Linux-Development-Sys Digest #3, Volume #7 Wed, 28 Jul 99 13:13:47 EDT
Contents:
f90 compiling error (Kamran Mohseni)
Re: help on gcc (Peter.vanHelden)
Re: MICROSOFT LINUX DISTRIBUTION (James Stevenson)
Changing source IP of outgoing UDP packets under RedHat 6.0 . ("Yuri G.")
Re: kernel compile (Igor Zlatkovic)
Re: Memory Management Bug ([EMAIL PROTECTED])
Re: Help! Cannot set thread's priority under Linux, even not as root! (Udo
Giacomozzi)
Re: Help! Cannot set thread's priority under Linux, even not as root! (Andreas
Jaeger)
Re: problems with two network cards. Fundamental limitation? (Chetan Ahuja)
Re: Why I can't compile my driver to SMP-Ready? (David Fox)
[Q] Bug with TCP/IP SACK in linux 2.3.3 ? (Julien Godard)
----------------------------------------------------------------------------
From: Kamran Mohseni <[EMAIL PROTECTED]>
Crossposted-To:
comp.lang.fortran,comp.os.linux.misc,comp.os.linux.setup,comp.os.linux.development.apps
Subject: f90 compiling error
Date: Wed, 28 Jul 1999 03:55:34 -0700
Reply-To: [EMAIL PROTECTED]
Hi guys;
I have a red hat linux 6 on my machine.
I recently instaled f90 Version 2.2(260) compiler from NAG.
I am trying to compile a f90 code that works fine on other machines.
It makes the object file initdns.o.
But when it try to link it with the library that I created, i.e.
/home/mohseni/mohseni/Br/FFTLibHome/libgpfa.a
it gets into trouble.
It seems that I have some linking problem. This is the message I get
Kami > make -f Makeinit
f90 -o initdns initdns.o /home/mohseni/mohseni/Br/FFTLibHome/libgpfa.a
/usr/local/lib/f90/libf90.a(open.o): In function `__NAGf90_open':
open.o(.text+0xc77): undefined reference to `_fxstat'
collect2: ld returned 1 exit status
make: *** [initdns] Error 1
Any idea how to fix this.
Please respond to [EMAIL PROTECTED]
Regards.
Kamran.
------------------------------
From: [EMAIL PROTECTED] (Peter.vanHelden)
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.misc
Subject: Re: help on gcc
Date: 28 Jul 1999 11:15:08 GMT
jievis ([EMAIL PROTECTED]) wrote:
: Hi, All:
: When I try to compile a c files, I want gcc output a middle files in
: which the expression is computed and enumeration is also computed, the
: output files should be also .c format. So who can tell me the switcher?
: Thanx in advance
man gcc?
Peter
------------------------------
From: [EMAIL PROTECTED] (James Stevenson)
Subject: Re: MICROSOFT LINUX DISTRIBUTION
Reply-To: [EMAIL PROTECTED]
Date: Wed, 28 Jul 1999 12:20:39 +0100
Hi
yea but knowing microsoft they will put there own kernel patch into it
and make the kernel as big as it can possibly be
and will also run 50% slower :)
cya
On Mon, 26 Jul 1999 23:02:11 +0200, Klamer Schutte <[EMAIL PROTECTED]> wrote:
>Samuel Brown wrote:
>
>> I had a someone tell me that Microsoft will sell their own linux
>> distribution. Is this true?
>
>> It will have word and excel 2000 also.
>
>If that would provide perfect file compatibility with Office'97 then I
>definitely would be interested in that distribution (most likely
>only in these two programs -- or is there also a powerpoint? ;-)
>
>Klamer
>
--
=============================================
Check Out: http://www.users.zetnet.co.uk/james/
E-Mail: [EMAIL PROTECTED]
12:10pm up 37 days, 13:13, 2 users, load average: 0.00, 0.00, 0.00
------------------------------
From: "Yuri G." <[EMAIL PROTECTED]>
Subject: Changing source IP of outgoing UDP packets under RedHat 6.0 .
Date: Wed, 28 Jul 1999 14:50:15 +0300
Hi!
I need to change the source IP address of outgoing packets (not for all,
just for certain destination IPs). If someone know which functions in kernel
source code I need to change to do this, please answer. If you know any
other way to do this - that would be greate if you tell me.
Thanks ahead,
Yuri.
------------------------------
From: Igor Zlatkovic <[EMAIL PROTECTED]>
Subject: Re: kernel compile
Date: Wed, 28 Jul 1999 15:31:28 +0200
Since you said "2.2.5-15", you probably use redhat 6.
If you want to do it the redhat way, then read on.
In addition to steps you described, consider the following:
Look how the existing files in /boot are organized. They all have the
version number encoded in the name of the file, and there is a symlink
without the version number that points to the newest one. This is how you
should place new files there.
Here are the files you should update in /boot:
- /boot/System.map (found in /usr/src/linux/ after kernel compilation)
- /boot/vmlinux (found in /usr/src/linux/ after kernel compilation
- /boot/module-info (or similar). Just make a copy of the existing file and
give it a new name.
If you have modular SCSI support, do these steps also, if not, skip them.
- ensure that your loopback device is loaded (insmod loop)
- create an empty .rhkvmtag file in /lib/modules/<version>/ (man touch)
- create a new module dependency file for the new kernel (man depmod)
- create a new initrd image (man mkinitrd)
Now create a new entry for the new kernel in /etc/lilo.conf, rerun lilo,
reboot and enjoy. After the new kernel boots, do a short cat /proc/version
> /lib/modules/<version>/.rhkvmtag
Everything is clear and nice, but I cannot explain you why did redhat make
this /lib/modules/<version>/.rhkvmtag file. It contains exactly the same as
/proc/version. The /etc/rc.system boot script checks for this file and
creates links to /boot/System.map-<version> and other stuff according to
what it finds there. I never figured out why. Maybe someone else knows? As
of 6.0, this file seems to be less important than before.
By the way, it is generally a bad idea to play around as root.
Ciao
Igor
root wrote:
> I want to upgrade from kernel version 2.2.5-15 to version 2.2.10. Is
> all I have to do is this:
>
> run either make xconfig or menuconfig or config
> then make dep
> make clean
> make bzImage or zImage
> make modules
> make modules_install
> copy the new kernel to the boot
> directory
> run lilo
> reboot ??
>
> Is this correct anyone ??
> are these the steps that it needs to take to compile a kernel with out
> errors and reboot the new kernel without errors ??
>
> I tried it ones like this and I think i got errors when booting the new
> kernel saying "incorrect version or system map"
>
> Oai Luong
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: Memory Management Bug
Date: Wed, 28 Jul 1999 12:46:43 GMT
Hello,
You maybe able to help.When i try to install
either
RH6.0 or SuSe6.1 from the bootdisk with more that
1gig of memory the system will not boot, i get
the error
�VFS: Cannot open root device 08:21
kernel panic : VFS : Unable to mount root fs on
08:21�
I can install fine with 768mb of memory, but when
i try yo add more memory and update the
/etc/lilo.conf. it still will not boot from the
my harddrive or disk.
I have tried this on several different servers
and i can not get around this problem.It is quiet
annoying as i need to be able to run with up to
4gigs of memory.
ANY suggestions
Cheers
Robert Smith
rticle <7meo7g$47r$[EMAIL PROTECTED]>,
Peter Ross <[EMAIL PROTECTED]> wrote:
> Stefan Proels <[EMAIL PROTECTED]>
writes:
>
> >Peter Samuelson wrote:
> >> [Stefan Proels <[EMAIL PROTECTED]>]
> >> > I have no way argued that it's a Bad Thing
to handle allocation this
> >> > way. I just don't think that it's a Good
Thing to enable every
> >> > ordinary user to crash the system.
> >>
> >> You may well have hit a bug then. You
should never be able to actually
> >> crash the system. What version of Linux?
(While I don't have resource
>
> >I'm running a 2.2.7 kernel from a SuSE 6.1
distribution.
>
> >> problems since I'm essentially single-user
here) Linus claims that late
> >> late 2.1.x releases should perform much more
sanely in tight-memory
> >> situations than before -- this was since
maybe 2.1.125 or so. In
> >> particular, Linux 2.0.x was pretty bad at
this.
> >>
> >> Do you have memory overcommit turned on or
off? Makes a difference.
>
> >It's turned off. The program I posted
allocated the memory in small
> >chunks; as I understand the docs memory
overcommit will have no effect
> >to this code because it only check for huge
individual allocations.
> >Anyway, it's turned off.
>
> I am not sure you understand what overcommit is.
>
> Say you have a system with 32Mb of Ram and 12Mb
of swap.
>
> Say at some instant of time 40Mb of memory has
been malloc'd. This
> leaves you with 4Mb of memory available.
>
> If you malloc 5Mb of memory the malloc will
fail with overcommit turned
> off. All that matters is that 5Mb is bigger
then 4Mb, not what the size
> of the requests are, it could easily be 5kb and
4kb.
>
> If overcommit is turned on then the the malloc
will succeed.
> However if you attempt to use all of that 5Mb
of memory then there will
> be problems.
>
> Pete.
> --
> +----------------------------------------------
========================+
> | Peter Ross M Sci/Eng Melbourne Uni
(change - to . for email) |
> | email: [EMAIL PROTECTED] WWW:
http://www.cs.mu.oz.au/~petdr |
> +----------------------------------------------
========================+
>
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: Udo Giacomozzi <[EMAIL PROTECTED]>
Subject: Re: Help! Cannot set thread's priority under Linux, even not as root!
Date: Wed, 28 Jul 1999 13:07:56 GMT
> In order to set the priorities, you have to be executing in real-time
> mode. That's about the extent of my knowledge of the problem
though....
>
Yes, I set the SCHED_FIFO scheduling policy at priority 50. The program
works on another machine (not mine) but it doesn't work on mine. So I
guess my machine is set up inproperly.
Someone told me I'm using too old glibc libraries. I'm using SuSE 5.3
(kernel 2.0.35). Where can I download new libraries?
Thanx
Udo
http://come.to/jampy
Sent via Deja.com http://www.deja.com/
Share what you know. Learn what you don't.
------------------------------
From: Andreas Jaeger <[EMAIL PROTECTED]>
Subject: Re: Help! Cannot set thread's priority under Linux, even not as root!
Date: 28 Jul 1999 15:55:03 +0200
>>>>> Udo Giacomozzi writes:
>> In order to set the priorities, you have to be executing in real-time
>> mode. That's about the extent of my knowledge of the problem
Udo> though....
>>
Udo> Yes, I set the SCHED_FIFO scheduling policy at priority 50. The program
Udo> works on another machine (not mine) but it doesn't work on mine. So I
Udo> guess my machine is set up inproperly.
Udo> Someone told me I'm using too old glibc libraries. I'm using SuSE 5.3
Udo> (kernel 2.0.35). Where can I download new libraries?
SuSE 5.3 comes with libc5. For better threading support, you should
get glibc (and glibc 2.1 has better thread support than glibc 2.0 -
thread support is still evolving).
Updating a libc5 based installation to glibc2 isn't a trivial task.
I'd advise to use a glibc based installation (all current ones are
glibc 2.0 or 2.1 based):
- SuSE 6.1 (glibc 2.0 based), SuSE 6.2 (glibc 2.1 based, should be out
in a week/fortnight?)
- RedHat 5.2 (glibc 2.0 based), RedHat 6.0 (glibc 2.1 based)
- Debian, Mandrake,...
Andreas
--
Andreas Jaeger [EMAIL PROTECTED] [EMAIL PROTECTED]
for pgp-key finger [EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Chetan Ahuja)
Subject: Re: problems with two network cards. Fundamental limitation?
Date: 28 Jul 1999 16:11:06 GMT
Tony Gale ([EMAIL PROTECTED]) wrote:
: In article <7nmcu2$l3d$[EMAIL PROTECTED]>,
: [EMAIL PROTECTED] (Chetan Ahuja) writes:
: > I have made a router out of a pentium 60 using the LRP (Linux Router
: > Project based on 2.0.36 modified a little) distro. It has two network
: > cards ( One Tulip based and the other via-rhine based) One talks to my
: > internal network and the other one talks to the external world. Now,
: > the external connection ( through xDSL) goes dead sometimes due to
: > problems beyond my control. The problem is, when that happens, I can't
: > even telnet into the box through it's internal interface. I can still
: > ping the box though. But no telnet. If a kernel hacker is reading
: > this, could you tell me where exactly is the problem. Is it a hardware
: > driver level problem, TCP/IP stack problem, or even a tcp-wrapper
: > problem ( all servers on the router box are invoked through inetd). My
: > guess is that the external connection is blocking on a read/write and
: > that somehow blocks the internal connection too.. ( or is it supposed
: > to be obvious and I am just exposing the limits of my knowledge here)
: > Is there a fix for it...(or has it been taken care in later kernels in
: > 2.2 series ) or is it a fundamental limitation of the kernel
: > architecture ??? If you can point me to a source file in the kernel tree,
: > or other sources of information, that will be good too...
: >
: Is, by any chance, your DNS server on the side that goes down?
: -tony
: --
Yes. But that shouldn't affect anything. I am telnetting directly
to the IP of the internal interface and hence no IP resolution is necessary.
And as I said, I can ping that (internal) IP alright.... just no telnet or
any other deamon connection.
Chetan
--
------------------------------
From: d s f o x @ c o g s c i . u c s d . e d u (David Fox)
Crossposted-To: comp.os.linux.development.apps,comp.os.linux.hardware
Subject: Re: Why I can't compile my driver to SMP-Ready?
Date: 28 Jul 1999 08:54:00 -0700
The header file package "kernel-headers" provided with Redhat 6.0
include only the non-SMP configuration variables and symbol names in
files like /usr/include/linux/autoconf.h and
/usr/include/linux/modules/*.ver. The easiest way to fix this is
to rebuild the kernel from the source RPM after modifying the spec
file to build the SMP kernel last:
%ifnarch i586 i686
# BOOT kernel
BuildKernel BOOT
%ifarch i386
BuildPCMCIA BOOT
%endif
%endif
# NORMAL KERNEL
BuildKernel
%ifarch i386 i586 i686
BuildiBCS no
BuildPCMCIA
%endif
#SMP-ENABLED KERNEL
BuildKernel smp
%ifarch i386 i586 i686
BuildiBCS yes smp
# pcmcia does not compile on the smp kernels
#BuildPCMCIA smp
%endif
I've mentioned this problem to redhat, and even done work on a kernel
rpm that creates a kernel-headers-smp subpackage, but at the moment it
doesn't set up the pcmcia stuff correctly.
If you're not using redhat I must have misinterpreted the problem.
"robert_c" <[EMAIL PROTECTED]> writes:
> Hello:
>
> Could someone tell me why I can't compile my UP-Ready Driver into SMP-Ready
> Driver. The following is my operations steps. Could you give me some advice?
> Thanks.
>
> 1. in the /usr/src/linux
> issue command: "make config"( just answer "Y" in support "CONFIG_SMP"
> option, others keep no change)
> 2. in the same directory
> issue command: "make dep"
> 3. go to my driver directory <-- Question 1: Is it needed to put our driver
> code into /usr/src/linux/drivers/misc? I think it is not necessary, isn't
> it?
> cd /robert/
> 4. compile our driver code
> gcc -D__SMP__ -D__KERNEL__ -DMODULE -DMODVERSIONS -O -w -c MyDriver.c
> <---- Question 2: Does it need another option? (and in my MyDriver.h, I have
> included <linux/modversions.h>)
> 5. Compile OK!
>
> 6. insmod in current directory, not in /lib/modules/2.2.5-15smp/misc <----
> Question 3: Does it be necessary to move to the certain directory?
> insmod -f MyDriver.o
>
> 7. It told me some error like following:
> Warning: kernel-module version mismatch
> MyDriver.o was compiled for kernel version 2.2.5-15<------ The same
> error like before
> while this kernel is version 2.2.5-15smp
>
> MyDriver.o: unresolved symbol schedule_timeout_R17d59d01
> MyDriver.o: unresolved symbol release_region_R43bde9b1
> MyDriver.o: unresolved symbol boot_cpu_data_R660bd466
> MyDriver.o: unresolved symbol kmalloc_R93d4cfe6
> MyDriver.o: unresolved symbol unregister_chrdev_Rc192d491
> MyDriver.o: unresolved symbol __verify_write_R203afbeb
> MyDriver.o: unresolved symbol register_chrdev_R0f33bd81
> MyDriver.o: unresolved symbol add_timer_Rbea990b2
> MyDriver.o: unresolved symbol jiffies_R0da02d67
> MyDriver.o: unresolved symbol request_region_R6d32b2d7
> MyDriver.o: unresolved symbol check_region_R522f4d72
> MyDriver.o: unresolved symbol kfree_R037a0cba
> MyDriver.o: unresolved symbol memcpy_fromfs
> MyDriver.o: unresolved symbol printk_R1b7d4074
> .....
>
> And I check my kernel supported function like schedule_timeout, issue
> command like following:
>
> [root@robert MyDriver]# ksyms -a|grep schedule_timeout
> c011215c schedule_timeout_Rsmp_17d59d01
> ^^^
> I got the reason why I can't insmod my SMP-driver into SMP-Kernel, because
> my driver still is UP-Driver. But how to change it to SMP-Ready Driver?
>
> 8. Sorry to bother you again. Could you give me a example driver (just say
> "Hello"), and a easy Makefile, and some operation steps. Because I always
> get failed for this problem.
>
> Appreciate for any your advices.
>
> Best Regards,
>
> Robert
>
>
>
--
David Fox http://hci.ucsd.edu/dsf xoF divaD
UCSD HCI Lab baL ICH DSCU
------------------------------
From: Julien Godard <[EMAIL PROTECTED]>
Subject: [Q] Bug with TCP/IP SACK in linux 2.3.3 ?
Date: Wed, 28 Jul 1999 17:33:49 +0100
Reply-To: [EMAIL PROTECTED]
====================================================
PLEASE CC YOUR REPLY TO MY PERSONAL ADDRESS :
[EMAIL PROTECTED]
====================================================
I am doing some test for TCP/IP with a large delay.
I use a 2Mbps link with 565 ms RTT.
In this test, I want to see what happend if I lose an IP datagram.
The TCP connection is fluent, receiver windows limited, and then I
generate an error, and lose an IP datagram (ONLY one). SACK seems to
work normally until 17 segments are received. Let me explain that with a
tcpdump output (I remove the timestamp and replace the computers name by
A and B) :
I use delayed Acks, large windows (scale factor 2), socket buffer size
147000 bytes on both sides
13:29:24.767481 A > B: . ack 3084064454 win 34390 < > (DF)
13:29:24.772009 B > A: P 3084118030:3084119478(1448) ack 2866015587 win
36562 < > (DF)
13:29:24.772026 B > A: P 3084119478:3084120926(1448) ack 2866015587 win
36562 < > (DF)
13:29:24.777481 A > B: . ack 3084067350 win 34390 < > (DF)
13:29:24.786096 B > A: P 3084120926:3084122374(1448) ack 2866015587 win
36562 < > (DF)
13:29:24.786113 B > A: P 3084122374:3084123822(1448) ack 2866015587 win
36562 < > (DF)
13:29:24.797481 A > B: . ack 3084070246 win 34390 < > (DF)
until here, all is normal...
13:29:24.800214 B > A: P 3084123822:3084125270(1448) ack 2866015587 win
36562 < > (DF)
13:29:24.800231 B > A: P 3084125270:3084126718(1448) ack 2866015587 win
36562 < > (DF)
13:29:24.807481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084073142> (DF)
duplicate ack, with sack : one datagram is missing but 1 datagram
received.
13:29:24.814293 B > A: P 3084126718:3084128166(1448) ack 2866015587 win
36562 < > (DF)
13:29:24.814311 B > A: P 3084128166:3084129614(1448) ack 2866015587 win
36562 < > (DF)
13:29:24.817481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084074590> (DF)
duplicate ack, with sack : one datagram is missing but 2 datagrams
received.
13:29:24.821351 B > A: P 3084129614:3084131062(1448) ack 2866015587 win
36562 < > (DF)
13:29:24.827481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084076038> (DF)
duplicate ack, with sack : one datagram is missing but 3 datagrams
received.
and so on...
13:29:24.828388 B > A: P 3084070246:3084071694(1448) ack 2866015587 win
36562 < > (DF)
13:29:24.827481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084077486> (DF)
13:29:24.837481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084078934> (DF)
13:29:24.867481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084080382> (DF)
13:29:24.997481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084081830> (DF)
13:29:25.007481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084083278> (DF)
13:29:25.017481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084084726> (DF)
13:29:25.017481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084086174> (DF)
13:29:25.027481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084087622> (DF)
13:29:25.037481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084089070> (DF)
13:29:25.037481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084090518> (DF)
13:29:25.047481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084091966> (DF)
13:29:25.057481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084093414> (DF)
13:29:25.067481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084094862> (DF)
duplicate ack, with sack : one datagram is missing but 16 datagrams
received.
13:29:25.067481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084071694 3084096310> (DF)
duplicate ack, with sack : one datagram is missing but 17 datagrams
received.
13:29:25.077481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084096310 3084097758> (DF)
HERE IS THE PROBLEM :
duplicate ack, with sack : 18 datagrams are missing but 1 datagram
received.
FYI
13:29:25.237481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084096310 3084099206> (DF)
13:29:25.247481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084096310 3084100654> (DF)
13:29:25.257481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084096310 3084102102> (DF)
13:29:25.260678 B > A: P 3084131062:3084132510(1448) ack 2866015587 win
36562 < > (DF)
13:29:25.267481 A > B: . ack 3084070246 win 34390 < nop,nop,sack
3084096310 3084103550> (DF)
13:29:25.267751 B > A: P 3084132510:3084133958(1448) ack 2866015587 win
36562 < > (DF)
This behaviour is periodic, every 17 SACK...The sender go down to slow
start and then recover, but instead of 1 retransmission, I have 86
retransmissions...
I try the same test with an other size of datagram (9128 bytes), and all
is fine (but I experienced only 14 SACKs)
Something seems to be wrong with SACK in linux 2.3.3 .Of course, it can
only be a limited buffer size (out_of_order_queue for example) or a too
small default value for this buffer, but can I have your opinion on that
(especially if you have a better understanding of C than me, and it's
not very difficult ;-)
Thank you to read all this message !
Julien
PS: FYI the SACK implementation of tcpdump 3.4 is RFC 1072 compliant,
not RFC 2018...
------------------------------
** 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
******************************
