Re: linux-kernel-digest V1 #2366

dave fournier (otter@otter-rsch.com)
Mon, 10 Aug 1998 16:49:06 +0000

Messages sorted by: [ date ][ thread ][ subject ][ author ]
Next message: Alan Cox: "Re: IDE-Driver Disk Accounting Missing, Maybe SCSI too......"
Previous message: Daniel Thom: "Re: 2.1.115 Panic Shortly After Boot (Interrupt related???)"

owner-linux-kernel-digest@vger.rutgers.edu wrote:
>
> linux-kernel-digest Monday, 10 August 1998 Volume 01 : Number 2366
>
> In this issue:
>
> Re: Strange interrupt behaviour
> Re: IDE-Driver Disk Accounting Missing, Maybe SCSI too......
> TCP variable plotting utilities
> Re: [OFFTOPIC] Stupid browser flamewar
> Re: DEVFSv50 and /dev/fb? (or /dev/fb/? ???)
> Re: core dump and threads
> Re: PPP --- TCP bad checksums
> RE: PPP --- TCP bad checksums
> in 2.1.115 umount freezes
> Re: DEVFSv50 and /dev/fb? (or /dev/fb/? ???)
> Device xx:xx not found by BIOS
> "ifconfig inet 0" vs. DHCP
> Re: problem with vgacon on Alphas
> Mounting with option "nodiratime" and "noatime"
> Re: Intel NLX motherboards
> Re: IDE-Driver Disk Accounting Missing, Maybe SCSI too......
> Re: 2.1.115 scsi funnies
> Re: pre-2.1.116-1 breaks Cyrix 6x86 support
> Re: 2.1.115 Extended VGA
> Re: std VGA text console funnies
> Re: PPP --- TCP bad checksums
> Re: Bug: PCI discovered primary peer bus
> Re: [OFFTOPIC]linux + WIN98
> Re: Why a find causes a flush?
> Re: 2.1.115 Extended VGA
> 2.1.112: Another page fault from irq handler
> Re: after-boot video mode changes...
> Can't build new linux kernel 2.0.35
> Is the Turtle Beach Montego PCI Sound Card recognized by Linux?
> Re: IDE-Driver Disk Accounting Missing, Maybe SCSI too......
> Re: IDE-Driver Disk Accounting Missing, Maybe SCSI too......
> lkml FAQ updates highlights; searching for new home...
> Re: vesafb-HOWTO v0.6 released
> Re: [OFFTOPIC] Stupid browser flamewar
> Re: [OFFTOPIC]linux + WIN98
> Re: [OFFTOPIC] Stupid browser flamewar
>
> See the end of the digest for information on subscribing to the linux-kernel
> or linux-kernel-digest mailing lists.
>
> ----------------------------------------------------------------------
>
> From: Andrea Arcangeli <arcangeli@mbox.queen.it>
> Date: Mon, 10 Aug 1998 03:17:11 +0200 (CEST)
> Subject: Re: Strange interrupt behaviour
>
> On Fri, 17 Jul 1998, MOLNAR Ingo wrote:
>
> >
> >or, (beware, it's going to be ugly), read-only map the task stack's
> >lower-4k as lower-4k of the IRQ handler. NOTE, there will be _no_ per-IRQ
> >MMU cost with this trick, and it's nice with SMP and threads too.
>
> I have implemented the per-irq thing with the help of mmu to handle
> current read only and rw(current) read/write. Booted and worked for some
> minutes then the machine locked with high net traffic. It' s too late to
> debug that lock now though ;-). Here the pointer if you want to take a
> look and give me hints.
>
> ftp://e-mind.com/pub/linux/kernel-patches/irq_stack-1-2.0.115.diff.gz
>
> >_much_ less write-access to 'current->'. (the only write-access from IRQ
> >contexts i know of is in update_process_times(), a per-timer-IRQ thing).
>
> I have just triggered the need_resched = 1 (look at the patch).
>
> Andrea[s] Arcangeli
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: Mark Lord <mlord@pobox.com>
> Date: Mon, 10 Aug 1998 09:52:48 -0400
> Subject: Re: IDE-Driver Disk Accounting Missing, Maybe SCSI too......
>
> Andre M. Hedrick wrote:
>
> > I am willing to make the assumption and statement that the lack of
> > disk_accounting based on HWGROUP add-request to disable interrupts,
> > could be the primary cause of fs-corruption on IDE drives.
>
> Exactly *what* "corruption" are you referring to?
> I have not seen any recent non-DMA-related reports of such,
> but then, it seems I'm usually the last person to be told of such.
> - --
> mlord@pobox.com
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: Ajit Anvekar <ajit@ece.iisc.ernet.in>
> Date: Mon, 10 Aug 1998 16:53:31 +0530 (GMT+05:30)
> Subject: TCP variable plotting utilities
>
> Hello,
> I am working on a project which involves graphically plotting the
> various TCP parameters like advertised window, congestion window, rtt
> values,throput calculation etc. Would anybody have information about the
> availibility of such a package/tool on Linux ?
> Utilities like iptraf only show output of the packets that are going on
> the network with out any of the internal variables. I suppose this would
> mean hacking the kernel to view these variables. Could this be done via
> the proc/net interface or is there a better way of doing this ?? I also
> want the variables to be plotted in real time. Which packages would be
> ideal ??
> Could somebody please shed light on this ??
> Thanks in advance for all the help and thank you for your time.
>
> Ajit
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: "Mike A. Harris" <mharris@ican.net>
> Date: Mon, 10 Aug 1998 09:58:05 -0400 (EDT)
> Subject: Re: [OFFTOPIC] Stupid browser flamewar
>
> On 10 Aug 1998, Jes Sorensen wrote:
>
> > Alex> Thank you, whilst I do appreciate constructive criticism, I will
> > Alex> NOT tolerate people levelling insults at me, over this. I'm not
> > Alex> going to change my mind until people starts being nicer and
> > Alex> putting forward constructive criticism instead of calling me an
> > Alex> "asshole", with an "one brain cell" to quote two people in this
> > Alex> mailing list. That definitely doesn't achieve anything except
> > Alex> piss me off big time, and especially when I've put a lot of work
> > Alex> into this HOWTO-framebuffer (which _does_ work on a text
> > Alex> browser) and this is what I get, a slap in my face.
> >
> > Whats insulting is that message your pages prints when people try to
> > access the pages with a non-frames browser.
>
> Although disappointing, I wouldn't say "insulting". It merely
> reports that non-frames browsers aren't supported, however I used
> LYNX to view the pages, and LYNX seemed to display the pages just
> fine IMHO...
>
> > I'd say you got two options: 1) make your pages stop printing that
> > crap or 2) stop doing the HOWTO (a HOWTO on a frames-only page is
> > useless).
>
> A non-frames page would be nice, but I think that he is free to
> do what he wants. Some people write documentation in a certain
> format, say as an ASCII text file, or as HTML, and refuse to do
> it in say SGML. That is their preference. It might be more
> useful to everyone to have it done in SGML, as it can be easily
> converted to many formats then, however if someone wants to
> writea helpful document, but doesn't want to learn SGML, or
> whatever, they just want to write it in a simple format they are
> familiar with, does that mean that they shouldn't bother?
>
> I think we should be more thankful to people that do things like
> this for free. Nobody is stoping anyone from copying such a
> HOWTO and putting it on their own NON-FRAMES pages.
>
> Don't get me wrong, I don't care for frames pages much either,
> however it is certainly not fair to complain about getting
> something free. If I did a frames only page, and got treated
> like this, I would do the same thing.
>
>
> > Just a side note, most of the frame buffer code comes from the m68k
> > port, however many m68k people do no run X because their boxes are too
> > slow and besides that there is no official Netscape (nor M$ Internet
> > Exploiter - fortunately) port for it.
>
> That is unfortunate. Time for an upgrade eh? I don't know a
> single person that has a computer that isn't capable of running a
> frames enabled web browser.
>
> > The only thing I know of that will be able to view your pages
> > is Mozilla with Lesstif .....
>
> Well then there shouldn't be any problem.
>
> > Jes
> >
> > PS: This of course doesn't have anything to do with the fact that
> > frames are a 43$@#$#@ pain in the a**.
>
> Well *that* we can CERTAINLY agree with! ;o) There are some
> useful web sites that use frames in a nice way though.
>
> - --
> Mike A. Harris - Computer Consultant - Linux advocate
>
> Escape from the confines of Microsoft's operating systems and push your
> PC to it's limits with LINUX - a real OS. http://www.redhat.com
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: "Michael L. Galbraith" <mikeg@weiden.de>
> Date: Mon, 10 Aug 1998 09:35:35 +0200 (CEST)
> Subject: Re: DEVFSv50 and /dev/fb? (or /dev/fb/? ???)
>
> On Sun, 9 Aug 1998, Shawn Leas wrote:
>
> > On Sat, 8 Aug 1998, Michael L. Galbraith wrote:
> >
> > > Damn poor excuse for calling a man who has devoted as much time and
> > > effort to Linux as Ted has an `idiot' IMNSHO. If you sent that gem
> > > from a Linux box, you used some of his code to slap him in the face.
> > >
> > > I just can't help myself.
> >
> > I can't help it if people refuse to embrace superior abstractions in favor
> > of old hacks, especially when not backing up claims made.
> >
>
> You say superior, everyone says superior.. or else.
>
> > This detracts from ANYONE's credability, especially if you are noteworthy.
> > Simply put, he should friggin know better than to say things without
> > backing them.
> >
> > -Shawn
>
> You conveniently forget the source code that he posted.
>
> These are my last words on the subject. I have expressed my opinion
> regarding your rude remark.. said opinion is not likely to change.
>
> EOT,
>
> -Mike
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: Philip Gladstone <philip@raptor.com>
> Date: Mon, 10 Aug 1998 10:33:59 -0400
> Subject: Re: core dump and threads
>
> I have a module that will give you multithreaded core dumps and
> a small patch to gdb-4.16 that allows it to read them. My module
> is just a patch against the 2.0.3x binfmt_elf.c. If you are interested
> then I'll try and package it up for you.
>
> Philip
>
> Andreas Jellinghaus wrote:
> >
> > can someone give me the status of coredumps for multithreaded programs ?
> > running and debuggin (useing the patches for gdb and glibc) works fine.
> >
> > the last information on core dumps i read on this list was :
> > the other threads continue to run. ps says, they don't.
> >
> > what can i do ?
> >
> > andreas
> >
> > -
> > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> > the body of a message to majordomo@vger.rutgers.edu
> > Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> - --
> Philip Gladstone +1 781 530 2461
> Raptor Systems / Axent Technologies
> Waltham, MA http://www.raptor.com/
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: Samuli Kaski <samkaski@cs.helsinki.fi>
> Date: Mon, 10 Aug 1998 17:39:38 +0300 (EEST)
> Subject: Re: PPP --- TCP bad checksums
>
> On Mon, 10 Aug 1998, Alan Cox wrote:
>
> > > There have been many reports that use if VJ compression triggers errors
> > > with some ISP equipment. You might try turning that option off an seeing
> > > what happens.
> >
> > Be that as it may (and its definitely the case) it doesn't cover the
> > "2.0.35 works perfectly" cases. All of these appear to be PPP - is anyone
> > seeing them wit SLIP (does anyone run SLIP any more)
>
> I am but I'm not seeing any with vj turned off. With vj on, I see them.
> This is with 2.1.105 but I think I saw them with 2.0.34 also, both SLIP
> and PPP. I'm running a leased line though...
>
> All it takes to trigger them is to fire up a network quake play (UDP)
> and watch most of open TCP connections (ssh, imap, irc, etc) cause them.
> I see some Socket destroy delayed's also, but they happen even when vj
> is off.
>
> - --
> Samuli Kaski, samkaski@cs.helsinki.fi
> Department of Computer Science, University of Helsinki, Finland.
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: "Manuel J. Galan" <manolow@step.es>
> Date: Mon, 10 Aug 1998 15:46:32 +0100 (WEST)
> Subject: RE: PPP --- TCP bad checksums
>
> For those having problems with ppp-2.3.5-1.i386.rpm I've put together
> one version (taken from redhat devel) that seems to fix the problem.
> It is under
>
> ftp://ftp.ulpgc.es/pub/ulpgc/ppp-2.3.5/standard/
>
> Another version under
>
> ftp://ftp.ulpgc.es/pub/ulpgc/ppp-2.3.5/nt-ras-callback/
>
> Is a hacked one that 'seems' to work with NT-RAS with callback.
>
> - ----------------------------------
> E-Mail: Manuel J. Galan <manolow@step.es>
> Date: 10-Aug-98
> Time: 15:40:24
>
> This message was sent by XFMail
> - ----------------------------------
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: Database <postgres@lstc.com>
> Date: Mon, 10 Aug 1998 07:47:17 -0700 (PDT)
> Subject: in 2.1.115 umount freezes
>
> Hi,
>
> I'm using Linux kernel 2.1.115, with mount 2.8a. When I attempt
> to unmount a filesystem using the "umount" command, or when
> a shutdown script does this the umount command locks up, and
> never returns. Umount cannot even be killed with kill -9.
> The system is a Dell Dimension XPS H266, running RedHat 5.1, I
> have upgraded all the software packages according to
> Documentation/Change. The kernel was compiled with gcc 2.7.2.3,
> it is a nonsmp kernel. All the selected drivers
> are non-experimental with the exception of /dev/fb. All of
> the filesystems, with the exception of one are ext2. The
> exception is my win95 disk which is a small vfat partition.
> I have four hard drives on my system, two of which are on
> an IDE bus, the other two are on a "Adaptec AIC-7860 (rev 1)"
> SCSI controller board. Umount did work fine under 2.0.34.
>
> PS
>
> I hope this isn't a stupid problem caused by user error
> *cringe*
>
> Thanks in advance,
> Nathan A. Hallquist,
> nathan@lstc.com
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: Raul Miller <rdm@test.legislate.com>
> Date: Mon, 10 Aug 1998 11:15:26 -0400
> Subject: Re: DEVFSv50 and /dev/fb? (or /dev/fb/? ???)
>
> Richard Gooch <Richard.Gooch@atnf.CSIRO.AU> wrote:
> > > > Note that if you have a configuration which uses devfs solely for
> > > > its role as an information service provider for devfsd, (i.e. you
> > > > keep using a disc-based /dev) then one of the things you loose is the
> > > > ability to lookup non-existent device nodes in /dev and have modules
> > > > automatically loaded by kmod.
> ...
> > I'm afraid I don't quite see what you're getting at here. Could you
> > please rephrase it somehow?
>
> My point is that once you've taken into account requirements for browsing
> (is the hardware installed correctly?) and security (who's allowed to
> bring up a piece of hardware?), you're way outside the scope of what
> you can do with the ability to lookup non-existent device nodes in /dev
> causing modules to be automatically loaded by kmod.
>
> - --
> Raul
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: David Bryant <d_bryant@lincoln.college.adelaide.edu.au>
> Date: Tue, 11 Aug 1998 00:59:48 +0930
> Subject: Device xx:xx not found by BIOS
>
> Hi,
> During boot up I get the following messages. These messages only
> appear in the 2.1.x series, they don't
> come up in the 2.0.x series:
>
> messages:Aug 8 00:47:55 whiskey kernel: PCI: PCI BIOS revision 2.00
> entry at 0xfc9ca
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Using configuration type 2
>
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Probing PCI hardware
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Device 00:80 not found by
> BIOS
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Device 00:88 not found by
> BIOS
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Device 00:90 not found by
> BIOS
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Device 00:98 not found by
> BIOS
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Device 00:a0 not found by
> BIOS
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Device 00:a8 not found by
> BIOS
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Device 00:b0 not found by
> BIOS
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Device 00:b8 not found by
> BIOS
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Device 00:c0 not found by
> BIOS
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Device 00:c8 not found by
> BIOS
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Device 00:d0 not found by
> BIOS
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Device 00:d8 not found by
> BIOS
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Device 00:e0 not found by
> BIOS
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Device 00:e8 not found by
> BIOS
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Device 00:f0 not found by
> BIOS
> messages:Aug 8 00:47:55 whiskey kernel: PCI: Device 00:f8 not found by
> BIOS
> messages:Aug 8 00:47:55 whiskey kernel: PCI: 00:00 [8086/04a3]: Bridge
> optimization (00)
>
> I (think I) have tried changing all the kernel config options,
> including:
>
> PCI BIOS support
> PCI direct access support
> PCI quirks
> PCI bridge optimization
>
> but the messages still appear.
> Anyone know what the problem is ?? or is my BIOS (PHOENIX) just a piece
> of crap ?
>
> The output of cat /proc/pci is:
>
> PCI devices found:
> Bus 0, device 0, function 0:
> Host bridge: Intel 82434LX Mercury/Neptune (rev 17).
> Slow devsel. Master Capable. Latency=96.
> Bus 0, device 2, function 0:
> Non-VGA device: Intel 82378IB (rev 3).
> Medium devsel. Master Capable. No bursts.
> Bus 0, device 3, function 0:
> IDE interface: CMD 640 (buggy) (rev 1).
> Medium devsel. IRQ e.
> I/O at 0x1f0 [0x1f1].
> I/O at 0x3f4 [0x3f5].
> I/O at 0xff00 [0xff01].
> I/O at 0xff08 [0xff09].
> Bus 0, device 14, function 0:
> Ethernet controller: Realtek 8029 (rev 0).
> Medium devsel. IRQ 9.
> I/O at 0xfce0 [0xfce1].
> Bus 0, device 15, function 0:
> Display controller: Texas Instruments TVP4020 Permedia 2 (rev 1).
> Medium devsel. Fast back-to-back capable. IRQ 9. Master
> Capable. Laten
> cy=96. Min Gnt=192.Max Lat=192.
> Non-prefetchable 32 bit memory at 0x40000000 [0x40000000].
> Non-prefetchable 32 bit memory at 0x40800000 [0x40800000].
> Non-prefetchable 32 bit memory at 0x41000000 [0x41000000].
> Bus 0, device 16, function 0:
> Display controller: Texas Instruments TVP4020 Permedia 2 (rev 1).
> Medium devsel. Fast back-to-back capable. IRQ 9. Master
> Capable. Laten
> cy=96. Min Gnt=192.Max Lat=192.
> Bus 0, device 17, function 0:
> Display controller: Texas Instruments TVP4020 Permedia 2 (rev 1).
> Medium devsel. Fast back-to-back capable. IRQ 9. Master
> Capable. Laten
> cy=96. Min Gnt=192.Max Lat=192.
> Bus 0, device 18, function 0:
> Display controller: Texas Instruments TVP4020 Permedia 2 (rev 1).
> Medium devsel. Fast back-to-back capable. IRQ 9. Master
> Capable. Laten
> cy=96. Min Gnt=192.Max Lat=192.
> Bus 0, device 19, function 0:
> Display controller: Texas Instruments TVP4020 Permedia 2 (rev 1).
> Medium devsel. Fast back-to-back capable. IRQ 9. Master
> Capable. Laten
> cy=96. Min Gnt=192.Max Lat=192.
> Bus 0, device 20, function 0:
> Display controller: Texas Instruments TVP4020 Permedia 2 (rev 1).
> Medium devsel. Fast back-to-back capable. IRQ 9. Master
> Capable. Laten
> cy=96. Min Gnt=192.Max Lat=192.
> Bus 0, device 21, function 0:
> Display controller: Texas Instruments TVP4020 Permedia 2 (rev 1).
> Medium devsel. Fast back-to-back capable. IRQ 9. Master
> Capable. Laten
> cy=96. Min Gnt=192.Max Lat=192.
> Bus 0, device 22, function 0:
> Display controller: Texas Instruments TVP4020 Permedia 2 (rev 1).
> Medium devsel. Fast back-to-back capable. IRQ 9. Master
> Capable. Laten
> cy=96. Min Gnt=192.Max Lat=192.
> Bus 0, device 23, function 0:
> Display controller: Texas Instruments TVP4020 Permedia 2 (rev 1).
> Medium devsel. Fast back-to-back capable. IRQ 9. Master
> Capable. Laten
> cy=96. Min Gnt=192.Max Lat=192.
> Bus 0, device 24, function 0:
> Display controller: Texas Instruments TVP4020 Permedia 2 (rev 1).
> Medium devsel. Fast back-to-back capable. IRQ 9. Master
> Capable. Laten
> cy=96. Min Gnt=192.Max Lat=192.
> Bus 0, device 25, function 0:
> Display controller: Texas Instruments TVP4020 Permedia 2 (rev 1).
> Medium devsel. Fast back-to-back capable. IRQ 9. Master
> Capable. Laten
> cy=96. Min Gnt=192.Max Lat=192.
> Bus 0, device 26, function 0:
> Display controller: Texas Instruments TVP4020 Permedia 2 (rev 1).
> Medium devsel. Fast back-to-back capable. IRQ 9. Master
> Capable. Laten
> cy=96. Min Gnt=192.Max Lat=192.
> Bus 0, device 27, function 0:
> Display controller: Texas Instruments TVP4020 Permedia 2 (rev 1).
> Medium devsel. Fast back-to-back capable. IRQ 9. Master
> Capable. Laten
> cy=96. Min Gnt=192.Max Lat=192.
> Bus 0, device 28, function 0:
> Display controller: Texas Instruments TVP4020 Permedia 2 (rev 1).
> Medium devsel. Fast back-to-back capable. IRQ 9. Master
> Capable. Laten
> cy=96. Min Gnt=192.Max Lat=192.
>
> warning: page-size limit reached!
>
> Thanks.
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: Chip Salzenberg <chip@perl.org>
> Date: Mon, 10 Aug 1998 11:33:55 -0400
> Subject: "ifconfig inet 0" vs. DHCP
>
> A difference between 2.0.x and 2.1.x is that setting a network
> interface IP address to 0.0.0.0 actually has the effect of setting it
> to _no_ address, i.e. disabling IP.
>
> Now that seems reasonable on the face of it, but I spent a few hours
> last night working around it -- because even though you can send DHCP
> requests over such an interface, absolutely _no_ incoming packets are
> processed, even if the packets are hardware broadcasts and there's an
> existing UDP socket ready and waiting.
>
> Now, you may say that you're supposed to set a _real_ IP address for
> an interface before you use it, and that's usually true -- but there's
> no way to do that here, because the whole point of DHCP is to *get*
> the IP address, because you don't know what it is yet.
>
> So, is "0.0.0.0" meaning "no IP" a bug or a feature?
> - --
> Chip Salzenberg - a.k.a. - <chip@perlsupport.com>
> "Take it to the bridge, Sinbad! ... drop it in the water ..." //MST3K
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: Martin Mares <mj@ucw.cz>
> Date: Mon, 10 Aug 1998 15:55:59 +0200
> Subject: Re: problem with vgacon on Alphas
>
> > after some reports about freezing Alpha machines with 2.1.113 as soon
> > as someone switches from X to a normal console. I've tried it myself and
> > my machine also stopped after I pressed CTRL-ALT-F1 under X.
> >
> > I've tracked the problem down (serial console is your friend) to
> > a memcpy_from(), where the source is a kernel address, which is
> > a no no on Alphas. Below is a patch, which fixes this problem.
> >
> > I hope the patch will make into one of the next kernels.
>
> OK, I've modified it a bit and added to my other console patches.
>
> Have a nice fortnight
> - --
> Martin `MJ' Mares <mj@ucw.cz> http://atrey.karlin.mff.cuni.cz/~mj/
> Faculty of Math and Physics, Charles University, Prague, Czech Rep., Earth
> "A student who changes the course of history is probably taking an exam."
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: "Mike A. Harris" <mharris@ican.net>
> Date: Mon, 10 Aug 1998 11:11:43 -0400 (EDT)
> Subject: Mounting with option "nodiratime" and "noatime"
>
> I have just finished upgrading to mount-2.8-4 and am using the
> mount options "noatime" and "nodiratime" on the following
> partitions:
>
> /dev/hda5 / ext2 defaults 1 1
> /dev/hda7 /home ext2 noatime,nodiratime 1 2
> /dev/hda2 /stuff ext2 noatime,nodiratime 1 2
> /dev/hda9 /usr ext2 noatime,nodiratime 1 2
> /dev/hda8 /usr/local ext2 noatime,nodiratime 1 2
>
> What possible things could break with this configuration? What
> actually USES the ATIME date stamp on files and directories? I
> don't use any backup software or anything like that. I'm just
> curious as to wether I can expect weird behaviour from any
> programs now or not?
>
> Is it safe to do noatime and nodiratime on "/" as well?
>
> Assuming nothing major breaks, are there any other wierdnesses I
> should be aware of?
>
> - --
> Mike A. Harris - Computer Consultant - Linux advocate
>
> Escape from the confines of Microsoft's operating systems and push your
> PC to it's limits with LINUX - a real OS. http://www.redhat.com
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: Joel Jaeggli <joelja@darkwing.uoregon.edu>
> Date: Mon, 10 Aug 1998 08:26:59 -0700 (PDT)
> Subject: Re: Intel NLX motherboards
>
> NLX describes the form factor/motherboard layout not the chipset. The
> databooks is a standard lx p-II chipset. Incedentaly If yooou really want
> a blast from the past the motherboard is connected to the riser card that
> holds the isa/pci slots with what looks like an s-100 bus connector...
>
> lx boards have been extensivly tested against 2.0.x and 2.1.x. Which fica
> board is this product built around. If I were you I'd would be looking at
> what the ide controller is and what the ide drive is as well. you may
> need to compile the kernel with the appropiate ide bugfix...
>
> joelja
>
> On 11 Aug 1998, Scott Hassan wrote:
>
> >
> > I have been having trouble with Linux on the new Intel NLX
> > chipset. More specifically on the Sahara II Databooks made
> > by FICA. Here is a URL for a specification of a Databook:
> >
> > http://www.fica.com/english/product/system/databook/sahara.htm
> >
> > The cool thing about these machines is the formfacter. They can
> > handle a Pentium 300 II, 256M of memory, two disks, and are
> > *only* 3.75" high. I figure I can put 12 on a standard 23" rack
> > shelf. It seems that I should be able to fit 72 of these in the
> > rack. Way, cool!
> >
> > RedHat 5.1 installs okay on the machine (PII300/256M/6.4G) but
> > after a day, it crashes with "bdflush, sleeping..." messages.
> > I am running 2.0.35.
> >
> > I saw this same problem with the Intel BX chipset and I found
> > that 2.0.35 fixed this problem.
> >
> > Has anyone tested Linux out with the Intel NLX chipset?
> > Does the 2.1.x kernels work with the NLX?
> >
> > Thanks,
> >
> > Scott Hassan
> > FindMail Communications
> >
> >
> >
> >
> > -
> > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> > the body of a message to majordomo@vger.rutgers.edu
> > Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
> >
>
> - --------------------------------------------------------------------------
> Joel Jaeggli joelja@darkwing.uoregon.edu
> Academic User Services consult@gladstone.uoregon.edu
> PGP Key Fingerprint: 1DE9 8FCA 51FB 4195 B42A 9C32 A30D 121E
> - --------------------------------------------------------------------------
> It is clear that the arm of criticism cannot replace the criticism of
> arms. Karl Marx -- Introduction to the critique of Hegel's Philosophy of
> the right, 1843.
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: "Andre M. Hedrick" <hedrick@Astro.Dyer.Vanderbilt.Edu>
> Date: Mon, 10 Aug 1998 10:43:04 -0500 (CDT)
> Subject: Re: IDE-Driver Disk Accounting Missing, Maybe SCSI too......
>
> On Mon, 10 Aug 1998, Mark Lord wrote:
>
> > Andre M. Hedrick wrote:
> >
> > > I am willing to make the assumption and statement that the lack of
> > > disk_accounting based on HWGROUP add-request to disable interrupts,
> > > could be the primary cause of fs-corruption on IDE drives.
> >
> > Exactly *what* "corruption" are you referring to?
> > I have not seen any recent non-DMA-related reports of such,
> > but then, it seems I'm usually the last person to be told of such.
>
> See "MESSAGE #1 :: Section #1"
>
> Mark, I know of no cases of "non-DMA-related reports", either.
> I am hunting for something to allow someone to save face, with
> an apology to you, but no real luck. I have yet to find a flaw.
>
> However IMHO, Linus must still believe that the IDE-DMA has problems or
> he would have removed the following.
>
> if [ "$CONFIG_BLK_DEV_IDEPCI" = "y" ]; then
> # Either the DMA code is buggy or the DMA hardware is unreliable. FIXME.
> bool ' Generic PCI bus-master DMA support' CONFIG_BLK_DEV_IDEDMA
>
> You know that I still back you, and the work that you have done.
> I am hunting through the source trying to find anything to question and
> confront Linus directly about his lack of forsight in this matter.
>
> OH DAMN, I just spouted HERSEY....So what!!!
>
> SINCE I AM THE ONE WHO SAID IT, DON'T BOTHER "Mark Lord"!!!!!
>
> I will stand alone, if needed to help change the direction of current
> kernel thinking.............Good Maintainers are hard to find, and I
> personally would like to see your return.
>
> I gather from previous "public" messages to the "linux-kernel" listserv
> that the IDE-DRIVER is going to be "Orphaned" soon........ This bites.
> See "MESSAGE #1 :: Section #2" "MESSAGE #2"
>
> Regards,
> Andre Hedrick
>
> References Below................
>
> MESSAGE #1
>
> Date: Tue, 28 Jul 1998 19:36:19 -0400
> From: Mark Lord <mlord@pobox.com>
> To: "Andre M. Hedrick" <hedrick@Astro.Dyer.Vanderbilt.Edu>
> Cc: linux-kernel@vger.rutgers.edu,
> <snip>,
> <snip>,
> Linus Torvalds <torvalds@transmeta.com>
> Subject: Re: 2.1.111: IDE DMA disabled...BLAH...BLAH...
>
> <snip>
>
> Section #1
>
> There are NO current documented cases of disk corruption due to DMA
> in the Linux IDE driver. If anyone thinks they know of such a case,
> I'd be happy to investigate it .. but all such cases thus far have
> been proven bogus.
>
> Section #2
>
> I will not be updating the IDE driver again
> until the linux/drivers/block/Config.in file
> is restored to its pre-111 state.
> - --
> mlord@pobox.com
> The Linux IDE guy
>
> MESSAGE #2
>
> Date: Wed, 29 Jul 1998 19:44:26 -0400
> From: Mark Lord <mlord@pobox.com>
> To: "Andre M. Hedrick" <hedrick@Astro.Dyer.Vanderbilt.Edu>
> Cc: linux-kernel@vger.rutgers.edu, Linus Torvalds <torvalds@transmeta.com>
> Subject: Re: 2.1.111: IDE DMA disabled...BLAH...BLAH...
>
> <snip>
>
> - --
> mlord@pobox.com
> The (ex-)Linux IDE guy
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: kwrohrer@ce.mediaone.net
> Date: Mon, 10 Aug 1998 10:45:21 -0500 (CDT)
> Subject: Re: 2.1.115 scsi funnies
>
> And lo, Ville Voutilainen saith unto me:
> >
> > Howdy people, I have noticed some scsi peculiarities;
> > my ncr53c810 claims that it does not support synchronous
> > transfers, but AFAIK it does.
> >
> > --clip--
> >
> > >From /proc/scsi/ncr53c8xx/0:
> >
> > Chip NCR53C810a, device id 0x1, revision id 0x12
> > IO port address 0xe000, IRQ number 11
> > Using memory mapped IO at virtual address 0xc4800000
> > Synchronous period factor 25, max commands per lun 32
> >
> > >From log:
> >
> > ncr53c810a-0-<4,*>: SYNC transfers not supported.
> board ^ ^ target
>
> This message says that device 4 on board 0 (all LUN's for that
> device) says it doesn't support SYNC transfers.
>
> > Aren't these controversial?
> Nope, they're not even contradictory. The 53c8xx can do synchronous;
> but your device number 4 says it can't. Therefore the board and driver
> will do synchronous transfers to the other devices (assuming they don't
> object), just not to that one.
>
> Keith
>
> - --
> "The avalanche has already started; |Linux: http://www.linuxhq.com |"Zooty,
> it is too late for the pebbles to |KDE: http://www.kde.org | zoot
> vote." Kosh, "Believers", Babylon 5 |Keith: kwrohrer@enteract.com | zoot!"
> www.midwinter.com/lurk/lurker.html |http://www.enteract.com/~kwrohrer | --Rebo
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: Linus Torvalds <torvalds@transmeta.com>
> Date: Mon, 10 Aug 1998 09:16:31 -0700 (PDT)
> Subject: Re: pre-2.1.116-1 breaks Cyrix 6x86 support
>
> On Mon, 10 Aug 1998, Rafael Reilova wrote:
> >
> > If there is no way to force gcc to keep the order right (in all cases),
> > could we keep the fixed inline asm functions for setC6x86() and getC6x86()
> > I sent you and leave this alone, (please!).
>
> Actually, gcc always keeps the order of IO instructions. The problem here
> was that the getcx/setcx stuff was changed to macros, and when used as
> macros the ordering was not well-defined for the case where you nest them.
>
> Fixed, I'll make a pre-116-2 soon for cyrix users to test..
>
> Linus
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: Martin Mares <mj@ucw.cz>
> Date: Mon, 10 Aug 1998 18:59:30 +0200
> Subject: Re: 2.1.115 Extended VGA
>
> Hello,
>
> > For 2.1.112 and 2.1.115, I have not been able to get the extended vga
> > console to work (80x43). Can somebody help me out. I'm sure I've selected
> > the right configuration options:
>
> Can you explain what do you try to do and what exactly happens?
>
> Have a nice fortnight
> - --
> Martin `MJ' Mares <mj@ucw.cz> http://atrey.karlin.mff.cuni.cz/~mj/
> Faculty of Math and Physics, Charles University, Prague, Czech Rep., Earth
> "ADA -- A Dumb Acronym"
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: Martin Mares <mj@ucw.cz>
> Date: Mon, 10 Aug 1998 19:02:04 +0200
> Subject: Re: std VGA text console funnies
>
> > For what it's worth, I've had this type of screen corruption with
> > every version of Linux from 1.2.13 through 2.0.0 - 2.0.35, and
> > 2.1.115 as well. This has happened on every computer I've used
> > Linux on, with any video card, so it is definitely not a hardware
> > problem.
>
> Are you really sure it happens with 2.1.115?
>
> Have a nice fortnight
> - --
> Martin `MJ' Mares <mj@ucw.cz> http://atrey.karlin.mff.cuni.cz/~mj/
> Faculty of Math and Physics, Charles University, Prague, Czech Rep., Earth
> "Error: Compiler unable to abort!"
>
> Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
>
> ------------------------------
>
> From: Michel Catudal <bbcat@netonecom.net>
> Date: Mon, 10 Aug 1998 13:07:09 -0400
> Subject: Re: PPP --- TCP bad checksums
>
> Alan Cox wrote:
> >
> > > There have been many reports that use if VJ compression triggers errors
> > > with some ISP equipment. You might try turning that option off an seeing
> > > what happens.
> >
> > Be that as it may (and its definitely the case) it doesn't cover the
> > "2.0.35 works perfectly" cases. All of these appear to be PPP - is anyone
> > seeing them wit SLIP (does anyone run SLIP any more)
> >
>
> I encountered a new problem with the PPP at noon and I was wondering
> if anyone has a clue as to what causes this.
>
> I connect as usual with ifup ppp0
> the line goes dead for some reason and I let it try to connect
> again after it's delay.
>
> I try to load Netscape which bombs out. I look for a locked file,
> no lock file. I try again, bombs out again.
> I get out of X and delete the lock file which happened to be there
> this time. I try to go back to X. I just get the starting screen, no
> blue background. It just sit there. I kill it with ctrl alt backspace.
>
> I call ifdown ppp0, line doesn't turn off.
>
> I look in the log and see messages about the modem being locked.
> I reboot and problem seems gone.
>
> The PC has been on all night, last I rebooted was yesterday afternoon.
>
> I have never had that problem with the 2.0.34 kernel or any other
I have compiled 2.1.115 with SMP ... all is OK except that
I can't print graphics to my HP printer.

I am using

cat xxx > /dev/par0

with 2.0.35 the equivlaent command

cat xxx > /dev/par1

works fine. since I am getting garbage on the printer I suspect that
some initialization bytes are not being sent properly.

Any help appreciated.

Dave

-- David A. Fournier Otter Research Ltd. PO Box 265, Station A Nanaimo, B.C. V9R 5K9, Canada voice/fax (250) 756-0956 http://otter-rsch.com

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html

Next message: Alan Cox: "Re: IDE-Driver Disk Accounting Missing, Maybe SCSI too......"
Previous message: Daniel Thom: "Re: 2.1.115 Panic Shortly After Boot (Interrupt related???)"