Linux-Development-Sys Digest #846, Volume #6 Sat, 19 Jun 99 15:13:51 EDT
Contents:
Re: How to dump printk() msgs in XFree86? (Peter Samuelson)
Re: Kernel Question ("Ross Vandegrift")
Re: RAID-1 and 2.2.9 revisited (bryan)
Re: TAO: the ultimate OS (Stefaan A Eeckels)
Re: TAO: the ultimate OS (Christopher B. Browne)
Re: How to unload module with screwy refcount? (Konrad Mieredorff)
Re: Translation of linux to minor languages (Fredrik Roubert)
Re: asus agp v3800 driver for linux (Timothy Murphy)
Re: Can Linux Boot and Run without a BIOS? ("H�rst Christian Wimmer-Rue")
Re: Can Linux Boot and Run without a BIOS? (Byron A Jeff)
Re: How to read files in a driver (Byron A Jeff)
Re: Can Linux Boot and Run without a BIOS? (Byron A Jeff)
how to get pathname from process ID ("Soohyung Lee")
Re: using C++ for linux device drivers (Philipp Thomas)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Peter Samuelson)
Subject: Re: How to dump printk() msgs in XFree86?
Date: 18 Jun 1999 21:15:47 -0500
Reply-To: Peter Samuelson <[EMAIL PROTECTED]>
[Eric Fowler <[EMAIL PROTECTED]>]
> <!doctype html public "-//w3c//dtd html 4.0 transitional//en">
Please... ...don't... ...do... ...that.
> I am starting my first device driver project in LInux, using
> xf86 as my dev environment. I have noticed that if I am not
> in xf86, my printk() msgs go to the console where I want them,
> but when I load or tweak the driver from the xterm, they don't
> appear. Howcum? How can I fix this?
xterm -C
or I suppose
tail -f /var/log/messages
--
Peter Samuelson
<sampo.creighton.edu!psamuels>
------------------------------
From: "Ross Vandegrift" <[EMAIL PROTECTED]>
Subject: Re: Kernel Question
Date: 19 Jun 1999 03:33:37 GMT
> 1) What is a jiffy? I assume it is some sort of time measurement. Is it
> 1 jiffy = 0.01 seconds?
Like you guessed, a jiffy is a unit of measure of time. I recall reading
it's about 1/60 of a second. However, as has already been posted, that's
not what Linux means by jiffy.
Ross
------------------------------
From: bryan <[EMAIL PROTECTED]>
Subject: Re: RAID-1 and 2.2.9 revisited
Date: Sat, 19 Jun 1999 10:18:20 GMT
John Hughes <[EMAIL PROTECTED]> wrote:
: John Burton <[EMAIL PROTECTED]> writes:
: > [EMAIL PROTECTED] wrote:
: > >
: > > I'm currently installing a RAID 1(hardware) system with 2 mirrored 9GB
: > > SCSI drives and a 4GB IDE sytem disk.
: > > I have simple question - what's the most appropriate partitionig sheme
: > > for a web server? Why I shouldn't put the swap on the RAID drives?
: > >
: >
: > Ummm as far as "why..." Do you really need to mirror you swap space?
: Umm... Of course you do?
: The whole point of mirroring is to let you keep running even if a disk
: dies.
good point!
and with good hardware raid-1, writes are still at 1x (no slower than
regular scsi) and reads are faster than 1x.
--
Bryan [at] Grateful.Net
http://www.Grateful.Net
------------------------------
From: [EMAIL PROTECTED] (Stefaan A Eeckels)
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: TAO: the ultimate OS
Date: 19 Jun 1999 12:42:55 GMT
In article <7kchs7$nvf$[EMAIL PROTECTED]>,
[EMAIL PROTECTED] (Terry Murphy) writes:
> In article <7kadij$6gp$[EMAIL PROTECTED]>,
> Stefaan A Eeckels <[EMAIL PROTECTED]> wrote:
>
>>So pulling knowledge of file structures (be it indexes or links)
>>into the OS is not a step forward, but a return to good, old,
>>mainframe concepts.
>
> On the contrary. You are making the grave mistake in considering
> that anything "newer" is "a step forward". In reality, Unix has
> set the computer industry back thirty years (and counting, of course).
If you read something disparaging in my reference to mainframes,
then I urge to to read the quote again. Revisiting existing
concepts is not a step forward. It also is not necessarily a
bad thing.
Now to the main part of your flame-bait (unless you *only* read
comp.os.misc, in which case you're forgiven) ;-)
> The only reason Unix was so massively commercially successful, was
> precisely because it was so minimally functional -- it was so easy to
> port because it didn't actually DO anything interesting. In the
> workstation boom of the early to mid '80's everyone and their mother
> made their own workstation and they ported Unix to it -- it was much
> cheaper to pay a few hackers to port an existing system to the machine,
> than to pay some professional engineers to design a serious system.
> The whole Unix "small is beautiful" philosophy won out for business,
> not technical reasons.
It was easier to port UNIX than to write an OS from scratch. The
fact that the principals of comapnies such as Sun cut their teeth
on UNIX at Berkeley (and liked it, I guess) was also a significant
factor. BTW, the people who did BSD and SunOS *are* professional
software engineers (that's at least what their degree says).
Today, hardware companies do not even dream of developing an OS, and
no longer consider porting UNIX. They design their hardware to be
compatible with Windows, and write drivers.
> In fact, many of the mainframe features that Unix original "rebelled"
> against have crept back into the system -- ACL's, for example, or
> file locking. Only they are much more kludgy and non-standard and
> non-functional than they are on the real systems. Mandatory locking,
> for example, is all but non-existent on Unix. Some systems have it,
> some don't, and you have to go through kludgy chmod sequences to
> turn it on. But then it still only works in special circumstances;
> it won't work over NFS for example. Real systems such as VMS and
> even Windows NT have no problem with simple concepts such as this.
NFS is not a great protocol, granted. Pretending that SMB is any
better displays your profound lack of knowledge.
> Then there is the problem of a record based file system. Since Unix
> doesn't know about anything but bags of bytes, there is no concept
> of records. Users have to resort to kludgy things such as seeks. Not
> so useful when you know the key of the record you want. Then there's
> record locking. Some Unix's may let you lock portions of files (not
> mandatorily of course though), but that seems very roundabout to locking
> a record. The basic Unix filesystem is so inefficient for databases that
> many database vendors have to implement their own filesystems. Again,
> real operating systems, such as VMS, have no problem here.
So what? The basic UNIX file system is designed to store lots of small
files. When your average mainframe OS was designed, disk space was so
expensive small text files were stored on paper --next you are going to
lament the fact that you needed to know how big a file was going to
be when you created it (after all, it stopped fragmentation ;-)
There's no guarantee that an OS provided index-sequential file will
be the best support for a relational DBMS (in fact, it isn't).
Similarly, the locking features of the OS might not be the best
choice for a DBMS (in fact, they aren't).
> Adding real file attributes and file types to Unix would be a great
> idea.
Do you really mean turning Oracle into a kernel module? Or do
you want to limit yourself to fixed-length records and index-
sequential? Oh yes, and if someone needs a structuring not
provided for by the OS, will you tell him to go and such eggs?
> Even though most people these days use their Unix machine like
> they do a Windows based pee-cee, file attributes will never become
> a force in Unix, because it couldn't fit into POSIX without some serious
> kludging. Had POSIX been a real API to begin with, there wouldn't be
> a problem.
Well done, Terry.
--
Stefaan
--
PGP key available from PGP key servers (http://www.pgp.net/pgpnet/)
___________________________________________________________________
Perfection is reached, not when there is no longer anything to add,
but when there is no longer anything to take away. -- Saint-Exup�ry
------------------------------
From: [EMAIL PROTECTED] (Christopher B. Browne)
Crossposted-To: alt.os.linux,comp.os.linux.advocacy,comp.os.misc,comp.unix.advocacy
Subject: Re: TAO: the ultimate OS
Reply-To: [EMAIL PROTECTED]
Date: Sat, 19 Jun 1999 13:55:09 GMT
On Sat, 19 Jun 1999 07:54:23 GMT, Vladimir Z. Nuri <[EMAIL PROTECTED]> posted:
>Christopher Browne ([EMAIL PROTECTED]) wrote:
>: More critically, there's little point to implementing *any* object
>: system if there isn't a clear picture of what the ontology of the base
>: object set is supposed to be.
>
>what do you mean by "the ontology of the base object set"..?
See page 200 of Structure and Interpretation of Computer Programs,
2nd Ed., Abelson and Sussman.
"Dealing with large numbers of interrelated types while still preserving
modularity in the design of large systems is very difficult, and is an
area of much current research."
"This statement, which also appears in the first edition of this book,
is just as true now as it was when we wrote it twelve years ago.
Developing a useful, general framework for expressing the relations
among different types of entities (what philosophers call "ontology")
seems intractably difficult. The main difference between the confusion
that existed ten years ago and the confusion that exists now is that
now a variety of inadequate ontological theories have been embodied in
a pletora of correspondingly inadequate programming languages."
Nobody really knows what your object system is about unless you elaborate
on what the hierarchy of object relationships actually are. And based
on the SICP comments, the problem of mapping object relationships onto
languages is, at this point, an intractable problem.
--
Those who do not understand Unix are condemned to reinvent it, poorly.
-- Henry Spencer <http://www.hex.net/~cbbrowne/lsf.html>
[EMAIL PROTECTED] - "What have you contributed to free software today?..."
------------------------------
From: Konrad Mieredorff <[EMAIL PROTECTED]>
Subject: Re: How to unload module with screwy refcount?
Date: Sat, 19 Jun 1999 17:42:27 +0200
Eric Fowler wrote:
>
> I am writing a driver with one eye on the ORA book (Rubini). When
> I try to unload my driver, rmmod whines, "device or resource busy".
> I think this is because I screwed up the ref counts with the MOD_[INC
> | DEC]_USE_COUNT macros. If so, how can I force this thing to die
> without rebooting? If it is some other cause, what might it be?
>
> --
> Vivez sans temps mort!
> (Live without dead time)
> -Situationist International
>
> sockeye [at] rmii [dot] com
>
As far as I remember what I read in Rubini's book it's impossible but at
the moment I'm too lazy to stand up and check it so you will have to
read again ... Sorry!!! ;-)
------------------------------
From: [EMAIL PROTECTED] (Fredrik Roubert)
Subject: Re: Translation of linux to minor languages
Date: 19 Jun 1999 16:02:48 GMT
In article <[EMAIL PROTECTED]>,
Nigel Andrews <[email protected].> wrote:
> The filesystem (ext2 anyway) will accept KOI8 characters for names. Only
> you'll run into problems with things like 'ls' not displaying them
> because of the 8th bit (I presume). In the past I've ended up using the
> correct number of wildcards to select files, not the best method but at
> least it works if it matches what you want.
GNU ls handles 8 bit characters all OK when invoked with the -N switch.
And because ISO-8859-1 is the standard character set in Linux I have
almost no problems at all using files with names like "�ngest",
"nytt_f�rs�k" and similar. The "almost" is that some older programs have
problems with 8 bit characters. But they aren't very common. In fact I
can't come to think of a single one right now.
Using another 8 bit character set than ISO-8859-1 should not imply too
many problems once the system is configured. GNU programs are in general
quite well prepared for internationalization.
H�lsningar // Fredrik Roubert
--
[EMAIL PROTECTED] � +46 46 188127
[EMAIL PROTECTED] � M�llev�ngsv�gen 6c
http://www.efd.lth.se/~d95fr/ � SE-222 40 Lund
------------------------------
From: [EMAIL PROTECTED] (Timothy Murphy)
Subject: Re: asus agp v3800 driver for linux
Date: 19 Jun 1999 17:31:03 +0100
[EMAIL PROTECTED] (Wez Furlong) writes:
>>i have a video card asus agp v3800 (nvidia riva 128/2) 32mb
>>who know where i can find correct driver ???
>You can grab an X-server for TNT and TNT2 cards from www.nvidia.com
>look for resource center then downloads/drivers.
I haven't tried nvidia's driver(s),
but I'm using the standard SVGA driver on RedHat-6.0
(you need a recent version of X like the one there).
Xconfigurator doesn't set up XF86Config properly for this card;
I saw an XF86Config.TNT2 file on one of the linux newsgroups
and this works quite well.
(Of course one has to make a few changes,
eg to the Monitor ranges and probably the Pointer [mouse] section.)
--
Timothy Murphy
e-mail: [EMAIL PROTECTED]
tel: +353-1-2842366
s-mail: School of Mathematics, Trinity College, Dublin 2, Ireland
------------------------------
From: "H�rst Christian Wimmer-Rue" <[EMAIL PROTECTED]>
Subject: Re: Can Linux Boot and Run without a BIOS?
Date: Sat, 19 Jun 1999 15:25:59 GMT
This is addressed to the original poster. (My news server sucks and doesn't
retain messages for more than a day or so, therefore I couldn't reply to the
original message.) You might want to check into the GRUB (GRand Unified
Boot-loader), which is MultiBoot-compliant. The URL is
www.uruk.org/grub
Speaking of which, does anyone know if Linux is MultiBoot-compliant or if
there have been any attempts to use GRUB with Linux? On the page, I could
only find what looks like very stale information. Any kernel gurus got any
fresher info?
<[EMAIL PROTECTED]> wrote in message news:7k893e$k6h$[EMAIL PROTECTED]...
> OK, if i have load the linux boot image into RAM, without
> using the BIOS, then, can i run linux normally? if linux have
> been on the way, will it never call any BIOS service (switch
> to real mode if required), or never use any BIOS data, true?
>
> regard
>
> In article <[EMAIL PROTECTED]>,
> Medical Electronics Lab <[EMAIL PROTECTED]> wrote:
> > [EMAIL PROTECTED] wrote:
> > >
> > > Can Linux boot from chaos and run, without a BIOS?
> >
> > In principle, yes. What boots the boot?
> > Usually a bios!!
> >
> > Suppose you have a system with only RAM, no
> > proms. How do you boot it? Somehow, you have
> > to load executable code into it, and when the
> > processor comes out of reset, it will begin
> > to execute that code. If it's a linux boot,
> > you're on the way. How'd you get the data into
> > RAM?
> >
> > Answer the last question, and you've done it!
> >
> > Patience, persistence, truth,
> > Dr. mike
> >
>
>
> Sent via Deja.com http://www.deja.com/
> Share what you know. Learn what you don't.
------------------------------
From: [EMAIL PROTECTED] (Byron A Jeff)
Subject: Re: Can Linux Boot and Run without a BIOS?
Date: 19 Jun 1999 12:01:46 -0400
In article <7k99kk$ivm$[EMAIL PROTECTED]>,
bill davidsen <[EMAIL PROTECTED]> wrote:
-In article <7k5cak$jaj$[EMAIL PROTECTED]>, <[EMAIL PROTECTED]> wrote:
-| Can Linux boot from chaos and run, without a BIOS?
-
-In theory? Sure. But you need something to load the kernel image into
-memory and jump to it. This is usually done by loading LILO part1 which
-uses the BIOS to load LILO part 2, which loads the kernel. Also, there
-are boot from net ROMs for many NICs, but they are started by the BIOS,
-too.
-
-Once the kernel is up it doesn't use the BIOS for i/o, but I think you
-may have to tell it not to read the config info the BIOS normally
-provides. I haven't looked at that in several years (1.3.x days), so my
-memory is way fuzzy.
-
-Seems to me someone posted about a boot ROM which just copied the kernel
-image into RAM and jumped, in regard to an embedded app. You might
-search on embedded and see if you get hits.
I can help you out here Bill. There was a Linux Journal article about an
embedded system that did what you described. Go to www.linuxjournal.com and
search for EPROM. Used a compressed filesystem that was loaded into a
ramdisk.
BAJ
------------------------------
From: [EMAIL PROTECTED] (Byron A Jeff)
Subject: Re: How to read files in a driver
Date: 19 Jun 1999 12:06:31 -0400
In article <gCB93.4$[EMAIL PROTECTED]>,
Brian Q Bian <[EMAIL PROTECTED]> wrote:
-Sorry for the cross-posting. I'm a newbie to the Linux device driver
-development so if this topic has been brought up numerous times before
-please still help :)
-
-I'm working on a driver that needs to read a files during initialization.
-I tried sys_open("filename", O_RDONLY, 0) and got the -14 which is
-EFAULT. Why am I getting this error message and what is the proper
-way to read files in the kernel mode?(Unfortunately, I couldn't find
-the answers in the "Linux Device Drivers" Book).
Well the answer isn't real simple. There is no easy mechanism for the kernel to
open and read files. The way that it's normally done is that the driver is
coupled with an application in user space, the application opens and reads
the file and passes the information to the driver. So for your app, the driver
can just sit idle until the config app comes along, opens up the config device,
and writes the configuration data.
Hope this helps,
BAJ
------------------------------
From: [EMAIL PROTECTED] (Byron A Jeff)
Subject: Re: Can Linux Boot and Run without a BIOS?
Date: 19 Jun 1999 11:59:55 -0400
In article <bEOa3.10288$[EMAIL PROTECTED]>,
H�rst Christian Wimmer-Rue <[EMAIL PROTECTED]> wrote:
>This is addressed to the original poster. (My news server sucks and doesn't
>retain messages for more than a day or so, therefore I couldn't reply to the
>original message.) You might want to check into the GRUB (GRand Unified
>Boot-loader), which is MultiBoot-compliant. The URL is
>
>www.uruk.org/grub
>
>Speaking of which, does anyone know if Linux is MultiBoot-compliant or if
>there have been any attempts to use GRUB with Linux? On the page, I could
>only find what looks like very stale information. Any kernel gurus got any
>fresher info?
GRUB is at the same level as LILO in terms of this discussion. It doesn't get
run until after the BIOS does its POST and after the loader is loaded from
the disk.
The original poster wanted to know if Linux could be run with no BIOS
available. The answer is yes. However something in the system must still do
initialization and bootloading.
BAJ
------------------------------
From: "Soohyung Lee" <[EMAIL PROTECTED]>
Subject: how to get pathname from process ID
Date: Sun, 20 Jun 1999 03:36:39 +0900
How can I get the full pathname of a process from its pid ?
Can you please help me ?
I really need your help .
Thanks in advance .
- Lee -
------------------------------
From: [EMAIL PROTECTED] (Philipp Thomas)
Subject: Re: using C++ for linux device drivers
Date: Sat, 19 Jun 1999 18:19:24 GMT
On 18 Jun 1999 13:18:14 -0400, [EMAIL PROTECTED] (Alexander Viro)
wrote:
>>complications with using C++. I heard that C++ needs some OS support,
>>especially for calls like "new", "delete" and stuff like that.
>
>It will not get it. It's beaten to death many, many times. Oh, and forget
>about try and catch - they are not going to work. Ditto for standard classes
>- runtime environment is not available too.
Well, I'd like to add: unless someone sits down and writes a C++
Framework for doing Drivers in Linux. I've seen Vireo do it for
Win9x/NT and something like that should be possible, although it would
need a specialist to implement it and I doubt that those capable of
doing it would do it for free.
Philipp
--
the reality is that having most things standardized at the expense
of a few gotchas is better than no standard at all, which is often
the only other choice.
Donn Terry <[EMAIL PROTECTED]>
------------------------------
** 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
******************************
