devfs. Well thought-out, yes. Finished, no. :) > I have *no* idea what is preventing you from making it the default > mechanism. 1. It has no persistance and a number of people have already told me directly (some of them at USENIX, immediately following Julian's talk on DEVFS) that we will be roundly castigated and screamed at in the various public forums for breaking /dev's historical behavior if we don't make persistance work and work *completely transparently*. Let's say somebody has a "securify" script that they run on a new system as part of their security auditing process. It's been running on FreeBSD systems since 2.0, making sure that tape drives and CDROMs and such all have sane perms (or maybe even some sort of "insane" perms which are part of a more elaborate ownership scheme which they like), and now suddenly with FreeBSD 4.0 or whatever they find that everything the script does on these new systems is undone after a reboot. Say what?? Sure, one can stick something in /etc/rc.local which does everything from beating on /dev to translating the catpages into Urdu, but the fact remains that someone shouldn't need to do that. As you yourself say, things should stay *backwards compatible*. :-) 2. It tends to panic the system fairly regularly on steady use. Why on earth would we make something this dangerous the default? It's simply not finished yet, and there are numerous open PRs against it in the PR database (#2033, #2034, #2738 and possibly #1211). Jordan From owner-freebsd-hackers Tue Apr 1 21:01:29 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id VAA05735 for hackers-outgoing; Tue, 1 Apr 1997 21:01:29 -0800 (PST) Received: from rosie.scsn.net (scsn.net [206.25.246.12]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id VAA05728 for ; Tue, 1 Apr 1997 21:01:26 -0800 (PST) Received: from cola134.scsn.net ([206.25.247.134]) by rosie.scsn.net (Post.Office MTA v3.0 release (Release 114) ID# 0-32322U5000L100S10000) with ESMTP id AAA135 for ; Tue, 1 Apr 1997 23:55:55 -0500 Received: (from root@localhost) by cola134.scsn.net (8.8.5/8.6.12) id XAA00487 for hackers@FreeBSD.org; Tue, 1 Apr 1997 23:59:53 -0500 (EST) From: "Donald J. Maddox" Message-Id: <199704020459.XAA00487@cola134.scsn.net> Subject: `find . -nogroup` doesn't work? To: hackers@FreeBSD.org Date: Tue, 1 Apr 1997 23:59:53 -0500 (EST) Reply-To: dmaddox@scsn.net X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk I was attempting to use find(1) to locate files/dirs on my system that have bogus owners/groups, but it appears that the '-nogroup' and '-noowner' options don't work.. Below is a partial listing of the files in my ghostscript fonts dir. You will notice that there are a number of entries with unknown owners/groups: # ls -l total 3620 -rw-r--r-- 1 root wheel 27617 Jun 28 1996 a010013l.afm -rw-r--r-- 1 root wheel 25994 Jun 28 1996 a010013l.pfb -rw-r--r-- 1 root wheel 4679 Jun 28 1996 a010013l.pfm -rw-r--r-- 1 root wheel 27601 Jun 28 1996 a010015l.afm -rw-r--r-- 1 root wheel 26980 Jun 28 1996 a010015l.pfb -rw-r--r-- 1 root wheel 4682 Jun 28 1996 a010015l.pfm -rw-r--r-- 1 root wheel 27756 Jun 28 1996 a010033l.afm -rw-r--r-- 1 root wheel 26882 Jun 28 1996 a010033l.pfb -rw-r--r-- 1 root wheel 4683 Jun 28 1996 a010033l.pfm -rw-r--r-- 1 root wheel 27796 Jun 28 1996 a010035l.afm -rw-r--r-- 1 root wheel 27530 Jun 28 1996 a010035l.pfb -rw-r--r-- 1 root wheel 4686 Jun 28 1996 a010035l.pfm -rw-r--r-- 1 root wheel 27608 Jun 28 1996 b018012l.afm -rw-r--r-- 1 root wheel 35522 Jun 28 1996 b018012l.pfb -rw-r--r-- 1 root wheel 4681 Jun 28 1996 b018012l.pfm -rw-r--r-- 1 root wheel 27432 Jun 28 1996 b018015l.afm -rw-r--r-- 1 root wheel 35453 Jun 28 1996 b018015l.pfb -rw-r--r-- 1 root wheel 4685 Jun 28 1996 b018015l.pfm -rw-r--r-- 1 root wheel 27920 Jun 28 1996 b018032l.afm -rw-r--r-- 1 root wheel 34617 Jun 28 1996 b018032l.pfb -rw-r--r-- 1 root wheel 4685 Jun 28 1996 b018032l.pfm -rw-r--r-- 1 root wheel 27680 Jun 28 1996 b018035l.afm -rw-r--r-- 1 root wheel 35218 Jun 28 1996 b018035l.pfb -rw-r--r-- 1 root wheel 4689 Jun 28 1996 b018035l.pfm -rw-r--r-- 1 bin bin 21626 Aug 2 1994 bchb.afm -rw-r--r-- 1 2461 2461 66591 Apr 24 1992 bchb.pfa -rw-r--r-- 1 2461 2461 22099 Aug 2 1994 bchbi.afm -rw-r--r-- 1 2461 2461 69478 Apr 24 1992 bchbi.pfa -rw-r--r-- 1 2461 2461 19764 Aug 2 1994 bchr.afm -rw-r--r-- 1 2461 2461 68769 Apr 24 1992 bchr.pfa -rw-r--r-- 1 2461 2461 19927 Aug 2 1994 bchri.afm -rw-r--r-- 1 2461 2461 69259 Apr 24 1992 bchri.pfa -rw-r--r-- 1 root wheel 27576 Jun 28 1996 c059013l.afm -rw-r--r-- 1 root wheel 37561 Jun 28 1996 c059013l.pfb -rw-r--r-- 1 root wheel 4665 Jun 28 1996 c059013l.pfm -rw-r--r-- 1 root wheel 27510 Jun 28 1996 c059016l.afm -rw-r--r-- 1 root wheel 38763 Jun 28 1996 c059016l.pfb -rw-r--r-- 1 root wheel 4665 Jun 28 1996 c059016l.pfm -rw-r--r-- 1 root wheel 27665 Jun 28 1996 c059033l.afm -rw-r--r-- 1 root wheel 38413 Jun 28 1996 c059033l.pfb -rw-r--r-- 1 root wheel 4665 Jun 28 1996 c059033l.pfm -rw-r--r-- 1 root wheel 27665 Jun 28 1996 c059036l.afm -rw-r--r-- 1 root wheel 38742 Jun 28 1996 c059036l.pfb -rw-r--r-- 1 root wheel 4669 Jun 28 1996 c059036l.pfm -rw-r--r-- 1 root wheel 9294 Jun 28 1996 d050000l.afm -rw-r--r-- 1 root wheel 44537 Jun 28 1996 d050000l.pfb -rw-r--r-- 1 root wheel 674 Jun 28 1996 d050000l.pfm -rw-r--r-- 1 2461 2461 36031 Sep 21 1993 fcyr.gsf -rw-r--r-- 1 2461 2461 44518 Sep 21 1993 fcyri.gsf -rw-r--r-- 1 2461 2461 38906 Sep 21 1993 fhirw.gsf -rw-r--r-- 1 2461 2461 40226 Sep 21 1993 fkarw.gsf -rw-r--r-- 1 2461 2461 58276 Oct 1 1993 hrger.pfa -rw-r--r-- 1 2461 2461 807 Sep 30 1993 hrgerb.gsf -rw-r--r-- 1 2461 2461 811 Sep 30 1993 hrgerd.gsf -rw-r--r-- 1 2461 2461 812 Sep 30 1993 hrgero.gsf -rw-r--r-- 1 2461 2461 8943 Oct 1 1993 hrgkc.gsf -rw-r--r-- 1 2461 2461 7015 Oct 1 1993 hrgks.gsf -rw-r--r-- 1 2461 2461 57602 Oct 1 1993 hrgrr.pfa However, this is the output of 'find . -nogroup -print': # find . -nogroup -print ./bchb.pfa Why does it find only the first file? The behavior of the '-nouser' option is exactly the same. Donald J. Maddox (dmaddox@scsn.net) From owner-freebsd-hackers Tue Apr 1 21:30:16 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id VAA07010 for hackers-outgoing; Tue, 1 Apr 1997 21:30:16 -0800 (PST) Received: from austin.polstra.com (austin.polstra.com [206.213.73.10]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id VAA07001 for ; Tue, 1 Apr 1997 21:30:13 -0800 (PST) Received: from austin.polstra.com (jdp@localhost) by austin.polstra.com (8.8.5/8.8.5) with ESMTP id VAA04918; Tue, 1 Apr 1997 21:29:52 -0800 (PST) Message-Id: <199704020529.VAA04918@austin.polstra.com> To: pmchen@eecs.umich.edu Subject: Re: question on buffer cache and VMIO Newsgroups: polstra.freebsd.hackers In-Reply-To: <199704020315.WAA03318@life.eecs.umich.edu> References: <199704020315.WAA03318@life.eecs.umich.edu> Organization: Polstra & Co., Seattle, WA Cc: hackers@freebsd.org Date: Tue, 01 Apr 1997 21:29:52 -0800 From: John Polstra Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > I've been amazed at how helpful and > responsive this mailing list has been to a detailed kernel hacking issue. > You all deserve a raise. Heh. Send us a nickel and it'll be the biggest raise we've ever gotten. :-) -- John Polstra jdp@polstra.com John D. Polstra & Co., Inc. Seattle, Washington USA "Self-knowledge is always bad news." -- John Barth From owner-freebsd-hackers Tue Apr 1 21:57:47 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id VAA08287 for hackers-outgoing; Tue, 1 Apr 1997 21:57:47 -0800 (PST) Received: from scanner.worldgate.com (scanner.worldgate.com [198.161.84.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id VAA08282 for ; Tue, 1 Apr 1997 21:57:44 -0800 (PST) Received: from znep.com (uucp@localhost) by scanner.worldgate.com (8.8.5/8.7.3) with UUCP id WAA04005; Tue, 1 Apr 1997 22:57:40 -0700 (MST) Received: from localhost (marcs@localhost) by alive.znep.com (8.7.5/8.7.3) with SMTP id WAA10946; Tue, 1 Apr 1997 22:57:29 -0700 (MST) Date: Tue, 1 Apr 1997 22:57:28 -0700 (MST) From: Marc Slemko To: dmaddox@scsn.net cc: hackers@FreeBSD.org Subject: Re: `find . -nogroup` doesn't work? In-Reply-To: <199704020459.XAA00487@cola134.scsn.net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk On Tue, 1 Apr 1997, Donald J. Maddox wrote: > > I was attempting to use find(1) to locate files/dirs on > my system that have bogus owners/groups, but it appears > that the '-nogroup' and '-noowner' options don't work.. > > Below is a partial listing of the files in my ghostscript > fonts dir. You will notice that there are a number of entries > with unknown owners/groups: Looks to me like a bug in the group_from_gid and user_from_uid functions in pwcache.c. They set the user name to be equal to the uid if it can't be found, but when using a cached name they completely disregard the second (nouser) parameter and return it anyway. This breaks find and looks like a bug in the library functions. To fix this, I think it would either need a flag that is set when the cp->name is generated from cp->uid (can't do a comparison later, since nothing says "111" can't be a valid user name for user 111, right?) or not put it the uid into cp->name if it isn't found and do the snprintf each time it is needed if called with nouser == 0. This problem annoyed me too, but haven't had time to do anything about it. From owner-freebsd-hackers Tue Apr 1 22:01:57 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id WAA08469 for hackers-outgoing; Tue, 1 Apr 1997 22:01:57 -0800 (PST) Received: from werple.net.au (melb.werple.net.au [203.9.190.18]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id WAA08463 for ; Tue, 1 Apr 1997 22:01:38 -0800 (PST) Received: (qmail 28671 invoked by uid 5); 2 Apr 1997 06:01:01 -0000 Received: (from jb@localhost) by freebsd1.cimlogic.com.au (8.7.5/8.7.3) id PAA12654; Wed, 2 Apr 1997 15:57:06 +1000 (EST) From: John Birrell Message-Id: <199704020557.PAA12654@freebsd1.cimlogic.com.au> Subject: Re: Bugs in libc_r (fwd) To: michaelh@cet.co.jp (Michael Hancock) Date: Wed, 2 Apr 1997 15:57:05 +1000 (EST) Cc: Hackers@freebsd.org In-Reply-To: from Michael Hancock at "Apr 2, 97 09:50:19 am" X-Mailer: ELM [version 2.4ME+ PL22 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Michael Hancock wrote: > Some C precedence errors. [...] I fixed these this morning, plus a few other things. 8-) -- John Birrell - jb@cimlogic.com.au; jb@netbsd.org; jb@freebsd.org CIMlogic Pty Ltd, 119 Cecil Street, South Melbourne Vic 3205, Australia Tel +61 3 9690 6900 Fax +61 3 9690 6650 Mob +61 418 353 137 From owner-freebsd-hackers Tue Apr 1 22:05:51 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id WAA08802 for hackers-outgoing; Tue, 1 Apr 1997 22:05:51 -0800 (PST) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id WAA08794 for ; Tue, 1 Apr 1997 22:05:44 -0800 (PST) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by mail.cdsnet.net (8.8.5/8.7.3) with SMTP id WAA29778 for ; Tue, 1 Apr 1997 22:05:42 -0800 (PST) Date: Tue, 1 Apr 1997 22:05:42 -0800 (PST) From: Jaye Mathisen To: hackers@freebsd.org Subject: Mildly interesting NFS numbers. Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Client is a Digital Prioris HX6000, P6-200/512cache, 32MB RAM, SMC 100Mbit ethernet. Server is a Network Appliance F540, 100GB disk, 8MB NVRAM, 128MB cache, 100MBit ethernet. I tried both TCP and UDP mounts, version 3 and version 2 NFS. The best performance was UDP version 3, for what that's worth. # iozone 500 Writing the 500 Megabyte file, 'iozone.tmp'...86.304688 seconds Reading the file...90.789062 seconds IOZONE performance measurements: 6074849 bytes/second for writing the file 5774792 bytes/second for reading the file using 64k block sizes resulted in about a 4% perf increase, however, it dropped CPU utilization on the Netapp from 100% to about 65%. Here's some bonnie numbers as well. Note that for bonnie, the best combo was UDP version 2, rather than version 3. Wish I knew why. In a nutshell, almost as fast as the fastest single spindles. # bonnie -s 300 File './Bonnie.4599', size: 314572800 Writing with putc()...done Rewriting...done Writing intelligently...done Reading with getc()...done Reading intelligently...done Seeker 3...Seeker 1...Seeker 2...start 'em...done...done...done... -------Sequential Output-------- ---Sequential Input-- --Random-- -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks--- Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU 300 5471 62.4 5810 12.9 1104 3.1 4471 52.1 5895 7.5 233.5 10.1 From owner-freebsd-hackers Tue Apr 1 22:23:38 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id WAA09723 for hackers-outgoing; Tue, 1 Apr 1997 22:23:38 -0800 (PST) Received: (from mpp@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id WAA09717; Tue, 1 Apr 1997 22:23:35 -0800 (PST) From: Mike Pritchard Message-Id: <199704020623.WAA09717@freefall.freebsd.org> Subject: Re: `find . -nogroup` doesn't work? To: dmaddox@scsn.net Date: Tue, 1 Apr 1997 22:23:34 -0800 (PST) Cc: hackers@FreeBSD.org In-Reply-To: <199704020459.XAA00487@cola134.scsn.net> from "Donald J. Maddox" at Apr 1, 97 11:59:53 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk Donald J. Maddox wrote: > > > I was attempting to use find(1) to locate files/dirs on > my system that have bogus owners/groups, but it appears > that the '-nogroup' and '-noowner' options don't work.. > > Below is a partial listing of the files in my ghostscript > fonts dir. You will notice that there are a number of entries > with unknown owners/groups: This is a libc problem. The user_from_{uid,gid} routines returned the wrong value when a non-existent user was in their internal caches. I just fixed it in rev 1.5 of libc/gen/pwcache.c. -- Mike Pritchard mpp@FreeBSD.org "Go that way. Really fast. If something gets in your way, turn" From owner-freebsd-hackers Tue Apr 1 23:03:05 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA11058 for hackers-outgoing; Tue, 1 Apr 1997 23:03:05 -0800 (PST) Received: from awfulhak.demon.co.uk (awfulhak.demon.co.uk [158.152.17.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA11049 for ; Tue, 1 Apr 1997 23:03:00 -0800 (PST) Received: from awfulhak.demon.co.uk (localhost.lan.awfulhak.org [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id HAA18005; Wed, 2 Apr 1997 07:54:47 +0100 (BST) Message-Id: <199704020654.HAA18005@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) cc: hackers@freebsd.org Subject: Re: DNS & Sendmail In-reply-to: Your message of "Wed, 02 Apr 1997 00:49:57 +0200." <19970402004957.UW00921@uriah.heep.sax.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 02 Apr 1997 07:54:46 +0100 From: Brian Somers Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As J"org wrote: > As Brian Somers wrote: > > > > When I kill named, I can start sendmail. > > [.....] > > > > Does FEATURE(nodns) and FEATURE(nocanonify) help ? > > It shouldn't. Sendmail has no reason to try canonifying anything else > than its own hostname on startup. The originator had a named set up > for his own stuff, as it seems. But these help tremendously if you have a SMART_HOST, *even* if you have full resolver capabilities. I agree though, it almost definitely won't hurt startup (unless you can't resolv your SMART_HOST or something). -- Brian , Don't _EVER_ lose your sense of humour.... From owner-freebsd-hackers Tue Apr 1 23:18:41 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA11834 for hackers-outgoing; Tue, 1 Apr 1997 23:18:41 -0800 (PST) Received: from siesta.cs.wustl.edu (nw1@siesta.cs.wustl.edu [128.252.165.3]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id XAA11829 for ; Tue, 1 Apr 1997 23:18:39 -0800 (PST) Received: from localhost by siesta.cs.wustl.edu (SMI-8.6/ECL-J1.00) id BAA07658; Wed, 2 Apr 1997 01:18:18 -0600 Message-Id: <199704020718.BAA07658@siesta.cs.wustl.edu> To: John Birrell cc: Hackers@freebsd.org Subject: Re: Bugs in libc_r (fwd) In-reply-to: Your message of "Wed, 02 Apr 1997 15:57:05 +1000." <199704020557.PAA12654@freebsd1.cimlogic.com.au> Date: Wed, 02 Apr 1997 01:18:17 -0600 From: Nanbor Wang Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Michael Hancock wrote: > > Some C precedence errors. > [...] > > I fixed these this morning, plus a few other things. 8-) Anyway we can get an update of this fix? (I don't want to get into -current at this moment ;-) nw > > -- > John Birrell - jb@cimlogic.com.au; jb@netbsd.org; jb@freebsd.org > CIMlogic Pty Ltd, 119 Cecil Street, South Melbourne Vic 3205, Australia > Tel +61 3 9690 6900 Fax +61 3 9690 6650 Mob +61 418 353 137 From owner-freebsd-hackers Tue Apr 1 23:22:59 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA12207 for hackers-outgoing; Tue, 1 Apr 1997 23:22:59 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id XAA12191 for ; Tue, 1 Apr 1997 23:22:56 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id JAA02929 for hackers@freebsd.org; Wed, 2 Apr 1997 09:22:54 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id JAA23046; Wed, 2 Apr 1997 09:04:34 +0200 (MET DST) Message-ID: <19970402090434.CQ60183@uriah.heep.sax.de> Date: Wed, 2 Apr 1997 09:04:34 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freebsd.org Subject: Re: Internal clock References: <199704012221.PAA09736@rocky.mt.sri.com> <199704012355.QAA12469@phaeton.artisoft.com> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199704012355.QAA12469@phaeton.artisoft.com>; from Terry Lambert on Apr 1, 1997 16:55:20 -0700 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Terry Lambert wrote: > We *have* a well designed and thought out device interface: devfs. > > I have *no* idea what is preventing you from making it the default > mechanism. We still miss Terry's bugfixes for it. If you had ever used it (with options DEVFS_ROOT, it's already there), you knew what the bugs are. Terry, the first 90 % of the work are done with it, that's right. We are now in the phase that takes the second 90 % of the development time... p.s.: No personal Cc's, please... -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Tue Apr 1 23:23:28 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA12267 for hackers-outgoing; Tue, 1 Apr 1997 23:23:28 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id XAA12255 for ; Tue, 1 Apr 1997 23:23:24 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id JAA02936 for hackers@freebsd.org; Wed, 2 Apr 1997 09:23:22 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id JAA23078; Wed, 2 Apr 1997 09:12:09 +0200 (MET DST) Message-ID: <19970402091209.IH24661@uriah.heep.sax.de> Date: Wed, 2 Apr 1997 09:12:09 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freebsd.org Subject: Re: Code maintenance References: <19970401225410.UY37092@uriah.heep.sax.de> <199704012351.QAA12448@phaeton.artisoft.com> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199704012351.QAA12448@phaeton.artisoft.com>; from Terry Lambert on Apr 1, 1997 16:51:39 -0700 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Terry Lambert wrote: > > Yes, see the ft(4) and aic(4) drivers for two good examples. Both > > sorely lack a maintainer, and nobody seems to be interested in doing > > this job. > > > Has this code suddenly mutated into unusability? No. It never worked excitingly well, but nobody bothered to fix the problems after the first imports. Also, in the ft(4) case, the hardware on the market evolved _a lot_ after the initial driver, making it only working for ancient crap nowadays. Terry, get real. Kernel interface changes have been the least part of the problem, and look into the CVS history, were often the only reason why someone has touched the source of that unmaintained code at all. And, one of these kernel interface changes should be well known to you: the inclusion of DEVFS. :) -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Tue Apr 1 23:28:04 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA12424 for hackers-outgoing; Tue, 1 Apr 1997 23:28:04 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id XAA12419 for ; Tue, 1 Apr 1997 23:28:01 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id JAA02963; Wed, 2 Apr 1997 09:27:18 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id JAA23103; Wed, 2 Apr 1997 09:20:28 +0200 (MET DST) Message-ID: <19970402092028.KM07161@uriah.heep.sax.de> Date: Wed, 2 Apr 1997 09:20:28 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: freebsd-hackers@freebsd.org Cc: adrian@obiwan.aceonline.com.au (Adrian Chadd) Subject: Re: Re : select() ? References: X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: ; from Adrian Chadd on Apr 2, 1997 09:32:43 +0800 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Adrian Chadd wrote: > I've found the select() call in the kernel source, but I was wondering if > there was a library call as a "front-end" to the kernel call ... I can't > seem to find it. Most of the syscall stubs in the library are being created automatically. > Also - does 2.2's select() support > 256 FDs out of the box? I think it supports any size. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Tue Apr 1 23:55:09 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA13608 for hackers-outgoing; Tue, 1 Apr 1997 23:55:09 -0800 (PST) Received: from pdx1.world.net (pdx1.world.net [192.243.32.18]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA13601 for ; Tue, 1 Apr 1997 23:55:06 -0800 (PST) From: proff@suburbia.net Received: from suburbia.net (suburbia.net [203.4.184.1]) by pdx1.world.net (8.7.5/8.7.3) with SMTP id XAA13035 for ; Tue, 1 Apr 1997 23:57:22 -0800 (PST) Received: (qmail 19723 invoked by uid 110); 2 Apr 1997 07:54:33 -0000 Message-ID: <19970402075433.19722.qmail@suburbia.net> Subject: devfs In-Reply-To: <28836.859954625@time.cdrom.com> from "Jordan K. Hubbard" at "Apr 1, 97 08:17:05 pm" To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Wed, 2 Apr 1997 17:54:33 +1000 (EST) Cc: hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > We *have* a well designed and thought out device interface: devfs. > > Well thought-out, yes. Finished, no. :) > > > I have *no* idea what is preventing you from making it the default > > mechanism. > > 1. It has no persistance and a number of people have already told me directly > (some of them at USENIX, immediately following Julian's talk on DEVFS) > that we will be roundly castigated and screamed at in the various > public forums for breaking /dev's historical behavior if we don't make > persistance work and work *completely transparently*. Let's say > somebody has a "securify" script that they run on a new system as part > of their security auditing process. It's been running on FreeBSD > systems since 2.0, making sure that tape drives and CDROMs and such > all have sane perms (or maybe even some sort of "insane" perms which are > part of a more elaborate ownership scheme which they like), and now > suddenly with FreeBSD 4.0 or whatever they find that everything the > script does on these new systems is undone after a reboot. Say what?? > > Jordan Your arguments about state are very true. I see a few ways around them. 1) accept that devfs in it's present form is a kludge. Move all the default permissions and device numbers into a sysctl hierarchy. This hierarchy is then used to create names in /dev. Add a new ufs file type: S_IFLNKDEV. When a namei occurs on what this pseudo-symlink points to, it looks up name in the sysctl and returns the inode of the "symlink" with device numbers filled in. The mode, ownership, time stamps, and everything but the major/minor numbers are taken from the LNKDEV entry itself. In effect this just replaces major/minor numbers with a string. 2) code a light "permission union" file system. this file system type would work like unionfs, but with file permissions only. Mount devfs on /dev and the permission union over the top. When ever permissions are changed from the default an entry is created in the upper layer. Problem: multiple permission views of individual devices are not supported using this scheme. devfs is an ill-thought out concept. If the desire is to return device driver status information then, then sysctl is the appropriate venue. If the desire is to bypass numeric (major/minor) numbering limitations then 1) is more appropriate. Cheers, Julian. From owner-freebsd-hackers Wed Apr 2 00:15:40 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA14714 for hackers-outgoing; Wed, 2 Apr 1997 00:15:40 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA14709 for ; Wed, 2 Apr 1997 00:15:37 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id AAA00278; Wed, 2 Apr 1997 00:15:18 -0800 (PST) To: proff@suburbia.net cc: hackers@freebsd.org Subject: Re: devfs In-reply-to: Your message of "Wed, 02 Apr 1997 17:54:33 +1000." <19970402075433.19722.qmail@suburbia.net> Date: Wed, 02 Apr 1997 00:15:18 -0800 Message-ID: <273.859968918@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > 1) accept that devfs in it's present form is a kludge. Move > all the default permissions and device numbers into a > sysctl hierarchy. This hierarchy is then used to create > names in /dev. Add a new ufs file type: S_IFLNKDEV. > When a namei occurs on what this pseudo-symlink points > to, it looks up name in the sysctl and returns the inode > of the "symlink" with device numbers filled in. The I'm not sure if I care much for this as a solution to the persistance problem, at least not unless you were also talking about having DEVFS manipulate this piece of the sysctl namespace on behalf of the user for any modifications they may make. It has to be transparent after setup, right? Also, I don't think that anyone has addressed the issue of sysctl MIB persistance either yet, so your changes would still go away on reboot. :-) Still, what you *do* raise here is an interesting, possibly even downright evil, idea (and I like the evil ones best ;). How about a new symlink type which is just like a classic symlink in every way except that its value, rather than being direct fodder for namei(), is instead used to look something up in sysctl, the value of THAT being then passed on to namei()? You could get variant symlink behavior by poking things into sysctl. :-) > 2) code a light "permission union" file system. this file > system type would work like unionfs, but with file permissions > only. Mount devfs on /dev and the permission union > over the top. When ever permissions are changed from the > default an entry is created in the upper layer. That idea has the most promise, I think, even with its limitations. People like Sean Eric Fagan have also suggested that devfs attempt to open a file at the mountpoint before proceeding to mount the devfs over it. If it succeeds, it's now got a handy "hidden" file handle in which to write a little transaction log to. When you remount it later, it sees the file already existing, does a quick sanity check on it and reapplies all the state (leaving it open for append afterwards, of course). Sort of like a journaling devfs or something. :-) That would also give you the multiple view model for different mount points. Jordan From owner-freebsd-hackers Wed Apr 2 00:36:20 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA15535 for hackers-outgoing; Wed, 2 Apr 1997 00:36:20 -0800 (PST) Received: from trifork.gu.net (trifork.gu.net [194.93.190.194]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA15530 for ; Wed, 2 Apr 1997 00:36:16 -0800 (PST) Received: from localhost (localhost.gu.kiev.ua [127.0.0.1]) by trifork.gu.net (8.8.5/8.8.5) with SMTP id LAA00385; Wed, 2 Apr 1997 11:35:10 +0300 (EEST) Date: Wed, 2 Apr 1997 11:35:10 +0300 (EEST) From: Andrew Stesin Reply-To: stesin@gu.net To: John Birrell cc: Michael Hancock , Hackers@freebsd.org Subject: Re: Bugs in libc_r (fwd) In-Reply-To: <199704020557.PAA12654@freebsd1.cimlogic.com.au> Message-ID: X-NCC-RegID: ua.gu MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Wed, 2 Apr 1997, John Birrell wrote: > > Some C precedence errors. > [...] > > I fixed these this morning, plus a few other things. 8-) Did the fix go into a 2.2 branch? Best regards, Andrew Stesin nic-hdl: ST73-RIPE From owner-freebsd-hackers Wed Apr 2 01:14:00 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA17164 for hackers-outgoing; Wed, 2 Apr 1997 01:14:00 -0800 (PST) Received: from ilink.nis.za (root@ilink.nis.za [196.6.121.10]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id BAA17158 for ; Wed, 2 Apr 1997 01:13:55 -0800 (PST) Received: (from keith4@localhost) by ilink.nis.za (8.6.12/8.6.6) id LAA03056 for freebsd-hackers@freebsd.org; Wed, 2 Apr 1997 11:13:45 +0200 From: keith waters Message-Id: <199704020913.LAA03056@ilink.nis.za> Subject: Unreasonable rebuild time on large password files To: freebsd-hackers@freebsd.org Date: Wed, 2 Apr 1997 11:13:44 +0200 (SAT) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi all.. We've got rather a lot of users in our /etc/passwd and every time we change or add a password/login, we have to wait over TEN MINUTES for the 'database to rebuild' [And yes, we have a pretty reasonably powerful machine] This must be one of the worst cases of code/resourse optimisation, since changing a password means rebuilding EVERY SINGLE entry in the password databases. Surely this can all be done with one quick write to the relevant files? Does anybody out there have a working solution ? Please! Thanks Regards, Keith Waters (Interlink Internet Services) [Please preferably use keith@nis.za] From owner-freebsd-hackers Wed Apr 2 01:18:02 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA17383 for hackers-outgoing; Wed, 2 Apr 1997 01:18:02 -0800 (PST) Received: from fgate.flevel.co.uk (root@fgate.flevel.co.uk [194.6.101.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA17378; Wed, 2 Apr 1997 01:17:58 -0800 (PST) Received: from localhost (dev@localhost) by fgate.flevel.co.uk (8.8.3/8.6.9) with SMTP id KAA09917; Wed, 2 Apr 1997 10:17:54 +0100 (BST) Date: Wed, 2 Apr 1997 10:17:54 +0100 (BST) From: Developer To: freebsd-current@freebsd.org, freebsd-hackers@freebsd.org Subject: PPP Problems -- The next chapter *grin* Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk After not being able to get the user ppp software to work I decided to give up and try pppd instead. After configuring correctly pppd works fine (Just use kermit to dial and then run it)?? I wonder what is different between ppp and pppd that would cause the problem? Regards, Trefor S. From owner-freebsd-hackers Wed Apr 2 01:51:37 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA19351 for hackers-outgoing; Wed, 2 Apr 1997 01:51:37 -0800 (PST) Received: from saba.kuentos.guam.net (root@saba.kuentos.guam.net [198.81.233.14]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id BAA19343 for ; Wed, 2 Apr 1997 01:51:29 -0800 (PST) Received: by saba.kuentos.guam.net (Smail3.1.29.1 #9) id m0wCMhP-002EuNC; Wed, 2 Apr 97 19:51 GST Date: Wed, 2 Apr 1997 19:51:23 +1000 (GST) From: T-I-GG-Rrr!! To: freebsd-hackers@freebsd.org Subject: unsubscribe Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk unsubscribe meltedice@kuentos.guam.net =*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*= =* Bill Shaw *= =* Kuentos Communications Inc. Guam USA *= =* Guams First Internet Provider *= =* meltedice@kuentos.guam.net *= =* http://www.guam.net/home/bhshaw2 *= =* IFORMS Chat Run On BSDI *= =* telnet buri.kuentos.guam.net 3000 *= =* *= =*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*= From owner-freebsd-hackers Wed Apr 2 02:03:37 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id CAA20000 for hackers-outgoing; Wed, 2 Apr 1997 02:03:37 -0800 (PST) Received: from trifork.gu.net (trifork.gu.net [194.93.190.194]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id CAA19976; Wed, 2 Apr 1997 02:03:21 -0800 (PST) Received: from localhost (localhost.gu.kiev.ua [127.0.0.1]) by trifork.gu.net (8.8.5/8.8.5) with SMTP id NAA01228; Wed, 2 Apr 1997 13:03:26 +0300 (EEST) Date: Wed, 2 Apr 1997 13:03:25 +0300 (EEST) From: Andrew Stesin Reply-To: stesin@gu.net To: hackers@freebsd.org, scsi@freebsd.org Subject: 2 PCI busses, 2 AIC chips, 2.2.1. Howto ? Message-ID: X-NCC-RegID: ua.gu MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk The box: Intel XXpress-, 2 AIC7870 chips onboard, EISA bus present. neither of AICs are recognized at boot. ??? I have the following guesses: 1. neither of the 2 SCSI busses has anything attached yet (disk drives will come in a day or two, they ordered drives separately, and no Wide drives here to try). 2. SCSI BIOSes for 2 chips aren't configured correctly (just now it has "BIOS enabled" in SCSIconfig for the first chip and "BIOS disabled" for the second, should be Ok?) 3. 2.2.1 has some problem with two PCI busses? (while both AICs are on the second bus??? or on EISA???) What I see at boot after the "Probing devices on the pci:0 bus" message: chip0 \ rev 2 on pci0:0 chip1 rev 5 on pci0:14 pci0:15:0: Intel Corporation device=0x0008, class=0xff, subclass=0x00 \ [no driver assigned] ...^^^^.... This last message is repeated exactly for all pci0:15:[0-7] values. What I am missing? Any comments and explanations are appreciated! Thanks a lot for your help! Best regards, Andrew Stesin nic-hdl: ST73-RIPE From owner-freebsd-hackers Wed Apr 2 02:16:50 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id CAA20829 for hackers-outgoing; Wed, 2 Apr 1997 02:16:50 -0800 (PST) Received: from werple.net.au (melb.werple.net.au [203.9.190.18]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id CAA20809 for ; Wed, 2 Apr 1997 02:16:44 -0800 (PST) Received: (qmail 20825 invoked by uid 5); 2 Apr 1997 10:16:39 -0000 Received: (from jb@localhost) by freebsd1.cimlogic.com.au (8.7.5/8.7.3) id UAA13233; Wed, 2 Apr 1997 20:15:08 +1000 (EST) From: John Birrell Message-Id: <199704021015.UAA13233@freebsd1.cimlogic.com.au> Subject: Re: Bugs in libc_r (fwd) To: stesin@gu.net Date: Wed, 2 Apr 1997 20:15:08 +1000 (EST) Cc: jb@cimlogic.com.au, michaelh@cet.co.jp, Hackers@freebsd.org In-Reply-To: from Andrew Stesin at "Apr 2, 97 11:35:10 am" X-Mailer: ELM [version 2.4ME+ PL22 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Andrew Stesin wrote: > On Wed, 2 Apr 1997, John Birrell wrote: > > > Some C precedence errors. > > [...] > > > > I fixed these this morning, plus a few other things. 8-) > > Did the fix go into a 2.2 branch? No. I don't have a 2.2 system (yet) to test it on. I'll mail diffs to hackers in reply to another message. Perhaps someone using 2.2 can try them? > > Best regards, > Andrew Stesin > > nic-hdl: ST73-RIPE Regards, -- John Birrell - jb@cimlogic.com.au; jb@netbsd.org; jb@freebsd.org CIMlogic Pty Ltd, 119 Cecil Street, South Melbourne Vic 3205, Australia Tel +61 3 9690 6900 Fax +61 3 9690 6650 Mob +61 418 353 137 From owner-freebsd-hackers Wed Apr 2 02:41:31 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id CAA22048 for hackers-outgoing; Wed, 2 Apr 1997 02:41:31 -0800 (PST) Received: from silvia.HIP.Berkeley.EDU (wck-ca16-19.ix.netcom.com [207.94.231.83]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id CAA22043; Wed, 2 Apr 1997 02:41:25 -0800 (PST) Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.8.5/8.6.9) id CAA03124; Wed, 2 Apr 1997 02:41:05 -0800 (PST) Date: Wed, 2 Apr 1997 02:41:05 -0800 (PST) Message-Id: <199704021041.CAA03124@silvia.HIP.Berkeley.EDU> To: stesin@gu.net CC: hackers@freebsd.org, scsi@freebsd.org In-reply-to: (message from Andrew Stesin on Wed, 2 Apr 1997 13:03:25 +0300 (EEST)) Subject: Re: 2 PCI busses, 2 AIC chips, 2.2.1. Howto ? From: asami@vader.cs.berkeley.edu (Satoshi Asami) Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk * 1. neither of the 2 SCSI busses has anything attached yet (disk drives will * come in a day or two, they ordered drives separately, and no Wide * drives here to try). This shouldn't matter. * 2. SCSI BIOSes for 2 chips aren't configured correctly (just now * it has "BIOS enabled" in SCSIconfig for the first chip and * "BIOS disabled" for the second, should be Ok?) This shouldn't matter either. You will never have to enable BIOS unless you plan to boot from a disk on that bus. * 3. 2.2.1 has some problem with two PCI busses? (while both AICs are * on the second bus??? or on EISA???) What I see at boot * after the "Probing devices on the pci:0 bus" message: It might be that it can't recognize the PCI-PCI bridge. What do the markings on the chip say? I know FreeBSD can deal with DEC and IBM chips. * chip0 \ * rev 2 on pci0:0 * chip1 rev 5 on pci0:14 * pci0:15:0: Intel Corporation device=0x0008, class=0xff, subclass=0x00 \ * [no driver assigned] Boot with "-v" and send the output to se@freebsd.org. You may want to hack /sys/pci/pcisupport.c by yourself (grep for "IBM") to see if you can get it to work. * ....^^^^.... * This last message is repeated exactly for all pci0:15:[0-7] * values. This is quite weird though. Sounds like a multi-function chip misprobed. Satoshi From owner-freebsd-hackers Wed Apr 2 02:50:33 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id CAA22578 for hackers-outgoing; Wed, 2 Apr 1997 02:50:33 -0800 (PST) Received: from werple.net.au (melb.werple.net.au [203.9.190.18]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id CAA22567 for ; Wed, 2 Apr 1997 02:50:27 -0800 (PST) Received: (qmail 23712 invoked by uid 5); 2 Apr 1997 10:50:24 -0000 Received: (from jb@localhost) by freebsd1.cimlogic.com.au (8.7.5/8.7.3) id UAA13376; Wed, 2 Apr 1997 20:49:55 +1000 (EST) From: John Birrell Message-Id: <199704021049.UAA13376@freebsd1.cimlogic.com.au> Subject: Re: Bugs in libc_r (fwd) To: nw1@cs.wustl.edu (Nanbor Wang) Date: Wed, 2 Apr 1997 20:49:54 +1000 (EST) Cc: jb@cimlogic.com.au, Hackers@freebsd.org In-Reply-To: <199704020718.BAA07658@siesta.cs.wustl.edu> from Nanbor Wang at "Apr 2, 97 01:18:17 am" X-Mailer: ELM [version 2.4ME+ PL22 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Nanbor Wang wrote: > > Michael Hancock wrote: > > > Some C precedence errors. > > [...] > > > > I fixed these this morning, plus a few other things. 8-) > > Anyway we can get an update of this fix? (I don't want to get into > -current at this moment ;-) I've checked the libc_r source that was tagged for 2.2 and there have been a lot of updates to uthread files since then. Enough to prevent patching the 2.2 tree (despite what I said earlier 8-). Sorry. We'll need to be in better shape for 3.0. Regards, -- John Birrell - jb@cimlogic.com.au; jb@netbsd.org; jb@freebsd.org CIMlogic Pty Ltd, 119 Cecil Street, South Melbourne Vic 3205, Australia Tel +61 3 9690 6900 Fax +61 3 9690 6650 Mob +61 418 353 137 From owner-freebsd-hackers Wed Apr 2 03:32:28 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id DAA24147 for hackers-outgoing; Wed, 2 Apr 1997 03:32:28 -0800 (PST) Received: from florence.pavilion.net (florence.pavilion.net [194.242.128.25]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id DAA24141 for ; Wed, 2 Apr 1997 03:32:22 -0800 (PST) Received: (from joe@localhost) by florence.pavilion.net (8.8.5/8.8.5) id MAA01784; Wed, 2 Apr 1997 12:31:12 +0100 (BST) Message-ID: <19970402123111.58518@pavilion.net> Date: Wed, 2 Apr 1997 12:31:11 +0100 From: Josef Karthauser To: Joerg Wunsch Cc: hackers@FreeBSD.ORG, gbeach@cybernet.com Subject: Re: Internal clock References: <3340C326.6150@cybernet.com> <19970401151947.61221@pavilion.net> <19970401203429.XR37988@uriah.heep.sax.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.64 In-Reply-To: <19970401203429.XR37988@uriah.heep.sax.de>; from J Wunsch on Tue, Apr 01, 1997 at 08:34:29PM +0200 Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Tue, Apr 01, 1997 at 08:34:29PM +0200, J Wunsch wrote: > As Josef Karthauser wrote: > > > I'm using gettimeofday which returns: > > > > struct timeval { > > long tv_sec; /* seconds since Jan. 1, 1970 */ > > long tv_usec; /* and microseconds */ > > }; > > > > I've no idea of the actual system resolution, ... > > RTFM clocks(7). I have read this, although I still wasn't too clear about the above. The fact that a system clock ticks in useconds doesn't mean that there's going to be an interrupt every usec. I didn't get a chance to look at the source either, as it was a late night hacking midi code. > > BTW a question to the hackers... does anyone have any plans for > > adding a real-time scheduling class, aka threads under solaris? > > There's already a pseudo-realtime scheduling class available, see > rtprio(1). This has, of course, nothing to do with threads at all, > nor will it be true realtime, since there's still the problem that > scheduling doesn't happen while running in kernel context (i.e., the > kernel cannot be preempted -- Solaris seems to have gone great lengths > to create some preemption points for long-time running kernel > functions). I found the solaris way a bit of a nightmare. I wanted to set up a cyclic midi in buffer, but unfortunately the serial devices seemed to run somewhere below the real time class :( I'm now reviving the old solaris project and porting it to FreeBSD (that now being my platform of choice ;) [and I can drive midi directly, instead of out of a 19200k serial port into an Amiga which did the baud rate translation.] Joe -- Josef Karthauser Technical Manager Email: joe@pavilion.net Pavilion Internet plc. [Tel: +44 1273 607072 Fax: +44 1273 607073] From owner-freebsd-hackers Wed Apr 2 04:19:05 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id EAA25903 for hackers-outgoing; Wed, 2 Apr 1997 04:19:05 -0800 (PST) Received: from shadows.aeon.net (bsdhack@shadows.aeon.net [194.100.41.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id EAA25880 for ; Wed, 2 Apr 1997 04:18:58 -0800 (PST) Received: (from bsdhack@localhost) by shadows.aeon.net (8.8.5/8.8.3) id PAA03426; Wed, 2 Apr 1997 15:19:20 +0300 (EET DST) From: mika ruohotie Message-Id: <199704021219.PAA03426@shadows.aeon.net> Subject: Re: p166 vs. p166mmx In-Reply-To: <33414845.51AD@db-net.com> from Wilson MacGyver at "Apr 1, 97 12:39:17 pm" To: macgyver@db-net.com (Wilson MacGyver) Date: Wed, 2 Apr 1997 15:19:20 +0300 (EET DST) Cc: hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > make -k world 7595.59s user 1843.94s system 87% cpu 3:00:48.28 total > > make -k world 5982.25s user 1636.02s system 87% cpu 2:24:21.88 total > Wouldn't this be because of the larger L1 cache? If memory > serves, L1 cache was increased from 16K to 32K... yes, but still i didnt expect the difference to be about 21% for user and 17% for total time... mickey From owner-freebsd-hackers Wed Apr 2 06:05:15 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA00957 for hackers-outgoing; Wed, 2 Apr 1997 06:05:15 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA00944 for ; Wed, 2 Apr 1997 06:05:06 -0800 (PST) Received: from dbws.etinc.com (dbws.etinc.com [204.141.95.130]) by etinc.com (8.8.3/8.6.9) with SMTP id JAA23279; Wed, 2 Apr 1997 09:08:02 -0500 (EST) Message-Id: <3.0.32.19970402091819.00695824@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Wed, 02 Apr 1997 09:18:26 -0500 To: Matt Thomas , Jaye Mathisen , hackers@FreeBSD.ORG From: dennis Subject: Re: Does de driver do 100MBIT Full Duplex? Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk At 10:10 PM 4/1/97 -0500, Matt Thomas wrote: >At 05:18 PM 4/1/97 -0800, Jaye Mathisen wrote: >> >> >>I'm not sure exactly how the process works, but the SMC card (dual >>etherpower 10/100), seems unable to kick into full-duplex mode with my >>Linkswitch. > >Is that the 9334bdt? What version of FreeBSD? > >>Does the driver support it? ifconfig shows a "Simplex" flag, but if >>I remember correctly, that was something to do with if a card could read >>it's own output on coax or somesuch. > >Correct. Note the driver in NetBSD is in much better shape than the >current FreeBSD driver. (it isn't hard to get the NetBSD driver running >under FreeBSD). >-- >Matt Thomas Internet: matt@3am-software.com >3am Software Foundry WWW URL: http://www.3am-software.com/bio/matt.html >Westford, MA Disclaimer: I disavow all knowledge of this message > Is the driver that you send me directly for 2.1.7 (that worked beautifully) in 2.2.1R? If not, WHY NOT! It seems that the 10/100 clone cards we have, that worked with that driver, do not work in 2.2.1R Dennis From owner-freebsd-hackers Wed Apr 2 06:15:51 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA01410 for hackers-outgoing; Wed, 2 Apr 1997 06:15:51 -0800 (PST) Received: from cheops.anu.edu.au (avalon@cheops.anu.edu.au [150.203.76.24]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA01403 for ; Wed, 2 Apr 1997 06:15:47 -0800 (PST) Message-Id: <199704021415.GAA01403@freefall.freebsd.org> Received: by cheops.anu.edu.au (1.37.109.16/16.2) id AA295340235; Thu, 3 Apr 1997 00:10:35 +1000 From: Darren Reed Subject: Re: devfs To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Thu, 3 Apr 1997 00:10:35 +1000 (EST) Cc: proff@suburbia.net, hackers@freebsd.org In-Reply-To: <273.859968918@time.cdrom.com> from "Jordan K. Hubbard" at Apr 2, 97 00:15:18 am X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In some mail from Jordan K. Hubbard, sie said: > > > 1) accept that devfs in it's present form is a kludge. Move > > all the default permissions and device numbers into a > > sysctl hierarchy. This hierarchy is then used to create > > names in /dev. Add a new ufs file type: S_IFLNKDEV. > > When a namei occurs on what this pseudo-symlink points > > to, it looks up name in the sysctl and returns the inode > > of the "symlink" with device numbers filled in. The > > I'm not sure if I care much for this as a solution to the persistance > problem, at least not unless you were also talking about having DEVFS > manipulate this piece of the sysctl namespace on behalf of the user > for any modifications they may make. It has to be transparent after > setup, right? > > Also, I don't think that anyone has addressed the issue of sysctl MIB > persistance either yet, so your changes would still go away on reboot. :-) > > Still, what you *do* raise here is an interesting, possibly even > downright evil, idea (and I like the evil ones best ;). How about a > new symlink type which is just like a classic symlink in every way > except that its value, rather than being direct fodder for namei(), is > instead used to look something up in sysctl, the value of THAT being > then passed on to namei()? You could get variant symlink behavior by > poking things into sysctl. :-) OSx5.1 (ye olde syvr3/4.2bsd mix) from Pyramid nhad a "conditional" symbolic link (if I remember right) worked something like this: ln -c att /.attbin -c ucb /.ucbbin /bin where 'att" and "ucb" where two "universes". Depending on which one you were in at the time (process property) /bin was either /.attbin or /.ucbbin. I think FreeBSD should start out with devfs under a different / directory, maybe /devices (;). Let devfs co-exist with /dev, getting a feel for it over time, and at some point, start moving things over to /devices using symbolic links. Eventually, /dev becomes a symlink to /devices. Or... if only name & permission persistance is sought, why can't devfs inherit permissions from /dev and just change major/minor #'s (or is persistance required there too) ? maybe mounting devfs needs to be a boot "procedure" where name->permissions ownership is transfered from /dev to /devices. Hmmm, what if /dev was a sym. link to /dev.old (or was reset to that at boot time), then after the devfs block was mounted under /devices and attributes transfered, changed to point to /devices ? darren From owner-freebsd-hackers Wed Apr 2 06:29:56 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA02063 for hackers-outgoing; Wed, 2 Apr 1997 06:29:56 -0800 (PST) Received: from trifork.gu.net (trifork.gu.net [194.93.190.194]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA02034; Wed, 2 Apr 1997 06:29:48 -0800 (PST) Received: from localhost (localhost.gu.kiev.ua [127.0.0.1]) by trifork.gu.net (8.8.5/8.8.5) with SMTP id RAA02371; Wed, 2 Apr 1997 17:29:42 +0300 (EEST) Date: Wed, 2 Apr 1997 17:29:42 +0300 (EEST) From: Andrew Stesin Reply-To: stesin@gu.net To: Satoshi Asami cc: hackers@freebsd.org, scsi@freebsd.org Subject: Re: 2 PCI busses, 2 AIC chips, 2.2.1. Howto ? In-Reply-To: <199704021041.CAA03124@silvia.HIP.Berkeley.EDU> Message-ID: X-NCC-RegID: ua.gu MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, thanks for comments. On Wed, 2 Apr 1997, Satoshi Asami wrote: > * 3. 2.2.1 has some problem with two PCI busses? (while both AICs are > * on the second bus??? or on EISA???) What I see at boot > * after the "Probing devices on the pci:0 bus" message: > > It might be that it can't recognize the PCI-PCI bridge. What do the > markings on the chip say? I know FreeBSD can deal with DEC and IBM > chips. Unfortunately (or vice versa? ;) I have no access to the box' internals, I'm not authorized to open the case and look inside. But: the box is "genuine Intel", the model is Magellan W/15", base board (kinda of backplane) is XXpress 15" Rel 2, with 1 P166 CPU module. I'll try to find what the bridge chip is. But what is the chance that _both_ AIC chips are "on the other side of the bridge", anyway? > * chip0 \ > * rev 2 on pci0:0 > * chip1 rev 5 on pci0:14 > * pci0:15:0: Intel Corporation device=0x0008, class=0xff, subclass=0x00 \ > * [no driver assigned] > > Boot with "-v" and send the output to se@freebsd.org. You may want to > hack /sys/pci/pcisupport.c by yourself (grep for "IBM") to see if you > can get it to work. Yep, I got the idea. > * ....^^^^.... > * This last message is repeated exactly for all pci0:15:[0-7] > * values. > > This is quite weird though. Sounds like a multi-function chip > misprobed. Gmmm... > > Satoshi > Best regards, Andrew Stesin nic-hdl: ST73-RIPE From owner-freebsd-hackers Wed Apr 2 06:57:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA03095 for hackers-outgoing; Wed, 2 Apr 1997 06:57:11 -0800 (PST) Received: from seagull.rtd.com (seagull.rtd.com [198.102.68.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA03090 for ; Wed, 2 Apr 1997 06:57:07 -0800 (PST) Received: (from dgy@localhost) by seagull.rtd.com (8.7.5/8.7.3) id HAA04618; Wed, 2 Apr 1997 07:57:00 -0700 (MST) From: Don Yuniskis Message-Id: <199704021457.HAA04618@seagull.rtd.com> Subject: Re: wd and funky geometry (understood?) To: bde@zeta.org.au (Bruce Evans) Date: Wed, 2 Apr 1997 07:57:00 -0700 (MST) Cc: freebsd-hackers@freefall.FreeBSD.org (FreeBSD hackers) In-Reply-To: <199703171346.AAA13488@godzilla.zeta.org.au> from "Bruce Evans" at Mar 18, 97 00:46:39 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk It seems that Bruce Evans said: > > > The problem appears to an inconsistent use of geometry information > >for each drive. During IPL, the drive is INITed with parameters from the > >BIOS's disk parameter table. This makes sense as I can't imagine any > >*other* place from which to get the disk geometry (reliably). From > >boot() in /usr/src/sys/i386/boot/biosboot/boot.c: > >... > > However, in wdattach(), wdgetctlr() is invoked to get the drive's > >geometry. It does this by issuing an IDENTIFY command (i.e. READP) > >to the drive. The data returned by the command is interpreted as > >a wdparams structure: > >... > >Herein lies the problem: these parameters reflect "the default geometry of > >the device" (paraphrased from the ATA specification). However, the BIOS's > >geometry may *NOT* agree with the drives "default geometry". For example, > >the probe of my IDE drives yields: > > > > wdc0 at 0x1f0-0x1f7 irq 14 on isa > > wdc0: unit 0 (wd0): > > wd0: 329MB (675450 sectors), 790 cyls, 15 heads, 57 S/T, 512 B/S > > wdc0: unit 1 (wd1): > > wd1: 321MB (659232 sectors), 654 cyls, 16 heads, 63 S/T, 512 B/S > > > >However, *both* drives have a BIOS geometry of 611/16/63. (Sorry, > >my ancient BIOS does not support "user configurable" geometries > >so this was the closest geometry I could find -- hence the reason > > What's the problem? The BIOS uses the fudged geometry that you told it, > but the FreeBSD driver knows better - there is no reason for it to use > a fudged geometry. *If* either (i.e. DOS vs. FBSD) lives alone on the disk, then fine. However, if they cohabitate, then they must agree on "what is where" on the disk. If DOS talks to the disk using a particular geometry, sets up it's partitions using that geometry, etc. then FBSD -- when it uses a different geometry -- must be sure to not stomp on those parts of the disk that DOS is using, etc. Does the drive (since that's the only "third party" involved here) guarantee that a sector that is *used* in one geometry is never regarded as "free" in another geometry (no, don't interpret this *literally*. my point is, if DOS uses a set of sectors in a given geometry, how is FBSD supposed to know that those sectors are "in use" when it adopts a different geometry to lay out it's slices?) > >Why is all this a problem? I don't *think* it causes any problems > >during IPL -- since the drive is INITed with the BIOS geometry > >during the boot and then switched to the "default geometry" > >during the attach/open. I think that the image loaded from the > >disk during boot is small enough that changes to the size/shape > >of the first track are safely ignored. (though, on second thought, > >I suspect this *can* screw you if your FBSD partition does not > >begin at the start of the disk! Comments?) > > > >However, sharing a partition might be a problem since DOS's > >FDISK (which you used to create the DOS partition) will talk > >to a disk shaped per the BIOS geometry. > > Of course you must use a consistently fudged geometry for everything > related to the BIOS. Yes, but when FBSD mounts a DOS "partition" that resides on the same disk as a FBSD "slice", how does it get to the "correct sectors" that the DOS partition was formatted to use since the drive is now configured for a different geometry? > >When FBSD takes over, > >the geometry of the disk is changed and track/cylinder boundaries > >can "move" (e.g., consider my case with a default geometry > >of 57 S/C vs. the BIOS's 63 S/C). > > The FreeBSD is unrelated to the BIOS. It doesn't even notice when > the boundaries move. See above. > >And, switching to/from DOS can screw you. Consider DOS and > >FBSD sharing wd0. FBSD reINITs the drive to use it's default > >geometry. Then, you mount a DOS partition FROM THE SAME DRIVE! > >The layout of the DOS partition expected the BIOS geometry > >yet the disk is now accessed using the FBSD geometry. > > No problem. If you mount the partition under FreeBSD, msdosfs > ignores everything related to geometries. If you reboot and > "mount" the partition under DOS, then the BIOS reinitializes the > geometry on reboot. Yes, I understand how/when the drive is reINITted. However, it doesn't explain how the two coexist AT RUN TIME. > >I suspect that LBA *could* be a redeeming grace. However, I don't > > Nope. > > >know enough about the *guarantees* within the drive to determine > >if a particular sector address will *allways* map to the same > >(PHYSICALLY EQUIVALENT) sector regardless of geometry (i.e., > >will sector "97" in a 611/16/63 geometry correspond to the > >exactl same physical sector when the drive is configured for > >790/15/57??) > > This is already guaranteed as follows: pick a geometry. Any physically > reasonable geometry will work with a modern IDE drive (but note that > BIOS geometries are usually not physically reasonable with modern BIOSes > for modern (non-small) IDE drives). Then issue a WDCC_IDC command to > tell the controller to translate using that geometry. Then do all > translations from blkno's to C/H/S values using that geometry. The Fine. So, for a DOS geometry, I "pick" 611/16/63. And, *DOS* faithfully uses that geometry for all disk accesses. Meanwhile, FBSD "picks" 790/15/57 and faithfully uses that geometry. But, FBSD doesn't *know* about the 611/16/63 that I "picked" for DOS! So, when *it* wants to go find a sector in that DOS filesystem that I've mounted, how does *it* get to the right sector -- given that the most recent INITIALIZE command specified the 790/15/57 geometry?? > driver could use LBA mode for modern drives, but that would complicate > the driver for little or negative benefit - it would still have to do > the translation for old drives. > > >[someone who really *knows* what they are talking about could probably > >determine the *real* effects of all this :-( ] > > Users get confused :-(. Well, from the followups to this original post, it appears "users" aren't the *only* ones who get confused! ;-) --don From owner-freebsd-hackers Wed Apr 2 07:04:56 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA03482 for hackers-outgoing; Wed, 2 Apr 1997 07:04:56 -0800 (PST) Received: from seagull.rtd.com (seagull.rtd.com [198.102.68.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA03477 for ; Wed, 2 Apr 1997 07:04:52 -0800 (PST) Received: (from dgy@localhost) by seagull.rtd.com (8.7.5/8.7.3) id IAA05043; Wed, 2 Apr 1997 08:03:10 -0700 (MST) From: Don Yuniskis Message-Id: <199704021503.IAA05043@seagull.rtd.com> Subject: Re: wd driver questions To: terry@lambert.org (Terry Lambert) Date: Wed, 2 Apr 1997 08:03:09 -0700 (MST) Cc: dgy@rtd.com, terry@lambert.org, bde@zeta.org.au, hackers@freebsd.org, helbig@MX.BA-Stuttgart.De In-Reply-To: <199703171804.LAA08120@phaeton.artisoft.com> from "Terry Lambert" at Mar 17, 97 11:04:24 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk It seems that Terry Lambert said: > > > > Q: Why does FreeBSD sometimes get it wrong? > > > A: Because FreeBS gets the INT 13 values in the boot loader, but then > > > discards them in favor of the potentially incorrect CMOS and/or > > > non-BIOS drive query return values. > > > > Yes, I don't understand why FBSD can't just use the same geometry > > settings that boot.c deduces? Are the BIOS ROMs no longer accessible > > once boot() has completed (sorry, my ignorance is showing...)? > > Yes. They are no longer accessable. Or, in particular, you are not > able to make INT 13 calls once you are in protected mode because there > is not VM86() support for doing so. > > As Bruce explained, there is no way to establish a correspondance > between BIOS device number (for which you have obtained a BIOS geometry, > the "correct" geometry) and a FreeBSD device. What he didn't explain is > that this is because the BIOS device order is a result of BIOS POST > order for the controllers and the devices, which hook INT 13, and the > FreeBSD device order is a result of probe order, generally a result > of the config file lines with which the kernel is built. Because Ahhhh... OK. *That* makes sense. I guess I had blindly assumed drives were logically tied to particular controller addresses, etc. (i.e. the "MASTER" IDE drive on the controller at 0xXXX is drive Y, the "SLAVE" drive on that same controller is drive Y+1, etc.) > people plug things into the bus in whatever order they feel like, there > is no way to buld a kernel that is guaranteed to probe devices in the > same order that INT 13 is chained. As a matter of fact, you can't > even guaranteed that a device with an INT 13 chain-in even *has* a > FreeBSD driver at all, so you may be out of order no matter what you > do (this is why I'm always calling for "a fallback driver using VM86()"). Presumably, INT13 is hooked in "slot order"? > > > I think that the correspondance can be largely established between > > > BIOS device and FreeBSD device using sector counts from protected > > > mode calls vds C*H*S from the BIOS. In the case where they are > > > equal, it doesn't matter. Where it seems to matter, you can look > > > for the FreeBSD partition, the DOS boot block to find the boot device, > > > or, if all else fails, ask the user (instead of asking by default). > > > > Why *should* there be a difference? I would assume that you would > > err on the side of adhering to the BIOS geometry -- wouldn't > > adopting some *other* geometry make it tough (impossible?) > > to have a DOS partition on the same drive as FBSD and be able to > > *access* that DOS partition from FBSD? > > Yes, that's true; coeexistance is impossible. Tony Overfield of Dell > has stated in no uncertain terms that FreeBSD is incapable of installing > on an increasing number of Dell systems because of this discrepancy > (which arises out of big disks and their use of the "service pack 2" > (OEMSR2) Win95 VFAT32 code). They *require* INT 13 LBA INIT'ing of the > drive. > > The reason there *should* be a difference is explained above: we can't > tell which INT 13 drive ID goes with which disk probed by FreeBSD. I still don't see why there "should" be the difference. The only reason (and I'm not trying to belittle this at all...) I can see that you would *not* want to use the geometry that the (IDE) drive has been initted to (by the "BIOS") would be to be able to support drives that the BIOS *can't*. But, you would do this at the expense of not being able to have DOS coexist with FBSD on those drives (that's still my impression -- though I could be quite wrong). --don From owner-freebsd-hackers Wed Apr 2 07:11:07 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA03795 for hackers-outgoing; Wed, 2 Apr 1997 07:11:07 -0800 (PST) Received: from pluto.plutotech.com (root@pluto.plutotech.com [206.168.67.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA03776; Wed, 2 Apr 1997 07:10:57 -0800 (PST) Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by pluto.plutotech.com (8.8.5/8.8.3) with ESMTP id IAA01337; Wed, 2 Apr 1997 08:10:35 -0700 (MST) Message-Id: <199704021510.IAA01337@pluto.plutotech.com> X-Mailer: exmh version 2.0beta 12/23/96 To: stesin@gu.net, davidg@FreeBSD.org cc: Satoshi Asami , hackers@FreeBSD.org, scsi@FreeBSD.org Subject: Re: 2 PCI busses, 2 AIC chips, 2.2.1. Howto ? In-reply-to: Your message of "Wed, 02 Apr 1997 17:29:42 +0300." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 02 Apr 1997 08:08:34 -0700 From: "Justin T. Gibbs" Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > I'll try to find what the bridge chip is. Don't XXpress motherboards have two PCI->Host bridges? I think that David ran into this same problem on an earlier incarnation of Wcarchive and solved it by telling the PCI code to go find more than 1 primary PCI bus. > But what is the chance that _both_ AIC chips are "on the > other side of the bridge", anyway? Very, very high. This is how the motherboard Wcarchive used to use was configured. -- Justin T. Gibbs =========================================== FreeBSD: Turning PCs into workstations =========================================== From owner-freebsd-hackers Wed Apr 2 07:11:54 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA03907 for hackers-outgoing; Wed, 2 Apr 1997 07:11:54 -0800 (PST) Received: from fyeung5 (netific.vip.best.com [205.149.182.145]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id HAA03885; Wed, 2 Apr 1997 07:11:48 -0800 (PST) Received: from fyeung8.netific.com (fyeung8 [204.238.125.8]) by fyeung5 (8.6.11/8.6.9) with SMTP id HAA03765; Wed, 2 Apr 1997 07:11:16 -0800 Received: by fyeung8.netific.com (5.x/SMI-SVR4) id AA16536; Wed, 2 Apr 1997 07:19:10 -0800 Date: Wed, 2 Apr 1997 07:19:10 -0800 From: fyeung@fyeung8.netific.com (Francis Yeung) Message-Id: <9704021519.AA16536@fyeung8.netific.com> To: dennis@etinc.com, brian@awfulhak.org Subject: Re: PPP Desperate for help! Cc: dev@fgate.flevel.co.uk, freebsd-current@freebsd.org, freebsd-hackers@freebsd.org X-Sun-Charset: US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Greetings, Will you have the magic number problem if you connect 2 FreeBSD machines directly via PPP ? (both machines are in LCP active) Francis > From brian@awfulhak.org Tue Apr 1 14:32 PST 1997 > To: dennis > Cc: Developer , freebsd-current@freebsd.org, > freebsd-hackers@freebsd.org > Subject: Re: PPP Desperate for help! > Mime-Version: 1.0 > Date: Tue, 01 Apr 1997 20:49:07 +0100 > From: Brian Somers > X-Loop: FreeBSD.org > X-Fetchmail-Warning: no recipient addresses matched declared local names > > > At 01:42 PM 4/1/97 +0100, Developer wrote: > > > > > > > > >On Tue, 1 Apr 1997, Andrzej Bialecki wrote: > > > > > >> On Tue, 1 Apr 1997, Developer wrote: > > >> > > >> > > > >> > We are really stuck trying to get the PPP user program (ppp) to > > connect to > > >> > our Perle 833 dial in server. We can connect using Windows NT fine but on > > >> > BSD ppp logs in and then data will travel in both directions (As I can > > see > > >> > the modem lights flash at both ends for both send/recieve) but no data > > >> > seems to actually get through as pings do not work. The packets from the > > >> > server to the user, but seem to be lost on the way back. > > >> > > >> Let me take a guess: maybe it's a known problem with tcp_extensions set to > > >> YES in /etc/sysconfig? Some broken TCP stacks don't like it, so the > > >> solution would be to turn it off. > > >> > > >> Andy > > > > > >We've tried this - no difference. Any more ideas please? > > > > Why not post a trace of the ppp negotiations? Perhaps there is a parameter > > conflict....are you getting to state IPCP_OPENED? If not, you cant pass > > traffic. > > > > Dennis > > I suspect you're way ahead of the problem. Has the original poster > tried "set openmode active" ? Without this, a client ppp will wait > for the server to initiate LCP. > > I think it may be frugal to make this the default for both client > *and* server. A lot of server implementations wait for the client > to start. Any comments ? > -- > Brian , > > Don't _EVER_ lose your sense of humour.... > > From owner-freebsd-hackers Wed Apr 2 07:21:59 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA04472 for hackers-outgoing; Wed, 2 Apr 1997 07:21:59 -0800 (PST) Received: from pdx1.world.net (pdx1.world.net [192.243.32.18]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA04465 for ; Wed, 2 Apr 1997 07:21:56 -0800 (PST) From: proff@suburbia.net Received: from suburbia.net (suburbia.net [203.4.184.1]) by pdx1.world.net (8.7.5/8.7.3) with SMTP id HAA18840 for ; Wed, 2 Apr 1997 07:24:08 -0800 (PST) Received: (qmail 27777 invoked by uid 110); 2 Apr 1997 15:21:14 -0000 Message-ID: <19970402152114.27776.qmail@suburbia.net> Subject: Re: devfs In-Reply-To: <199704021415.GAA01403@freefall.freebsd.org> from Darren Reed at "Apr 3, 97 00:10:35 am" To: avalon@coombs.anu.edu.au (Darren Reed) Date: Thu, 3 Apr 1997 01:21:14 +1000 (EST) Cc: hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > In some mail from Jordan K. Hubbard, sie said: > > > > > 1) accept that devfs in it's present form is a kludge. Move > > > all the default permissions and device numbers into a > > > sysctl hierarchy. This hierarchy is then used to create > > > names in /dev. Add a new ufs file type: S_IFLNKDEV. > > > When a namei occurs on what this pseudo-symlink points > > > to, it looks up name in the sysctl and returns the inode > > > of the "symlink" with device numbers filled in. The > > > > I'm not sure if I care much for this as a solution to the persistance > > problem, at least not unless you were also talking about having DEVFS > > manipulate this piece of the sysctl namespace on behalf of the user > > for any modifications they may make. It has to be transparent after > > setup, right? > > > > Also, I don't think that anyone has addressed the issue of sysctl MIB > > persistance either yet, so your changes would still go away on reboot. :-) Perhaps I didn't make it clear enough. sysctl provides the names of the devices only. Since this information can't be changed, it doesn't need persistance. All persistent information (i.e everything you normally see in a struct stat) is stored "on the disk" in LNKDEV inodes. > > then passed on to namei()? You could get variant symlink behaviour by > > poking things into sysctl. :-) > > OSx5.1 (ye olde syvr3/4.2bsd mix) from Pyramid nhad a "conditional" > symbolic link (if I remember right) worked something like this: > > ln -c att /.attbin -c ucb /.ucbbin /bin > > darren Yes, I remember my Pyramid fondly. It not only provided two universes, but heated the whole house as well, drinking around $10 a day in power :) The kernel shouldn't be handling these name space hacks. One world view of mount points isn't a good thing. This should all be done in libc, ala plan9/VSTa. Such a thing has already started with the introduction of modifications to readdir(3) for unions. Once again though, this is more in the manner of a gross hack than any consistent or modular approach. There are very good solutions to this, but as far as I see -core is too conservative entertain such a deviation from the BSD path. You could say it is not the place for FreeBSD to be a research platform - but I'd have to disagree here and say that the research here is quite mature and proven. As for your variant on the sysctl idea, I like it, but I don't think it has any benefit for the devfs problem :) Cheers, Julian. From owner-freebsd-hackers Wed Apr 2 07:33:47 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA04907 for hackers-outgoing; Wed, 2 Apr 1997 07:33:47 -0800 (PST) Received: from seagull.rtd.com (seagull.rtd.com [198.102.68.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA04898 for ; Wed, 2 Apr 1997 07:33:37 -0800 (PST) Received: (from dgy@localhost) by seagull.rtd.com (8.7.5/8.7.3) id IAA07133; Wed, 2 Apr 1997 08:32:00 -0700 (MST) From: Don Yuniskis Message-Id: <199704021532.IAA07133@seagull.rtd.com> Subject: Re: wd driver questions To: terry@lambert.org (Terry Lambert) Date: Wed, 2 Apr 1997 08:31:59 -0700 (MST) Cc: msmith@atrad.adelaide.edu.au, terry@lambert.org, bde@zeta.org.au, dgy@rtd.com, hackers@FreeBSD.org, helbig@MX.BA-Stuttgart.De In-Reply-To: <199703171835.LAA08169@phaeton.artisoft.com> from "Terry Lambert" at Mar 17, 97 11:35:00 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk It seems that Terry Lambert said: > > > > Q: Why does FreeBSD sometimes get it wrong? > > > A: Because FreeBS gets the INT 13 values in the boot loader, but then > > > discards them in favor of the potentially incorrect CMOS and/or > > > non-BIOS drive query return values. > > > > This is not answering the question. > > Yes it is. If FreeBSD did not discard the boot loader values, it would I *still* fail to see why the drive isn't left INITted the way it already *has* been INITted -- i.e. why these values are discarded. And, why is it necessary to "pair up" an already INITted drive with it's FBSD counterpart? > have the "correct" geometry for the device. The "correctness" of the > geometry is a result of: > > 1) Being able to determine the absolute sector offset for a > given C/H/S value in the partition table. > > 2) Being able to, from protected mode, write a DOS partition > table C/H/S value so that when the DOS MBR goes to that > C/H/S location and reads a track, it gets the FreeBSD second > stage boot. > > 3) The FreeBSD kernel (which should use the LBA address, not > the C/H/S address) being able to turn the C/H/S address > back into an absolute sector offset. I had assumed that this would be the "easy" way to "equalize" all translation schemes. However, apparently this is not a universally implemented feature (?) > > > Q: Why does FreeBSD do this? > > > A: No one knows. > > > > Several, perhaps even many people know. Some of these have tried to explain > > the problem, but it would appear that you don't actually want to know the > > answer. > > > > One answer is that the values returned by the int13 call in the bootloader > > may not match the actual layout of the disk. The CMOS values allow the > > user to override a BIOS that might be getting it wrong. The direct drive Hmmm... I hadn't seen how the CMOS parameters factored into this. I was under the impression that the BIOS would INIT the drive using these CMOS parameters (e.g., "user defined geometry") > > query allows a properly-configured disk to be moved from one system to > > another with a reasonable chance of it still working. Hmmm... why are the "direct drive query" parameters "correct"? In my case, they aren't. > As long as you have no other OS's on the drive which don't re-INIT > the drive as well. Perhaps I'm misunderstanding this comment. While FBSD is running, no "other OS" *can* reINIT the drive, right? So, is this the same as my statement that if another OS *cohabits* the drive with FBSD, then you have a (potential) problem? > > It is not sensible to try to match sector counts against c/h/s values > > from the BIOS in many cases; often the discrepancy can be more than > > one track multiple out. A signature approach might perhaps be better. > > o Multiply the C/H/S geometry of the INT 13 ID's in question by > the C/H/S limit values to get the absolute drive size. Compare > the size against the IDENTIFY size. This will establish disk > identity in the vast majority of cases. > > o Multiply the C/H/S geometry of the INT 13 ID's in question by > the C/H/S value from the partition table for bootable OS types, > starting with FreeBSD and MS OS's, and look for the "bootable > signature" word that the DOS MBR looks for to establish disk > identity. This will cover the vast majority of the remaining > cases. > > o Modify the second stage boot to fill in the correct LBA address > in the DOS partition table for a given C/H/S value at boot time; > have BSD *always* use the LBA values in protected mode and > *always* use the C/H/S values in the INT 13 using boot code. > This should cover all other cases except where the drive is > identical. > > o When the drive is identical, ask the user the first time the > protected mode kernel is run, before root is mounted, which > device is correct. Record this value in the "bootbias" in > the second stage boot block *after* a successful mount of > the root FS. Never ask again(*). > > (*) Ideally, this last would not be necessary if the root > mount modifications I submitted in June of 1994 were > integrated, since multiple root mounts could be attempted, > and the one which was successful could be taken as the > right one. In the case of identical disk devices, the > "last mounted" time stamp should be examined, and the most > recent one taken, and/or the drive serial number should be > recorded in the boot block and retrieved using IDENTIFY or > SCSI operands on the devices before they are booted, so > that subsequent mounts following a duplication including > timestamp would not fail. (sigh) Perhaps this would make sense to me if I understood why the "need" to match BIOS devices with FBSD devices... Sorry, when I originally started this thread, I thought it was a "simple little thing" (TM Reg.) <:-( --don From owner-freebsd-hackers Wed Apr 2 07:44:35 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA05650 for hackers-outgoing; Wed, 2 Apr 1997 07:44:35 -0800 (PST) Received: from seagull.rtd.com (seagull.rtd.com [198.102.68.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA05643 for ; Wed, 2 Apr 1997 07:44:31 -0800 (PST) Received: (from dgy@localhost) by seagull.rtd.com (8.7.5/8.7.3) id IAA08149; Wed, 2 Apr 1997 08:43:06 -0700 (MST) From: Don Yuniskis Message-Id: <199704021543.IAA08149@seagull.rtd.com> Subject: Re: wd and funky geometry (understood?) To: terry@lambert.org (Terry Lambert) Date: Wed, 2 Apr 1997 08:43:05 -0700 (MST) Cc: bde@zeta.org.au, freebsd-hackers@freefall.FreeBSD.org (FreeBSD hackers) In-Reply-To: <199703171843.LAA08211@phaeton.artisoft.com> from "Terry Lambert" at Mar 17, 97 11:43:45 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk It seems that Terry Lambert said: > > > >However, *both* drives have a BIOS geometry of 611/16/63. (Sorry, > > >my ancient BIOS does not support "user configurable" geometries > > >so this was the closest geometry I could find -- hence the reason > > > > What's the problem? The BIOS uses the fudged geometry that you told it, > > but the FreeBSD driver knows better - there is no reason for it to use > > a fudged geometry. > > The disk is shared with DOS and uses the DOS MBR. THe DOS MBR will > use the fudged geometry, and the BSD DOS partition table entry must > be in the fudged geometry to make the DOS MBR happy to boot it. *Yes*! (whew, I *thought* that's what I was saying :-( > > Of course you must use a consistently fudged geometry for everything > > related to the BIOS. > > Like multiplying the C/H/S value by the fudged geometry in protected > mode to find the BSD boot and disktab data. And doesn't FBSD need to know the DOS geometry (or "virtually" reINIT the drive) to be able to find all of the DOS sectors when it mounts the msdosfs? --don From owner-freebsd-hackers Wed Apr 2 07:49:12 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA05847 for hackers-outgoing; Wed, 2 Apr 1997 07:49:12 -0800 (PST) Received: from seagull.rtd.com (seagull.rtd.com [198.102.68.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA05830 for ; Wed, 2 Apr 1997 07:48:57 -0800 (PST) Received: (from dgy@localhost) by seagull.rtd.com (8.7.5/8.7.3) id IAA07901; Wed, 2 Apr 1997 08:40:04 -0700 (MST) From: Don Yuniskis Message-Id: <199704021540.IAA07901@seagull.rtd.com> Subject: Re: wd driver questions To: bde@zeta.org.au (Bruce Evans) Date: Wed, 2 Apr 1997 08:40:04 -0700 (MST) Cc: freebsd-hackers@freefall.FreeBSD.org (FreeBSD hackers), terry@lambert.org In-Reply-To: <199703171835.FAA22959@godzilla.zeta.org.au> from "Bruce Evans" at Mar 18, 97 05:35:57 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk It seems that Bruce Evans said: > > >> The ROMs are completely inaccessible, but that isn't the problem since > >> the geometry of all drives known to the BIOS (actually, only the first > >> 10) is passed from boot.c to the kernel. The problems are: > > > >Hmmm... I thought only the drives *configured* were passed along > >(i.e. two drives on each of two controllers)? > > 10 always. Some might be SCSI drives. Ah, OK... I had assumed :-( they were just the IDE drives since I was looking through the wd driver... > >> 2. the kernel has no way of knowing to correspondence between BIOS drive > >> numbers and FreeBSD drive names. It could do the same thing as boot.c, > > > >Huh? I thought wd0 was the first drive (i.e. the IDE "master") on the first > >controller, wd1 was the "slave" on that controller, etc. > > > >Ahhhh... this isn't *guaranteed* to be the case? > > Yes, but BIOS drive 0 is just the first drive found. It may be wd2 if there > is nothing on the first controller. Then there are SCSI drives. But the wd probe shouldn't (?) need to reINIT the drive regardless of which BIOS drive it is, should it? Why does it even *care* which BIOS drive it is? > >> I don't see any problem. The BIOS geometry is 611/16/63 for some reason > >> (probably the BIOS is old and can't translate to > 16 heads). You must > > > >[if you look at the actual geometry -- 790/15/57 -- it has nothing to > >do with translating for > 16 heads or "EIDE" stuff. Rather, the BIOS > >just doesn't support a "user configurable" disk type so I'm stuck with > >whatever was hardcoded into the ROMs -- hence the reason I'm reburning > >the system ROMs! :>] > > OK. The BIOS's lack of translation is only a problem when the 16/63 > translation gives more than 1024 (translated) cylinders. I guess this is the IDE/EIDE issue? > >> use the BIOS geometry in fdisk. The geometry used elsewhere is almost > >> arbitrary. The driver uses 790/15/57 for simplicity. It prints 790/15/57 > >> because that's the only one that it knows about. Don't use this in fdisk > >> if it differs from the BIOS geometry. OTOH, you may note that the BIOS > >> goemetry wastes about 30MB: > >> > >> 790*15*57 = 675450 > >> 611*16*63 = 615888 > >> wasted = 59562 > > > >Yes, I have been aware of that for far too long :-( Each machine has a pair > >of 4G SCSI drives so the 30M I lose on the IDE drives is pretty insignificant. > > So is the 2*300MB on the drives :-). Um, yes. But, the 300M drives are internal and allow the system to run nicely in the event the external drives are not attached, broke or SCSI failures, etc. > >> You can probably use these sectors for a non-DOS non-bootable partition > >> by telling fdisk that the geometry is 612/16/63 and being careful not > >> to use the nonexistent part of the last "cylinder". > > Oops, there is more than one "cylinder" past the last, and only a small > amount of the wastage is on the partial cylinder. > > 790*15*57 / (16*63) = 670 > > so you have 670 full "cylinders" to use. You can probably use all these > cylinders, perhaps even for DOS and bootable partitions by telling fdisk > that the geometry is 670/16/63 and not using the partial cylinder. This > depends on the softare being either stupid enough to not check cylinder > numbers against the 612 limit or smart enough to know that this limit > is wrong. I'm working on an "easier" solution: patch the ROMs to support the "correct" geometries! :> > >> Your probably may actually be that the drive is jumpered to force > problem > >> 611/16/63 and broken so that it still reports 790/15/57. Some ESDI > > > >No, the native drive geometry is 790/15/57. It is INITed by the > >BIOS to 611/16/63. This is the geometry that it actually *uses*. > >The IDENTIFY command returns the *default* geometry **AND** the > >*apparent* geometry. The existing code ignores the apparent > >geometry in favor of the default geometry (in fact, the wdparams > >struct doesn't even acknowledge the presence of the apparent geometry > >parameters). > > This is because the driver only knows about old standards where the > apparent geometry parameters didn't exist. Using these parameters > would either break the support for old drives or complicate the driver > (it's not easy to determine drive capabilities, because old standards > didn't have reliable capability bits). Ah, OK. But, why does the probe even *care* what the drive's geometry happens to be? *If* you assume that the drive is already INITted, is there any real need to "guess" at the correct geometry and reINIT the drive? > >As such, the drive is doing exactly what it is supposed > >to do -- reporting it's *default* geometry even though it is configured > >with a different geometry. > > And the driver is doing exactly what it is supposed to do - ignoring > the current geometry. Again, *why*? --don From owner-freebsd-hackers Wed Apr 2 07:56:08 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA06299 for hackers-outgoing; Wed, 2 Apr 1997 07:56:08 -0800 (PST) Received: from tor-adm1.nbc.netcom.ca (taob@tor-adm1.nbc.netcom.ca [207.181.89.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA06262 for ; Wed, 2 Apr 1997 07:56:02 -0800 (PST) Received: from localhost (taob@localhost) by tor-adm1.nbc.netcom.ca (8.8.5/8.8.5) with SMTP id KAA05494; Wed, 2 Apr 1997 10:54:47 -0500 (EST) Date: Wed, 2 Apr 1997 10:54:47 -0500 (EST) From: Brian Tao To: keith waters cc: freebsd-hackers@freebsd.org Subject: Re: Unreasonable rebuild time on large password files In-Reply-To: <199704020913.LAA03056@ilink.nis.za> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Wed, 2 Apr 1997, keith waters wrote: > > This must be one of the worst cases of code/resourse optimisation, > since changing a password means rebuilding EVERY SINGLE entry in the > password databases. We await your optimized code diffs to pwd_mkdb(8) then. :) > Does anybody out there have a working solution ? Please! In the meantime, look through the pwd_mkdb source and increase the default hash table size (it is fixed at 2MB, IIRC). Increase this as necessary to minimize disk thrashing. The time needed to rebuild the databases is almost inversely proportional to the amount of memory you allow pwd_mkdb to allocate, up to a point. -- Brian Tao (BT300, taob@netcom.ca) "Though this be madness, yet there is method in't" From owner-freebsd-hackers Wed Apr 2 08:01:58 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA06633 for hackers-outgoing; Wed, 2 Apr 1997 08:01:58 -0800 (PST) Received: from out2.ibm.net (out2.ibm.net [165.87.201.252]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id IAA06606; Wed, 2 Apr 1997 08:01:49 -0800 (PST) Received: (from uucp@localhost) by out2.ibm.net (8.6.9/8.6.9) id QAA710624; Wed, 2 Apr 1997 16:01:39 GMT Message-Id: <199704021601.QAA710624@out2.ibm.net> Received: from slip166-72-229-226.va.us.ibm.net(166.72.229.226) by out2.ibm.net via smap (V1.3mjr) id smanecDFW; Wed Apr 2 15:58:48 1997 Reply-To: From: "Steve Sims" To: "Hackers" , Subject: -current 'make world' chokes Date: Wed, 2 Apr 1997 10:58:25 -0500 X-MSMail-Priority: Normal X-Priority: 3 X-Mailer: Microsoft Internet Mail 4.70.1161 MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk How come -current has, for the last couple of days (at least, that I've *noticed*) chokes while making the world. I do 'cd /usr/src ; make update world' to pull current and build it and get: [blah, blah, blah] cd /usr/src/lib/csu/i386 && make beforeinstall install -C -o bin -g bin -m 444 /usr/src/lib/csu/i386/dlfcn.h /usr/include install -c -o bin -g bin -m 444 crt0.o c++rt0.o gcrt0.o scrt0.o sgcrt0.o /usr/lib install: crt0.o: No such file or directory *** Error code 71 Stop. *** Error code 1 Stop. *** Error code 1 Stop. # _ cd'ing to /usr/src/lib and doing 'make' and/or 'make install' builds crt0.o just fine, but it looks like it's getting skipped in the world context. It's kind of a pain to iteratively make this branch of the tree and then make world with NOCLOBBER, NOCLEAN, etc... Am I totally missing the big picture here? ...sjs... From owner-freebsd-hackers Wed Apr 2 08:09:09 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA07006 for hackers-outgoing; Wed, 2 Apr 1997 08:09:09 -0800 (PST) Received: from siesta.cs.wustl.edu (nw1@siesta.cs.wustl.edu [128.252.165.3]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id IAA06997 for ; Wed, 2 Apr 1997 08:09:03 -0800 (PST) Received: from localhost by siesta.cs.wustl.edu (SMI-8.6/ECL-J1.00) id KAA17928; Wed, 2 Apr 1997 10:08:54 -0600 Message-Id: <199704021608.KAA17928@siesta.cs.wustl.edu> To: John Birrell cc: Hackers@freebsd.org Subject: Re: Bugs in libc_r (fwd) In-reply-to: Your message of "Wed, 02 Apr 1997 20:49:54 +1000." <199704021049.UAA13376@freebsd1.cimlogic.com.au> Date: Wed, 02 Apr 1997 10:08:54 -0600 From: Nanbor Wang Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > > Some C precedence errors. > > > [...] > > > > > > I fixed these this morning, plus a few other things. 8-) > > > > Anyway we can get an update of this fix? (I don't want to get into > > -current at this moment ;-) > > I've checked the libc_r source that was tagged for 2.2 and there have > been a lot of updates to uthread files since then. Enough to prevent > patching the 2.2 tree (despite what I said earlier 8-). Sorry. We'll > need to be in better shape for 3.0. Sorry, John. What I really meant was Lite2 merge. I am using -current before lite2 merge. (before 2/9/97 ?) Can I patch the pthread library? Thanks. nw > Regards, > > -- > John Birrell - jb@cimlogic.com.au; jb@netbsd.org; jb@freebsd.org > CIMlogic Pty Ltd, 119 Cecil Street, South Melbourne Vic 3205, Australia > Tel +61 3 9690 6900 Fax +61 3 9690 6650 Mob +61 418 353 137 From owner-freebsd-hackers Wed Apr 2 08:35:38 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA08057 for hackers-outgoing; Wed, 2 Apr 1997 08:35:38 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA08041; Wed, 2 Apr 1997 08:35:26 -0800 (PST) Received: from ntws (ntws.etinc.com [204.141.95.142]) by etinc.com (8.8.3/8.6.9) with SMTP id LAA24142; Wed, 2 Apr 1997 11:39:42 -0500 (EST) Message-Id: <3.0.32.19970402113410.00b716e0@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Wed, 02 Apr 1997 11:34:17 -0500 To: fyeung@fyeung8.netific.com (Francis Yeung), brian@awfulhak.org From: dennis Subject: Re: PPP Desperate for help! Cc: dev@fgate.flevel.co.uk, freebsd-current@FreeBSD.org, freebsd-hackers@FreeBSD.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk At 07:19 AM 4/2/97 -0800, Francis Yeung wrote: > > >Greetings, > > Will you have the magic number problem if you >connect 2 FreeBSD machines directly via PPP ? (both >machines are in LCP active) > > Francis What "magic number" problem are you referring to? db > >> From brian@awfulhak.org Tue Apr 1 14:32 PST 1997 >> To: dennis >> Cc: Developer , freebsd-current@freebsd.org, >> freebsd-hackers@freebsd.org >> Subject: Re: PPP Desperate for help! >> Mime-Version: 1.0 >> Date: Tue, 01 Apr 1997 20:49:07 +0100 >> From: Brian Somers >> X-Loop: FreeBSD.org >> X-Fetchmail-Warning: no recipient addresses matched declared local names >> >> > At 01:42 PM 4/1/97 +0100, Developer wrote: >> > > >> > > >> > >On Tue, 1 Apr 1997, Andrzej Bialecki wrote: >> > > >> > >> On Tue, 1 Apr 1997, Developer wrote: >> > >> >> > >> > >> > >> > We are really stuck trying to get the PPP user program (ppp) to >> > connect to >> > >> > our Perle 833 dial in server. We can connect using Windows NT fine but on >> > >> > BSD ppp logs in and then data will travel in both directions (As I can >> > see >> > >> > the modem lights flash at both ends for both send/recieve) but no data >> > >> > seems to actually get through as pings do not work. The packets from the >> > >> > server to the user, but seem to be lost on the way back. >> > >> >> > >> Let me take a guess: maybe it's a known problem with tcp_extensions set to >> > >> YES in /etc/sysconfig? Some broken TCP stacks don't like it, so the >> > >> solution would be to turn it off. >> > >> >> > >> Andy >> > > >> > >We've tried this - no difference. Any more ideas please? >> > >> > Why not post a trace of the ppp negotiations? Perhaps there is a parameter >> > conflict....are you getting to state IPCP_OPENED? If not, you cant pass >> > traffic. >> > >> > Dennis >> >> I suspect you're way ahead of the problem. Has the original poster >> tried "set openmode active" ? Without this, a client ppp will wait >> for the server to initiate LCP. >> >> I think it may be frugal to make this the default for both client >> *and* server. A lot of server implementations wait for the client >> to start. Any comments ? >> -- >> Brian , >> >> Don't _EVER_ lose your sense of humour.... >> >> > > From owner-freebsd-hackers Wed Apr 2 09:13:19 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA10882 for hackers-outgoing; Wed, 2 Apr 1997 09:13:19 -0800 (PST) Received: from labinfo.iet.unipi.it (labinfo.iet.unipi.it [131.114.9.5]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id JAA10774 for ; Wed, 2 Apr 1997 09:11:18 -0800 (PST) Received: from localhost (luigi@localhost) by labinfo.iet.unipi.it (8.6.5/8.6.5) id SAA27172; Wed, 2 Apr 1997 18:19:22 +0200 From: Luigi Rizzo Message-Id: <199704021619.SAA27172@labinfo.iet.unipi.it> Subject: Multiple versions of packages (was: Re: bin/3165...) To: jmz@cabri.obs-besancon.fr (Jean-Marc Zucconi) Date: Wed, 2 Apr 1997 18:19:21 +0200 (MET DST) Cc: hackers@freebsd.org, tege@nada.kth.se In-Reply-To: <9704021608.AA02684@cabri.obs-besancon.fr> from "Jean-Marc Zucconi" at Apr 2, 97 05:07:53 pm X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > >> Description: > > > A file `kpsetool' is required by `MakeTeXPK' but does not exist in > > the tex-3.14159.tgz package. > > I don't see any occurence of ``kpsetool'' in MakeTeXPK nor any program > with this name in the build directory. Are you sure to use the > MakeTeXPK script which comes with tex? I think I saw the above problem with the teTeX port as well (the latter also lacks symlinks for latex, tex and mf). BTW it is a bit confusing to have teTeX, tex-3.14159, latex2e, xdvi, dvips, etc. as separate packages when teTeX includes almost everything one needs. IMHO, at least for the precompiled packages, it would be better to have just one -- the most recent and complete one, teTeX in this case -- and just leave the option of ports for those with unusual needs. The same applies probably to other packages such as tk/tcl, ghostscript, gv/ghostview... in all cases where the most recent version is just a superset of the previous one and does not suffer suffer from what I call the "Netscape syndrome" (application size grows exponentially with major version numbers. And minor version numbers are never greater than 0). Cheers Luigi -----------------------------+-------------------------------------- Luigi Rizzo | Dip. di Ingegneria dell'Informazione email: luigi@iet.unipi.it | Universita' di Pisa tel: +39-50-568533 | via Diotisalvi 2, 56126 PISA (Italy) fax: +39-50-568522 | http://www.iet.unipi.it/~luigi/ _____________________________|______________________________________ From owner-freebsd-hackers Wed Apr 2 09:31:08 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA11700 for hackers-outgoing; Wed, 2 Apr 1997 09:31:08 -0800 (PST) Received: from austin.polstra.com (austin.polstra.com [206.213.73.10]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA11695 for ; Wed, 2 Apr 1997 09:31:06 -0800 (PST) Received: (from jdp@localhost) by austin.polstra.com (8.8.5/8.8.5) id JAA08926; Wed, 2 Apr 1997 09:31:04 -0800 (PST) To: freebsd-hackers@freebsd.org Path: not-for-mail From: jdp@polstra.com (John Polstra) Newsgroups: polstra.freebsd.hackers Subject: Re: select() ? Date: 2 Apr 1997 09:31:04 -0800 Organization: Polstra & Co., Seattle, WA Lines: 17 Distribution: local Message-ID: <5hu54o$8mr@austin.polstra.com> References: <19970402092028.KM07161@uriah.heep.sax.de> Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In article <19970402092028.KM07161@uriah.heep.sax.de>, J Wunsch wrote: > As Adrian Chadd wrote: > > > I've found the select() call in the kernel source, but I was wondering if > > there was a library call as a "front-end" to the kernel call ... I can't > > seem to find it. > > Most of the syscall stubs in the library are being created > automatically. And you can find the magic that generates them in "src/lib/libc/sys/Makefile.inc". -- John Polstra jdp@polstra.com John D. Polstra & Co., Inc. Seattle, Washington USA "Self-knowledge is always bad news." -- John Barth From owner-freebsd-hackers Wed Apr 2 09:42:52 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA12310 for hackers-outgoing; Wed, 2 Apr 1997 09:42:52 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id JAA12305 for ; Wed, 2 Apr 1997 09:42:49 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id KAA13803; Wed, 2 Apr 1997 10:24:18 -0700 From: Terry Lambert Message-Id: <199704021724.KAA13803@phaeton.artisoft.com> Subject: Re: Internal clock To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Wed, 2 Apr 1997 10:24:17 -0700 (MST) Cc: terry@lambert.org, nate@mt.sri.com, proff@suburbia.net, hackers@FreeBSD.ORG In-Reply-To: <28836.859954625@time.cdrom.com> from "Jordan K. Hubbard" at Apr 1, 97 08:17:05 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > I have *no* idea what is preventing you from making it the default > > mechanism. > > 1. It has no persistance and a number of people have already told me directly > (some of them at USENIX, immediately following Julian's talk on DEVFS) > that we will be roundly castigated and screamed at in the various > public forums for breaking /dev's historical behavior if we don't make > persistance work and work *completely transparently*. Let's say > somebody has a "securify" script that they run on a new system as part > of their security auditing process. It's been running on FreeBSD > systems since 2.0, making sure that tape drives and CDROMs and such > all have sane perms (or maybe even some sort of "insane" perms which are > part of a more elaborate ownership scheme which they like), and now > suddenly with FreeBSD 4.0 or whatever they find that everything the > script does on these new systems is undone after a reboot. Say what?? > > Sure, one can stick something in /etc/rc.local which does everything > from beating on /dev to translating the catpages into Urdu, but > the fact remains that someone shouldn't need to do that. As you yourself > say, things should stay *backwards compatible*. :-) Frankly, I don't care about persistance other than the default permission, which should be set correctly already so that most everyone besides me doesn't care about persistance either. As far as putting code into rc.local & rc.shutdown: why doesn't that count as transparent? Because we can't upgrade the rc scripts out from under people because we haven't provided sufficient abstraction of the data representing the configuration from the code that implements it? That brings us to the other thing I have no idea why it hasn't been integrated: directory based rc processing for per configuration item scripts. It doesn't matter if you replace only the rc components that you control by default anyway, as long as you can do so without damaging third-party and post-install configuration. It seems to me that the same people whining for persistance of non-default settings that they aren't willing to set in the device registration for the devices they want non-default should be whining for directory-based rc code. > 2. It tends to panic the system fairly regularly on steady use. Why on > earth would we make something this dangerous the default? It's > simply not finished yet, and there are numerous open PRs against it in > the PR database (#2033, #2034, #2738 and possibly #1211). These are generally on unmount. The ones that aren't, weren't there until there was implementation of this ill-concieved notion of supplying persistance in the devfs code instead of as an external elective for those people who wanted hacker changes to their system to survive reboot, for God only knows why. I keep telling everyone how to fix the VFS code, as I have for the past four (*4*) years (*YEARS*), and keep offering code, but it's always turned down as doing too much at once. The issues are in the layering of the VFS code, and how you define a VFS consumer (including a set of system calls). Now that I've scaled back the code into a "gee, a third grader could understand this"-sized pieces, it looks like something might happen (understandably, one third-grade mentality sized piece at a time is unlikely to implement all the necessary fixes at once). But if you are going to demand "a fix by Tuesday", you must be prepared to take a fix that is non-trivial and may take real effort to understand. Keep your eyes on your goal, not the obstacles, or all you will see is obstacles. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Wed Apr 2 10:00:12 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA13146 for hackers-outgoing; Wed, 2 Apr 1997 10:00:12 -0800 (PST) Received: from fyeung5 (netific.vip.best.com [205.149.182.145]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA13057; Wed, 2 Apr 1997 10:00:00 -0800 (PST) Received: from fyeung8.netific.com (fyeung8 [204.238.125.8]) by fyeung5 (8.6.11/8.6.9) with SMTP id JAA26102; Wed, 2 Apr 1997 09:59:28 -0800 Received: by fyeung8.netific.com (5.x/SMI-SVR4) id AA17015; Wed, 2 Apr 1997 10:07:17 -0800 Date: Wed, 2 Apr 1997 10:07:17 -0800 From: fyeung@fyeung8.netific.com (Francis Yeung) Message-Id: <9704021807.AA17015@fyeung8.netific.com> To: fyeung@fyeung8.netific.com, brian@awfulhak.org, dennis@etinc.com Subject: Re: PPP Desperate for help! Cc: dev@fgate.flevel.co.uk, freebsd-current@freebsd.org, freebsd-hackers@freebsd.org X-Sun-Charset: US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Dennis, The LCP magic number which is to be exchanged between both sides during LCP negotiation. Francis > From dennis@etinc.com Wed Apr 2 09:42 PST 1997 > X-Sender: dennis@etinc.com > Date: Wed, 02 Apr 1997 11:34:17 -0500 > To: fyeung@fyeung8.netific.com (Francis Yeung), brian@awfulhak.org > From: dennis > Subject: Re: PPP Desperate for help! > Cc: dev@fgate.flevel.co.uk, freebsd-current@freebsd.org, > freebsd-hackers@freebsd.org > Mime-Version: 1.0 > X-Loop: FreeBSD.org > X-Fetchmail-Warning: no recipient addresses matched declared local names > > At 07:19 AM 4/2/97 -0800, Francis Yeung wrote: > > > > > >Greetings, > > > > Will you have the magic number problem if you > >connect 2 FreeBSD machines directly via PPP ? (both > >machines are in LCP active) > > > > Francis > > What "magic number" problem are you referring to? > > db > > > >> From brian@awfulhak.org Tue Apr 1 14:32 PST 1997 > >> To: dennis > >> Cc: Developer , freebsd-current@freebsd.org, > >> freebsd-hackers@freebsd.org > >> Subject: Re: PPP Desperate for help! > >> Mime-Version: 1.0 > >> Date: Tue, 01 Apr 1997 20:49:07 +0100 > >> From: Brian Somers > >> X-Loop: FreeBSD.org > >> X-Fetchmail-Warning: no recipient addresses matched declared local names > >> > >> > At 01:42 PM 4/1/97 +0100, Developer wrote: > >> > > > >> > > > >> > >On Tue, 1 Apr 1997, Andrzej Bialecki wrote: > >> > > > >> > >> On Tue, 1 Apr 1997, Developer wrote: > >> > >> > >> > >> > > >> > >> > We are really stuck trying to get the PPP user program (ppp) to > >> > connect to > >> > >> > our Perle 833 dial in server. We can connect using Windows NT > fine but on > >> > >> > BSD ppp logs in and then data will travel in both directions (As > I can > >> > see > >> > >> > the modem lights flash at both ends for both send/recieve) but no > data > >> > >> > seems to actually get through as pings do not work. The packets > from the > >> > >> > server to the user, but seem to be lost on the way back. > >> > >> > >> > >> Let me take a guess: maybe it's a known problem with tcp_extensions > set to > >> > >> YES in /etc/sysconfig? Some broken TCP stacks don't like it, so the > >> > >> solution would be to turn it off. > >> > >> > >> > >> Andy > >> > > > >> > >We've tried this - no difference. Any more ideas please? > >> > > >> > Why not post a trace of the ppp negotiations? Perhaps there is a > parameter > >> > conflict....are you getting to state IPCP_OPENED? If not, you cant pass > >> > traffic. > >> > > >> > Dennis > >> > >> I suspect you're way ahead of the problem. Has the original poster > >> tried "set openmode active" ? Without this, a client ppp will wait > >> for the server to initiate LCP. > >> > >> I think it may be frugal to make this the default for both client > >> *and* server. A lot of server implementations wait for the client > >> to start. Any comments ? > >> -- > >> Brian , > >> > >> Don't _EVER_ lose your sense of humour.... > >> > >> > > > > From owner-freebsd-hackers Wed Apr 2 10:10:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA13593 for hackers-outgoing; Wed, 2 Apr 1997 10:10:11 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA13588 for ; Wed, 2 Apr 1997 10:10:07 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id KAA13869; Wed, 2 Apr 1997 10:53:04 -0700 From: Terry Lambert Message-Id: <199704021753.KAA13869@phaeton.artisoft.com> Subject: Re: question on buffer cache and VMIO To: pmchen@eecs.umich.edu (Peter M. Chen) Date: Wed, 2 Apr 1997 10:53:04 -0700 (MST) Cc: freebsd-hackers@freebsd.org In-Reply-To: <199704020315.WAA03318@life.eecs.umich.edu> from "Peter M. Chen" at Apr 1, 97 10:15:03 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > >One big issue: I'm not sure how this will interact with the "soft > >updates" facility described in the Ganger/Patt paper. This facility > >is a planned future feature on nearly everyone's whiteboards. If you > >have read the paper, I can discuss the idea of a generic implementation > >that is not FS dependent (like the paper's Appendix A UFS example is) > >and the ramifications that would have on FS/cache interaction. > > Soft updates allow you to delay metadata disk writes (just like data disk > writes) without losing file system integrity. However, you will lose more > metadata in a crash than before. E.g. if you mkdir then crash a few seconds > later, soft updates will lose that directory (but in a way that doesn't > compromise file system integrity). This is the same tradeoff that's made > with data. Think of it as allowing more "data loss" without "integrity loss". I was actually thinking more in terms of where the soft updates must be implements so as to not fall back to DOW (Delayed Ordered Write) performance in the unified cache case. Effectively, they must be implemented at the VFS/BIO boundry layer (I've made a couple other postings on this on the -current list, if you are interested). I actually believe the FS-level soft updates implementation that is currently being worked on by a former CSRG member will have at most a 160% of current speed (the same as a DOW implementation in FreeBSD). The effect on BSDI and other non-cache unified BSD's should be significantly higher -- but will be offset by the fact that cache unification is a significant win (it saves a bmap per copyin/out, for one thing). The problem is that soft updates implemented in an FS that does not do physical block I/O as a result of a BIO request (instead, writing a page, marking it dirty, etc.) is somewhat antithetical to the involvement of the BIO subsystem in the node dependency scheduling. The Ganger/Patt paper Appendix A is a non-generic implementation of soft updates, and statically orders operations on an assumption of how a BIO will traditionally behave. The dependency graph is "closed" to the implementation. Effectively, this makes the graph edge-opaque. We can do a graph reduction on the edge-opaque graph, and what we get out will look remarkably like the graph we would get if we were to intentionally implement DOW instead of soft updates. Unfortunately, the Ganger/Patt paper did not take a graph theory approach to modelling the FS event/handler/order-dependency for SVR4 UFS, so it's not easily relocated without another analysis of the FS (on the plus side, a system based on such a model can be used on any FS type, if the model is seperated from the implementation of the Soft Updates itself). All this leads to the question "How intrusive is Rio at the BIO level, and what are the actual interface impacts that will result?". Effectively, those of us without battery-backed RAM, or RAM that is somehow not persistant over a crash/boot cycle, will need to be able to "switch Rio off", and we'll still want to get the benefit of soft updates when we do. Ideally, we would want a well-defined bottom end interface for VFS's that need to read and write at the block level, and the Rio code, the Soft Update code, and the system-supplied BIO would all use the same interface (the BIO only on the top end, the VFS only on the bottom end, and Rio and SU on both the top and bottom, so they can be stacked). Since this probably isn't going to happen in the near future without some architectural control concessions by FreeBSD, the next best thing to do is to understand the impact that integrating both will have on each other. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Wed Apr 2 10:14:54 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA13801 for hackers-outgoing; Wed, 2 Apr 1997 10:14:54 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA13794 for ; Wed, 2 Apr 1997 10:14:49 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id KAA13882 for hackers@freebsd.org; Wed, 2 Apr 1997 10:57:17 -0700 From: Terry Lambert Message-Id: <199704021757.KAA13882@phaeton.artisoft.com> Subject: Re: Internal clock To: hackers@freebsd.org Date: Wed, 2 Apr 1997 10:57:17 -0700 (MST) X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > We *have* a well designed and thought out device interface: devfs. > > > > I have *no* idea what is preventing you from making it the default > > mechanism. > > We still miss Terry's bugfixes for it. You aren't missing them, you're just not integrating them for fear of Lite2 (a fear now put to rest) and fear of their scope. > If you had ever used it (with options DEVFS_ROOT, it's already there), > you knew what the bugs are. Terry, the first 90 % of the work are > done with it, that's right. We are now in the phase that takes the > second 90 % of the development time... I've used it. For what it's worth, prior to the Lite2 merge in 3.0, my system didn't crash when I unmounted things. I've tried to make your system the same as mine, but you won't let me. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Wed Apr 2 10:15:08 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA13851 for hackers-outgoing; Wed, 2 Apr 1997 10:15:08 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA13841 for ; Wed, 2 Apr 1997 10:15:05 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id KAA13892 for hackers@freebsd.org; Wed, 2 Apr 1997 10:58:50 -0700 From: Terry Lambert Message-Id: <199704021758.KAA13892@phaeton.artisoft.com> Subject: Re: Code maintenance To: hackers@freebsd.org Date: Wed, 2 Apr 1997 10:58:50 -0700 (MST) In-Reply-To: <19970402091209.IH24661@uriah.heep.sax.de> from "J Wunsch" at Apr 2, 97 09:12:09 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > No. It never worked excitingly well, but nobody bothered to fix the > problems after the first imports. Also, in the ft(4) case, the > hardware on the market evolved _a lot_ after the initial driver, > making it only working for ancient crap nowadays. > > Terry, get real. Kernel interface changes have been the least part of > the problem, and look into the CVS history, were often the only reason > why someone has touched the source of that unmaintained code at all. > And, one of these kernel interface changes should be well known to > you: the inclusion of DEVFS. :) So if the kernel interfaces have not changed, the code still works, just not excitingly well? Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Wed Apr 2 10:28:56 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA14569 for hackers-outgoing; Wed, 2 Apr 1997 10:28:56 -0800 (PST) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA14562 for ; Wed, 2 Apr 1997 10:28:53 -0800 (PST) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id LAA14450; Wed, 2 Apr 1997 11:28:49 -0700 (MST) Date: Wed, 2 Apr 1997 11:28:49 -0700 (MST) Message-Id: <199704021828.LAA14450@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: hackers@freebsd.org Subject: 21" Monitor recommendations? X-Mailer: VM 6.22 under 19.15 XEmacs Lucid Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I'm in the market for some 21" monitors. I'm biased towards the Nanao/Eizo models (my 21" at work and my 17" at home are gorgeous), but the model I'm using at work is discontinued, and the replacement isn't nearly as nice. Does anyone have any recommendations for a *real* 21" monitor that will handle 1600x1200@72Hz, with a decent dot-pitch? My current monitor will handle 1600@1200@76Hz, and has a .25 dot-pitch, and I'd like something that is comparable. I'd like to spend under $2K/monitor if possible. Thanks! Nate From owner-freebsd-hackers Wed Apr 2 10:29:06 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA14601 for hackers-outgoing; Wed, 2 Apr 1997 10:29:06 -0800 (PST) Received: from Campino.Informatik.RWTH-Aachen.DE (campino.Informatik.RWTH-Aachen.DE [137.226.116.240]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA14592; Wed, 2 Apr 1997 10:29:04 -0800 (PST) Received: from gilberto.physik.rwth-aachen.de (gilberto.physik.rwth-aachen.de [137.226.31.2]) by Campino.Informatik.RWTH-Aachen.DE (RBI-Z-5/8.6.12) with ESMTP id UAA08403; Wed, 2 Apr 1997 20:29:12 +0200 (MET DST) Received: (from kuku@localhost) by gilberto.physik.rwth-aachen.de (8.8.5/8.6.9) id UAA00819; Wed, 2 Apr 1997 20:42:24 +0200 (MET DST) From: Christoph Kukulies Message-Id: <199704021842.UAA00819@gilberto.physik.rwth-aachen.de> Subject: Re: -current 'make world' chokes In-Reply-To: <199704021601.QAA710624@out2.ibm.net> from Steve Sims at "Apr 2, 97 10:58:25 am" To: SimsS@IBM.Net Date: Wed, 2 Apr 1997 20:42:24 +0200 (MET DST) Cc: Hackers@FreeBSD.ORG, freebsd-current@FreeBSD.ORG Reply-To: Christoph Kukulies X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > How come -current has, for the last couple of days (at least, that I've > *noticed*) chokes while making the world. > > I do 'cd /usr/src ; make update world' to pull current and build it and get: > > [blah, blah, blah] > > cd /usr/src/lib/csu/i386 && make beforeinstall > install -C -o bin -g bin -m 444 /usr/src/lib/csu/i386/dlfcn.h /usr/include > install -c -o bin -g bin -m 444 crt0.o c++rt0.o gcrt0.o scrt0.o sgcrt0.o > /usr/lib > install: crt0.o: No such file or directory > *** Error code 71 Got this too, today. Just that you're not feeling alone :-) > > Stop. > *** Error code 1 > > Stop. > *** Error code 1 > > Stop. > > # _ > > > cd'ing to /usr/src/lib and doing 'make' and/or 'make install' builds crt0.o > just fine, but it looks like it's getting skipped in the world context. > > It's kind of a pain to iteratively make this branch of the tree and then make > world with NOCLOBBER, NOCLEAN, etc... > > Am I totally missing the big picture here? > > ...sjs... > -- Christoph P. U. Kukulies kuku@gil.physik.rwth-aachen.de From owner-freebsd-hackers Wed Apr 2 10:31:28 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA14795 for hackers-outgoing; Wed, 2 Apr 1997 10:31:28 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA14770 for ; Wed, 2 Apr 1997 10:31:20 -0800 (PST) Received: from ntws (ntws.etinc.com [204.141.95.142]) by etinc.com (8.8.3/8.6.9) with SMTP id NAA24847; Wed, 2 Apr 1997 13:35:27 -0500 (EST) Message-Id: <3.0.32.19970402132955.00b708f0@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Wed, 02 Apr 1997 13:29:57 -0500 To: fyeung@fyeung8.netific.com (Francis Yeung) From: dennis Subject: Re: PPP Desperate for help! Cc: hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 10:07 AM 4/2/97 -0800, you wrote: >Dennis, > > The LCP magic number which is to be exchanged between both sides during >LCP negotiation. you said *magic number PROBLEM*...whats the problem? db > > Francis > > >> From dennis@etinc.com Wed Apr 2 09:42 PST 1997 >> X-Sender: dennis@etinc.com >> Date: Wed, 02 Apr 1997 11:34:17 -0500 >> To: fyeung@fyeung8.netific.com (Francis Yeung), brian@awfulhak.org >> From: dennis >> Subject: Re: PPP Desperate for help! >> Cc: dev@fgate.flevel.co.uk, freebsd-current@freebsd.org, >> freebsd-hackers@freebsd.org >> Mime-Version: 1.0 >> X-Loop: FreeBSD.org >> X-Fetchmail-Warning: no recipient addresses matched declared local names >> >> At 07:19 AM 4/2/97 -0800, Francis Yeung wrote: >> > >> > >> >Greetings, >> > >> > Will you have the magic number problem if you >> >connect 2 FreeBSD machines directly via PPP ? (both >> >machines are in LCP active) >> > >> > Francis >> >> What "magic number" problem are you referring to? >> >> db >> > >> >> From brian@awfulhak.org Tue Apr 1 14:32 PST 1997 >> >> To: dennis >> >> Cc: Developer , freebsd-current@freebsd.org, >> >> freebsd-hackers@freebsd.org >> >> Subject: Re: PPP Desperate for help! >> >> Mime-Version: 1.0 >> >> Date: Tue, 01 Apr 1997 20:49:07 +0100 >> >> From: Brian Somers >> >> X-Loop: FreeBSD.org >> >> X-Fetchmail-Warning: no recipient addresses matched declared local names >> >> >> >> > At 01:42 PM 4/1/97 +0100, Developer wrote: >> >> > > >> >> > > >> >> > >On Tue, 1 Apr 1997, Andrzej Bialecki wrote: >> >> > > >> >> > >> On Tue, 1 Apr 1997, Developer wrote: >> >> > >> >> >> > >> > >> >> > >> > We are really stuck trying to get the PPP user program (ppp) to >> >> > connect to >> >> > >> > our Perle 833 dial in server. We can connect using Windows NT >> fine but on >> >> > >> > BSD ppp logs in and then data will travel in both directions (As >> I can >> >> > see >> >> > >> > the modem lights flash at both ends for both send/recieve) but no >> data >> >> > >> > seems to actually get through as pings do not work. The packets >> from the >> >> > >> > server to the user, but seem to be lost on the way back. >> >> > >> >> >> > >> Let me take a guess: maybe it's a known problem with tcp_extensions >> set to >> >> > >> YES in /etc/sysconfig? Some broken TCP stacks don't like it, so the >> >> > >> solution would be to turn it off. >> >> > >> >> >> > >> Andy >> >> > > >> >> > >We've tried this - no difference. Any more ideas please? >> >> > >> >> > Why not post a trace of the ppp negotiations? Perhaps there is a >> parameter >> >> > conflict....are you getting to state IPCP_OPENED? If not, you cant pass >> >> > traffic. >> >> > >> >> > Dennis >> >> >> >> I suspect you're way ahead of the problem. Has the original poster >> >> tried "set openmode active" ? Without this, a client ppp will wait >> >> for the server to initiate LCP. >> >> >> >> I think it may be frugal to make this the default for both client >> >> *and* server. A lot of server implementations wait for the client >> >> to start. Any comments ? >> >> -- >> >> Brian , >> >> >> >> Don't _EVER_ lose your sense of humour.... >> >> >> >> >> > >> > > > From owner-freebsd-hackers Wed Apr 2 10:32:10 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA14895 for hackers-outgoing; Wed, 2 Apr 1997 10:32:10 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA14881 for ; Wed, 2 Apr 1997 10:32:06 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id LAA13944; Wed, 2 Apr 1997 11:13:45 -0700 From: Terry Lambert Message-Id: <199704021813.LAA13944@phaeton.artisoft.com> Subject: Re: devfs To: avalon@coombs.anu.edu.au (Darren Reed) Date: Wed, 2 Apr 1997 11:13:45 -0700 (MST) Cc: jkh@time.cdrom.com, proff@suburbia.net, hackers@FreeBSD.org In-Reply-To: <199704021415.GAA01403@freefall.freebsd.org> from "Darren Reed" at Apr 3, 97 00:10:35 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > if only name & permission persistance is sought, why can't devfs inherit > permissions from /dev and just change major/minor #'s (or is persistance > required there too) ? 1) What do you do with a "persistance" record for a pluggable (PCMCIA, etc.) device that is not currently present? 2) This requires that you maintain specfs for normal FS's. One of the main benefits of a devfs is that I can mount my root NFS from a system which does not support FreeBSD's bizarre idea of the number of bits in a major and minor number. Like, oh, all of them but FreeBSD. 3) This requires that the root FS be writeable. This is unsuitable for embedded applications, POS systems, and NC systems, among other applications. > maybe mounting devfs needs to be a boot "procedure" where name->permissions > ownership is transfered from /dev to /devices. Hmmm, what if /dev was a > sym. link to /dev.old (or was reset to that at boot time), then after > the devfs block was mounted under /devices and attributes transfered, > changed to point to /devices ? This persistance crap is just crap. If people cared about it, it should be possible to simply add an ioctl() that would return the location in KVM of the entry which created the devfs entry you are interested in changing, and then mung the real kernel image at that offset for the struct entries mode_t, uid_t, and gid_t to implement "persistance" as a permanent change to boot image data instead of something that's stored seperately from the kernel, for no good reason. The problem is that the people who care about persistance aren't the people writing the code -- the same problem FreeBSD has always had with IDE device, floppy tape devices, and anything else that works fine in another free OS but not in FreeBSD. If FreeBSD is to act historically consistent, then it would integrate devfs, and let the people who cared about persistance of permissions other than the default, either submit patches for the default values set on device registration (losing the need for persistance) or they can implement persistance and submit it on their own. If this doesn't seem to be a good idea, then getting rid of netiso was probably a bad idea with exactly the same character. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Wed Apr 2 10:32:55 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA14945 for hackers-outgoing; Wed, 2 Apr 1997 10:32:55 -0800 (PST) Received: from life.eecs.umich.edu (pmchen@life.eecs.umich.edu [141.213.8.32]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA14938 for ; Wed, 2 Apr 1997 10:32:49 -0800 (PST) Received: (from pmchen@localhost) by life.eecs.umich.edu (8.8.5/8.8.0) id NAA07782 for freebsd-hackers@freebsd.org; Wed, 2 Apr 1997 13:31:58 -0500 (EST) Date: Wed, 2 Apr 1997 13:31:58 -0500 (EST) From: "Peter M. Chen" Message-Id: <199704021831.NAA07782@life.eecs.umich.edu> To: freebsd-hackers@freebsd.org Subject: Re: question on buffer cache and VMIO Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >From terry@lambert.org Wed Apr 2 13:08:47 1997 > >All this leads to the question "How intrusive is Rio at the BIO level, >and what are the actual interface impacts that will result?". > >Effectively, those of us without battery-backed RAM, or RAM that is >somehow not persistant over a crash/boot cycle, will need to be >able to "switch Rio off", and we'll still want to get the benefit >of soft updates when we do. You're getting ahead of me--I'm still just investigating what it would take to put Rio in FreeBSD or another free Unix. Our implementation of Rio in Digital Unix caused minimal changes 1) turn off reliability-induced disk writes a) redirect bwrite calls to bdwrite b) the few places that need to write to disk (e.g. getnewbuf) call the original bwrite 2) maintain a self-contained registry of permanent data in memory. We may or may not do this in FreeBSD. 3) syncing to disk during reboot. We'd probably change this to sync during the crash in FreeBSD. 4) disable/enable write protection for the buffer cache. Our Digital Unix implementation made it really easy to turn on/off Rio while the system was running (basically via something like sysctl). Rio should have little effect on soft updates. Pete From owner-freebsd-hackers Wed Apr 2 10:34:36 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA15108 for hackers-outgoing; Wed, 2 Apr 1997 10:34:36 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA15096 for ; Wed, 2 Apr 1997 10:34:31 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id LAA13958; Wed, 2 Apr 1997 11:17:00 -0700 From: Terry Lambert Message-Id: <199704021817.LAA13958@phaeton.artisoft.com> Subject: Re: wd and funky geometry (understood?) To: dgy@rtd.com (Don Yuniskis) Date: Wed, 2 Apr 1997 11:17:00 -0700 (MST) Cc: bde@zeta.org.au, freebsd-hackers@freefall.freebsd.org In-Reply-To: <199704021457.HAA04618@seagull.rtd.com> from "Don Yuniskis" at Apr 2, 97 07:57:00 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk [ ... INIT'ed vs. reINIT'ed geometry ... ] I agree with Don. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Wed Apr 2 10:51:12 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA15903 for hackers-outgoing; Wed, 2 Apr 1997 10:51:12 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA15896 for ; Wed, 2 Apr 1997 10:51:08 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id UAA10025; Wed, 2 Apr 1997 20:50:42 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id UAA24374; Wed, 2 Apr 1997 20:35:10 +0200 (MET DST) Message-ID: <19970402203510.RR52594@uriah.heep.sax.de> Date: Wed, 2 Apr 1997 20:35:10 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@FreeBSD.ORG Cc: joe@pavilion.net (Josef Karthauser) Subject: Re: Internal clock References: <3340C326.6150@cybernet.com> <19970401151947.61221@pavilion.net> <19970401203429.XR37988@uriah.heep.sax.de> <19970402123111.58518@pavilion.net> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <19970402123111.58518@pavilion.net>; from Josef Karthauser on Apr 2, 1997 12:31:11 +0100 Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk As Josef Karthauser wrote: > > RTFM clocks(7). > > I have read this, although I still wasn't too clear about the above. > The fact that a system clock ticks in useconds doesn't mean that there's > going to be an interrupt every usec. You haven't read very carefully? :-) o The scheduling clock. This is a real clock with frequency that happens to be 100. It isn't available to applications. o The statistics clock. This is a real clock with frequency that happens to be 128. It isn't directly available to applications. So this explains the basic clock tick rates. There are exceptions, for example if profiling is active (tick rate 1024), or if a kernel device driver like pcaudio(4) (man page yet to be written by Søren :) raises the clock rate to 16000. The resolution for gettimeofday(2) is much higher than the tick rate of the scheduling clock since the values are interpolated. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Wed Apr 2 11:05:00 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA16736 for hackers-outgoing; Wed, 2 Apr 1997 11:05:00 -0800 (PST) Received: from DNS.Lamb.net (root@DNS.Lamb.net [207.90.181.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA16723 for ; Wed, 2 Apr 1997 11:04:55 -0800 (PST) Received: from Gatekeeper.Lamb.net (ulf@gatekeeper.Lamb.net [207.90.181.2]) by DNS.Lamb.net (8.8.5/20.74.3.14) with ESMTP id LAA29910; Wed, 2 Apr 1997 11:05:02 -0800 (PST) Received: (from ulf@localhost) by Gatekeeper.Lamb.net (8.8.5/8.7.6) id LAA21642; Wed, 2 Apr 1997 11:04:51 -0800 (PST) From: Ulf Zimmermann Message-Id: <199704021904.LAA21642@Gatekeeper.Lamb.net> Subject: Re: 21" Monitor recommendations? To: nate@mt.sri.com (Nate Williams) Date: Wed, 2 Apr 1997 11:04:51 -0800 (PST) Cc: hackers@freebsd.org In-Reply-To: <199704021828.LAA14450@rocky.mt.sri.com> from Nate Williams at "Apr 2, 97 11:28:49 am" X-Mailer: ELM [version 2.4ME+ PL22 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > I'm in the market for some 21" monitors. I'm biased towards the > Nanao/Eizo models (my 21" at work and my 17" at home are gorgeous), but > the model I'm using at work is discontinued, and the replacement isn't > nearly as nice. > > Does anyone have any recommendations for a *real* 21" monitor that will > handle 1600x1200@72Hz, with a decent dot-pitch? My current monitor will > handle 1600@1200@76Hz, and has a .25 dot-pitch, and I'd like something > that is comparable. I'd like to spend under $2K/monitor if possible. > > Thanks! > > > > Nate > Take a look at the Sony GDM-20E(?), SGI uses it now as standard monitor. It does 1600x1200@76 (96KHz maximum). I am not sure about the dot pitch. It has 2 inputs. Right I am using one at home with a Matrox Millenium. It runs fine at tha resolution. Ulf. -------------------------------------------------------------------------- Ulf Zimmermann, 1525 Pacific Ave., Alameda, CA-94501, #: 510-769-2936 Alameda Networks, Inc. | http://www.Alameda.net From owner-freebsd-hackers Wed Apr 2 11:31:01 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA18170 for hackers-outgoing; Wed, 2 Apr 1997 11:31:01 -0800 (PST) Received: from softway95.softway.com (softway95.softway.com [206.80.1.38]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA18164; Wed, 2 Apr 1997 11:30:59 -0800 (PST) Received: (from renaud@localhost) by softway95.softway.com (8.8.5/8.6.12) id LAA28404; Wed, 2 Apr 1997 11:04:11 -0800 (PST) From: Renaud Waldura Message-Id: <199704021904.LAA28404@softway95.softway.com> Subject: Re: Apache lingering_close error To: marcs@znep.com (Marc Slemko) Date: Wed, 2 Apr 1997 11:04:10 -0800 (PST) Cc: rw@softway.com, freebsd-isp@freebsd.org, freebsd-hackers@freebsd.org In-Reply-To: from "Marc Slemko" at Apr 1, 97 06:39:30 pm Reply-To: Renaud Waldura Organization: Softway, Inc. (San Francisco) X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Marc Slemko wrote: > > There is no reason to disable linging_close for FreeBSD[1]. The message > is simply a combination between a warning and a not-yet-removed debugging > proxy code that I haven't had time to follow, but nothing that hurts more > than outputting that message sometimes. It should not hurt anything. We OK. You're a little bit difficult to follow but I think I got the main message; sounds like a complicated issue. Thanks. I commented out the printing; error messages bug me. *** http_main.c Wed Apr 2 10:54:35 1997 --- http_main.c.old Fri Mar 14 08:38:53 1997 *************** *** 338,348 **** if (((shutdown(lsd, 1)) != 0) || (fcntl(lsd, F_SETFL, FNDELAY) == -1)) { /* if it fails, no need to go through the rest of the routine */ if (errno != ENOTCONN) - /* RW Wed Apr 2 10:52:36 PST 1997 - This error message is actually useless (from the author) log_unixerr("shutdown", NULL, "lingering_close", r->server); - */ - ; bclose(r->connection->client); kill_timeout(r); return; --- 338,344 ---- -- -- Renaud Waldura -- -- Softway International, Inc -- -- rw@softway.com -- -- 185 Berry Street, Suite 5514 -- -- San Francisco, CA 94107 -- -- USA -- -- Tel (415) 896-0708 -- Fax (415) 896-0709 -- http://www.softway.com/ -- -- From owner-freebsd-hackers Wed Apr 2 11:31:06 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA18203 for hackers-outgoing; Wed, 2 Apr 1997 11:31:06 -0800 (PST) Received: from crh.cl.msu.edu (crh.cl.msu.edu [35.8.1.24]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA18145 for ; Wed, 2 Apr 1997 11:30:43 -0800 (PST) Received: (from henrich@localhost) by crh.cl.msu.edu (8.8.5/8.8.4) id OAA16118; Wed, 2 Apr 1997 14:30:22 -0500 (EST) Date: Wed, 2 Apr 1997 14:30:22 -0500 (EST) From: Charles Henrich Message-Id: <199704021930.OAA16118@crh.cl.msu.edu> To: nate@mt.sri.com, freebsd-hackers@freebsd.org Subject: Re: 21" Monitor recommendations? Newsgroups: lists.freebsd.hackers References: <5hubcs$els$1@msunews.cl.msu.edu> X-Newsreader: NN version 6.5.0 #1 (NOV) Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In lists.freebsd.hackers you write: >I'm in the market for some 21" monitors. I'm biased towards the >Nanao/Eizo models (my 21" at work and my 17" at home are gorgeous), but >the model I'm using at work is discontinued, and the replacement isn't >nearly as nice. >Does anyone have any recommendations for a *real* 21" monitor that will >handle 1600x1200@72Hz, with a decent dot-pitch? My current monitor will >handle 1600@1200@76Hz, and has a .25 dot-pitch, and I'd like something >that is comparable. I'd like to spend under $2K/monitor if possible. I run a NEC XP21 at 1600x1200@72 (maybe @70 ?) in any case, its extremely nice. Very crisp clean picture, and has been for several years now. The NEC's are still a bit pricey I think. It has a .25dp if I remember correctly. -Crh -- Charles Henrich Michigan State University henrich@msu.edu http://pilot.msu.edu/~henrich From owner-freebsd-hackers Wed Apr 2 11:33:10 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA18360 for hackers-outgoing; Wed, 2 Apr 1997 11:33:10 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA18342 for ; Wed, 2 Apr 1997 11:33:03 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id VAA10443 for hackers@freebsd.org; Wed, 2 Apr 1997 21:33:01 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id VAA24609; Wed, 2 Apr 1997 21:08:17 +0200 (MET DST) Message-ID: <19970402210817.YJ08249@uriah.heep.sax.de> Date: Wed, 2 Apr 1997 21:08:17 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freebsd.org Subject: Re: Internal clock References: <199704021757.KAA13882@phaeton.artisoft.com> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199704021757.KAA13882@phaeton.artisoft.com>; from Terry Lambert on Apr 2, 1997 10:57:17 -0700 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Terry Lambert wrote: > > If you had ever used it (with options DEVFS_ROOT, it's already there), > > you knew what the bugs are. > I've used it. For what it's worth, prior to the Lite2 merge in 3.0, > my system didn't crash when I unmounted things. That's not the biggest problem. Right now, it suffers serious aliasing problems that even prevent the system to start at all (it doesn't find its root filesystem entry, or rather, the vnode is being revoked while opening the root device, resulting in an EIO when trying to mount it). Once i've sorted these problems out, i'll continue to make DEVFS usable as a default. If persistancy is once the last remaining problem, and nobody is stepping forward to implement the all-singing- all-dancing union layer shadowing solution, i'll probably implement a startup/shutdown script-based solution. If something better comes up later, it can still be replaced. I think FreeBSD 3.0 might be shipping with DEVFS as the default, but it's still ``quarter o'mile'' to go. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Wed Apr 2 11:33:13 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA18386 for hackers-outgoing; Wed, 2 Apr 1997 11:33:13 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA18361 for ; Wed, 2 Apr 1997 11:33:10 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id VAA10450 for hackers@freebsd.org; Wed, 2 Apr 1997 21:33:08 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id VAA24629; Wed, 2 Apr 1997 21:11:57 +0200 (MET DST) Message-ID: <19970402211157.WU57156@uriah.heep.sax.de> Date: Wed, 2 Apr 1997 21:11:57 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freebsd.org Subject: Re: Code maintenance References: <19970402091209.IH24661@uriah.heep.sax.de> <199704021758.KAA13892@phaeton.artisoft.com> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199704021758.KAA13892@phaeton.artisoft.com>; from Terry Lambert on Apr 2, 1997 10:58:50 -0700 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Terry Lambert wrote: > So if the kernel interfaces have not changed, the code still works, > just not excitingly well? Yes, aic(4) and ft(4) still work, more or less. We'd probably have removed them otherwise. But in particular ft(4) suffers from a not very elegant design, and its unmaintained state resulted in it being almost worthless over time, since it doesn't handle the more modern drives that are available now at all. If it remains unmaintained, the day will come when all the previously existing drives it used to support on the earth are dead. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Wed Apr 2 11:34:18 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA18512 for hackers-outgoing; Wed, 2 Apr 1997 11:34:18 -0800 (PST) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA18505 for ; Wed, 2 Apr 1997 11:34:13 -0800 (PST) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by mail.cdsnet.net (8.8.5/8.7.3) with SMTP id LAA18161; Wed, 2 Apr 1997 11:32:56 -0800 (PST) Date: Wed, 2 Apr 1997 11:32:56 -0800 (PST) From: Jaye Mathisen To: Terry Lambert cc: hackers@freebsd.org Subject: Re: devfs In-Reply-To: <199704021813.LAA13944@phaeton.artisoft.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk And in a suprising move, as one of those people who unfortunately doesn't have the necessary talent to contribute code, but does use FreeBSD extensively, I agree with terry from the standpoint of letting the users put in scripts in rc or whatever to maintain the persistence, and not incorporating it into devfs. Even though it would mean a small amount of extra work for myself tweaking perms and such. From owner-freebsd-hackers Wed Apr 2 11:40:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA19031 for hackers-outgoing; Wed, 2 Apr 1997 11:40:11 -0800 (PST) Received: from austin.polstra.com (austin.polstra.com [206.213.73.10]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA18980 for ; Wed, 2 Apr 1997 11:40:00 -0800 (PST) Received: from austin.polstra.com (jdp@localhost) by austin.polstra.com (8.8.5/8.8.5) with ESMTP id LAA09985; Wed, 2 Apr 1997 11:39:38 -0800 (PST) Message-Id: <199704021939.LAA09985@austin.polstra.com> To: nate@mt.sri.com Subject: Re: 21" Monitor recommendations? Newsgroups: polstra.freebsd.hackers In-Reply-To: <199704021828.LAA14450@rocky.mt.sri.com> References: <199704021828.LAA14450@rocky.mt.sri.com> Organization: Polstra & Co., Seattle, WA Cc: hackers@freebsd.org Date: Wed, 02 Apr 1997 11:39:37 -0800 From: John Polstra Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In article <199704021828.LAA14450@rocky.mt.sri.com>, Nate Williams wrote: > Does anyone have any recommendations for a *real* 21" monitor that will > handle 1600x1200@72Hz, with a decent dot-pitch? My current monitor will > handle 1600@1200@76Hz, and has a .25 dot-pitch, and I'd like something > that is comparable. I'd like to spend under $2K/monitor if possible. The Nokia monitors are very nice. My 21" one can handle 1600x1200@80Hz. It's about 1.5 years old -- model 445X. I've been happy with it. I think they might cost more than $2K, though. -- John Polstra jdp@polstra.com John D. Polstra & Co., Inc. Seattle, Washington USA "Self-knowledge is always bad news." -- John Barth From owner-freebsd-hackers Wed Apr 2 11:50:42 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA19477 for hackers-outgoing; Wed, 2 Apr 1997 11:50:42 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA19470 for ; Wed, 2 Apr 1997 11:50:36 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id VAA10646; Wed, 2 Apr 1997 21:50:35 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id VAA24750; Wed, 2 Apr 1997 21:24:43 +0200 (MET DST) Message-ID: <19970402212443.OT00003@uriah.heep.sax.de> Date: Wed, 2 Apr 1997 21:24:43 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: freebsd-hackers@freebsd.org Cc: keith4@ilink.nis.za (keith waters), taob@nbc.netcom.ca (Brian Tao) Subject: Re: Unreasonable rebuild time on large password files References: <199704020913.LAA03056@ilink.nis.za> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: ; from Brian Tao on Apr 2, 1997 10:54:47 -0500 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Brian Tao wrote: > > This must be one of the worst cases of code/resourse optimisation, > > since changing a password means rebuilding EVERY SINGLE entry in the > > password databases. > > We await your optimized code diffs to pwd_mkdb(8) then. :) No need. It's already there. Guido once implemented incremental updates to the password database if only a single user's entry has been changed. They seem to be non-optional (much to my surprise :), and the default in FreeBSD 2.2 and above. Nobody thought of this change when assembling the release notes for 2.2, not even the author. ;-) -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Wed Apr 2 11:52:04 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA19668 for hackers-outgoing; Wed, 2 Apr 1997 11:52:04 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA19658 for ; Wed, 2 Apr 1997 11:52:00 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id VAA10658 for hackers@FreeBSD.ORG; Wed, 2 Apr 1997 21:51:57 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id VAA24807; Wed, 2 Apr 1997 21:46:27 +0200 (MET DST) Message-ID: <19970402214627.DT15556@uriah.heep.sax.de> Date: Wed, 2 Apr 1997 21:46:27 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@FreeBSD.ORG Subject: Re: Does de driver do 100MBIT Full Duplex? References: <3.0.32.19970402091819.00695824@etinc.com> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <3.0.32.19970402091819.00695824@etinc.com>; from dennis on Apr 2, 1997 09:18:26 -0500 Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk As dennis wrote: > Is the driver that you send me directly for 2.1.7 (that worked > beautifully) in 2.2.1R? If not, WHY NOT! No need to shout. We aren't deaf exactly. Since Matt released it _after_ 2.2 has been cut. I've got a very explicit message from Matt when i've been asking him earlier (right in time to get something into 2.2 still) about a new version, that he considered the stuff that sneaked into NetBSD by that time too buggy to see it officially in FreeBSD. That's why we've been integrating the little hack still to make at least the DE21140A supported. We originally deferred the inclusion of this patch in anticipation of the new driver version. As you can see, it's not that we're sleeping or wouldn't care about something that can be considered one of the major ethernet NIC drivers in *BSD. Also, as i understand it, Matt's new driver won't fit without a little work into 2.2 or higher, due to the new media selection method. This work simply needs to be done, and it requires a volunteer. (I've got one potential volunteer, but he wouldn't be able to do it anytime soon. There are not many people volunteering for work, and these are seldom the people who shout in a mailing list. :) -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Wed Apr 2 12:27:10 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA21678 for hackers-outgoing; Wed, 2 Apr 1997 12:27:10 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA21670 for ; Wed, 2 Apr 1997 12:27:02 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id NAA14311; Wed, 2 Apr 1997 13:10:27 -0700 From: Terry Lambert Message-Id: <199704022010.NAA14311@phaeton.artisoft.com> Subject: Re: Code maintenance To: joerg_wunsch@uriah.heep.sax.de Date: Wed, 2 Apr 1997 13:10:26 -0700 (MST) Cc: hackers@freebsd.org In-Reply-To: <19970402211157.WU57156@uriah.heep.sax.de> from "J Wunsch" at Apr 2, 97 09:11:57 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk [ ... ft driver ... ] > If it remains unmaintained, the > day will come when all the previously existing drives it used to > support on the earth are dead. Heh. If the end of ft is coming on that fast, then we must be nearing the time when we can delete all support for non-PnP ISA machines. 8-). Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Wed Apr 2 12:36:13 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA22200 for hackers-outgoing; Wed, 2 Apr 1997 12:36:13 -0800 (PST) Received: from iworks.InterWorks.org (deischen@iworks.interworks.org [128.255.18.10]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA22195 for ; Wed, 2 Apr 1997 12:36:09 -0800 (PST) Received: (from deischen@localhost) by iworks.InterWorks.org (8.7.5/) id OAA00583; Wed, 2 Apr 1997 14:36:06 -0600 (CST) Message-Id: <199704022036.OAA00583@iworks.InterWorks.org> Date: Wed, 2 Apr 1997 14:36:06 -0600 (CST) From: "Daniel M. Eischen" To: freebsd-hackers@freefall.freebsd.org, nate@mt.sri.com Subject: Re: 21" Monitor recommendations? Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > I'm in the market for some 21" monitors. I'm biased towards the > Nanao/Eizo models (my 21" at work and my 17" at home are gorgeous), but > the model I'm using at work is discontinued, and the replacement isn't > nearly as nice. > > Does anyone have any recommendations for a *real* 21" monitor that will > handle 1600x1200@72Hz, with a decent dot-pitch? My current monitor will > handle 1600@1200@76Hz, and has a .25 dot-pitch, and I'd like something > that is comparable. I'd like to spend under $2K/monitor if possible. I've had good luck with ViewSonic. I have a 21PS at home and work and have no problems with it. We also have a Nokia 447x at work that looks really nice. The new ViewSonic P815 is suppose to do 1800x1440 at 75Hz with .25dp. I've seen it for under $2K. Dan Eischen deischen@iworks.InterWorks.org From owner-freebsd-hackers Wed Apr 2 12:43:32 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA22808 for hackers-outgoing; Wed, 2 Apr 1997 12:43:32 -0800 (PST) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA22798 for ; Wed, 2 Apr 1997 12:43:27 -0800 (PST) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id NAA15377; Wed, 2 Apr 1997 13:43:19 -0700 (MST) Date: Wed, 2 Apr 1997 13:43:19 -0700 (MST) Message-Id: <199704022043.NAA15377@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Jaye Mathisen Cc: hackers@freebsd.org Subject: Re: devfs In-Reply-To: References: <199704021813.LAA13944@phaeton.artisoft.com> X-Mailer: VM 6.22 under 19.15 XEmacs Lucid Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Jaye Mathisen writes: > > > And in a suprising move, as one of those people who unfortunately doesn't > have the necessary talent to contribute code I for one won't stand for that statement. You know as well as I do that you have all the necessary talent (more than I for sure, and even I've done some kernel hacking), you're just too darn lazy to help. (Or was that busy? *grin*) Nate From owner-freebsd-hackers Wed Apr 2 12:45:15 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA22961 for hackers-outgoing; Wed, 2 Apr 1997 12:45:15 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA22953 for ; Wed, 2 Apr 1997 12:45:12 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id MAA13374; Wed, 2 Apr 1997 12:44:51 -0800 (PST) To: Darren Reed cc: proff@suburbia.net, hackers@freebsd.org Subject: Re: devfs In-reply-to: Your message of "Thu, 03 Apr 1997 00:10:35 +1000." <199704021415.GAA12419@time.cdrom.com> Date: Wed, 02 Apr 1997 12:44:50 -0800 Message-ID: <13370.860013890@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > OSx5.1 (ye olde syvr3/4.2bsd mix) from Pyramid nhad a "conditional" > symbolic link (if I remember right) worked something like this: Yeah, and you also had this stupid "universe" command for selecting the symlink value and you could only have a predefined # of universes. No thanks, I wanted something a little more general. :-) Jordan From owner-freebsd-hackers Wed Apr 2 12:50:38 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA23319 for hackers-outgoing; Wed, 2 Apr 1997 12:50:38 -0800 (PST) Received: from smtp.enteract.com (qmailr@char-star.rdist.org [206.54.252.22]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA23302 for ; Wed, 2 Apr 1997 12:50:34 -0800 (PST) Received: (qmail 17705 invoked from network); 2 Apr 1997 20:43:49 -0000 Received: from enteract.com (stox@206.54.252.1) by char-star.rdist.org with SMTP; 2 Apr 1997 20:43:49 -0000 Date: Wed, 2 Apr 1997 14:43:48 -0600 (CST) From: Ken Stox To: Charles Henrich cc: nate@mt.sri.com, freebsd-hackers@freebsd.org Subject: Re: 21" Monitor recommendations? In-Reply-To: <199704021930.OAA16118@crh.cl.msu.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I'm running the Panasonic E21 1600x1200x72Hz. Looks great, and was about $00 cheaper than the NEC. I paid ~$1500 at Best Buy. I'd recommend it. -Ken Stox On Wed, 2 Apr 1997, Charles Henrich wrote: > In lists.freebsd.hackers you write: > > >I'm in the market for some 21" monitors. I'm biased towards the > >Nanao/Eizo models (my 21" at work and my 17" at home are gorgeous), but > >the model I'm using at work is discontinued, and the replacement isn't > >nearly as nice. > > >Does anyone have any recommendations for a *real* 21" monitor that will > >handle 1600x1200@72Hz, with a decent dot-pitch? My current monitor will > >handle 1600@1200@76Hz, and has a .25 dot-pitch, and I'd like something > >that is comparable. I'd like to spend under $2K/monitor if possible. > > I run a NEC XP21 at 1600x1200@72 (maybe @70 ?) in any case, its extremely nice. > Very crisp clean picture, and has been for several years now. The NEC's are > still a bit pricey I think. It has a .25dp if I remember correctly. > > -Crh > -- > > Charles Henrich Michigan State University henrich@msu.edu > > http://pilot.msu.edu/~henrich > From owner-freebsd-hackers Wed Apr 2 12:55:52 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA23645 for hackers-outgoing; Wed, 2 Apr 1997 12:55:52 -0800 (PST) Received: from squirrel.tgsoft.com (squirrel.tgsoft.com [207.167.64.183]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA23632 for ; Wed, 2 Apr 1997 12:55:47 -0800 (PST) Received: (qmail 11073 invoked by uid 128); 2 Apr 1997 20:55:13 -0000 Date: 2 Apr 1997 20:55:13 -0000 Message-ID: <19970402205513.11072.qmail@squirrel.tgsoft.com> From: mark thompson To: hackers@freefall.freebsd.org In-reply-to: message from Terry Lambert on Tue, 1 Apr 1997 16:51:39 -0700 (MST) Subject: Re: Code maintenance Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk From: Terry Lambert Date: Tue, 1 Apr 1997 16:51:39 -0700 (MST) > > > Then adding the code into the tree is a 'bad thing', since it becomes > > > unsupported. If no-one is willing to incorporate/support the code, then > > > it shouldn't be incorporated. > > > > Better that it be lost forever? > > Yes, see the ft(4) and aic(4) drivers for two good examples. Both > sorely lack a maintainer, and nobody seems to be interested in doing > this job. Hmmm.... i have volunteered before to maintain ft, since tapes are kind of a (twisted) specialty of mine. My only request is that someone donate the drives to test on, since i don't use QIC format any more. My offer to help kind of got blown off. I understand that everyone is up to their respective keisters in release engineering, but it seems pretty clear to me that there is not percieved to be a crying need for this alleged maintenance. -mark From owner-freebsd-hackers Wed Apr 2 13:14:09 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA24428 for hackers-outgoing; Wed, 2 Apr 1997 13:14:09 -0800 (PST) Received: from awfulhak.demon.co.uk (awfulhak.demon.co.uk [158.152.17.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA24388; Wed, 2 Apr 1997 13:13:53 -0800 (PST) Received: from awfulhak.demon.co.uk (localhost.lan.awfulhak.org [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id WAA14851; Wed, 2 Apr 1997 22:07:01 +0100 (BST) Message-Id: <199704022107.WAA14851@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: Developer cc: freebsd-current@freebsd.org, freebsd-hackers@freebsd.org Subject: Re: PPP Problems -- The next chapter *grin* In-reply-to: Your message of "Wed, 02 Apr 1997 10:17:54 BST." Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 02 Apr 1997 22:07:01 +0100 From: Brian Somers Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > After not being able to get the user ppp software to work I decided to > give up and try pppd instead. After configuring correctly pppd works fine > (Just use kermit to dial and then run it)?? > > I wonder what is different between ppp and pppd that would cause the > problem? > > Regards, > > Trefor S. The answer would be shorter if you asked what isn't different. Were you using the pmdemand example and reading the handbook tutorial ? Was it >= 2.2 ? -- Brian , Don't _EVER_ lose your sense of humour.... From owner-freebsd-hackers Wed Apr 2 13:14:09 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA24432 for hackers-outgoing; Wed, 2 Apr 1997 13:14:09 -0800 (PST) Received: from awfulhak.demon.co.uk (awfulhak.demon.co.uk [158.152.17.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA24398 for ; Wed, 2 Apr 1997 13:14:03 -0800 (PST) Received: from awfulhak.demon.co.uk (localhost.lan.awfulhak.org [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id WAA14871; Wed, 2 Apr 1997 22:09:44 +0100 (BST) Message-Id: <199704022109.WAA14871@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: mika ruohotie cc: macgyver@db-net.com (Wilson MacGyver), hackers@freebsd.org Subject: Re: p166 vs. p166mmx In-reply-to: Your message of "Wed, 02 Apr 1997 15:19:20 +0300." <199704021219.PAA03426@shadows.aeon.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 02 Apr 1997 22:09:44 +0100 From: Brian Somers Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > make -k world 7595.59s user 1843.94s system 87% cpu 3:00:48.28 total > > > make -k world 5982.25s user 1636.02s system 87% cpu 2:24:21.88 total > > Wouldn't this be because of the larger L1 cache? If memory > > serves, L1 cache was increased from 16K to 32K... > > yes, but still i didnt expect the difference to be about 21% for user and 17% > for total time... > > > mickey Hmm, I don't know. I heard reports that a PP150 w/ 512k cache was almost as good as a PP200 2/ 256k. -- Brian , Don't _EVER_ lose your sense of humour.... From owner-freebsd-hackers Wed Apr 2 13:15:49 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA24603 for hackers-outgoing; Wed, 2 Apr 1997 13:15:49 -0800 (PST) Received: from awfulhak.demon.co.uk (awfulhak.demon.co.uk [158.152.17.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA24580; Wed, 2 Apr 1997 13:15:43 -0800 (PST) Received: from awfulhak.demon.co.uk (localhost.lan.awfulhak.org [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id VAA14484; Wed, 2 Apr 1997 21:51:59 +0100 (BST) Message-Id: <199704022051.VAA14484@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: dennis cc: Brian Somers , Developer , freebsd-current@freebsd.org, freebsd-hackers@freebsd.org Subject: Re: PPP Desperate for help! In-reply-to: Your message of "Tue, 01 Apr 1997 19:07:32 CDT." <3.0.32.19970401190724.00b1d924@etinc.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 02 Apr 1997 21:51:58 +0100 From: Brian Somers Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > >I think it may be frugal to make this the default for both client > >*and* server. A lot of server implementations wait for the client > >to start. Any comments ? > > It shouldnt make a difference, if implemented properly. There is > no master-slave relationship in ppp...the ends are peers. Upon > starting the interface it should send "n" configs, and then stop > if no reply is received. When a config is received it starts again. The > state machine allows for this nicely. One problem is when the electrical > interface (UP) is not implemented (ie, receipt of DSR) properly. You should > not be able to get into both ends being dormant if its done properly. > > Dennis But the problem is (and I experienced this myself with my ISP) that ppp in "call-up" mode won't start LCP. My ISP will not start either, so I make the call and we both sit there looking at eachother, grinning in a way that only ppp connections can. With "set openmode active", LCP is initiated irrespective. -- Brian , Don't _EVER_ lose your sense of humour.... From owner-freebsd-hackers Wed Apr 2 13:15:59 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA24659 for hackers-outgoing; Wed, 2 Apr 1997 13:15:59 -0800 (PST) Received: from awfulhak.demon.co.uk (awfulhak.demon.co.uk [158.152.17.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA24627 for ; Wed, 2 Apr 1997 13:15:53 -0800 (PST) Received: from awfulhak.demon.co.uk (localhost.lan.awfulhak.org [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id VAA14510; Wed, 2 Apr 1997 21:59:35 +0100 (BST) Message-Id: <199704022059.VAA14510@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: Darren Reed cc: jkh@time.cdrom.com (Jordan K. Hubbard), proff@suburbia.net, hackers@freebsd.org Subject: Re: devfs In-reply-to: Your message of "Thu, 03 Apr 1997 00:10:35 +1000." <199704021415.GAA01403@freefall.freebsd.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 02 Apr 1997 21:59:35 +0100 From: Brian Somers Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > In some mail from Jordan K. Hubbard, sie said: > > > > > 1) accept that devfs in it's present form is a kludge. Move > > > all the default permissions and device numbers into a > > > sysctl hierarchy. This hierarchy is then used to create > > > names in /dev. Add a new ufs file type: S_IFLNKDEV. > > > When a namei occurs on what this pseudo-symlink points > > > to, it looks up name in the sysctl and returns the inode > > > of the "symlink" with device numbers filled in. The > > > > I'm not sure if I care much for this as a solution to the persistance > > problem, at least not unless you were also talking about having DEVFS > > manipulate this piece of the sysctl namespace on behalf of the user > > for any modifications they may make. It has to be transparent after > > setup, right? > > > > Also, I don't think that anyone has addressed the issue of sysctl MIB > > persistance either yet, so your changes would still go away on reboot. :-) > > > > Still, what you *do* raise here is an interesting, possibly even > > downright evil, idea (and I like the evil ones best ;). How about a > > new symlink type which is just like a classic symlink in every way > > except that its value, rather than being direct fodder for namei(), is > > instead used to look something up in sysctl, the value of THAT being > > then passed on to namei()? You could get variant symlink behavior by > > poking things into sysctl. :-) > > OSx5.1 (ye olde syvr3/4.2bsd mix) from Pyramid nhad a "conditional" > symbolic link (if I remember right) worked something like this: > > ln -c att /.attbin -c ucb /.ucbbin /bin ln -s /bin att=/.attbin ucb=/.ucbbin It was horrible because att didn't know about symlinks. All it takes is someone to symlink . to .. and att can't hack it. > where 'att" and "ucb" where two "universes". Depending on which one > you were in at the time (process property) /bin was either /.attbin > or /.ucbbin. [.....] You could (if it worked) mount a unionfs on top of devfs if you want any sort of persistence. It'd be nice and easy to "clean up" dev too ! -- Brian , Don't _EVER_ lose your sense of humour.... From owner-freebsd-hackers Wed Apr 2 13:16:26 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA24772 for hackers-outgoing; Wed, 2 Apr 1997 13:16:26 -0800 (PST) Received: from awfulhak.demon.co.uk (awfulhak.demon.co.uk [158.152.17.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA24767; Wed, 2 Apr 1997 13:16:22 -0800 (PST) Received: from awfulhak.demon.co.uk (localhost.lan.awfulhak.org [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id VAA14184; Wed, 2 Apr 1997 21:48:17 +0100 (BST) Message-Id: <199704022048.VAA14184@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: fyeung@fyeung8.netific.com (Francis Yeung) cc: dennis@etinc.com, brian@awfulhak.org, dev@fgate.flevel.co.uk, freebsd-current@freebsd.org, freebsd-hackers@freebsd.org Subject: Re: PPP Desperate for help! In-reply-to: Your message of "Wed, 02 Apr 1997 07:19:10 -0800." <9704021519.AA16536@fyeung8.netific.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 02 Apr 1997 21:48:16 +0100 From: Brian Somers Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > Greetings, > > Will you have the magic number problem if you > connect 2 FreeBSD machines directly via PPP ? (both > machines are in LCP active) > > Francis Nope - check out GenerateMagic in lcp.c. It's not exactly random, but you'd be pretty lucky to have problems. -- Brian , Don't _EVER_ lose your sense of humour.... From owner-freebsd-hackers Wed Apr 2 13:17:30 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA24842 for hackers-outgoing; Wed, 2 Apr 1997 13:17:30 -0800 (PST) Received: from relay-7.mail.demon.net (relay-7.mail.demon.net [194.217.242.9]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id NAA24836 for ; Wed, 2 Apr 1997 13:17:27 -0800 (PST) Received: from awfulhak.demon.co.uk ([158.152.17.1]) by relay-5.mail.demon.net id aa0528772; 2 Apr 97 22:14 BST Received: from awfulhak.demon.co.uk (localhost.lan.awfulhak.org [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id WAA14837; Wed, 2 Apr 1997 22:04:52 +0100 (BST) Message-Id: <199704022104.WAA14837@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: proff@suburbia.net cc: Darren Reed , hackers@freebsd.org Subject: Re: devfs In-reply-to: Your message of "Thu, 03 Apr 1997 01:21:14 +1000." <19970402152114.27776.qmail@suburbia.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 02 Apr 1997 22:04:52 +0100 From: Brian Somers Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Yes, I remember my Pyramid fondly. It not only provided two universes, but > heated the whole house as well, drinking around $10 a day in power :) [.....] > Cheers, > Julian. Hey, don't knock it ! The company I work for has just signed a contract with pyramid (gulp). It's sad, they used to have all Sequent boxes 'till I finally managed to get them to agree to replace the development machine with a P100 running FreeBSD. Builds (using my mk program, not make) weren't just an order of magnitude quicker, they were 30 times quicker (mostly due to the unified vm stuff). Ok, the old machine was a 4x386 box... -- Brian , Don't _EVER_ lose your sense of humour.... From owner-freebsd-hackers Wed Apr 2 13:18:43 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA25029 for hackers-outgoing; Wed, 2 Apr 1997 13:18:43 -0800 (PST) Received: from awfulhak.demon.co.uk (awfulhak.demon.co.uk [158.152.17.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA25020; Wed, 2 Apr 1997 13:18:36 -0800 (PST) Received: from awfulhak.demon.co.uk (localhost.lan.awfulhak.org [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id VAA14200; Wed, 2 Apr 1997 21:49:22 +0100 (BST) Message-Id: <199704022049.VAA14200@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: dennis cc: fyeung@fyeung8.netific.com (Francis Yeung), brian@awfulhak.org, dev@fgate.flevel.co.uk, freebsd-current@FreeBSD.org, freebsd-hackers@FreeBSD.org Subject: Re: PPP Desperate for help! In-reply-to: Your message of "Wed, 02 Apr 1997 11:34:17 CDT." <3.0.32.19970402113410.00b716e0@etinc.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 02 Apr 1997 21:49:22 +0100 From: Brian Somers Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk > At 07:19 AM 4/2/97 -0800, Francis Yeung wrote: > > > > > >Greetings, > > > > Will you have the magic number problem if you > >connect 2 FreeBSD machines directly via PPP ? (both > >machines are in LCP active) > > > > Francis > > What "magic number" problem are you referring to? > > db >From rfc1663: [.....] Every link in a single machine MUST have different Magic Numbers, and each end of every link between two peers SHOULD have Magic Numbers which are unique to those peers. This protects against patch-panel errors in addition to looped-back links. [.....] -- Brian , Don't _EVER_ lose your sense of humour.... From owner-freebsd-hackers Wed Apr 2 14:00:34 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA27988 for hackers-outgoing; Wed, 2 Apr 1997 14:00:34 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id OAA27978 for ; Wed, 2 Apr 1997 14:00:28 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id OAA14577; Wed, 2 Apr 1997 14:41:38 -0700 From: Terry Lambert Message-Id: <199704022141.OAA14577@phaeton.artisoft.com> Subject: Re: devfs To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Wed, 2 Apr 1997 14:41:38 -0700 (MST) Cc: avalon@coombs.anu.edu.au, proff@suburbia.net, hackers@freebsd.org In-Reply-To: <13370.860013890@time.cdrom.com> from "Jordan K. Hubbard" at Apr 2, 97 12:44:50 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > OSx5.1 (ye olde syvr3/4.2bsd mix) from Pyramid nhad a "conditional" > > symbolic link (if I remember right) worked something like this: > > Yeah, and you also had this stupid "universe" command for > selecting the symlink value and you could only have a predefined > # of universes. No thanks, I wanted something a little more general. :-) Move the process environment into the kernel on the other side of a system call, and you can have all the variant links your crooked little heart desires. The big problem is the POSIX execve() families envp parameter; if you can decide to do away with that, or to handle it by importing it into the template process, then the last remaining issue is resolved. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Wed Apr 2 14:07:39 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA28339 for hackers-outgoing; Wed, 2 Apr 1997 14:07:39 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA28333 for ; Wed, 2 Apr 1997 14:07:35 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id OAA17730; Wed, 2 Apr 1997 14:05:37 -0800 (PST) To: Terry Lambert cc: nate@mt.sri.com, proff@suburbia.net, hackers@FreeBSD.ORG Subject: Re: Internal clock In-reply-to: Your message of "Wed, 02 Apr 1997 10:24:17 MST." <199704021724.KAA13803@phaeton.artisoft.com> Date: Wed, 02 Apr 1997 14:05:37 -0800 Message-ID: <17726.860018737@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Frankly, I don't care about persistance other than the default > permission, which should be set correctly already so that most > everyone besides me doesn't care about persistance either. Coming from the guy who constantly exhorts us to remain *programmer interface* compatible, much less user compatible, with everything, I find this to be a rather blatant instance of holding firm principles only for as long as they suit the purposes of the arguments you're trying to advance, to be quickly discarded the moment they work against you. I expected better from you, Terry. :-) > As far as putting code into rc.local & rc.shutdown: why doesn't that > count as transparent? Because we can't upgrade the rc scripts out Because it's not. It's not even close. Jordan From owner-freebsd-hackers Wed Apr 2 14:21:30 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA29139 for hackers-outgoing; Wed, 2 Apr 1997 14:21:30 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id OAA29127 for ; Wed, 2 Apr 1997 14:21:24 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id AAA12726; Thu, 3 Apr 1997 00:21:02 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id AAA25382; Thu, 3 Apr 1997 00:08:56 +0200 (MET DST) Message-ID: <19970403000856.EU42233@uriah.heep.sax.de> Date: Thu, 3 Apr 1997 00:08:56 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freefall.freebsd.org Cc: thompson@squirrel.tgsoft.com (mark thompson) Subject: Re: Code maintenance References: <19970402205513.11072.qmail@squirrel.tgsoft.com> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <19970402205513.11072.qmail@squirrel.tgsoft.com>; from mark thompson on Apr 2, 1997 20:55:13 -0000 Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk As mark thompson wrote: > > Yes, see the ft(4) and aic(4) drivers for two good examples. Both > > sorely lack a maintainer, and nobody seems to be interested in doing > > this job. > > Hmmm.... i have volunteered before to maintain ft, since tapes are kind Hmm, i can't muchly remember your name, nor having seen any offer to maintain ft(4). If you're serious about this, go ahead! > of a (twisted) specialty of mine. My only request is that someone donate > the drives to test on, since i don't use QIC format any more. Chuck Robey once contributed his old Colorado tape to the project, which is currently sitting in my scratchbox. I happily send it to you, but please understand that i'd like to see some sort of, well not quite `guarantee', but at least promise, that you're serious about this. Remember, there have been three maintainers of this code before who all went off-ship since. This is an old Colorado QIC-40 drive. I've also got a supposedly broken QIC-80 one, but i'm totally uncertain about its status. > My offer to help kind of got blown off. By whom? Certainly not by me. Nor can i remember that you ever wrote a mail to the core team applying for this job, at least not during the time i've been a member of that team (which must go into its 2nd anniversary anytime soon). -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Wed Apr 2 14:23:47 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA29293 for hackers-outgoing; Wed, 2 Apr 1997 14:23:47 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA29286 for ; Wed, 2 Apr 1997 14:23:44 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id OAA17913; Wed, 2 Apr 1997 14:23:40 -0800 (PST) To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) cc: hackers@freebsd.org Subject: Re: Does de driver do 100MBIT Full Duplex? In-reply-to: Your message of "Wed, 02 Apr 1997 21:46:27 +0200." <19970402214627.DT15556@uriah.heep.sax.de> Date: Wed, 02 Apr 1997 14:23:39 -0800 Message-ID: <17909.860019819@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > soon. There are not many people volunteering for work, and these are > seldom the people who shout in a mailing list. :) ^^^^^^ surprisingly almost never :-) From owner-freebsd-hackers Wed Apr 2 14:34:54 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA29929 for hackers-outgoing; Wed, 2 Apr 1997 14:34:54 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA29923 for ; Wed, 2 Apr 1997 14:34:52 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id OAA17953; Wed, 2 Apr 1997 14:33:05 -0800 (PST) To: Terry Lambert cc: avalon@coombs.anu.edu.au, proff@suburbia.net, hackers@freebsd.org Subject: Re: devfs In-reply-to: Your message of "Wed, 02 Apr 1997 14:41:38 MST." <199704022141.OAA14577@phaeton.artisoft.com> Date: Wed, 02 Apr 1997 14:33:05 -0800 Message-ID: <17949.860020385@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Move the process environment into the kernel on the other side of a > system call, and you can have all the variant links your crooked > little heart desires. I think indirecting through sysctl() would be a lot easier. I didn't WANT to rearchitect the entire process environment handling mechanism, perturbing everything from the kernel to libc, I was looking for a lower-impact change. And besides, if I were going to redo environment handling, I wouldn't just shove the default environment handling into the kernel and yell "Done!", I'd implement a whole multi-tier namespace with process local, group local and systemwide logical names. Why bother adding more abstraction to that mechanism otherwise? All that work just to achieve an equivalent status quo (with variant symlinks as the only added benefit) would be just silly. Jordan From owner-freebsd-hackers Wed Apr 2 14:38:00 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA00218 for hackers-outgoing; Wed, 2 Apr 1997 14:38:00 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id OAA00207 for ; Wed, 2 Apr 1997 14:37:57 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id PAA14691; Wed, 2 Apr 1997 15:19:33 -0700 From: Terry Lambert Message-Id: <199704022219.PAA14691@phaeton.artisoft.com> Subject: Re: Internal clock To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Wed, 2 Apr 1997 15:19:33 -0700 (MST) Cc: terry@lambert.org, nate@mt.sri.com, proff@suburbia.net, hackers@FreeBSD.ORG In-Reply-To: <17726.860018737@time.cdrom.com> from "Jordan K. Hubbard" at Apr 2, 97 02:05:37 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > Frankly, I don't care about persistance other than the default > > permission, which should be set correctly already so that most > > everyone besides me doesn't care about persistance either. > > Coming from the guy who constantly exhorts us to remain *programmer > interface* compatible, much less user compatible, with everything, I > find this to be a rather blatant instance of holding firm principles > only for as long as they suit the purposes of the arguments you're > trying to advance, to be quickly discarded the moment they work > against you. I expected better from you, Terry. :-) What? I don't exhort you to maintain compatability with everything -- just commercial binaries that can't be easily recompiled, and (less so) statically linked binaries linked against comeercial libraries which can not be redistributed (ie: Motif). Most of that falls into the "ABI compatability" arena. I also expect you to comply with standards, though that hardly locks you into compatability with anything, as Linux passing POSIX ceritifcation proves. I'd like to see you maintain ABI compatability, of course, but if you'll remember, I was one of the first people on the "unmap page 0" bandwagon (I even whined load enough that it became an option in SVR4; page 0 is mapped on fault unless a configuration option is specified), so a good reason is good enough. Creating "well defined kernel interfaces" is a far cry from mandating ABI interfaces: kernel interfaces are not exported. Any place they are ("ps", "w", "ifconfig", "route", "netstat", etc.) is an error. There is a big difference between "intra-kernel" calls by other kernel code and "inter-kernel" calls by user code. > > As far as putting code into rc.local & rc.shutdown: why doesn't that > > count as transparent? Because we can't upgrade the rc scripts out > > Because it's not. It's not even close. It's perfectly transparent to people who don't goo around in their rc files. It's not my fault they aren't data driven an functionally seperated enough that that's a larger number than I'd like it to be... If you'll remember, I don't want people mucking with the rc files enough that I advocate mounting root read-only. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Wed Apr 2 14:41:51 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA00495 for hackers-outgoing; Wed, 2 Apr 1997 14:41:51 -0800 (PST) Received: from awfulhak.demon.co.uk (awfulhak.demon.co.uk [158.152.17.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA00475 for ; Wed, 2 Apr 1997 14:41:37 -0800 (PST) Received: from awfulhak.demon.co.uk (localhost.lan.awfulhak.org [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id WAA16943; Wed, 2 Apr 1997 22:46:17 +0100 (BST) Message-Id: <199704022146.WAA16943@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: Brian Somers cc: mika ruohotie , macgyver@db-net.com (Wilson MacGyver), hackers@freebsd.org Subject: Re: p166 vs. p166mmx In-reply-to: Your message of "Wed, 02 Apr 1997 22:09:44 BST." <199704022109.WAA14871@awfulhak.demon.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Wed, 02 Apr 1997 22:46:17 +0100 From: Brian Somers Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Hmm, I don't know. I heard reports that a PP150 w/ 512k cache was > almost as good as a PP200 2/ 256k. ^^ oops, that's "w/" ! > -- > Brian , > > Don't _EVER_ lose your sense of humour.... From owner-freebsd-hackers Wed Apr 2 15:08:50 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA02508 for hackers-outgoing; Wed, 2 Apr 1997 15:08:50 -0800 (PST) Received: from tor-adm1.nbc.netcom.ca (taob@tor-adm1.nbc.netcom.ca [207.181.89.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA02496 for ; Wed, 2 Apr 1997 15:08:45 -0800 (PST) Received: from localhost (taob@localhost) by tor-adm1.nbc.netcom.ca (8.8.5/8.8.5) with SMTP id SAA07257; Wed, 2 Apr 1997 18:07:50 -0500 (EST) Date: Wed, 2 Apr 1997 18:07:50 -0500 (EST) From: Brian Tao To: Joerg Wunsch cc: freebsd-hackers@freebsd.org, keith waters Subject: Re: Unreasonable rebuild time on large password files In-Reply-To: <19970402212443.OT00003@uriah.heep.sax.de> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Wed, 2 Apr 1997, J Wunsch wrote: > > No need. It's already there. Guido once implemented incremental > updates to the password database if only a single user's entry has > been changed. They seem to be non-optional (much to my surprise :), > and the default in FreeBSD 2.2 and above. Ah, geez, that's right... I even remember pestering him about it a long time ago. :) So the patches have been integrated? I didn't see any mention of them in the man pages, so I presumed it was still necessary to patch it separately. How do you enable the feature? -- Brian Tao (BT300, taob@netcom.ca) "Though this be madness, yet there is method in't" From owner-freebsd-hackers Wed Apr 2 15:13:26 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA03024 for hackers-outgoing; Wed, 2 Apr 1997 15:13:26 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id PAA02983 for ; Wed, 2 Apr 1997 15:12:57 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id PAA14823; Wed, 2 Apr 1997 15:54:06 -0700 From: Terry Lambert Message-Id: <199704022254.PAA14823@phaeton.artisoft.com> Subject: Re: devfs To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Wed, 2 Apr 1997 15:54:05 -0700 (MST) Cc: terry@lambert.org, avalon@coombs.anu.edu.au, proff@suburbia.net, hackers@freebsd.org In-Reply-To: <17949.860020385@time.cdrom.com> from "Jordan K. Hubbard" at Apr 2, 97 02:33:05 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > And besides, if I were going to redo environment handling, I wouldn't > just shove the default environment handling into the kernel and yell > "Done!", I'd implement a whole multi-tier namespace with process > local, group local and systemwide logical names. Why bother adding > more abstraction to that mechanism otherwise? All that work just to > achieve an equivalent status quo (with variant symlinks as the only > added benefit) would be just silly. What you are asking for doesn't require "terrible forethought and consideration"... it's FREE FREE FREE with the most obvious and simplistic implementation! The group and system LNM's are trivial; the automatically fall out of the design: Process LNM's: The environment space hung off the calling process Group LNM's: The environment space hung off the process group leader for the calling process System LNM's: The environment hung off the "init" process. For backward compatability, all you have to do is add a wrapper that specifies "ANY" on getenv and "PROCESS" on setenv/unsetenv. Here's the replacement getenv.c and syslnm.h header... like I said, *TRIVIAL*. It's pretty darn obvious that you can implement the syslnm() call by migrating most of the existing libc environment code down. If you want to pass down a pid_t instead of a lnm_type_t, and call getpgrp() explicitly for "LNM_TYPE_GROUP" and pass an explicit 1 for "LNM_TYPE_SYSTEM", you can solve the "get/modify the parent/other environment" problem at the same time. I would recommend doing that, but didn't want to clutter up the initial rev. of the syslnm.h header with inline functions to do the conversion between constants. You may need to add an "LNM_CMD_GET_ENVP" to get a command to user space for use with the execve(2) envp argument, assuming you don't move that to execve(3) and keep POSIX compatability that way. *ANY* brain-damaged code hack should be able to do this in a day, including all necessary testing... hell, it didn't take me that long to write everything, including the library replacements, so I'm giving you a head start. PS: if you need the changes to namei() for variant symbolic links, ask me nicely, and I will disentangle them from my other changes to namei() for layering fixes, Unicode, and alternate namespace support (used by a modified (CIFS enhanced) Samba server which needs to have the DOS short name remain constant across directory searches). Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. --- BEGIN syslnm.h ----------------------------------------------------- /* * Copyright (c) 1996 * Terrence Lambert. All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * 3. All advertising materials mentioning features or use of this software * must display the following acknowledgement: * This product includes software developed by Terrence Lambert. * 4. Neither the name of the author nor the names of subsequent contributors * may be used to endorse or promote products derived from this software * without specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. */ #ifndef _SYSLNM_H_ #define _SYSLNM_H_ /* * Length of static buffer for getenv(3) data returns */ #define LNM_MAX_GET 1024 /* 1k max for using getenv()*/ /* * Command codes */ typedef enum { LNM_CMD_GET, /* get a logical name*/ LNM_CMD_SET, /* set a logical name*/ LNM_CMD_UNSET /* unset a logical name*/ } lnm_cmd_t; /* * Flags */ typedef enum { LNM_FLAG_NONE = 0, /* no flags*/ LNM_FLAG_INHERIT, /* in inheritance order*/ LNM_FLAG_REPLACE, /* replace if existing*/ } lnm_flag_t; /* * Tables (bitmap) */ typedef enum { LNM_TYPE_PROCESS = 0x01, /* process logical name table*/ LNM_TYPE_GROUP = 0x02, /* group logical name table*/ LNM_TYPE_SYSTEM = 0x04, /* system logical name table*/ LNM_TYPE_ANY = 0x07 /* any logical name table*/ } lnm_type_t; /* * Structures for commands */ struct lnm_get_arg { char *name; /* name to get*/ char *buf; /* return buffer*/ int bufsize; /* return buffer size*/ lnm_flags_t flags; /* option flags*/ lnm_type_t table; /* table(s) to involve*/ }; struct lnm_set_arg { char *name; /* name to set*/ char *buf; /* value buffer*/ lnm_flags_t flags; /* option flags*/ lnm_type_t table; /* table(s) to involve*/ }; struct lnm_del_arg { char *name; /* name to set*/ lnm_flags_t flags; /* option flags*/ lnm_type_t table; /* table(s) to involve*/ }; #endif /* _SYSLNM_H_*/ /* * EOF */ --- END syslnm.h ------------------------------------------------------- --- BEGIN getenv.c ----------------------------------------------------- /* * Copyright (c) 1996, Terrence Lambert * All rights reserved. * * Redistribution and use in source and binary forms, with or without * modification, are permitted provided that the following conditions * are met: * 1. Redistributions of source code must retain the above copyright * notice, this list of conditions and the following disclaimer. * 2. Redistributions in binary form must reproduce the above copyright * notice, this list of conditions and the following disclaimer in the * documentation and/or other materials provided with the distribution. * 3. All advertising materials mentioning features or use of this software * must display the following acknowledgement: * This product includes software developed by Terrence Lambert. * 4. The name of the author may not be used to endorse or promote products * derived from this software without specific prior written permission. * * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE * ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF * SUCH DAMAGE. * * $FreeBSD$ */ #if defined(LIBC_SCCS) && !defined(lint) static char sccsid[] = "@(#)getenv.c 1.0 (Lambert) 22 Jun 1996"; #endif /* LIBC_SCCS and not lint */ #include #include #include #include /* * getenv -- * Returns ptr to value associated with name, if any, else NULL. */ char * getenv(name) const char *name; { static char ret[ LNM_GET_MAX]; struct lnm_get_arg get; get.name = name; /* name to get*/ get.buf = ret; /* store it here*/ get.bufsize = LNM_MAX_GET; /* max of this many*/ get.flags = LNM_FLAG_INHERIT; /* inherit values*/ get.table = LNM_TYPE_ANY; /* name table to use*/ if( syslnm( LNM_CMD_GET, (caddr_t)&get)) return( NULL); else return( ret); } int putenv(str) const char *str; { struct lnm_set_arg set; char *p, *equal; int rval; if ((p = strdup(str)) == NULL) return (-1); if ((equal = index(p, '=')) == NULL) { (void)free(p); return (-1); } *equal = '\0'; set.name = name; /* name to set*/ set.buf = equal + 1; /* value to assign*/ set.flags = LNM_FLAG_REPLACE; /* replae if exists*/ set.table = LNM_TYPE_PROCESS; /* process name table*/ rval = syslnm( LNM_CMD_SET, (caddr_t)&set); (void)free(p); return (rval); } /* * setenv -- * Set the value of the environmental variable "name" to be * "value". If rewrite is set, replace any current value. */ setenv(name, value, rewrite) register const char *name; register const char *value; int rewrite; { struct lnm_set_arg set; set.name = name; /* name to set*/ set.buf = value; /* value to assign*/ set.flags = ( rewrite ? LNM_FLAG_REPLACE : 0); /* conditionally replace*/ set.table = LNM_TYPE_PROCESS; /* process name table*/ return( syslnm( LNM_CMD_SET, (caddr_t)&set)); } /* * unsetenv(name) -- * Delete environmental variable "name". */ void unsetenv(name) const char *name; { struct lnm_del_arg del; del.name = name; /* name to delete*/ del.flags = 0; /* no flags*/ del.table = LNM_TYPE_PROCESS; /* delete from process*/ (void)syslnm( LNM_CMD_UNSET, (caddr_t)&del); } /* * EOF */ --- END getenv.c ------------------------------------------------------- From owner-freebsd-hackers Wed Apr 2 15:21:37 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA03922 for hackers-outgoing; Wed, 2 Apr 1997 15:21:37 -0800 (PST) Received: from life.eecs.umich.edu (pmchen@life.eecs.umich.edu [141.213.8.32]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA03913 for ; Wed, 2 Apr 1997 15:21:33 -0800 (PST) Received: (from pmchen@localhost) by life.eecs.umich.edu (8.8.5/8.8.0) id SAA19709 for freebsd-hackers@freebsd.org; Wed, 2 Apr 1997 18:20:42 -0500 (EST) Date: Wed, 2 Apr 1997 18:20:42 -0500 (EST) From: "Peter M. Chen" Message-Id: <199704022320.SAA19709@life.eecs.umich.edu> To: freebsd-hackers@freebsd.org Subject: VOP_PUTPAGES Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I feel stupid for asking this, but what does VOP_PUTPAGES call? I tried tracing through the maze of macros but ended up at vp->v_op[0], which I think is an error function. VOP_PUTPAGES (kern/vnode_if.h) calls VCALL(vp, VOFFSET(vop_putpages), &a) which becomes VOCALL(vp->v_op, VOFFSET(vop_putpages),&a) which becomes ( *(vp->v_op[VOFFSET(vop_putpages)])) (&a) which becomes ( *(vp->v_op[((&vop_putpages_desc)->vdesc_offset) ])) (&a) which becomes ( *(vp->v_op[0])) (&a) I'm interested in the paging out behavior of files. E.g. a file block that has been mmap'ed and dirtied needs to be written back to its file. Thanks, Pete From owner-freebsd-hackers Wed Apr 2 15:32:34 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA05202 for hackers-outgoing; Wed, 2 Apr 1997 15:32:34 -0800 (PST) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA05188 for ; Wed, 2 Apr 1997 15:32:30 -0800 (PST) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id QAA17367; Wed, 2 Apr 1997 16:32:24 -0700 (MST) Date: Wed, 2 Apr 1997 16:32:24 -0700 (MST) Message-Id: <199704022332.QAA17367@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: Brian Somers Cc: mika ruohotie , macgyver@db-net.com (Wilson MacGyver), hackers@freebsd.org Subject: Re: p166 vs. p166mmx In-Reply-To: <199704022109.WAA14871@awfulhak.demon.co.uk> References: <199704021219.PAA03426@shadows.aeon.net> <199704022109.WAA14871@awfulhak.demon.co.uk> X-Mailer: VM 6.22 under 19.15 XEmacs Lucid Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Hmm, I don't know. I heard reports that a PP150 w/ 512k cache was > almost as good as a PP200 2/ 256k. I'd have a *really* hard time believing it, especially given Intel's numbers on their WWW site. I've been going with the PP200-256K chip vs. the 512K chip due to the $800 difference in chip costs and the 4-5% difference in performance seen by the benchmarks on Intels site. Nate From owner-freebsd-hackers Wed Apr 2 15:57:19 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA07537 for hackers-outgoing; Wed, 2 Apr 1997 15:57:19 -0800 (PST) Received: from mail.vlsi.fi (w6DXJo9SDvOLhNNqlmb5XFmSEtzOpb6H@mail.vlsi.fi [195.74.10.147]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA07530 for ; Wed, 2 Apr 1997 15:57:13 -0800 (PST) Received: (from smap@localhost) by mail.vlsi.fi (8.7.6/8.7.3) id CAA12332; Thu, 3 Apr 1997 02:56:48 +0300 (EET DST) Received: from vlsi1.vlsi.fi(193.64.2.2) by mail.vlsi.fi via smap (V1.3) id sma012328; Thu Apr 3 02:56:29 1997 Received: from layout.vlsi.fi by vlsi1.vlsi.fi with ESMTP (1.37.109.16/16.2) id AA190985387; Thu, 3 Apr 1997 02:56:28 +0300 Received: by layout.vlsi.fi (1.37.109.15/16.2) id AA145495387; Thu, 3 Apr 1997 02:56:27 +0300 Date: Thu, 3 Apr 1997 02:56:27 +0300 Message-Id: <199704022356.AA145495387@layout.vlsi.fi> From: Ville Eerola To: Nate Williams Cc: hackers@FreeBSD.ORG Subject: Re: 21" Monitor recommendations? In-Reply-To: <199704021828.LAA14450@rocky.mt.sri.com> References: <199704021828.LAA14450@rocky.mt.sri.com> X-Mailer: VM Version 5.93 (beta) under GNU Emacs 19.29.6 Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Nate Williams writes: > Does anyone have any recommendations for a *real* 21" monitor that will > handle 1600x1200@72Hz, with a decent dot-pitch? My current monitor will > handle 1600@1200@76Hz, and has a .25 dot-pitch, and I'd like something > that is comparable. I'd like to spend under $2K/monitor if possible. We use Hitachi CM-21xx monitors at work. The current model number seems to be CM-2112M. They are quite good performers, the picture quality is good, and the prices at least here in Finland are reasonable. They are definitely bigger in size than some of the 20" monitors. The dot pitch is 0.26mm. The max resolution rating for given in the WWW-site of a local supplier seems to be 1800x1440@73Hz, and the recommended one is 1600x1200@88Hz! Regards, Ville -- Ville.Eerola@vlsi.fi VLSI Solution Oy Tel:+358 3 3165579 Hermiankatu 6-8 C Fax:+358 3 3165220 FIN-33720 Tampere, Finland From owner-freebsd-hackers Wed Apr 2 16:09:29 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA08219 for hackers-outgoing; Wed, 2 Apr 1997 16:09:29 -0800 (PST) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA08210 for ; Wed, 2 Apr 1997 16:09:26 -0800 (PST) Received: (from root@localhost) by dyson.iquest.net (8.8.4/8.6.9) id TAA03796; Wed, 2 Apr 1997 19:09:17 -0500 (EST) From: "John S. Dyson" Message-Id: <199704030009.TAA03796@dyson.iquest.net> Subject: Re: VOP_PUTPAGES In-Reply-To: <199704022320.SAA19709@life.eecs.umich.edu> from "Peter M. Chen" at "Apr 2, 97 06:20:42 pm" To: pmchen@eecs.umich.edu (Peter M. Chen) Date: Wed, 2 Apr 1997 19:09:17 -0500 (EST) Cc: freebsd-hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > I feel stupid for asking this, but what does VOP_PUTPAGES call? I tried > tracing through the maze of macros but ended up at vp->v_op[0], which I > think is an error function. > You are right!!! :-). If there is no special putpages routine for a filesystem, a default one is used in the pager. This allows for either layering for special filesystems, for higher performance pageins/pageouts, or for the default usage of VOP_WRITE... The default routine is near the end of vnode_pager.c. John From owner-freebsd-hackers Wed Apr 2 16:23:52 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA09496 for hackers-outgoing; Wed, 2 Apr 1997 16:23:52 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id QAA09491 for ; Wed, 2 Apr 1997 16:23:48 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id RAA15015; Wed, 2 Apr 1997 17:06:49 -0700 From: Terry Lambert Message-Id: <199704030006.RAA15015@phaeton.artisoft.com> Subject: Re: VOP_PUTPAGES To: pmchen@eecs.umich.edu (Peter M. Chen) Date: Wed, 2 Apr 1997 17:06:48 -0700 (MST) Cc: freebsd-hackers@freebsd.org In-Reply-To: <199704022320.SAA19709@life.eecs.umich.edu> from "Peter M. Chen" at Apr 2, 97 06:20:42 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > I feel stupid for asking this, but what does VOP_PUTPAGES call? I tried > tracing through the maze of macros but ended up at vp->v_op[0], which I > think is an error function. > > VOP_PUTPAGES (kern/vnode_if.h) calls > VCALL(vp, VOFFSET(vop_putpages), &a) yes. > which becomes > VOCALL(vp->v_op, VOFFSET(vop_putpages),&a) yes. > which becomes > ( *(vp->v_op[VOFFSET(vop_putpages)])) (&a) yes. And VOFFSET(vop_putpages) is at offset 43 in vfs_op_descs[]. > which becomes > ( *(vp->v_op[((&vop_putpages_desc)->vdesc_offset) ])) (&a) yes. > which becomes > ( *(vp->v_op[0])) (&a) BZZZZZZZZZZTTTT. For FS's where the op is supported, it's the offset of the per FS specific *_putpage. For most FS's, it's not supported, but getpages is -- see ffs_getpages() routine in the source file /sys/ufs/ufs/ufs_readwrite.c For those where it is not, a "failed" call will be called. I don't believe the code is ever really called anywhere -- instead, you should grep for "putpages" in /sys/vm/*.c, and look around there (mostly "default_pager.c", device_pager.c", "swap_pager.c", and "vn_pager.c"). These are called explicity instead of the VOP's. One of the most useful places to look would be in the mmap() code, since that is where the VOP_PUTPAGES() would be most easily implemented (the most interesting code in that case is the vn_pager.c). In general, the getpages() is only used for paging in pages from a file being used as a swap backing store for an executable -- in other words, *clean* pages. The putpages() routine is for writing dirty pages out... and is not currently used. For more details, you should see the posting in -current about John Dyson's modifications to add the op's into the list of VOPS supported as "standard", about 6 moths to a year ago. There is every intent to actually use the putpages VOP -- eventually. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Wed Apr 2 17:21:15 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA14428 for hackers-outgoing; Wed, 2 Apr 1997 17:21:15 -0800 (PST) Received: from parkplace.cet.co.jp (parkplace.cet.co.jp [202.32.64.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA14412 for ; Wed, 2 Apr 1997 17:21:10 -0800 (PST) Received: from localhost (michaelh@localhost) by parkplace.cet.co.jp (8.8.5/CET-v2.1) with SMTP id BAA01306 for ; Thu, 3 Apr 1997 01:20:59 GMT Date: Thu, 3 Apr 1997 10:20:58 +0900 (JST) From: Michael Hancock To: FreeBSD Hackers Subject: factoring of vput/vrele not good Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk I don't think the factoring of some of the common code in vput and vrele into vputrele was a very good idea. The problem is that the locking semantics are very different which makes it tricky. Tricky code is *evil*. Factoring is a good habit in many cases, but sometimes we need to look at other aspects of code quality than processor cache efficiencies or whatever benefit that was used to make this decision. Regards, Mike Hancock From owner-freebsd-hackers Wed Apr 2 17:35:13 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA15258 for hackers-outgoing; Wed, 2 Apr 1997 17:35:13 -0800 (PST) Received: from colin.Muc.DE (root@colin.muc.de [193.174.4.1]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id RAA15253 for ; Wed, 2 Apr 1997 17:35:09 -0800 (PST) Received: from tor by colin.muc.de with UUCP id <86045-1>; Thu, 3 Apr 1997 03:34:49 +0200 Received: from odin.muc.de (odin.muc.de [193.31.20.18]) by tor.muc.de (8.8.5/8.6.6) with SMTP id QAA03730 for ; Wed, 2 Apr 1997 16:16:50 +0200 (CEST) Received: (from eilts@localhost) by odin.muc.de (8.6.8/8.6.6) id QAA00255 for freebsd-hackers@freebsd.org; Wed, 2 Apr 1997 16:16:46 +0200 Date: Wed, 2 Apr 1997 16:16:46 +0200 From: Hinrich Eilts Message-Id: <199704021416.QAA00255@odin.muc.de> Received: by NeXT.Mailer (1.100) Received: by NeXT Mailer (1.100) To: freebsd-hackers@freebsd.org Subject: Server-site patch for nn-current Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hello, if I tried to use nn (6.5.0) from ports on the same 2.2-FreeBSD PC as the NNTP server (INN-1.4sec) is located, it doesn't work (-> bug report in freebsd-bugs). The nn-current from ports will not use NNTP if running on the server what is unwanted in some cases (e.g. statistics). A small patch which is to be applied after the ports-patches "patch-a[abc]' add the ability to switch this behavior off, i.e. NNTP will be used regardless from nn running on NNTP server or remote. The nn-current will run on FreeBSD-2.2. Hinrich *** config.h.orig Wed Apr 2 15:43:55 1997 --- config.h Wed Apr 2 15:47:40 1997 *************** *** 56,61 **** --- 56,68 ---- #define NNTP_SERVER "/usr/local/lib/news/nntp_server" /* + * Define USE_NNTP_ON_SERVER if NNTP shall be used if nn is running on the + * nntp server itself. Else it will use direct access to files. + */ + + #define USE_NNTP_ON_SERVER + + /* * * Defining NO_NNTP_SERVER_READ causes NNTP to ignore the NNTP_SERVER * file for reading news but still allows the -R command to specify *** nntp.c.orig Wed Apr 2 15:43:40 1997 --- nntp.c Wed Apr 2 15:40:01 1997 *************** *** 175,180 **** --- 175,181 ---- * the rope. If he wants to hang himself, then let him! :-) * Let the user worry about keeping his .newsrc straight. */ + fprintf(stderr,"NNTP Server\n"); if ((cp = getenv("NNTPSERVER")) != NULL) { strncpy(nntp_server, cp, sizeof nntp_server); return 1; *************** *** 706,712 **** || find_server() #endif ) { ! nn_gethostname(host, sizeof host); tmp = gethostbyname(host); if (tmp) { strncpy(host, tmp->h_name, sizeof host); --- 707,716 ---- || find_server() #endif ) { ! #ifdef USE_NNTP_ON_SERVER ! use_nntp = 1; ! #else ! nn_gethostname(host, sizeof host); tmp = gethostbyname(host); if (tmp) { strncpy(host, tmp->h_name, sizeof host); *************** *** 718,724 **** } else { server_real_name = nntp_server; } ! use_nntp = (strcmp(host, server_real_name) != 0); } else { use_nntp = 0; } --- 722,729 ---- } else { server_real_name = nntp_server; } ! use_nntp = (strcmp(host, server_real_name) != 0); ! #endif } else { use_nntp = 0; } From owner-freebsd-hackers Wed Apr 2 17:46:07 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA15604 for hackers-outgoing; Wed, 2 Apr 1997 17:46:07 -0800 (PST) Received: from awfulhak.demon.co.uk (awfulhak.demon.co.uk [158.152.17.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA15595 for ; Wed, 2 Apr 1997 17:46:00 -0800 (PST) Received: from awfulhak.demon.co.uk (localhost.lan.awfulhak.org [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id AAA21160 for ; Thu, 3 Apr 1997 00:48:43 +0100 (BST) Message-Id: <199704022348.AAA21160@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: freebsd-hackers@freebsd.org Subject: Bustek FlashPointLT Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 03 Apr 1997 00:48:43 +0100 From: Brian Somers Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Is anyone developing a driver for the above ? It's a SCSI III PCI controler for half the price of a 3940. -- Brian , Don't _EVER_ lose your sense of humour.... From owner-freebsd-hackers Wed Apr 2 17:50:41 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA16036 for hackers-outgoing; Wed, 2 Apr 1997 17:50:41 -0800 (PST) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA16030 for ; Wed, 2 Apr 1997 17:50:36 -0800 (PST) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.5/8.7.3) id LAA05710; Thu, 3 Apr 1997 11:20:10 +0930 (CST) From: Michael Smith Message-Id: <199704030150.LAA05710@genesis.atrad.adelaide.edu.au> Subject: Re: Maxpeed SS8 drivers... In-Reply-To: <333D9299.31DFF4F5@pascal.org> from "Freeman P. Pascal IV" at "Mar 29, 97 02:07:21 pm" To: pascal@pascal.org (Freeman P. Pascal IV) Date: Thu, 3 Apr 1997 11:20:10 +0930 (CST) Cc: freebsd-hackers@freebsd.org, pascal@calvary.pascal.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Freeman P. Pascal IV stands accused of saying: > > I was curious if anyone has written Maxpeed SS8 drivers for 2.1.x or > 2.2? I've recently come into possession of four boards and being > able to run them with FreeBSD would be a boon to me. I don't know of anyone that has. If you are interested in having a driver written/modified, I am available for that sort of work and my rates are quite reasonable. > As a side questions. Just how hard is it to translate Linux drivers > for serial equipment for use with FreeBSD. I do have the latest > Linux drivers from ftp://ftp.maxpeed.com. Generally, not easy at all. The best way to go is normally to use the Linux driver as a reference for "how to make the hardware dance", and rework one of the BSD serial drivers to make it match. > -Freeman -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ From owner-freebsd-hackers Wed Apr 2 18:09:55 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id SAA17020 for hackers-outgoing; Wed, 2 Apr 1997 18:09:55 -0800 (PST) Received: (from mpp@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id SAA17013 for hackers; Wed, 2 Apr 1997 18:09:53 -0800 (PST) From: Mike Pritchard Message-Id: <199704030209.SAA17013@freefall.freebsd.org> Subject: Re: Internal clock To: hackers Date: Wed, 2 Apr 1997 18:09:53 -0800 (PST) In-Reply-To: <19970402210817.YJ08249@uriah.heep.sax.de> from "J Wunsch" at Apr 2, 97 09:08:17 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk J Wunsch wrote: > > Once i've sorted these problems out, i'll continue to make DEVFS > usable as a default. If persistancy is once the last remaining > problem, and nobody is stepping forward to implement the all-singing- > all-dancing union layer shadowing solution, i'll probably implement a > startup/shutdown script-based solution. If something better comes up > later, it can still be replaced. > > I think FreeBSD 3.0 might be shipping with DEVFS as the default, but > it's still ``quarter o'mile'' to go. I like the idea that someone posted about keeping some type of data file in the mount-point directory. This could be implemented similar to how the quota data files are handled. Manage it as a sparse file so that unused information isn't taking up disk space. On attribute change of a devfs device, the file system code seeks to something like major*minor*structsize in the devfs file and writes the information out. A zero entry means use the defaults. At boot time it reads the information structure for each major/minor that is configured into the system. The overhead of this should be even less than that of the quota system, since only attribute changes need to be tracked, not allocations. With the proper tools, a system like this should also allow you to set permissions for devices that are not currently configured into the system. You get the idea. The quota system also allows you to specify an alternate location/file name for the quota file, so devfs could do something similar. -- Mike Pritchard mpp@FreeBSD.org "Go that way. Really fast. If something gets in your way, turn" From owner-freebsd-hackers Wed Apr 2 18:31:28 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id SAA20175 for hackers-outgoing; Wed, 2 Apr 1997 18:31:28 -0800 (PST) Received: from werple.net.au (melb.werple.net.au [203.9.190.18]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id SAA20169 for ; Wed, 2 Apr 1997 18:31:23 -0800 (PST) Received: (qmail 19439 invoked by uid 5); 3 Apr 1997 02:31:20 -0000 Received: (from jb@localhost) by freebsd1.cimlogic.com.au (8.7.5/8.7.3) id LAA14864; Thu, 3 Apr 1997 11:33:05 +1000 (EST) From: John Birrell Message-Id: <199704030133.LAA14864@freebsd1.cimlogic.com.au> Subject: Re: Bugs in libc_r (fwd) To: nw1@cs.wustl.edu (Nanbor Wang) Date: Thu, 3 Apr 1997 11:33:04 +1000 (EST) Cc: jb@cimlogic.com.au, Hackers@freebsd.org In-Reply-To: <199704021608.KAA17928@siesta.cs.wustl.edu> from Nanbor Wang at "Apr 2, 97 10:08:54 am" X-Mailer: ELM [version 2.4ME+ PL22 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Nanbor Wang wrote: > > > Anyway we can get an update of this fix? (I don't want to get into > > > -current at this moment ;-) > > > > I've checked the libc_r source that was tagged for 2.2 and there have > > been a lot of updates to uthread files since then. Enough to prevent > > patching the 2.2 tree (despite what I said earlier 8-). Sorry. We'll > > need to be in better shape for 3.0. > > Sorry, John. What I really meant was Lite2 merge. I am using > -current before lite2 merge. (before 2/9/97 ?) Can I patch the > pthread library? Thanks. Just CVSup the libc_r tree. I doubt that there are any Lite2 issues that would get in the way. Regards, -- John Birrell - jb@cimlogic.com.au; jb@netbsd.org; jb@freebsd.org CIMlogic Pty Ltd, 119 Cecil Street, South Melbourne Vic 3205, Australia Tel +61 3 9690 6900 Fax +61 3 9690 6650 Mob +61 418 353 137 From owner-freebsd-hackers Wed Apr 2 18:38:00 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id SAA20464 for hackers-outgoing; Wed, 2 Apr 1997 18:38:00 -0800 (PST) Received: from calvary.pascal.org ([207.21.96.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA20459 for ; Wed, 2 Apr 1997 18:37:55 -0800 (PST) Received: from calvary.pascal.org (localhost [127.0.0.1]) by calvary.pascal.org (8.8.5/8.8.5) with SMTP id SAA09971; Wed, 2 Apr 1997 18:37:23 -0800 (PST) Message-ID: <334317E2.F3615C1@pascal.org> Date: Wed, 02 Apr 1997 18:37:22 -0800 From: "Freeman P. Pascal IV" Organization: The Pascal Family X-Mailer: Mozilla 3.01 (X11; U; FreeBSD 2.1.7-RELEASE i386) MIME-Version: 1.0 To: Michael Smith CC: freebsd-hackers@freebsd.org, pascal@calvary.pascal.org Subject: Re: Maxpeed SS8 drivers... References: <199704030150.LAA05710@genesis.atrad.adelaide.edu.au> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I was asking to avoid duplication of effort. I'm more than interested in writing and maintaining the driver myself. I've written drivers before for NetBSD, but not FreeBSD (although there shouldn't be any real differences between the two). Thanks for the offer anyway. -Freeman Michael Smith wrote: > > Freeman P. Pascal IV stands accused of saying: > > > > I was curious if anyone has written Maxpeed SS8 drivers for 2.1.x or > > 2.2? I've recently come into possession of four boards and being > > able to run them with FreeBSD would be a boon to me. > > I don't know of anyone that has. If you are interested in having > a driver written/modified, I am available for that sort of work > and my rates are quite reasonable. > > > As a side questions. Just how hard is it to translate Linux drivers > > for serial equipment for use with FreeBSD. I do have the latest > > Linux drivers from ftp://ftp.maxpeed.com. > > Generally, not easy at all. The best way to go is normally to use the > Linux driver as a reference for "how to make the hardware dance", and > rework one of the BSD serial drivers to make it match. > > > -Freeman > > -- > ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ > ]] Genesis Software genesis@gsoft.com.au [[ > ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ > ]] realtime instrument control. (ph) +61-8-8267-3493 [[ > ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ -- --------------------------------------------------------------------------- C O M P U T E I N T E N S I V E , I N C . --------------------------------------------------------------------------- Freeman P. Pascal IV Phone Work: (800) 273-5600 Compute Intensive, Inc. Home: (510) 232-0914 8001 Irvine Center Drive FAX: (510) 689-5405 Suite 1130 Email Work: pascal@compute.com Irvine, CA 92618 Home: pascal@pascal.org URL: Work: http://www.compute.com Home: http://www.pascal.org --------------------------------------------------------------------------- For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life. - John 3:16 (KJV) From owner-freebsd-hackers Wed Apr 2 19:10:45 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id TAA21901 for hackers-outgoing; Wed, 2 Apr 1997 19:10:45 -0800 (PST) Received: from squirrel.tgsoft.com (squirrel.tgsoft.com [207.167.64.183]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id TAA21888 for ; Wed, 2 Apr 1997 19:10:40 -0800 (PST) Received: (qmail 11557 invoked by uid 128); 3 Apr 1997 03:10:35 -0000 Date: 3 Apr 1997 03:10:35 -0000 Message-ID: <19970403031035.11556.qmail@squirrel.tgsoft.com> From: mark thompson To: joerg_wunsch@uriah.heep.sax.de CC: hackers@freefall.freebsd.org In-reply-to: <19970403000856.EU42233@uriah.heep.sax.de> (j@uriah.heep.sax.de) Subject: Re: Code maintenance Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Date: Thu, 3 Apr 1997 00:08:56 +0200 From: j@uriah.heep.sax.de (J Wunsch) As mark thompson wrote: > > Yes, see the ft(4) and aic(4) drivers for two good examples. Both > > sorely lack a maintainer, and nobody seems to be interested in doing > > this job. > > Hmmm.... i have volunteered before to maintain ft, since tapes are kind > of a (twisted) specialty of mine. My only request is that someone donate > the drives to test on, since i don't use QIC format any more. > My offer to help kind of got blown off. Hmm, i can't muchly remember your name, nor having seen any offer to maintain ft(4). If you're serious about this, go ahead! I apologize for the bitchy tone of my posting. I should have thought twice about what i was saying before posting. The other thing, i will take up privately. -mark From owner-freebsd-hackers Wed Apr 2 20:31:14 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA25007 for hackers-outgoing; Wed, 2 Apr 1997 20:31:14 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA25000; Wed, 2 Apr 1997 20:31:10 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id UAA19205; Wed, 2 Apr 1997 20:31:14 -0800 (PST) To: Mike Pritchard cc: hackers@freefall.freebsd.org Subject: Re: Internal clock In-reply-to: Your message of "Wed, 02 Apr 1997 18:09:53 PST." <199704030209.SAA17013@freefall.freebsd.org> Date: Wed, 02 Apr 1997 20:31:14 -0800 Message-ID: <19202.860041874@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > I like the idea that someone posted about keeping some type > of data file in the mount-point directory. This could be implemented I posted it but it was Sean Eric Fagan's idea. :-) > similar to how the quota data files are handled. Manage it as a > sparse file so that unused information isn't taking up disk space. > On attribute change of a devfs device, the file system code seeks to > something like major*minor*structsize in the devfs file and writes the > information out. A zero entry means use the defaults. I like this idea better than the "history log" idea since it never gets any larger than the max # of entries in /dev. Yes! Go Mike Go! :-) Jordan From owner-freebsd-hackers Wed Apr 2 22:51:52 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id WAA01231 for hackers-outgoing; Wed, 2 Apr 1997 22:51:52 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id WAA01223 for ; Wed, 2 Apr 1997 22:51:42 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id IAA17955 for freebsd-hackers@freebsd.org; Thu, 3 Apr 1997 08:50:58 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id IAA27833; Thu, 3 Apr 1997 08:21:03 +0200 (MET DST) Message-ID: <19970403082103.GT06676@uriah.heep.sax.de> Date: Thu, 3 Apr 1997 08:21:03 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: freebsd-hackers@freebsd.org Subject: Re: Unreasonable rebuild time on large password files References: <19970402212443.OT00003@uriah.heep.sax.de> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: ; from Brian Tao on Apr 2, 1997 18:07:50 -0500 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Brian Tao wrote: > Ah, geez, that's right... I even remember pestering him about it a > long time ago. :) So the patches have been integrated? I didn't see > any mention of them in the man pages, so I presumed it was still > necessary to patch it separately. How do you enable the feature? Guido doesn't seem to be very well in writing man pages. :-| The -u option to pwd_mkdb(8) ain't mentioned at all, although it's being used _by default_ if only a single user's entry is updated (passwd(1), or chpass(1), as opposed to vipw(8)). -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Wed Apr 2 23:09:58 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA02198 for hackers-outgoing; Wed, 2 Apr 1997 23:09:58 -0800 (PST) Received: from pcpsj.pfcs.com (TvjDhA3ClEX/rh83l3Jwce2YTwI3d7rd@harlan.fred.net [205.252.219.31]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id XAA02191 for ; Wed, 2 Apr 1997 23:09:53 -0800 (PST) Received: from mumps.pfcs.com (mumps.pfcs.com [192.52.69.11]) by pcpsj.pfcs.com (8.6.12/8.6.9) with SMTP id CAA17944 for ; Thu, 3 Apr 1997 02:09:43 -0500 Received: from localhost by mumps.pfcs.com with SMTP id AA29345 (5.67b/IDA-1.5 for ); Thu, 3 Apr 1997 02:09:41 -0500 To: hackers@freebsd.org Subject: Re: 21" Monitor recommendations? Date: Thu, 03 Apr 1997 02:09:39 -0500 Message-Id: <29343.860051379@mumps.pfcs.com> From: Harlan Stenn Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk After seeing the answers posted, I'm inclined to ask the opposite question. What 21" monitors have folks had trouble with? Thanks... H From owner-freebsd-hackers Wed Apr 2 23:21:12 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA03046 for hackers-outgoing; Wed, 2 Apr 1997 23:21:12 -0800 (PST) Received: from haldjas.folklore.ee (Haldjas.folklore.ee [193.40.6.121]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA03012 for ; Wed, 2 Apr 1997 23:20:56 -0800 (PST) Received: from localhost (narvi@localhost) by haldjas.folklore.ee (8.8.4/8.8.4) with SMTP id KAA25747; Thu, 3 Apr 1997 10:23:48 +0300 (EEST) Date: Thu, 3 Apr 1997 10:23:32 +0300 (EEST) From: Narvi To: Joerg Wunsch cc: hackers@freebsd.org Subject: Re: Does de driver do 100MBIT Full Duplex? In-Reply-To: <19970402214627.DT15556@uriah.heep.sax.de> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Wed, 2 Apr 1997, J Wunsch wrote: > As dennis wrote: > > > Is the driver that you send me directly for 2.1.7 (that worked > > beautifully) in 2.2.1R? If not, WHY NOT! > > No need to shout. We aren't deaf exactly. > > Since Matt released it _after_ 2.2 has been cut. I've got a very > explicit message from Matt when i've been asking him earlier (right in > time to get something into 2.2 still) about a new version, that he > considered the stuff that sneaked into NetBSD by that time too buggy > to see it officially in FreeBSD. That's why we've been integrating > the little hack still to make at least the DE21140A supported. We > originally deferred the inclusion of this patch in anticipation of the > new driver version. > > As you can see, it's not that we're sleeping or wouldn't care about > something that can be considered one of the major ethernet NIC drivers > in *BSD. > > Also, as i understand it, Matt's new driver won't fit without a little > work into 2.2 or higher, due to the new media selection method. This > work simply needs to be done, and it requires a volunteer. (I've got > one potential volunteer, but he wouldn't be able to do it anytime > soon. There are not many people volunteering for work, and these are > seldom the people who shout in a mailing list. :) > I am ready to volunteer and I will also have time (string sometime today, ending sometime on Sunday). I do have a 21140A board (actually AE) that does not work with the current driver. The problem I *do* have is that the ftp transfer rate to the closest place with NetBSD source tree is too low and the connection breaks before it ever reaches me. As mail still seems to reach me, could someone please tar+gzip the needed files together and forward them to me? Sander > -- > cheers, J"org > > joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE > Never trust an operating system you don't have sources for. ;-) > From owner-freebsd-hackers Thu Apr 3 00:35:14 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA06463 for hackers-outgoing; Thu, 3 Apr 1997 00:35:14 -0800 (PST) Received: from fgate.flevel.co.uk (root@fgate.flevel.co.uk [194.6.101.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA06457; Thu, 3 Apr 1997 00:35:10 -0800 (PST) Received: from localhost (dev@localhost) by fgate.flevel.co.uk (8.8.3/8.6.9) with SMTP id JAA29929; Thu, 3 Apr 1997 09:35:06 +0100 (BST) Date: Thu, 3 Apr 1997 09:35:06 +0100 (BST) From: Developer To: Brian Somers cc: freebsd-current@freebsd.org, freebsd-hackers@freebsd.org Subject: Re: PPP Problems -- The next chapter *grin* In-Reply-To: <199704022107.WAA14851@awfulhak.demon.co.uk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Wed, 2 Apr 1997, Brian Somers wrote: > The answer would be shorter if you asked what isn't different. Were you > using the pmdemand example and reading the handbook tutorial ? Was it > >= 2.2 ? No, we are not using pmdemand. No I wasn't reading the tutorial in the handbook either *grin* We are using a 3.0-current snap. Regards, Trefor S. > Brian , > > Don't _EVER_ lose your sense of humour.... > > > From owner-freebsd-hackers Thu Apr 3 00:40:29 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA06732 for hackers-outgoing; Thu, 3 Apr 1997 00:40:29 -0800 (PST) Received: from trinity.radio-do.de (fn@trinity.Radio-do.de [193.101.164.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA06727 for ; Thu, 3 Apr 1997 00:40:25 -0800 (PST) From: Frank Nobis Message-Id: <199704030840.KAA24859@trinity.radio-do.de> Received: by trinity.radio-do.de (8.8.5/CLIENT-1.2.7-i) via EUnet id KAA24859; Thu, 3 Apr 1997 10:40:15 +0200 (CEST) Subject: Solved my 2940 timeout problem. To: freebsd-hackers@freebsd.org Date: Thu, 3 Apr 1997 10:40:15 +0200 (CEST) X-Mailer: ELM [version 2.4 PL24 ME8a] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk It sounds curious, but it may be worth a try for you, who have problems with their adaptec 2940 controllers. Last night I added a third disk to my system. The new drive is the last one on the bus, therefor I pulled the terminators off the second drive. Now my box is running 2.2.1-RELEASE for more then 10 hours with no problems so far. I checked the old terminators and voila one was broken :-( So it may be worth checking all the cables and terminators. Regards Frank From owner-freebsd-hackers Thu Apr 3 00:41:09 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA06775 for hackers-outgoing; Thu, 3 Apr 1997 00:41:09 -0800 (PST) Received: from florence.pavilion.net (florence.pavilion.net [194.242.128.25]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA06770 for ; Thu, 3 Apr 1997 00:41:06 -0800 (PST) Received: (from joe@localhost) by florence.pavilion.net (8.8.5/8.8.5) id JAA29887; Thu, 3 Apr 1997 09:40:30 +0100 (BST) Message-ID: <19970403094030.30623@pavilion.net> Date: Thu, 3 Apr 1997 09:40:30 +0100 From: Josef Karthauser To: Joerg Wunsch Cc: hackers@FreeBSD.ORG Subject: Re: Internal clock References: <3340C326.6150@cybernet.com> <19970401151947.61221@pavilion.net> <19970401203429.XR37988@uriah.heep.sax.de> <19970402123111.58518@pavilion.net> <19970402203510.RR52594@uriah.heep.sax.de> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Mailer: Mutt 0.64 In-Reply-To: <19970402203510.RR52594@uriah.heep.sax.de>; from J Wunsch on Wed, Apr 02, 1997 at 08:35:10PM +0200 Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Wed, Apr 02, 1997 at 08:35:10PM +0200, J Wunsch wrote: > > o The scheduling clock. This is a real clock with frequency that > happens to be 100. It isn't available to applications. > > o The statistics clock. This is a real clock with frequency that > happens to be 128. It isn't directly available to applications. > > So this explains the basic clock tick rates. There are exceptions, > for example if profiling is active (tick rate 1024), or if a kernel > device driver like pcaudio(4) (man page yet to be written by Søren :) > raises the clock rate to 16000. > Would you mind added the units to this man page. i.e. is it 100Hz, or 100KHz, or 100 camels? ;) Joe -- Josef Karthauser Technical Manager Email: joe@pavilion.net Pavilion Internet plc. [Tel: +44 1273 607072 Fax: +44 1273 607073] From owner-freebsd-hackers Thu Apr 3 01:40:22 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA09585 for hackers-outgoing; Thu, 3 Apr 1997 01:40:22 -0800 (PST) Received: from critter.dk.tfs.com ([140.145.230.252]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA09553; Thu, 3 Apr 1997 01:40:08 -0800 (PST) Received: from critter (localhost [127.0.0.1]) by critter.dk.tfs.com (8.8.5/8.8.5) with ESMTP id LAA00344; Thu, 3 Apr 1997 11:39:06 +0200 (CEST) Date: Thu, 03 Apr 1997 11:39:05 +0200 Message-ID: <338.860060345.1@critter> From: Poul-Henning Kamp Subject: New CTM meister... MIME-Version: 1.0 Content-Type: multipart/digest; boundary="----- =_aaaaaaaaaa" Content-Description: Blind Carbon Copy Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk To: undisclosed-recipients:; ------- =_aaaaaaaaaa Content-Type: message/rfc822 Content-Description: Original Message To: ctm-announce@freebsd.org Subject: New CTM meister... Reply-to: phk@freebsd.org Date: Thu, 03 Apr 1997 11:39:05 +0200 Message-ID: <338.860060345@critter> From: Poul-Henning Kamp Just this bit to make it official: Richard Wackerbarth Has taken over the job of CTM meister for FreeBSD, he is now the man behind the "ctm@FreeBSD.org" alias, which as always is the place to send email on the subject of CTM. Thanks for taking this over Richard! Poul-Henning -- Poul-Henning Kamp | phk@FreeBSD.ORG FreeBSD Core-team. http://www.freebsd.org/~phk | phk@login.dknet.dk Private mailbox. whois: [PHK] | phk@tfs.com TRW Financial Systems, Inc. Future will arrive by its own means, progress not so. ------- =_aaaaaaaaaa-- From owner-freebsd-hackers Thu Apr 3 01:49:37 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA10064 for hackers-outgoing; Thu, 3 Apr 1997 01:49:37 -0800 (PST) Received: from labinfo.iet.unipi.it (labinfo.iet.unipi.it [131.114.9.5]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id BAA10051 for ; Thu, 3 Apr 1997 01:49:10 -0800 (PST) Received: from localhost (luigi@localhost) by labinfo.iet.unipi.it (8.6.5/8.6.5) id KAA29446 for hackers@freebsd.org; Thu, 3 Apr 1997 10:57:04 +0200 From: Luigi Rizzo Message-Id: <199704030857.KAA29446@labinfo.iet.unipi.it> Subject: CD format for backups ? To: hackers@freebsd.org Date: Thu, 3 Apr 1997 10:57:04 +0200 (MET DST) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, a curiosity. Assuming I want to backup a filesystem to a CD and don't want to bother with mkisofs (it is not only a matter of time, I often do not have the space for the additional cd image!) can I do the following: - unmount the partition - use the partition (/dev/wd0f or /dev/rwd0f ?) as the source for the CD data then in order to read the data I would only want to mount /dev/wcd0c as FFS instead of cd9660 -- any hope that it works ? And also, should I use the char or block device ? (just to save burning a couple of CD's in the trial&error process...) Thanks Luigi -----------------------------+-------------------------------------- Luigi Rizzo | Dip. di Ingegneria dell'Informazione email: luigi@iet.unipi.it | Universita' di Pisa tel: +39-50-568533 | via Diotisalvi 2, 56126 PISA (Italy) fax: +39-50-568522 | http://www.iet.unipi.it/~luigi/ _____________________________|______________________________________ From owner-freebsd-hackers Thu Apr 3 02:50:47 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id CAA14171 for hackers-outgoing; Thu, 3 Apr 1997 02:50:47 -0800 (PST) Received: from plum.cyber.com.au (plum.cyber.com.au [203.7.155.24]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id CAA14166 for ; Thu, 3 Apr 1997 02:50:41 -0800 (PST) Received: (from darrenr@localhost) by plum.cyber.com.au (8.6.12/8.6.6) id UAA18193 for hackers@freebsd.org; Thu, 3 Apr 1997 20:50:17 +1000 From: Darren Reed Message-Id: <199704031050.UAA18193@plum.cyber.com.au> Subject: IP Filter in -current. To: hackers@freebsd.org Date: Thu, 3 Apr 1997 20:50:16 +1000 (EST) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I've imported IP Filter 3.2alpha4 into the -current sources for FreeBSD just now. This latest update should successfully build a LKM for use with FreeBSD 2.2 or later(?). It DOESN'T support being built into the kernel itself. Darren From owner-freebsd-hackers Thu Apr 3 05:17:07 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA21742 for hackers-outgoing; Thu, 3 Apr 1997 05:17:07 -0800 (PST) Received: from cheops.anu.edu.au (avalon@cheops.anu.edu.au [150.203.76.24]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA21733 for ; Thu, 3 Apr 1997 05:17:04 -0800 (PST) Message-Id: <199704031317.FAA21733@freefall.freebsd.org> Received: by cheops.anu.edu.au (1.37.109.16/16.2) id AA028673124; Thu, 3 Apr 1997 23:12:04 +1000 From: Darren Reed Subject: securelevel & IP filter To: hackers@freebsd.org Date: Thu, 3 Apr 1997 23:12:04 +1000 (EST) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I'm aware of the kernel recognising securelevel having a value of -1, 0, 1 or 2 and above but are there any plans for implementing something more than this ? It has been suggested that IP Filter disallow changes to filter rules if securelevel is set to some level...(I think 3 was the suggestion). However, before doing this, I want to poll people on whether they think three is appropriate, or should 10 be used (say) as a synonym for "firewall security level". Thoughts ? Darren From owner-freebsd-hackers Thu Apr 3 05:45:22 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA23020 for hackers-outgoing; Thu, 3 Apr 1997 05:45:22 -0800 (PST) Received: from jazz.snu.ac.kr (jazz.snu.ac.kr [147.46.102.36]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA23015 for ; Thu, 3 Apr 1997 05:45:20 -0800 (PST) Received: (from junker@localhost) by jazz.snu.ac.kr (8.8.5/8.8.4-procmail) id WAA04111 for freebsd-hackers@freebsd.org; Thu, 3 Apr 1997 22:39:11 +0900 (KST) From: Choi Jun Ho Message-Id: <199704031339.WAA04111@jazz.snu.ac.kr> Subject: How to make kernel boot from DOS to install? To: freebsd-hackers@freebsd.org Date: Thu, 3 Apr 1997 22:39:10 +0900 (KST) MIME-Version: 1.0 Content-Type: text/plain; charset=EUC-KR Content-Transfer-Encoding: 8bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I don't have seen WC FreeBSD CDROMs, but it should include some kind of boot image which can launch boot kernel and install program from DOS(using fbsdboot.exe. right?). I am currently working some kind of FreeBSD CD-ROM, so I want to install it from DOS with making install floppy... /usr/src/release seems to make boot floppies(I haven't run yet) only. How can I make some kernel image bootable from DOS, for installation? I wish your help... (I orderd FreeBSD CD-ROM from WC 10 days ago, but it is not arrived yet.. :< Oh so big world!) -- --------------------------------------------------------------^^--- Judgement Uninfected Naked Kind & Executive Ranger - J U N K E R from KONAMI 1990 "SD-Snatcher" in MSX2 Choi Jun Ho http://jazz.snu.ac.kr/~junker Distributed Computing System Lab, CS Dept.,Seoul National Univ.,ROK From owner-freebsd-hackers Thu Apr 3 05:57:26 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA23544 for hackers-outgoing; Thu, 3 Apr 1997 05:57:26 -0800 (PST) Received: from gvr.win.tue.nl (root@gvr.win.tue.nl [131.155.210.19]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA23539 for ; Thu, 3 Apr 1997 05:57:21 -0800 (PST) Received: (from guido@localhost) by gvr.win.tue.nl (8.8.5/8.8.2) id PAA10893 for FreeBSD-hackers@freebsd.org; Thu, 3 Apr 1997 15:57:08 +0200 (MET DST) From: Guido van Rooij Message-Id: <199704031357.PAA10893@gvr.win.tue.nl> Subject: apache like preforking apps and high loads To: FreeBSD-hackers@freebsd.org (FreeBSD-hackers) Date: Thu, 3 Apr 1997 15:57:08 +0200 (MET DST) X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk When looking at Apacha like applications, one often sees extermely high load averages. Apache preforks a number of processes that all block on accept(). When a request comes in, all process are woken up and the scheduler chooses one of the now runnable processes taht will succeed in the accept(). The other go back to sleep. There is a max amount of free processes (i.e.e processes doing an accept) and a min amunt. When less the minimum amount of processes are available, new ones are forked. When more then the maximum are available, some are killed. Now, as told, I regularly see loads like 20 on an otherwise idle machine, causing e.g. sendmail to not deliver mail anymore. One could argue that then the values of min and max above are wrong, but that is not completely true. These values should be optimised for those times where the amount of connections is the highest. But it will bite you when the server is relatively idle. My proposition would be to dynamically set the values of min and max according to the current load profile on the server. If the amount of connections in the last minut (or so) is 20, min should be set to 20 and max to 2*min. Of course we will still have the already present `real maximum' where max<`real maximum'. I'm wondering if this is the right approach...Waht do you guys think? -Guido From owner-freebsd-hackers Thu Apr 3 05:58:18 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA23597 for hackers-outgoing; Thu, 3 Apr 1997 05:58:18 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA23592 for ; Thu, 3 Apr 1997 05:58:16 -0800 (PST) Received: from dbws.etinc.com (dbws.etinc.com [204.141.95.130]) by etinc.com (8.8.3/8.6.9) with SMTP id JAA01244; Thu, 3 Apr 1997 09:00:21 -0500 (EST) Message-Id: <3.0.32.19970403091025.006a20c4@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Thu, 03 Apr 1997 09:10:28 -0500 To: Narvi , Joerg Wunsch From: dennis Subject: Re: Does de driver do 100MBIT Full Duplex? Cc: hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 10:23 AM 4/3/97 +0300, Narvi wrote: > > >On Wed, 2 Apr 1997, J Wunsch wrote: > >> As dennis wrote: >> >> > Is the driver that you send me directly for 2.1.7 (that worked >> > beautifully) in 2.2.1R? If not, WHY NOT! >> >> No need to shout. We aren't deaf exactly. >> >> Since Matt released it _after_ 2.2 has been cut. I've got a very >> explicit message from Matt when i've been asking him earlier (right in >> time to get something into 2.2 still) about a new version, that he >> considered the stuff that sneaked into NetBSD by that time too buggy >> to see it officially in FreeBSD. That's why we've been integrating >> the little hack still to make at least the DE21140A supported. We >> originally deferred the inclusion of this patch in anticipation of the >> new driver version. Where/what is this *patch*? db From owner-freebsd-hackers Thu Apr 3 06:09:41 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA24199 for hackers-outgoing; Thu, 3 Apr 1997 06:09:41 -0800 (PST) Received: from lattice.latte.it (line08.globalnet.it [194.185.53.40]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA24192 for ; Thu, 3 Apr 1997 06:09:32 -0800 (PST) Received: from localhost (ssigala@localhost) by lattice.latte.it (8.8.4/8.8.4) with SMTP id QAA00647 for ; Thu, 3 Apr 1997 16:08:42 +0200 (CEST) X-Authentication-Warning: lattice.latte.it: ssigala owned process doing -bs Date: Thu, 3 Apr 1997 16:08:41 +0200 (CEST) From: S Sigala X-Sender: ssigala@lattice.latte.it To: freebsd-hackers@freebsd.org Subject: Making holes in files with lseek() Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hello, i have just written this little program that replaces sequences of nul bytes with holes (using lseek()). This seem to work, but i would like to know the ideal length of the nul bytes sequence where a hole is better (requires less space on disk) than the sequence. In other words, how much disk space is wasted by a hole? Does every lseek() call (a seek below the end of the file) create a hole? Thanks in advance. Regards, -sandro -- CUT HERE ------------------------------------------------------------------ #define USEFUL_LENGTH 16 #include #include #include #include #include #include int lflag = USEFUL_LENGTH; void usage __P((void)); void setfile __P((char *path, struct stat *fs)); void process_file __P((char *filename)); int main(argc, argv) int argc; char *argv[]; { int c; while ((c = getopt(argc, argv, "l:")) != -1) switch (c) { case 'l': if ((lflag = atoi(optarg)) <= 0) errx(1, "invalid -l value"); break; case '?': default: usage(); /* NOTREACHED */ } argc -= optind; argv += optind; if (argc < 1) usage(); while (*argv) process_file(*argv++); return 0; } void usage() { fprintf(stderr, "usage: holeify [-l length] filename ...\n"); exit(1); } void setfile(path, fs) char *path; struct stat *fs; { static struct timeval tv[2]; /* Set file access and modification times. */ TIMESPEC_TO_TIMEVAL(&tv[0], &fs->st_atimespec); TIMESPEC_TO_TIMEVAL(&tv[1], &fs->st_mtimespec); utimes(path, tv); fs->st_mode &= S_ISUID | S_ISGID | S_ISVTX | S_IRWXU | S_IRWXG | S_IRWXO; /* Set owner and group. */ if (chown(path, fs->st_uid, fs->st_gid)) fs->st_mode &= ~(S_ISUID | S_ISGID); /* Set mode. */ chmod(path, fs->st_mode); /* Set flags. */ chflags(path, fs->st_flags); } void process_file(filename) char *filename; { unsigned char ibuf[BUFSIZ], obuf[BUFSIZ]; char tempfile[12]; struct stat fstat_s; int ifd, ofd, isize, osize; int i, num0; ifd = open(filename, O_RDONLY); ofd = mkstemp(strcpy(tempfile, "temp.XXXX")); num0 = osize = 0; while ((isize = read(ifd, ibuf, BUFSIZ)) > 0) for (i = 0; i < isize; i++) { if (ibuf[i] == '\0') num0++; else { if (num0 > 0) { if (osize > 0) { /* Flush the output buffer. */ write(ofd, obuf, osize); osize = 0; } if (num0 >= lflag) { /* Create a hole. */ lseek(ofd, num0, SEEK_CUR); num0 = 0; } else { /* Fill with zeros. */ for (; num0; num0--) { obuf[osize++] = '\0'; if (osize == BUFSIZ) { write(ofd, obuf, BUFSIZ); osize = 0; } } } } obuf[osize++] = ibuf[i]; /* Flush the output buffer if full. */ if (osize == BUFSIZ) { write(ofd, obuf, BUFSIZ); osize = 0; } } } if (osize > 0) write(ofd, obuf, osize); if (num0 > 0) { /* * Create a hole if required. */ if (num0 > 1) { if (num0 - 1 >= lflag) { /* Large enough: create a hole. */ lseek(ofd, num0 - 1, SEEK_CUR); } else { /* Fill with zeros. */ for (; num0 - 1; num0--) write(ofd, "", 1); } } /* * Put a null byte at the end to make the hole. */ write(ofd, "", 1); } fstat(ifd, &fstat_s); close(ifd); close(ofd); /* * Restore old file attributes and rename to old file name. */ setfile(tempfile, &fstat_s); rename(tempfile, filename); } From owner-freebsd-hackers Thu Apr 3 06:42:02 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA25695 for hackers-outgoing; Thu, 3 Apr 1997 06:42:02 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA25689 for ; Thu, 3 Apr 1997 06:41:58 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id GAA08793; Thu, 3 Apr 1997 06:41:59 -0800 (PST) To: Choi Jun Ho cc: freebsd-hackers@FreeBSD.ORG Subject: Re: How to make kernel boot from DOS to install? In-reply-to: Your message of "Thu, 03 Apr 1997 22:39:10 +0900." <199704031339.WAA04111@jazz.snu.ac.kr> Date: Thu, 03 Apr 1997 06:41:59 -0800 Message-ID: <8789.860078519@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > I don't have seen WC FreeBSD CDROMs, but it should include > some kind of boot image which can launch boot kernel and > install program from DOS(using fbsdboot.exe. right?). Right. You can just pass it any kernel to boot, though one which has the MFS root compiled in is necessary if you want the whole thing to be stand-alone. > I am currently working some kind of FreeBSD CD-ROM, so I want to > install it from DOS with making install floppy... /usr/src/release > seems to make boot floppies(I haven't run yet) only. How can I make > some kernel image bootable from DOS, for installation? The /usr/src/release stuff will also copy the MFS kernel aside during the release build and put it in ${CHROOTDIR}/R/stage/cdrom/disc1 (as well as ${CHROOTDIR}/R/stage/kernels). You could grab it from there. Jordan From owner-freebsd-hackers Thu Apr 3 07:06:09 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA26789 for hackers-outgoing; Thu, 3 Apr 1997 07:06:09 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA26781 for ; Thu, 3 Apr 1997 07:06:04 -0800 (PST) Received: from ntws (ntws.etinc.com [204.141.95.142]) by etinc.com (8.8.3/8.6.9) with SMTP id KAA01667; Thu, 3 Apr 1997 10:10:38 -0500 (EST) Message-Id: <3.0.32.19970403100503.00b885f0@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Thu, 03 Apr 1997 10:05:06 -0500 To: fyeung@fyeung8.netific.com (Francis Yeung) From: dennis Subject: Re: PPP Desperate for help! Cc: hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 02:20 PM 4/2/97 -0800, you wrote: > >Dennis, > > If you put 2 freebsd machines back to back connecting via ppp, >if both are LCP active, the probability that both magic numbers end >up to be identical is very likely. If the magic numbers are identical, >they won't do further LCP negotiation. There is always a probability and >the retry won't help either. Only if the implementation is defective.. the magic number for a given server should be random and the likelihood of duplication infinitesimally small. db From owner-freebsd-hackers Thu Apr 3 07:14:03 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA27073 for hackers-outgoing; Thu, 3 Apr 1997 07:14:03 -0800 (PST) Received: from sasami.jurai.net (winter@sasami.jurai.net [207.172.25.144]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA27067 for ; Thu, 3 Apr 1997 07:14:00 -0800 (PST) Received: from localhost (winter@localhost) by sasami.jurai.net (8.8.5/8.8.3) with SMTP id KAA03601; Thu, 3 Apr 1997 10:15:07 -0500 (EST) Date: Thu, 3 Apr 1997 10:15:07 -0500 (EST) From: "Matthew N. Dodd" To: Nate Williams cc: hackers@freebsd.org Subject: Re: 21" Monitor recommendations? In-Reply-To: <199704021828.LAA14450@rocky.mt.sri.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Wed, 2 Apr 1997, Nate Williams wrote: > Does anyone have any recommendations for a *real* 21" monitor that will > handle 1600x1200@72Hz, with a decent dot-pitch? My current monitor will > handle 1600@1200@76Hz, and has a .25 dot-pitch, and I'd like something > that is comparable. I'd like to spend under $2K/monitor if possible. We're looking at the Viewsonic P815s. 1800x1440 @ 75Hz 1600x1200 @ 91Hz .25mm ~$1600 /* Matthew N. Dodd | A memory retaining a love you had for life winter@jurai.net | As cruel as it seems nothing ever seems to http://www.jurai.net/~winter | go right - FLA M 3.1:53 */ From owner-freebsd-hackers Thu Apr 3 07:26:29 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA27442 for hackers-outgoing; Thu, 3 Apr 1997 07:26:29 -0800 (PST) Received: from root.com (implode.root.com [198.145.90.17]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA27437 for ; Thu, 3 Apr 1997 07:26:26 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by root.com (8.8.5/8.6.5) with SMTP id HAA03657; Thu, 3 Apr 1997 07:27:26 -0800 (PST) Message-Id: <199704031527.HAA03657@root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: Guido van Rooij cc: FreeBSD-hackers@freeBSD.org (FreeBSD-hackers) Subject: Re: apache like preforking apps and high loads In-reply-to: Your message of "Thu, 03 Apr 1997 15:57:08 +0200." <199704031357.PAA10893@gvr.win.tue.nl> From: David Greenman Reply-To: dg@root.com Date: Thu, 03 Apr 1997 07:27:26 -0800 Sender: owner-hackers@freeBSD.org X-Loop: FreeBSD.org Precedence: bulk >When looking at Apacha like applications, one often sees extermely high >load averages. Apache preforks a number of processes that all block >on accept(). When a request comes in, all process are woken up and the >scheduler chooses one of the now runnable processes taht will succeed in >the accept(). The other go back to sleep. Not any more. I changed this a few days ago. Now only one process is woken up. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-hackers Thu Apr 3 07:31:21 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA27751 for hackers-outgoing; Thu, 3 Apr 1997 07:31:21 -0800 (PST) Received: from bmccane.uit.net (bmccane.uit.net [208.129.189.48]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA27745 for ; Thu, 3 Apr 1997 07:31:14 -0800 (PST) Received: (from root@localhost) by bmccane.uit.net (8.8.5/8.7.3) id JAA14545; Thu, 3 Apr 1997 09:30:24 -0600 (CST) Date: Thu, 3 Apr 1997 09:30:22 -0600 (CST) From: Wm Brian McCane To: "Kenneth D. Merry" cc: Charles Henrich , freebsd-hackers@freebsd.org Subject: Re: netscape communicator versions? In-Reply-To: <199703160620.BAA10332@r74h25.res.gatech.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Sun, 16 Mar 1997, Kenneth D. Merry wrote: > Charles Henrich wrote: > > There are two bsd versions of communicator now, one file identifies as: > > > > netscape: 386BSD demand paged executable > > > > and one as > > > > netscape: BSD/386 demand paged (first page unmapped) pure ex > > > > Which is the first? > > > I suspect that the former, from: > > netscape-v40b2-export.x86-unknown-bsd.tar.gz > > is the BSD/386 1.1 version. > > I would guess that the latter, from: > > netscape-v40b2-export.x86-unknown-bsd2.tar.gz > > is the BSD/OS 2.x version. The first one was a little more stable > on my machine (SMP kernel, -current from around the end of January) than > the second, so I went with that one. > > There are still a number of bugs with it, though. For instance, it > doesn't seem to respond to -geometry, and crashes whenever you try to look > at the source to a page with a java applet in it. (check out the non-java > and java versions of the SGI home page -- it only crashes on the Java > version) > > > Ken > -- > Kenneth Merry > ken@r74h25.res.gatech.edu > Disclaimer: I don't speak for GTRI, GT, or Elvis. > Communicator has various bugs involving JAVA. I have about 6 graphics intensive applets I have been playing with (trying to design a NEAT logo 8). All of them operate correctly in `appletviewer' and look right on MSIE3. I am playing with a double-buffered video scheme, and something in Communicator keeps blanking my background buffer. brian +-------------------------------------+----------------------------------------+ He rides a cycle of mighty days, and \ Wm Brian and Lori McCane he represents the last great schizm \ McCane Consulting among the gods. Evil though he obviously \ root@bmccane.uit.net is, he is a mighty figure, this father of \ http://bmccane.uit.net/ my spirit, and I respect him as the sons \ http://bmccane.uit.net/~pictures/ of old did the fathers of their bodies. \ http://bmccane.uit.net/~bmccane/ Roger Zelazny - "Lord of Light" \ http://bmccane.uit.net/~bbs/ +---------------------------------------------+--------------------------------+ From owner-freebsd-hackers Thu Apr 3 07:57:54 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA29231 for hackers-outgoing; Thu, 3 Apr 1997 07:57:54 -0800 (PST) Received: from gw.itfs.nsk.su (ns.itfs.nsk.su [193.124.36.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id HAA29200 for ; Thu, 3 Apr 1997 07:57:06 -0800 (PST) Received: from itfs.UUCP (uucp@localhost) by gw.itfs.nsk.su (8.6.12/8.6.12) with UUCP id WAA25270 for hackers@freebsd.org; Thu, 3 Apr 1997 22:45:50 +0700 Received: by itfs.nsk.su; Thu, 3 Apr 97 23:06:02 +0600 (NSK) Received: (from daemon@localhost) by news.itfs.nsk.su (8.7.5/8.6.12) id WAA21246; Thu, 3 Apr 1997 22:40:30 +0700 (NSD) From: "Nickolay N. Dudorov" To: hackers@freebsd.org Subject: FreeBSD Elf-Kit and dynamic loading Date: 3 Apr 1997 15:40:29 GMT Message-ID: <5i0j1d$jtk@news.itfs.nsk.su> Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Today I try to install ELF-Kit on my 2.2-RELEASE system. I can succesfully (?) build binutils-2.7, gcc-2.7.2.1, almost all libraries from /usr/src/lib and libg++-2.7.2. After that I can test elf-cc and elf-g++ on provided in kit hello.c and hello.cc - AND they works. And after that I discovered that 'file hello' says: hello: ELF 32-bit LSB executable, Intel 80386, version 1 (FreeBSD) and file 'hello' is 47761 bytes long. So, there is a question - is there any possibility to build "dynamic ELF" program ? My elf-gcc produce identical results with and without '-static' flag. N.Dudorov From owner-freebsd-hackers Thu Apr 3 08:21:39 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA00756 for hackers-outgoing; Thu, 3 Apr 1997 08:21:39 -0800 (PST) Received: from nevis.oss.uswest.net (nevis.oss.uswest.net [204.147.85.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA00744 for ; Thu, 3 Apr 1997 08:21:33 -0800 (PST) Received: (from greg@localhost) by nevis.oss.uswest.net (8.8.2/8.8.2) id KAA29101; Thu, 3 Apr 1997 10:20:56 -0600 (CST) From: "Greg Rowe" Message-Id: <9704031020.ZM29099@nevis.oss.uswest.net> Date: Thu, 3 Apr 1997 10:20:55 -0600 In-Reply-To: Frank Nobis "Solved my 2940 timeout problem." (Apr 3, 10:40am) References: <199704030840.KAA24859@trinity.radio-do.de> X-Mailer: Z-Mail (3.2.1 10oct95) To: Frank Nobis , freebsd-hackers@freebsd.org Subject: Re: Solved my 2940 timeout problem. Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I think there's still a problem somewhere. I had a server that had been running with no problems for over a year die yesterday with: Apr 2 15:50:45 tahiti /kernel: sd1(ahc0:1:0): timed out while idle, LASTPHASE = = 0x1, SCSISIGI == 0x0 Apr 2 15:50:45 tahiti /kernel: SEQADDR == 0xd Apr 2 15:55:45 tahiti /kernel: FreeBSD 2.1.7-RELEASE #0: Mon Mar 3 02:22:25 CS T 1997 Greg On Apr 3, 10:40am, Frank Nobis wrote: > Subject: Solved my 2940 timeout problem. > It sounds curious, but it may be worth a try for you, who have > problems with their adaptec 2940 controllers. > > Last night I added a third disk to my system. > > The new drive is the last one on the bus, therefor I pulled the > terminators off the second drive. Now my box is running 2.2.1-RELEASE > for more then 10 hours with no problems so far. > > I checked the old terminators and voila one was broken :-( > > So it may be worth checking all the cables and terminators. > > Regards > Frank >-- End of excerpt from Frank Nobis -- Greg Rowe | U S West - Interact Services | INTERNET greg@uswest.net 111 Washington Ave. South | Fax: (612) 672-8537 Minneapolis, MN USA 55401 | Voice: (612) 672-8535 Never trust an operating system you don't have source for.... From owner-freebsd-hackers Thu Apr 3 08:52:50 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA02067 for hackers-outgoing; Thu, 3 Apr 1997 08:52:50 -0800 (PST) Received: from chai.plexuscom.com (chai.plexuscom.com [207.87.46.100]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA02061 for ; Thu, 3 Apr 1997 08:52:47 -0800 (PST) Received: from chai.plexuscom.com (localhost [127.0.0.1]) by chai.plexuscom.com (8.8.5/8.8.5) with ESMTP id LAA20772; Thu, 3 Apr 1997 11:53:34 -0500 (EST) Message-Id: <199704031653.LAA20772@chai.plexuscom.com> To: dg@root.com Cc: Guido van Rooij , FreeBSD-hackers@freebsd.org (FreeBSD-hackers) Subject: Re: apache like preforking apps and high loads In-reply-to: Your message of "Thu, 03 Apr 1997 07:27:26 PST." <199704031527.HAA03657@root.com> Date: Thu, 03 Apr 1997 11:53:34 -0500 From: Bakul Shah Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > >When looking at Apacha like applications, one often sees extermely high > >load averages. Apache preforks a number of processes that all block > >on accept(). When a request comes in, all process are woken up and the > >scheduler chooses one of the now runnable processes taht will succeed in > >the accept(). The other go back to sleep. > > Not any more. I changed this a few days ago. Now only one process is > woken up. Fairness is probably not an issue when an app. consists of a number of anonymous servers but in general one would want to make sure that if N processes are waiting on accept() on the same socket, no one process is starved of accepting. How do you ensure that? To guido: For apache like apps one idea is to have one process be the acceptor and have it pass a new socket to individual server processes (including doing some initial processing to determine which process should get this socket if not all server processes are equivalent). -- bakul From owner-freebsd-hackers Thu Apr 3 09:01:42 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA02545 for hackers-outgoing; Thu, 3 Apr 1997 09:01:42 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA02537 for ; Thu, 3 Apr 1997 09:01:32 -0800 (PST) Received: from ntws (ntws.etinc.com [204.141.95.142]) by etinc.com (8.8.3/8.6.9) with SMTP id MAA02283; Thu, 3 Apr 1997 12:04:45 -0500 (EST) Message-Id: <3.0.32.19970403115910.00b89450@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Thu, 03 Apr 1997 11:59:12 -0500 To: Narvi , Joerg Wunsch From: dennis Subject: Re: Does de driver do 100MBIT Full Duplex? Cc: hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 09:10 AM 4/3/97 -0500, dennis wrote: >At 10:23 AM 4/3/97 +0300, Narvi wrote: >> >> >>On Wed, 2 Apr 1997, J Wunsch wrote: >> >>> As dennis wrote: >>> >>> > Is the driver that you send me directly for 2.1.7 (that worked >>> > beautifully) in 2.2.1R? If not, WHY NOT! >>> >>> No need to shout. We aren't deaf exactly. >>> >>> Since Matt released it _after_ 2.2 has been cut. I've got a very >>> explicit message from Matt when i've been asking him earlier (right in >>> time to get something into 2.2 still) about a new version, that he >>> considered the stuff that sneaked into NetBSD by that time too buggy >>> to see it officially in FreeBSD. That's why we've been integrating >>> the little hack still to make at least the DE21140A supported. We >>> originally deferred the inclusion of this patch in anticipation of the >>> new driver version. It seems to me, that the fact that the most popular driver for FreeBSD doesn't work is a sufficient condition to hold up the release until it does. What you have now is a release that was supposed to be a "great saviour" feature-wise that is fundamentally unusable in its released form for a large number of users..... Dennis From owner-freebsd-hackers Thu Apr 3 09:05:25 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA02839 for hackers-outgoing; Thu, 3 Apr 1997 09:05:25 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA02833 for ; Thu, 3 Apr 1997 09:05:18 -0800 (PST) Received: from ntws (ntws.etinc.com [204.141.95.142]) by etinc.com (8.8.3/8.6.9) with SMTP id MAA02313 for ; Thu, 3 Apr 1997 12:10:01 -0500 (EST) Message-Id: <3.0.32.19970403120426.00b78860@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Thu, 03 Apr 1997 12:04:28 -0500 To: hackers@freebsd.org From: dennis Subject: 2.2.1R hangs on NFS load Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Using a 21041-based Ethernet card (not at 21040), NFS loads consistently hang with 2.2.1R (same hardware/system as worked with 2.1.5,6,7). Is this to be expected? Dennis From owner-freebsd-hackers Thu Apr 3 09:36:25 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA04559 for hackers-outgoing; Thu, 3 Apr 1997 09:36:25 -0800 (PST) Received: from boulder.syr.servtech.com (boulder.syr.servtech.com [206.106.144.94]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA04547 for ; Thu, 3 Apr 1997 09:36:17 -0800 (PST) Received: from boulder.syr.servtech.com (localhost [127.0.0.1]) by boulder.syr.servtech.com (8.8.5/8.8.5) with SMTP id MAA02604; Thu, 3 Apr 1997 12:41:32 -0500 (EST) Message-ID: <3343EBCC.2F1CF0FB@servtech.com> Date: Thu, 03 Apr 1997 12:41:32 -0500 From: Shawn Carey X-Mailer: Mozilla 3.0 (X11; I; FreeBSD 2.2.1-RELEASE i386) MIME-Version: 1.0 To: dennis CC: hackers@freebsd.org Subject: Re: 2.2.1R hangs on NFS load References: <3.0.32.19970403120426.00b78860@etinc.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk dennis wrote: > > Using a 21041-based Ethernet card (not at 21040), NFS loads consistently > hang with 2.2.1R (same hardware/system as worked with 2.1.5,6,7). Is > this to be expected? > I've seen problems with NFS on 2.2.1 also, though nothing to hang the machine. We have a remote site connected via 56k that occasionally mounts a filesystem or two. Quite often, an operation (cp, rm, etc) will just hang, or so the reports tell me. Yet another NFS server (on a different 56k to the client) running 2.2-ALPHA serves it's filesystems reliably. I've recently tried mounting with TCP transport, but it hasn't been long enough yet to know if this solves the problem. While we're on the subject of NFS, I see in the release notes for 2.2.1 that you can set NFS up with async writes? I've looked through the documentation, but I don't see anything referring to async NFS. Am i looking in the wrong places? -Shawn Carey From owner-freebsd-hackers Thu Apr 3 09:42:44 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA04867 for hackers-outgoing; Thu, 3 Apr 1997 09:42:44 -0800 (PST) Received: from mexico.brainstorm.eu.org (mexico.brainstorm.fr [193.56.58.253]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA04859 for ; Thu, 3 Apr 1997 09:42:38 -0800 (PST) Received: from brasil.brainstorm.eu.org (brasil.brainstorm.fr [193.56.58.33]) by mexico.brainstorm.eu.org (8.8.4/8.8.4) with ESMTP id TAA30390; Thu, 3 Apr 1997 19:42:25 +0200 Received: (from uucp@localhost) by brasil.brainstorm.eu.org (8.8.4/8.6.12) with UUCP id TAA06758; Thu, 3 Apr 1997 19:42:24 +0200 Received: (from roberto@localhost) by keltia.freenix.fr (8.8.5/keltia-uucp-2.9) id TAA29827; Thu, 3 Apr 1997 19:12:10 +0200 (CEST) Message-ID: <19970403191209.52889@keltia.freenix.fr> Date: Thu, 3 Apr 1997 19:12:09 +0200 From: Ollivier Robert To: "Nickolay N. Dudorov" Cc: "FreeBSD Hackers' list" Subject: Re: FreeBSD Elf-Kit and dynamic loading References: <5i0j1d$jtk@news.itfs.nsk.su> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.67 In-Reply-To: <5i0j1d$jtk@news.itfs.nsk.su>; from Nickolay N. Dudorov on Thu, Apr 03, 1997 at 03:40:29PM +0000 X-Operating-System: FreeBSD 3.0-CURRENT ctm#3153 Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk According to Nickolay N. Dudorov: > And after that I discovered that 'file hello' > says: > hello: ELF 32-bit LSB executable, Intel 80386, version 1 (FreeBSD) > > and file 'hello' is 47761 bytes long. > > So, there is a question - is there any possibility to > build "dynamic ELF" program ? My elf-gcc produce identical results > with and without '-static' flag. Are you sure you have the right naming for the libraries ? 258 [19:05] roberto@keltia:/tmp> cat foo.c main(){} 260 [19:06] roberto@keltia:/tmp> cc foo.c -o foo -v Reading specs from /usr/local/elf/lib/gcc-lib/i386-unknown-freebsdelf/2.7.2.1/specs gcc version 2.7.2.1 /usr/local/elf/lib/gcc-lib/i386-unknown-freebsdelf/2.7.2.1/cpp -lang-c -v -undef -D__GNUC__=2 -D__GNUC_MINOR__=7 -Dunix -Di386 -D__ELF__ -D__FreeBSD__=2 -D__unix__ -D__i386__ -D__ELF__ -D__FreeBSD__=2 -D__unix -D__i386 -Asystem(unix) -Asystem(FreeBSD) -Acpu(i386) -Amachine(i386) foo.c /tmp/cc029804.i GNU CPP version 2.7.2.1 (i386 FreeBSD/ELF) #include "..." search starts here: #include <...> search starts here: /usr/local/elf/lib/gcc-lib/i386-unknown-freebsdelf/2.7.2.1/include /usr/local/elf/lib/gcc-lib/i386-unknown-freebsdelf/2.7.2.1/sys-include /usr/local/elf/i386-unknown-freebsdelf/include End of search list. /usr/local/elf/lib/gcc-lib/i386-unknown-freebsdelf/2.7.2.1/cc1 /tmp/cc029804.i -quiet -dumpbase foo.c -version -o /tmp/cc029804.s GNU C version 2.7.2.1 (i386 FreeBSD/ELF) compiled by GNU C version 2.7.2.1. /usr/local/elf/i386-unknown-freebsdelf/bin/as -V -Qy -o /tmp/cc0298041.o /tmp/cc029804.s GNU assembler version 2.7 (i386-unknown-freebsdelf), using BFD version 2.7 /usr/local/elf/i386-unknown-freebsdelf/bin/ld -m elf_i386 -dynamic-linker /usr/libexec/ld-elf.so.1 -o foo /usr/local/elf/i386-unknown-freebsdelf/lib/crt1.o /usr/local/elf/i386-unknown-freebsdelf/lib/crti.o /usr/local/elf/i386-unknown-freebsdelf/lib/crtbegin.o -L/usr/local/elf/lib/gcc-lib/i386-unknown-freebsdelf/2.7.2.1 -L/usr/local/elf/i386-unknown-freebsdelf/lib /tmp/cc0298041.o -lgcc -lc -lgcc /usr/local/elf/i386-unknown-freebsdelf/lib/crtend.o /usr/local/elf/i386-unknown-freebsdelf/lib/crtn.o 261 [19:06] roberto@keltia:/tmp> ll foo -rwxr-xr-x 1 roberto wheel 4124 Apr 3 19:06 foo 262 [19:07] roberto@keltia:/tmp> size foo text data bss dec hex filename 652 811 8 1471 5bf foo In /usr/local/elf/lib: -r--r--r-- 1 bin bin 939192 Apr 3 19:01 libc.a -r--r--r-- 2 bin bin 465883 Apr 3 18:59 libc.so -r--r--r-- 2 bin bin 465883 Apr 3 18:59 libc.so.1 -r--r--r-- 1 bin bin 1009454 Apr 3 19:01 libc_pic.a With -static the executable is -rwxr-xr-x 1 roberto wheel 44788 Apr 3 19:11 foo objdump --all-headers gives: foo: file format elf32-i386 foo architecture: i386, flags 0x00000112: EXEC_P, HAS_SYMS, D_PAGED start address 0x080482e0 ... Dynamic Section: NEEDED libc.so.1 INIT 0x8048294 FINI 0x8048520 ... Sections: Idx Name Size VMA LMA File off Algn 0 .interp 00000019 080480d4 080480d4 000000d4 2**0 CONTENTS, ALLOC, LOAD, READONLY, DATA 1 .hash 00000048 080480f0 080480f0 000000f0 2**2 CONTENTS, ALLOC, LOAD, READONLY, DATA 2 .dynsym 000000d0 08048138 08048138 00000138 2**2 CONTENTS, ALLOC, LOAD, READONLY, DATA 3 .dynstr 00000071 08048208 08048208 00000208 2**0 CONTENTS, ALLOC, LOAD, READONLY, DATA 4 .rel.plt 00000018 0804827c 0804827c 0000027c 2**2 CONTENTS, ALLOC, LOAD, READONLY, DATA ... -- Ollivier ROBERT -=- FreeBSD: There are no limits -=- roberto@keltia.freenix.fr Usenet Canal Historique From owner-freebsd-hackers Thu Apr 3 09:42:45 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA04874 for hackers-outgoing; Thu, 3 Apr 1997 09:42:45 -0800 (PST) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA04865 for ; Thu, 3 Apr 1997 09:42:42 -0800 (PST) Received: from austin.polstra.com (austin.polstra.com [206.213.73.10]) by who.cdrom.com (8.8.5/8.6.11) with ESMTP id JAA18411 for ; Thu, 3 Apr 1997 09:42:40 -0800 (PST) Received: from austin.polstra.com (jdp@localhost) by austin.polstra.com (8.8.5/8.8.5) with ESMTP id JAA16810; Thu, 3 Apr 1997 09:41:05 -0800 (PST) Message-Id: <199704031741.JAA16810@austin.polstra.com> To: dennis@etinc.com Subject: Re: Does de driver do 100MBIT Full Duplex? Newsgroups: polstra.freebsd.hackers In-Reply-To: <3.0.32.19970403091025.006a20c4@etinc.com> References: <3.0.32.19970403091025.006a20c4@etinc.com> Organization: Polstra & Co., Seattle, WA Cc: hackers@freebsd.org Date: Thu, 03 Apr 1997 09:41:05 -0800 From: John Polstra Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In article <3.0.32.19970403091025.006a20c4@etinc.com>, dennis wrote: > >> That's why we've been integrating > >> the little hack still to make at least the DE21140A supported. We > >> originally deferred the inclusion of this patch in anticipation of the > >> new driver version. > > Where/what is this *patch*? In revision 1.60 of src/sys/pci/if_de.c: revision 1.60 date: 1997/02/23 10:57:30; author: joerg; state: Exp; lines: +15 -1 Add support for the SMC9332BDT that's using the DE21140A chip. This is merely a stop-gap measure until we can import an upgraded driver from Matt Thomas. Closes PR # 2696, and most likely also 2767. OKed by: core John -- John Polstra jdp@polstra.com John D. Polstra & Co., Inc. Seattle, Washington USA "Self-knowledge is always bad news." -- John Barth From owner-freebsd-hackers Thu Apr 3 09:56:36 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA05771 for hackers-outgoing; Thu, 3 Apr 1997 09:56:36 -0800 (PST) Received: from scanner.worldgate.com (scanner.worldgate.com [198.161.84.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA05766 for ; Thu, 3 Apr 1997 09:56:34 -0800 (PST) Received: from znep.com (uucp@localhost) by scanner.worldgate.com (8.8.5/8.7.3) with UUCP id KAA06075; Thu, 3 Apr 1997 10:56:06 -0700 (MST) Received: from localhost (marcs@localhost) by alive.znep.com (8.7.5/8.7.3) with SMTP id KAA01937; Thu, 3 Apr 1997 10:50:20 -0700 (MST) Date: Thu, 3 Apr 1997 10:50:19 -0700 (MST) From: Marc Slemko To: Bakul Shah cc: Guido van Rooij , FreeBSD-hackers Subject: Re: apache like preforking apps and high loads In-Reply-To: <199704031653.LAA20772@chai.plexuscom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Thu, 3 Apr 1997, Bakul Shah wrote: > > >When looking at Apacha like applications, one often sees extermely high > > >load averages. Apache preforks a number of processes that all block > > >on accept(). When a request comes in, all process are woken up and the > > >scheduler chooses one of the now runnable processes taht will succeed in > > >the accept(). The other go back to sleep. > > > > Not any more. I changed this a few days ago. Now only one process is > > woken up. > > Fairness is probably not an issue when an app. consists of a number > of anonymous servers but in general one would want to make sure that > if N processes are waiting on accept() on the same socket, no one > process is starved of accepting. How do you ensure that? For something like Apache you want the _least_ equal distribution possible, ie. a stack of most recently used processes from which you pop a server to process a request. Right now, if you have 100 servers running and get requests at a rate that one could handle, all of them will still be used. This is bad; it hurts locality of reference a lot. On some architectures, it has a significant impact WRT caching, and also means that if you don't have enough RAM for all server processes then they will each end up being swapped out and in again. The reason why Apache doesn't do this is the implementation details. It is difficult to get a good implementation that is efficient (so you don't create more overhead than you remove) and portable. Sort of timed destruction of child processes could help; right now you are limited to how it is hard coded in Apache and if you set MaxSpareServers too low you run into problems with short term load fluctuations. > To guido: For apache like apps one idea is to have one process be > the acceptor and have it pass a new socket to individual server > processes (including doing some initial processing to determine > which process should get this socket if not all server processes are > equivalent). This has a lot of portability issues and can lead to running into things such as per process fd limits more quickly on some architectures. From owner-freebsd-hackers Thu Apr 3 09:56:42 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA05800 for hackers-outgoing; Thu, 3 Apr 1997 09:56:42 -0800 (PST) Received: from labinfo.iet.unipi.it (labinfo.iet.unipi.it [131.114.9.5]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id JAA05749 for ; Thu, 3 Apr 1997 09:56:11 -0800 (PST) Received: from localhost (luigi@localhost) by labinfo.iet.unipi.it (8.6.5/8.6.5) id TAA00593; Thu, 3 Apr 1997 19:03:19 +0200 From: Luigi Rizzo Message-Id: <199704031703.TAA00593@labinfo.iet.unipi.it> Subject: Re: Does de driver do 100MBIT Full Duplex? To: dennis@etinc.com (dennis) Date: Thu, 3 Apr 1997 19:03:18 +0200 (MET DST) Cc: narvi@haldjas.folklore.ee, joerg_wunsch@uriah.heep.sax.de, hackers@FreeBSD.ORG In-Reply-To: <3.0.32.19970403115910.00b89450@etinc.com> from "dennis" at Apr 3, 97 11:58:53 am X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > It seems to me, that the fact that the most popular driver for FreeBSD ^^^^^^^^^^^^^^^^^^^^^^^ you mean the "ed" driver ? :) Seriously, the 21x4x cards are nice and cheap (considering their performance), but I suspect there are far more NE2000 clones around (both PCI and ISA) than the total of other network cards. What you say (delay the release of 2.2 until we had a stable "de" driver) would have only been possible if there were someone paid for fixing the driver with some reasonable deadline. Nothing like that in our project. Cheers Luigi -----------------------------+-------------------------------------- Luigi Rizzo | Dip. di Ingegneria dell'Informazione email: luigi@iet.unipi.it | Universita' di Pisa tel: +39-50-568533 | via Diotisalvi 2, 56126 PISA (Italy) fax: +39-50-568522 | http://www.iet.unipi.it/~luigi/ _____________________________|______________________________________ From owner-freebsd-hackers Thu Apr 3 10:16:45 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA06593 for hackers-outgoing; Thu, 3 Apr 1997 10:16:45 -0800 (PST) Received: from austin.polstra.com (austin.polstra.com [206.213.73.10]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA06586 for ; Thu, 3 Apr 1997 10:16:39 -0800 (PST) Received: from austin.polstra.com (jdp@localhost) by austin.polstra.com (8.8.5/8.8.5) with ESMTP id KAA17009; Thu, 3 Apr 1997 10:14:32 -0800 (PST) Message-Id: <199704031814.KAA17009@austin.polstra.com> To: nnd@info.itfs.nsk.su Subject: Re: FreeBSD Elf-Kit and dynamic loading Newsgroups: polstra.freebsd.hackers In-Reply-To: <5i0j1d$jtk@news.itfs.nsk.su> References: <5i0j1d$jtk@news.itfs.nsk.su> Organization: Polstra & Co., Seattle, WA Cc: hackers@freebsd.org Date: Thu, 03 Apr 1997 10:14:32 -0800 From: John Polstra Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In article <5i0j1d$jtk@news.itfs.nsk.su>, Nickolay N. Dudorov wrote: > Today I try to install ELF-Kit on my 2.2-RELEASE > system. I can succesfully (?) build binutils-2.7, gcc-2.7.2.1, > almost all libraries from /usr/src/lib and libg++-2.7.2. > > After that I can test elf-cc and elf-g++ on > provided in kit hello.c and hello.cc - AND they works. > > And after that I discovered that 'file hello' > says: > hello: ELF 32-bit LSB executable, Intel 80386, version 1 (FreeBSD) ElfKit _should_ be producing dynamically linked executables. Hmm ... Except on FreeBSD-current, the "file" command does not report whether an ELF file is dynamically linked or not. For example, on a dynamically-linked ELF file "hello", the -2.2 "file" command says: hello: ELF 32-bit LSB executable, Intel 80386, version 1 (FreeBSD) The same command on a -current system says: hello: ELF 32-bit LSB executable, Intel 80386, version 1 (FreeBSD), dynamically linked, not stripped (on a single line). At first, I thought you had a dynamically linked program but did not realize it. But ... > and file 'hello' is 47761 bytes long. > > So, there is a question - is there any possibility to > build "dynamic ELF" program ? My elf-gcc produce identical results > with and without '-static' flag. 47761 bytes is much too large (my "hello" is only 4121 bytes even before stripping it). So I think you really are getting a statically linked executable, for some reason. Are you sure your ELF shared libc is installed properly? You should have this in /usr/local/elf/i386-unknown-freebsdelf/lib: lrwxr-xr-x 1 root bin 9 Apr 16 1996 libc.so -> libc.so.3 -rwxr-xr-x 1 root bin 404921 Apr 15 1996 libc.so.3 If that looks OK, then please run the following command, and send me the output it produces: elf-cc -v -o hello hello.c Thanks, John -- John Polstra jdp@polstra.com John D. Polstra & Co., Inc. Seattle, Washington USA "Self-knowledge is always bad news." -- John Barth From owner-freebsd-hackers Thu Apr 3 10:20:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA06854 for hackers-outgoing; Thu, 3 Apr 1997 10:20:11 -0800 (PST) Received: from aries.bb.cc.wa.us (root@[208.8.136.11]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA06835 for ; Thu, 3 Apr 1997 10:20:08 -0800 (PST) Received: from localhost (chris@localhost) by aries.bb.cc.wa.us (8.8.3/8.6.9) with SMTP id KAA20666; Thu, 3 Apr 1997 10:15:25 -0800 (PST) Date: Thu, 3 Apr 1997 10:15:25 -0800 (PST) From: Chris Coleman To: Luigi Rizzo cc: hackers@freebsd.org Subject: Re: Does de driver do 100MBIT Full Duplex? In-Reply-To: <199704031703.TAA00593@labinfo.iet.unipi.it> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Thu, 3 Apr 1997, Luigi Rizzo wrote: > > It seems to me, that the fact that the most popular driver for FreeBSD > ^^^^^^^^^^^^^^^^^^^^^^^ > > you mean the "ed" driver ? :) > I agree, of the 500 computers on Campus, all most all of them have the NE2000 cards in them, Including some of our FreeBSD servers. We use the "fxp" driver otherwise. Christopher J. Coleman (chris@aries.bb.cc.wa.us) Computer Support Technician I (509)-766-8873 Big Bend Community College Internet Instructor FreeBSD Book Project: http://vinyl.quickweb.com/~chrisc/book.html Disclaimer: Even Though it has My Name on it, Doesn't mean I said it. From owner-freebsd-hackers Thu Apr 3 10:41:32 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA08000 for hackers-outgoing; Thu, 3 Apr 1997 10:41:32 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA07982 for ; Thu, 3 Apr 1997 10:41:22 -0800 (PST) Received: from ntws (ntws.etinc.com [204.141.95.142]) by etinc.com (8.8.3/8.6.9) with SMTP id NAA02871; Thu, 3 Apr 1997 13:44:15 -0500 (EST) Message-Id: <3.0.32.19970403133840.00af2770@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Thu, 03 Apr 1997 13:38:42 -0500 To: Luigi Rizzo From: dennis Subject: Re: Does de driver do 100MBIT Full Duplex? Cc: narvi@haldjas.folklore.ee, joerg_wunsch@uriah.heep.sax.de, hackers@FreeBSD.ORG Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk At 07:03 PM 4/3/97 +0200, Luigi Rizzo wrote: >> It seems to me, that the fact that the most popular driver for FreeBSD > ^^^^^^^^^^^^^^^^^^^^^^^ > >you mean the "ed" driver ? :) > Sorry...most popular PCI driver....... Although at this point I believe that the problems I am having have more to do with NFS than the de driver. Alas, the plans to seriously promote FreeBSD that we had clearly have to go on hold..... Dennis >Seriously, the 21x4x cards are nice and cheap (considering their >performance), but I suspect there are far more NE2000 clones around >(both PCI and ISA) than the total of other network cards. > >What you say (delay the release of 2.2 until we had a stable "de" >driver) would have only been possible if there were someone paid for >fixing the driver with some reasonable deadline. Nothing like that in >our project. > > Cheers > Luigi >-----------------------------+-------------------------------------- >Luigi Rizzo | Dip. di Ingegneria dell'Informazione >email: luigi@iet.unipi.it | Universita' di Pisa >tel: +39-50-568533 | via Diotisalvi 2, 56126 PISA (Italy) >fax: +39-50-568522 | http://www.iet.unipi.it/~luigi/ >_____________________________|______________________________________ > > Emerging Technologies, Inc. Router cards for BSD/OS, FreeBSD, NetBSD, OPenBSD and Linux Standalone Routers Bandwidth Allocation/Limiter Manager http://www.etinc.com sales@etinc.com (516) 271-4525 From owner-freebsd-hackers Thu Apr 3 10:42:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA08050 for hackers-outgoing; Thu, 3 Apr 1997 10:42:11 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA08042 for ; Thu, 3 Apr 1997 10:42:05 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id LAA16584; Thu, 3 Apr 1997 11:24:38 -0700 From: Terry Lambert Message-Id: <199704031824.LAA16584@phaeton.artisoft.com> Subject: Re: Making holes in files with lseek() To: ssigala@globalnet.it (S Sigala) Date: Thu, 3 Apr 1997 11:24:38 -0700 (MST) Cc: freebsd-hackers@freebsd.org In-Reply-To: from "S Sigala" at Apr 3, 97 04:08:41 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Hello, i have just written this little program that replaces sequences > of nul bytes with holes (using lseek()). This seem to work, but i would > like to know the ideal length of the nul bytes sequence where a hole is > better (requires less space on disk) than the sequence. > In other words, how much disk space is wasted by a hole? Does every > lseek() call (a seek below the end of the file) create a hole? > > Thanks in advance. This requires that there be enough space for the file and a "holified" version of the file. I think one could use a script containing a tar in of the source file and a tar out of the destination file, using the "--sparse" or "-S" argument, and achieve the same effect... May I suggest implementing F_FREESP for fcntl(2) instead, and operating on the existing file? This would allow you to make an existing file sparse by "marking" holes with F_FREESP. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Thu Apr 3 10:52:52 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA08606 for hackers-outgoing; Thu, 3 Apr 1997 10:52:52 -0800 (PST) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA08598; Thu, 3 Apr 1997 10:52:47 -0800 (PST) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by mail.cdsnet.net (8.8.5/8.7.3) with SMTP id KAA11632; Thu, 3 Apr 1997 10:52:23 -0800 (PST) Date: Thu, 3 Apr 1997 10:52:22 -0800 (PST) From: Jaye Mathisen To: Gianmarco Giovannelli cc: questions@freebsd.org, hackers@freebsd.org Subject: Re: ... routed[47]: punt RTM_LOSING without gateway In-Reply-To: <3.0.1.32.19970404080223.00718b2c@scotty.masternet.it> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I could've sworn Bill Fenner said he fixed this, that it was a pathological case or something that couldn't happen, but I still get these as well. On Fri, 4 Apr 1997, Gianmarco Giovannelli wrote: > > Hello, what does this messsage stand for ? > > spock routed[47]: punt RTM_LOSING without gateway > > I have written this questions few mounths ago, but I hadn't receive any > answer, but the message still continue to fill my screens ... > I looked in the man page but with no luck.... perhaps I search in the wrong > place ? :-) > > Thanks again.... > > > Regards... > Gianmarco ( gmarco@masternet.it ) > "Unix expert since yesterday" > > Home page: http://www2.masternet.it/~gmarco > Server page: http://www2.masternet.it/ > From owner-freebsd-hackers Thu Apr 3 11:02:05 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA09125 for hackers-outgoing; Thu, 3 Apr 1997 11:02:05 -0800 (PST) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA09119 for ; Thu, 3 Apr 1997 11:02:03 -0800 (PST) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by mail.cdsnet.net (8.8.5/8.7.3) with SMTP id LAA12934; Thu, 3 Apr 1997 11:01:47 -0800 (PST) Date: Thu, 3 Apr 1997 11:01:46 -0800 (PST) From: Jaye Mathisen To: Frank Nobis cc: freebsd-hackers@freebsd.org Subject: Re: Solved my 2940 timeout problem. In-Reply-To: <199704030840.KAA24859@trinity.radio-do.de> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I have gone to just yanking all terminators, and putting active terminators on the ends of all my scsi chains. I haven't had any significant SCSI related problems since. > > So it may be worth checking all the cables and terminators. > > Regards > Frank > From owner-freebsd-hackers Thu Apr 3 11:10:24 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA09623 for hackers-outgoing; Thu, 3 Apr 1997 11:10:24 -0800 (PST) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA09616 for ; Thu, 3 Apr 1997 11:10:22 -0800 (PST) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by mail.cdsnet.net (8.8.5/8.7.3) with SMTP id LAA13297; Thu, 3 Apr 1997 11:04:05 -0800 (PST) Date: Thu, 3 Apr 1997 11:04:04 -0800 (PST) From: Jaye Mathisen To: Narvi cc: Joerg Wunsch , hackers@freebsd.org Subject: Re: Does de driver do 100MBIT Full Duplex? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I can mail this stuff off if nobody else has, I took a whack at porting the driver to 2.2.1, and reached brain drain on the IF_MEDIA stuff. But I have all the parts sitting here. Just let me know. > and the connection breaks before it ever reaches me. As mail still seems > to reach me, could someone please tar+gzip the needed files together and > forward them to me? > > Sander From owner-freebsd-hackers Thu Apr 3 11:41:25 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA11394 for hackers-outgoing; Thu, 3 Apr 1997 11:41:25 -0800 (PST) Received: from gordini.inf.ufrgs.br (maspohn@[143.54.83.8]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA11389 for ; Thu, 3 Apr 1997 11:41:19 -0800 (PST) Received: from localhost (maspohn@localhost) by gordini.inf.ufrgs.br (8.7.5/8.7.3) with SMTP id QAA01147 for ; Thu, 3 Apr 1997 16:38:29 -0300 Date: Thu, 3 Apr 1997 16:38:27 -0300 (EST) From: Marco Aurelio Spohn To: freebsd-hackers@FreeBSD.ORG Subject: How to obtain time during the packet filtering process... Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi! I'm working with the packet filter "ipfw". I intend to measure the time spent during the packet filter process but it's not possible, for instance, using "gettimeofday" to obtain the time before and after the filtering. Do anybody know how i could do it? Thanks in advance! Marco Aurelio Spohn maspohn@inf.ufrgs.br Universidade Federal do Rio Grande do Sul Brasil - Porto Alegre - RS From owner-freebsd-hackers Thu Apr 3 11:44:07 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA11528 for hackers-outgoing; Thu, 3 Apr 1997 11:44:07 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA11523 for ; Thu, 3 Apr 1997 11:44:02 -0800 (PST) Received: from ntws (ntws.etinc.com [204.141.95.142]) by etinc.com (8.8.3/8.6.9) with SMTP id OAA03259 for ; Thu, 3 Apr 1997 14:48:40 -0500 (EST) Message-Id: <3.0.32.19970403144304.00b8d1f0@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Thu, 03 Apr 1997 14:43:06 -0500 To: hackers@freebsd.org From: dennis Subject: NFS problems with 2.2.1R Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I've concluded that the problems that I've been having are *not* due to a buggy de driver but NFS. I have the same symptoms on a totally different machine with an ne2000 card as I had with the de PCI card. My server is a 2.1.6R box which has been reliably been working for quite some time (previously with 2.1.5 of course). When loading 2.2.1R via NFS, it consistantly hangs loading bin, however not in the same place. The system completely hangs (ie alt-F keys do not work, cant ctl-alt-del...). Anyone with any ideas on how to get around this? I loaded 2.2R without incident on the same hardware. Dennis From owner-freebsd-hackers Thu Apr 3 11:56:23 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA12361 for hackers-outgoing; Thu, 3 Apr 1997 11:56:23 -0800 (PST) Received: from alpha.xerox.com (alpha.Xerox.COM [13.1.64.93]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id LAA12336; Thu, 3 Apr 1997 11:56:14 -0800 (PST) Received: from crevenia.parc.xerox.com ([13.2.116.11]) by alpha.xerox.com with SMTP id <60066(9)>; Thu, 3 Apr 1997 11:31:24 PST Received: from localhost by crevenia.parc.xerox.com with SMTP id <177486>; Thu, 3 Apr 1997 11:30:20 -0800 To: Jaye Mathisen cc: Gianmarco Giovannelli , questions@freebsd.org, hackers@freebsd.org Subject: Re: ... routed[47]: punt RTM_LOSING without gateway In-reply-to: Your message of "Thu, 03 Apr 97 10:52:22 PST." Date: Thu, 3 Apr 1997 11:30:15 PST From: Bill Fenner Message-Id: <97Apr3.113020pst.177486@crevenia.parc.xerox.com> Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk It means that a TCP connection to a host to which you're directly connected retransmitted a packet. e.g. it's nothing to worry about. I thought that it was only a high-level-debugging message in the new routed but I must not have been paying enough attention. In any case, it can be safely ignored; at worst it's a symptom of some other problem that you may be having. Bill From owner-freebsd-hackers Thu Apr 3 12:22:47 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA13770 for hackers-outgoing; Thu, 3 Apr 1997 12:22:47 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA13761 for ; Thu, 3 Apr 1997 12:22:39 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id WAA26975 for hackers@freebsd.org; Thu, 3 Apr 1997 22:22:23 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id WAA29940; Thu, 3 Apr 1997 22:18:49 +0200 (MET DST) Message-ID: <19970403221849.DL32306@uriah.heep.sax.de> Date: Thu, 3 Apr 1997 22:18:49 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freebsd.org Subject: Re: Internal clock References: <3340C326.6150@cybernet.com> <19970401151947.61221@pavilion.net> <19970401203429.XR37988@uriah.heep.sax.de> <19970402123111.58518@pavilion.net> <19970402203510.RR52594@uriah.heep.sax.de> <19970403094030.30623@pavilion.net> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <19970403094030.30623@pavilion.net>; from Josef Karthauser on Apr 3, 1997 09:40:30 +0100 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Josef Karthauser wrote: > Would you mind added the units to this man page. i.e. is it 100Hz, > or 100KHz, or 100 camels? ;) Since it's a rate, i think 100 camels per second. :-) -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Thu Apr 3 12:34:57 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA14457 for hackers-outgoing; Thu, 3 Apr 1997 12:34:57 -0800 (PST) Received: from scanner.worldgate.com (scanner.worldgate.com [198.161.84.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA14452 for ; Thu, 3 Apr 1997 12:34:55 -0800 (PST) Received: from znep.com (uucp@localhost) by scanner.worldgate.com (8.8.5/8.7.3) with UUCP id NAA12801; Thu, 3 Apr 1997 13:34:48 -0700 (MST) Received: from localhost (marcs@localhost) by alive.znep.com (8.7.5/8.7.3) with SMTP id NAA02837; Thu, 3 Apr 1997 13:34:30 -0700 (MST) Date: Thu, 3 Apr 1997 13:34:29 -0700 (MST) From: Marc Slemko To: David Greenman cc: FreeBSD-hackers Subject: Re: apache like preforking apps and high loads In-Reply-To: <199704031527.HAA03657@root.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Thu, 3 Apr 1997, David Greenman wrote: > >When looking at Apacha like applications, one often sees extermely high > >load averages. Apache preforks a number of processes that all block > >on accept(). When a request comes in, all process are woken up and the > >scheduler chooses one of the now runnable processes taht will succeed in > >the accept(). The other go back to sleep. > > Not any more. I changed this a few days ago. Now only one process is > woken up. Looking at these changes and the other code around them, it appears to me like the net effect is that processes are added to the tail of the queue when they block on accept() and removed from (near) the head with wakeup_one. Am I reading this right, or am I getting my queue ends mixed up somewhere? For code that does similar things to what Apache does, it would be far better to treat it as a stack and have wakeup_one() remove from the tail since you want the same processes to serve as many requests as possible. In the grand scheme of things it may be a bad idea to have wakeup_one() behave this way, but it looks like it would be useful in this case. From owner-freebsd-hackers Thu Apr 3 12:50:49 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA15365 for hackers-outgoing; Thu, 3 Apr 1997 12:50:49 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA15360 for ; Thu, 3 Apr 1997 12:50:46 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id WAA27321 for hackers@FreeBSD.ORG; Thu, 3 Apr 1997 22:50:44 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id WAA00117; Thu, 3 Apr 1997 22:32:56 +0200 (MET DST) Message-ID: <19970403223255.SE29929@uriah.heep.sax.de> Date: Thu, 3 Apr 1997 22:32:55 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@FreeBSD.ORG Subject: Re: Does de driver do 100MBIT Full Duplex? References: <3.0.32.19970403115910.00b89450@etinc.com> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <3.0.32.19970403115910.00b89450@etinc.com>; from dennis on Apr 3, 1997 11:59:12 -0500 Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk As dennis wrote: > It seems to me, that the fact that the most popular driver for FreeBSD > doesn't work is a sufficient condition to hold up the release until it > does. Even if Matt told me before that it's uncertain whether and when he will be able to continue his work on this, given his current other obligations? Btw., i've just delivered a customer's machine running FreeBSD 2.2 with a dual-channel SMC card using the DEC chips. It was really plug&play. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Thu Apr 3 12:58:59 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA15744 for hackers-outgoing; Thu, 3 Apr 1997 12:58:59 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA15728 for ; Thu, 3 Apr 1997 12:58:54 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id WAA27453 for hackers@FreeBSD.ORG; Thu, 3 Apr 1997 22:58:52 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id WAA00148; Thu, 3 Apr 1997 22:44:55 +0200 (MET DST) Message-ID: <19970403224454.SP47619@uriah.heep.sax.de> Date: Thu, 3 Apr 1997 22:44:54 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@FreeBSD.ORG Subject: Re: CD format for backups ? References: <199704030857.KAA29446@labinfo.iet.unipi.it> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199704030857.KAA29446@labinfo.iet.unipi.it>; from Luigi Rizzo on Apr 3, 1997 10:57:04 +0200 Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk As Luigi Rizzo wrote: > - use the partition (/dev/wd0f or /dev/rwd0f ?) as the source for the > CD data > > then in order to read the data I would only want to mount /dev/wcd0c as > FFS instead of cd9660 -- any hope that it works ? Well, UFS is currently still a little iffy when it comes to devices that are not using 512 bytes blocksize. So, ``it depends''. > And also, should I use the char or block device ? Use them as their names suggest: the raw device for writing raw data (i.e., reading from your harddisk), the buffered device for mounting filesystems using the buffer cache. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Thu Apr 3 13:13:18 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA16438 for hackers-outgoing; Thu, 3 Apr 1997 13:13:18 -0800 (PST) Received: from trinity.radio-do.de (fn@trinity.Radio-do.de [193.101.164.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA16429 for ; Thu, 3 Apr 1997 13:13:11 -0800 (PST) From: Frank Nobis Message-Id: <199704032112.XAA24684@trinity.radio-do.de> Received: by trinity.radio-do.de (8.8.5/CLIENT-1.2.7-i) via EUnet id XAA24684; Thu, 3 Apr 1997 23:12:29 +0200 (CEST) Subject: Re: Solved my 2940 timeout problem. To: mrcpu@cdsnet.net (Jaye Mathisen) Date: Thu, 3 Apr 1997 23:12:29 +0200 (CEST) Cc: fn@radio-do.de, freebsd-hackers@freebsd.org In-Reply-To: from "Jaye Mathisen" at Apr 3, 97 11:01:46 am X-Mailer: ELM [version 2.4 PL24 ME8a] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > I have gone to just yanking all terminators, and putting active > terminators on the ends of all my scsi chains. I haven't had any > significant SCSI related problems since. > Often the simple solutions are the best. I will keep an eye on my little computer box. Frank From owner-freebsd-hackers Thu Apr 3 13:51:45 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA18896 for hackers-outgoing; Thu, 3 Apr 1997 13:51:45 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id NAA18891 for ; Thu, 3 Apr 1997 13:51:42 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id XAA27997 for hackers@freebsd.org; Thu, 3 Apr 1997 23:51:40 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id XAA00411; Thu, 3 Apr 1997 23:37:38 +0200 (MET DST) Message-ID: <19970403233738.KY42145@uriah.heep.sax.de> Date: Thu, 3 Apr 1997 23:37:38 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freebsd.org Subject: Re: securelevel & IP filter References: <199704031317.FAA21733@freefall.freebsd.org> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <199704031317.FAA21733@freefall.freebsd.org>; from Darren Reed on Apr 3, 1997 23:12:04 +1000 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As Darren Reed wrote: > It has been suggested that IP Filter disallow changes to filter rules if > securelevel is set to some level...(I think 3 was the suggestion). I personally think securelevel 2 would be sufficient. It blocks already enough things, like running an Xserver :). But the most important is that you make this consistent throughout all BSDs, including BSD/OS, if possible. -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Thu Apr 3 13:58:54 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA19149 for hackers-outgoing; Thu, 3 Apr 1997 13:58:54 -0800 (PST) Received: from ravenock.cybercity.dk (ravenock.cybercity.dk [194.16.57.32]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA19144 for ; Thu, 3 Apr 1997 13:58:49 -0800 (PST) Received: (from sos@localhost) by ravenock.cybercity.dk (8.8.5/8.7.3) id XAA01245; Thu, 3 Apr 1997 23:58:31 +0200 (MET DST) From: Søren Schmidt Message-Id: <199704032158.XAA01245@ravenock.cybercity.dk> Subject: Re: CD format for backups ? In-Reply-To: <19970403224454.SP47619@uriah.heep.sax.de> from J Wunsch at "Apr 3, 97 10:44:54 pm" To: joerg_wunsch@uriah.heep.sax.de Date: Thu, 3 Apr 1997 23:58:31 +0200 (MET DST) Cc: hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL30 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In reply to J Wunsch who wrote: > As Luigi Rizzo wrote: > > > - use the partition (/dev/wd0f or /dev/rwd0f ?) as the source for the > > CD data > > > > then in order to read the data I would only want to mount /dev/wcd0c as > > FFS instead of cd9660 -- any hope that it works ? > > Well, UFS is currently still a little iffy when it comes to devices > that are not using 512 bytes blocksize. So, ``it depends''. Well I use UFS on 1024byte sector disks, and I know it also works on 2048 byte sectors, so... -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Søren Schmidt (sos@FreeBSD.org) FreeBSD Core Team Even more code to hack -- will it ever end .. From owner-freebsd-hackers Thu Apr 3 14:12:43 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA19739 for hackers-outgoing; Thu, 3 Apr 1997 14:12:43 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id OAA19734 for ; Thu, 3 Apr 1997 14:12:39 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id OAA17387; Thu, 3 Apr 1997 14:55:48 -0700 From: Terry Lambert Message-Id: <199704032155.OAA17387@phaeton.artisoft.com> Subject: Re: CD format for backups ? To: joerg_wunsch@uriah.heep.sax.de Date: Thu, 3 Apr 1997 14:55:48 -0700 (MST) Cc: hackers@freebsd.org In-Reply-To: <19970403224454.SP47619@uriah.heep.sax.de> from "J Wunsch" at Apr 3, 97 10:44:54 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > - use the partition (/dev/wd0f or /dev/rwd0f ?) as the source for the > > CD data > > > > then in order to read the data I would only want to mount /dev/wcd0c as > > FFS instead of cd9660 -- any hope that it works ? > > Well, UFS is currently still a little iffy when it comes to devices > that are not using 512 bytes blocksize. So, ``it depends''. You should be able to pick a block size of 16k (with a frag size of 1k) or a smaller block size, as long as the frag size does not go below 1k. There is a small issue on directory modifications for a directory block size smaller than the physical device block size (this is really a bug in /sys/ufs/ufs/ufs_dir.c), but it should not bite you as long as you are dealing with read-only media, where insertions are unlikely. 8-). Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Thu Apr 3 14:17:31 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA20150 for hackers-outgoing; Thu, 3 Apr 1997 14:17:31 -0800 (PST) Received: from glacier.wise.edt.ericsson.se (glacier-ext.wise.edt.ericsson.se [193.180.251.38]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA20143 for ; Thu, 3 Apr 1997 14:17:25 -0800 (PST) Received: from mailhost.epa.ericsson.se (epa.epa.ericsson.se [146.11.8.1]) by glacier.wise.edt.ericsson.se (8.7.5/8.7.3/glacier-0.9) with ESMTP id AAA17670; Fri, 4 Apr 1997 00:17:09 +0200 (MET DST) Received: from brpc795.epa.ericsson.se (brpc795.epa.ericsson.se [146.11.13.2]) by mailhost.epa.ericsson.se (8.8.5/8.8.5) with SMTP id IAA18349; Fri, 4 Apr 1997 08:17:05 +1000 (EST) Message-ID: <33453785.273B@epa.ericsson.se> Date: Fri, 04 Apr 1997 09:16:53 -0800 From: Mark Hannon Organization: Ericsson Australia X-Mailer: Mozilla 3.01Gold (Win16; I) MIME-Version: 1.0 To: hackers@freebsd.org CC: "Hannon, Mark (seeware)" Subject: Locale problems with CDE/FreeBSD !Linux References: <199704032101.HAA01341@putte.seeware.DIALix.oz.au> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi all, I have been having some more problems with the Xinside CDE implementation. As is their practice, the technical support department have been unresponsive. I received a reply this morning from Thomas Roell that they were looking at my problem but didn't think they could fix it. So, maybe someone on the lists can give me some clues.... FreeBSD-2.2-RELEASE (with sgid patch to get dtmail working) | FreeBSD-2.1.7 Locale set from dtlogin is en_AU | sv_SE. Swedish keyboard installed. LANG is set to en_AU by the login program. My basic problem is that I cannot enter swedish characters ÅÅÖ in the dtmail compose window. The same problem in the dtpad editor. I can however read emails with these characters & can even paste the characters into dtmail. The dtterm terminal emulator running tcsh allows me to enter the characters. Xemacs allows their input. The problem is the same under both the swedish (sv_SE) and english (en_AU) locales. If I try the default 'C' locale then dtmail does not even printout the swedish characters in received mails. I even removed my FreeBSD partition a couple of weekends ago, installed a Redhat Linux distribution and CDE on top of it. In that configuration it was possible to input the swedish characters! It seemed like a rather drastic solution to muck around recompiling/reconfiging my machines to Linux so I reinstalled FreeBSD. Is anybody using Xig CDE and a non-ascii locale? Do you have any problems? Are there many CDE users out there? It seems that Xig are concentrating on the Linux market. I have seen a number of problems in CDE which seem to indicate very little testing on FreeBSD (dtaction core dumps, dthelpview core dumps, programs won't start from front panel, dtmail core dumps, dtappintegrate doesn't work etc.) Are there any other vendors thinking about a CDE port to FreeBSD? Can anybody give me any clues where to look further? Regards/Mark > > From mark Sun Feb 9 16:44:44 1997 > > To: support@xinside.com > > Subject: More language problems with CDE > > Mime-Version: 1.0 > > Content-Type: text/plain; charset=C.us-ascii > > Content-Transfer-Encoding: quoted-printable > > Content-MD5: KUuqh1i7l2IBAtyKTB5pQg== > > Content-Length: 1138 > > > > > > Hi again, > > > > Running FreeBSD-2.2, and trying to use Swedish fonts & characters > > I am having some problems. > > > > I can type the Swedish national characters from dtterm but not > > from dtmail. dtmail also gives the following errors on startup: > > > > putte: {2} dtmail > > Warning: Cannot convert string > > "-dt-application-medium-r-normal--14-*-*-*-*-*-dtsymbol-*" to type > > FontStruct > > Warning: Cannot convert string "....Missing message #104" to type Short > > Warning: Cannot convert string "Ö" to type KeySym > > Warning: Cannot convert string "Ö" to type KeySym > > Warning: Cannot convert string "Å" to type KeySym > > Warning: Cannot convert string "ä" to type KeySym > > _DtXlateGetXlateEnv: Platform: FreeBSD; Execution Ver: 0; Compiled Ver: > > 210 > > Warning: Cannot convert string "....Missing message #103" to type Short > > Warning: Cannot convert string "Å" to type KeySym > > Warning: Cannot convert string "....Missing message #102" to type KeySym > > Warning: Cannot convert string "Å" to type KeySym > > Warning: Cannot convert string "Å" to type KeySym > > Warning: Cannot convert string "Å" to type KeySym > > > > I am using the Swedish locale. > > > > Any ideas? > > > > Regards/Mark From owner-freebsd-hackers Thu Apr 3 14:18:13 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA20206 for hackers-outgoing; Thu, 3 Apr 1997 14:18:13 -0800 (PST) Received: from news.IAEhv.nl (root@news.IAEhv.nl [194.151.64.4]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id OAA20201 for ; Thu, 3 Apr 1997 14:18:10 -0800 (PST) Received: from LOCAL (uucp@localhost) by news.IAEhv.nl (8.6.13/1.63) with IAEhv.nl; pid 8112 on Fri, 4 Apr 1997 00:18:00 +0200; id AAA08112 efrom: devet@adv.IAEhv.nl; eto: UNKNOWN Received: by adv.IAEhv.nl (8.7.5/1.63) id VAA10059; Thu, 3 Apr 1997 21:10:27 +0200 (MET DST) Date: Thu, 3 Apr 1997 21:10:27 +0200 (MET DST) From: Arjan.deVet@adv.IAEhv.nl (Arjan de Vet) Message-Id: <199704031910.VAA10059@adv.IAEhv.nl> To: bakul@torrentnet.com Subject: Re: apache like preforking apps and high loads X-Newsgroups: list.freebsd.hackers In-Reply-To: <199704031653.LAA20772@chai.plexuscom.com> References: <199704031527.HAA03657@root.com> Organization: Internet Access Eindhoven, the Netherlands Cc: freebsd-hackers@freebsd.org Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In article <199704031653.LAA20772@chai.plexuscom.com> you write: >To guido: For apache like apps one idea is to have one process be >the acceptor and have it pass a new socket to individual server >processes (including doing some initial processing to determine >which process should get this socket if not all server processes are >equivalent). That was one of the first approaches NCSA used for preforked HTTP daemons: one parent process which does an accept() and which passes the filedescriptor to one of the child processes. I don't know about more recent versions of NCSA but Apache now uses one parent process which only watches the number of free child processes. The child processes all block on accept(), waiting for a connection to arrive. Arjan From owner-freebsd-hackers Thu Apr 3 15:44:42 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA26879 for hackers-outgoing; Thu, 3 Apr 1997 15:44:42 -0800 (PST) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA26874 for ; Thu, 3 Apr 1997 15:44:38 -0800 (PST) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by who.cdrom.com (8.8.5/8.6.11) with ESMTP id PAA18895 for ; Thu, 3 Apr 1997 15:44:36 -0800 (PST) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.5/8.7.3) id JAA15356; Fri, 4 Apr 1997 09:12:14 +0930 (CST) From: Michael Smith Message-Id: <199704032342.JAA15356@genesis.atrad.adelaide.edu.au> Subject: Re: Does de driver do 100MBIT Full Duplex? In-Reply-To: <3.0.32.19970403115910.00b89450@etinc.com> from dennis at "Apr 3, 97 11:59:12 am" To: dennis@etinc.com (dennis) Date: Fri, 4 Apr 1997 09:12:14 +0930 (CST) Cc: narvi@haldjas.folklore.ee, joerg_wunsch@uriah.heep.sax.de, hackers@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk dennis stands accused of saying: > > What you have now is a release that was supposed to be a "great > saviour" feature-wise that is fundamentally unusable in its released > form for a large number of users..... Because, as has been observed beforehand, this "large number" of users were either too timid, too indifferent, or too _stupid_ to participate in the testing process. I sure as hell don't recall any observations on your part on 2.2 during the prerelease testing on this or a number of other points you've subsequently griped about; I would suggest that you could probably have justified spending just a few hours a couple of times installing and testing the various fully-bundled prerelease snapshots and feeding back on your woes. If you, and a few other serious, "production" users had just _thought_ahead_ a little, 2.2 would have been less of a debacle than it has currently turned out to be. Without that sort of input _before_ the final release is rolled, it's _not_ going to be possible to make you happy. Ever. And that would be Bad. > Dennis -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ From owner-freebsd-hackers Thu Apr 3 15:53:24 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA27560 for hackers-outgoing; Thu, 3 Apr 1997 15:53:24 -0800 (PST) Received: from root.com (implode.root.com [198.145.90.17]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA27552 for ; Thu, 3 Apr 1997 15:53:22 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by root.com (8.8.5/8.6.5) with SMTP id PAA05634; Thu, 3 Apr 1997 15:54:17 -0800 (PST) Message-Id: <199704032354.PAA05634@root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: Bakul Shah cc: Guido van Rooij , FreeBSD-hackers@freebsd.org (FreeBSD-hackers) Subject: Re: apache like preforking apps and high loads In-reply-to: Your message of "Thu, 03 Apr 1997 11:53:34 EST." <199704031653.LAA20772@chai.plexuscom.com> From: David Greenman Reply-To: dg@root.com Date: Thu, 03 Apr 1997 15:54:17 -0800 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >> >When looking at Apacha like applications, one often sees extermely high >> >load averages. Apache preforks a number of processes that all block >> >on accept(). When a request comes in, all process are woken up and the >> >scheduler chooses one of the now runnable processes taht will succeed in >> >the accept(). The other go back to sleep. >> >> Not any more. I changed this a few days ago. Now only one process is >> woken up. > >Fairness is probably not an issue when an app. consists of a number >of anonymous servers but in general one would want to make sure that >if N processes are waiting on accept() on the same socket, no one >process is starved of accepting. How do you ensure that? The processes blocked on accept are handled in a round-robin fashion, oldest first. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-hackers Thu Apr 3 15:59:19 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA27759 for hackers-outgoing; Thu, 3 Apr 1997 15:59:19 -0800 (PST) Received: from root.com (implode.root.com [198.145.90.17]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA27752 for ; Thu, 3 Apr 1997 15:59:16 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by root.com (8.8.5/8.6.5) with SMTP id QAA05674; Thu, 3 Apr 1997 16:00:09 -0800 (PST) Message-Id: <199704040000.QAA05674@root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: Marc Slemko cc: FreeBSD-hackers Subject: Re: apache like preforking apps and high loads In-reply-to: Your message of "Thu, 03 Apr 1997 13:34:29 MST." From: David Greenman Reply-To: dg@root.com Date: Thu, 03 Apr 1997 16:00:09 -0800 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >On Thu, 3 Apr 1997, David Greenman wrote: > >> >When looking at Apacha like applications, one often sees extermely high >> >load averages. Apache preforks a number of processes that all block >> >on accept(). When a request comes in, all process are woken up and the >> >scheduler chooses one of the now runnable processes taht will succeed in >> >the accept(). The other go back to sleep. >> >> Not any more. I changed this a few days ago. Now only one process is >> woken up. > >Looking at these changes and the other code around them, it appears to me >like the net effect is that processes are added to the tail of the queue >when they block on accept() and removed from (near) the head with >wakeup_one. Am I reading this right, or am I getting my queue ends mixed >up somewhere? You have it right, and that is the way it needs to work. I don't know how Apache does it's # processes load balancing, but fairness can be important in general. >For code that does similar things to what Apache does, it would be far >better to treat it as a stack and have wakeup_one() remove from the tail >since you want the same processes to serve as many requests as possible. >In the grand scheme of things it may be a bad idea to have wakeup_one() >behave this way, but it looks like it would be useful in this case. On the x86 and most other architectures, the cache is of physical memory and the majority of the Apache code is in shared pages. There isn't any significant difference in terms of cache effects whether you wake the same process all the time or you wake different processes. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-hackers Thu Apr 3 16:28:42 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA29995 for hackers-outgoing; Thu, 3 Apr 1997 16:28:42 -0800 (PST) Received: (from mpp@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA29988; Thu, 3 Apr 1997 16:28:37 -0800 (PST) From: Mike Pritchard Message-Id: <199704040028.QAA29988@freefall.freebsd.org> Subject: Re: Unreasonable rebuild time on large password files To: joerg_wunsch@uriah.heep.sax.de Date: Thu, 3 Apr 1997 16:28:37 -0800 (PST) Cc: freebsd-hackers@freebsd.org In-Reply-To: <19970403082103.GT06676@uriah.heep.sax.de> from "J Wunsch" at Apr 3, 97 08:21:03 am X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk J Wunsch wrote: > > As Brian Tao wrote: > > > Ah, geez, that's right... I even remember pestering him about it a > > long time ago. :) So the patches have been integrated? I didn't see > > any mention of them in the man pages, so I presumed it was still > > necessary to patch it separately. How do you enable the feature? > > Guido doesn't seem to be very well in writing man pages. :-| The -u > option to pwd_mkdb(8) ain't mentioned at all, although it's being > used _by default_ if only a single user's entry is updated (passwd(1), > or chpass(1), as opposed to vipw(8)). What? Someone added an option to a command and didn't document it? Consider this a first warning from the man page police :-). The next violation may require community service in the form if me banging on your door until you cough up the appropriate diff :-). At least it wasn't the more serious crime of adding a new command without a man page. For that one I may forward your e-mail address to all of the stupid junk e-mail soliciters who seem to have recently found my address :-). -- Mike Pritchard mpp@FreeBSD.org "Go that way. Really fast. If something gets in your way, turn" From owner-freebsd-hackers Thu Apr 3 18:16:14 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id SAA07090 for hackers-outgoing; Thu, 3 Apr 1997 18:16:14 -0800 (PST) Received: from pdx1.world.net (pdx1.world.net [192.243.32.18]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA07077 for ; Thu, 3 Apr 1997 18:16:11 -0800 (PST) From: proff@suburbia.net Received: from suburbia.net (suburbia.net [203.4.184.1]) by pdx1.world.net (8.7.5/8.7.3) with SMTP id SAA20679 for ; Thu, 3 Apr 1997 18:18:15 -0800 (PST) Received: (qmail 14227 invoked by uid 110); 4 Apr 1997 01:17:35 -0000 Message-ID: <19970404011735.14225.qmail@suburbia.net> Subject: Re: IP Filter in -current. In-Reply-To: <199704031050.UAA18193@plum.cyber.com.au> from Darren Reed at "Apr 3, 97 08:50:16 pm" To: darrenr@cyber.com.au (Darren Reed) Date: Fri, 4 Apr 1997 11:17:35 +1000 (EST) Cc: hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > I've imported IP Filter 3.2alpha4 into the -current sources for FreeBSD > just now. This latest update should successfully build a LKM for use > with FreeBSD 2.2 or later(?). It DOESN'T support being built into the > kernel itself. > > Darren The module/userland code won't compile under -current. I've ported it however. major issues were: #include brain-damange. Bruce's include file philosophy continues to make FreeBSD progressivly more compile time incompatible with everything else, including other BSD's. Use of TAILQ's for handling ifaddr chains. VOP_LOCK() in lite2 takes three arguments. At run-time it seems to panic quickly if logging is going on and ipmon isn't there to read the log device. Cheers, Julian. From owner-freebsd-hackers Thu Apr 3 18:43:58 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id SAA09308 for hackers-outgoing; Thu, 3 Apr 1997 18:43:58 -0800 (PST) Received: from zen.nash.org (nash.pr.mcs.net [204.95.47.72]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA09284 for ; Thu, 3 Apr 1997 18:43:50 -0800 (PST) Received: from zen.nash.org (localhost [127.0.0.1]) by zen.nash.org (8.8.5/8.6.12) with SMTP id UAA00446; Thu, 3 Apr 1997 20:42:57 -0600 (CST) Message-ID: <33446AB1.41C67EA6@mcs.com> Date: Thu, 03 Apr 1997 20:42:57 -0600 From: Alex Nash X-Mailer: Mozilla 3.01Gold (X11; I; FreeBSD 2.2-RELEASE i386) MIME-Version: 1.0 To: Joerg Wunsch CC: hackers@freebsd.org, avalon@coombs.anu.edu.au Subject: Re: securelevel & IP filter References: <199704031317.FAA21733@freefall.freebsd.org> <19970403233738.KY42145@uriah.heep.sax.de> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk J Wunsch wrote: > > As Darren Reed wrote: > > > It has been suggested that IP Filter disallow changes to filter rules if > > securelevel is set to some level...(I think 3 was the suggestion). > > I personally think securelevel 2 would be sufficient. It blocks > already enough things, like running an Xserver :). > > But the most important is that you make this consistent throughout all > BSDs, including BSD/OS, if possible. There's some (albeit arbitrary) precedence for using 3 already in ipfw. The main reason 2 was avoided was principle of least surprise. Alex From owner-freebsd-hackers Thu Apr 3 19:20:47 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id TAA11203 for hackers-outgoing; Thu, 3 Apr 1997 19:20:47 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id TAA11190 for ; Thu, 3 Apr 1997 19:20:41 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA15400; Thu, 3 Apr 1997 22:20:05 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Thu, 3 Apr 1997 22:20 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id WAA15955; Thu, 3 Apr 1997 22:04:49 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id WAA09093; Thu, 3 Apr 1997 22:10:40 -0500 (EST) Date: Thu, 3 Apr 1997 22:10:40 -0500 (EST) From: Thomas David Rivers Message-Id: <199704040310.WAA09093@lakes.water.net> To: ponds!freefall.cdrom.com!freebsd-hackers, ponds!dg.com!root, ponds!idi.ntnu.no!Tor.Egge Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Wow! I was _very_ excited by this report; especially since it seemed to match the 'dup alloc' panic I've reported, and it was a timing issue, and this is certainly the area where I get "dup alloc"s and "free vnode isn't" panics... So, I applied the following diff to a 2.1.7 tree, which basically moves the call to VOP_INACTIVE() until after the vnode has been added to the free list (as Tor had described) [Tor - does this seem to be what you intended? The source in 2.1.7, of course, doesn't have the SMP locking, etc....]: *** vfs_subr.c.ori Thu Aug 15 13:08:20 1996 --- vfs_subr.c Thu Apr 3 20:07:46 1997 *************** *** 835,840 **** --- 835,843 ---- panic("vrele: ref cnt"); } #endif + + VOP_INACTIVE(vp); + if (vp->v_flag & VAGE) { TAILQ_INSERT_HEAD(&vnode_free_list, vp, v_freelist); vp->v_flag &= ~VAGE; *************** *** 843,849 **** } freevnodes++; - VOP_INACTIVE(vp); } /* --- 846,851 ---- Unfortunately, I am still able to demonstrate my particular problem :-( So, I'm beginning to wonder... is there a way a vnode can get onto the freelist; but still be "busy" somehow... Of hand, is there a way to check for this condition I could add to the code to see if that's what's going on? - Thanks - - Dave Rivers - From owner-freebsd-hackers Thu Apr 3 20:09:12 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA13070 for hackers-outgoing; Thu, 3 Apr 1997 20:09:12 -0800 (PST) Received: from hamby1 (hamby1.lightside.net [207.67.176.17]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id UAA13065 for ; Thu, 3 Apr 1997 20:09:07 -0800 (PST) Received: (from jehamby@localhost) by hamby1 (SMI-8.6/SMI-SVR4) id UAA00644 for hackers@freebsd.org; Thu, 3 Apr 1997 20:09:25 -0800 Date: Thu, 3 Apr 1997 20:09:25 -0800 From: jehamby@lightside.com (Jake Hamby) Message-Id: <199704040409.UAA00644@hamby1> To: hackers@freebsd.org Subject: Embedding Motif lib inside another lib (licensing issues?) Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-MD5: Vz+LKKBtJGZc6PcBE4muIA== Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk This question is a bit premature to ask, because the program I'm working on is still in the early stages, but I'm curious to the answer anyway. I'm writing Yet Another GUI API for X, but the difference is that this one is highly multithreaded and based on C++. The initial prototype will be on Solaris, simply because it seems to have the most mature threads library (and threadsafe libraries) of all the UNIXs. Afterwards, I'm porting to FreeBSD and Linux. Most of the actual GUI functions will probably map to Xlib calls (for font drawing and colormap handling, for example). But for buttons, menus, and other widgets, I thought the simplest approach would be to map them to Motif calls. Please, no comments about Motif sucking, I'm familiar with it, it does the job, and with CDE, it's cemented a place as the standard UNIX look-and-feel. On Solaris, Motif is bundled, but of course, on Linux and FreeBSD one must pay extra for it. Suppose I pay $150 for a copy of Motif 2.0 for FreeBSD (I already have an old MetroLink Motif license for Linux). Is there any way for me to embed the Motif library itself statically INSIDE of my graphics library legally? Of course, I won't be shipping any Motif header files, and none of the Xm functions themselves will be exposed to the user. I remember seeing a toolkit several years ago that was actually a server process that clients would connect through via a TCP connection, and the server would handle all of the Motif commands on behalf of the client. Because the server was a standalone Motif program, linked statically, and the clients only communicated to it, this is perfectly legal. But with my idea, where the clients dynamically link to my shared library, which embeds Motif, this sounds legally suspect? Anyone care to comment? If this is not kosher, then rather than give up Motif, I'll probably just link with LessTif instead. The promise of good CDE integration on the Solaris platform is just too useful to pass up Motif altogether. Cheers, Jake From owner-freebsd-hackers Thu Apr 3 20:09:54 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA13138 for hackers-outgoing; Thu, 3 Apr 1997 20:09:54 -0800 (PST) Received: from root.com (implode.root.com [198.145.90.17]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA13133 for ; Thu, 3 Apr 1997 20:09:50 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by root.com (8.8.5/8.6.5) with SMTP id UAA08780; Thu, 3 Apr 1997 20:10:45 -0800 (PST) Message-Id: <199704040410.UAA08780@root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: Thomas David Rivers cc: freebsd-hackers@freebsd.org, Tor.Egge@idi.ntnu.no Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) In-reply-to: Your message of "Thu, 03 Apr 1997 22:10:40 EST." <199704040310.WAA09093@lakes.water.net> From: David Greenman Reply-To: dg@root.com Date: Thu, 03 Apr 1997 20:10:45 -0800 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk You know, your mailer does absolutely disgusting things to the To: and cc: lines of messages. I either have to delete the stuff or rewrite it to something RFC 822 compliant. This last message would have tried to send replies to ucbvax.berkeley.edu...a machine which doesn't even exist anymore. >Wow! > > I was _very_ excited by this report; especially since it seemed >to match the 'dup alloc' panic I've reported, and it was a timing >issue, and this is certainly the area where I get "dup alloc"s and >"free vnode isn't" panics... > > So, I applied the following diff to a 2.1.7 tree, which basically >moves the call to VOP_INACTIVE() until after the vnode has been >added to the free list (as Tor had described) [Tor - does this seem >to be what you intended? The source in 2.1.7, of course, doesn't >have the SMP locking, etc....]: > >*** vfs_subr.c.ori Thu Aug 15 13:08:20 1996 >--- vfs_subr.c Thu Apr 3 20:07:46 1997 >*************** >*** 835,840 **** >--- 835,843 ---- > panic("vrele: ref cnt"); > } > #endif >+ >+ VOP_INACTIVE(vp); >+ > if (vp->v_flag & VAGE) { > TAILQ_INSERT_HEAD(&vnode_free_list, vp, v_freelist); > vp->v_flag &= ~VAGE; >*************** >*** 843,849 **** > } > freevnodes++; > >- VOP_INACTIVE(vp); > } > > /* >--- 846,851 ---- Uh, this is wrong since VOP_INACTIVE really wants a '0' usecount vnode, and there are assumptions throughout the code that a '0' usecount also implies that the vnode is on the free list. A quick code review of Tor's suggested fix shows that it will fail in several places in the kernel and basically needs to be re-thought...which is why it hasn't been committed yet. >Unfortunately, I am still able to demonstrate my particular problem :-( Not a surpise to me, actually. You problem appears not to have anything to do with vnodes...it appears to be a buffer problem (or FFS allocation bug) of some kind. >So, I'm beginning to wonder... is there a way a vnode can get >onto the freelist; but still be "busy" somehow... Of hand, >is there a way to check for this condition I could add to the code >to see if that's what's going on? No, a vnode on the free list is always "free" - it will be removed from the free list prior to any I/O occuring. The only caveat to this is that entries in the namei cache can point to "free" vnodes and there is some kludgewear to deal with this. There is already code in the kernel to check for "free vnode isn't" conditions where a vnode usecount is non-zero while the vnode is on the free list...but I'm not aware of any bugs related to this in the current code. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-hackers Thu Apr 3 21:58:43 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id VAA18633 for hackers-outgoing; Thu, 3 Apr 1997 21:58:43 -0800 (PST) Received: from aries.bb.cc.wa.us (root@[208.8.136.11]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id VAA18628 for ; Thu, 3 Apr 1997 21:58:39 -0800 (PST) Received: from localhost (chris@localhost) by aries.bb.cc.wa.us (8.8.3/8.6.9) with SMTP id VAA25028; Thu, 3 Apr 1997 21:54:12 -0800 (PST) Date: Thu, 3 Apr 1997 21:54:12 -0800 (PST) From: Chris Coleman To: Mike Pritchard cc: joerg_wunsch@uriah.heep.sax.de, freebsd-hackers@FreeBSD.ORG Subject: Re: Unreasonable rebuild time on large password files In-Reply-To: <199704040028.QAA29988@freefall.freebsd.org> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Thu, 3 Apr 1997, Mike Pritchard wrote: > What? Someone added an option to a command and didn't document it? > Consider this a first warning from the man page police :-). > -- > Mike Pritchard > mpp@FreeBSD.org ^^^ Is that what mpp stands for "Man Page Police" ? :-) > Christopher J. Coleman (chris@aries.bb.cc.wa.us) Computer Support Technician I (509)-766-8873 Big Bend Community College Internet Instructor FreeBSD Book Project: http://vinyl.quickweb.com/~chrisc/book.html Disclaimer: Even Though it has My Name on it, Doesn't mean I said it. From owner-freebsd-hackers Thu Apr 3 23:36:49 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA23873 for hackers-outgoing; Thu, 3 Apr 1997 23:36:49 -0800 (PST) Received: from plum.cyber.com.au (plum.cyber.com.au [203.7.155.24]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id XAA23866 for ; Thu, 3 Apr 1997 23:36:45 -0800 (PST) Received: (from darrenr@localhost) by plum.cyber.com.au (8.6.12/8.6.6) id RAA23038; Fri, 4 Apr 1997 17:36:09 +1000 From: Darren Reed Message-Id: <199704040736.RAA23038@plum.cyber.com.au> Subject: Re: IP Filter in -current. To: proff@suburbia.net Date: Fri, 4 Apr 1997 17:36:09 +1000 (EST) Cc: darrenr@cyber.com.au, hackers@freebsd.org In-Reply-To: <19970404011735.14225.qmail@suburbia.net> from "proff@suburbia.net" at Apr 4, 97 11:17:35 am X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In some mail I received from proff@suburbia.net, sie wrote > > The module/userland code won't compile under -current. I've ported it > however. major issues were: > > #include brain-damange. Bruce's include file philosophy > continues to make FreeBSD progressivly more compile time > incompatible with everything else, including other BSD's. Someone should give this guy a labotamy if he's the one who came up with the "include every other include file in this include file" idea. > Use of TAILQ's for handling ifaddr chains. Is this consistant with NetBSD ? > VOP_LOCK() in lite2 takes three arguments. 3 instead of 4 or 3 instead of 2 ? > At run-time it seems to panic quickly if logging is going on and > ipmon isn't there to read the log device. heh...I guess you'll be able to track this bug then :) From owner-freebsd-hackers Thu Apr 3 23:49:18 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA24686 for hackers-outgoing; Thu, 3 Apr 1997 23:49:18 -0800 (PST) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA24680 for ; Thu, 3 Apr 1997 23:49:16 -0800 (PST) Received: from mail.cdsnet.net (mail.cdsnet.net [204.118.244.5]) by mail.cdsnet.net (8.8.5/8.7.3) with SMTP id XAA14133 for ; Thu, 3 Apr 1997 23:49:15 -0800 (PST) Date: Thu, 3 Apr 1997 23:49:15 -0800 (PST) From: Jaye Mathisen To: hackers@freebsd.org Subject: MUSBUS? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Anybody have a copy of the old musbus benchmark? I think musbus stood for "Monash University Software for Benchmarking Unix Systems". I was thinking it might make a reasonable start for a "FreeBSD system thrasher". I would like to be able to take a new box, pop FreeBSD on it, then run some package that thrashes the daylights out of the darn thing, for a time period that I specify. Currently, the only thing I can think of that really bashes on it is the classic make world, coupled with news. Solaris has something like that in their "Hardware Certification Toolset", or whatever the heck that things called. From owner-freebsd-hackers Fri Apr 4 00:05:33 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA25261 for hackers-outgoing; Fri, 4 Apr 1997 00:05:33 -0800 (PST) Received: from sunrise.pg.gda.pl (sunrise.pg.gda.pl [153.19.40.230]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA25246 for ; Fri, 4 Apr 1997 00:05:07 -0800 (PST) Received: from localhost (libja@localhost [127.0.0.1]) by sunrise.pg.gda.pl (8.8.5/8.8.5) with SMTP id KAA19461 for ; Fri, 4 Apr 1997 10:03:54 +0200 (MET DST) Date: Fri, 4 Apr 1997 10:03:53 +0200 (MET DST) From: Jaros/law Libert X-Sender: libja@sunrise To: freebsd-hackers@FreeBSD.org Subject: subscribe Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk From owner-freebsd-hackers Fri Apr 4 00:29:04 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA26163 for hackers-outgoing; Fri, 4 Apr 1997 00:29:04 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA26158 for ; Fri, 4 Apr 1997 00:29:02 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id AAA19005; Fri, 4 Apr 1997 00:23:35 -0800 (PST) To: Darren Reed cc: proff@suburbia.net, hackers@freebsd.org Subject: Re: IP Filter in -current. In-reply-to: Your message of "Fri, 04 Apr 1997 17:36:09 +1000." <199704040736.RAA23038@plum.cyber.com.au> Date: Fri, 04 Apr 1997 00:23:34 -0800 Message-ID: <18999.860142214@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Someone should give this guy a labotamy if he's the one who came up > with the "include every other include file in this include file" idea. "Labotomy", and I think Bruce is too valuable to even consider such a radical treatment for him - how about trying direct debate with him over this issue instead? ;-) Jordan From owner-freebsd-hackers Fri Apr 4 00:59:46 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id AAA27177 for hackers-outgoing; Fri, 4 Apr 1997 00:59:46 -0800 (PST) Received: from gluon.mep.ruhr-uni-bochum.de (gluon.mep.ruhr-uni-bochum.de [134.147.160.165]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id AAA27169 for ; Fri, 4 Apr 1997 00:59:43 -0800 (PST) Received: (from roberte@localhost) by gluon.mep.ruhr-uni-bochum.de (8.6.12/8.6.12) id KAA10630 for hackers@freebsd.org; Fri, 4 Apr 1997 10:59:41 +0200 From: Robert Eckardt Message-Id: <199704040859.KAA10630@gluon.mep.ruhr-uni-bochum.de> Subject: ipfw configuration To: hackers@freebsd.org Date: Fri, 4 Apr 1997 10:59:41 +0200 (MET DST) X-Mailer: ELM [version 2.4ME+ PL25 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi, is there any configuration tool to generate ipfw rules from a simple policy matrix e.g. OUT->IN OUT->FW IN->OUT IN->FW FW->OUT FW->IN telnet|tcp deny deny allow allow allow allow ftp|tcp allow allow from allow deny allow allow a.b.c.d (including multiple adresses for FW) o If yes, where can I get it ? o If not, I would need to write one (prob. Tcl/Tk based). Any comments, recommendations, ... ? Robert -- Dr. Robert Eckardt ( Ruhr-Universitaet Bochum, Inst.f.Theor.Physik, NB6/169 ) Universitaetsstrasse 150, D-44780 Bochum, Germany ----X---8---- Telefon: +49 234 700-3709, Telefax: +49 234 7094-574 8 E-Mail: RobertE@MEP.Ruhr-Uni-Bochum.de --------8---- URL: http://WWW.MEP.Ruhr-Uni-Bochum.de/~roberte >>> To be successful one needs friends, <<< >>> To be very successful one needs enemies. <<< From owner-freebsd-hackers Fri Apr 4 01:00:56 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA27281 for hackers-outgoing; Fri, 4 Apr 1997 01:00:56 -0800 (PST) Received: from florence.pavilion.net (florence.pavilion.net [194.242.128.25]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA27275 for ; Fri, 4 Apr 1997 01:00:51 -0800 (PST) Received: (from joe@localhost) by florence.pavilion.net (8.8.5/8.8.5) id KAA28456; Fri, 4 Apr 1997 10:00:09 +0100 (BST) Message-ID: <19970404100009.41395@pavilion.net> Date: Fri, 4 Apr 1997 10:00:09 +0100 From: Josef Karthauser To: Joerg Wunsch Cc: hackers@freebsd.org Subject: Re: Internal clock References: <3340C326.6150@cybernet.com> <19970401151947.61221@pavilion.net> <19970401203429.XR37988@uriah.heep.sax.de> <19970402123111.58518@pavilion.net> <19970402203510.RR52594@uriah.heep.sax.de> <19970403094030.30623@pavilion.net> <19970403221849.DL32306@uriah.heep.sax.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.64 In-Reply-To: <19970403221849.DL32306@uriah.heep.sax.de>; from J Wunsch on Thu, Apr 03, 1997 at 10:18:49PM +0200 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Thu, Apr 03, 1997 at 10:18:49PM +0200, J Wunsch wrote: > As Josef Karthauser wrote: > > > Would you mind added the units to this man page. i.e. is it 100Hz, > > or 100KHz, or 100 camels? ;) > > Since it's a rate, i think 100 camels per second. :-) Ooops, so it is. I've been wondering why my code doesn't work! ;) Joe -- Josef Karthauser Technical Manager Email: joe@pavilion.net Pavilion Internet plc. [Tel: +44 1273 607072 Fax: +44 1273 607073] From owner-freebsd-hackers Fri Apr 4 01:06:08 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA27475 for hackers-outgoing; Fri, 4 Apr 1997 01:06:08 -0800 (PST) Received: from robin.camelot.de (root@robin.camelot.de [194.97.87.3]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id BAA27462 for ; Fri, 4 Apr 1997 01:06:01 -0800 (PST) Received: by robin.camelot.de via sendmail with stdio id for freebsd-hackers@freebsd.org; Fri, 4 Apr 1997 11:05:53 +0200 (MET DST) (Smail-3.2 1996-Jul-4 #1 built DST-Jul-28) Message-Id: From: knarf@camelot.de (Frank Bartels) Subject: Re: kern/2923: panic: vm_fault: fault on nofault entry, addr: f6e21000 In-Reply-To: <199703091138.GAA19040@lakes.water.net> from Thomas David Rivers at "Mar 9, 97 06:38:17 am" To: ponds!rivers@dg-rtp.dg.com (Thomas David Rivers) Date: Fri, 4 Apr 1997 11:05:53 +0200 (MET DST) Cc: freebsd-hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL29 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hello, Thomas David Rivers wrote: > I believe it to be a timing problem in the kernel. A heavy load just > gives you a greater opportunity (as a percentage of time) to trip over > the problem. Again; this also happens on my extremely lightly loaded > (i.e. only a shell) system I put together just to reproduce this problem. I have the same problem here on a medium loaded newsserver (6 GB newsspool). I get "panic: vm_fault: fault on nofault entry, addr: f6718000" with different values for addr. I updated from 2.1.5 to 2.2.1 about 48 hours before I've seen this problem the first time. I really have no idea what to do now. The machine always reboots some minutes after the crash. Running fsck on 10 GB disks all the time is no fun... :/ If you have any idea how I can make the uptime 6 hrs or higher, please tell me! Thank you, Knarf -- Frank Bartels |UUCP/ZModem/Fax: +49 89 8948040| "Captain, why not just knarf@camelot.de | http://www.camelot.de/~knarf/ | give the Borg Windows?" From owner-freebsd-hackers Fri Apr 4 01:21:43 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA28331 for hackers-outgoing; Fri, 4 Apr 1997 01:21:43 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA28325 for ; Fri, 4 Apr 1997 01:21:39 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id BAA20428; Fri, 4 Apr 1997 01:16:16 -0800 (PST) cc: Darren Reed , proff@suburbia.net, hackers@freebsd.org Subject: Re: IP Filter in -current. In-reply-to: Your message of "Fri, 04 Apr 1997 00:23:34 PST." <18999.860142214@time.cdrom.com> Date: Fri, 04 Apr 1997 01:16:15 -0800 Message-ID: <20424.860145375@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > "Labotomy", and I think Bruce is too valuable to even consider such a Doh! I just love it when I "correct" someone's spelling like this. "Lobotomy", OK? "Lobotomy!" It's what I thought but not what I typed. :-) From owner-freebsd-hackers Fri Apr 4 01:44:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA29308 for hackers-outgoing; Fri, 4 Apr 1997 01:44:11 -0800 (PST) Received: from panda.hilink.com.au (panda.hilink.com.au [203.2.144.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA29297 for ; Fri, 4 Apr 1997 01:44:06 -0800 (PST) Received: (from danny@localhost) by panda.hilink.com.au (8.8.5/8.7.3) id TAA17587; Fri, 4 Apr 1997 19:43:59 +1000 (EST) Date: Fri, 4 Apr 1997 19:43:58 +1000 (EST) From: "Daniel O'Callaghan" To: "Jordan K. Hubbard" cc: Darren Reed , proff@suburbia.net, hackers@FreeBSD.ORG Subject: Re: IP Filter in -current. In-Reply-To: <18999.860142214@time.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Fri, 4 Apr 1997, Jordan K. Hubbard wrote: > > Someone should give this guy a labotamy if he's the one who came up > > with the "include every other include file in this include file" idea. > > "Labotomy", and I think Bruce is too valuable to even consider such a > radical treatment for him - how about trying direct debate with him > over this issue instead? ;-) A "labotomy" is probably the removal of lips. Bruce doesn't talk much anyway, so it might not make too much difference. (Hi Bruce! :-)) A "Frontal lobotomy" is the removal of the frontal lobes of the cerebral cortex. Now that *would* make a difference. Danny From owner-freebsd-hackers Fri Apr 4 01:56:31 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id BAA29706 for hackers-outgoing; Fri, 4 Apr 1997 01:56:31 -0800 (PST) Received: from polaris.we.lc.ehu.es (polaris.we.lc.ehu.es [158.227.6.43]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id BAA29699 for ; Fri, 4 Apr 1997 01:56:25 -0800 (PST) Received: from sirius.we.lc.ehu.es by polaris.we.lc.ehu.es (SMI-8.6/SMI-SVR4) id LAA01442; Fri, 4 Apr 1997 11:54:49 +0200 From: borjam@we.lc.ehu.es (Borja Marcos) Received: by sirius.we.lc.ehu.es (4.1/SMI-4.1) id AA17319; Fri, 4 Apr 97 12:02:52 +0200 Message-Id: <9704041002.AA17319@sirius.we.lc.ehu.es> Subject: new malloc To: freebsd-hackers@freebsd.org Date: Fri, 4 Apr 1997 12:02:51 +0200 (MET DST) X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hello, I have read with great interest Poul Henning-Kamp's article on the new malloc implementation for FreeBSD. It's funny to find a "twin" who has been solving a similar problem and has reached similar conclusions. In my case, I have been solving a more restricted problem, because what I have done is a buffer pool which offers buffers of a fixed size. Most of the techniques user in the new malloc are similar to mine, but I have had an idea that could be useful to the new malloc. My main concern when writing my "fixed-block_size_malloc" was accessing the least number of pages possible. In my case, it's used in a computer based voice logging system, with lots of processes and threads (the number of threads can reach 100) and a lot of IPC activity. The conclusion was: Why not organizing the memory blocks in a LIFO structure? If I do a malloc and malloc can give me the most recently used of all the possible blocks, the probability of causing a page-in is decreased. Depending on the system load, a great number of page-ins can be avoided. (I have observed this in my case) Of course, the same idea cn be applied adding a MRU policy to the allocation routine. Borja. -- *********************************************************************** Borja Marcos * Internet: borjam@we.lc.ehu.es Alangoeta, 11 1 izq * borjam@well.com 48990 - Algorta (Vizcaya) * borjamar@sarenet.es SPAIN * CompuServe: 100015,3502 *********************************************************************** NOTE TO BULK EMAIL SENDERS: When I receive an unsolicited advertisement, I take some time to send complaints not only to your service provider, but to the next level providers who serve your ISP. From owner-freebsd-hackers Fri Apr 4 02:04:20 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id CAA00117 for hackers-outgoing; Fri, 4 Apr 1997 02:04:20 -0800 (PST) Received: from polaris.we.lc.ehu.es (polaris.we.lc.ehu.es [158.227.6.43]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id CAA00109 for ; Fri, 4 Apr 1997 02:04:05 -0800 (PST) Received: from sirius.we.lc.ehu.es by polaris.we.lc.ehu.es (SMI-8.6/SMI-SVR4) id MAA01543; Fri, 4 Apr 1997 12:02:19 +0200 From: borjam@we.lc.ehu.es (Borja Marcos) Received: by sirius.we.lc.ehu.es (4.1/SMI-4.1) id AA17376; Fri, 4 Apr 97 12:10:22 +0200 Message-Id: <9704041010.AA17376@sirius.we.lc.ehu.es> Subject: new malloc To: freebsd-hackers@freebsd.org Date: Fri, 4 Apr 1997 12:02:51 +0200 (MET DST) X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hello, I have read with great interest Poul Henning-Kamp's article on the new malloc implementation for FreeBSD. It's funny to find a "twin" who has been solving a similar problem and has reached similar conclusions. In my case, I have been solving a more restricted problem, because what I have done is a buffer pool which offers buffers of a fixed size. Most of the techniques user in the new malloc are similar to mine, but I have had an idea that could be useful to the new malloc. My main concern when writing my "fixed-block_size_malloc" was accessing the least number of pages possible. In my case, it's used in a computer based voice logging system, with lots of processes and threads (the number of threads can reach 100) and a lot of IPC activity. The conclusion was: Why not organizing the memory blocks in a LIFO structure? If I do a malloc and malloc can give me the most recently used of all the possible blocks, the probability of causing a page-in is decreased. Depending on the system load, a great number of page-ins can be avoided. (I have observed this in my case) Of course, the same idea cn be applied adding a MRU policy to the allocation routine. Borja. -- *********************************************************************** Borja Marcos * Internet: borjam@we.lc.ehu.es Alangoeta, 11 1 izq * borjam@well.com 48990 - Algorta (Vizcaya) * borjamar@sarenet.es SPAIN * CompuServe: 100015,3502 *********************************************************************** NOTE TO BULK EMAIL SENDERS: When I receive an unsolicited advertisement, I take some time to send complaints not only to your service provider, but to the next level providers who serve your ISP. From owner-freebsd-hackers Fri Apr 4 02:26:28 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id CAA00764 for hackers-outgoing; Fri, 4 Apr 1997 02:26:28 -0800 (PST) Received: from root.com (implode.root.com [198.145.90.17]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id CAA00759 for ; Fri, 4 Apr 1997 02:26:25 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by root.com (8.8.5/8.6.5) with SMTP id CAA10928; Fri, 4 Apr 1997 02:27:01 -0800 (PST) Message-Id: <199704041027.CAA10928@root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: borjam@we.lc.ehu.es (Borja Marcos) cc: freebsd-hackers@freebsd.org Subject: Re: new malloc In-reply-to: Your message of "Fri, 04 Apr 1997 12:02:51 +0200." <9704041002.AA17319@sirius.we.lc.ehu.es> From: David Greenman Reply-To: dg@root.com Date: Fri, 04 Apr 1997 02:27:01 -0800 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > The conclusion was: Why not organizing the memory blocks in >a LIFO structure? If I do a malloc and malloc can give me the most >recently used of all the possible blocks, the probability of causing >a page-in is decreased. Depending on the system load, a great >number of page-ins can be avoided. (I have observed this in my case) > > > Of course, the same idea cn be applied adding a MRU policy to the >allocation routine. Both Poul-Henning's and the FreeBSD kernel's malloc allocate and insert chunks at the head of the free queue - thus they already are LIFO. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-hackers Fri Apr 4 02:44:26 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id CAA01224 for hackers-outgoing; Fri, 4 Apr 1997 02:44:26 -0800 (PST) Received: from gw.itfs.nsk.su (ns.itfs.nsk.su [193.124.36.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id CAA01168 for ; Fri, 4 Apr 1997 02:43:54 -0800 (PST) Received: from itfs.UUCP (uucp@localhost) by gw.itfs.nsk.su (8.6.12/8.6.12) with UUCP id RAA29284 for hackers@freebsd.org; Fri, 4 Apr 1997 17:32:28 +0700 Received: by itfs.nsk.su; Fri, 4 Apr 97 17:36:21 +0600 (NSK) Received: (from daemon@localhost) by news.itfs.nsk.su (8.7.5/8.6.12) id RAA09668; Fri, 4 Apr 1997 17:12:03 +0700 (NSD) From: "Nickolay N. Dudorov" To: hackers@freebsd.org Subject: Re: FreeBSD Elf-Kit and dynamic loading Date: 4 Apr 1997 10:12:01 GMT Message-ID: <5i2k5h$4jb@news.itfs.nsk.su> References: <5i0j1d$jtk@news.itfs.nsk.su> <19970403191209.52889@keltia.freenix.fr> Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Ollivier Robert wrote: > According to Nickolay N. Dudorov: > > And after that I discovered that 'file hello' > > says: > > hello: ELF 32-bit LSB executable, Intel 80386, version 1 (FreeBSD) > > > > and file 'hello' is 47761 bytes long. > > > > So, there is a question - is there any possibility to > > build "dynamic ELF" program ? My elf-gcc produce identical results > > with and without '-static' flag. > Are you sure you have the right naming for the libraries ? Yes - it turns out that because I build and install elf-libc from (copy of) FreeBSD-2.2 sources there was no libc.so --> libc.so.3.0 link. After I make all libXX.so --> libXX.so.M.N links my test programs builds as dynamic. There is still some problems unsolved by me today: - If I build libmsun from RELENG_2_2 sources there is an error in file 'msun/i387/s_finite.S'. I blindly replace 'setnel %al' with 'setne %al' (error message says - register does not match opcode suffix) and library successfuly builds ;-) but is this a right "fix" ? - I have not found the way to build g++ shared libraries (from GNU libg++-2.7.2 sources) So, I erased all my elf-related bins and libs in order to make it all from scratch and write down all the steps and problems I'll encounter in the way to 'make (almost) all user-land libraries and binaries in ELF' ;-) The most serious problem now is - where can I get another 16 hours in a day to perform this very interesting but not very profitable experiment :-( N.Dudorov From owner-freebsd-hackers Fri Apr 4 03:50:48 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id DAA03171 for hackers-outgoing; Fri, 4 Apr 1997 03:50:48 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id DAA03162 for ; Fri, 4 Apr 1997 03:50:44 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA25194; Fri, 4 Apr 1997 06:50:12 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Fri, 4 Apr 1997 06:50 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id GAA25907; Fri, 4 Apr 1997 06:11:58 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id GAA10676; Fri, 4 Apr 1997 06:17:51 -0500 (EST) Date: Fri, 4 Apr 1997 06:17:51 -0500 (EST) From: Thomas David Rivers Message-Id: <199704041117.GAA10676@lakes.water.net> To: ponds!camelot.de!knarf, ponds!lakes.water.net!rivers Subject: Re: kern/2923: panic: vm_fault: fault on nofault entry, addr: f6e21000 Cc: ponds!freebsd.org!freebsd-hackers Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > Hello, > > Thomas David Rivers wrote: > > > I believe it to be a timing problem in the kernel. A heavy load just > > gives you a greater opportunity (as a percentage of time) to trip over > > the problem. Again; this also happens on my extremely lightly loaded > > (i.e. only a shell) system I put together just to reproduce this problem. > > I have the same problem here on a medium loaded newsserver (6 GB > newsspool). I get "panic: vm_fault: fault on nofault entry, addr: > f6718000" with different values for addr. I updated from 2.1.5 to > 2.2.1 about 48 hours before I've seen this problem the first time. > > I really have no idea what to do now. The machine always reboots > some minutes after the crash. Running fsck on 10 GB disks all the > time is no fun... :/ Well - this doesn't sound like exactly the same problem - although I can easily imagine it's just a frustrating. > > If you have any idea how I can make the uptime 6 hrs or higher, please > tell me! Since I'm guessing this is a different problem; if you'll provide the kgdb traceback and an nm of your kernel (are you running a 2.2.1 GENERIC kernel?) and details of your system - it's quite possible your problem isn't as thorny as mine and can be much more readily repaired. > > Thank you, > Knarf > -- > Frank Bartels |UUCP/ZModem/Fax: +49 89 8948040| "Captain, why not just > knarf@camelot.de | http://www.camelot.de/~knarf/ | give the Borg Windows?" > - Dave Rivers - From owner-freebsd-hackers Fri Apr 4 03:50:49 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id DAA03172 for hackers-outgoing; Fri, 4 Apr 1997 03:50:49 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id DAA03161 for ; Fri, 4 Apr 1997 03:50:44 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA25148; Fri, 4 Apr 1997 06:50:06 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Fri, 4 Apr 1997 06:50 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id FAA25826; Fri, 4 Apr 1997 05:58:40 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id GAA10560; Fri, 4 Apr 1997 06:04:32 -0500 (EST) Date: Fri, 4 Apr 1997 06:04:32 -0500 (EST) From: Thomas David Rivers Message-Id: <199704041104.GAA10560@lakes.water.net> To: ponds!etinc.com!dennis, ponds!uriah.heep.sax.de!joerg_wunsch, ponds!haldjas.folklore.ee!narvi Subject: pay-for support? (was: Re: Does de driver do 100MBIT Full Duplex?) Cc: ponds!FreeBSD.ORG!hackers Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk dennis writes: > > At 09:10 AM 4/3/97 -0500, dennis wrote: > >At 10:23 AM 4/3/97 +0300, Narvi wrote: > >> > >> > >>On Wed, 2 Apr 1997, J Wunsch wrote: > >> > >>> As dennis wrote: > >>> > >>> > Is the driver that you send me directly for 2.1.7 (that worked > >>> > beautifully) in 2.2.1R? If not, WHY NOT! > >>> > >>> No need to shout. We aren't deaf exactly. > >>> > >>> Since Matt released it _after_ 2.2 has been cut. I've got a very > >>> explicit message from Matt when i've been asking him earlier (right in > >>> time to get something into 2.2 still) about a new version, that he > >>> considered the stuff that sneaked into NetBSD by that time too buggy > >>> to see it officially in FreeBSD. That's why we've been integrating > >>> the little hack still to make at least the DE21140A supported. We > >>> originally deferred the inclusion of this patch in anticipation of the > >>> new driver version. > > It seems to me, that the fact that the most popular driver for FreeBSD > doesn't work is a sufficient condition to hold up the release until it > does. > > What you have now is a release that was supposed to be a "great > saviour" feature-wise that is fundamentally unusable in its released > form for a large number of users..... > > Dennis > Hmmm... why do you say that's the most popular driver. Out of the all of the machines I can physically touch, only one of them uses that driver. I'd say the most popular driver was the syscons one :-) For my purposes, second to syscons is aha2940, which coincidently, suffers a little as well. This begs a question I've been considering for some time. Just how much would you be willing to pay for a release that was supported on your hardware? Would you accept purchasing the hardware from a vendor that "guaranteed" (in some sense) that FreeBSD worked, and provided FreeBSD updates as necessary? If that's true; what's the limit you'd put on that. Certainly a million dollars (US) is too great; but would you go for, say, a few hundred above what Gateway is selling equivalent hardware? Or, would a better model be the way Red Hat is selling Linux; less that $50 a user; with phone technical support? [I figure, since I live in the same county as Red Hat, and could maybe even steal some of their technical support people :-), I could possibly start up this business...] - Dave Rivers - From owner-freebsd-hackers Fri Apr 4 04:50:48 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id EAA06039 for hackers-outgoing; Fri, 4 Apr 1997 04:50:48 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id EAA06034 for ; Fri, 4 Apr 1997 04:50:36 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA27582; Fri, 4 Apr 1997 07:50:04 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Fri, 4 Apr 1997 07:50 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id GAA26369; Fri, 4 Apr 1997 06:22:04 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id GAA10738; Fri, 4 Apr 1997 06:27:56 -0500 (EST) Date: Fri, 4 Apr 1997 06:27:56 -0500 (EST) From: Thomas David Rivers Message-Id: <199704041127.GAA10738@lakes.water.net> To: ponds!freefall.cdrom.com!freebsd-hackers, ponds!iafrica.com!khetan, ponds!lakes.water.net!rivers Subject: ucbvax.Berkely.EDU (was Re: Uninvited usage for sendmail) Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > >Get sendmail 8.8.5 (seems stable now, been out for a couple of months) > >and install the check_* rules. check_relay is the one you want. > > Just to make like easier, I'm including my freebsd.mc file, which you can > use to build your "own" sendmail.cf (in /etc). > ... > divert(0)dnl > VERSIONID(`@(#)freebsd.mc $Revision: 1.3 $') > OSTYPE(bsd4.4)dnl > DOMAIN(generic)dnl > MAILER(local)dnl > MAILER(smtp)dnl > FEATURE(mailertable, `hash -o /etc/mailertable')dnl > define(`UUCP_RELAY', ucbvax.Berkeley.EDU)dnl > define(`BITNET_RELAY', mailhost.Berkeley.EDU)dnl > define(`CSNET_RELAY', mailhost.Berkeley.EDU)dnl > define(`confCW_FILE', `-o /etc/mail/whoarewe')dnl You likely want to fix that UUCP_RELAY definition; ucbvax.Berkeley.EDU no longer exists... (uunet.uu.net is probably still a good alternative.) Believe me, appearances of ucbvax.Berkeley.EDU in uucp !-paths is quite an annoyance. In fact, you can dispense with the XXX_RELAYs all-together if you go to a SMART_HOST-based definition... - Dave Rivers - From owner-freebsd-hackers Fri Apr 4 05:30:08 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA07562 for hackers-outgoing; Fri, 4 Apr 1997 05:30:08 -0800 (PST) Received: from mailhost1.BayNetworks.COM (ns3.BayNetworks.COM [134.177.3.16]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA07542 for ; Fri, 4 Apr 1997 05:30:03 -0800 (PST) Received: from ns1.BayNetworks.COM ([134.177.1.107]) by mailhost1.BayNetworks.COM (8.8.5/BNET-97/03/26-E) with ESMTP id FAA18311; Fri, 4 Apr 1997 05:29:29 -0800 (PST) for Received: from pobox.engeast.BayNetworks.COM (pobox.corpeast.baynetworks.com [192.32.151.199]) by ns1.BayNetworks.COM (8.8.5/BNET-97/03/12-I) with ESMTP id FAA00256; Fri, 4 Apr 1997 05:29:27 -0800 (PST) for Posted-Date: Fri, 4 Apr 1997 05:29:27 -0800 (PST) Received: from tuva.engeast.baynetworks.com (tuva [192.32.180.119]) by pobox.engeast.BayNetworks.COM (SMI-8.6/BNET-SVR4-S) with ESMTP id IAA02096; Fri, 4 Apr 1997 08:29:25 -0500 for Received: from tuva.engeast.baynetworks.com (localhost [127.0.0.1]) by tuva.engeast.baynetworks.com (8.8.3/8.8.3) with ESMTP id IAA06810 for ; Fri, 4 Apr 1997 08:29:19 -0500 (EST) Message-Id: <199704041329.IAA06810@tuva.engeast.baynetworks.com> X-Mailer: exmh version 2.0gamma 1/27/96 To: hackers@freebsd.org Subject: Suggested change to rc Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 04 Apr 1997 08:29:14 -0500 From: Robert Withrow Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In order to have amd get all of its information from nis, the following change needs to be made in rc. This allows amdflags to be set like this: amdflags='-p -c 3600 -l syslog `ypcat -k amd.master`' *** rc~ Tue Mar 25 09:49:40 1997 --- rc Fri Apr 4 08:25:32 1997 *************** *** 269,275 **** if [ "X${amdflags}" != X"NO" ]; then echo -n ' amd' ! amd -p ${amdflags} > /var/run/amd.pid fi # $rwhod is imported from /etc/sysconfig; --- 269,275 ---- if [ "X${amdflags}" != X"NO" ]; then echo -n ' amd' ! eval amd -p ${amdflags} > /var/run/amd.pid fi # $rwhod is imported from /etc/sysconfig; -- Robert Withrow -- (+1 508 436 8256) BWithrow@BayNetworks.com From owner-freebsd-hackers Fri Apr 4 05:36:07 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA08018 for hackers-outgoing; Fri, 4 Apr 1997 05:36:07 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA07965; Fri, 4 Apr 1997 05:35:51 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id FAA01852; Fri, 4 Apr 1997 05:35:49 -0800 (PST) To: hackers@freebsd.org cc: www@freebsd.org Subject: Better late than never... The FreeBSD News! Date: Fri, 04 Apr 1997 05:35:49 -0800 Message-ID: <1848.860160949@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk OK folks, sorry for the delay on this.. As I intimated last week, the "FreeBSD News" newsletter publication thingy is now past the early planning & layout stages and into the "who will write the articles?" stage. We're not looking for anything super ambitious in the way of articles here, just something which describes an interesting use you're making of FreeBSD or aims itself at educating the new user about some aspect of FreeBSD's infrastructure (Davidg talking about wcarchive, mpp talking about GNATs and how to use it, jmb talking about the mailing lists, etc). Generic "how to configure your " sorts of tutorials would be nice too. Length should be 300-400 words, though shorter "quick tip" articles are always welcome too. We can make the panels any size we want. :) This introductory issue isn't going to be very large (maybe 6 pages, counting both front and back covers) and it's intended more to give a taste of things to come than anything else, also setting the stage with some of the "what is FreeBSD?" boilerplate that will take up permanent residence in the front). That doesn't mean you shouldn't all submit articles like crazy, of course, since there's always the next issue(s) and one can *never have enough articles*! :) "How much does it pay?" you ask me. "Erm, $0.00" I answer. Sorry 'bout that, but Walnut Creek CDROM is looking for strictly volunteer writers for TFN and for good reason - anything published in the Nooze will be treated the same as BSD-copyrighted code going into the source tree - you can redistribute, republish, do whatever you like with it. Sell it to other magazines, we don't care! The good news: You don't have to sweat the formatting because the whole thing is going to be typeset in Quark, that being what our art dept. uses and is familiar with. Just raw ascii text files will do, along with any .gif or .jpeg files for figures and drawings (Fig 1: "new growth in FreeBSD userbase indexed against coffee sales world-wide") and they'll make it all look neat. As the editor, I will also run a spell checker over it and make suggestions about any particularly odd passages I find (:-). I'll also make an archive of all the submissions available via FTP, of course, and wosch, I'm sure, will also have an HTML-ifying script put together for it all before the first issue reaches the subscribers. :) Walnut Creek CDROM will be printing and mailing some 30,000 copies of this issue, and when you add to that the # of people who will probably read it on the web, that's actually a pretty fair audience. I only now hope that we can get content from one part of our user base for the other part to read! :-) The deadline for submissions is April 21st. Again, we're not talking War & Peace here, just ~300 words or less, subject of course to negotiation (if you're being positively fascinating, I suppose we can always include more words and knock the point size down a notch or try to beg another page out of the publisher [WC]). What have you got to lose? Go ahead! Send me some text! It doesn't have to be pretty, we'll doctor it up before the deadline! Let's get this publication on the road and get some serious articles going! Yeah! Rah rah rah!! [cue inspirational theme music from "Rocky"] :-) Some example topics just off the top of my head for those suffering from writer's block and/or deep existential angst: o The FreeBSD mailing lists and how to make them work for you. [jmb? :)] o The FreeBSD web site - an overview of services offered. o The FreeBSD problem reporting mechanism - how to submit reports and how to search for them. [ mpp is signed up for this ]. o Staying up to date with {CVSup | CTM} [jdp? phk? :)] o Setting up an effective X desktop environment. o How FreeBSD is developed [subtitle: What are core team members and committers and how does one become one?]. o Installing and managing your Apache web server. o Installing and managing SAMBA for Win95/NT interoperability o Running Linux binaries under FreeBSD - what to do and where to get the libraries. o A comparison between FreeBSD and Linux [special note: This topic reserved for someone with cojones the size of coconuts]. Also don't fall into the trap of thinking that you have to write something brand-spanking-new for the newsletter in order for it to have any value - you don't! Think of your readers as newborn babes who just found FreeBSD under a rock and haven't the faintest idea that there is even a web page for it, much less looked at it. Perhaps a previously written handbook section can find new life as a short column? Perhaps so! Also, as I told Mike Pritchard when I asked him about doing a "GNATs & how to use it" article, much of this material can also, with a little editing, do double-duty as periodic posting fodder for comp.unix.bsd.freebsd.announce. I've really neglected the whole "periodic posting" thing for awhile now, and if I had short, up-to-date descriptions of the mailing lists and bug reporting mechanisms as well as a web resource summary, I'd post each once a month for the benefit of new users. All correspondence to me please (and watch your cc lines! I don't think all of -hackers wants to read your first draft :-). Thanks! Jordan From owner-freebsd-hackers Fri Apr 4 05:51:20 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id FAA08790 for hackers-outgoing; Fri, 4 Apr 1997 05:51:20 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA08785 for ; Fri, 4 Apr 1997 05:51:15 -0800 (PST) Received: from dbws.etinc.com (dbws.etinc.com [204.141.95.130]) by etinc.com (8.8.3/8.6.9) with SMTP id IAA09019; Fri, 4 Apr 1997 08:55:56 -0500 (EST) Message-Id: <3.0.32.19970404090548.006b121c@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Fri, 04 Apr 1997 09:05:57 -0500 To: Doug Rabson From: dennis Subject: Re: NFS problems with 2.2.1R Cc: hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 12:40 AM 4/4/97 +0100, you wrote: >On Thu, 3 Apr 1997, dennis wrote: > >> >> I've concluded that the problems that I've been having are *not* due to >> a buggy de driver but NFS. I have the same symptoms on a totally >> different machine with an ne2000 card as I had with the de PCI >> card. >> >> My server is a 2.1.6R box which has been reliably been working for >> quite some time (previously with 2.1.5 of course). When loading >> 2.2.1R via NFS, it consistantly hangs loading bin, however not >> in the same place. The system completely hangs (ie alt-F keys >> do not work, cant ctl-alt-del...). >> >> Anyone with any ideas on how to get around this? I loaded 2.2R >> without incident on the same hardware. > >When the system is hung, is there any net traffic? I just fixed a bug in >current which caused it to loop, continually re-reading the same directory >block. This is really obvious when you use tcpdump. I can make a patch >for this specific problem against 2.2 but I don't (yet) have a 2.2 machine >to test. No, there is no network traffic. I have a LAN monitor on the network and the server sends a window of frames that do not get ack'd by the hung machine. As this is occurring with the boot floppy I'm not able to do much diagnosis :( Dennis > >-- >Doug Rabson Mail: dfr@nlsystems.com >Nonlinear Systems Ltd. Phone: +44 181 951 1891 > > > From owner-freebsd-hackers Fri Apr 4 06:31:53 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA10161 for hackers-outgoing; Fri, 4 Apr 1997 06:31:53 -0800 (PST) Received: from labinfo.iet.unipi.it (labinfo.iet.unipi.it [131.114.9.5]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id GAA10142 for ; Fri, 4 Apr 1997 06:31:48 -0800 (PST) Received: from localhost (luigi@localhost) by labinfo.iet.unipi.it (8.6.5/8.6.5) id PAA03129; Fri, 4 Apr 1997 15:37:39 +0200 From: Luigi Rizzo Message-Id: <199704041337.PAA03129@labinfo.iet.unipi.it> Subject: Re: NFS problems with 2.2.1R To: dennis@etinc.com (dennis) Date: Fri, 4 Apr 1997 15:37:38 +0200 (MET DST) Cc: dfr@nlsystems.com, hackers@freebsd.org In-Reply-To: <3.0.32.19970404090548.006b121c@etinc.com> from "dennis" at Apr 4, 97 09:05:38 am X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > >When the system is hung, is there any net traffic? I just fixed a bug in > > No, there is no network traffic. I have a LAN monitor on the network and the > server sends a window of frames that do not get ack'd by the hung machine. > > As this is occurring with the boot floppy I'm not able to do much diagnosis :( you can try to ping the system and see if it at least responds. Luigi From owner-freebsd-hackers Fri Apr 4 06:35:37 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA10420 for hackers-outgoing; Fri, 4 Apr 1997 06:35:37 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA10412 for ; Fri, 4 Apr 1997 06:35:34 -0800 (PST) Received: from dbws.etinc.com (dbws.etinc.com [204.141.95.130]) by etinc.com (8.8.3/8.6.9) with SMTP id JAA09266; Fri, 4 Apr 1997 09:40:22 -0500 (EST) Message-Id: <3.0.32.19970404095015.006b3514@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Fri, 04 Apr 1997 09:50:18 -0500 To: Michael Smith From: dennis Subject: Re: Does de driver do 100MBIT Full Duplex? Cc: hackers@FreeBSD.ORG Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk At 09:12 AM 4/4/97 +0930, Michael Smith wrote: >dennis stands accused of saying: >> >> What you have now is a release that was supposed to be a "great >> saviour" feature-wise that is fundamentally unusable in its released >> form for a large number of users..... > >Because, as has been observed beforehand, this "large number" of users >were either too timid, too indifferent, or too _stupid_ to participate >in the testing process. Please dont blame us for our genetic problems....... :-) > >I sure as hell don't recall any observations on your part on 2.2 >during the prerelease testing on this or a number of other points >you've subsequently griped about; I would suggest that you could >probably have justified spending just a few hours a couple of times >installing and testing the various fully-bundled prerelease snapshots >and feeding back on your woes. As much as I would have like to download and test the countless alphas, betas and gamas, I kind of have my hands full at the moment. Life isn't always so simple...and unlike most of you, I have to support 5 O/S, some of which have muliple "releases", and until they get that 30 hour day thing going I'm sunk. The point here is that you *knew* that de wasn' t ready, and if NFS loads really dont work then it seems incomprehensible that no-one knew about it... unless it was simply never tested. > >If you, and a few other serious, "production" users had just >_thought_ahead_ a little, 2.2 would have been less of a debacle than >it has currently turned out to be. Without that sort of input >_before_ the final release is rolled, it's _not_ going to be possible >to make you happy. Ever. And that would be Bad. The question is, how far ahead do we think? Ever since 1.0 there's been the promise of "finality" in the next release, but every time the next release rolls around there are sufficient problems to consider waiting for the next one. The "expectation" is that things that worked before will still work with a new release, and that there may be problems with new stuff. When there are problems with old stuff, you have a real perception problem with your product. There has been much talk about get publicity; getting corporate users to use FreeBSD, but until you get to a point where loading a new release is much less than a crapshoot It's just not going to happen on a widespread basis. db From owner-freebsd-hackers Fri Apr 4 06:43:29 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA10751 for hackers-outgoing; Fri, 4 Apr 1997 06:43:29 -0800 (PST) Received: from chai.plexuscom.com (chai.plexuscom.com [207.87.46.100]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA10744 for ; Fri, 4 Apr 1997 06:43:26 -0800 (PST) Received: from chai.plexuscom.com (localhost [127.0.0.1]) by chai.plexuscom.com (8.8.5/8.8.5) with ESMTP id JAA23469; Fri, 4 Apr 1997 09:44:16 -0500 (EST) Message-Id: <199704041444.JAA23469@chai.plexuscom.com> To: Marc Slemko , David Greenman Cc: FreeBSD-hackers Subject: Re: apache like preforking apps and high loads In-reply-to: Your message of "Thu, 03 Apr 1997 10:50:19 MST." Date: Fri, 04 Apr 1997 09:44:16 -0500 From: Bakul Shah Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk David Greenman writes: > The processes blocked on accept are handled in a round-robin fashion, > oldest first. Thanks! In response to > > Fairness is probably not an issue when an app. consists of a number > > of anonymous servers but in general one would want to make sure that > > if N processes are waiting on accept() on the same socket, no one > > process is starved of accepting. How do you ensure that? Marc Slemko writes: > For something like Apache you want the _least_ equal distribution > possible, ie. a stack of most recently used processes from which you pop a > server to process a request. Right now, if you have 100 servers running > and get requests at a rate that one could handle, all of them will still > be used. This is bad; it hurts locality of reference a lot. On some > architectures, it has a significant impact WRT caching, and also means > that if you don't have enough RAM for all server processes then they will > each end up being swapped out and in again. Granted that Apache like apps don't care about fairness to _server_ processes and may even provide worse performance with fair scheduling. Which is why such apps should do their own scheduling if at all possible. But without some hints (and a mechanism to specify such hints) about *how* some processes are cooperating, a generic kernel must treat all processes equally. If providing equal service to all _accepted_ connections is one's goal, handling accepts in a stack fashion is not ideal either: clients serveed by a process deep in the stack will get worse service compared to one handled by a process near the top of the stack. > The reason why Apache doesn't do this is the implementation details. It > is difficult to get a good implementation that is efficient (so you don't > create more overhead than you remove) and portable. I don't know Apache details but handling one client at-a-time per process does not seem like the most efficient use of resources given Unix'es heavyweight processes. As for portability *and* performance, you pretty much need _some_ OS/version/processor specific code. > > To guido: For apache like apps one idea is to have one process be > > the acceptor and have it pass a new socket to individual server > > processes (including doing some initial processing to determine > > which process should get this socket if not all server processes are > > equivalent). > > This has a lot of portability issues and can lead to running into things > such as per process fd limits more quickly on some architectures. Even Linux handles passing file descriptors via sendmsg(). Running out of fds should not be a problem if the acceptor process simply passes on the fd to another process. For initial processing things get a bit trickier and you may need more acceptor processes (but N/FD_SETSIZE cuts down your accept() load considerably). -- bakul From owner-freebsd-hackers Fri Apr 4 06:50:30 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA11042 for hackers-outgoing; Fri, 4 Apr 1997 06:50:30 -0800 (PST) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id GAA11037 for ; Fri, 4 Apr 1997 06:50:27 -0800 (PST) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.5/8.7.3) id AAA21916; Sat, 5 Apr 1997 00:20:10 +0930 (CST) From: Michael Smith Message-Id: <199704041450.AAA21916@genesis.atrad.adelaide.edu.au> Subject: Re: Does de driver do 100MBIT Full Duplex? In-Reply-To: <3.0.32.19970404095015.006b3514@etinc.com> from dennis at "Apr 4, 97 09:50:18 am" To: dennis@etinc.com (dennis) Date: Sat, 5 Apr 1997 00:20:10 +0930 (CST) Cc: msmith@atrad.adelaide.edu.au, hackers@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk dennis stands accused of saying: > > > >Because, as has been observed beforehand, this "large number" of users > >were either too timid, too indifferent, or too _stupid_ to participate > >in the testing process. > > Please dont blame us for our genetic problems....... :-) If I thought the problem was genetic, I'd be off abusing your parents 8) > As much as I would have like to download and test the countless > alphas, betas and gamas, I kind of have my hands full at the moment. > Life isn't always so simple...and unlike most of you, I have to support > 5 O/S, some of which have muliple "releases", and until they get > that 30 hour day thing going I'm sunk. Please note that many of the rest of us have the same sort of time restrictions that you do; we depend on FreeBSD working for us and so I have integrated testing FreeBSD into our product test process. It means that when something busts in a prerelease we can often feed the problem right back and integrate the FreeBSD fix cycle into our own. We, too, support (to admittedly a much lesser degree than you do) our software on a pile of different platforms, most hostile. I delegate that work. > The point here is that you *knew* that de wasn' t ready, and if NFS loads > really dont work then it seems incomprehensible that no-one knew about it... > unless it was simply never tested. I didn't "know" the 'de' driver wasn't ready - every 'de' card I could get my hands on worked flawlessly with the driver in 2.2. During the intensive training we have been carrying on for our field techs they have been averaging 2 or 3 FreeBSD installs per day, via NFS from either their notebooks, our build server, or one of our GP workstations off an MOD, or via FTP from the same source(s). > The question is, how far ahead do we think? Ever since 1.0 there's > been the promise of "finality" in the next release, but every time > the next release rolls around there are sufficient problems to > consider waiting for the next one. Yup, and until the wider userbase hops on the bandwagon by testing the releases before they are cut, this will continue. If you can spare a few hours to install a BETA- or GAMMA- prerelease on a test machine and pound it for a while like you would a production machine, you're helping. Downloading the release and saying "when I do XXX it doesn't work" is just going to get someone like me, who has torn his hair out trying to find anything to complain about, _very_cross_. I and my employer donate my/our time and resources to this; it'd be nice to feel we were being appreciated just a little. > There has been much talk about get publicity; getting corporate users to > use FreeBSD, but until you get to a point where loading a new release is much > less than a crapshoot It's just not going to happen on a widespread basis. Major vendors achieve this by spending lots of money on test hardware and "beta sites". We don't have money; we depend on the generosity and enlightened self-interest of the user community. Sadly, this is one area in which that's still wanting 8( > db -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ From owner-freebsd-hackers Fri Apr 4 06:56:58 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA11363 for hackers-outgoing; Fri, 4 Apr 1997 06:56:58 -0800 (PST) Received: from wgold.demon.co.uk (wgold.demon.co.uk [158.152.96.124]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id GAA11353 for ; Fri, 4 Apr 1997 06:56:55 -0800 (PST) Received: from wgold.demon.co.uk by wgold.demon.co.uk (NTMail 3.02.10) with ESMTP id ga001306 for ; Fri, 4 Apr 1997 11:11:31 +0100 Message-ID: <3344D3D3.618A@wgold.demon.co.uk> Date: Fri, 04 Apr 1997 11:11:31 +0100 From: James Mansion Organization: Westongold Ltd X-Mailer: Mozilla 3.01Gold (WinNT; I) MIME-Version: 1.0 To: FreeBSD-hackers@freebsd.org Subject: Re: apache like preforking apps and high loads References: <199704031653.LAA20772@chai.plexuscom.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Info: Westongold Ltd: +44 1992 620025 www.westongold.com Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Bakul Shah wrote: > Fairness is probably not an issue when an app. consists of a number > of anonymous servers but in general one would want to make sure that > if N processes are waiting on accept() on the same socket, no one > process is starved of accepting. How do you ensure that? > Why would you necessarily want this, apart from aesthetics? I don't think this behaviour is mandated anywhere. In any case, presumably (hopefully) the wakeup goes to the process with the highest scheduling priority. If the scheduler does dynamic adjustment based on CPU time consumed 'recently' then waiters will be favoured over processes that ran recently, and you'll get th eeffect that you want. James From owner-freebsd-hackers Fri Apr 4 06:56:59 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id GAA11371 for hackers-outgoing; Fri, 4 Apr 1997 06:56:59 -0800 (PST) Received: from wgold.demon.co.uk (wgold.demon.co.uk [158.152.96.124]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id GAA11354 for ; Fri, 4 Apr 1997 06:56:55 -0800 (PST) Received: from wgold.demon.co.uk by wgold.demon.co.uk (NTMail 3.02.10) with ESMTP id fa001305 for ; Fri, 4 Apr 1997 10:54:55 +0100 Message-ID: <3344CFEE.589F@wgold.demon.co.uk> Date: Fri, 04 Apr 1997 10:54:54 +0100 From: James Mansion Organization: Westongold Ltd X-Mailer: Mozilla 3.01Gold (WinNT; I) MIME-Version: 1.0 To: FreeBSD-hackers@FreeBSD.org Subject: Re: apache like preforking apps and high loads References: <199704031357.PAA10893@gvr.win.tue.nl> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Info: Westongold Ltd: +44 1992 620025 www.westongold.com Sender: owner-hackers@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk Guido van Rooij wrote: > > When looking at Apacha like applications, one often sees extermely high > load averages. Apache preforks a number of processes that all block > on accept(). When a request comes in, all process are woken up and the > scheduler chooses one of the now runnable processes taht will succeed in > the accept(). The other go back to sleep. Silly question - why do they all get woken up? (Not 'why does the current implementation do this' but 'is this behaviour required'?) James From owner-freebsd-hackers Fri Apr 4 07:04:38 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA12071 for hackers-outgoing; Fri, 4 Apr 1997 07:04:38 -0800 (PST) Received: from pat.idt.unit.no (0@pat.idt.unit.no [129.241.103.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA12064 for ; Fri, 4 Apr 1997 07:04:34 -0800 (PST) Received: from idt.unit.no (26850@kamelia.idt.unit.no [129.241.111.27]) by pat.idt.unit.no (8.8.5/8.8.5) with ESMTP id RAA05693; Fri, 4 Apr 1997 17:03:20 +0200 (MET DST) Message-Id: <199704041503.RAA05693@pat.idt.unit.no> To: dg@root.com Cc: ponds!rivers@dg-rtp.dg.com, freebsd-hackers@freebsd.org Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) In-Reply-To: Your message of "Thu, 03 Apr 1997 20:10:45 -0800" References: <199704040410.UAA08780@root.com> X-Mailer: Mew version 1.06 on Emacs 19.34.1 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Date: Fri, 04 Apr 1997 17:03:15 +0200 From: Tor Egge Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Uh, this is wrong since VOP_INACTIVE really wants a '0' usecount vnode, > and there are assumptions throughout the code that a '0' usecount also > implies that the vnode is on the free list. A quick code review of Tor's > suggested fix shows that it will fail in several places in the kernel and > basically needs to be re-thought...which is why it hasn't been committed > yet. I'm running with the modified suggested fix now, and have not seen any failures due to that suggested fix. The original suggested fix failed due to the assumptions that a `0' usecount meant that it was on the free list, and a NULL pointer was dereferenced when trying to move the vnode to the head of the free list. Adding a kludge (magic number 0xdeadb, used elsewhere in the code to mark that the vnode was not on the freelist) made the code work for my tests. > No, a vnode on the free list is always "free" - it will be removed from > the free list prior to any I/O occuring. The only caveat to this is that > entries in the namei cache can point to "free" vnodes and there is some > kludgewear to deal with this. The -current code calls VOP_INACTIVE on vnodes on the free list having `0' as usecount, and fails when the vnode is reused for other purposes while VOP_INACTIVE is blocked. Before the Lite/2 merge, nfs_inactive did not try to unlock the node. The vnode was probably not referenced after the blocking of VOP_INACTIVE (due to nfs_removeit() blocking, waiting for RPC answer). - Tor Egge From owner-freebsd-hackers Fri Apr 4 07:43:26 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA14524 for hackers-outgoing; Fri, 4 Apr 1997 07:43:26 -0800 (PST) Received: from nlsystems.com (nlsys.demon.co.uk [158.152.125.33]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA14516 for ; Fri, 4 Apr 1997 07:43:20 -0800 (PST) Received: from herring.nlsystems.com (herring.nlsystems.com [10.0.0.2]) by nlsystems.com (8.8.5/8.8.5) with SMTP id QAA11191; Fri, 4 Apr 1997 16:43:08 +0100 (BST) Date: Fri, 4 Apr 1997 16:43:08 +0100 (BST) From: Doug Rabson To: Tor Egge cc: dg@root.com, ponds!rivers@dg-rtp.dg.com, freebsd-hackers@FreeBSD.ORG Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) In-Reply-To: <199704041503.RAA05693@pat.idt.unit.no> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Fri, 4 Apr 1997, Tor Egge wrote: > > Uh, this is wrong since VOP_INACTIVE really wants a '0' usecount vnode, > > and there are assumptions throughout the code that a '0' usecount also > > implies that the vnode is on the free list. A quick code review of Tor's > > suggested fix shows that it will fail in several places in the kernel and > > basically needs to be re-thought...which is why it hasn't been committed > > yet. > > I'm running with the modified suggested fix now, and have not seen any > failures due to that suggested fix. The original suggested fix failed > due to the assumptions that a `0' usecount meant that it was on the > free list, and a NULL pointer was dereferenced when trying to move the > vnode to the head of the free list. Adding a kludge (magic number > 0xdeadb, used elsewhere in the code to mark that the vnode was not on > the freelist) made the code work for my tests. I tried testing your fix this morning and the 0xdeadb stuff just caused vget to fault a couple of minutes into my test (simultaneous rm -rf largetree and cvs co src, both remote). This problem really has little to do with nfs_inactive. What is happening is a race between vgone and vget which would normally be solved by the vnode locks. Since NFS doesn't have vnode locks, the race happens. I am most of the way there in implementing the right solution for NFS which is to used shared locks for NFS; vgone can then use the lock manager to wait for all the shared locks to drain before recycling the vnode. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 181 951 1891 From owner-freebsd-hackers Fri Apr 4 07:51:09 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA15012 for hackers-outgoing; Fri, 4 Apr 1997 07:51:09 -0800 (PST) Received: from boulder.syr.servtech.com (boulder.syr.servtech.com [206.106.144.94]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA14994; Fri, 4 Apr 1997 07:50:52 -0800 (PST) Received: from boulder.syr.servtech.com (localhost [127.0.0.1]) by boulder.syr.servtech.com (8.8.5/8.8.5) with SMTP id KAA01245; Fri, 4 Apr 1997 10:54:44 -0500 (EST) Message-ID: <33452443.41C67EA6@servtech.com> Date: Fri, 04 Apr 1997 10:54:43 -0500 From: Shawn Carey X-Mailer: Mozilla 3.0 (X11; I; FreeBSD 2.2.1-RELEASE i386) MIME-Version: 1.0 To: John Polstra CC: dyson@freebsd.org, hackers@freebsd.org Subject: Re: Anyone else seen this? References: <199703270427.XAA04344@dyson.iquest.net> <333AA089.41C67EA6@servtech.com> <199703280313.TAA28286@austin.polstra.com> <333B5B51.41C67EA6@servtech.com> <199703282034.MAA04175@austin.polstra.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk John Polstra wrote: > > > If I run the same executable under GDB, it runs for 10 seconds at the > > most and then GDB kills it before reporting "Process terminated due to > > text file modification". No breakpoints, no nuthin. Not even a > > .gdbinit. > > > > Now, if I link with -static, the exectuable retains its timestamp, even > > whe run under GDB, and GDB lets it run peacefully. > > It's worth noting that there are hooks between gdb and the dynamic > linker. The dynamic linker can tell whether the program is being run > [...] Indeed it is. I just found out today that if I set a breakpoint on the static binary *before* running the program, I get the same result as when the binary is lunk dynamically. Sorry for the confusion on my part... Is there anything I can do to help solve this problem? If someone gave me a probable starting point, I'd be more than happy to thrash around with it... -Shawn From owner-freebsd-hackers Fri Apr 4 07:56:03 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA15370 for hackers-outgoing; Fri, 4 Apr 1997 07:56:03 -0800 (PST) Received: from boulder.syr.servtech.com (boulder.syr.servtech.com [206.106.144.94]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA15312 for ; Fri, 4 Apr 1997 07:55:56 -0800 (PST) Received: from boulder.syr.servtech.com (localhost [127.0.0.1]) by boulder.syr.servtech.com (8.8.5/8.8.5) with SMTP id LAA01258 for ; Fri, 4 Apr 1997 11:00:57 -0500 (EST) Message-ID: <334525B9.167EB0E7@servtech.com> Date: Fri, 04 Apr 1997 11:00:57 -0500 From: Shawn Carey X-Mailer: Mozilla 3.0 (X11; I; FreeBSD 2.2.1-RELEASE i386) MIME-Version: 1.0 To: hackers@freebsd.org Subject: Re: FreeBSD Elf-Kit and dynamic loading References: <5i0j1d$jtk@news.itfs.nsk.su> <19970403191209.52889@keltia.freenix.fr> <5i2k5h$4jb@news.itfs.nsk.su> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Nickolay N. Dudorov wrote: > Ollivier Robert wrote: > [...] > > Are you sure you have the right naming for the libraries ? > > Yes - it turns out that because I build and install elf-libc > from (copy of) FreeBSD-2.2 sources there was no libc.so --> libc.so.3.0 > link. After I make all libXX.so --> libXX.so.M.N links my test > programs builds as dynamic. > As an ELF newbie, it seems curious to me that the ELF linker requires a symlink to the shared lib that does not have the version in its name. Does ELF keep track of the shared lib versions? -Shawn From owner-freebsd-hackers Fri Apr 4 08:27:57 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA17515 for hackers-outgoing; Fri, 4 Apr 1997 08:27:57 -0800 (PST) Received: from pat.idt.unit.no (0@pat.idt.unit.no [129.241.103.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA17508 for ; Fri, 4 Apr 1997 08:27:53 -0800 (PST) Received: from idt.unit.no (26850@kamelia.idt.unit.no [129.241.111.27]) by pat.idt.unit.no (8.8.5/8.8.5) with ESMTP id SAA07632; Fri, 4 Apr 1997 18:27:20 +0200 (MET DST) Message-Id: <199704041627.SAA07632@pat.idt.unit.no> To: dfr@nlsystems.com Cc: dg@root.com, ponds!rivers@dg-rtp.dg.com, freebsd-hackers@FreeBSD.ORG Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) In-Reply-To: Your message of "Fri, 4 Apr 1997 16:43:08 +0100 (BST)" References: X-Mailer: Mew version 1.06 on Emacs 19.34.1 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Date: Fri, 04 Apr 1997 18:27:18 +0200 From: Tor Egge Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > I tried testing your fix this morning and the 0xdeadb stuff just caused > vget to fault a couple of minutes into my test (simultaneous rm -rf > largetree and cvs co src, both remote). A different part of the Lite/2 commit removed an explicit check for the 0xdeadb stuff in vget. Perhaps the kludge is still needed there. > > This problem really has little to do with nfs_inactive. What is > happening is a race between vgone and vget which would normally be solved > by the vnode locks. Since NFS doesn't have vnode locks, the race > happens. In my opinion, the problem has to do with marking objects as free and reusable while they are still in use. When the vnode is on the freelist, with 0 as v_usecount, and the vp->v_interlock is not locked, how should getnewvnode decide that the vnode is not reusable ? VOP_ISLOCKED ? Not currently used in getnewvnode nor in vget. > I am most of the way there in implementing the right solution for NFS > which is to used shared locks for NFS; vgone can then use the lock > manager to wait for all the shared locks to drain before recycling the > vnode. Good. - Tor Egge From owner-freebsd-hackers Fri Apr 4 08:44:32 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA18592 for hackers-outgoing; Fri, 4 Apr 1997 08:44:32 -0800 (PST) Received: from nlsystems.com (nlsys.demon.co.uk [158.152.125.33]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA18575 for ; Fri, 4 Apr 1997 08:44:24 -0800 (PST) Received: from herring.nlsystems.com (herring.nlsystems.com [10.0.0.2]) by nlsystems.com (8.8.5/8.8.5) with SMTP id RAA11989; Fri, 4 Apr 1997 17:43:38 +0100 (BST) Date: Fri, 4 Apr 1997 17:43:38 +0100 (BST) From: Doug Rabson To: Tor Egge cc: dg@root.com, ponds!rivers@dg-rtp.dg.com, freebsd-hackers@FreeBSD.ORG Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) In-Reply-To: <199704041627.SAA07632@pat.idt.unit.no> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Fri, 4 Apr 1997, Tor Egge wrote: > > I tried testing your fix this morning and the 0xdeadb stuff just caused > > vget to fault a couple of minutes into my test (simultaneous rm -rf > > largetree and cvs co src, both remote). > > A different part of the Lite/2 commit removed an explicit check for > the 0xdeadb stuff in vget. Perhaps the kludge is still needed there. > > > > > This problem really has little to do with nfs_inactive. What is > > happening is a race between vgone and vget which would normally be solved > > by the vnode locks. Since NFS doesn't have vnode locks, the race > > happens. > > In my opinion, the problem has to do with marking objects as free and > reusable while they are still in use. > > When the vnode is on the freelist, with 0 as v_usecount, and the > vp->v_interlock is not locked, how should getnewvnode decide that the > vnode is not reusable ? VOP_ISLOCKED ? Not currently used in > getnewvnode nor in vget. The vnode in question is being recycled from VT_NFS to VT_UFS. Getnewvnode picks a vnode of the free list and calls vgone to extract it from the previous owner. This calls VOP_LOCK(.., LK_DRAIN, ..) which is intended to block until the vnode is inactive. For vnodes which are not being recycled but just picked up from the cache, there is no race but still either vget or its caller will lock the vnode before using it for anything important. > > > I am most of the way there in implementing the right solution for NFS > > which is to used shared locks for NFS; vgone can then use the lock > > manager to wait for all the shared locks to drain before recycling the > > vnode. > > Good. My test is running... Lets hope it passes this time :-). -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 181 951 1891 From owner-freebsd-hackers Fri Apr 4 08:48:05 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA18914 for hackers-outgoing; Fri, 4 Apr 1997 08:48:05 -0800 (PST) Received: from labinfo.iet.unipi.it (labinfo.iet.unipi.it [131.114.9.5]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id IAA18893 for ; Fri, 4 Apr 1997 08:48:00 -0800 (PST) Received: from localhost (luigi@localhost) by labinfo.iet.unipi.it (8.6.5/8.6.5) id RAA03521; Fri, 4 Apr 1997 17:56:24 +0200 From: Luigi Rizzo Message-Id: <199704041556.RAA03521@labinfo.iet.unipi.it> Subject: Re: Does de driver do 100MBIT Full Duplex? To: dennis@etinc.com (dennis) Date: Fri, 4 Apr 1997 17:56:24 +0200 (MET DST) Cc: msmith@atrad.adelaide.edu.au, hackers@freebsd.org In-Reply-To: <3.0.32.19970404095015.006b3514@etinc.com> from "dennis" at Apr 4, 97 09:49:59 am X-Mailer: ELM [version 2.4 PL23] Content-Type: text Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > The "expectation" is that things that worked before will still work with a new > release, and that there may be problems with new stuff. When there are > problems with old stuff, you have a real perception problem with your product. > There has been much talk about get publicity; getting corporate users to > use FreeBSD, but until you get to a point where loading a new release is much > less than a crapshoot It's just not going to happen on a widespread basis. It is quite explicit that 2.1.7 is the stable branch and 2.2 is the experimental one. By choosing 2.2 you must know you are taking some risks. Unfortunately 2.2 has quite a few interesting features (such as WORM support, better PCI support, more supported hw etc.); this might be tempting for people and make them forget what is supposed to be stable and what is not. As for extensive testings: troubles often come out in a production environment, not in quick tests done between ALPHA and BETA. And I (as probably others) did not feel like putting a snap on a production machine (with some exceptions; our web server is running a pre 2.0.5 version, but that was just a lucky experiment and the server did not exist before). So people who really want to help should IMHO move to 2.2 and test the release as much as possible, so that the next one can have bugs fixed. Luigi -----------------------------+-------------------------------------- Luigi Rizzo | Dip. di Ingegneria dell'Informazione email: luigi@iet.unipi.it | Universita' di Pisa tel: +39-50-568533 | via Diotisalvi 2, 56126 PISA (Italy) fax: +39-50-568522 | http://www.iet.unipi.it/~luigi/ _____________________________|______________________________________ From owner-freebsd-hackers Fri Apr 4 08:52:28 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA19160 for hackers-outgoing; Fri, 4 Apr 1997 08:52:28 -0800 (PST) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA19152 for ; Fri, 4 Apr 1997 08:52:25 -0800 (PST) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id JAA28254; Fri, 4 Apr 1997 09:52:16 -0700 (MST) Date: Fri, 4 Apr 1997 09:52:16 -0700 (MST) Message-Id: <199704041652.JAA28254@rocky.mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: "Jordan K. Hubbard" Cc: Darren Reed , proff@suburbia.net, hackers@freebsd.org Subject: Re: IP Filter in -current. In-Reply-To: <18999.860142214@time.cdrom.com> References: <199704040736.RAA23038@plum.cyber.com.au> <18999.860142214@time.cdrom.com> X-Mailer: VM 6.22 under 19.15 XEmacs Lucid Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > Someone should give this guy a labotamy if he's the one who came up > > with the "include every other include file in this include file" idea. > > "Labotomy", and I think Bruce is too valuable to even consider such a > radical treatment for him - how about trying direct debate with him > over this issue instead? ;-) The way I read it, Darren agrees with Bruce and disagrees that what he's doing is a 'bad thing'. Nate From owner-freebsd-hackers Fri Apr 4 09:19:03 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA20432 for hackers-outgoing; Fri, 4 Apr 1997 09:19:03 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA20425 for ; Fri, 4 Apr 1997 09:18:58 -0800 (PST) Received: from dialup-usr11.etinc.com (dialup-usr11.etinc.com [204.141.95.132]) by etinc.com (8.8.3/8.6.9) with SMTP id MAA10176; Fri, 4 Apr 1997 12:23:13 -0500 (EST) Message-Id: <3.0.32.19970404121523.00b3375c@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Fri, 04 Apr 1997 12:15:27 -0500 To: Luigi Rizzo From: dennis Subject: Re: NFS problems with 2.2.1R Cc: dfr@nlsystems.com, hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 03:37 PM 4/4/97 +0200, Luigi Rizzo wrote: >> >When the system is hung, is there any net traffic? I just fixed a bug in >> >> No, there is no network traffic. I have a LAN monitor on the network and the >> server sends a window of frames that do not get ack'd by the hung machine. >> >> As this is occurring with the boot floppy I'm not able to do much diagnosis :( > >you can try to ping the system and see if it at least responds. The machine is DEAD! No ping response...Hard disk light on solid. Dennis > > Luigi > > From owner-freebsd-hackers Fri Apr 4 09:21:45 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA20559 for hackers-outgoing; Fri, 4 Apr 1997 09:21:45 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA20554 for ; Fri, 4 Apr 1997 09:21:39 -0800 (PST) Received: from dialup-usr11.etinc.com (dialup-usr11.etinc.com [204.141.95.132]) by etinc.com (8.8.3/8.6.9) with SMTP id MAA10192; Fri, 4 Apr 1997 12:24:53 -0500 (EST) Message-Id: <3.0.32.19970404121704.00b3b858@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Fri, 04 Apr 1997 12:17:07 -0500 To: Luigi Rizzo From: dennis Subject: Re: Does de driver do 100MBIT Full Duplex? Cc: msmith@atrad.adelaide.edu.au, hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 05:56 PM 4/4/97 +0200, Luigi Rizzo wrote: >> The "expectation" is that things that worked before will still work with a new >> release, and that there may be problems with new stuff. When there are >> problems with old stuff, you have a real perception problem with your product. >> There has been much talk about get publicity; getting corporate users to >> use FreeBSD, but until you get to a point where loading a new release is much >> less than a crapshoot It's just not going to happen on a widespread basis. > >It is quite explicit that 2.1.7 is the stable branch and 2.2 is >the experimental one. By choosing 2.2 you must know you are taking >some risks. Ho, ho hoooooold on now! When the -RELEASE is added to 2.2 it becomes the stable branch. Stable, as i understand it, is a branch of the latest release. Dennis From owner-freebsd-hackers Fri Apr 4 09:23:38 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA20670 for hackers-outgoing; Fri, 4 Apr 1997 09:23:38 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA20663 for ; Fri, 4 Apr 1997 09:23:32 -0800 (PST) Received: from dialup-usr11.etinc.com (dialup-usr11.etinc.com [204.141.95.132]) by etinc.com (8.8.3/8.6.9) with SMTP id MAA10218; Fri, 4 Apr 1997 12:28:17 -0500 (EST) Message-Id: <3.0.32.19970404122027.00b32834@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Fri, 04 Apr 1997 12:20:30 -0500 To: Michael Smith From: dennis Subject: FTP install Cc: hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 02:07 AM 4/5/97 +0930, Michael Smith wrote: >dennis stands accused of saying: >> >> You're the first one to tell me that it actually does work....What is your >> FTP server? >> Freebsd or some other animal? If freebsd, what version? > >Our build machine was obviously running 2.2, the GP workstation a >2.2-GAMMA-vintage from early this year (maybe Jan or so) and the >notebooks were running an old 2.2 (october or so last year), but were >upgraded as part of the testing. > >It's perhaps worth mentioning that we decided to standardise on FTP, >rather than NFS, for the install as it's somewhat faster and requires >less setup on the source machine, particularly if the source is a >notebook that you've just temporarily thrown onto a customer's >network. We did, however, do several dozen NFS installs before coming >to this decision. I cant get a local FTP load to work (this would be just dandy for us). It doesnt find the directory on a non-anonymous FTP to the server. Whats the trick if your using the "other URL" option? Dennis From owner-freebsd-hackers Fri Apr 4 09:39:04 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA21394 for hackers-outgoing; Fri, 4 Apr 1997 09:39:04 -0800 (PST) Received: from haldjas.folklore.ee (Haldjas.folklore.ee [193.40.6.121]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA21362 for ; Fri, 4 Apr 1997 09:38:51 -0800 (PST) Received: from localhost (narvi@localhost) by haldjas.folklore.ee (8.8.4/8.8.4) with SMTP id UAA11850; Fri, 4 Apr 1997 20:40:31 +0300 (EEST) Date: Fri, 4 Apr 1997 20:40:30 +0300 (EEST) From: Narvi To: "Jordan K. Hubbard" cc: hackers@freebsd.org Subject: Re: Better late than never... The FreeBSD News! In-Reply-To: <1848.860160949@time.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Fri, 4 Apr 1997, Jordan K. Hubbard wrote: > OK folks, sorry for the delay on this.. As I intimated last week, the > "FreeBSD News" newsletter publication thingy is now past the early > planning & layout stages and into the "who will write the articles?" > stage. > > We're not looking for anything super ambitious in the way of articles > here, just something which describes an interesting use you're making > of FreeBSD or aims itself at educating the new user about some aspect > of FreeBSD's infrastructure (Davidg talking about wcarchive, mpp > talking about GNATs and how to use it, jmb talking about the mailing > lists, etc). Generic "how to configure your " sorts of tutorials > would be nice too. Length should be 300-400 words, though shorter > "quick tip" articles are always welcome too. We can make the panels > any size we want. :) > [snip] > > o The FreeBSD mailing lists and how to make them work > for you. [jmb? :)] > > o The FreeBSD web site - an overview of services offered. > > o The FreeBSD problem reporting mechanism - how to submit reports and > how to search for them. [ mpp is signed up for this ]. > > o Staying up to date with {CVSup | CTM} [jdp? phk? :)] > > o Setting up an effective X desktop environment. > > o How FreeBSD is developed [subtitle: What are core team > members and committers and how does one become one?]. > > o Installing and managing your Apache web server. > > o Installing and managing SAMBA for Win95/NT interoperability > > o Running Linux binaries under FreeBSD - what to do and where > to get the libraries. > > o A comparison between FreeBSD and Linux [special note: This topic > reserved for someone with cojones the size of coconuts]. > Well, someone could write about /etc/make.conf... (No, this is not a promise). Sander [snip again] > > > All correspondence to me please (and watch your cc lines! I don't > think all of -hackers wants to read your first draft :-). > > Thanks! > > Jordan > From owner-freebsd-hackers Fri Apr 4 09:41:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA21702 for hackers-outgoing; Fri, 4 Apr 1997 09:41:11 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id JAA21683 for ; Fri, 4 Apr 1997 09:41:04 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id KAA19463; Fri, 4 Apr 1997 10:22:15 -0700 From: Terry Lambert Message-Id: <199704041722.KAA19463@phaeton.artisoft.com> Subject: Re: IP Filter in -current. To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Fri, 4 Apr 1997 10:22:15 -0700 (MST) Cc: darrenr@cyber.com.au, proff@suburbia.net, hackers@freebsd.org In-Reply-To: <18999.860142214@time.cdrom.com> from "Jordan K. Hubbard" at Apr 4, 97 00:23:34 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > Someone should give this guy a labotamy if he's the one who came up > > with the "include every other include file in this include file" idea. > > "Labotomy", and I think Bruce is too valuable to even consider such a > radical treatment for him - how about trying direct debate with him > over this issue instead? ;-) Yes, don't take away his "lab". Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Fri Apr 4 09:44:22 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA21960 for hackers-outgoing; Fri, 4 Apr 1997 09:44:22 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id JAA21952 for ; Fri, 4 Apr 1997 09:44:15 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id KAA19477; Fri, 4 Apr 1997 10:26:19 -0700 From: Terry Lambert Message-Id: <199704041726.KAA19477@phaeton.artisoft.com> Subject: Re: new malloc To: dg@root.com Date: Fri, 4 Apr 1997 10:26:18 -0700 (MST) Cc: borjam@we.lc.ehu.es, freebsd-hackers@freebsd.org In-Reply-To: <199704041027.CAA10928@root.com> from "David Greenman" at Apr 4, 97 02:27:01 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > Of course, the same idea cn be applied adding a MRU policy to the > >allocation routine. > > Both Poul-Henning's and the FreeBSD kernel's malloc allocate and insert > chunks at the head of the free queue - thus they already are LIFO. I assume this is for things with zero locality of reference (ie: no second chancing, etc.)? Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Fri Apr 4 09:56:17 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA22781 for hackers-outgoing; Fri, 4 Apr 1997 09:56:17 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id JAA22774 for ; Fri, 4 Apr 1997 09:56:12 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id KAA19497; Fri, 4 Apr 1997 10:39:07 -0700 From: Terry Lambert Message-Id: <199704041739.KAA19497@phaeton.artisoft.com> Subject: Re: apache like preforking apps and high loads To: james@wgold.demon.co.uk (James Mansion) Date: Fri, 4 Apr 1997 10:39:06 -0700 (MST) Cc: FreeBSD-hackers@freebsd.org In-Reply-To: <3344D3D3.618A@wgold.demon.co.uk> from "James Mansion" at Apr 4, 97 11:11:31 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Bakul Shah wrote: > > Fairness is probably not an issue when an app. consists of a number > > of anonymous servers but in general one would want to make sure that > > if N processes are waiting on accept() on the same socket, no one > > process is starved of accepting. How do you ensure that? > > > > Why would you necessarily want this, apart from aesthetics? > > I don't think this behaviour is mandated anywhere. > > In any case, presumably (hopefully) the wakeup goes to the > process with the highest scheduling priority. If the scheduler > does dynamic adjustment based on CPU time consumed 'recently' then > waiters will be favoured over processes that ran recently, and you'll > get th eeffect that you want. In point of fact, you often *do* want this, unless you are attempting to load balance work-to-do between processors. One of the most clever things I came up with for the NetWare for UNIX product (Steve and Marty implemented it, not me) was the idea that the recv requests to the streams NCP MUX should be LIFO'ed. This increases the probability that the next work-to-do engine to which you give work will have all of its pages already in core and in its map (the biggest danger here is the loss of transaction data pages, which must be, by definition, per engine). If you were attempting to balance between processors, then you really want two effective queues, and you select which "hot" queue head by managing the insertion order by processor-by-time (the top is the hottest engine on processor A, the next is the hottest on processor B, the next is the second hottest on processor A, etc.. This assumes a high or fixed CPU affinity for the engines). Because you can control whose call returns immediately and whose doesn't, you can effect a private scheduling policy using a MUX in a work-to-do process model. This is one of the reasons NetWare for UNIX beat native NetWare by up to 12% on most benchmarks when running on the exact same hardware, even after you take UnixWare's high latency stack (mostly because of the ODI drivers) into account. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Fri Apr 4 10:13:48 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA23489 for hackers-outgoing; Fri, 4 Apr 1997 10:13:48 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id KAA23481 for ; Fri, 4 Apr 1997 10:13:45 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id KAA19541; Fri, 4 Apr 1997 10:54:19 -0700 From: Terry Lambert Message-Id: <199704041754.KAA19541@phaeton.artisoft.com> Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) To: Tor.Egge@idi.ntnu.no (Tor Egge) Date: Fri, 4 Apr 1997 10:54:19 -0700 (MST) Cc: dfr@nlsystems.com, dg@root.com, ponds!rivers@dg-rtp.dg.com, freebsd-hackers@FreeBSD.ORG In-Reply-To: <199704041627.SAA07632@pat.idt.unit.no> from "Tor Egge" at Apr 4, 97 06:27:18 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > When the vnode is on the freelist, with 0 as v_usecount, and the > vp->v_interlock is not locked, how should getnewvnode decide that the > vnode is not reusable ? VOP_ISLOCKED ? Not currently used in > getnewvnode nor in vget. This is all terrible code, IMO. It's both confusing, and totally lacking in uniformity of application. That said, you can test to see if the VFS the vnode references is deadfs. Vnodes which are invalid have deadfs as their VFS. The problem with this is that the deadfs is an indicator of invalidation of the vnode, not the fact that the vnode does not have refernces to it (deadfs is specifically built to deal with the now-invalid references by causing VOP calls on the vnode to fail). You should join the discussion on the -current list regarding this issue. Historically, I've seen the vclean() and the locking which must be duplicated per FS as the primary evil. I think maybe that the vnodes need to be more carefully reference counted and released on a per FS basis, and deadfs needs to go away. This has a number of implications for kernel file I/O: if you take away the ability to invalidate a vnode (by getting rid of deadfs, and by eventually removing the ability to migrate a vnode between FS types), then you need another mechanism to serve as a "handle" for the vnode so that it can still be invalidated. I'v nominated the system open file table, and that's about where the discussion currently stands: the merits and demerits of doing all this work, and what we expect to get out of it. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Fri Apr 4 10:14:56 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA23581 for hackers-outgoing; Fri, 4 Apr 1997 10:14:56 -0800 (PST) Received: from mangle.dcs.qmw.ac.uk (mangle-main.dcs.qmw.ac.uk [138.37.88.250]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA23576 for ; Fri, 4 Apr 1997 10:14:53 -0800 (PST) Received: from md@ruby [138.37.88.139]; by mangle.dcs.qmw.ac.uk (8.8.5/8.8.5/S-4.0) with ESMTP; id TAA25879; Fri, 4 Apr 1997 19:14:41 +0100 (BST) Received: from md@localhost; by ruby.dcs.qmw.ac.uk (8.8.4/8.8.4/C-3.2); id TAA07859; Fri, 4 Apr 1997 19:14:39 +0100 (BST) Received: from Messages.8.5.N.CUILIB.3.45.SNAP.NOT.LINKED.ruby.cs.qmw.ac.uk.sun4.41 via MS.5.6.ruby.cs.qmw.ac.uk.sun4_41; Fri, 4 Apr 1997 19:14:39 +0100 (BST) Message-ID: <8nFIIDS_9JY1FiIzR5@dcs.qmw.ac.uk> Date: Fri, 4 Apr 1997 19:14:39 +0100 (BST) From: Mark Dawson To: freebsd-hackers@freebsd.org Subject: NetBSD's if_de.c Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk NetBSD's de driver (1.83 1997/03/25 21:12:17) has a fix for 21140A hangup's which is absent in FreeBSD's version... in tulip_pci_attach(): if (chipid == TULIP_21140A && revinfo <= 0x22) sc->tulip_features |= TULIP_HAVE_RXBUGGY; in tulip_intr_handler(): /* * Pass 2.[012] of the 21140A-A[CDE] may hang and/or corrupt data * on receive overflows. */ if ((misses & 0x0FFE0000) && (sc->tulip_features & TULIP_HAVE_RXBUGGY)) { /* * Stop the receiver process and spin until it's stopped. * Tell rx_intr to drop the packets it dequeues. */ TULIP_CSR_WRITE(sc, csr_command, sc->tulip_cmdmode & ~TULIP_CMD_RXRUN); while ((TULIP_CSR_READ(sc, csr_status) & TULIP_STS_RXSTOPPED) == 0) ; TULIP_CSR_WRITE(sc, csr_status, TULIP_STS_RXSTOPPED); sc->tulip_flags |= TULIP_RXBAD; } Mark From owner-freebsd-hackers Fri Apr 4 10:20:45 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA24104 for hackers-outgoing; Fri, 4 Apr 1997 10:20:45 -0800 (PST) Received: from lister.bogon.net (500@gw.bogon.net [204.137.132.49]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA24099 for ; Fri, 4 Apr 1997 10:20:41 -0800 (PST) Received: (from wes@localhost) by lister.bogon.net (8.8.5/8.8.3) id KAA19125 for hackers@freebsd.org; Fri, 4 Apr 1997 10:20:27 -0800 (PST) From: Wes Santee Message-Id: <199704041820.KAA19125@lister.bogon.net> Subject: Uses for divert sockets? To: hackers@freebsd.org Date: Fri, 4 Apr 1997 10:20:26 -0800 (PST) X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk 'Lo all. I asked this on questions but never received an answer. Perhaps folks here would know? If so, please include me in replies as I'm not quite sophisticated enough with the source to follow this list (but I'm working on it!). :) My apologies if this falls under the 'abuse' category. The new ipdivert sockets in 2.2 sound great, but I'm not quite sure just what I can accomplish with them outside of address translation. Is it just for packet altering, or can entire connections be redirected? For example, let's say I want some incomming connections to port 25 to be diverted to another "special" MTA running on another port (to deal with known spam sites "properly", for instance). Is it possible using divert and ipfw to do this? Or is divert pretty much only for dealing with raw IP packets that get injected back into the stream to end up at their intended destination? If divert sockets aren't capable of doing this, is there any other mechanism available to perform these kinds of tasks? Cheers, -- ( Wes Santee PGP: e-mail w/Subject: "Send PGP Key" ) ( mailto:wes@bogon.net ) From owner-freebsd-hackers Fri Apr 4 10:35:29 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id KAA24834 for hackers-outgoing; Fri, 4 Apr 1997 10:35:29 -0800 (PST) Received: from austin.polstra.com (austin.polstra.com [206.213.73.10]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA24829 for ; Fri, 4 Apr 1997 10:35:24 -0800 (PST) Received: from austin.polstra.com (jdp@localhost) by austin.polstra.com (8.8.5/8.8.5) with ESMTP id KAA24555; Fri, 4 Apr 1997 10:35:08 -0800 (PST) Message-Id: <199704041835.KAA24555@austin.polstra.com> To: smc@servtech.com Subject: Re: FreeBSD Elf-Kit and dynamic loading Newsgroups: polstra.freebsd.hackers In-Reply-To: <334525B9.167EB0E7@servtech.com> References: <5i0j1d$jtk@news.itfs.nsk.su> <19970403191209.52889@keltia.freenix.fr> <5i2k5h$4jb@news.itfs.nsk.su> <334525B9.167EB0E7@servtech.com> Organization: Polstra & Co., Seattle, WA Cc: hackers@freebsd.org Date: Fri, 04 Apr 1997 10:35:07 -0800 From: John Polstra Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In article <334525B9.167EB0E7@servtech.com>, Shawn Carey wrote: > Nickolay N. Dudorov wrote: > > Yes - it turns out that because I build and install elf-libc > > from (copy of) FreeBSD-2.2 sources there was no libc.so --> libc.so.3.0 > > link. After I make all libXX.so --> libXX.so.M.N links my test > > programs builds as dynamic. > > > > As an ELF newbie, it seems curious to me that the ELF linker requires a > symlink to the shared lib that does not have the version in its name. > Does ELF keep track of the shared lib versions? It's kind of a screwy system, but it's the way ELF does things. Here's how it works. Say you've got three versions of libc: libc.so.3 (the newest) libc.so.2 libc.so.1 (the oldest) (ELF doesn't use minor version numbers.) You also arrange to have a symlink pointing to the newest version: libc.so -> libc.so.3 The static linker "ld" does not do any searching based on version numbers. It always opens its libraries under the unversioned name, e.g., "libc.so". If the link is set up right, that means that "ld" always uses the newest version of the library. Recorded inside each shared library is its fully-versioned name, e.g., "libc.so.3". That name is stored into the executable file that "ld" produces, as a "shared object dependency." When you run the program, the dynamic linker uses the shared object dependencies to form a list of libraries that it has to open and link in. So in this case, it will always look for "libc.so.3", and won't use the symbolic link. Now suppose you add a new version of the library, "libc.so.4". You change the symlink "libc.so" to point to the new version. Things work out the way they should. Existing executables that were linked against libc.so.3 still will use that same library, because the versioned name is recorded in them. When you build new programs, though, they'll use the newest library, because that's where the unversioned symlink points now. I didn't invent it. I just implemented it. :-) -- John Polstra jdp@polstra.com John D. Polstra & Co., Inc. Seattle, Washington USA "Self-knowledge is always bad news." -- John Barth From owner-freebsd-hackers Fri Apr 4 12:29:02 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA01750 for hackers-outgoing; Fri, 4 Apr 1997 12:29:02 -0800 (PST) Received: from sumatra.americantv.com (sumatra.americantv.com [199.184.181.250]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA01741 for ; Fri, 4 Apr 1997 12:28:57 -0800 (PST) Received: from right.PCS (right.pcs. [148.105.10.31]) by sumatra.americantv.com (8.7.6/8.7.3) with ESMTP id OAA05328; Fri, 4 Apr 1997 14:35:31 -0600 (CST) Received: (jlemon@localhost) by right.PCS (8.6.13/8.6.4) id OAA25445; Fri, 4 Apr 1997 14:29:57 -0600 Message-ID: <19970404142957.41778@right.PCS> Date: Fri, 4 Apr 1997 14:29:57 -0600 From: Jonathan Lemon To: John Polstra Cc: smc@servtech.com, hackers@FreeBSD.ORG Subject: Re: FreeBSD Elf-Kit and dynamic loading References: <5i0j1d$jtk@news.itfs.nsk.su> <19970403191209.52889@keltia.freenix.fr> <5i2k5h$4jb@news.itfs.nsk.su> <334525B9.167EB0E7@servtech.com> <199704041835.KAA24555@austin.polstra.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.61.1 In-Reply-To: <199704041835.KAA24555@austin.polstra.com>; from John Polstra on Apr 04, 1997 at 10:35:07AM -0800 Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Apr 04, 1997 at 10:35:07AM -0800, John Polstra wrote: > > Now suppose you add a new version of the library, "libc.so.4". You > change the symlink "libc.so" to point to the new version. Things > work out the way they should. Existing executables that were linked > against libc.so.3 still will use that same library, because the > versioned name is recorded in them. When you build new programs, > though, they'll use the newest library, because that's where the > unversioned symlink points now. > > I didn't invent it. I just implemented it. :-) My first reaction: "Oh, yuck. You mean I have to remember to update a symlink whenever I put in a new version of the shared libraries?!" My second reaction: "Oh, good. Now I have a way of dealing with certain l^Husers who insist on having a libc.so.261 on some systems." I guess there are benefits to this approach. -- Jonathan From owner-freebsd-hackers Fri Apr 4 13:12:59 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA03727 for hackers-outgoing; Fri, 4 Apr 1997 13:12:59 -0800 (PST) Received: from mail12.digital.com (mail12.digital.com [192.208.46.20]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA03718 for ; Fri, 4 Apr 1997 13:12:54 -0800 (PST) Received: from muggsy.lkg.dec.com by mail12.digital.com (8.7.5/UNX 1.5/1.0/WV) id QAA31106; Fri, 4 Apr 1997 16:04:14 -0500 (EST) Received: from whydos.lkg.dec.com by muggsy.lkg.dec.com (5.65/DEC-Ultrix/4.3) with SMTP id AA20026; Fri, 4 Apr 1997 16:04:06 -0500 Received: from localhost.lkg.dec.com (localhost.lkg.dec.com [127.0.0.1]) by whydos.lkg.dec.com (8.6.12/8.6.12) with SMTP id QAA28898 for ; Fri, 4 Apr 1997 16:13:44 GMT Message-Id: <199704041613.QAA28898@whydos.lkg.dec.com> X-Authentication-Warning: whydos.lkg.dec.com: Host localhost.lkg.dec.com didn't use HELO protocol X-Mailer: exmh version 1.6.5 12/11/95 To: hackers@freebsd.org Subject: How to get the latest de driver under FreeBSD... Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 04 Apr 1997 16:13:43 +0000 From: Matt Thomas Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I just wrote a web page on how to upgrade/update your FreeBSD so it can run the latest de driver (which is in NetBSD). It's at http://www.3am-software.com/ifmedia.html . Note that the NetBSD is in much better shape than the current FreeBSD driver. (www.3am-software.com is now 206.107.21.156 if you have problems accessing that page) -- Matt Thomas Internet: matt@3am-software.com 3am Software Foundry WWW URL: http://www.3am-software.com/bio/matt.html Westford, MA Disclaimer: I disavow all knowledge of this message From owner-freebsd-hackers Fri Apr 4 13:21:37 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA04393 for hackers-outgoing; Fri, 4 Apr 1997 13:21:37 -0800 (PST) Received: from sax.sax.de (sax.sax.de [193.175.26.33]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id NAA04385 for ; Fri, 4 Apr 1997 13:21:33 -0800 (PST) Received: (from uucp@localhost) by sax.sax.de (8.6.12/8.6.12-s1) with UUCP id XAA02482 for hackers@freebsd.org; Fri, 4 Apr 1997 23:21:25 +0200 Received: (from j@localhost) by uriah.heep.sax.de (8.8.5/8.8.5) id XAA05129; Fri, 4 Apr 1997 23:03:43 +0200 (MET DST) Message-ID: <19970404230343.ME62630@uriah.heep.sax.de> Date: Fri, 4 Apr 1997 23:03:43 +0200 From: j@uriah.heep.sax.de (J Wunsch) To: hackers@freebsd.org Subject: Re: FTP install References: <3.0.32.19970404122027.00b32834@etinc.com> X-Mailer: Mutt 0.60_p2-3,5,8-9 Mime-Version: 1.0 X-Phone: +49-351-2012 669 X-PGP-Fingerprint: DC 47 E6 E4 FF A6 E9 8F 93 21 E0 7D F9 12 D6 4E Reply-To: joerg_wunsch@uriah.heep.sax.de (Joerg Wunsch) In-Reply-To: <3.0.32.19970404122027.00b32834@etinc.com>; from dennis on Apr 4, 1997 12:20:30 -0500 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk As dennis wrote: > I cant get a local FTP load to work (this would be just dandy for us). > It doesnt find the directory on a non-anonymous FTP to the server. > Whats the trick if your using the "other URL" option? The directory is (as is documented) relative to the home directory of the non-anon user. Also, i think sysinstall looks into a subdirectory identical to the release name relative to the given URL (so you can just say ``.../pub/FreeBSD/'' as the default, and it will automatic- ally pick up the right release from the official mirrors), unless you specify the release name in the options menu as "none". If you turn on debugging, you can easily watch the looked up files on the second screen. It's not all that magic... -- cheers, J"org joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE Never trust an operating system you don't have sources for. ;-) From owner-freebsd-hackers Fri Apr 4 13:39:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA05376 for hackers-outgoing; Fri, 4 Apr 1997 13:39:11 -0800 (PST) Received: from relay-7.mail.demon.net (relay-7.mail.demon.net [194.217.242.9]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id NAA05371 for ; Fri, 4 Apr 1997 13:39:08 -0800 (PST) Received: from awfulhak.demon.co.uk ([158.152.17.1]) by relay-6.mail.demon.net id aa0627558; 4 Apr 97 22:35 BST Received: from awfulhak.demon.co.uk (localhost.lan.awfulhak.org [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id WAA15753; Fri, 4 Apr 1997 22:15:03 +0100 (BST) Message-Id: <199704042115.WAA15753@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: Robert Withrow cc: hackers@freebsd.org Subject: Re: Suggested change to rc In-reply-to: Your message of "Fri, 04 Apr 1997 08:29:14 CDT." <199704041329.IAA06810@tuva.engeast.baynetworks.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Fri, 04 Apr 1997 22:15:03 +0100 From: Brian Somers Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > In order to have amd get all of its information from nis, the following > change needs to be made in rc. This allows amdflags to be set like this: > [.....] > ! amd -p ${amdflags} > /var/run/amd.pid [.....] > ! eval amd -p ${amdflags} > /var/run/amd.pid With a sentence warning that the line is "eval"d in /etc/sysconfig, I'd say it's fine. > -- > Robert Withrow -- (+1 508 436 8256) > BWithrow@BayNetworks.com -- Brian , Don't _EVER_ lose your sense of humour.... From owner-freebsd-hackers Fri Apr 4 14:22:44 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA07404 for hackers-outgoing; Fri, 4 Apr 1997 14:22:44 -0800 (PST) Received: from pat.idt.unit.no (0@pat.idt.unit.no [129.241.103.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA07394 for ; Fri, 4 Apr 1997 14:22:34 -0800 (PST) Received: from idt.unit.no (26850@kamelia.idt.unit.no [129.241.111.27]) by pat.idt.unit.no (8.8.5/8.8.5) with ESMTP id AAA14484; Sat, 5 Apr 1997 00:22:12 +0200 (MET DST) Message-Id: <199704042222.AAA14484@pat.idt.unit.no> To: dfr@nlsystems.com Cc: dg@root.com, ponds!rivers@dg-rtp.dg.com, freebsd-hackers@FreeBSD.ORG Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) In-Reply-To: Your message of "Fri, 4 Apr 1997 17:43:38 +0100 (BST)" References: X-Mailer: Mew version 1.06 on Emacs 19.34.1 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Date: Sat, 05 Apr 1997 00:22:09 +0200 From: Tor Egge Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > On Fri, 4 Apr 1997, Tor Egge wrote: > > > When the vnode is on the freelist, with 0 as v_usecount, and the > > vp->v_interlock is not locked, how should getnewvnode decide that the > > vnode is not reusable ? VOP_ISLOCKED ? Not currently used in > > getnewvnode nor in vget. > > The vnode in question is being recycled from VT_NFS to VT_UFS. > Getnewvnode picks a vnode of the free list and calls vgone to extract it > from the previous owner. This calls VOP_LOCK(.., LK_DRAIN, ..) which is > intended to block until the vnode is inactive. > > For vnodes which are not being recycled but just picked up from the cache, > there is no race but still either vget or its caller will lock the vnode > before using it for anything important. OK. I thought that vnode_free_list was meant to be a free list, due to the list name. I was wrong. Proper locking on NFS will help against the crash I experienced, but if the NFS server goes down at the wrong moment, a call to getnewvnode may hang until the NFS server is back again. - Tor Egge From owner-freebsd-hackers Fri Apr 4 14:47:22 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA08573 for hackers-outgoing; Fri, 4 Apr 1997 14:47:22 -0800 (PST) Received: from pdx1.world.net (pdx1.world.net [192.243.32.18]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA08566 for ; Fri, 4 Apr 1997 14:47:18 -0800 (PST) From: proff@suburbia.net Received: from suburbia.net (suburbia.net [203.4.184.1]) by pdx1.world.net (8.7.5/8.7.3) with SMTP id OAA01628 for ; Fri, 4 Apr 1997 14:49:24 -0800 (PST) Received: (qmail 13290 invoked by uid 110); 4 Apr 1997 19:15:03 -0000 Message-ID: <19970404191503.13288.qmail@suburbia.net> Subject: Re: IP Filter 3.2alpha4 In-Reply-To: <199704021314.XAA13307@plum.cyber.com.au> from Darren Reed at "Apr 2, 97 11:14:33 pm" To: darrenr@cyber.com.au (Darren Reed) Date: Sat, 5 Apr 1997 05:15:03 +1000 (EST) Cc: ipfilter@postbox.anu.edu.au X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > 3.2alpha4 is primarily concerned with one thing: port to FreeBSD-2.2 Here are some bugs, some particular to freebsd, and some not: **fil.c: (general) #ifdef _KERNEL if (pass & FR_RETICMP) { # if SOLARIS ICMP_ERROR(q, ip, ICMP_UNREACH, fin->fin_icode, qif, ip->ip_src); # else ICMP_ERROR(m, ip, ICMP_UNREACH, fin->fin_icode, ifp, ip->ip_src); m = *mp = NULL; /* freed by icmp_error() */ ^^^^^^^^^^^^^^^ (1) # endif frstats[0].fr_ret++; } else if ((pass & FR_RETRST) && !(fin->fin_fi.fi_fl & FI_SHORT)) { if (SEND_RESET(ip, qif, q, ifp) == 0) frstats[1].fr_ret++; } #else if (pass & FR_RETICMP) { verbose("- ICMP unreachable sent\n"); frstats[0].fr_ret++; } else if ((pass & FR_RETRST) && !(fin->fin_fi.fi_fl & FI_SHORT)) { verbose("- TCP RST sent\n"); frstats[1].fr_ret++; } #endif } } #ifdef _KERNEL # if !SOLARIS if (pass & FR_DUP) mc = m_copy(m, 0, M_COPYALL); ^^^^^^^^^^^^^^^^^^^^^^^^ (2) Any return-icmp rule will panic if dup-to is enabled, excepting Solaris. **ip_fil.c: (code in the >= 220000 ifdef is mine) # if __FreeBSD_version >= 220000 ;{ struct route ro; bzero (&ro, sizeof ro); (void) ip_output(m, (struct mbuf *)0, &ro, 0, 0); if (ro.ro_rt) RTFREE(ro.ro_rt); }; #else (void) ip_output(m, (struct mbuf *)0, 0, 0, 0); #endif return 0; } This code is the tail end of send_reset(), and is called whenever a tcp connection is RST'd. The original code will panic under FreeBSD 2.2 and above. There is a severe mbuf leak in/triggered by the state following code. Using: while true do clear netstat -m sleep 1 done and only: pass out quick on ed0 proto tcp from any to any flags S/SAFR keep state pass out quick on ed0 proto udp from any to any port = 53 keep state I saw an exponential increase in mbuf usage, over a period of around 5 minutes. This led to 100% network memory usage, at which point fr_check() started behaving very strangly, calling send_reset(), and triggering the previously mentioned panic condition with ip_output(). Unfortunately I haven't as yet been able to track down the mbuf leak. If it aids you it all, here is the technique I am using for examining ipfilter lkm crash dumps: root@sentry:/S# kgdb GDB is free software and you are welcome to distribute copies of it under certain conditions; type "show copying" to see the conditions. There is absolutely no warranty for GDB; type "show warranty" for details. GDB 4.16 (i386-unknown-freebsd), Copyright 1996 Free Software Foundation, Inc. (kgdb) symbol-file kernel Reading symbols from kernel...done. (kgdb) exec-file /var/crash/kernel.21 (kgdb) core-file /var/crash/vmcore.21 IdlePTD 1e5000 current pcb at 1c29d4 panic: page fault #0 boot (howto=256) at ../../kern/kern_shutdown.c:244 244 dumppcb.pcb_cr3 = rcr3(); (kgdb) bt #0 boot (howto=256) at ../../kern/kern_shutdown.c:244 #1 0xf010dc5a in panic (fmt=0xf017f4ff "page fault") at ../../kern/kern_shutdown.c:368 #2 0xf0180073 in trap_fatal (frame=0xf01b2eb4) at ../../i386/i386/trap.c:742 #3 0xf017fb59 in trap_pfault (frame=0xf01b2eb4, usermode=0) at ../../i386/i386/trap.c:653 #4 0xf017f82f in trap (frame={tf_es = 16, tf_ds = 16, tf_edi = 0, tf_esi = 0, tf_ebp = -266653936, tf_isp = -266653988, tf_ebx = 200967105, tf_edx = 1073610751, tf_ecx = -266653792, tf_eax = -2147483648, tf_trapno = 12, tf_err = 0, tf_eip = -267249626, tf_cs = 8, tf_eflags = 66118, tf_esp = 200967105, tf_ss = -266653884}) at ../../i386/i386/trap.c:311 #5 0xf0121826 in m_copym (m=0x0, off0=0, len=1000000000, wait=1) at ../../kern/uipc_mbuf.c:419 #6 0xf4b40e5f in ?? () #7 0xf014d650 in ip_input (m=0xf0cf8a00) at ../../netinet/ip_input.c:330 #8 0xf014da18 in ipintr () at ../../netinet/ip_input.c:555 (kgdb) p *fr_checkp $1 = {int ()} 0xf4b409f4 (kgdb) [1]+ Suspended kgdb root@profane:/S# nm /lkm/if_ipl.o|grep fr_check 000019d4 T _fr_check U _fr_checkp 000041cc T _fr_checkstate root@profane:/S# % kgdb (kgdb) add-symbol-file /local2/src/ip_fil3.2a4/BSD/i386/if_ipl.o *fr_checkp-0x19d4 add symbol table from file "/local2/src/ip_fil3.2a4/BSD/i386/if_ipl.o" at text_addr = 0xf4b3f020? (y or n) y (kgdb) bt #0 boot (howto=256) at ../../kern/kern_shutdown.c:244 #1 0xf010dc5a in panic (fmt=0xf017f4ff "page fault") at ../../kern/kern_shutdown.c:368 #2 0xf0180073 in trap_fatal (frame=0xf01b2eb4) at ../../i386/i386/trap.c:742 #3 0xf017fb59 in trap_pfault (frame=0xf01b2eb4, usermode=0) at ../../i386/i386/trap.c:653 #4 0xf017f82f in trap (frame={tf_es = 16, tf_ds = 16, tf_edi = 0, tf_esi = 0, tf_ebp = -266653936, tf_isp = -266653988, tf_ebx = 200967105, tf_edx = 1073610751, tf_ecx = -266653792, tf_eax = -2147483648, tf_trapno = 12, tf_err = 0, tf_eip = -267249626, tf_cs = 8, tf_eflags = 66118, tf_esp = 200967105, tf_ss = -266653884}) at ../../i386/i386/trap.c:311 #5 0xf0121826 in m_copym (m=0x0, off0=0, len=1000000000, wait=1) at ../../kern/uipc_mbuf.c:419 #6 0xf4b40e5f in fr_check (ip=0xf0cf8a2c, hlen=20, ifp=0xf01cbf9c, out=0, mp=0xf01b2fa0) at ../../fil.c:719 #7 0xf014d650 in ip_input (m=0xf0cf8a00) at ../../netinet/ip_input.c:330 #8 0xf014da18 in ipintr () at ../../netinet/ip_input.c:555 (kgdb) p pass $3 = 200967105 -- Prof. Julian Assange |If you want to build a ship, don't drum up people |together to collect wood and don't assign them tasks proff@suburbia.net |and work, but rather teach them to long for the endless proff@gnu.ai.mit.edu |immensity of the sea. -- Antoine de Saint Exupery From owner-freebsd-hackers Fri Apr 4 14:47:28 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA08594 for hackers-outgoing; Fri, 4 Apr 1997 14:47:28 -0800 (PST) Received: from pdx1.world.net (pdx1.world.net [192.243.32.18]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA08589 for ; Fri, 4 Apr 1997 14:47:26 -0800 (PST) From: proff@suburbia.net Received: from suburbia.net (suburbia.net [203.4.184.1]) by pdx1.world.net (8.7.5/8.7.3) with SMTP id OAA01659 for ; Fri, 4 Apr 1997 14:49:42 -0800 (PST) Received: (qmail 7413 invoked by uid 110); 4 Apr 1997 13:14:08 -0000 Message-ID: <19970404131408.7412.qmail@suburbia.net> Subject: Re: IP Filter in -current. In-Reply-To: <199704040736.RAA23038@plum.cyber.com.au> from Darren Reed at "Apr 4, 97 05:36:09 pm" To: darrenr@cyber.com.au (Darren Reed) Date: Fri, 4 Apr 1997 23:14:07 +1000 (EST) Cc: darrenr@cyber.com.au, hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Someone should give this guy a labotamy if he's the one who came up > with the "include every other include file in this include file" idea. > > > Use of TAILQ's for handling ifaddr chains. > > Is this consistent with NetBSD ? I'm not sure, but it wasn't consistent with ipfilter :) Note that the below is a not production; I haven't as yet isolated the FreeBSD_version when these changes were introduced (so use >=2). --- /usr/src/contrib/ipfilter/ip_nat.c Fri Apr 4 13:07:54 1997 +++ ip_nat.c Thu Apr 3 05:02:24 1997 @@ -10,8 +12,8 @@ static char rcsid[] = "$Id: ip_nat.c,v 2.0.2.8 1997/04/02 12:23:23 darrenr Exp $"; #endif +#ifdef IPFILTER_LKM +#define ACTUALLY_LKM_NOT_KERNEL #endif #if !defined(_KERNEL) && !defined(KERNEL) @@ -27,6 +27,7 @@ #include #include #include +#include #include #include #include @@ -42,7 +43,12 @@ # include #endif +#ifdef __FreeBSD__ +# include +#endif + #include +#include #ifdef sun #include #endif @@ -425,15 +431,22 @@ #else struct ifaddr *ifa; struct sockaddr_in *sin; - +# if __FreeBSD__ >= 2 + ifa = TAILQ_FIRST(&ifp->if_addrhead); +# else ifa = ifp->if_addrlist; +# endif # if BSD < 199306 sin = (struct sockaddr_in *)&ifa->ifa_addr; # else sin = (struct sockaddr_in *)ifa->ifa_addr; while (sin && ifa && sin->sin_family != AF_INET) { +# if __FreeBSD__ >=2 + ifa = TAILQ_NEXT(ifa, ifa_link); +# else ifa = ifa->ifa_next; +# endif sin = (struct sockaddr_in *)ifa->ifa_addr; } if (!ifa) > > > VOP_LOCK() in lite2 takes three arguments. > > 3 instead of 4 or 3 instead of 2 ? Three instead of one :) I'm still not clear if I'm using the correct locking arguments. mln_ipl.c uses this construct: VOP_LOCK(nd.ni_vp); Which i've turned into: VOP_LOCK(nd.ni_vp, LK_INTERLOCK|LK_SHARED, curproc); Although I feel: VOP_LOCK(nd.ni_vp, 0, curproc); is a more direct translation. This code is only called when the module is unloaded by hand, and so far all the module unloading has been courtesy of panic, so testing opportunities have been limited :) > > At run-time it seems to panic quickly if logging is going on and > > ipmon isn't there to read the log device. > > heh...I guess you'll be able to track this bug then :) I actually strongly suspect my initial views on this were wrong. The problem doesn't seem to lay with the log buffer filling, but with the connection state following code, which seems to crash after a certain period of time (memory?). Cheers, Julian. From owner-freebsd-hackers Fri Apr 4 14:48:04 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA08683 for hackers-outgoing; Fri, 4 Apr 1997 14:48:04 -0800 (PST) Received: from pdx1.world.net (pdx1.world.net [192.243.32.18]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA08678 for ; Fri, 4 Apr 1997 14:48:01 -0800 (PST) From: proff@suburbia.net Received: from suburbia.net (suburbia.net [203.4.184.1]) by pdx1.world.net (8.7.5/8.7.3) with SMTP id OAA01707 for ; Fri, 4 Apr 1997 14:50:13 -0800 (PST) Received: (qmail 6879 invoked by uid 110); 4 Apr 1997 12:49:13 -0000 Message-ID: <19970404124913.6878.qmail@suburbia.net> Subject: Re: new malloc In-Reply-To: <199704041027.CAA10928@root.com> from David Greenman at "Apr 4, 97 02:27:01 am" To: dg@root.com Date: Fri, 4 Apr 1997 22:49:13 +1000 (EST) Cc: borjam@we.lc.ehu.es, freebsd-hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > The conclusion was: Why not organizing the memory blocks in > >a LIFO structure? If I do a malloc and malloc can give me the most > >recently used of all the possible blocks, the probability of causing > >a page-in is decreased. Depending on the system load, a great > >number of page-ins can be avoided. (I have observed this in my case) > > > > > > Of course, the same idea cn be applied adding a MRU policy to the > >allocation routine. > > Both Poul-Henning's and the FreeBSD kernel's malloc allocate and insert > chunks at the head of the free queue - thus they already are LIFO. > > -DG You also get the benefit of a greater chance of the page being in cache. Cheers, Julian. From owner-freebsd-hackers Fri Apr 4 14:49:37 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA08858 for hackers-outgoing; Fri, 4 Apr 1997 14:49:37 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA08840 for ; Fri, 4 Apr 1997 14:49:32 -0800 (PST) Received: from dialup-usr11.etinc.com (dialup-usr11.etinc.com [204.141.95.132]) by etinc.com (8.8.3/8.6.9) with SMTP id RAA12114; Fri, 4 Apr 1997 17:54:00 -0500 (EST) Message-Id: <3.0.32.19970404174607.00aa89a4@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Fri, 04 Apr 1997 17:46:10 -0500 To: Michael Smith From: dennis Subject: 2.2.1R NFS and FTP load problem FOUND Cc: msmith@atrad.adelaide.edu.au, hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Any of you who have chosen not to ignore my rantings of the past 2 days might be aware that I couldn't get 2.2.1-RELEASE to load with an NFS load. The symptoms that I was having was a complete system hang during the transfers, at random points, but always rather quickly (usually within the bin load). I tried both a dec chip PCI card (de driver) and an old reliable ne2000 (ed driver) with identical results. I also tried 2 different PCs with 2 different types of hard drives...again identical results. I then thought that NFS was broken, and shifted to FTP. I was surprised to find that FTP had exactly the same problem; that is a random hang during the transfer. I then began thinking that it was the distribution diskette, and noticed that a new one had been put up on April 3...so I tried it. Same result. I then remembered that I had noticed that 2.2R was a bit clunky with 8 meg of ram, so I popped in another 8 meg, and the problems disappeared. So, it seems, ftp and nfs loads cant be done on at 8 meg system. Now I know that RAM is cheap, but there's something wrong when a system running no services crashes with 8 meg of ram during a file transfer. There must be a hole somewhere....unfortunately its pretty hard to monitor anything when booting from the install disk. Well, its 5:30 on a Friday, and Im outta here...so I guess its not THAT bad! :-) Dennis From owner-freebsd-hackers Fri Apr 4 15:23:27 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA10723 for hackers-outgoing; Fri, 4 Apr 1997 15:23:27 -0800 (PST) Received: from huset.fm.unit.no (huset.fm.unit.no [129.241.211.212]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id PAA10718 for ; Fri, 4 Apr 1997 15:23:15 -0800 (PST) Message-Id: <199704042323.PAA10718@freefall.freebsd.org> Received: (qmail 11990 invoked from network); 4 Apr 1997 23:23:07 -0000 Received: from huset.fm.unit.no (HELO stud.fim.ntnu.no) (129.241.211.212) by huset.fm.unit.no with SMTP; 4 Apr 1997 23:23:07 -0000 To: bwithrow@BayNetworks.COM Cc: hackers@freebsd.org Subject: Re: Suggested change to rc In-Reply-To: Your message of "Fri, 04 Apr 1997 08:29:14 -0500" References: <199704041329.IAA06810@tuva.engeast.baynetworks.com> X-Mailer: Mew version 1.06 on Emacs 19.34.1 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Date: Sat, 05 Apr 1997 01:23:07 +0200 From: Arne Henrik Juul Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk From: Robert Withrow Date: Fri, 04 Apr 1997 08:29:14 -0500 > In order to have amd get all of its information from nis, the following > change needs to be made in rc. This allows amdflags to be set like this: > > amdflags='-p -c 3600 -l syslog `ypcat -k amd.master`' Why aren't you doing amdflags="-p -c 3600 -l syslog `ypcat -k amd.master`" instead, if that's what you want? I can't see that you should need an eval for this. - Arne H. J. From owner-freebsd-hackers Fri Apr 4 15:51:04 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA11810 for hackers-outgoing; Fri, 4 Apr 1997 15:51:04 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id PAA11805 for ; Fri, 4 Apr 1997 15:51:00 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA18606; Fri, 4 Apr 1997 18:50:28 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Fri, 4 Apr 1997 18:50 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id SAA06940; Fri, 4 Apr 1997 18:17:28 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id SAA11584; Fri, 4 Apr 1997 18:23:22 -0500 (EST) Date: Fri, 4 Apr 1997 18:23:22 -0500 (EST) From: Thomas David Rivers Message-Id: <199704042323.SAA11584@lakes.water.net> To: ponds!nlsystems.com!dfr, ponds!idi.ntnu.no!Tor.Egge Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) Cc: ponds!root.com!dg, ponds!FreeBSD.ORG!freebsd-hackers, ponds!lakes.water.net!rivers Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk The discussion about NFS and vnode looking is very good to see... However, Tor, you had mentioned that if ufs ever blocks for some reason there would be other problems... can you describe where/why these problems would exist? [My problem doesn't involve NFS at all, just UFS.] - Dave Rivers - From owner-freebsd-hackers Fri Apr 4 16:23:51 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA12983 for hackers-outgoing; Fri, 4 Apr 1997 16:23:51 -0800 (PST) Received: from robin.camelot.de (root@robin.camelot.de [194.97.87.3]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id QAA12973 for ; Fri, 4 Apr 1997 16:23:46 -0800 (PST) From: robin!knarf@camelot.de Received: from robin by robin.camelot.de via sendmail with uucp id for hackers@freebsd.org; Sat, 5 Apr 1997 02:23:39 +0200 (MET DST) (Smail-3.2 1996-Jul-4 #1 built DST-Jul-28) Message-Id: Subject: Re: kern/2923: panic: vm_fault: fault on nofault entry, addr: f6e21000 In-Reply-To: <199704041117.GAA10676@lakes.water.net> from Thomas David Rivers at "Apr 4, 97 06:17:51 am" To: ponds!rivers@dg-rtp.dg.com (Thomas David Rivers) Date: Fri, 4 Apr 1997 19:22:08 +0200 (MET DST) X-Mailer: ELM [version 2.4ME+ PL29 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, You, Thomas David Rivers wrote: > > I have the same problem here on a medium loaded newsserver (6 GB > > newsspool). I get "panic: vm_fault: fault on nofault entry, addr: > > f6718000" with different values for addr. I updated from 2.1.5 to > > 2.2.1 about 48 hours before I've seen this problem the first time. > > > > I really have no idea what to do now. The machine always reboots > > some minutes after the crash. Running fsck on 10 GB disks all the > > time is no fun... :/ > > Well - this doesn't sound like exactly the same problem - although > I can easily imagine it's just a frustrating. It is... lancelot:/home/knarf> last reboot reboot ~ Fri Apr 4 18:52 reboot ~ Fri Apr 4 17:52 reboot ~ Fri Apr 4 16:13 reboot ~ Fri Apr 4 10:27 reboot ~ Fri Apr 4 10:17 reboot ~ Fri Apr 4 10:06 reboot ~ Fri Apr 4 09:47 reboot ~ Fri Apr 4 09:37 reboot ~ Fri Apr 4 09:27 reboot ~ Fri Apr 4 07:52 reboot ~ Fri Apr 4 06:17 reboot ~ Fri Apr 4 01:35 reboot ~ Thu Apr 3 22:18 reboot ~ Thu Apr 3 22:04 reboot ~ Thu Apr 3 21:48 reboot ~ Thu Apr 3 20:03 reboot ~ Thu Apr 3 19:46 reboot ~ Thu Apr 3 19:24 reboot ~ Thu Apr 3 19:10 Before 2.2.1 I had uptimes of 10 to 30 days... > > If you have any idea how I can make the uptime 6 hrs or higher, please > > tell me! > > Since I'm guessing this is a different problem; if you'll provide > the kgdb traceback and an nm of your kernel (are you running a 2.2.1 > GENERIC kernel?) and details of your system - it's quite possible your > problem isn't as thorny as mine and can be much more readily repaired. I am not familiar with kernel debugging, but I think I have to build a kernel with "options DDB" and/or KTRACE set. I'll do that right now and hope to receive further instructions from you. :) I have 128 MB of RAM and 2 x 64 MB swap partitions. Will this be a problem (crash dump...)? Bye, Knarf -- Frank Bartels |UUCP/ZModem/Fax: +49 89 8948040| "Captain, why not just knarf@camelot.de | http://www.camelot.de/~knarf/ | give the Borg Windows?" From owner-freebsd-hackers Fri Apr 4 16:23:58 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA13004 for hackers-outgoing; Fri, 4 Apr 1997 16:23:58 -0800 (PST) Received: from robin.camelot.de (root@robin.camelot.de [194.97.87.3]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id QAA12990 for ; Fri, 4 Apr 1997 16:23:52 -0800 (PST) From: robin!knarf@camelot.de Received: from robin by robin.camelot.de via sendmail with uucp id for hackers@freebsd.org; Sat, 5 Apr 1997 02:23:48 +0200 (MET DST) (Smail-3.2 1996-Jul-4 #1 built DST-Jul-28) Message-Id: Subject: Re: kern/2923: panic: vm_fault: fault on nofault entry, addr: f6e21000 In-Reply-To: <199704041117.GAA10676@lakes.water.net> from Thomas David Rivers at "Apr 4, 97 06:17:51 am" To: ponds!rivers@dg-rtp.dg.com (Thomas David Rivers) Date: Sat, 5 Apr 1997 01:05:17 +0200 (MET DST) X-Mailer: ELM [version 2.4ME+ PL29 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, You, Thomas David Rivers wrote: > > If you have any idea how I can make the uptime 6 hrs or higher, please > > tell me! > > Since I'm guessing this is a different problem; if you'll provide > the kgdb traceback and an nm of your kernel (are you running a 2.2.1 > GENERIC kernel?) and details of your system - it's quite possible your > problem isn't as thorny as mine and can be much more readily repaired. Ok, I have DDB in the kernel and had another panic. I typed `trace' and saw nfs_bioread, nfs_readdir, getdirentries, syscall, Xsyscall. Does this mean the panic happend during an nfs_readdir as a client? This would be really a bad thing. [later] Ooops, next reboot. After that I said "uustat -a" with /var/spool/uucp mounted via nfs from a 2.1.5 server. The command finished, but short after that it paniced again with the same (or similar) trace output in the debugger. Damn. Hmm, any ideas? How can I provide more detailed information? Bye, Knarf -- Frank Bartels |UUCP/ZModem/Fax: +49 89 8948040| "Captain, why not just knarf@camelot.de | http://www.camelot.de/~knarf/ | give the Borg Windows?" From owner-freebsd-hackers Fri Apr 4 17:20:24 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA14281 for hackers-outgoing; Fri, 4 Apr 1997 17:20:24 -0800 (PST) Received: from bookcase.com (root@notes.bookcase.com [207.22.115.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA14273 for ; Fri, 4 Apr 1997 17:20:16 -0800 (PST) Received: from localhost (chadf@localhost) by bookcase.com (8.8.5/8.7.3) with SMTP id TAA17720 for ; Fri, 4 Apr 1997 19:23:55 -0500 (EST) Date: Fri, 4 Apr 1997 20:23:54 -0400 (EST) From: "Chad M. Fraleigh" X-Sender: chadf@notes To: hackers@FreeBSD.ORG Subject: Re: Code maintenance In-Reply-To: <19970402211157.WU57156@uriah.heep.sax.de> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Yes, aic(4) and ft(4) still work, more or less. We'd probably have > removed them otherwise. But in particular ft(4) suffers from a not > very elegant design, and its unmaintained state resulted in it being > almost worthless over time, since it doesn't handle the more modern > drives that are available now at all. If it remains unmaintained, the > day will come when all the previously existing drives it used to > support on the earth are dead. I was once doing a little work on trying to fix the ft driver (ok, more like rewrite the whole thing). But to do so the way I wanted would of required rewritting half the fdc/fd code into independent, generalized moduals (like the scsi ones). Of course in the process it would of broke the floppy drive (at least for awhile) and annoyed many, many people. Just what I need.. an electronic linch mob after me. ;) Eventually somebody else said they wanted to work on it, and since I didn't have the time, and no docs (was doing it by reverse enginering the current source with *alot* of guess work), I just gave up. Anyway.. now I'm working on a way to trace kernel memory leaks. -Chad From owner-freebsd-hackers Fri Apr 4 17:23:23 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA14382 for hackers-outgoing; Fri, 4 Apr 1997 17:23:23 -0800 (PST) Received: from mail.burlco.lib.nj.us (mail.burlco.lib.nj.us [204.91.160.99]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA14373 for ; Fri, 4 Apr 1997 17:23:19 -0800 (PST) Received: from localhost (rcummins@localhost) by mail.burlco.lib.nj.us (8.8.5/8.8.5) with SMTP id UAA00878; Fri, 4 Apr 1997 20:22:53 -0500 (EST) Date: Fri, 4 Apr 1997 20:22:52 -0500 (EST) From: Ray Cummins To: dennis cc: Michael Smith , hackers@freebsd.org Subject: Re: 2.2.1R NFS and FTP load problem FOUND In-Reply-To: <3.0.32.19970404174607.00aa89a4@etinc.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Fri, 4 Apr 1997, dennis wrote: > I then remembered that I had noticed that 2.2R was a bit clunky with > 8 meg of ram, so I popped in another 8 meg, and the problems > disappeared. So, it seems, ftp and nfs loads cant be done on at > 8 meg system. > Hm, I upgraded to 2.2.1-RELEASE from 2.1.0-RELEASE with no problems on a Dell 486sx-20 w/ 8MB RAM, using NFS. From owner-freebsd-hackers Fri Apr 4 18:20:42 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id SAA16029 for hackers-outgoing; Fri, 4 Apr 1997 18:20:42 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id SAA16019 for ; Fri, 4 Apr 1997 18:20:38 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA12630; Fri, 4 Apr 1997 21:20:05 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Fri, 4 Apr 1997 21:20 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id UAA09291 for ; Fri, 4 Apr 1997 20:53:10 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id UAA12259 for freebsd-hackers@freefall.cdrom.com; Fri, 4 Apr 1997 20:59:05 -0500 (EST) Date: Fri, 4 Apr 1997 20:59:05 -0500 (EST) From: Thomas David Rivers Message-Id: <199704050159.UAA12259@lakes.water.net> To: ponds!freefall.cdrom.com!freebsd-hackers Subject: "dup alloc" problem and 2.2.1. Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Just to keep up-to-date; I've just verified that the "dup alloc" problem exists in 2.2.1-RELEASE... (I guess I'm just not that lucky :-) ) - Dave Rivers - From owner-freebsd-hackers Fri Apr 4 18:20:42 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id SAA16028 for hackers-outgoing; Fri, 4 Apr 1997 18:20:42 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id SAA16018 for ; Fri, 4 Apr 1997 18:20:37 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA12625; Fri, 4 Apr 1997 21:20:03 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Fri, 4 Apr 1997 21:20 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id UAA09286 for ; Fri, 4 Apr 1997 20:51:04 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id UAA12255 for freebsd-hackers@freefall.cdrom.com; Fri, 4 Apr 1997 20:56:59 -0500 (EST) Date: Fri, 4 Apr 1997 20:56:59 -0500 (EST) From: Thomas David Rivers Message-Id: <199704050156.UAA12255@lakes.water.net> To: ponds!freefall.cdrom.com!freebsd-hackers Subject: Some significant (daresay serious) problems with 2.2.1 install & keyboard Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk I've (at long last) gotten the 2.2.1 distribution down to do test installs. I'm using the 2.2.1-RELEASE from ftp.freebsd.org; with the boot.flp found in floppies/newer. [There is no checksum there, so I have no way of validating my copy...] If I boot this on a 386dx-33, with an intel 387; 12 megs of memory and an MGA screen.., circa 1988. I find that I cannot use the cursor keys on the keyboard to navigate the options... the down-arrow does nothing. Furthermore; even though the menu items imply you can use numbers to select an option; pressing the numbers on the keyboard yields nothing - or incorrect results. For example, pressing the '3' on the keyboard moves the highlighted selection to "l" for "Load Config"; pressing '5' gets you "c" for Configure . Pressing '7' pops up the "Are you sure you wish to exit?" diaglog box; which immediately pops back down... Cntrl-Alt-Del doesn't do anything... I'm fairly confident the keyboard driver is working; since I can go configure the kernel at boot time and enter things like: port ed0 0x300 irq ed0 9 [which has plenty of digits, etc...] Now; here's the "kicker" - If if don't configure the kernel in any way (i.e. don't go into CLI mode to enter those commands) the keyboard works just great! [Of course, to install 2.2.1, I need to get to the ethernet, which means I need to reconfigure ed0... sigh...] I'm happy to work with anyone/test boot floppies, to resolve this problem. - Dave Rivers - From owner-freebsd-hackers Fri Apr 4 19:00:51 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id TAA17239 for hackers-outgoing; Fri, 4 Apr 1997 19:00:51 -0800 (PST) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id TAA17234 for ; Fri, 4 Apr 1997 19:00:48 -0800 (PST) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id TAA02609; Fri, 4 Apr 1997 19:00:44 -0800 (PST) To: Thomas David Rivers Cc: hackers@freebsd.org Subject: Re: Some significant (daresay serious) problems with 2.2.1 install & keyboard In-reply-to: Your message of "Fri, 04 Apr 1997 20:56:59 EST." <199704050156.UAA12255@lakes.water.net> Date: Fri, 04 Apr 1997 19:00:43 -0800 Message-ID: <2605.860209243@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > I'm using the 2.2.1-RELEASE from ftp.freebsd.org; with the > boot.flp found in floppies/newer. [There is no checksum there, > so I have no way of validating my copy...] And I assume that the "older" boot.flp does the same thing on this 386? I'm just trying to establish an effective timeframe for the breakage. Thanks! Jordan P.S. Fix your mailer please - it emits very bogus addresses! I also don't see the need for a UUCP account in this day and age. Why not get a *real* email address for $5/mo or so? It's cheap if you already have internet access through some other means (and folks like hotmail or bigger.net will even give it to you for free if you're willing to receive commercial adverts occasionally). :-) From owner-freebsd-hackers Fri Apr 4 20:03:55 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA19153 for hackers-outgoing; Fri, 4 Apr 1997 20:03:55 -0800 (PST) Received: from vinyl.quickweb.com (vinyl.quickweb.com [206.222.77.8]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA19147 for ; Fri, 4 Apr 1997 20:03:52 -0800 (PST) Received: from localhost (mark@localhost) by vinyl.quickweb.com (8.8.5/8.6.12) with SMTP id WAA23994; Fri, 4 Apr 1997 22:57:32 -0500 (EST) Date: Fri, 4 Apr 1997 22:57:32 -0500 (EST) From: Mark Mayo To: dennis cc: Michael Smith , hackers@FreeBSD.ORG Subject: Re: 2.2.1R NFS and FTP load problem FOUND In-Reply-To: <3.0.32.19970404174607.00aa89a4@etinc.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Fri, 4 Apr 1997, dennis wrote: > > I then remembered that I had noticed that 2.2R was a bit clunky with > 8 meg of ram, so I popped in another 8 meg, and the problems > disappeared. So, it seems, ftp and nfs loads cant be done on at > 8 meg system. > That's odd.. I had no problems installing 2.2R or 2.2.1R over NFS with an old ISA-only 486 with 8MB of RAM. What are some more specifics of the machine? It must be a flaky piece of hardware or something bizarre. Or maybe just bad luck ;-) -Mark ---------------------------------------------------------------------------- Mark Mayo mark@quickweb.com RingZero Comp. http://vinyl.quickweb.com/mark finger mark@quickweb.com for my PGP key and GCS code ---------------------------------------------------------------------------- University degrees are a bit like adultery: you may not want to get involved with that sort of thing, but you don't want to be thought incapable. -Sir Peter Imbert From owner-freebsd-hackers Fri Apr 4 20:40:41 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA20301 for hackers-outgoing; Fri, 4 Apr 1997 20:40:41 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id UAA20280 for ; Fri, 4 Apr 1997 20:40:18 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id VAA20745; Fri, 4 Apr 1997 21:21:42 -0700 From: Terry Lambert Message-Id: <199704050421.VAA20745@phaeton.artisoft.com> Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) To: Tor.Egge@idi.ntnu.no (Tor Egge) Date: Fri, 4 Apr 1997 21:21:42 -0700 (MST) Cc: dfr@nlsystems.com, dg@root.com, ponds!rivers@dg-rtp.dg.com, freebsd-hackers@freebsd.org In-Reply-To: <199704042222.AAA14484@pat.idt.unit.no> from "Tor Egge" at Apr 5, 97 00:22:09 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Proper locking on NFS will help against the crash I experienced, but > if the NFS server goes down at the wrong moment, a call to getnewvnode > may hang until the NFS server is back again. This will always be the case, right? Or do you mean the case where an attempt to vclean() a vnode is against an NFS vnode with the server down? Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Fri Apr 4 20:42:42 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA20395 for hackers-outgoing; Fri, 4 Apr 1997 20:42:42 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id UAA20390 for ; Fri, 4 Apr 1997 20:42:40 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id VAA20754; Fri, 4 Apr 1997 21:24:16 -0700 From: Terry Lambert Message-Id: <199704050424.VAA20754@phaeton.artisoft.com> Subject: Re: 2.2.1R NFS and FTP load problem FOUND To: dennis@etinc.com (dennis) Date: Fri, 4 Apr 1997 21:24:16 -0700 (MST) Cc: msmith@atrad.adelaide.edu.au, hackers@freebsd.org In-Reply-To: <3.0.32.19970404174607.00aa89a4@etinc.com> from "dennis" at Apr 4, 97 05:46:10 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > I then remembered that I had noticed that 2.2R was a bit clunky with > 8 meg of ram, so I popped in another 8 meg, and the problems > disappeared. So, it seems, ftp and nfs loads cant be done on at > 8 meg system. More likely, your RAM is flakey, and you are simply changing the usage pattern such that the flakey RAM is no longer critical path. Any chance of you playing a game of "shuffle the SIMMs" to verify this? Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Fri Apr 4 21:05:37 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id VAA21091 for hackers-outgoing; Fri, 4 Apr 1997 21:05:37 -0800 (PST) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id VAA21086 for ; Fri, 4 Apr 1997 21:05:33 -0800 (PST) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.5/8.7.3) id OAA26477; Sat, 5 Apr 1997 14:35:20 +0930 (CST) From: Michael Smith Message-Id: <199704050505.OAA26477@genesis.atrad.adelaide.edu.au> Subject: Re: Code maintenance In-Reply-To: from "Chad M. Fraleigh" at "Apr 4, 97 08:23:54 pm" To: chadf@bookcase.com (Chad M. Fraleigh) Date: Sat, 5 Apr 1997 14:35:20 +0930 (CST) Cc: hackers@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Chad M. Fraleigh stands accused of saying: > > Anyway.. now I'm working on a way to trace kernel memory leaks. 'vmstat -m' is your friend! > -Chad -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ From owner-freebsd-hackers Fri Apr 4 21:50:39 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id VAA22747 for hackers-outgoing; Fri, 4 Apr 1997 21:50:39 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id VAA22742 for ; Fri, 4 Apr 1997 21:50:35 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA13767; Sat, 5 Apr 1997 00:50:03 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Sat, 5 Apr 1997 00:50 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id XAA11906; Fri, 4 Apr 1997 23:18:39 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id XAA14880; Fri, 4 Apr 1997 23:24:34 -0500 (EST) Date: Fri, 4 Apr 1997 23:24:34 -0500 (EST) From: Thomas David Rivers Message-Id: <199704050424.XAA14880@lakes.water.net> To: ponds!zeta.org.au!bde, ponds!root.com!dg, ponds!freefall.cdrom.com!freebsd-hackers, ponds!lakes.water.net!rivers Subject: "dup alloc"/interrupts and Xresume11() - splbio() not working???? Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Ok - I've finally learned how to use ddb - and discovered I was looking in the totally wrong place for my "dup alloc" problem. [I should, of course, simply been looking at physio(), since newfs uses the raw device... duh.] Anyway; I'm seeing things like this, and want to make sure I understand everything that's going on: 1) Xresume11() is the vector that is taken when interrupt 11 is triggered - right? 2) A cpl of 0xc0000840 should mask-out interrupt 11... right? Well - look at the following from a ddb session: [fjord]$ Debugger("serial console break") Stopped at _Debugger+0x2b: movb $0,_in_Debugger.110 db> br _scsi_done db> cont [fjord]$ newfs -b 8192 -f 1024 /dev/rsd0a Breakpoint at _scsi_done: pushl %ebp db> cont Breakpoint at _scsi_done: pushl %ebp db> print *(_cpl) c0000840 db> trace _scsi_done(f047db00,f05d9000,f05d9b18,f0) at _scsi_done _aha_done(f05d9000,f05d9b18,f0190404,f05d9b18) at _aha_done+0x1ab _ahaintr(0,80000000,f05e0010,10,0) at _ahaintr+0xd8 Xresume11() at Xresume11+0x25 --- interrupt, eip = 0xf018b5e0, ebp = 0x0 --- idle_loop() at idle_loop+0x38 db> doesn't this show that I have entered scsi_done() (because of an interrupt 11) - while the cpl should have IRQ 11 masked???? Now - looking at Xresume/**/irq_num: in vector.s, shouldn't the following have skipped over this: movl _cpl,%eax ; \ testb $IRQ_BIT(irq_num),%reg ; \ jne 2f ; \ which should test bit #3 of the 'ah' register since %reg expands to %ah (as defined by the INTR(11,..) expansion later in vector.s.) Which, loads up _cpl into %eax and tests the appropriate bit... Now; what follows is a contradiction (remember those proofs in college?) that I can't reconcile: Ok, lets say _cpl isn't at splbio(), so we take the interrupt and jump to ahaintr() ; which calls aha_done() which calls scsi_done(); which is where I hit my kernel breakpoint. As far as I can tell; none of the intervening routines changed _cpl. How could _cpl possibly be splbio() when we hit the prolog to scsi_done(), and not be splbio() in ahaintr. And, then; how could you have gotten into ahaintr() if the testb above failed??? Why have I entered scsi_done when I don't believe I should have? And, if this is all correct; how would an splbio() matter to just about any driver (that is, why aren't more drivers falling over?) Can someone explain to me what's going on here? Has some other function mysteriously done something with/to _cpl? - Thanks - - Dave Rivers - From owner-freebsd-hackers Fri Apr 4 23:14:41 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA25266 for hackers-outgoing; Fri, 4 Apr 1997 23:14:41 -0800 (PST) Received: from mailsrv.cc.utsunomiya-u.ac.jp (mailsrv.cc.utsunomiya-u.ac.jp [160.12.200.20]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA25261 for ; Fri, 4 Apr 1997 23:14:35 -0800 (PST) Received: from zodiac.mech.utsunomiya-u.ac.jp (CQRPLer9UypBLbgolu5p7uplbBLmkX4+@zodiac.mech.utsunomiya-u.ac.jp [160.12.33.1]) by mailsrv.cc.utsunomiya-u.ac.jp (8.8.5+2.7Wbeta5/3.5Wpl4) with ESMTP id QAA13551; Sat, 5 Apr 1997 16:13:52 +0900 (JST) Received: from zodiac.mech.utsunomiya-u.ac.jp (zenith.mech.utsunomiya-u.ac.jp [160.12.33.60]) by zodiac.mech.utsunomiya-u.ac.jp (8.7.6+2.6Wbeta7/3.4W/zodiac-May96) with ESMTP id QAA25642; Sat, 5 Apr 1997 16:18:22 +0900 (JST) Message-Id: <199704050718.QAA25642@zodiac.mech.utsunomiya-u.ac.jp> To: Thomas David Rivers cc: freebsd-hackers@freebsd.org, yokota@zodiac.mech.utsunomiya-u.ac.jp Subject: Re: Some significant (daresay serious) problems with 2.2.1 install & keyboard In-reply-to: Your message of "Fri, 04 Apr 1997 20:56:59 EST." <199704050156.UAA12255@lakes.water.net> References: <199704050156.UAA12255@lakes.water.net> Date: Sat, 05 Apr 1997 16:18:21 +0900 From: Kazutaka YOKOTA Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >I'm using the 2.2.1-RELEASE from ftp.freebsd.org; with the >boot.flp found in floppies/newer. [There is no checksum there, >so I have no way of validating my copy...] > >If I boot this on a 386dx-33, with an intel 387; 12 megs of >memory and an MGA screen.., circa 1988. I find that I cannot use >the cursor keys on the keyboard to navigate the options... the >down-arrow does nothing. > >Furthermore; even though the menu items imply you can use numbers >to select an option; pressing the numbers on the keyboard yields >nothing - or incorrect results. For example, pressing the '3' >on the keyboard moves the highlighted selection to "l" for "Load Config"; >pressing '5' gets you "c" for Configure . Pressing '7' pops >up the "Are you sure you wish to exit?" diaglog box; which immediately >pops back down... Do regular alpha and numeric keys and any other keys (such as ENTER, ESC, TAB, space and function keys) work? How about Num-lock, Caps-lock and Ctrl? Do you see any error messages such as "sc0: keyboard reset failed" after the kernel configuration menu. >Now; here's the "kicker" - If if don't configure the kernel >in any way (i.e. don't go into CLI mode to enter those commands) >the keyboard works just great! [Of course, to install 2.2.1, I >need to get to the ethernet, which means I need to reconfigure >ed0... sigh...] Hmm, strange. Kazu >I'm happy to work with anyone/test boot floppies, to resolve this problem. > > - Dave Rivers - From owner-freebsd-hackers Sat Apr 5 02:45:17 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id CAA01669 for hackers-outgoing; Sat, 5 Apr 1997 02:45:17 -0800 (PST) Received: from pdx1.world.net (pdx1.world.net [192.243.32.18]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id CAA01663 for ; Sat, 5 Apr 1997 02:45:14 -0800 (PST) From: proff@suburbia.net Received: from suburbia.net (suburbia.net [203.4.184.1]) by pdx1.world.net (8.7.5/8.7.3) with SMTP id CAA12439 for ; Sat, 5 Apr 1997 02:47:24 -0800 (PST) Received: (qmail 22328 invoked by uid 110); 5 Apr 1997 10:44:22 -0000 Message-ID: <19970405104422.22327.qmail@suburbia.net> Subject: ipfilter To: hackers@freebsd.org Date: Sat, 5 Apr 1997 20:44:22 +1000 (EST) X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I have almost completed my ipfilter-current integration. If anyone has any outstanding code can they please send it to me. ps. darren: I'm considering breaking out nearly all of your #includes into seven different files: conf.h dns.h netkern.h kernel.h types.h netinet.h user.h At the moment the are a mess and not easily maintainable across platforms. The changes in FreeBSD's include files don't help, but ipfilter should be able to adapt more easily to such conflicts. I still haven't nailed the mbuf leak in the tcp-state following code. I've stopped the various panics that occur when pass returns strange values, nonetheless ipf was happy to eat over a 1000k in data mbuf's with only a few dozen concurrently active connections last night, and a high-water (according to ipfstat) of 78 connections. e.g 1408/1440 mbufs in use: 1391 mbufs allocated to data 1 mbufs allocated to packet headers 13 mbufs allocated to protocol control blocks 3 mbufs allocated to socket names and addresses 313/318 mbuf clusters in use 816 Kbytes allocated to network (98% in use) 0 requests for memory denied 0 requests for memory delayed 0 calls to protocol drain routines -- Prof. Julian Assange |If you want to build a ship, don't drum up people |together to collect wood and don't assign them tasks proff@suburbia.net |and work, but rather teach them to long for the endless proff@gnu.ai.mit.edu |immensity of the sea. -- Antoine de Saint Exupery From owner-freebsd-hackers Sat Apr 5 03:50:39 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id DAA03115 for hackers-outgoing; Sat, 5 Apr 1997 03:50:39 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id DAA03106 for ; Sat, 5 Apr 1997 03:50:35 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA19546; Sat, 5 Apr 1997 06:50:03 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Sat, 5 Apr 1997 06:50 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id GAA21282; Sat, 5 Apr 1997 06:11:49 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id GAA17222; Sat, 5 Apr 1997 06:17:46 -0500 (EST) Date: Sat, 5 Apr 1997 06:17:46 -0500 (EST) From: Thomas David Rivers Message-Id: <199704051117.GAA17222@lakes.water.net> To: ponds!time.cdrom.com!jhk, ponds!lakes.water.net!rivers Subject: Re: Some significant (daresay serious) problems with 2.2.1 install & keyboard Cc: ponds!freebsd.org!hackers Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > > I'm using the 2.2.1-RELEASE from ftp.freebsd.org; with the > > boot.flp found in floppies/newer. [There is no checksum there, > > so I have no way of validating my copy...] > > And I assume that the "older" boot.flp does the same thing on this > 386? I'm just trying to establish an effective timeframe for the > breakage. Thanks! Well, after I discovered that if I didn't change the kernel config, everything worked - I didn't try the older 2.2.1-RELEASE boot.flp. I'll go do that now and report back... > > Jordan > > P.S. Fix your mailer please - it emits very bogus addresses! It's not me - really :-) When things leave hear they look great... For example, I believe our own sendmail.cf examples still reference ucbvax.Berkely.edu as the UUCP_RELAY; which is where a lot of this creeps in. But; to help me track this down - what bogus address to you get for this return mail? And, of course, since this is all FreeBSD - it's probably worthwhile for someone to get it right for some else to use... > I also > don't see the need for a UUCP account in this day and age. Why not > get a *real* email address for $5/mo or so? It's cheap if you already > have internet access through some other means (and folks like hotmail > or bigger.net will even give it to you for free if you're willing to > receive commercial adverts occasionally). :-) I've been working on this; but it's predicated on a 2.2.X box to run natd (or ipfw) that stays up for a while... I've been trying to get the "dup alloc" problem fixed first. In the mean time; I've hacked together some things (based on jmb's recommendation) that should allow me to forward my mail to my ISP... - Dave Rivers - From owner-freebsd-hackers Sat Apr 5 03:50:41 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id DAA03135 for hackers-outgoing; Sat, 5 Apr 1997 03:50:41 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id DAA03110 for ; Sat, 5 Apr 1997 03:50:37 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA19564; Sat, 5 Apr 1997 06:50:05 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Sat, 5 Apr 1997 06:50 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id GAA21914; Sat, 5 Apr 1997 06:25:39 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id GAA17438; Sat, 5 Apr 1997 06:31:35 -0500 (EST) Date: Sat, 5 Apr 1997 06:31:35 -0500 (EST) From: Thomas David Rivers Message-Id: <199704051131.GAA17438@lakes.water.net> To: ponds!lakes.water.net!rivers, ponds!zodiac.mech.utsunomiya-u.ac.jp!yokota Subject: Re: Some significant (daresay serious) problems with 2.2.1 install & keyboard Cc: ponds!freebsd.org!freebsd-hackers Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > >I'm using the 2.2.1-RELEASE from ftp.freebsd.org; with the > >boot.flp found in floppies/newer. [There is no checksum there, > >so I have no way of validating my copy...] > > > >If I boot this on a 386dx-33, with an intel 387; 12 megs of > >memory and an MGA screen.., circa 1988. I find that I cannot use > >the cursor keys on the keyboard to navigate the options... the > >down-arrow does nothing. > > > >Furthermore; even though the menu items imply you can use numbers > >to select an option; pressing the numbers on the keyboard yields > >nothing - or incorrect results. For example, pressing the '3' > >on the keyboard moves the highlighted selection to "l" for "Load Config"; > >pressing '5' gets you "c" for Configure . Pressing '7' pops > >up the "Are you sure you wish to exit?" diaglog box; which immediately > >pops back down... > > Do regular alpha and numeric keys and any other keys (such as ENTER, > ESC, TAB, space and function keys) work? How about Num-lock, Caps-lock > and Ctrl? Just about the only keys that "work" are the ones I mentioned... > > Do you see any error messages such as "sc0: keyboard reset failed" > after the kernel configuration menu. Nope... > > >Now; here's the "kicker" - If if don't configure the kernel > >in any way (i.e. don't go into CLI mode to enter those commands) > >the keyboard works just great! [Of course, to install 2.2.1, I > >need to get to the ethernet, which means I need to reconfigure > >ed0... sigh...] > > Hmm, strange. We've seen this before (it happened in one of the 2.2-GAMMA releases, if I remember correctly...) > > Kazu - Dave Rivers - From owner-freebsd-hackers Sat Apr 5 03:50:42 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id DAA03143 for hackers-outgoing; Sat, 5 Apr 1997 03:50:42 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id DAA03112 for ; Sat, 5 Apr 1997 03:50:38 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA19570; Sat, 5 Apr 1997 06:50:06 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Sat, 5 Apr 1997 06:50 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id GAA22041; Sat, 5 Apr 1997 06:28:17 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id GAA17448; Sat, 5 Apr 1997 06:34:14 -0500 (EST) Date: Sat, 5 Apr 1997 06:34:14 -0500 (EST) From: Thomas David Rivers Message-Id: <199704051134.GAA17448@lakes.water.net> To: ponds!zeta.org.au!bde, ponds!root.com!dg, ponds!freefall.cdrom.com!freebsd-hackers, ponds!lakes.water.net!rivers Subject: Re: "dup alloc"/interrupts and Xresume11() - splbio() not working???? Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > > Anyway; I'm seeing things like this, and want to make sure > >I understand everything that's going on: > > > > 1) Xresume11() is the vector that is taken when interrupt > > 11 is triggered - right? > > Wrong - _Xintr11 is the vector. > > Bruce > Well - umm; Ok - sure... But then, why does _Xresume11 call ahaintr()? Which eventually gets me down to scsi_done() with _cpl set to splbio? (Or, have I misunderstood something else and this isn't necessarily a "bad" thing to happen?) - Dave Rivers - From owner-freebsd-hackers Sat Apr 5 03:50:43 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id DAA03148 for hackers-outgoing; Sat, 5 Apr 1997 03:50:43 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id DAA03118 for ; Sat, 5 Apr 1997 03:50:39 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA19608; Sat, 5 Apr 1997 06:50:08 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Sat, 5 Apr 1997 06:50 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id GAA22055; Sat, 5 Apr 1997 06:30:02 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id GAA17475; Sat, 5 Apr 1997 06:36:00 -0500 (EST) Date: Sat, 5 Apr 1997 06:36:00 -0500 (EST) From: Thomas David Rivers Message-Id: <199704051136.GAA17475@lakes.water.net> To: ponds!time.cdrom.com!jhk Subject: Re: Some significant (daresay serious) problems with 2.2.1 install & keyboard Cc: ponds!freebsd.org!hackers Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > > I'm using the 2.2.1-RELEASE from ftp.freebsd.org; with the > > boot.flp found in floppies/newer. [There is no checksum there, > > so I have no way of validating my copy...] > > And I assume that the "older" boot.flp does the same thing on this > 386? I'm just trying to establish an effective timeframe for the > breakage. Thanks! I just checked; it seems the "older" boot.flp works fine. So, the problem must have been introduced between "older" and "newer." :-) - Dave Rivers - > > Jordan From owner-freebsd-hackers Sat Apr 5 07:09:05 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA12041 for hackers-outgoing; Sat, 5 Apr 1997 07:09:05 -0800 (PST) Received: from nlsystems.com (nlsys.demon.co.uk [158.152.125.33]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA12030 for ; Sat, 5 Apr 1997 07:08:59 -0800 (PST) Received: from herring.nlsystems.com (herring.nlsystems.com [10.0.0.2]) by nlsystems.com (8.8.5/8.8.5) with SMTP id QAA22412 for ; Sat, 5 Apr 1997 16:08:48 +0100 (BST) Date: Sat, 5 Apr 1997 16:08:48 +0100 (BST) From: Doug Rabson To: hackers@freebsd.org Subject: Debugging lkms using GDB Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Ever been frustrated by the lack of symbols in LKMs when debugging? Just use kgdb over a serial line: On debugging machine # cd /usr/src/lkm/linux # make clean; make COPTS=-g # gdb -k /...path to kernel/kernel On test machine # ... install debug version of linux_mod.o ... # linux # modstat Type Id Off Loadaddr Size Info Rev Module Name EXEC 0 4 f5109000 001c f510f010 1 linux_mod DEV 1 30 f5114000 0154 f514804c 1 oss_mod ... attach to remote gdb ... On debugging machine (kgdb) add-symbol-file /usr/src/lkm/linux/linux_mod.o 0xf5109020 add symbol table from file "/usr/src/lkm/linux/linux_mod.o" at text_addr = 0xf5109020? (y or n) y (kgdb) ... debug the lkm with full debugging symbols! ... The important thing is to add 0x20 to the address the module was loaded at (I guess this is the size of the a.out header). -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 181 951 1891 From owner-freebsd-hackers Sat Apr 5 07:11:28 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA12198 for hackers-outgoing; Sat, 5 Apr 1997 07:11:28 -0800 (PST) Received: from spooky.rwwa.com (rwwa.com [198.115.177.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA12189 for ; Sat, 5 Apr 1997 07:11:24 -0800 (PST) Received: from spooky.rwwa.com (localhost.rwwa.com [127.0.0.1]) by spooky.rwwa.com (8.7.5/8.7.3) with ESMTP id KAA16932; Sat, 5 Apr 1997 10:10:52 -0500 (EST) Message-Id: <199704051510.KAA16932@spooky.rwwa.com> X-Mailer: exmh version 2.0gamma 1/27/96 To: Arne Henrik Juul cc: bwithrow@BayNetworks.COM, hackers@freebsd.org Subject: Re: Suggested change to rc In-reply-to: Your message of "Sat, 05 Apr 1997 01:23:07 +0200." <199704042323.PAA10718@freefall.freebsd.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 05 Apr 1997 10:10:52 -0500 From: Robert Withrow Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk arnej@stud.fim.ntnu.no said: :- I can't see that you should need an eval for this. Because then the ypcat gets evaluated when /etc/sysconf is sourced, *before* yp is up... ----------------------------------------------------------------------------- Robert Withrow, Tel: +1 617 592 8935, Net: witr@rwwa.COM From owner-freebsd-hackers Sat Apr 5 07:38:15 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA13092 for hackers-outgoing; Sat, 5 Apr 1997 07:38:15 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA13083 for ; Sat, 5 Apr 1997 07:38:10 -0800 (PST) Received: from ntws (ntws.etinc.com [204.141.95.142]) by etinc.com (8.8.3/8.6.9) with SMTP id KAA18581; Sat, 5 Apr 1997 10:42:53 -0500 (EST) Message-Id: <3.0.32.19970405103644.00b0bbd0@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Sat, 05 Apr 1997 10:36:52 -0500 To: Luigi Rizzo From: dennis Subject: Re: 2.2.1R NFS and FTP load problem FOUND Cc: hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 03:37 AM 4/5/97 +0200, you wrote: >> I then remembered that I had noticed that 2.2R was a bit clunky with >> 8 meg of ram, so I popped in another 8 meg, and the problems >> disappeared. So, it seems, ftp and nfs loads cant be done on at >> 8 meg system. > >the curios thing is that I installed 2.2.1-R on a Pentium with 8MB >using FTP and a PCI ne2000 clone. No problems there, I even rebuilt >the kernel several times on that system. Are you sure your original >8MB of RAM is ok ? Did you load from a LOCAL FTP server, or over the net? I suspect that a net load would be a LOT slower and thus have different load requirements int terms of buffers and memory. I didnt remove the orig 8 mb...I just added 8 more. Plus, this occurred on 2 different machines....with clearly different ram modules. I've also been using these machines before, so its nothing new. Dennis > > Luigi > > From owner-freebsd-hackers Sat Apr 5 07:48:10 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA13461 for hackers-outgoing; Sat, 5 Apr 1997 07:48:10 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA13456 for ; Sat, 5 Apr 1997 07:48:06 -0800 (PST) Received: from ntws (ntws.etinc.com [204.141.95.142]) by etinc.com (8.8.3/8.6.9) with SMTP id KAA18644; Sat, 5 Apr 1997 10:53:01 -0500 (EST) Message-Id: <3.0.32.19970405104652.00b0a830@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Sat, 05 Apr 1997 10:46:55 -0500 To: Mark Mayo From: dennis Subject: Re: 2.2.1R NFS and FTP load problem FOUND Cc: Michael Smith , hackers@FreeBSD.ORG Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk At 10:57 PM 4/4/97 -0500, Mark Mayo wrote: >On Fri, 4 Apr 1997, dennis wrote: > >> >> I then remembered that I had noticed that 2.2R was a bit clunky with >> 8 meg of ram, so I popped in another 8 meg, and the problems >> disappeared. So, it seems, ftp and nfs loads cant be done on at >> 8 meg system. >> > >That's odd.. I had no problems installing 2.2R or 2.2.1R over NFS with an >old ISA-only 486 with 8MB of RAM. What are some more specifics of the >machine? It must be a flaky piece of hardware or something bizarre. Or >maybe just bad luck ;-) Well, I might think that also, except that I had exactly the same problem on 2 very different machines. I also tried on 2 completely different network segments, 1 rather busy and 1 practically dedicated. The specifics of the machine(s) were: Box 1: '486 100 PCI MB w/8 MB ram 600MB IDE HDD Dec-chip PCI card AND ne2000 clone (tried both) Cheap VGA card Box 2: 100Mhz Pentium 600MB IDE HDD and 130MB HDD (tried both with large and small swap spaces) Dec-chip PCI and ne2000 clone (tried both) monochrome adapter Nothing fancy, and the same MBs we use for lots of other systems and for the systems that we sell...so the MBs are (or at least have been) compatible. Dennis From owner-freebsd-hackers Sat Apr 5 07:50:53 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA13564 for hackers-outgoing; Sat, 5 Apr 1997 07:50:53 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA13559 for ; Sat, 5 Apr 1997 07:50:49 -0800 (PST) Received: from ntws (ntws.etinc.com [204.141.95.142]) by etinc.com (8.8.3/8.6.9) with SMTP id KAA18656; Sat, 5 Apr 1997 10:54:33 -0500 (EST) Message-Id: <3.0.32.19970405104824.00a54100@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Sat, 05 Apr 1997 10:48:26 -0500 To: Terry Lambert From: dennis Subject: Re: 2.2.1R NFS and FTP load problem FOUND Cc: msmith@atrad.adelaide.edu.au, hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 09:24 PM 4/4/97 -0700, Terry Lambert wrote: >> I then remembered that I had noticed that 2.2R was a bit clunky with >> 8 meg of ram, so I popped in another 8 meg, and the problems >> disappeared. So, it seems, ftp and nfs loads cant be done on at >> 8 meg system. > >More likely, your RAM is flakey, and you are simply changing the >usage pattern such that the flakey RAM is no longer critical path. > >Any chance of you playing a game of "shuffle the SIMMs" to verify >this? As I said, 2 different machines were involved, so the likelihood of it being flakey ram is rather unlikely. Plus the same ram was used when it works (just added another bank)..... Dennis > > > Regards, > Terry Lambert > terry@lambert.org >--- >Any opinions in this posting are my own and not those of my present >or previous employers. > > From owner-freebsd-hackers Sat Apr 5 07:52:20 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA13626 for hackers-outgoing; Sat, 5 Apr 1997 07:52:20 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA13612 for ; Sat, 5 Apr 1997 07:52:14 -0800 (PST) Received: from ntws (ntws.etinc.com [204.141.95.142]) by etinc.com (8.8.3/8.6.9) with SMTP id KAA18677 for ; Sat, 5 Apr 1997 10:57:15 -0500 (EST) Message-Id: <3.0.32.19970405105105.00a53e70@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Sat, 05 Apr 1997 10:51:08 -0500 To: hackers@freebsd.org From: dennis Subject: Re: 2.2.1R NFS and FTP load problem FOUND Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 10:57 PM 4/4/97 -0500, you wrote: >On Fri, 4 Apr 1997, dennis wrote: > >> >> I then remembered that I had noticed that 2.2R was a bit clunky with >> 8 meg of ram, so I popped in another 8 meg, and the problems >> disappeared. So, it seems, ftp and nfs loads cant be done on at >> 8 meg system. >> > >That's odd.. I had no problems installing 2.2R or 2.2.1R over NFS with an >old ISA-only 486 with 8MB of RAM. Since there seems to be widespread doubt about this conclusion, if someone can get me a boot disk with some diagnostics for memory usage I'd be more than happy to try to track this down (as its no doubt going to bite me in the future)....theres a limited amount of info that I can gather with the install disk unfortunatly..... Dennis From owner-freebsd-hackers Sat Apr 5 08:00:19 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA13946 for hackers-outgoing; Sat, 5 Apr 1997 08:00:19 -0800 (PST) Received: from root.com (implode.root.com [198.145.90.17]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA13938; Sat, 5 Apr 1997 08:00:10 -0800 (PST) Received: from localhost (localhost [127.0.0.1]) by root.com (8.8.5/8.6.5) with SMTP id IAA17284; Sat, 5 Apr 1997 08:00:52 -0800 (PST) Message-Id: <199704051600.IAA17284@root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: Thomas David Rivers cc: bde@freebsd.org, freebsd-hackers@freebsd.org Subject: Re: "dup alloc"/interrupts and Xresume11() - splbio() not working???? In-reply-to: Your message of "Sat, 05 Apr 1997 06:34:14 EST." <199704051134.GAA17448@lakes.water.net> From: David Greenman Reply-To: dg@root.com Date: Sat, 05 Apr 1997 08:00:52 -0800 Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk >> >> > Anyway; I'm seeing things like this, and want to make sure >> >I understand everything that's going on: >> > >> > 1) Xresume11() is the vector that is taken when interrupt >> > 11 is triggered - right? >> >> Wrong - _Xintr11 is the vector. >> >> Bruce >> > > Well - umm; Ok - sure... > > But then, why does _Xresume11 call ahaintr()? Which eventually >gets me down to scsi_done() with _cpl set to splbio? (Or, have >I misunderstood something else and this isn't necessarily a "bad" >thing to happen?) Bruce has probably already responded to this, but I'll jump in... The cpl is set to the mask that is associated with the interrupt (splbio for int 11 in this case) very early in the interrupt processing code, before the "Xresume11" in the traceback. See the code in i386/isa/vector.s - the non- fast interrupt case. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-freebsd-hackers Sat Apr 5 08:02:27 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA14063 for hackers-outgoing; Sat, 5 Apr 1997 08:02:27 -0800 (PST) Received: from bookcase.com (root@notes.bookcase.com [207.22.115.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA14058 for ; Sat, 5 Apr 1997 08:02:23 -0800 (PST) Received: from localhost (chadf@localhost) by bookcase.com (8.8.5/8.7.3) with SMTP id KAA21060; Sat, 5 Apr 1997 10:05:54 -0500 (EST) Date: Sat, 5 Apr 1997 11:05:53 -0400 (EST) From: "Chad M. Fraleigh" X-Sender: chadf@notes To: Michael Smith cc: hackers@FreeBSD.ORG Subject: Re: Code maintenance In-Reply-To: <199704050505.OAA26477@genesis.atrad.adelaide.edu.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > Anyway.. now I'm working on a way to trace kernel memory leaks. > > 'vmstat -m' is your friend! That seems somewhat useful, but I was thinking along the lines of exact details.. like what function the memory was allocated it (stack backtrace). Perhaps even to the source line number (assuming the -g kernel info was avaliable). -Chad From owner-freebsd-hackers Sat Apr 5 08:14:34 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA15022 for hackers-outgoing; Sat, 5 Apr 1997 08:14:34 -0800 (PST) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA15016 for ; Sat, 5 Apr 1997 08:14:29 -0800 (PST) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.5/8.7.3) id BAA27630; Sun, 6 Apr 1997 01:44:18 +0930 (CST) From: Michael Smith Message-Id: <199704051614.BAA27630@genesis.atrad.adelaide.edu.au> Subject: Re: Code maintenance In-Reply-To: from "Chad M. Fraleigh" at "Apr 5, 97 11:05:53 am" To: chadf@bookcase.com (Chad M. Fraleigh) Date: Sun, 6 Apr 1997 01:44:17 +0930 (CST) Cc: msmith@atrad.adelaide.edu.au, hackers@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Chad M. Fraleigh stands accused of saying: > > > > Anyway.. now I'm working on a way to trace kernel memory leaks. > > > > 'vmstat -m' is your friend! > > That seems somewhat useful, but I was thinking along the lines of > exact details.. like what function the memory was allocated it (stack > backtrace). Perhaps even to the source line number (assuming the -g kernel > info was avaliable). Hmm, nasty. You could add an extra field to the memory block header with the address of the caller in it, which would help a lot. (With that you could nominate caller and possibly line number, but a full backtrace for every block is a big ask). > -Chad -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ From owner-freebsd-hackers Sat Apr 5 08:33:32 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA15537 for hackers-outgoing; Sat, 5 Apr 1997 08:33:32 -0800 (PST) Received: from pat.idt.unit.no (0@pat.idt.unit.no [129.241.103.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA15527 for ; Sat, 5 Apr 1997 08:33:25 -0800 (PST) Received: from idt.unit.no (26850@kamelia.idt.unit.no [129.241.111.27]) by pat.idt.unit.no (8.8.5/8.8.5) with ESMTP id SAA00567; Sat, 5 Apr 1997 18:29:45 +0200 (MET DST) Message-Id: <199704051629.SAA00567@pat.idt.unit.no> To: terry@lambert.org Cc: dfr@nlsystems.com, dg@root.com, ponds!rivers@dg-rtp.dg.com, freebsd-hackers@freebsd.org Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) In-Reply-To: Your message of "Fri, 4 Apr 1997 21:21:42 -0700 (MST)" References: <199704050421.VAA20745@phaeton.artisoft.com> X-Mailer: Mew version 1.06 on Emacs 19.34.1 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Date: Sat, 05 Apr 1997 18:29:37 +0200 From: Tor Egge Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > Proper locking on NFS will help against the crash I experienced, but > > if the NFS server goes down at the wrong moment, a call to getnewvnode > > may hang until the NFS server is back again. > > This will always be the case, right? > > Or do you mean the case where an attempt to vclean() a vnode is against > an NFS vnode with the server down? When the first vnode on the so-called free list is a NFS vnode, where VOP_INACTIVE is blocked (due to nfs_remove blocking when the NFS server is down), that vnode cannot be reused for a long time (until the nfs server is up again). When getnewvnode picks that vnode it waits for the lock to drain (in vclean). if the routine that called getnewvnode is ffs_vget, the ffs_inode_hash_lock is set while getnewvnode blocks. This has the side effect that any further calls to ffs_vget will block. This might block any creation of new files on local ufs file systems while the NFS server is down. - Tor Egge From owner-freebsd-hackers Sat Apr 5 08:42:24 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA15899 for hackers-outgoing; Sat, 5 Apr 1997 08:42:24 -0800 (PST) Received: from nlsystems.com (nlsys.demon.co.uk [158.152.125.33]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA15891 for ; Sat, 5 Apr 1997 08:42:19 -0800 (PST) Received: from herring.nlsystems.com (herring.nlsystems.com [10.0.0.2]) by nlsystems.com (8.8.5/8.8.5) with SMTP id RAA25813; Sat, 5 Apr 1997 17:40:31 +0100 (BST) Date: Sat, 5 Apr 1997 17:40:31 +0100 (BST) From: Doug Rabson To: Tor Egge cc: terry@lambert.org, dg@root.com, ponds!rivers@dg-rtp.dg.com, freebsd-hackers@freebsd.org Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) In-Reply-To: <199704051629.SAA00567@pat.idt.unit.no> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Sat, 5 Apr 1997, Tor Egge wrote: > > > Proper locking on NFS will help against the crash I experienced, but > > > if the NFS server goes down at the wrong moment, a call to getnewvnode > > > may hang until the NFS server is back again. > > > > This will always be the case, right? > > > > Or do you mean the case where an attempt to vclean() a vnode is against > > an NFS vnode with the server down? > > When the first vnode on the so-called free list is a NFS vnode, where > VOP_INACTIVE is blocked (due to nfs_remove blocking when the NFS > server is down), that vnode cannot be reused for a long time (until > the nfs server is up again). > > When getnewvnode picks that vnode it waits for the lock to drain (in > vclean). if the routine that called getnewvnode is ffs_vget, the > ffs_inode_hash_lock is set while getnewvnode blocks. This has the > side effect that any further calls to ffs_vget will block. This might > block any creation of new files on local ufs file systems while the > NFS server is down. I know about this problem. Unfortunately the only way of fixing it is probably to keep private pools of vnodes for each filesystem (which Terry has been badgering me about for *years*). It might happen but it is a fair amount of work. For the moment, I will settle with a system which is stable, even it does hang when a server fails. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 181 951 1891 From owner-freebsd-hackers Sat Apr 5 08:54:48 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA16352 for hackers-outgoing; Sat, 5 Apr 1997 08:54:48 -0800 (PST) Received: from bookcase.com (root@notes.bookcase.com [207.22.115.2]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA16345 for ; Sat, 5 Apr 1997 08:54:44 -0800 (PST) Received: from localhost (chadf@localhost) by bookcase.com (8.8.5/8.7.3) with SMTP id KAA21188; Sat, 5 Apr 1997 10:58:07 -0500 (EST) Date: Sat, 5 Apr 1997 11:58:05 -0400 (EST) From: "Chad M. Fraleigh" X-Sender: chadf@notes To: Michael Smith cc: hackers@FreeBSD.ORG Subject: Re: Code maintenance In-Reply-To: <199704051614.BAA27630@genesis.atrad.adelaide.edu.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Hmm, nasty. You could add an extra field to the memory block header > with the address of the caller in it, which would help a lot. (With > that you could nominate caller and possibly line number, but a full > backtrace for every block is a big ask). Now really, if you compress redundent stack info. The principle is simple.. Log each kernel malloc and free to a file (actually buffering it in memory first and removing as much malloc/free's as possible). You would only log the stack changes (since probably 50-80% might be redundent). For example you have the following call structure: func1->func2->func3->func4->line20->malloc(20) = func1->func2->func3->func4->line42->malloc(541) = func1->func2->func3->func4->line53->free() func1->func2->func3->func5->line14->malloc(256) = This would produce a log of: malloc size = 20 addr = stack repeat = 0 stack = func1->func2->func3->func4->line20 malloc size = 541 addr = stack repeat = 4 stack = line42 free addr = malloc size = 256 addr = stack repeat = 3 stack = func5->line14 And if all of this happened in the buffer before writting to disk, only the second and fourth malloc would be logged (the other two would cancel out). Then a utility program would read the file.. cancel out anything that was free-ed. Given a list of debugging info (from nm -a kernel) it would produce a list of what was allocated the memory. Then you could visually see that such-and-such allocted some memory over and over, but never released it. There will of course be some allocation at the begining that will never be released through the life of the boot, but this shouldn't be too hard for someone to reconize. It would have a set file size limit (so you could say don't use more then 32Mb in logging for example). The hard parts in the implimentation would be hooking the mount function so it knew when it could start writting to disk... getting the backtrace info (it's probably not that hard.. I just never tried to do it before), and last would be so you could boot once, have it log, and boot it the next time and not have it log anymore (since now you'd want to anylize it).. some kernel flag I guess or maybe it would only log if the file didn't exist already. I mean a memory tracer like this wouldn't be a standard thing.. it would be purely development that someone could enable (right before a version release for example) to get out some memory glitches. Heck.. if someone has a big enough drive to log on (Gb or so), you could run it for awhile and catch alot of things. Still think it's not viable? -Chad From owner-freebsd-hackers Sat Apr 5 08:58:02 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id IAA16498 for hackers-outgoing; Sat, 5 Apr 1997 08:58:02 -0800 (PST) Received: from pat.idt.unit.no (0@pat.idt.unit.no [129.241.103.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id IAA16489 for ; Sat, 5 Apr 1997 08:57:57 -0800 (PST) Received: from idt.unit.no (26850@kamelia.idt.unit.no [129.241.111.27]) by pat.idt.unit.no (8.8.5/8.8.5) with ESMTP id SAA01155; Sat, 5 Apr 1997 18:54:22 +0200 (MET DST) Message-Id: <199704051654.SAA01155@pat.idt.unit.no> To: dfr@nlsystems.com Cc: terry@lambert.org, dg@root.com, ponds!rivers@dg-rtp.dg.com, freebsd-hackers@freebsd.org Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) In-Reply-To: Your message of "Sat, 5 Apr 1997 17:40:31 +0100 (BST)" References: X-Mailer: Mew version 1.06 on Emacs 19.34.1 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Date: Sat, 05 Apr 1997 18:54:21 +0200 From: Tor Egge Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > I know about this problem. Unfortunately the only way of fixing it is > probably to keep private pools of vnodes for each filesystem (which Terry > has been badgering me about for *years*). It might happen but it is a > fair amount of work. For the moment, I will settle with a system which is > stable, even it does hang when a server fails. Two other ways: - Let getnewvnode call VOP_ISLOCKED and skip locked nodes on the so-called free list. - Delay putting vnodes onto the so-called free list until VOP_INACTIVE has been called, and use a flag to indicate that a vnode is not on the so-called free list even when v_usecount is 0. Add needed checks where the code currently assumes that v_usecount==0 means that the vnode is on the so-called free list. - Tor Egge From owner-freebsd-hackers Sat Apr 5 09:13:51 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id JAA17109 for hackers-outgoing; Sat, 5 Apr 1997 09:13:51 -0800 (PST) Received: from nlsystems.com (nlsys.demon.co.uk [158.152.125.33]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA17101 for ; Sat, 5 Apr 1997 09:13:47 -0800 (PST) Received: from herring.nlsystems.com (herring.nlsystems.com [10.0.0.2]) by nlsystems.com (8.8.5/8.8.5) with SMTP id SAA15191; Sat, 5 Apr 1997 18:11:13 +0100 (BST) Date: Sat, 5 Apr 1997 18:11:13 +0100 (BST) From: Doug Rabson To: Tor Egge cc: terry@lambert.org, dg@root.com, ponds!rivers@dg-rtp.dg.com, freebsd-hackers@freebsd.org Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) In-Reply-To: <199704051654.SAA01155@pat.idt.unit.no> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Sat, 5 Apr 1997, Tor Egge wrote: > > I know about this problem. Unfortunately the only way of fixing it is > > probably to keep private pools of vnodes for each filesystem (which Terry > > has been badgering me about for *years*). It might happen but it is a > > fair amount of work. For the moment, I will settle with a system which is > > stable, even it does hang when a server fails. > > Two other ways: > > - Let getnewvnode call VOP_ISLOCKED and skip locked nodes on the > so-called free list. > > - Delay putting vnodes onto the so-called free list until > VOP_INACTIVE has been called, and use a flag to indicate that a > vnode is not on the so-called free list even when v_usecount is > 0. Add needed checks where the code currently assumes that > v_usecount==0 means that the vnode is on the so-called free list. I think I prefer the first alternative, because it seems simpler. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 181 951 1891 From owner-freebsd-hackers Sat Apr 5 11:14:53 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id LAA22576 for hackers-outgoing; Sat, 5 Apr 1997 11:14:53 -0800 (PST) Received: from who.cdrom.com (who.cdrom.com [204.216.27.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA22571 for ; Sat, 5 Apr 1997 11:14:52 -0800 (PST) Received: from vinyl.quickweb.com (vinyl.quickweb.com [206.222.77.8]) by who.cdrom.com (8.8.5/8.6.11) with ESMTP id LAA23582 for ; Sat, 5 Apr 1997 11:14:51 -0800 (PST) Received: from localhost (mark@localhost) by vinyl.quickweb.com (8.8.5/8.6.12) with SMTP id OAA28036; Sat, 5 Apr 1997 14:07:19 -0500 (EST) Date: Sat, 5 Apr 1997 14:07:19 -0500 (EST) From: Mark Mayo To: dennis cc: Michael Smith , hackers@FreeBSD.ORG Subject: Re: 2.2.1R NFS and FTP load problem FOUND In-Reply-To: <3.0.32.19970405104652.00b0a830@etinc.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Sat, 5 Apr 1997, dennis wrote: > At 10:57 PM 4/4/97 -0500, Mark Mayo wrote: > >On Fri, 4 Apr 1997, dennis wrote: > > [TALK OF 8MB FAILURES CUT] > > Well, I might think that also, except that I had exactly the same problem > on 2 very > different machines. I also tried on 2 completely different network > segments, 1 rather > busy and 1 practically dedicated. The specifics of the machine(s) were: Weird. Maybe I'm the lucky one here. Was the ne2000 clone a PCI or ISA card? The 8MB 486 I used has an SMC Ultra 16 (ISA) in it. That's the only differnece I can see (and of course, the SMC and the ne2000 use the same ed driver..). Maybe I'll pull some RAM out of the other 2 machines I have access to and try the install. -Mark ---------------------------------------------------------------------------- Mark Mayo mark@quickweb.com RingZero Comp. http://vinyl.quickweb.com/mark finger mark@quickweb.com for my PGP key and GCS code ---------------------------------------------------------------------------- University degrees are a bit like adultery: you may not want to get involved with that sort of thing, but you don't want to be thought incapable. -Sir Peter Imbert From owner-freebsd-hackers Sat Apr 5 12:01:55 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA24741 for hackers-outgoing; Sat, 5 Apr 1997 12:01:55 -0800 (PST) Received: from pdx1.world.net (pdx1.world.net [192.243.32.18]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA24736 for ; Sat, 5 Apr 1997 12:01:51 -0800 (PST) From: proff@suburbia.net Received: from suburbia.net (suburbia.net [203.4.184.1]) by pdx1.world.net (8.7.5/8.7.3) with SMTP id MAA20231 for ; Sat, 5 Apr 1997 12:04:13 -0800 (PST) Received: (qmail 20082 invoked by uid 110); 5 Apr 1997 20:01:18 -0000 Message-ID: <19970405200118.20081.qmail@suburbia.net> Subject: ipfilter build system complete To: hackers@freebsd.org Date: Sun, 6 Apr 1997 06:01:18 +1000 (EST) X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk I've finished writing the build (and module loading) system for ipfilter. A few other other code changes were also necessarily. The patches (only about 30k) are available from GNATS as `kern/3207'. wollman/DG/avalon, can you review this for me? Cheers, Julian. -- Prof. Julian Assange |If you want to build a ship, don't drum up people |together to collect wood and don't assign them tasks proff@suburbia.net |and work, but rather teach them to long for the endless proff@gnu.ai.mit.edu |immensity of the sea. -- Antoine de Saint Exupery From owner-freebsd-hackers Sat Apr 5 12:18:35 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA25425 for hackers-outgoing; Sat, 5 Apr 1997 12:18:35 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA25420 for ; Sat, 5 Apr 1997 12:18:30 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id MAA23416; Sat, 5 Apr 1997 12:59:46 -0700 From: Terry Lambert Message-Id: <199704051959.MAA23416@phaeton.artisoft.com> Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) To: Tor.Egge@idi.ntnu.no (Tor Egge) Date: Sat, 5 Apr 1997 12:59:46 -0700 (MST) Cc: terry@lambert.org, dfr@nlsystems.com, dg@root.com, ponds!rivers@dg-rtp.dg.com, freebsd-hackers@freebsd.org In-Reply-To: <199704051629.SAA00567@pat.idt.unit.no> from "Tor Egge" at Apr 5, 97 06:29:37 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > When the first vnode on the so-called free list is a NFS vnode, where > VOP_INACTIVE is blocked (due to nfs_remove blocking when the NFS > server is down), that vnode cannot be reused for a long time (until > the nfs server is up again). > > When getnewvnode picks that vnode it waits for the lock to drain (in > vclean). if the routine that called getnewvnode is ffs_vget, the > ffs_inode_hash_lock is set while getnewvnode blocks. This has the > side effect that any further calls to ffs_vget will block. This might > block any creation of new files on local ufs file systems while the > NFS server is down. OK. "Yes, this is what will happen, if the vnode at the top of the freelist has writeable data associated with it". The only real fix is to seperate the pools and murder vclean() and the other associated paraphernalia for the gloval vnode pool. This isn't a trivial undertaking with the current way the system open file table and vnode revocation works. Currently, a "revoked" vnode changes allegiance from its original FS to "deadfs" to implement the revocation at a vnode rather than a system open file table entry level. Effectively, you would impact every kernel subsystem that depended on vnode-type "open" calls (except swap, which counts as a node reference and prevents node deletion): quotas, execution classes, etc.. This isn't as bad as it sounds, since it probably needs to be done anyway, but it can be done incrementally by first removing the dependencies on the global pool before going in and hacking out the global pool altogether. That's the substance ofsome of the recent discussions. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Sat Apr 5 12:22:09 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA25607 for hackers-outgoing; Sat, 5 Apr 1997 12:22:09 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA25599 for ; Sat, 5 Apr 1997 12:22:05 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id NAA23438; Sat, 5 Apr 1997 13:03:20 -0700 From: Terry Lambert Message-Id: <199704052003.NAA23438@phaeton.artisoft.com> Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) To: Tor.Egge@idi.ntnu.no (Tor Egge) Date: Sat, 5 Apr 1997 13:03:20 -0700 (MST) Cc: dfr@nlsystems.com, terry@lambert.org, dg@root.com, ponds!rivers@dg-rtp.dg.com, freebsd-hackers@freebsd.org In-Reply-To: <199704051654.SAA01155@pat.idt.unit.no> from "Tor Egge" at Apr 5, 97 06:54:21 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > I know about this problem. Unfortunately the only way of fixing it is > > probably to keep private pools of vnodes for each filesystem (which Terry > > has been badgering me about for *years*). It might happen but it is a > > fair amount of work. For the moment, I will settle with a system which is > > stable, even it does hang when a server fails. > > Two other ways: > > - Let getnewvnode call VOP_ISLOCKED and skip locked nodes on the > so-called free list. There's no holder, so it isn't locked. If it were locked via VOP_LOCK, then you wouldn'r be having this problem in the first place, since a VOP_LOCK will prevent a node from being vclean()'ed (that's the whole point of the NFS changes Doug has been trying to deal with, if you followed his postings on them). > - Delay putting vnodes onto the so-called free list until > VOP_INACTIVE has been called, and use a flag to indicate that a > vnode is not on the so-called free list even when v_usecount is > 0. Add needed checks where the code currently assumes that > v_usecount==0 means that the vnode is on the so-called free list. This is hard. It would much easier to go to a reference as a counting semaphore (vp->v_count++;) in all cases where the reference was held, and establish a "hold" on behalf of the recycler. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Sat Apr 5 12:48:30 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA26782 for hackers-outgoing; Sat, 5 Apr 1997 12:48:30 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA26777 for ; Sat, 5 Apr 1997 12:48:27 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id NAA23575; Sat, 5 Apr 1997 13:29:45 -0700 From: Terry Lambert Message-Id: <199704052029.NAA23575@phaeton.artisoft.com> Subject: Re: 2.2.1R NFS and FTP load problem FOUND To: dennis@etinc.com (dennis) Date: Sat, 5 Apr 1997 13:29:45 -0700 (MST) Cc: terry@lambert.org, msmith@atrad.adelaide.edu.au, hackers@freebsd.org In-Reply-To: <3.0.32.19970405104824.00a54100@etinc.com> from "dennis" at Apr 5, 97 10:48:26 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > >More likely, your RAM is flakey, and you are simply changing the > >usage pattern such that the flakey RAM is no longer critical path. > > > >Any chance of you playing a game of "shuffle the SIMMs" to verify > >this? > > As I said, 2 different machines were involved, so the likelihood of > it being flakey ram is rather unlikely. Plus the same ram was used > when it works (just added another bank)..... The same code was not necessarily loaded into the same spot. If there was more RAM, the code exhibiting the problem could have been loaded into the additional RAM instead of the flakey RAM. This would depend highly on the exact usage patter by processes invoked during install, including, but not limited to, their invocation order, etc.. There is no evidence, short of a shuffle, that will guarantee you that the problem is resolved, not masked by the additional RAM. You will need to do a full shuffle so that an 8M situation without the possibly flakey RAM is also tested (and verified to fail). Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Sat Apr 5 12:50:14 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA26896 for hackers-outgoing; Sat, 5 Apr 1997 12:50:14 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id MAA26891 for ; Sat, 5 Apr 1997 12:50:11 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id NAA23589; Sat, 5 Apr 1997 13:32:09 -0700 From: Terry Lambert Message-Id: <199704052032.NAA23589@phaeton.artisoft.com> Subject: Re: 2.2.1R NFS and FTP load problem FOUND To: dennis@etinc.com (dennis) Date: Sat, 5 Apr 1997 13:32:09 -0700 (MST) Cc: hackers@freebsd.org In-Reply-To: <3.0.32.19970405105105.00a53e70@etinc.com> from "dennis" at Apr 5, 97 10:51:08 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > >That's odd.. I had no problems installing 2.2R or 2.2.1R over NFS with an > >old ISA-only 486 with 8MB of RAM. > > Since there seems to be widespread doubt about this conclusion, > if someone can get me a boot disk with some diagnostics for > memory usage I'd be more than happy to try to track this down (as > its no doubt going to bite me in the future)....theres a limited amount > of info that I can gather with the install disk unfortunatly..... A good memory diagnostic is to boot the FreeBSD install; if it fails in 8M or more, you probably have flakey RAM. 8-). Seriously, short of a hardware tester of some kind, there are no real good RAM diagnostics that I know of, and certainly no software ones that are anything but high-level (and therefore mostly useless). Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Sat Apr 5 12:56:26 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA27178 for hackers-outgoing; Sat, 5 Apr 1997 12:56:26 -0800 (PST) Received: from peedub.gj.org (newpc.muc.ditec.de [194.120.126.33]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA27172 for ; Sat, 5 Apr 1997 12:56:21 -0800 (PST) Received: from peedub.gj.org (localhost [127.0.0.1]) by peedub.gj.org (8.8.5/8.6.9) with ESMTP id WAA13307 for ; Sat, 5 Apr 1997 22:56:30 GMT Message-Id: <199704052256.WAA13307@peedub.gj.org> X-Mailer: exmh version 2.0gamma 1/27/96 To: freebsd-hackers@freefall.FreeBSD.org Subject: Re: ipfilter build system complete Reply-To: Gary Jennejohn In-reply-to: Your message of "Sun, 06 Apr 1997 06:01:18 +1000." <19970405200118.20081.qmail@suburbia.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 05 Apr 1997 22:56:30 +0000 From: Gary Jennejohn Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk proff@suburbia.net writes: > >I've finished writing the build (and module loading) system for >ipfilter. A few other other code changes were also necessarily. >The patches (only about 30k) are available from GNATS as `kern/3207'. >wollman/DG/avalon, can you review this for me? > Great ! The only comment I have right off the bat is that including osreldate.h doesn't work if you're trying to compile ipfilter into the kernel. So, I think ipfconf.h should look like: #ifdef __FreeBSD__ # if defined(KERNEL) # ifndef _KERNEL # define _KERNEL # endif # if __FreeBSD__=3 # define __FreeBSD_version 300000 # endif # if __FreeBSD__=2 # define __FreeBSD_version 220000 # endif # else # include # endif #endif this isn't absolutely correct, unfortunatley, since it doesn't handle 2.1.X. For that we'd need to use BSD from param.h, but ipconf.h gets included much too soon for that. --- Gary Jennejohn Home - Gary.Jennejohn@munich.netsurf.de Work - gjennejohn@frt.dec.com From owner-freebsd-hackers Sat Apr 5 12:59:53 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA27305 for hackers-outgoing; Sat, 5 Apr 1997 12:59:53 -0800 (PST) Received: from pat.idt.unit.no (0@pat.idt.unit.no [129.241.103.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA27300 for ; Sat, 5 Apr 1997 12:59:49 -0800 (PST) Received: from idt.unit.no (26850@kamelia.idt.unit.no [129.241.111.27]) by pat.idt.unit.no (8.8.5/8.8.5) with ESMTP id WAA06400; Sat, 5 Apr 1997 22:56:11 +0200 (MET DST) Message-Id: <199704052056.WAA06400@pat.idt.unit.no> To: terry@lambert.org Cc: dfr@nlsystems.com, dg@root.com, ponds!rivers@dg-rtp.dg.com, freebsd-hackers@freebsd.org Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) In-Reply-To: Your message of "Sat, 5 Apr 1997 13:03:20 -0700 (MST)" References: <199704052003.NAA23438@phaeton.artisoft.com> X-Mailer: Mew version 1.06 on Emacs 19.34.1 Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Date: Sat, 05 Apr 1997 22:56:08 +0200 From: Tor Egge Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > - Let getnewvnode call VOP_ISLOCKED and skip locked nodes on the > > so-called free list. > > There's no holder, so it isn't locked. If it were locked via VOP_LOCK, > then you wouldn'r be having this problem in the first place, since a > VOP_LOCK will prevent a node from being vclean()'ed (that's the whole > point of the NFS changes Doug has been trying to deal with, if you > followed his postings on them). Wrong. Under some circumstances, the vnode is locked. See the end of vputrele. - Tor Egge From owner-freebsd-hackers Sat Apr 5 13:33:21 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA28418 for hackers-outgoing; Sat, 5 Apr 1997 13:33:21 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id NAA28412 for ; Sat, 5 Apr 1997 13:33:19 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id OAA23667; Sat, 5 Apr 1997 14:14:39 -0700 From: Terry Lambert Message-Id: <199704052114.OAA23667@phaeton.artisoft.com> Subject: Re: kern/3184: vnodes are used after they are freed. (dup alloc?) To: Tor.Egge@idi.ntnu.no (Tor Egge) Date: Sat, 5 Apr 1997 14:14:39 -0700 (MST) Cc: terry@lambert.org, dfr@nlsystems.com, dg@root.com, ponds!rivers@dg-rtp.dg.com, freebsd-hackers@freebsd.org In-Reply-To: <199704052056.WAA06400@pat.idt.unit.no> from "Tor Egge" at Apr 5, 97 10:56:08 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > - Let getnewvnode call VOP_ISLOCKED and skip locked nodes on the > > > so-called free list. > > > > There's no holder, so it isn't locked. If it were locked via VOP_LOCK, > > then you wouldn'r be having this problem in the first place, since a > > VOP_LOCK will prevent a node from being vclean()'ed (that's the whole > > point of the NFS changes Doug has been trying to deal with, if you > > followed his postings on them). > > Wrong. Under some circumstances, the vnode is locked. See the end of vputrele. I should have said that there's no *guarantee* that there is a holder or that it is locked. The approach will fix some circumstances, but will fail for others without the *guarantee*. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Sat Apr 5 13:51:01 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA29215 for hackers-outgoing; Sat, 5 Apr 1997 13:51:01 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA29210 for ; Sat, 5 Apr 1997 13:50:57 -0800 (PST) Received: from dialup-usr11.etinc.com (dialup-usr11.etinc.com [204.141.95.132]) by etinc.com (8.8.3/8.6.9) with SMTP id QAA20568; Sat, 5 Apr 1997 16:54:16 -0500 (EST) Message-Id: <3.0.32.19970405164610.00ac1574@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Sat, 05 Apr 1997 16:46:15 -0500 To: Terry Lambert From: dennis Subject: Re: 2.2.1R NFS and FTP load problem FOUND Cc: terry@lambert.org, msmith@atrad.adelaide.edu.au, hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 01:29 PM 4/5/97 -0700, Terry Lambert wrote: >> >More likely, your RAM is flakey, and you are simply changing the >> >usage pattern such that the flakey RAM is no longer critical path. >> > >> >Any chance of you playing a game of "shuffle the SIMMs" to verify >> >this? >> >> As I said, 2 different machines were involved, so the likelihood of >> it being flakey ram is rather unlikely. Plus the same ram was used >> when it works (just added another bank)..... > >The same code was not necessarily loaded into the same spot. If there >was more RAM, the code exhibiting the problem could have been loaded >into the additional RAM instead of the flakey RAM. This would depend >highly on the exact usage patter by processes invoked during install, >including, but not limited to, their invocation order, etc.. > >There is no evidence, short of a shuffle, that will guarantee you >that the problem is resolved, not masked by the additional RAM. > >You will need to do a full shuffle so that an 8M situation without the >possibly flakey RAM is also tested (and verified to fail). Ok, but why can I load any OTHER version of Freebsd on these machines, and what is the chance of a random ram problem having the exact same failure on 2 different machines? I dont buy it.....sounds like a "mine works so you're wrong" answer to me. db From owner-freebsd-hackers Sat Apr 5 13:52:05 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA29301 for hackers-outgoing; Sat, 5 Apr 1997 13:52:05 -0800 (PST) Received: from etinc.com (et-gw-fr1.etinc.com [204.141.244.98]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA29296 for ; Sat, 5 Apr 1997 13:52:02 -0800 (PST) Received: from dialup-usr11.etinc.com (dialup-usr11.etinc.com [204.141.95.132]) by etinc.com (8.8.3/8.6.9) with SMTP id QAA20583; Sat, 5 Apr 1997 16:55:49 -0500 (EST) Message-Id: <3.0.32.19970405164743.00ac0524@etinc.com> X-Sender: dennis@etinc.com X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Sat, 05 Apr 1997 16:47:46 -0500 To: Terry Lambert From: dennis Subject: Re: 2.2.1R NFS and FTP load problem FOUND Cc: terry@lambert.org, msmith@atrad.adelaide.edu.au, hackers@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk At 01:29 PM 4/5/97 -0700, Terry Lambert wrote: >> >More likely, your RAM is flakey, and you are simply changing the >> >usage pattern such that the flakey RAM is no longer critical path. >> > >> >Any chance of you playing a game of "shuffle the SIMMs" to verify >> >this? >> >> As I said, 2 different machines were involved, so the likelihood of >> it being flakey ram is rather unlikely. Plus the same ram was used >> when it works (just added another bank)..... > >The same code was not necessarily loaded into the same spot. If there >was more RAM, the code exhibiting the problem could have been loaded >into the additional RAM instead of the flakey RAM. This would depend >highly on the exact usage patter by processes invoked during install, >including, but not limited to, their invocation order, etc.. I think you missed the "2 different machines" part.... > >There is no evidence, short of a shuffle, that will guarantee you >that the problem is resolved, not masked by the additional RAM. > >You will need to do a full shuffle so that an 8M situation without the >possibly flakey RAM is also tested (and verified to fail). > > > Regards, > Terry Lambert > terry@lambert.org >--- >Any opinions in this posting are my own and not those of my present >or previous employers. > > From owner-freebsd-hackers Sat Apr 5 15:07:20 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA05429 for hackers-outgoing; Sat, 5 Apr 1997 15:07:20 -0800 (PST) Received: from almond.elite.net (root@almond.elite.net [205.199.220.1]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA05421 for ; Sat, 5 Apr 1997 15:07:17 -0800 (PST) Received: (from jpm@localhost) by almond.elite.net (8.8.3/ELITE) id PAA09272 for freebsd-hackers@freebsd.org; Sat, 5 Apr 1997 15:07:11 -0800 (PST) From: Jon Moldenhauer Message-Id: <199704052307.PAA09272@almond.elite.net> Subject: mail archive? To: freebsd-hackers@freebsd.org Date: Sat, 5 Apr 1997 15:07:11 -0800 (PST) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Is there a way to get direct access to the mail archive used by the search engine on www.freebsd.org? I find the search engine unusable when looking for recent messages or specific threads. Jonathon Elite.Net System Administrator From owner-freebsd-hackers Sat Apr 5 15:23:27 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA07443 for hackers-outgoing; Sat, 5 Apr 1997 15:23:27 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id PAA07431 for ; Sat, 5 Apr 1997 15:23:23 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id QAA27120; Sat, 5 Apr 1997 16:04:42 -0700 From: Terry Lambert Message-Id: <199704052304.QAA27120@phaeton.artisoft.com> Subject: Re: 2.2.1R NFS and FTP load problem FOUND To: dennis@etinc.com (dennis) Date: Sat, 5 Apr 1997 16:04:42 -0700 (MST) Cc: terry@lambert.org, msmith@atrad.adelaide.edu.au, hackers@freebsd.org In-Reply-To: <3.0.32.19970405164610.00ac1574@etinc.com> from "dennis" at Apr 5, 97 04:46:15 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > >There is no evidence, short of a shuffle, that will guarantee you > >that the problem is resolved, not masked by the additional RAM. > > > >You will need to do a full shuffle so that an 8M situation without the > >possibly flakey RAM is also tested (and verified to fail). > > Ok, but why can I load any OTHER version of Freebsd on these machines, Possibly because it is a bug in FreeBSD, but since you are in the vast minority, probably because the usage patterns are different between revisions and you are just "lucky". 8-(. > and what is the chance of a random ram problem having the exact > same failure on 2 different machines? Small, but not non-existant. > I dont buy it.....sounds like a "mine works so you're wrong" > answer to me. Do the shuffle; if it fails in all 8M configurations on your hardware, you have proven it is a FreeBSD bug. Right now, all you have is a strong feeling that it is a FreeBSD bug, and we have a strong feeling that "a million other users not having the problem can't be wrong". Circumstantially, our feeling is simpler than yours (Occam's razor) until you generate contradictory evidence. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. From owner-freebsd-hackers Sat Apr 5 15:50:44 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA09178 for hackers-outgoing; Sat, 5 Apr 1997 15:50:44 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id PAA09173 for ; Sat, 5 Apr 1997 15:50:40 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA19482; Sat, 5 Apr 1997 18:50:05 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Sat, 5 Apr 1997 18:50 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id SAA06885; Sat, 5 Apr 1997 18:11:56 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id SAA18135; Sat, 5 Apr 1997 18:17:55 -0500 (EST) Date: Sat, 5 Apr 1997 18:17:55 -0500 (EST) From: Thomas David Rivers Message-Id: <199704052317.SAA18135@lakes.water.net> To: ponds!zeta.org.au!bde, ponds!root.com!dg, ponds!freebsd.org!hackers, ponds!lakes.water.net!rivers Subject: Re: "dup alloc"/interrupts and Xresume11() - splbio() not working???? Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > >> > 1) Xresume11() is the vector that is taken when interrupt > >> > 11 is triggered - right? > >> > >> Wrong - _Xintr11 is the vector. > > > But then, why does _Xresume11 call ahaintr()? Which eventually > >gets me down to scsi_done() with _cpl set to splbio? (Or, have > >I misunderstood something else and this isn't necessarily a "bad" > >thing to happen?) > > There is no _Xresume11. Xresume11 is a private label in _Xintr11(). > It's normal for the traceback to show private labels. Nothing bad > there. Oh; I see... I wouldn't have expected a private label to show up in the debugger... That explains that... Now; as to why ahaintr() is called with _cpl shows splbio... any ideas? I was thinking that maybe I'm all confused about as well. I had understood that an splbio() would "mask" all I/O interrupts; but that doesn't make sense as something like the aha1542 isn't just going to interrupt again; is it? Seems there would want to be some queueing mechanism that would notice the _cpl and 'remember' somehow that an interrupt was pending... but I don't see that anywhere. [Of course, I could be staring it in the face and just not know what I'm looking at...] With all of that said - is my understanding that ahaintr() should not be called with I/O interrupts masked by the _cpl - or should it? - Dave Rivers - > > Bruce > From owner-freebsd-hackers Sat Apr 5 15:50:47 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA09195 for hackers-outgoing; Sat, 5 Apr 1997 15:50:47 -0800 (PST) Received: from dg-rtp.dg.com (dg-rtp.rtp.dg.com [128.222.1.2]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id PAA09185 for ; Sat, 5 Apr 1997 15:50:45 -0800 (PST) Received: by dg-rtp.dg.com (5.4R3.10/dg-rtp-v02) id AA19513; Sat, 5 Apr 1997 18:50:07 -0500 Received: from ponds by dg-rtp.dg.com.rtp.dg.com; Sat, 5 Apr 1997 18:50 EST Received: from lakes.water.net (lakes [10.0.0.3]) by ponds.water.net (8.8.3/8.7.3) with ESMTP id SAA06892; Sat, 5 Apr 1997 18:13:36 -0500 (EST) Received: (from rivers@localhost) by lakes.water.net (8.8.3/8.6.9) id SAA18145; Sat, 5 Apr 1997 18:19:36 -0500 (EST) Date: Sat, 5 Apr 1997 18:19:36 -0500 (EST) From: Thomas David Rivers Message-Id: <199704052319.SAA18145@lakes.water.net> To: khetan@iafrica.com, ponds!freefall.cdrom.com!freebsd-hackers Subject: Re: ucbvax.Berkely.EDU (was Re: Uninvited usage for sendmail) Content-Type: text Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Khetan Gajjar writes: > > On Fri, 4 Apr 1997, Thomas David Rivers wrote: > > > You likely want to fix that UUCP_RELAY definition; ucbvax.Berkeley.EDU > >no longer exists... (uunet.uu.net is probably still a good alternative.) > >Believe me, appearances of ucbvax.Berkeley.EDU in uucp !-paths is quite > >an annoyance. > > Sorry about that. Afaik, this is the stock standard freebsd.mc file > that ships with the sendmail source for FreeBSD. > > I'm afraid I don't admin that file :-( > > --- > Khetan Gajjar [ http://www.iafrica.com/~khetan] Hmmm... perhaps we want to change that; or do we simply get this from the official sendmail? - Dave Rivers - From owner-freebsd-hackers Sat Apr 5 17:14:48 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA12977 for hackers-outgoing; Sat, 5 Apr 1997 17:14:48 -0800 (PST) Received: from pcpsj.pfcs.com (tIh5NLv2V2ywjEHLZ8tCH/Wj73BB/h4W@harlan.fred.net [205.252.219.31]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id RAA12969 for ; Sat, 5 Apr 1997 17:14:42 -0800 (PST) Received: from mumps.pfcs.com (mumps.pfcs.com [192.52.69.11]) by pcpsj.pfcs.com (8.6.12/8.6.9) with SMTP id UAA04878 for ; Sat, 5 Apr 1997 20:14:30 -0500 Received: from localhost by mumps.pfcs.com with SMTP id AA18335 (5.67b/IDA-1.5 for ); Sat, 5 Apr 1997 20:14:29 -0500 To: hackers@freebsd.org Subject: Re: Suggested change to rc In-Reply-To: wirt@rwwa.com's message of "Sat, 05 Apr 1997 10:10:52 EST." <199704051510.KAA16932@spooky.rwwa.com> Date: Sat, 05 Apr 1997 20:14:28 -0500 Message-Id: <18333.860289268@mumps.pfcs.com> From: Harlan Stenn Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Robert> Because then the ypcat gets evaluated when /etc/sysconf is Robert> sourced, *before* yp is up... Well, other alternatives are: - source /etc/sysconf again, after yp is up - break up the rc file into smaller bits. Eventualy, we'll have enough small bits that we'll put them into a subdirectory and then we'll invent a scheme to invoke them in a useful order and ... H From owner-freebsd-hackers Sat Apr 5 20:07:45 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA20388 for hackers-outgoing; Sat, 5 Apr 1997 20:07:45 -0800 (PST) Received: from whistle.com (s205m131.whistle.com [207.76.205.131]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA20383 for ; Sat, 5 Apr 1997 20:07:39 -0800 (PST) Received: (from smap@localhost) by whistle.com (8.7.5/8.6.12) id UAA14692; Sat, 5 Apr 1997 20:07:08 -0800 (PST) Received: from bubba.whistle.com(207.76.205.7) by whistle.com via smap (V1.3) id sma014690; Sat Apr 5 20:06:53 1997 Received: (from archie@localhost) by bubba.whistle.com (8.7.5/8.6.12) id UAA06266; Sat, 5 Apr 1997 20:06:52 -0800 (PST) From: Archie Cobbs Message-Id: <199704060406.UAA06266@bubba.whistle.com> Subject: Re: Uses for divert sockets? In-Reply-To: <199704041820.KAA19125@lister.bogon.net> from Wes Santee at "Apr 4, 97 10:20:26 am" To: wes@bogon.net (Wes Santee) Date: Sat, 5 Apr 1997 20:06:52 -0800 (PST) Cc: hackers@freebsd.org X-Mailer: ELM [version 2.4ME+ PL25 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > The new ipdivert sockets in 2.2 sound great, but I'm not quite sure > just what I can accomplish with them outside of address translation. > Is it just for packet altering, or can entire connections be > redirected? > > For example, let's say I want some incomming connections to port 25 to > be diverted to another "special" MTA running on another port (to deal > with known spam sites "properly", for instance). Is it possible using > divert and ipfw to do this? Or is divert pretty much only for dealing > with raw IP packets that get injected back into the stream to end up > at their intended destination? You should be able to do this... pretty much anything is "possible" because you can completely change a packet, drop it, optionally have it bypass the normal routing code, create new packets, etc... -Archie ___________________________________________________________________________ Archie Cobbs * Whistle Communications, Inc. * http://www.whistle.com From owner-freebsd-hackers Sat Apr 5 21:02:47 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id VAA23460 for hackers-outgoing; Sat, 5 Apr 1997 21:02:47 -0800 (PST) Received: from scanner.worldgate.com (scanner.worldgate.com [198.161.84.3]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id VAA23439 for ; Sat, 5 Apr 1997 21:02:42 -0800 (PST) Received: from znep.com (uucp@localhost) by scanner.worldgate.com (8.8.5/8.7.3) with UUCP id WAA27674; Sat, 5 Apr 1997 22:02:38 -0700 (MST) Received: from localhost (marcs@localhost) by alive.znep.com (8.7.5/8.7.3) with SMTP id WAA20698; Sat, 5 Apr 1997 22:02:55 -0700 (MST) Date: Sat, 5 Apr 1997 22:02:54 -0700 (MST) From: Marc Slemko To: David Greenman cc: FreeBSD-hackers Subject: Re: apache like preforking apps and high loads In-Reply-To: <199704040000.QAA05674@root.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Thu, 3 Apr 1997, David Greenman wrote: > >On Thu, 3 Apr 1997, David Greenman wrote: > > > >> >When looking at Apacha like applications, one often sees extermely high > >> >load averages. Apache preforks a number of processes that all block > >> >on accept(). When a request comes in, all process are woken up and the > >> >scheduler chooses one of the now runnable processes taht will succeed in > >> >the accept(). The other go back to sleep. > >> > >> Not any more. I changed this a few days ago. Now only one process is > >> woken up. > > > >Looking at these changes and the other code around them, it appears to me > >like the net effect is that processes are added to the tail of the queue > >when they block on accept() and removed from (near) the head with > >wakeup_one. Am I reading this right, or am I getting my queue ends mixed > >up somewhere? > > You have it right, and that is the way it needs to work. I don't know > how Apache does it's # processes load balancing, but fairness can be important > in general. Of course; lots of things rely on that behavior and it can't really be changed by default. Apache's method of load balancing over processes is to not bother; just has all of them blocking in accept[1] and whatever one gets it (ie. what the kernel decides) gets it. > > >For code that does similar things to what Apache does, it would be far > >better to treat it as a stack and have wakeup_one() remove from the tail > >since you want the same processes to serve as many requests as possible. > >In the grand scheme of things it may be a bad idea to have wakeup_one() > >behave this way, but it looks like it would be useful in this case. > > On the x86 and most other architectures, the cache is of physical memory > and the majority of the Apache code is in shared pages. There isn't any > significant difference in terms of cache effects whether you wake the same > process all the time or you wake different processes. The code yes, but obviously not most of the data which should be less significant, but possibly still of some importance. I have done some testing on these things and it did appear that something was effecting a noticable slowdown when there were more children handling requests; it is true that there are other things impacting this and I was not looking at this from the OS level but from the level of the Apache code. I'll have to look at this again when I'm doing some testing and actually try the change. Would people consider it too big of a hack to add an option (setsockopt()?) to allow a process to override the default behavior on a socket, provided it actually has a useful performance impact? [1] On non-broken platforms it works this way; an amazingly large number of platforms can't handle multiple processes in accept() on the same socket at the same time, so a lockfile is used to control who accepts()s at any time. From owner-freebsd-hackers Sat Apr 5 22:51:17 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id WAA27867 for hackers-outgoing; Sat, 5 Apr 1997 22:51:17 -0800 (PST) Received: from obiwan.aceonline.com.au (obiwan.aceonline.com.au [203.103.90.67]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id WAA27862; Sat, 5 Apr 1997 22:51:06 -0800 (PST) Received: from localhost (adrian@localhost) by obiwan.aceonline.com.au (8.8.5/8.8.5) with SMTP id OAA02120; Sun, 6 Apr 1997 14:43:52 +0800 (WST) Date: Sun, 6 Apr 1997 14:43:52 +0800 (WST) From: Adrian Chadd To: hackers@freebsd.org cc: jkh@freebsd.org Subject: Re: Better late than never... The FreeBSD News! In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > > > > o A comparison between FreeBSD and Linux [special note: This topic > > reserved for someone with cojones the size of coconuts]. Hehe.. :) Anyone else wanna write something along these lines with me? I run Linux and FreeBSD at my work, if someone would like to throw me some benchmarks so I can give some speed tests, it would be great. (Although I wouldn't JUST compare speed figures..:) Cya -- Adrian Chadd | UNIX, MS-DOS and Windows ... | (also known as the Good, the bad and the | ugly..) From owner-freebsd-hackers Sat Apr 5 23:10:11 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA28522 for hackers-outgoing; Sat, 5 Apr 1997 23:10:11 -0800 (PST) Received: from narcissus.ml.org (root@brosenga.Pitzer.edu [134.173.120.201]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA28511; Sat, 5 Apr 1997 23:10:04 -0800 (PST) Received: from localhost (fullermd@localhost) by narcissus.ml.org (8.8.5/8.7.3) with SMTP id XAA16440; Sat, 5 Apr 1997 23:09:40 -0800 (PST) Date: Sat, 5 Apr 1997 23:09:40 -0800 (PST) From: The Devil Himself To: Adrian Chadd cc: hackers@freebsd.org, jkh@freebsd.org Subject: Re: Better late than never... The FreeBSD News! In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Sun, 6 Apr 1997, Adrian Chadd wrote: > > > > > > > o A comparison between FreeBSD and Linux [special note: This topic > > > reserved for someone with cojones the size of coconuts]. > > Hehe.. :) > > Anyone else wanna write something along these lines with me? > > I run Linux and FreeBSD at my work, if someone would like to throw me some > benchmarks so I can give some speed tests, it would be great. > > (Although I wouldn't JUST compare speed figures..:) > > Cya > > -- > Adrian Chadd | UNIX, MS-DOS and Windows ... > | (also known as the Good, the bad and the > | ugly..) Well, I don't have much experience w/ Linux directly, but I do have access to a linux machine here. plus, I've been saving all the FreeBSD vs Linux mails that have come through the questions list lately; I have a link to them off my home page, and I can always ... ahem... 'nudge' the results. Although, comparing FreeBSD to Linux, I don't think the results would even NEED nudging... :-} *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* |FreeBSD is good. FreeBSD is our friend. UNIX is our god.| *Micro$oft is bad. Micro$oft causes problems.* |MicroBSD??? I DON'T THINK SO!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!| |"I hate quotes in signature files" :-} MAtthew Fuller| *fullermd@narcissus.ml.org FreeBSD junkie* |http://keystone.westminster.edu/~fullermd Westminster College| *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*