Linux-Development-Sys Digest #941, Volume #6 Thu, 8 Jul 99 17:14:21 EDT
Contents:
ASSEMBLER problem (Vincent)
Re: Why we are still holding on to MS Windows ("Thomas Steffen")
Re: RedHat 6.0 Installation crashes... (Marc Jauvin)
Re: Can Linux Boot and Run without a BIOS? (Jim Robertson)
Re: CD-ROM File Time Bug (Peter Samuelson)
ASSEMBLER SYNTAXE in C (Vincent)
Re: 2.2.10-ac8 won't link on sparc (Pete Zaitcev)
2 Sound cards (Caleb Lyness)
Re: ASSEMBLER problem (Marcus Sundberg)
Re: Writing sound driver for Linux (ellis)
Re: gethostbyname from kernel (Andrew Lipnitsky)
Writing sound driver for Linux (lefebvre herve)
Re: Filesize larger than 2 GB on Intel machines an Linux 2.0.36
([EMAIL PROTECTED])
2.2.10-ac8 won't link on sparc (Sverker Wiberg)
Re: Why not C++ (Greg Comeau)
Re: Index of known issues/bugs? (Konrad Mieredorff)
Re: kernel (Konrad Mieredorff)
Re: 2 Sound cards (Konrad Mieredorff)
Re: Why not C++ (Josh Stern)
----------------------------------------------------------------------------
From: Vincent <[EMAIL PROTECTED]>
Crossposted-To: linux.dev.kernel
Subject: ASSEMBLER problem
Date: Thu, 08 Jul 1999 11:49:55 +0200
How can I put a varaible in a register
void toto()
{
int val=0;
__asm__("mov val,ax");
printf("%d\n", val);
}
When I compil, I've this message: unindentifer ax !!!
I've tryed with %ax but the variable val is always 0 but it's compil and
run.
Thanks.
--
=========================================
DEVERRE Vincent - MCII : [EMAIL PROTECTED]
=========================================
------------------------------
From: "Thomas Steffen" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: Why we are still holding on to MS Windows
Date: 08 Jul 1999 14:29:26 +0200
Mads Dydensborg <[EMAIL PROTECTED]> writes:
> My coworker has a setup, where when he plugs his linux portable into
> the network and secondary screen, it automatically uses the mouse and
> keyboard from the workstation at his desk. He can then copy and paste
> between the machines, etc.
the program is called x2x, and it is basically just a small hack that
forwards events on the very right (eg) side of the screen to another X
server (using generic events). nothing special really, apart from
being another demonstration of excellent design.
btw, you can even control a windows screen from your UNIX/X X server,
the program is called x2vnc. this is something that really surprised
me (but it is *very* convenient). so there way be a place of windows
after all: running legacy applications in a "shell" controlled from a
real system... :-)
--
linux, linuctis - f, das beste Betriebssystem ;-)
------------------------------
From: Marc Jauvin <[EMAIL PROTECTED]>
Subject: Re: RedHat 6.0 Installation crashes...
Date: Thu, 08 Jul 1999 08:05:21 -0400
Daniel Robert Franklin wrote:
>
> Marc Jauvin <[EMAIL PROTECTED]> writes:
>
> >John Bell wrote:
> >>
> >> Marc Jauvin wrote:
> >> >
> >> > I tried many times to install RedHat 6.0 with 160MB of RAM and the
> >> > installation keeps crashing when creating the filesystems...
> >> >
> >> > I tried with 32MB of RAM and the install worked fine.
> >> >
> >> > After installation, every time I "e2fsck" my 11GB /usr/local partition
> >> > with more that 32MB of RAM, the kernel core dumps.
> >> >
> >> > Any idea what might be causing that?
> >> >
> >>
> >> Bad memory. Have it replaced.
>
> >That's what I thought at first, but I swapped the Harddisk in another machine
> >with 64MB and same thing happened.
>
> Bizarre. I have installed RH6 successfully on a machine with 128 MB.
>
> However, I *did* have trouble some time ago with formatting large
> partitions with a 2.0.31 kernel (or was it 30?). This turned out to have
> a number of serious bugs. I initially blamed the HDD (the infamous Quantum
> Bigfoot) but I repeated the operation with a 2.0.36 kernel and it was
> fine.
Yes, it seems it only happens with LARGE partitions. I will do more testing
but I am havin trouble with a 11GB partition.
Every time I try to use mke2fs or e2fsck on that partition it fails with a
core dump of the kernel or the program (e2fsck/mke2fs).
-- marc
------------------------------
From: Jim Robertson <[EMAIL PROTECTED]>
Subject: Re: Can Linux Boot and Run without a BIOS?
Date: Thu, 08 Jul 1999 12:12:59 GMT
Byron A Jeff wrote:
>
> In article <[EMAIL PROTECTED]>,
> Joe Pfeiffer <[EMAIL PROTECTED]> wrote:
> -Bill Anderson <[EMAIL PROTECTED]> writes:
> -
> -> Jim Robertson wrote:
> -> ...
> -> > However, I'm pretty sure that once Linux is running, it does not depend
> -> > on the BIOS at all. Of course, the root file system must reside on a
> -> > device whose driver is not in a module, but compiled into the kernel.
> ->
> -> Nope, it can be a module, and for many people it is.
> -
> -How do you load the module before you have a driver that can read the
> -root file system? The config help (for 2.2.5) specifies
> -
> -``Do not
> -compile this driver as a module if your root filesystem (the one
> -containing the directory /) is located on an IDE device.''
> -
> -for CONFIG_BLK_DEV_IDE, and makes similar comments about SCSI.
>
> By using a initial Ramdisk to load the initiall root filesystem, run insmod
> to load the appropriate module, then switch to the real filesystem.
>
> I use this exact process to run Linux out of a single file in a DOS/Windows
> FAT filesystem without using UMSDOS, which under a FAT 16 FS sucks up an
> enormous amount of wated space due to bad clustering. I create a single file
> for the filesystem, attach it to a loopback device, format and mount the
> loopback device, and install Linux into it. Using loadlin to boot the kernel
> and a initial ramdisk, I mount the DOS filesys, load the loop driver, reattach
> the file to the loop device, and make that loop device the real root filesystem
> after which everything proceeds as normal. Save space, doesn't require
> repartitioning, and in fact the IDE, SCSI, and all the filesystems could be
> loaded as modules. I don't do it out of simple convenience.
>
> BAJ
Ramdisk is a module in my kernel. If I included it at all, can't
remember now. When my system boots, it mounts root using an ext2
filesystem on a partition of an IDE drive. Anyhow, with my present
kernel configuration, ramdisk can't be /. This rule is not because of
lack of BIOS support, it's just that the kernel needs to have at least
one filesystem running before it can insmod any others.
I would venture a guess that your unique configuration includes ramdisk
as a native driver, not as a module.
-- JRob
------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Subject: Re: CD-ROM File Time Bug
Date: 8 Jul 1999 02:52:59 -0500
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[Terry Boldt <[EMAIL PROTECTED]>]
> How does the 'system' know that the time stored on the CD-ROM is UTC?
Because you used a Unix variant to record it. (Actually a CD-burning
program wouldn't *have* to use UTC but I think mkisofs is sensible
enough to do so.)
If you didn't burn the CD in Unix, then Heisenberg's Uncertainty
Principle gives you a 12-hour window each way for when the CD was
actually burnt. Note that this has nothing to do with the Unix client;
you have that *same* 12-hour window if you read the CD in DOS, OS/2,
Winblows, whatever.
In other words, the Unix scheme is *predictable*. Other schemes rely
on knowing where the CD came from.
> Does the standard for writing CD-ROM state that the stored time is
> UTC time? And this is universally recognized by all?
1) Dunno, I've never read ISO-9660. 2) I think so, in the Unix world.
Outside Unix, YMMV. But why would you want to go outside Unix? (:
Think about this: do you care *exactly* when a CD file was created? If
you *don't* really care, there is no issue here. If you *do* care, why
are you willing to be off by as much as 23 hours because you don't know
the original time zone?
> And if the CD-ROM was recorded in Paris, both differ from the time
> the person in Paris thinks it was recorded.
Not if the person in Paris used a Unix variant.
> For a hard disk, this 'correction' probably makes sense
It *does* make sense. And implementing arbitrary special cases is
definitely not the Unix way, even if it were a good idea here, which I
contend that it isn't.
> If I backup a file to a floppy at Home on the east coast and then fly
> (with my laptop and floppy) to the west coast. Correct the TZ setting
> when I arrive so that the time displayed makes sense locally. I then
> display the contents of the floppy directory to verify the time on
> the file to ascertain that I have the correct floppy and backup, the
> file time reported is going to appear completely wrong since it has
> now been 'corrected' to west coast time with a 3 hour difference.
No. You're looking at the problem wrong. Try this perspective:
TZ=EST5EDT for the East Coast, we're in DST so time is -0400.
- Back up file at 3:00 PM which is 19:00 UTC.
- Fly to the West Coast. Six hours later you are settled in your
hotel room, set your watch back three hours and have dug out your
laptop.
TZ=PST8PDT now, so time is now -0700.
- This means it is 6:00 PM now, local time.
- Look at file. `ls' notes your time zone and says you created the
file at 12:00 noon.
- Hmmm, `date' tells you it's 6:00 PM now, so that was six hours ago.
- Sure enough, you created the file about six hours ago.
- Yup, it's the same file.
It's entirely consistent. What is the problem?
Now the problems with what you are wanting:
(a) You are assuming that a computer will only have *one* time zone
active at any given time. What if five users are logged in from five
different time zones? Which time zone should the system report
removable media files in? The present system works fine, because the
kernel does *everything* in UTC and the conversion happens in userspace
with the help of your custom TZ variable.
(b) What's "removable" and what isn't? You can't always tell. Thanks
to loopback mounting, you can mount a floppy disk image almost the same
as a floppy ... so do you treat the disk image as "removable" or not?
(If not, that's an ugly special case.) How about SCSI ZIP drives?
They look (to the computer) a lot like other SCSI drives -- are you
going to insert a kludge into the kernel to detect the difference?
(c) Why stop with removable media? Why not correct for network
filesystems, by where they're being mounted from? (Not necessary,
actually, since NFS servers usually use UTC and the SMB protocol is
explicitly TZ-aware, but one can imagine a network filesystem you
couldn't make assumptions about.)
------------------------------
From: Vincent <[EMAIL PROTECTED]>
Subject: ASSEMBLER SYNTAXE in C
Date: Thu, 08 Jul 1999 15:23:16 +0200
J'ai trouv� la syntaxe ASM � utiliser en C sous Linux Plateform INTEL,
voir
I've found the ASM syntax to use in C language under Linux Plateform
INTEL, see
http://www.castle.net/~avly/djasm.html
--
=========================================
DEVERRE Vincent - MCII : [EMAIL PROTECTED]
=========================================
------------------------------
Subject: Re: 2.2.10-ac8 won't link on sparc
From: [EMAIL PROTECTED] (Pete Zaitcev)
Date: Thu, 08 Jul 1999 17:02:37 GMT
>When I tried to compile linux 2.2.10-ac8 on a sun4m (quad ross cpus),
>the final link stage failed with the error
> fs/filesystems.a(proc.o): In function `mem_read':
> proc.o(.text+0x23b0): undefined reference to `pte_read'
> proc.o(.text+0x23b0): relocation truncated to fit: R_SPARC_WDISP30
>pte_read
>[...]
>/Sverker
Many people complain about that... I do not use 2.2.x myself anymore
so all I can suggest is to rip off the definition from 2.2.5 which
was known to work.
Goog luck,
--Pete
------------------------------
From: Caleb Lyness <[EMAIL PROTECTED]>
Subject: 2 Sound cards
Date: 8 Jul 1999 17:29:15 GMT
Does anyone know how to use 2 sound cards under
Linux. Is it possible to run them both using
modules? Or does the support need to be compiled
into the kernel?
Cheers
Caleb.
------------------------------
From: Marcus Sundberg <[EMAIL PROTECTED]>
Crossposted-To: linux.dev.kernel
Subject: Re: ASSEMBLER problem
Date: Thu, 08 Jul 1999 17:02:19 +0200
Vincent wrote:
>
> How can I put a varaible in a register
>
> void toto()
> {
> int val=0;
>
> __asm__("mov val,ax");
> printf("%d\n", val);
> }
>
> When I compil, I've this message: unindentifer ax !!!
>
> I've tryed with %ax but the variable val is always 0 but it's compil and
> run.
That code is completely bogus.
Read the "Extended Asm" section in the gcc manual, as well as the
i386 section in the GNU as manual.
//Marcus
--
===============================+====================================
Marcus Sundberg | http://www.stacken.kth.se/~mackan/
Royal Institute of Technology | Phone: +46 707 295404
Stockholm, Sweden | E-Mail: [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (ellis)
Subject: Re: Writing sound driver for Linux
Date: 8 Jul 1999 17:46:11 GMT
In article <[EMAIL PROTECTED]>,
lefebvre herve <[EMAIL PROTECTED]> wrote:
>Where can I find some docs/samples about writing linux drivers for sound
>cards ?
Are there plenty of samples in the kernel source tree?
--
http://www.fnet.net/~ellis/photo/linux.html
------------------------------
From: Andrew Lipnitsky <[EMAIL PROTECTED]>
Subject: Re: gethostbyname from kernel
Date: Thu, 08 Jul 1999 20:47:44 +0300
[email protected] wrote:
> Is there a way to do a gethostbyname within the kernel? I am currently
> hacking nfs to support transparent replication, but i need to convert
> hostname->ip to do this. any ptrs? thanks.
>
> ben
>
I have more general question. Is it way to call glibc2.1- function
(for example gethostbyname) from linux kernel?
Thanks in advance.
Andrew.
------------------------------
From: lefebvre herve <[EMAIL PROTECTED]>
Subject: Writing sound driver for Linux
Date: Thu, 08 Jul 1999 16:49:16 +0100
Hi,
Where can I find some docs/samples about writing linux drivers for sound
cards ?
Thanks.
--
H.Lefebvre
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.misc,comp.os.linux.setup
Subject: Re: Filesize larger than 2 GB on Intel machines an Linux 2.0.36
Date: Thu, 8 Jul 1999 18:20:45 +0059
Get a real computer :-).
>From an overbearingly smug Alpha owner!
--
Richard Simpson
Farnborough, Hants, Uk Fax: 01252 392118
[EMAIL PROTECTED]
I am not aware of any views shared by myself and my employers.
------------------------------
From: Sverker Wiberg <[EMAIL PROTECTED]>
Subject: 2.2.10-ac8 won't link on sparc
Date: Thu, 08 Jul 1999 18:06:41 +0200
When I tried to compile linux 2.2.10-ac8 on a sun4m (quad ross cpus),
the final link stage failed with the error
fs/filesystems.a(proc.o): In function `mem_read':
proc.o(.text+0x23b0): undefined reference to `pte_read'
proc.o(.text+0x23b0): relocation truncated to fit: R_SPARC_WDISP30
pte_read
Some searching revealed that .../include/asm-*/pgtable.h defines
`pte_read()' for most platforms. asm-sparc (and asm-generic) lacks this
definition. It should be fixed somehow (either in
`include/asm-sparc/pgtable.h' or in `fs/proc/mem.c'). A quick check
showed that it had not been fixed in ac9.
Has anybody else been hit by this? Has anybody come up with a
fix/workaround?
/Sverker
------------------------------
From: [EMAIL PROTECTED] (Greg Comeau)
Crossposted-To: comp.os.linux.development.apps
Subject: Re: Why not C++
Date: 8 Jul 1999 15:37:05 -0400
Reply-To: [EMAIL PROTECTED]
In article <[EMAIL PROTECTED]>
[EMAIL PROTECTED] writes:
>Greg Comeau wrote:
>>
>> In article <[EMAIL PROTECTED]>
>[EMAIL PROTECTED] writes:
>> >In my opinion C++ is not the most powerful, but the first hyped OOPL.
>>
>> Wow, if there ever was a statement that was wrong, this is it.
>> This completely ignores the history of OOPL's and C++.
>> This does not consider the OOPL's at the time there _were_ hyped.
>> C++ had _every_ opportunity to fail, exactly because it was NOT hyped.
>> It was strictly though it's own merits that it is where is it today.
>
>You can believe that all you want but I know of at least two major
>government contracts where that is not true. I admit it is only my
>personal experience and it may be a minority experience. But I have
>experienced the hype first hand. I've had managers and civil servants
>pushing C++ not because of merit, but because they either read some
>article or went to a seminar pushing OO this and OO that. They come away
>from the experience a true convert, saying that C++ is the one and only
>OOPL.
>
>I also worked with engineers that didn't want to learn Ada and argued
>forcefully that C++ was easy to learn, its based on C don't you know. It
>was sold that way to management. I think that is hype, I guess you don't
>or never experienced it.
>
>So if my mistake was claiming is was the first hyped OOPL. So be it.
>Doesn't change the fact that C++ was hyped big time, at least on the
>government contracts I've worked on.
My remark was not an opinion. Let's just take this one example:
It is a fact that while Smalltalk was being hyped, that every effort
was being made to destroy C++ even before it barely got out the door.
There were even internal naysers in AT&T. Probably these attempts were
part of the reason for the C++ synergy at the time. It is a fact that
C++ received NO marketing backing for its first 8 years or so.
Even after AT&T released it, not much was done. There are not beliefs,
that are what happened.
Now, it is true that _after that_ time C++ _was_ hyped. Quite so.
I seem to recall this happening after late 1988 or so, and
from that point on totally believe your examples above, as I experienced
it myself.
Anyway, yes, I was challenging the "first" claim. C++ was for sure
hyped big time Xth'ed. :)
This same blind pushing is part of the current Java phenomenon.
Kudos to Sun for their excellent marketing machine.
- Greg
--
Comeau Computing, 91-34 120th Street, Richmond Hill, NY, 11418-3214
Producers of Comeau C/C++ 4.2.38 -- New Release! We now do Windows too.
Email: [EMAIL PROTECTED] / Voice:718-945-0009 / Fax:718-441-2310
*** WEB: http://www.comeaucomputing.com ***
------------------------------
From: Konrad Mieredorff <[EMAIL PROTECTED]>
Subject: Re: Index of known issues/bugs?
Date: Thu, 08 Jul 1999 21:33:56 +0200
Peter Samuelson wrote:
>
> [Konrad Mieredorff <[EMAIL PROTECTED]>]
> > Is there some sort of index which lists
> > proplems/bugs/doubts/whatsoever?
>
> Bugs in what?
>
> If you mean kernel bugs, no, not really. For very serious bugs, see
> Richard's "Kernel Newsflash":
Your guess was right ... i just forgot to mention it
> http://www.atnf.csiro.au/~rgooch/linux/docs/kernel-newsflash.html
>
> For some other bugs, see Alan's "jobs page" which, despite what it
> says, may not be *quite* up-to-date:
>
> http://roadrunner.swansea.uk.linux.org/jobs.shtml
>
> If you want to volunteer to run a "real" bug repository based on (for
> example) Debian's BTS or Samba's Jitterbug, there has been some noise
> in the past that some kernel developers would be willing to use it.
> But nobody seems to think it worth *their* time to set up and run.
Well, i have no idea how much time i would have to spend on such a thing
but on the other hand, someone should do it ... so why not me ;-}. I'll
think about it.
- Konrad Mierendorff
------------------------------
From: Konrad Mieredorff <[EMAIL PROTECTED]>
Subject: Re: kernel
Date: Thu, 08 Jul 1999 21:35:28 +0200
Astro wrote:
>
> What is that, kernel?
> Astro.
Hey! your're in the wrong group.
------------------------------
From: Konrad Mieredorff <[EMAIL PROTECTED]>
Subject: Re: 2 Sound cards
Date: Thu, 08 Jul 1999 21:37:13 +0200
Caleb Lyness wrote:
>
> Does anyone know how to use 2 sound cards under
> Linux. Is it possible to run them both using
> modules? Or does the support need to be compiled
> into the kernel?
>
> Cheers
> Caleb.
See http://www.alsa-project.org
------------------------------
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.networking
Subject: Re: Why not C++
From: [EMAIL PROTECTED] (Josh Stern)
Date: Thu, 08 Jul 1999 16:28:27 GMT
Craig Graham <[EMAIL PROTECTED]> wrote:
>[EMAIL PROTECTED] (Mike McDonald) wrote:
>> =09Craig Graham <[EMAIL PROTECTED]> writes:
>> > Compiled Java would have extra overheads relative to C++,
>> > due mostly to the absense of a delete operator. You have to either:
I'm no expert in Java - especially state-of-the-art java compilers - but
my thought would be that it is the much higher percentage of non-local
variables that would cause the greatest performance hit in Java.
For local variables, of course, there is no delete or garbage collection
required. I'd be curious to learn about whether Java compilers are able
to transform a lot of stuff to local variables behind the scenes.
If they can do that, then I would think that they could use explicit
allocation and deallocation as well.
>> You think delete() is quick?
>
>Delete can be quick. It depends how it's implemented.
Like most performance discussions, a lot probably depends on context.
Part of the relevant context would probably include program(mer) style.
Even such a simple thing as freeing dynamic storage in the reverse order
from allocation might turn out to be play a decisive role in the
comparison.
- Josh
------------------------------
** 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: [EMAIL PROTECTED]
You can send mail to the entire list (and comp.os.linux.development.system) via:
Internet: [EMAIL PROTECTED]
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Development-System Digest
******************************
