Linux-Development-Sys Digest #728, Volume #8 Thu, 17 May 01 06:13:12 EDT
Contents:
Re: AIX coff binaries (Chris Cox)
Re: Raid A0 patch with Kernel 2.2.17 and ReiserFS (=?iso-8859-1?Q?Andr=E9?= David)
Re: Help me write a driver (Grant Edwards)
Link to an object. (".")
Re: SIGSEGV is not blocking ("Arthur H. Gold")
Re: SIGSEGV is not blocking (Eric P. McCoy)
Re: Newbie programmer ("Paul Jones")
serial output and keyboard input (Javi Roman)
Re: Raid A0 patch with Kernel 2.2.17 and ReiserFS ("peter.katzmann")
[kernel] packet filtering/modifying (Vyacheslav Burdjanadze)
interrupts too frequent? (Barry Smyth)
Re: Execve and execvp (Josef Moellers)
Re: Raid A0 patch with Kernel 2.2.17 and ReiserFS (Thomas Steffen)
Re: Help me write a driver (Tahar)
xml api ("Karim A.")
Java vs memory ([EMAIL PROTECTED])
Re: xml api (Vyacheslav Burdjanadze)
Re: Path to kernel include directory? (Kasper Dupont)
Re: Newbie programmer (Kasper Dupont)
Re: Raid A0 patch with Kernel 2.2.17 and ReiserFS ("peter.katzmann")
Re: transpareny (Kasper Dupont)
----------------------------------------------------------------------------
From: Chris Cox <[EMAIL PROTECTED]>
Subject: Re: AIX coff binaries
Date: Thu, 17 May 2001 03:10:57 GMT
iBCS != coff
Just fyi...
Also, I think AIX 1.3 predates iBCS... I mean, we're
talking about IBM-RT days with that version level of
AIX aren't we? Maybe they had it for the 386 version
of their PS/2 platforms (even that might be a
stretch)?? Pretty old stuff... my guess. Might
as well be PC-IX.
If it is that old... I don't know of anything it
may have provided that would be worthwile running
under Linux anyway. AIX didn't really get it legs
until about AIX 3.x.
Feel free to prove me wrong....
Chris
BrentRBrian wrote:
>
> No, it's an IBM 'non-free' OS.
>
> the iBCS is "included" into the 2.4 kernel .... I have it turned on,
> still it fails to execute (says it can't).
>
> I had iBCS as a separte package in 7.0, still didn't work.
>
> I think AIX coffs were not "finished".
>
> B
>
> In article <zU2M6.124$[EMAIL PROTECTED]>, "Cameron Kerr"
> <[EMAIL PROTECTED]> wrote:
>
> > In article <[EMAIL PROTECTED]>, "BrentRBrian"
> > <[EMAIL PROTECTED]> wrote:
> >
> >> I have a friend running AIX 1.3 on x86. I noticed the AIX COFF
> >> binaries mentioned in the kernel sources for 2.4, but after #define'ing
> >> the AIX magic number in the header and rebuilding, still no luck.
> >>
> >> Anybody have any idea what would be involved in getting the AIX
> >> binaries to execute under Linux ?
> >>
> >> B
> >
> > You probably want to have a look at something called IBCS. It lets you
> > run various Unix binary formats under Linux.
> >
> > Slackware comes with it, although I've never used it.
> >
> > PS. AIX for x86, is that free by any chance?
> >
------------------------------
From: =?iso-8859-1?Q?Andr=E9?= David <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: Raid A0 patch with Kernel 2.2.17 and ReiserFS
Date: Thu, 17 May 2001 05:03:20 +0200
This is a multi-part message in MIME format.
==============ECC38BC9FC90D9802F6ACAE1
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
[EMAIL PROTECTED] wrote:
>
> I'll add another question to the thread. 2 of the developers I work with say
> that reiserFS should be used for development because it is more robust then
> ext2 and in the embedded system we are developed, the power frequently is
> removed from the target system running Linux. As a consequence of the fact
> that power can be removed at any time, there is no graceful shutdown of this
> system. I am told by one of the developers that reiserFS will be the new
> default file system in the next release of Linux. I wonder what some of the
> developers here have to say about reiserFS and its reliability in a
Check http://www.namesys.com/ for Reiserfs
> situation where power can and will be removed frequently from an embedded
> system.
Power shouldn't be removed frequently from any system relying on it.
If you have it going that way, I'd think it's a bad design ...
But it's your project.
Cheers, Andre
--
"Share the code. If you hide it ain't good."
Popular knowledge
==============ECC38BC9FC90D9802F6ACAE1
Content-Type: text/x-vcard; charset=us-ascii;
name="Andre.David.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Andr� David
Content-Disposition: attachment;
filename="Andre.David.vcf"
begin:vcard
n:David;Andr�
tel;cell:+41792013849
tel;work:+41227676147
x-mozilla-html:FALSE
org:CERN - European Centre for Nuclear Research;EP/NA60
adr:;;;;;;
version:2.1
email;internet:[EMAIL PROTECTED]
note:Geneva, Switzerland
x-mozilla-cpt:;-16000
fn:Andr� David
end:vcard
==============ECC38BC9FC90D9802F6ACAE1==
------------------------------
From: [EMAIL PROTECTED] (Grant Edwards)
Subject: Re: Help me write a driver
Date: Thu, 17 May 2001 03:35:04 GMT
On Wed, 16 May 2001 16:24:17 -0600, Curtis Russell Nielson <[EMAIL PROTECTED]> wrote:
>I am trying to write a very simple Kernel Loadable device driver for a
>character device and I built. The card is a very simple 8255 PPI. I
>am not very familiar with any kind of kernel programming and would
>appreciate some instructions or example code I could look at. Thanks in
>Advance
Two answers:
1) You really ought to buy Rubini's book "Linux Device
Drivers". The second edition that covers kernel version 2.4
wont' be out until July, but most of the basic info will
still be useful.
2) Find a driver for something similar (similar functionality,
same bus), and start modifying it.
--
Grant Edwards grante Yow! Actually, what
at I'd like is a little toy
visi.com spaceship!!
------------------------------
From: "." <[EMAIL PROTECTED]>
Subject: Link to an object.
Date: Wed, 16 May 2001 20:40:16 -0700
This may be a naive question, (this may already exist) but may be not:
I am wondering if Unix/Linux provides at the file system a link to an
object, rather than a file.
Or this may be available at the system call level?
This occured to me, thinking that I would like in a directory, to have a
file wich is a link to a bziped version of a file or set of files,
created only when used.
Of course this reasonning could apply way beyond this.
The potential trouble that I see is the unability to know the size of an
object until created, but this does not seem overwelming.
What do you think?
Thanks for your answers.
Andre G-
------------------------------
Date: Wed, 16 May 2001 23:04:09 -0500
From: "Arthur H. Gold" <[EMAIL PROTECTED]>
Subject: Re: SIGSEGV is not blocking
Linus Torvalds wrote:
>
> In article <[EMAIL PROTECTED]>,
> Michael J. Saletnik <[EMAIL PROTECTED]> wrote:
> >
> >If I explicitly block the SIGSEGV from being delivered, I would not
> >expect the process still to terminate on a SIGSEGV.
>
> Well, the kernel has no alternative. It has to do something. You told it
> not to send the SEGV, so _what_ do you expect it to do?
>
> What Linux will do (an dyou should expect other UNIXes to do the same,
> AIX notwithstanding), is to say "if we generate a fault that we can't
> just continue on, and the fault is blocked, we just ignore that anyway".
>
> Reasonable, no?
>
> >For example, in the example code shown it *is* perfectly acceptable to
> >continue down the codeline since we don't do anything with the null
> >value that we've tried to dereference and faulted on.
>
> It is NOT POSSIBLE!
>
> If the kernel just returns, the process will take the same fault over
> and over and over again. The fault is not going away. The CPU does not
> advance the instruction pointer past the instruction that faulted, and
> the kernel is sure as hell not going to try to decode the instruction to
> see what it should do.
>
> >In other scenarios I might just keep SIGSEGV'ing until
> >unblocked/pending-delivered or it died for another reason.
>
> It's never ever going to be unblocked. The process will never make any
> progress, and thus the process will never even have the _possibility_ of
> unblocking the signal.
>
> >What bothers me is that SIGSEGV was blocked, and yet still
> >delivered, and I want to know why.
>
> Because the kernel had two choices: deliver it despite the programmer
> telling it not to, or killing the process outright because the process
> is no longer doing anything useful.
>
> Now, you might say that killing is better. I might agree. We can vote on
> it, I don't personally particularly care one way or the other. But you
> only get those two choices: kill the dang thing outright, or just ignore
> the blocking. We just don't have any other good choices.
As long as we're talking about SIGSEGVs, a question (if you have
the time to get to it):
Is there a Good Reason why valid addresses (i.e. addresses within
the VM space of a process) but with incompatible page protections
for the purpose, cause system calls to just fail with EINVALs as
opposed to causing a SIGSEGV?
Many+ thanks,
--ag
--
Artie Gold, Austin, TX (finger the cs.utexas.edu account for more
info)
mailto:[EMAIL PROTECTED] or mailto:[EMAIL PROTECTED]
--
I am looking for work. Contact me.
------------------------------
Crossposted-To: comp.os.linux.development.apps,comp.programming.threads
Subject: Re: SIGSEGV is not blocking
From: [EMAIL PROTECTED] (Eric P. McCoy)
Date: 17 May 2001 00:32:49 -0400
[EMAIL PROTECTED] () writes:
> >It could be worse than that. The continuously generated interrupt
> >could prevent the scheduler from ever running, and the machine would
> >be effectively locked-up.
> But it's a trap not an interrupt. The timer interrupt should still
> get serviced.
I don't think so. The fault handler would return, the CPU would
re-execute the faulting instruction (under the assumption that it
would be corrected), and would fault again.
I guess some timer ints could get in during the handler or whatever,
depending on how Linux does things (I have no idea), but the system
would likely be effectively frozen, even if not technically. I think,
but am not sure, that exceptions in userland go through gates to move
up the privilege ring, and I don't know how IA32 CPUs handle
interrupts during state changes like those.
--
Eric McCoy <[EMAIL PROTECTED]>
"Knowing that a lot of people across the world with Geocities sites
absolutely despise me is about the only thing that can add a positive
spin to this situation." - Something Awful, 1/11/2001
------------------------------
From: "Paul Jones" <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux.mandrake,comp.os.linux.development.apps
Subject: Re: Newbie programmer
Date: Thu, 17 May 2001 15:18:33 +1000
Sorry for being vague!! I was after c++ and preferably an ide to go with it.
Thanks for all the help so far!!!
__
P.j
"Paul Jones" <[EMAIL PROTECTED]> wrote in message
news:9dtpfd$550$[EMAIL PROTECTED]...
> I was wondering if anyone could point me in the direction of some help for
> people moving from windows to Linux development. I can't seem to find
> anything on the subject. I'm ok at windows programming, but I'm assuming
> that linux has totally different libraries and stuff. Also I'm still
> learning Linux which makes it a bit hard :)
>
> All help appreciated!
>
> Paul Jones.
>
>
------------------------------
From: Javi Roman <[EMAIL PROTECTED]>
Subject: serial output and keyboard input
Date: Thu, 17 May 2001 07:00:31 +0200
Hi:
I need help about this (I can't find information):
I need display all messages over a serial display (serial
console) but I need
take keyboard input from the PC keyboard (not from serial terminal). I
have a linux box
whithout vga device.
I have compiled kernel with serial-console support and I put this
line in my lilo.conf:
append = "console=ttyS2,9600"
And I put in inittab
...... :/sbin/getty ttyS2
Well, I can see al messages on ttyS2 display and I obtain a login
prompt
but the keyboard PC don�t work. I think that kernel wait inputs from
serial line.
Who can I do that?? please any suggestion??
Best regards, Javi.
------------------------------
From: "peter.katzmann" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: Raid A0 patch with Kernel 2.2.17 and ReiserFS
Date: Thu, 17 May 2001 09:30:11 +0200
Reply-To: [EMAIL PROTECTED]
Cristov wrote:
>
> Does the Raid A0 patch for the 2.2.17 kernel fix whether or not ReiserFS
> will work with Raid5?
When you also will acces the data with nfs, no !
peter
------------------------------
Date: Thu, 17 May 2001 12:22:56 +0400
From: Vyacheslav Burdjanadze <[EMAIL PROTECTED]>
Subject: [kernel] packet filtering/modifying
Not so far i asked about packed capturing in Linux kernel. All you
replied to me how to solve my problem, but now I got similar problem.
I write kernel module which should perform two things:
1. grab IP packet from specified interface and take some action on it
(modify/drop/pass)
2. inject modified packed back into interface queue.
It seems like TUN/TAP driver does something similar but it operates in
user space and I need to configure virtual network interfaces.My task is
to drop/pass/modify incoming/outcoming IP packets transparently to user.
I spent two days looking into kernel source but can't understand a lot
of network stuff and there is a lack of documentation on it. It is
possible to solve my problem in Linux kernel IP? Is there something like
hook mechanism like in Win32?
please , help me. :(
------------------------------
From: Barry Smyth <[EMAIL PROTECTED]>
Subject: interrupts too frequent?
Date: Thu, 17 May 2001 09:34:04 +0100
Reply-To: [EMAIL PROTECTED]
Hi,
I am writing a driver for a PCI card which generate interrupts at
intervals of about 1 millisecond. The interrupt has been requested with
the SA_INTERRUPT flag. The interrupts occur in batches of 256. However
on some occasions when running, some interrupts are skipped leading to
lost data.
Could this be because the operating system is doing something in the
background which lasts a few milliseconds leading to missing my
interrupts? If so is there any way to give my interrupts a high priority
to ensure that they are all acted upon? Is there a stated limit on the
frequency of interrupts in linux?
Thanks
Barry
------------------------------
From: Josef Moellers <[EMAIL PROTECTED]>
Subject: Re: Execve and execvp
Date: Thu, 17 May 2001 10:37:11 +0200
override wrote:
> =
> Hi! I'm programming a Linux shell; i need to convert the following
> execvp() syscall to an execve() syscall... i've been trying to do
> this many ways, but it didn' work... can you please help me?
For one, it would have been helpfull if you had described what "many
ways" you have tried. Most certainly, people will respond with
suggestions that you have already tried.
Here's my sugestion:
the main difference between execvp and execve is that execvp will
automagically traverse your PATH, so one possibility is (no checks, not
tried!)
char *p =3D strdup(getenv("PATH"));
char *pe;
char *cmd; /* Holds command to execute, e.g. "ls" */
char **argv; /* argument vector */
argv[0] =3D cmd;
for (pe =3D strtok(p, ":"); pe; pe =3D strtok(NULL, ":"))
{
char *pathcmd;
pathcmd =3D concat(pe, cmd);
if (execve(pathcmd, argv, envp) =3D=3D -1 && errno !=3D ENOENT)
break;
}
It's just a skeleton, so you need to fill in lots and lots of additional
code.
-- =
Josef M=F6llers (Pinguinpfleger bei FSC)
If failure had no penalty success would not be a prize
-- T. Pratchett
------------------------------
From: Thomas Steffen <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: Raid A0 patch with Kernel 2.2.17 and ReiserFS
Date: 17 May 2001 10:51:17 +0200
<[EMAIL PROTECTED]> writes:
> I'll add another question to the thread. 2 of the developers I work
> with say that reiserFS should be used for development because it is
> more robust then ext2 and in the embedded system we are developed,
> the power frequently is removed from the target system running
> Linux.
As far as I can see ReiserFS is more advanced than ext2 in a few ways:
* more efficient with small files
* no inode-limit
* journal -> no fsck
> As a consequence of the fact that power can be removed at any time,
> there is no graceful shutdown of this system.
ReiserFS might have a slight advantage there, at least the recovery is
faster (not necessarily safer or even safe). And even if the
filesystem is in a consistent state after recovery, your data may not
be :-(
But: ReiserFS is really complex. Source and code is rather large,
computation complexity, too, I think. And it takes about 50 MB of
every partition just for administration (and journaling).
Thomas
------------------------------
From: Tahar <[EMAIL PROTECTED]>
Subject: Re: Help me write a driver
Date: Thu, 17 May 2001 10:58:51 +0200
Reply-To: [EMAIL PROTECTED]
As an example, take a look at:
http://www.scs.ch/~frey/linux/
Curtis Russell Nielson wrote:
>
> I am trying to write a very simple Kernel Loadable device driver for a
> character device and I built. The card is a very simple 8255 PPI. I
> am not very familiar with any kind of kernel programming and would
> appreciate some instructions or example code I could look at. Thanks in
> Advance
Tahar
------------------------------
From: "Karim A." <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: xml api
Date: Thu, 17 May 2001 11:13:56 +0200
Hi all,
I'm looking for a XML api that exist both on Linux/Unix and Windows.
Does anyone know ?
Thanks,
Karim
------------------------------
Subject: Java vs memory
From: [EMAIL PROTECTED]
Date: Thu, 17 May 2001 09:15:45 GMT
Hi,
this might be offtopic here but I haven't got a better idea where to ask ...
I read somewhere, when I start several instances of the same program, then the
loader puts only one instance of the code segment into the memory, and all the
processes will refer to this.
I wonder if this allso works with java code.
Thanks in advance
--
Imre Palik
Microsoft Certified Angry OS Rebooter
------------------------------
Date: Thu, 17 May 2001 13:28:36 +0400
From: Vyacheslav Burdjanadze <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.apps
Subject: Re: xml api
"Karim A." wrote:
>
> Hi all,
>
> I'm looking for a XML api that exist both on Linux/Unix and Windows.
>
> Does anyone know ?
>
> Thanks,
>
> Karim
search for libxml
------------------------------
From: Kasper Dupont <[EMAIL PROTECTED]>
Subject: Re: Path to kernel include directory?
Date: Thu, 17 May 2001 09:46:44 +0000
Rene Herman wrote:
>
> Jim Cochrane wrote:
>
> > I think (and I hope someone more knowledgeable will correct me if I'm
> > wrong) that it's standard for linux distributions to have a symbolic
> > link from the linux include directory to /usr/include/linux.
>
> It's not only standard, it's pretty much *required* to have a set of
> kernel headers in /usr/include/{linux,asm}. Sometimes these directories
> will be just symlinks to /usr/src/linux/include/{linux,asm}, sometimes
> they will be real directories containing the headers from some fixed
> kernel version that the rest of the system was compiled against.
They usually exist but in RH7.0 they are another
version than the kernel installed. This means that
using this under RH7.0 would compile for another
kernel version than installed. I don't know if any
other distribution has this problem as well.
>
> I'm not a kernel and/or driver developper, so somebody please correct
> if *I* am wrong, but I believe that driver code should just rely on
>
> #include <linux/...>
> #include <asm/...>
The c files use this kind of includes, but the
Makefile specifies the path where the files are
to be read from.
>
> The kernel supplied drivers do. If 3rd party code relies on a very
> specific kernel version (and/or kernel headers from, say, net or scsi),
> I guess it should tell the the person that's building it to dump the
> driver source into that specific tree and use the master kernel
> Makefile to build. The kernel build process will make sure that the
> #include will pick up the version specific headers before looking
> through /usr/include/{linux,asm}
It would be a lot easier to require the user to
create a symlink in the build directory pointing
to the kernel directory to compile against.
>
> Rene.
--
Kasper Dupont
------------------------------
From: Kasper Dupont <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux.mandrake,comp.os.linux.development.apps
Subject: Re: Newbie programmer
Date: Thu, 17 May 2001 09:56:24 +0000
William Dandreta wrote:
> The first release costs $1000 but a GPL version should be available via
> download this summer. You might want to checkout the Kylix info on Borland's
> website and the Kylix newsgroup as well.
Where do you have that GPL information from,
the only information I could found was an
article from november 13th november 2000
with the title:
"Borland to Release CLX Under Dual License,
Kylix Will Not be Open Source"
--
Kasper Dupont
------------------------------
From: "peter.katzmann" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup
Subject: Re: Raid A0 patch with Kernel 2.2.17 and ReiserFS
Date: Thu, 17 May 2001 11:50:53 +0200
Reply-To: [EMAIL PROTECTED]
[EMAIL PROTECTED] wrote:
>
> I'll add another question to the thread. 2 of the developers I work with say
> that reiserFS should be used for development because it is more robust then
> ext2 and in the embedded system we are developed, the power frequently is
> removed from the target system running Linux. As a consequence of the fact
> that power can be removed at any time, there is no graceful shutdown of this
> system. I am told by one of the developers that reiserFS will be the new
> default file system in the next release of Linux. I wonder what some of the
> developers here have to say about reiserFS and its reliability in a
> situation where power can and will be removed frequently from an embedded
> system.
Well i would say: Let your developers rethink their design!I think they
refer to the journaled feature of reiser. They are thinking thet they
will never have a chance to get a corrupted filesystems and speed up in
power up.
To say it simple, yes you will have a speed up in filesystem check
No you will have also a chance to get a corrupted filesystem, especially
in an embedded design where power loss doesnt mean complete power loss !
there is anything in between possible
peter
------------------------------
From: Kasper Dupont <[EMAIL PROTECTED]>
Crossposted-To:
alt.linux,comp.os.linux.development.apps,linux.redhat.devel,linux.redhat.development
Subject: Re: transpareny
Date: Thu, 17 May 2001 10:07:30 +0000
Darren LS wrote:
>
[...]
>
> Hi Kasper, thanks but what I would like to do is dump the output to a vacant
> tty terminal for debugging purposes
>
> Darren
In that case the easiest probably is just to
start it with the open command from the
command line. "man 1 open"
Notice that in some distributions open is not
installed by default. It is really a handy
tool, so just install it from your CD if it
is not already installed.
You could also use /dev/ttyXX instead of
/dev/null in the code from my last post, the
problem is to find the appropriate value for
XX. There is an ioctl to get that number, I
can post some code if you need it.
--
Kasper Dupont
------------------------------
** 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 by posting to the
comp.os.linux.development.system newsgroup.
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
******************************
