Subject: Linux-Development Digest #533 From: Digestifier To: Linux-Development@senator-bedfellow.MIT.EDU Reply-To: Linux-Development@senator-bedfellow.MIT.EDU Date: Wed, 9 Mar 94 18:13:08 EST Linux-Development Digest #533, Volume #1 Wed, 9 Mar 94 18:13:08 EST Contents: Re: select (Finn Arne Gangstad) Install Slackware from (SCSI) Tape? (Hans Vermeulen) Re: GOD SPEAKS ON LINUX! (Glen Buhlmann) Re: AMD 486DX problem (with Linux?) (Romano Giannetti) Re: Amiga FileSystem, Anyone? Re: Amiga FileSystem, Anyone? Re: Amiga FileSystem, Anyone? (Matthias Urlichs) Re: UDP report card (Chris Anderson) Problems with memory (BERNARD Sebastien [93-94]) Re: rarpd for Linux ?? (Florian La Roche) pre-1 kernel crash (Filip M Gieszczykiewicz) ---------------------------------------------------------------------------- From: finnag@pvv.unit.no (Finn Arne Gangstad) Subject: Re: select Date: 9 Mar 1994 13:40:47 GMT In article , fgm@doc.ic.ac.uk (Frank McCabe) writes: |> I have come across an apparent problem with the select system call. |> |> According to the specification, if select is given a non-zero timeout |> then the system call is supposed to wait for the appropriate interval |> before terminating. |> |> Well it doesnt! If you give a non-zero timeout the nit comes back |> immediately. |> |> I know that this is not my problem, because the same (i.e. identical) |> program behaves as expected on a sun sparc under suno 4.1.13. |> |> Are there any known fixes for this? It's hard to comment on this as you don't include any source, but select seems to work as it 'should' on linux. This little program will print 'test' about every other second, it will also print the new values of the timeval struct you pass to select: #include #include #include main () { struct timeval t; while (1) { t.tv_sec = 2; t.tv_usec = 0; select (0, NULL, NULL, NULL, &t); printf ("test tv_sec:%d, tv_usec:%d\n", t.tv_sec, t.tv_usec); } } On a Linux system this program will typically produce this output: test tv_sec:0, tv_usec:0 test tv_sec:0, tv_usec:0 etc... While on SunOs you'll get test tv_sec:2, tv_usec:0 test tv_sec:2, tv_usec:0 etc... Therefore, to make portable code, you should set each time you call select, since select may or may not modify the passed timeval struct depending on what system you are on. ====================================== Finn Arne Gangstad, finnag@pvv.unit.no ------------------------------ From: vermeule@wi.leidenuniv.nl (Hans Vermeulen) Crossposted-To: comp.os.linux.help,comp.os.linux.admin Subject: Install Slackware from (SCSI) Tape? Date: 9 Mar 1994 13:28:01 GMT Reply-To: vermeule@wi.leidenuniv.nl Hello everybody, Is is possible to install Slackware from tape? I have a 150Mb SCSI tape drive connected to a Adapect 1542C. Running Slackware 1.1.2 I can read and write the tape (even read a tape written on a Sun running SunOs 4.1.3). Do I need a special boot floppy, or what? Thanks in advance, ---Hans. #------------------------------------------------------------------------------# Hans Vermeulen, Scientific Programmer Dept. of Computer Science, Leiden University, The Netherlands P.O. Box 9512, 2300 RA Leiden, The Netherlands Fax: +31 71 276985, Voice: +31 71 277106 e-mail: Hans.Vermeulen@wi.LeidenUniv.nl #------------------------------------------------------------------------------# ------------------------------ From: gbuhlman@uoguelph.ca (Glen Buhlmann) Crossposted-To: comp.os.linux,comp.os.linux.admin,comp.os.linux.help,comp.os.linux.misc Subject: Re: GOD SPEAKS ON LINUX! Date: 9 Mar 1994 14:20:43 GMT Grant Taylor (gtaylor@god.ext.tufts.edu) wrote: : I'll have you know I'm sitting right here in front of god, and god is : running Linux. : -grant : -- : Grant Taylor gtaylor@cs.tufts.edu : Read the linux Printing-HOWTO -- get it from sunsite or mail server: : To: listserv@god.ext.tufts.edu : with message body: -- I am God......and I use an Amiga...... +-------------------------------------------------------------------------+ | Glen Buhlmann | | | C.I.S. Department | | | University of Guelph | | | Guelph, ON | The opinions expressed in this | | CANADA | document are not my own....they are | | (519) 836-5612 | the result of extensive brainwashing | | (519) 821-5453 | by IBM, Microsoft, and others. | +-------------------------------------------------------------------------+ ------------------------------ From: romano@pimac2.iet.unipi.it (Romano Giannetti) Subject: Re: AMD 486DX problem (with Linux?) Date: 9 Mar 1994 14:54:42 GMT In comp.os.linux.development, Charles T Wilson -- Personal Account (ctwilson@rock.concert.net) wrote: > My apologies if it's bad netiquitte to post a file like this in > this newsgroup, but I was glancing at this thread, and thought > you might be interested in this program, which tells you a lot > about how your processor/compiler handles lots of things, floating > point representation among them. It's kinda old, but I think it's > still valid...enjoy. But _before_ comment out the following lines around line#450: while (size!=0) { while (malloc(size)!=(char *)NULL) total+=(size/2); size/=2; } that drive my Linux box to a quiet dead :-) after a lot of swapping. BTW: is this normal? I cannot afford test it on another Unix. My conf is Linux pre-1.0, 8M ram, 16M swap. The box don't crash nor panic, only get more and more slow if I don't ctrl-c the program. Bye, Romano -- ******************************************************************************* Romano Giannetti * DII-EIT, University of Pisa(E stands for Electronics) romano@iet.unipi.it * Dpto Electr. y Electronica, Facultad de Fisica * Universidad Complutense de Madrid ******************************************************************************* ------------------------------ From: cprakt2@acpx5.exp.univie.ac.at () Subject: Re: Amiga FileSystem, Anyone? Date: 9 Mar 1994 14:59:35 GMT Reply-To: cprakt2@acpx5.exp.univie.ac.at () In article <2lj2u9$hsd@ilink1.i-link.com>, leeh@i-link.com (Lee Heins) writes: > [stuff about mac floppies deleted] >The Amiga is also GCR, although they use variable bit rate >density zones (similar to that used on many IDE hard drives) instead of >variable spindle speed. Wrong. The Amiga hardware is capable of reading/writing GCR as well as MFM, but at a single defined bit rate. I never heard about a program using GCR, although it might be used in copy protection. The standard format for Amiga floppies is MFM with 11 sectors/track. The problem for PC hardware, as I see it, is the fact that there are no gaps between the sectors except after the last written. Still, I would assume that reading an Amiga disk should be possible. [ More stuff about Mac deleted.] >-- > > Lee Heins, leeh@i-link.com ------------------------------ From: cprakt2@acpx5.exp.univie.ac.at () Subject: Re: Amiga FileSystem, Anyone? Date: 9 Mar 1994 15:02:48 GMT Reply-To: cprakt2@acpx5.exp.univie.ac.at () In article <1994Mar3.174849.2359@swan.pyr>, iiitac@swan.pyr (Alan Cox) writes: > >The amiga floppy is a single sector MFM encoded 80 track double sided disk. >It's beyond the standard PC hardware to drive > >[And before Amiga people go no no its 11 sectors/track read the hardware >manual - its 11 _software_ sectors per physical sector] > >Alan I would guess from reading what is written on my disk package that PC disks have 9 _software_ sectors - it says "soft sectored" - so I don't see the difference here. Markus Milleder ------------------------------ From: urlichs@smurf.noris.de (Matthias Urlichs) Subject: Re: Amiga FileSystem, Anyone? Date: 9 Mar 1994 17:43:03 +0100 In comp.os.linux.development, article <2lj2u9$hsd@ilink1.i-link.com>, leeh@i-link.com (Lee Heins) writes: > > 1.4M floppies. For Mac 800k floppies, they use a variable spindle speed > (like CD-ROM drives) which most PC clone floppy drives aren't capable of. Actually, they replaced this with variable bit density zones. (It's compatible.) > Mac 800k disks are also GCR encoded instead of MFM, but that is something > that it is probably possible to do in software with PC-clone style floppy > mechanisms. I don't think so. (Not with variable bit densities.) I vaguely recall that somebody tried to fake variable speed by quickly turning the motor on and off, and succeeded ... in frying at least one vital part of their disk hardware. :-/ > Also untrue for the Mac for 1.4M floppies at least, the MFM hardware is > built into the SWIM (floppy controller) chip. I think what he meant was SIWM ("Super Integrated Woz Machine"). The IWM was the first integrated version of the WM, which was a very magical piece of circuitry, created (of course) by Steve Wozniak... totally incomprehensible, but it did its job with the absolutely fewest possible number of parts. ;-) -- The artist does not illustrate science [but] he frequently responds to the same interests that a scientist does. -- Lewis Mumford -- Matthias Urlichs \ XLink-POP N|rnberg | EMail: urlichs@smurf.noris.de Schleiermacherstra_e 12 \ Unix+Linux+Mac | Phone: ...please use email. 90491 N|rnberg (Germany) \ Consulting+Networking+Programming+etc'ing 42 Click here. ------------------------------ From: christop@access3.digex.net (Chris Anderson) Subject: Re: UDP report card Date: 9 Mar 1994 12:54:53 -0500 In article <1994Mar9.101239.4955@swan.pyr>, Alan Cox wrote: >In article <2lj8f2$gis@access1.digex.net> christop@access1.digex.net (Chris Anderson) writes: >> >>3. A recvfrom trashes the 8 bytes at the end of a sockaddr_in. This seems >> kinda sloppy. The code on line 484 of net/udp.c is where this happens. > >I'm not sure I know what you are seeing here. Care to elaborate. My code hashes on addresses. Because the addrlen returned from recvfrom includes the pad at the end of inet addrs, my hash function hashes on it. The code in inet/udp.c doesn't zero the pad, so I get stack garbage. udp.c, line 484: /* Copy the address. */ if (sin) { struct sockaddr_in addr; addr.sin_family = AF_INET; addr.sin_port = skb->h.uh->source; addr.sin_addr.s_addr = skb->daddr; memcpy_tofs(sin, &addr, sizeof(*sin)); } > >Alan >iiitac@pyr.swan.ac.uk >Networking Bod > Thanks for the cool linux work, Chris ------------------------------ From: bernard@zephyr (BERNARD Sebastien [93-94]) Subject: Problems with memory Date: 8 Mar 1994 14:09:06 GMT Hi folks, I'am trying to write a linux console driver for the Atari machine. I'm based upon the Amiga kernel sources. But I was wondering something. For printing on the screen, I have to allocate a screen buffer that should be coutiguous (I should'nt be split into the memory nor should it be swapped ) . Is the kmalloc doing such thing or how to do it ? ( the Amiga kernel is a port of the pl14 kernel). Second the real adresses seems to be different from the virtual adresses. How to compute them ? Thanks a lot for showing me the way... S. Bernard ------------------------------ From: rzsfl@sbusol.rz.uni-sb.de (Florian La Roche) Subject: Re: rarpd for Linux ?? Date: 9 Mar 1994 00:51:21 GMT Cheryl Dematteis (cdematt@ucrengr.ucr.edu) wrote: : -- : Has anyone out there ported rarpd to Linux. We have a couple of old Sun3's : that we need to boot off of a Linux machine. Any pointers or suggestions : would be greatly appreciated. Someone has done rarp in the kernel for pl12, I think. I have mailed him a copy of the BSD rarpd, and he tries to port it to linux. Maybe he is soon ready with a first version... Florian La Roche : Please respond via email, I will post a summary of responses : Thanks : ------------------------------------------------------------------------------ : Cheryl K. DeMatteis cdematt@cs.ucr.edu : Graduate Student (909)787-3824 : Department of Computer Science : University of California : Riverside, CA 92521 : ------------------------------------------------------------------------------ ------------------------------ From: filip@alpha.smi.med.pitt.edu (Filip M Gieszczykiewicz) Subject: pre-1 kernel crash Date: 9 Mar 94 13:39:22 GMT Greetings. I just had a weird but reproducible kernel krash. I reproduce it by running /usr/lib/makewhatis of all things. Nothing else does it! The first time it happened (a day ago), I just got a "Segment fault" after makewhatis executed for 20-40 seconds. After the first message, re-running the command would immedietly produce the error. I would have to reboot to get the 20-40 second "run time" before the error error. I though, "I'm going to upgrade to Slack 1.1.2, screw makewhatis" :-) Today, I made the mistake of trying it again this morning: I got 2+ pages of "Unable to handle kernel paging request at address 24440336".... So, I did the next logical thing.... I switched to the next VC (I have 14), hit ENTER, my prompt scrolled up again, I though, "Oh, goody", enter "root" and it just hung. So I switch to a VC in which I was logged in: It gets weird here... naturally, the internal shell commands work but any command fails with the "unable to handle kernel paging" and a "Segment fault" to boot at the end of the stack dump. Yes, I did copy a few complete entries. I then switched to the VC that got the 2+ pages of dumps and noticed a pattern: all the running jobs (normally 20 or so) got dumped. I was a bit shocked to see "init" spill its guts :-) While I was looking over the above and twiddling my thumbs in anticipation of the [of course] unbacked up FS(es), after ~20 seconds, crond spilled its guts with a stack dump.... Now, it gets even weirder, when I rebooted (had to hit reset!), I checked the /usr/adm/ system logs (all of them) NOTHING. Actually, after trying it a few times, I got this from syslog: =================chop=with=axe================chop=with=axe===================== Mar 9 07:25:05 alpha ecx: 0000000a edx: 00000400 Mar 9 07:25:05 alpha ernel: <6>Code: 8b 94 f5 2c 03 44 24 14 8d 44 02 ff d3\ f8 89 44 24 24 8d 6c (the \ is mine) =================chop=with=axe================chop=with=axe===================== There's more. I changed "update" to "update 10" just to feel safer :-), I prepare the VC's for easy switching, got top running in 2 VCs, etc. and re-ran /usr/lib/makewhatis. Sure enough, 2 minutes later the VC I was watching filled up with "unable to handle".... but, I was kinda hoping this time... What changed was that when I switched to a non-logged in VC and pressed ENTER, I would get: =================chop=with=axe================chop=with=axe===================== Invalid operator: 0000 EIP: 0010:0000000f EFLAGS: 00010202 (EAX thru SS follow) db1c20, c473c0, 246, db1cb8, db1cb8, db1d7c, db1cfc, e2d00c, 18, 18, 18, 2b, 18 Pid: 1512.... (getty) Code: f0 97 ea 00 f0 54 ff 00 f0 43 eb 00 f0 eb ea 00 f0 a5 fe 00 Stack: 00000008 00000000 (* 4) =================chop=with=axe================chop=with=axe===================== In case it helps, I switched to 4 VC's in a row and hit ENTER in each (getting the guts of getty in each :-) The only flags in the dump that changed in each were the esp & Stack: field #1. BTW, makewhatis DID exit this time with "Error -117" right after the dumps in the other VCs... Some misc dumps: (First time around) =================chop=with=axe================chop=with=axe===================== PID: 17 Process nr 4 (syslogd) Code: ea 00 f0 c3 e2 00 f0 97 ea 1c f0 97 ea 00 f0 54 ff 00 f0 43 Stack: 00000000 (* 5) Trying to free kernel page-directory: Not good Invalid operand: 0000 EIP: 0010:0000000f EFLAGS: 00010202 I have the eax-ss flags too .... (gawk) Code: general protection: 000c =================chop=with=axe================chop=with=axe===================== (Second time around) =================chop=with=axe================chop=with=axe===================== Unable to handle kernel paging request at address 24440336 Oops: 0000 EIP: 0010:0013d77d EFLAGS: 00010212 =================chop=with=axe================chop=with=axe===================== What should I do? I keep 5 previos kernels so down-grading is not a problem. Anyone? BTW, I don't mind. This was kinda fun in its own little way :-) Take care. P.S. I've had Linux sine 96pl1 and this is the FIRST crash... That's a pretty good record... SETUP: 486 DX2/66 VLB 16MB 100MB IDE, pre-1 kernel, gcc 2.4.5 & 4.4.4 libs (so shoot me!), xiafs (no bad blocks), iozone 690K write 270K read (sucks), kernel bare except for SCSI support (no net) base is slack 1.1.1 Copy of my /usr/src/linux/.config file: =================chop=with=axe================chop=with=axe===================== begin 644 config.gz M'XL(".Z*?2T``V-O;F9I9P!]5<%NHS`0O>X`/.3I+LBC=1#'1[`1 MB9!!+D*P"[8"K-/D+=K\!F&:?,N!"*-\=%?L*=;R)(Q!)O+#_@QI-C'PTK<@ M7_G(4+JW$JXR;3XZLM!?MP+\#V,LNT]R(.] MN`>:+M/=CY9/74:F*#=KE5M;T3/&;_2,484)59@LU<>,T\^?L?@0,0AE]"%D MMXBEQ(JKV7>C9M%N9&E58/+[AYT`VF0_3#,$EY&A8P"?3YV'6-3F%C-MT6/(T'H\-R!COE9^(_#.5 M6Q"*CV@MVO?MXH#X"E$-4IWU=M?MU=V'"52_.DW]3-G"N6:*942QR(Y9+G99 MZZI:F&#/-UCK#59(*TF/EPI5J(*:7G$5XV'8L3U_7\(RW6;=L!52*QQJ:QU! M!32QB<&&TL+,_C6?CX?HE3.\M9Q//:Y8Z_=`!NMI[Z%S@[/8[;W%Q/=,I:0^#S*1=OI;:)YO#34V'#K(_XNU\509Q2CY M_GA)#TG8G2:V2_HW[T7(I!@_[\%Z&R6;3MUL`YNHT^,2Z$R?XC?CD'!V-7\N +^.@+58&1]E0'```W ` end =================chop=with=axe================chop=with=axe===================== P.P.S. If you read this far... was this an "OK" post of a kernel panic? what flags are mandatory to be posted? How much info to include? This is fun... it just "might" happend again :-) -- +-->Filip "I'll buy a vowel" Gieszczykiewicz | E-mail: filip@alpha.med.pitt.edu | Send e-mail with Subject: "get FAQ index" for index to sci.electronics oldFAQ | Checkout: ftp bode.ee.ualberta.ca and get file /pub/cookbook/contents. Neat. | Making money with CS and spending it on EE, robotics, windsurfing, & dreams. ------------------------------ ** FOR YOUR REFERENCE ** The service address, to which questions about the list itself and requests to be added to or deleted from it should be directed, is: Internet: Linux-Development-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux.development) via: Internet: Linux-Development@NEWS-DIGESTS.MIT.EDU Linux may be obtained via one of these FTP sites: nic.funet.fi pub/OS/Linux tsx-11.mit.edu pub/linux sunsite.unc.edu pub/Linux End of Linux-Development Digest ******************************