Re: Jessie, problem compiling kernel 4.3

[Curt Howland]

I spoke too soon, someone in the ubuntu forums had the same problem,
installed libssl-dev and it worked.

Sorry to trouble y'all.

Curt-

On Fri, Nov 6, 2015 at 3:06 PM, Curt Howland  wrote:
> So, there I was, doing a compile of the new kernel, 4.3
>
> I get the following interesting error:
>
> ===
> scripts/extract-cert.c:21:25: fatal error: openssl/bio.h: No such file
> or directory
>  #include 
>  ^
> compilation terminated.
> ===
>
> I'm accustomed to getting compile time errors that "You have no
> certs!", this is the first time this has failed. (4.2.5 compiled just
> fine, for example)
>
> Searching has turned nothing up, am I the only person having this problem?
>
> Curt-
>
>
> --
> The secret of happiness is freedom,
> and the secret of freedom is courage.
> - Thucydides



-- 
The secret of happiness is freedom,
and the secret of freedom is courage.
- Thucydides

Jessie, problem compiling kernel 4.3

[Curt Howland]

So, there I was, doing a compile of the new kernel, 4.3

I get the following interesting error:

===
scripts/extract-cert.c:21:25: fatal error: openssl/bio.h: No such file
or directory
 #include 
 ^
compilation terminated.
===

I'm accustomed to getting compile time errors that "You have no
certs!", this is the first time this has failed. (4.2.5 compiled just
fine, for example)

Searching has turned nothing up, am I the only person having this problem?

Curt-


-- 
The secret of happiness is freedom,
and the secret of freedom is courage.
- Thucydides

Re: problem compiling kernel

[Amax]

On Tue, 25 Aug 2009 00:50:10 +0200, Bernard wrote:

- snip --

 The initrd.img that I have on my working system, as well as those
 initrd.img that 'mkinitrd' generates when requested, are not compressed
 files. Filenames are : initrd.img-2.6.20-16-386 for instance. No .gz
 behind. I still tried to gunzip one, just in case that would still be a
 compressed file without usual extension, but no, it is not handled by
 gunzip or zcat. I tried cpio on that file as is, but I got : 'cpio:
 premature end of file'. a 'vi filename' shows that this is a binary
 file. No point to edit then.
 
 So, at this point, I don't have a clue of how to build an initrd.img
 file that would allow my newly compiled 2.6.30.4 kernel to boot on my
 system.
 
 then edit init to match my needs i.e. depmod, modprobe, cryptsetup etc
 and finally put a line to run the real init. I then zip it
find . ! -name *~ | cpio -H newc --create | gzip -9 
../test-initrd.gz
 I can install then the new initrd (cp ../test-initrd.gz
 /boot/initrdgz)

Once you've done it it's very simple and easy ... before it was a big
trouble for me too.

Just look positive as way to learn something new about your operating
system.

reagrds




Sorry if I'm misunderstanding, but are you saying that although you've 
included 'make --initrd . . ' in your compile command sequence, you are 
still not getting an appropriate initrd.img?

This appears to be some kind of bug in the 2.6.30 kernel  I've managed 
to solve this by using the update-initramfs -c -k linux-2.6.30-x-custom 
command.  After doing this, check your /boot to see if there is, in fact, 
a new 'initrd.img-2.6.30-x-custom' file.  If there is, update your grub 
or lilo  reboot.

I've compiled 5 different versions of the 2.6.30-x kernel  had to do 
this in each one.

The developers know about this but so far haven't done squat about it - 
maybe they consider it a low priority or something.

Hope this helps . . .


   ~A~

-- 
A person needs only two tools: WD-40 and duct tape. If it doesn't move
and it should, use WD-40. If it moves and shouldn't, use the tape.
 -- Red Green

Registered Linux User No. 306834


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: problem compiling kernel

[Bernard]

Amax wrote:

On Tue, 25 Aug 2009 00:50:10 +0200, Bernard wrote:

- snip --

  

The initrd.img that I have on my working system, as well as those
initrd.img that 'mkinitrd' generates when requested, are not compressed
files. Filenames are : initrd.img-2.6.20-16-386 for instance. No .gz
behind. I still tried to gunzip one, just in case that would still be a
compressed file without usual extension, but no, it is not handled by
gunzip or zcat. I tried cpio on that file as is, but I got : 'cpio:
premature end of file'. a 'vi filename' shows that this is a binary
file. No point to edit then.

So, at this point, I don't have a clue of how to build an initrd.img
file that would allow my newly compiled 2.6.30.4 kernel to boot on my
system.



then edit init to match my needs i.e. depmod, modprobe, cryptsetup etc
and finally put a line to run the real init. I then zip it
   find . ! -name *~ | cpio -H newc --create | gzip -9 
   ../test-initrd.gz
I can install then the new initrd (cp ../test-initrd.gz
/boot/initrdgz)

Once you've done it it's very simple and easy ... before it was a big
trouble for me too.

Just look positive as way to learn something new about your operating
system.

reagrds




  
Sorry if I'm misunderstanding, but are you saying that although you've 
included 'make --initrd . . ' in your compile command sequence, you are 
still not getting an appropriate initrd.img?


This appears to be some kind of bug in the 2.6.30 kernel  I've managed 
to solve this by using the update-initramfs -c -k linux-2.6.30-x-custom 
command.  After doing this, check your /boot to see if there is, in fact, 
a new 'initrd.img-2.6.30-x-custom' file.  If there is, update your grub 
or lilo  reboot.


I've compiled 5 different versions of the 2.6.30-x kernel  had to do 
this in each one.


The developers know about this but so far haven't done squat about it - 
maybe they consider it a low priority or something.


Hope this helps . . .


   ~A~

  
Thanks for your input. Ever since I wrote about those late problems, 
things have changed a lot here. I got convinced that I would go no 
further with my old Sarge distro : old kernels could no longer get 
compiled with newer tools, and, as far as building newer kernels, the 
initrd image that I got using mkinitrd did not fit my system and 
wouldn't allow boot. So, I decided to get a more recent distro. Since 
there was 'Etch' in between Sarge and Lenny, I figured that I could not 
just upgrade, maybe I was wrong, in any case, I just saved whatever had 
to be kept, and I installed Lenny from scratch. This was about 6 days 
ago. On my new Lenny, I was able to recompile my kernel, but it has not 
been so easy. I could not get the official process to work here, using 
make-kpkg with --initrd. It did compile indeed, and gave a kernel image 
and an initrd image, but that did not boot either. However, using plain 
old 'make', then make_install, then make_modules_install, then, not 
mkinitrd, but initramfs, I got what I needed, and my new kernel boots 
all right.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: problem compiling kernel

[Emanoil Kotsev]

Bernard wrote:


You really could use the recent 2.6.30.4. There were different problems
with 2.6.20 to 2.6.30. I find 2.6.30.4 the best I've had since 2.6.20.
  

 I tried 2.6.30.4. Same result as with 2.6.26.2 : compiles without
 errors, but crashes on boot.
 

so you are missing some essential part of it

So, what I would do (if I were you) is that I would download latest
2.6.30.4, and compile all I need to access my boot partition (as you
already did with md in the kernel),

 
 I just did that again

You sure you picked up _all_ you need to boot?

 
then compile and rebuild or build by
hand initramfs.
Build by hand I pretty simple- it's actually hacking the one used. I do
unzip it
cd /tmp; mkdir test; cd test
zcat /boot/initrdgz | cpio -Hnewc -i
  

 The initrd.img that I have on my working system, as well as those
 initrd.img that 'mkinitrd' generates when requested, are not compressed
 files. Filenames are : initrd.img-2.6.20-16-386 for instance. No .gz
 behind. I still tried to gunzip one, just in case that would still be a
 compressed file without usual extension, but no, it is not handled by
 gunzip or zcat. I tried cpio on that file as is, but I got : 'cpio:
 premature end of file'. a 'vi filename' shows that this is a binary
 file. No point to edit then.

Did you try bzip?

 
 So, at this point, I don't have a clue of how to build an initrd.img
 file that would allow my newly compiled 2.6.30.4 kernel to boot on my
 system.

You know pretty much already. You can do it yourself.

What about using a already working image like I've suggested.
You can take i.e. ubuntu knoppix or debian live - boot with it and copy the
relevant parts from

/boot/ and /lib/modules to your hard drive

I now had also another idea. Is your boot partition may be full, so initrd
can not be written completely when generated? it happens more often then
you can imagine

regards


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: problem compiling kernel

[Bernard]

Emanoil Kotsev wrote:


Bernard wrote:

 


Emanoil Kotsev wrote:

   


Bernard wrote:




Compiling md in the kernel is the right approach to boot from raided root
without initrd. You can try this just skipping (deleteing the line in grub
temporary)


 


I just tried that. Raid compiled into the kernel instead of modules. No
initrd. Still crashes at boot.
   



most probably you are missing other modules (like ide/ata lvm etc)
You said your boot is on md but not on lvm. you can build a working initrd
easily - this is actually all you need.

 


Also done another test:
in the /boot/grub/menu.lst file, replaced root=/dev/mapper/vg00-root by
/dev/sda2. Still crashed : cannot open root device 'sda2' or unknown
block(0,0).
   



this can not work as your root is on lvm. what did you expect?

try passing the kernel option init=/bin/sh

 


There is another test that I would like to run, but I need help for
this, since I don't know the whole package list:

apt-get purge kernel-building gcc make kernel-utils etc...

then edit my /etc/apt/sources.list and comment out lines that refer to
package directories that are too recent, uncomment old lines referring
to debian sarge packages only, excluding 'testing' etc..

then

apt-get install kernel-building gcc make kernel-utils etc...

and, from there on, trying to recompile, not newer kernels, but my good
old running kernel 2.6.20-16-386 into a custom version without any sound
options in it.

What I need is the list of all packages that I should purge and
re-install in their former version.

regards
   



You really could use the recent 2.6.30.4. There were different problems with
2.6.20 to 2.6.30. I find 2.6.30.4 the best I've had since 2.6.20.
 

I tried 2.6.30.4. Same result as with 2.6.26.2 : compiles without 
errors, but crashes on boot.



So, what I would do (if I were you) is that I would download latest
2.6.30.4, and compile all I need to access my boot partition (as you
already did with md in the kernel), 



I just did that again


then compile and rebuild or build by
hand initramfs.
Build by hand I pretty simple- it's actually hacking the one used. I do
unzip it
   cd /tmp; mkdir test; cd test
   zcat /boot/initrdgz | cpio -Hnewc -i
 

The initrd.img that I have on my working system, as well as those 
initrd.img that 'mkinitrd' generates when requested, are not compressed 
files. Filenames are : initrd.img-2.6.20-16-386 for instance. No .gz 
behind. I still tried to gunzip one, just in case that would still be a 
compressed file without usual extension, but no, it is not handled by 
gunzip or zcat. I tried cpio on that file as is, but I got : 'cpio: 
premature end of file'. a 'vi filename' shows that this is a binary 
file. No point to edit then.


So, at this point, I don't have a clue of how to build an initrd.img 
file that would allow my newly compiled 2.6.30.4 kernel to boot on my 
system.



then edit init to match my needs i.e. depmod, modprobe, cryptsetup etc
and finally put a line to run the real init. I then zip it
   find . ! -name *~ | cpio -H newc --create | gzip -9  ../test-initrd.gz
I can install then the new initrd (cp ../test-initrd.gz /boot/initrdgz)

Once you've done it it's very simple and easy ... before it was a big
trouble for me too.

Just look positive as way to learn something new about your operating
system.

reagrds


 




--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: problem compiling kernel

[Bernard]

Emanoil Kotsev wrote:


Bernard wrote:

 



Compiling md in the kernel is the right approach to boot from raided root
without initrd. You can try this just skipping (deleteing the line in grub
temporary)
 

I just tried that. Raid compiled into the kernel instead of modules. No 
initrd. Still crashes at boot.


Also done another test:
in the /boot/grub/menu.lst file, replaced root=/dev/mapper/vg00-root by 
/dev/sda2. Still crashed : cannot open root device 'sda2' or unknown 
block(0,0).


There is another test that I would like to run, but I need help for 
this, since I don't know the whole package list:


apt-get purge kernel-building gcc make kernel-utils etc...

then edit my /etc/apt/sources.list and comment out lines that refer to 
package directories that are too recent, uncomment old lines referring 
to debian sarge packages only, excluding 'testing' etc..


then

apt-get install kernel-building gcc make kernel-utils etc...

and, from there on, trying to recompile, not newer kernels, but my good 
old running kernel 2.6.20-16-386 into a custom version without any sound 
options in it.


What I need is the list of all packages that I should purge and 
re-install in their former version.


regards


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: problem compiling kernel

[Emanoil Kotsev]

Bernard wrote:

 Emanoil Kotsev wrote:
 
Bernard wrote:

  


Compiling md in the kernel is the right approach to boot from raided root
without initrd. You can try this just skipping (deleteing the line in grub
temporary)
  

 I just tried that. Raid compiled into the kernel instead of modules. No
 initrd. Still crashes at boot.

most probably you are missing other modules (like ide/ata lvm etc)
You said your boot is on md but not on lvm. you can build a working initrd
easily - this is actually all you need.

 
 Also done another test:
 in the /boot/grub/menu.lst file, replaced root=/dev/mapper/vg00-root by
 /dev/sda2. Still crashed : cannot open root device 'sda2' or unknown
 block(0,0).

this can not work as your root is on lvm. what did you expect?

try passing the kernel option init=/bin/sh

 
 There is another test that I would like to run, but I need help for
 this, since I don't know the whole package list:
 
 apt-get purge kernel-building gcc make kernel-utils etc...
 
 then edit my /etc/apt/sources.list and comment out lines that refer to
 package directories that are too recent, uncomment old lines referring
 to debian sarge packages only, excluding 'testing' etc..
 
 then
 
 apt-get install kernel-building gcc make kernel-utils etc...
 
 and, from there on, trying to recompile, not newer kernels, but my good
 old running kernel 2.6.20-16-386 into a custom version without any sound
 options in it.
 
 What I need is the list of all packages that I should purge and
 re-install in their former version.
 
 regards

You really could use the recent 2.6.30.4. There were different problems with
2.6.20 to 2.6.30. I find 2.6.30.4 the best I've had since 2.6.20.

I was also very sad when I found out I can not compile 2.6.20 anymore. Put
let's believe it's for the sake of the progress.

So, what I would do (if I were you) is that I would download latest
2.6.30.4, and compile all I need to access my boot partition (as you
already did with md in the kernel), then compile and rebuild or build by
hand initramfs.
Build by hand I pretty simple- it's actually hacking the one used. I do
unzip it
cd /tmp; mkdir test; cd test
zcat /boot/initrdgz | cpio -Hnewc -i
 then edit init to match my needs i.e. depmod, modprobe, cryptsetup etc
 and finally put a line to run the real init. I then zip it
find . ! -name *~ | cpio -H newc --create | gzip -9  ../test-initrd.gz
 I can install then the new initrd (cp ../test-initrd.gz /boot/initrdgz)

Once you've done it it's very simple and easy ... before it was a big
trouble for me too.

Just look positive as way to learn something new about your operating
system.

reagrds


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: problem compiling kernel

[Bernard]

Emanoil Kotsev wrote:


Emanoil Kotsev wrote:

 


Sorry forgot to write

Yes there is problem compiling the 2.6.20 with recent gcc

The problem is the compiler. If you are compiling just grab the last
version from kernel.org.

2.6.30.4 seems to be working fine

   



Just to be objective the gnu compiler people said kernel people are wrong
and vice versa. I didn't follow the discussion. I'm glad next kernels
compile

regards




 



2.6.30.4 does compile all right, so does 2.6.26, but 2.6.20 does not. Problem is that I still can't boot those I compiled, i.e. 2.6.26. because the initrd.img is buggy. I did find something, still it is not enough to get the process to work. In my 


/etc/mkinitrd/mkinitrd.conf file, I found a modified line :

# Command to generate the initrd image.
# MKIMAGE='mkcramfs %s %s  /dev/null' this has been changed august 19, 2009

MKIMAGE='genromfs -d %s -f %s'

The change date has not been written by me, so this must be a conf file that 
came with a recent package upgrade that I did. I tried uncommenting the old 
line, commenting the new one instead. MKIMAGE='mkcramfs... became active. What 
gave me this idea, is that in those error messages that I could see at crash, 
it was matter of cramfs.

Well that change made mkinitrd to produce smaller images. I tried installing 
them in the grub boot menu, and then, now, the boot crashes do not happen at 
the same time as before... but it still crashes !

I could do nothing else than catch photos of my screen, since no log file are 
recorded in such cases.

http://www.teaser.fr/~bdebreil/bootcrash1.jpg

and

http://www.teaser.fr/~bdebreil/bootcrash2.jpg

will show you what I got

The first crash screen is not very informative :

could not load '/lib/modules... no such files

(these files exist, but at this point in time it is not in the /boot partition, 
therefore not mounted as yet). This crash came from a kernel which I had 
configure to have RAID inside, not as modules. While watching the boot logs of 
my working kernel, I could see that RAID was as modules. So, I recompiled a new 
kernel with modules for RAID, and then boot went a little bit further, as can 
be seen in the screen picture at crash :

'raid1 set md1 active with 2 out of 2 mirrors

mdadm : /dev/md1 has been started with 2 drives'

but then :

'mkdir cannot create directory '/devfs/vg00' : read only filesystem
..

failure to communicate to kernel device mapper driver
incompatible libdevmapper 1.01.00-ioctl (2005-01-17) (compat) and kernel driver'

I think that this last quoted line does most explain that the tools I am using 
are not appropriate.

I have good grounds to think that the problems are in my initrd.img file... but 
there may also be something wrong in the compiled kernel image.

Could someone please tell me what tool packages to purge and what to install 
instead so that I can recompile a 2.6.26 or 2.6.30 kernel that will boot on my 
Debian 3.1 system with raid 1 ?

Thanks in advance for your help


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: problem compiling kernel

[Emanoil Kotsev]

Bernard wrote:

 
 2.6.30.4 does compile all right, so does 2.6.26, but 2.6.20 does not.

you find out why in the archives

 Problem is that I still can't boot those I compiled, i.e. 2.6.26. because
 the initrd.img is buggy. I did find something, still it is not enough to
 get the process to work. In my
 
 /etc/mkinitrd/mkinitrd.conf file, I found a modified line :
 
 # Command to generate the initrd image.
 # MKIMAGE='mkcramfs %s %s  /dev/null' this has been changed august 19,
 # 2009
 
 MKIMAGE='genromfs -d %s -f %s'
 
 The change date has not been written by me, so this must be a conf file
 that came with a recent package upgrade that I did. I tried uncommenting
 the old line, commenting the new one instead. MKIMAGE='mkcramfs... became
 active. What gave me this idea, is that in those error messages that I
 could see at crash, it was matter of cramfs.
 
 Well that change made mkinitrd to produce smaller images. I tried
 installing them in the grub boot menu, and then, now, the boot crashes do
 not happen at the same time as before... but it still crashes !
 
 I could do nothing else than catch photos of my screen, since no log file
 are recorded in such cases.
 
 http://www.teaser.fr/~bdebreil/bootcrash1.jpg

so this is the old error, and you don't need a fix for it

 
 and
 
 http://www.teaser.fr/~bdebreil/bootcrash2.jpg
 
 will show you what I got
 
 The first crash screen is not very informative :
 
 could not load '/lib/modules... no such files

I don't think so it's as informative as it should be. It can not mount sdb2
(is it your root?)

 
 (these files exist, but at this point in time it is not in the /boot
 partition, therefore not mounted as yet). This crash came from a kernel
 which I had configure to have RAID inside, not as modules. While watching
 the boot logs of my working kernel, I could see that RAID was as modules.
 So, I recompiled a new kernel with modules for RAID, and then boot went a
 little bit further, as can be seen in the screen picture at crash :
 
 'raid1 set md1 active with 2 out of 2 mirrors
 
 mdadm : /dev/md1 has been started with 2 drives'
 
 but then :
 
 'mkdir cannot create directory '/devfs/vg00' : read only filesystem
 ..
 
 failure to communicate to kernel device mapper driver
 incompatible libdevmapper 1.01.00-ioctl (2005-01-17) (compat) and kernel
 driver'
 

The problem is as far as I remember that devfs was given up ... was it
something that worked with hotplug ... I really don't remember right now,
but there was a change affecting devmapper. I think you have to read about
it, perhaps replace it and recreate initrd.

Compiling md in the kernel is the right approach to boot from raided root
without initrd. You can try this just skipping (deleteing the line in grub
temporary)

 I think that this last quoted line does most explain that the tools I am
 using are not appropriate.
 
 I have good grounds to think that the problems are in my initrd.img
 file... but there may also be something wrong in the compiled kernel
 image.

try without initrd (with custom kernel, you can put everything you need
inside it (i.e filesystem support ide/ata etc) you then can access your
root partition and the boot process will continue from there. The initrd is
only needed to load drivers which helps you do the above.
Because you are using lvm, if not using initrd you need to compile also lvm
inside the kernel.

 
 Could someone please tell me what tool packages to purge and what to
 install instead so that I can recompile a 2.6.26 or 2.6.30 kernel that
 will boot on my Debian 3.1 system with raid 1 ?
 
There are good howtos for upgrading from sarg - etch and etch - lenny.
You definitely better use udev ... devmapper is not needed anymore as far as
I know.

I did it last year ... and yes there were some troubles with the initrds ...

I could send you my scripts for building your own initrd ( I have used them
to build initrd for crypted root - before it started working in debian),
though I've already posted a 5step howto fix broken boot initrd - try
init=/bin/sh option ;-) and fix the boot by hand - you'll see what you are
missing

You could just copy over a working image and initrd (from some live cd/dvd)
edit grub and reboot - this should work.

regards


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

problem compiling kernel

[Bernard]

Hi to Everyone,

I need to re-compile my kernel so that it does not include sound support 
inside. I am running Debian 3.1 (Sarge). My system is on RAID1. My /boot 
partition is from /dev/sda1 (mirror on /dev/sdb1) installed on /dev/md0 
(ext3), while my '/' partition is from /dev/sda2 and /dev/sdb2 installed 
on /dev/md1 mounted on /dev/mapper/vg00-root (LVM2). It's been working 
OK for several years.


The easiest way would be to try re-compiling from 
/boot/config-2.6.20-16-386 using make oldconfig. But this repeatedly 
fails. 'make' soon returns error messages saying that such and such 
function have not been declared. Since the process works with more 
recent kernels, I suppose that the Makefile that I have in 
/usr/src/linux-2.6.20-16-386 is buggy, or else, maybe the version of 
gcc/make that I get is no longer compatible with said Makefile : I must 
admit that I have sometimes ran 'apt-get install' on various packages 
with a /etc/apt/sources.list that contained lines referring to testing 
directories.


So, upon my failures to recompile kernel 2.6.20-16-386, I tried 
downloading 2.6.20-17-386, but I got the same results. However, with 
2.6.26.2, it did compile without errors... but in the end the image 
won't boot !


My compiling process does not generate initrd.img, so I did generate one 
using 'mkinitrd', and I wrote its path in /boot/grub/menu.lst with the 
kernel image.


Here is what I get on booting trials :

boot starts, as usual text displays very fast and you cannot read until 
it stops. When this happen, I can read this :



md: raid0 personality registered for level 0
md: raid1 personality registered for level 1
device mapper
.
no filesystem could mount root. Tried: cramfs
kernel panic - not syncing: VFS unable to mount root fs on unknown 
block(0,0)



So, I re-tried compiling after de-activating raid0 in the config, 
leaving only raid1... to the same end result.


Could someone tell me what options I should select in the config so as 
to obtain a new kernel that will boot my RAID architecture ?


I know that there are other ways to compile Debian kernels, using 
'make-kpkg', but this does not work here, likely because this tool 
version is too old or too new for my system : it keeps saying that 
parameters are missing, while the howtos that I saw did not mention any 
such parameters for make-kpkg.


By the way, I noticed that the initrd.img that I obtain using mkinitrd 
are about 5 times bigger than that which is used to boot my usual 
kernel. I am not sure that I am using mkinitrd correctly.


mkinitrd -o initrd.img-2.6.26.2 2.6.26 done from /boot issues an error 
messages saying that this image will not take raid into account, and 
that therefore it won't boot, unless the main image contains what is 
missing here. In any case, it doesn't boot. The same command issued in 
/usr/src/linux-2.6.26.2 does not send any error message, but in the end 
the image doesn't boot either.


Thanks in advance for your help


--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: problem compiling kernel

[Emanoil Kotsev]

Bernard wrote:

 Hi to Everyone,
 
 I need to re-compile my kernel so that it does not include sound support
 inside. I am running Debian 3.1 (Sarge). My system is on RAID1. My /boot
 partition is from /dev/sda1 (mirror on /dev/sdb1) installed on /dev/md0
 (ext3), while my '/' partition is from /dev/sda2 and /dev/sdb2 installed
 on /dev/md1 mounted on /dev/mapper/vg00-root (LVM2). It's been working
 OK for several years.
 
 The easiest way would be to try re-compiling from
 /boot/config-2.6.20-16-386 using make oldconfig. But this repeatedly
 fails. 'make' soon returns error messages saying that such and such
 function have not been declared. Since the process works with more
 recent kernels, I suppose that the Makefile that I have in
 /usr/src/linux-2.6.20-16-386 is buggy, or else, maybe the version of
 gcc/make that I get is no longer compatible with said Makefile : I must
 admit that I have sometimes ran 'apt-get install' on various packages
 with a /etc/apt/sources.list that contained lines referring to testing
 directories.
 
 So, upon my failures to recompile kernel 2.6.20-16-386, I tried
 downloading 2.6.20-17-386, but I got the same results. However, with
 2.6.26.2, it did compile without errors... but in the end the image
 won't boot !
 
 My compiling process does not generate initrd.img, so I did generate one
 using 'mkinitrd', and I wrote its path in /boot/grub/menu.lst with the
 kernel image.
 
 Here is what I get on booting trials :
 
 boot starts, as usual text displays very fast and you cannot read until
 it stops. When this happen, I can read this :
 
 
 md: raid0 personality registered for level 0
 md: raid1 personality registered for level 1
 device mapper
 .
 no filesystem could mount root. Tried: cramfs
 kernel panic - not syncing: VFS unable to mount root fs on unknown
 block(0,0)
 
 
 So, I re-tried compiling after de-activating raid0 in the config,
 leaving only raid1... to the same end result.
 
 Could someone tell me what options I should select in the config so as
 to obtain a new kernel that will boot my RAID architecture ?
 
 I know that there are other ways to compile Debian kernels, using
 'make-kpkg', but this does not work here, likely because this tool
 version is too old or too new for my system : it keeps saying that
 parameters are missing, while the howtos that I saw did not mention any
 such parameters for make-kpkg.
 
 By the way, I noticed that the initrd.img that I obtain using mkinitrd
 are about 5 times bigger than that which is used to boot my usual
 kernel. I am not sure that I am using mkinitrd correctly.
 
 mkinitrd -o initrd.img-2.6.26.2 2.6.26 done from /boot issues an error
 messages saying that this image will not take raid into account, and
 that therefore it won't boot, unless the main image contains what is
 missing here. In any case, it doesn't boot. The same command issued in
 /usr/src/linux-2.6.26.2 does not send any error message, but in the end
 the image doesn't boot either.
 
 Thanks in advance for your help

why not just compile it on your notebook (or copy a compiled kernel) ?! 

you also can just disable the loading of the sound modules to make it more
simple.

regards


-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: problem compiling kernel

[Emanoil Kotsev]

Sorry forgot to write

Yes there is problem compiling the 2.6.20 with recent gcc

The problem is the compiler. If you are compiling just grab the last version
from kernel.org.

2.6.30.4 seems to be working fine

 
 So, upon my failures to recompile kernel 2.6.20-16-386, I tried
 downloading 2.6.20-17-386, but I got the same results. However, with
 2.6.26.2, it did compile without errors... but in the end the image
 won't boot !
 
 My compiling process does not generate initrd.img, so I did generate one
 using 'mkinitrd', and I wrote its path in /boot/grub/menu.lst with the
 kernel image.
 
 Here is what I get on booting trials :
 
 boot starts, as usual text displays very fast and you cannot read until
 it stops. When this happen, I can read this :
 
 
 md: raid0 personality registered for level 0
 md: raid1 personality registered for level 1
 device mapper
 .
 no filesystem could mount root. Tried: cramfs
 kernel panic - not syncing: VFS unable to mount root fs on unknown
 block(0,0)
 

try compiling the necessary modules _in_ the kernel. 

 
 So, I re-tried compiling after de-activating raid0 in the config,
 leaving only raid1... to the same end result.

In the config it should be [*] not [M] if booting from raid - do the same
for LVM

make
make install
make modules_install - do you really need kpkg ?

 
 mkinitrd -o initrd.img-2.6.26.2 2.6.26 done from /boot issues an error
 messages saying that this image will not take raid into account, and
 that therefore it won't boot, unless the main image contains what is
 missing here. In any case, it doesn't boot. The same command issued in
 /usr/src/linux-2.6.26.2 does not send any error message, but in the end
 the image doesn't boot either.
 

do you have a not raid boot partition, where you can put the initrd image?




-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: problem compiling kernel

[Emanoil Kotsev]

Emanoil Kotsev wrote:

 Sorry forgot to write
 
 Yes there is problem compiling the 2.6.20 with recent gcc
 
 The problem is the compiler. If you are compiling just grab the last
 version from kernel.org.
 
 2.6.30.4 seems to be working fine
 

Just to be objective the gnu compiler people said kernel people are wrong
and vice versa. I didn't follow the discussion. I'm glad next kernels
compile

regards




-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: problem compiling kernel

[Bernard]

Emanoil Kotsev wrote:


Bernard wrote:

 






why not just compile it on your notebook (or copy a compiled kernel) ?! 


you also can just disable the loading of the sound modules to make it more
simple.
 



Things would be easy if all sound support were in modules. But some 
functions are part of the kernel and load with it. Because of this, I 
cannot compile a new sound system (OSS) without errors : it says that I 
have conflicting problems, even though I have blacklisted all sound modules.



--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: problem compiling kernel

[Bernard]

Emanoil Kotsev wrote:



 



do you have a not raid boot partition, where you can put the initrd image?
 



My boot partition is not raid, or, at least, even though it is mirrored, 
it remains in ext2fs, while the rest is in LVM2. So, the initrd image 
that I am trying is available at start, same with the one that works at 
every boot. If, for instance, I boot my system using a rescue disc or 
CD, I can't mount my '/' partition, but I mount '/boot' without problem.






 




--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: problem compiling kernel

[Bernard]

Emanoil Kotsev wrote:




try compiling the necessary modules _in_ the kernel. 
 



This is the way it has been done.

 


So, I re-tried compiling after de-activating raid0 in the config,
leaving only raid1... to the same end result.
 



In the config it should be [*] not [M] if booting from raid - do the same
for LVM

make
make install
make modules_install - do you really need kpkg ?
 



No, except if it generates an initrd image easier than with mkinitrd

 




--
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Re: problem compiling kernel

[Emanoil Kotsev]

Bernard wrote:

 Emanoil Kotsev wrote:
 
Bernard wrote:

  

 
 

why not just compile it on your notebook (or copy a compiled kernel) ?!

you also can just disable the loading of the sound modules to make it more
simple.
  

 
 Things would be easy if all sound support were in modules. But some
 functions are part of the kernel and load with it. Because of this, I
 cannot compile a new sound system (OSS) without errors : it says that I
 have conflicting problems, even though I have blacklisted all sound
 modules.

ah, I understand right now what you're trying to do - you need basic
OSS/ALSA removed completely.

which compiler versions do you have installed? On sarge the 2.6.20 should
compile as far as I remember.

You might have to set the right compiler I'm not sure I think 4.X is the one
with the problem, so I would try with something older.

The other thing would be to compile on another machine (you need to check
options/Makefile for this) and move the images and drivers over (because
the new OSS driver will probably fail with the old compiler ;) that would
compile the older kernel )

regards




-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

RE: problem compiling kernel

[Kevin Ross]

 From: news [mailto:n...@ger.gmane.org] On Behalf Of Emanoil Kotsev
 Sent: Friday, August 21, 2009 1:55 PM
 
 Bernard wrote:
 
  Emanoil Kotsev wrote:
 
 Bernard wrote:
 
 
 
 
 
 
 why not just compile it on your notebook (or copy a compiled kernel)
 ?!
 
 you also can just disable the loading of the sound modules to make it
 more
 simple.
 
 
 
  Things would be easy if all sound support were in modules. But some
  functions are part of the kernel and load with it. Because of this, I
  cannot compile a new sound system (OSS) without errors : it says that
 I
  have conflicting problems, even though I have blacklisted all sound
  modules.
 
 ah, I understand right now what you're trying to do - you need basic
 OSS/ALSA removed completely.
 
 which compiler versions do you have installed? On sarge the 2.6.20
 should
 compile as far as I remember.
 
 You might have to set the right compiler I'm not sure I think 4.X is
 the one
 with the problem, so I would try with something older.
 
 The other thing would be to compile on another machine (you need to
 check
 options/Makefile for this) and move the images and drivers over
 (because
 the new OSS driver will probably fail with the old compiler ;) that
 would
 compile the older kernel )
 
 regards
 

The kernel in Lenny is compiled with the gcc-4.1 version.  You should
compile with whatever version the stock kernel was compiled with.  If you're
using something older than Lenny, the way I check is, open any module under
/lib/modules/`uname -r` with vi, and search for GCC.  You will see the
version used.  There might be an easier way, but I don't know it.

-- Kevin



-- 
To UNSUBSCRIBE, email to debian-user-requ...@lists.debian.org 
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

problem compiling kernel-headers-2.6.8-2-smp

[Ivan Paganini]

Hello everybody. I am trying to install anbd on a PIII cluster, and for
that I have to compile the module that they provide. But unfortunately,
I am not being capable of compile this module agaist the
kernel-headers-2.6.8-2-686-smp. make menuconfig works nicely, but when I
try to make or make modules, I have this two messages:

make
make[1]: *** No rule to make target `init/main.o', needed by
`init/built-in.o'.  Stop.
make: *** [init] Error 2

or
make modules
make[1]: *** No rule to make target `arch/i386/kernel/msr.c', needed by
`arch/i386/kernel/msr.o'.  Stop.
make: *** [arch/i386/kernel] Error 2
++
I have made make mrproper, make oldconfig, make menuconfig and
configured everything, make clean to be sure that nothing is hanging,
but with no luck. I have copied the nbd.h and nbd.c to the right places
(include and drivers/block). So, what can be happening? Is there other
way to compile this module? When I try to compile standalone, I have
tons of error messages of missing libraries and dependencies.

Thanks in advance.

Ivan Marin
--
---
Ivan S. P. Marin
Laboratório de Física Computacional
lfc.ifsc.usp.br
Instituto de Física de São Carlos - USP
--

Re: Problem Compiling kernel-source-2.4.18

[Jonathan Pietkiewicz]

On Fri, Dec 05, 2003 at 10:00:48AM -0600, Hugo Vanwoerkom wrote:
 In the file included from ksym.c:50:
 /usr/src/kernel-source-2.4.18/include/asm/checksum.h:72:30: missing 
 termintating  character
 
 If you are using gcc 3.3, which you are if you ran dist-upgrade, it 
 *will* complain of that. You can apt-get install gcc-2.95 and that will 
 do away with the error.


What I had to do was a little more than simply installing gcc 2.95.

Change the lines in the makefile which set CC and HOSTCC.

At the start they are set to gcc, but what I had to do was set them
explicitly to gcc-2.95 like this:

In file 'Makefile'

...

HOSTCC = gcc-2.95

...

CC = gcc-2.95

...

End of Makefile representation.

This fixed the problem for me and everything appears to work great.

Best of luck to all finding this by google (like I did).

jdp


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Problem Compiling kernel-source-2.4.18

[Thomas H. George]

My system is testing fully up-to-date.  I just ran apt-get update and 
apt-get dist-upgrade which installed kernel-source-2.4.18..  

My current kernel was built from kernel-source-2.4.22 but I have been 
experiencing irratic behavior - specifically many 
p80211/knetdev_hard-start_xinit: messages on shutdown followed by 
various problems on rebooting.  Genrally a second reboot works.  I 
decided to backup to kernel-2.4.18 in hopes of eliminating the problems.

After make menuconfig and make dep when I run make bzImage the compile 
aborts with a series of messages the first of which is in

In the file included from ksym.c:50:
/usr/src/kernel-source-2.4.18/include/asm/checksum.h:72:30: missing 
termintating  character

I am concerned that the system is compromised.  I have run chkrootkit 
and even tried chkrootkit -x SucKIT but chkrootkit doesn't know SucKIT.

I would appreciate any insights about the nature of the problems and 
what to do about them.

Tom

--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

Re: Problem Compiling kernel-source-2.4.18

[Hugo Vanwoerkom]

Thomas H. George wrote:
My system is testing fully up-to-date.  I just ran apt-get update and 
apt-get dist-upgrade which installed kernel-source-2.4.18.. 
My current kernel was built from kernel-source-2.4.22 but I have been 
experiencing irratic behavior - specifically many 
p80211/knetdev_hard-start_xinit: messages on shutdown followed by 
various problems on rebooting.  Genrally a second reboot works.  I 
decided to backup to kernel-2.4.18 in hopes of eliminating the problems.

After make menuconfig and make dep when I run make bzImage the compile 
aborts with a series of messages the first of which is in

In the file included from ksym.c:50:
/usr/src/kernel-source-2.4.18/include/asm/checksum.h:72:30: missing 
termintating  character


And is there a missing terminating character?
Checking I note that the line in question is:
	__asm__ __volatile__(

which indeed does have a missing  character.

If you are using gcc 3.3, which you are if you ran dist-upgrade, it 
*will* complain of that. You can apt-get install gcc-2.95 and that will 
do away with the error.

Also apt-get install kernel-package and instead do:

make-kpkg --revision 1 kernel_image

and that will create a deb file for you that is installable as the new 
kernel: it will do all that make dep for you.

I would not be afraid that the system is compromised until you really 
see something that is not of your own doing...

Hugo

I am concerned that the system is compromised.  I have run chkrootkit 
and even tried chkrootkit -x SucKIT but chkrootkit doesn't know SucKIT.

I would appreciate any insights about the nature of the problems and 
what to do about them.

Tom




--
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

problem compiling kernel

[Richard Kimber]

I tried to compile a new kernel with
CC=gcc-2.95 make-kpkg --added_patches=lowlatency,preempt
--append-to-version=.rk1103

but got the following:-
In file included from ide-cd.c:318:
ide-cd.h:440: error: long, short, signed or unsigned used invalidly for
`slot_tablelen'
make[4]: *** [ide-cd.o] Error 1
make[4]: Leaving directory `/usr/src/kernel-source-2.4.19/drivers/ide'
make[3]: *** [first_rule] Error 2
make[3]: Leaving directory `/usr/src/kernel-source-2.4.19/drivers/ide'
make[2]: *** [_subdir_ide] Error 2
make[2]: Leaving directory `/usr/src/kernel-source-2.4.19/drivers'
make[1]: *** [_dir_drivers] Error 2
make[1]: Leaving directory `/usr/src/kernel-source-2.4.19'
make: *** [stamp-build] Error 2

Is this a bug, or have I done something silly?

I've been using 2.4.19.  Would I be better off with another version?

- Richard
-- 
Richard Kimber
http://www.psr.keele.ac.uk/


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED] 
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]

lm_sensors: first problem compiling kernel

[Julio Merino]

Hi all,

Here it comes my first problem...
I've unpacked a freshly kernel-2.4.9 sources on /usr/src, and copied my old
configuration there (.config). I've run make menuconfig and deselected the
bttv drivers and i2c. (I've reviewed all the configuration and seems ok).

So, I run:
make-kpkg configure --revision=juli2
Which works fine.

And then, I issue:
make-kpkg
to build the kernel and it fails:

-
/usr/bin/make all_targets
make[4]: Entering directory `/usr/src/linux-2.4.9/drivers/char'
gcc -D__KERNEL__ -I/usr/src/linux-2.4.9/include -Wall -Wstrict-prototypes 
-Wno-trigraphs -O2 -fomit-frame-pointer -fno-strict-aliasing -fno-common -pipe 
-mpreferred-stack-boundary=2 -march=i686-c -o vt.o vt.c
vt.c: In function `vt_ioctl':
vt.c:507: `kbd_rate' undeclared (first use in this function)
vt.c:507: (Each undeclared identifier is reported only once
vt.c:507: for each function it appears in.)
vt.c:514: `kbd_rate' used prior to declaration
vt.c:514: warning: implicit declaration of function `kbd_rate'
make[4]: *** [vt.o] Error 1
make[4]: Leaving directory `/usr/src/linux-2.4.9/drivers/char'
make[3]: *** [first_rule] Error 2
make[3]: Leaving directory `/usr/src/linux-2.4.9/drivers/char'
make[2]: *** [_subdir_char] Error 2
make[2]: Leaving directory `/usr/src/linux-2.4.9/drivers'
make[1]: *** [_dir_drivers] Error 2
make[1]: Leaving directory `/usr/src/linux-2.4.9'
make: *** [stamp-build] Error 2
juli:/usr/src/linux# 
-

What happens? I've had the same problem last they when trying to compile
a very tiny kernel...

Thanks!

-- 
FreeBSD is the power--
Julio Merino [EMAIL PROTECTED] ICQ: 18961975


pgpJtIM0CvdF4.pgp
Description: PGP signature

Problem compiling kernel

[Georg Werthner]

hi,
i installed debian v2.2rc0.
i have a problem compiling the kernel 2.2.17:
the makefile in /usr/src/linux/arch/i386/boot wants to call as86 with
options -0  -a  and ld86.
but as doesnt recognize the option -0 .
funny thing is that bulding kernel 2.4.0test8 works.
i also tried 2.2.15 but i had the same problem. 
i hope that somebody can help me.
thx georg

dselect problem/compiling kernel

[KTB]

Hi, I'm attempting my first kernel compiling and was reading in the
Debian User's Guide that an easy way to do this is to install,
kernel-source and kernel-package.  I have also heard there is an X
configuration tool for this also.  Anyway I tried installing the kernel
packages with dselect.  I have used dselect to install programs off the
official cd before.  This time I selected my packages with + I hit
enter to accept the default dependent packages it selected.  Then
enter again and then install.  The problem is, shortly after
selecting install I'm told that the installation is OK hit return to
continue.  Usually there is a ton of info that scrolls by before I get
installation is OK..., that is not happening.  Dselect says the
programs have not been installed under select.  Anyone know what is
going on here.  I tried a reboot to clear dselect but that didn't help.
I have gone through these steps twice.
Thanks,
Kent

Re: dselect problem/compiling kernel

[KTB]

David Stern wrote:

 On Tue, 09 Feb 1999 10:44:24 CST, KTB wrote:
  Hi, I'm attempting my first kernel compiling and was reading in the
  Debian User's Guide that an easy way to do this is to install,
  kernel-source and kernel-package.  I have also heard there is an X
  configuration tool for this also.  Anyway I tried installing the kernel
  packages with dselect.  I have used dselect to install programs off the
  official cd before.  This time I selected my packages with + I hit
  enter to accept the default dependent packages it selected.  Then
  enter again and then install.  The problem is, shortly after
  selecting install I'm told that the installation is OK hit return to
  continue.  Usually there is a ton of info that scrolls by before I get
  installation is OK..., that is not happening.  Dselect says the
  programs have not been installed under select.  Anyone know what is
  going on here.  I tried a reboot to clear dselect but that didn't help.
  I have gone through these steps twice.

 It sounds like the most recent available version of packages are
 installed.  Some questions to consider:
   o  Did you successfully configure the method using the [A]ccess menu
 item?

I haven't changed the access.  When I selected install the cdrom takes off
just as it has before.


   o  Did you successfully [U]pdate the available packages using the
 menu?

No, if I'm reading you correctly that is what I'm trying to do.  Usually when
I want something off the cd I go into dselect, do a search for whatever I'm
looking for, in this case kernel-packages, deselect displays the line where
the package is, I hit shift + and then select install and config  This
time it didn't work.


   o  What do the ton of messages that scroll by say? (when you quit
 dselect, hold the shift key and press the page-up key to scroll
 backwards)

There were no messages scrolling by when I tried to install the packages in
question.  In the past I had a lot of messages scrolling by when I installed
other packages.  This was my first clue that the installation didn't happen.


   o  Are the packages already installed? (dpkg -l kernel-package, look
 for i i )

I tried this and the packages I'm trying to install are not installed.


   o  Are you logged in as root?

Yes I was.  By default I can't do anything in dselect except view packages if
I'm not in root.

Thanks,
Kent

Re: dselect problem/compiling kernel

[KTB]

Thanks I have the packages installed:)  I had forgotten a while back I was 
trying to figure out how to get kde off the extras cd and I was messing around 
with the access menu then and must have screwed it up so that I couldn't access 
the
binary cd.
Thanks,
Kent

David Stern wrote:

 On Tue, 09 Feb 1999 12:54:32 CST, KTB wrote:
  David Stern wrote:

   It sounds like the most recent available version of packages are
   installed.  Some questions to consider:
 o  Did you successfully configure the method using the [A]ccess menu
   item?
 
  I haven't changed the access.  When I selected install the cdrom takes off
  just as it has before.

 Generally, the cdrom access method is straightforward, but make sure
 your cdrom is specified as a particular block device, and not as a
 symlink (/dev/cdrom can be a symlink to /dev/hdaX)

 o  Did you successfully [U]pdate the available packages using the
   menu?
 
  No, if I'm reading you correctly that is what I'm trying to do.  Usually 
  when
  I want something off the cd I go into dselect, do a search for whatever I'm
  looking for, in this case kernel-packages, deselect displays the line where
  the package is, I hit shift + and then select install and config  This
  time it didn't work.

 Unless your available packages have changed (unlikely on a cdrom), you
 don't ordinarily need to update the list.  However, it is somewhat
 common for available package lists to become corrupt every once in a
 while, and the solution is to blow em away and start over.  dpkg
 --clear-avail

 o  What do the ton of messages that scroll by say? (when you quit
   dselect, hold the shift key and press the page-up key to scroll
   backwards)
 
  There were no messages scrolling by when I tried to install the packages in
  question.  In the past I had a lot of messages scrolling by when I installed
  other packages.  This was my first clue that the installation didn't happen.

 Maybe this was one of those dependency related issues, where certain
 packages have to be installed before others can be installed.
 Sometimes it helps to run though install a couple times, and configure
 at least once.

 If none of this helps, then more specific wording of messages during the 
 attempted install phase would be helpful.  Act like you're going to install, 
 and after it does whatever it's doing, quit, then cut and paste the messages 
 into a mail.
 --
 David
 [EMAIL PROTECTED]

Re: problem compiling kernel 2.0.34 under Debian 1.3.1

[Patrick Olson]

On Mon, 6 Jul 1998, Shaleh wrote:

  gcc: Internal compiler error: program cc1 got fatal signal 11
 
 sig11 (signal 11) is often a sign of a hardware problem.  Either you
 machine is over/under clocked, over heating, has a memory glitch or
 something.  Sig 11 can also be one of the problems that appears and then
 never re-appears.

The sysetm is not over or under clocked as far as I know.  I will double
check.  However, I think it is overheating because I have been looking at
http://www.bitwizard.nl/sig11/ (as suggested by someone else on the list)
and got the idea to try make repeatedly.

Sure enough, it stalls quicker and quicker every time I try it, but if I
leave the system idle for a few minutes then try again it goes a little
further.

Thank you to all who responded.  Now I just have to figure out WHAT is
overheating.



--  
Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED]  /dev/null

problem compiling kernel 2.0.34 under Debian 1.3.1

[Patrick Olson]

I've been running Debian 1.3.1 with kernel 2.0.29

I decided to upgrade to kernel 2.0.34 but it fails during make zImage with
an error message.  Can anyone help?

Here's the error message and a few of the lines before it:

gcc -D__KERNEL__ -I/usr/src/linux-2.0.34/include -Wall -Wstrict-prototypes
-O2 -fomit-frame-pointer -fno-strength-reduce -pipe -m486 -malign-loops=2
-malign-jumps=2 -malign-functions=2 -DCPU=586  -c -o file.o file.c
gcc: Internal compiler error: program cc1 got fatal signal 11
cpp: output pipe has been closed
make[3]: *** [file.o] Error 1
make[3]: Leaving directory `/usr/src/linux-2.0.34/fs/ext2'
make[2]: *** [first_rule] Error 2
make[2]: Leaving directory `/usr/src/linux-2.0.34/fs/ext2'
make[1]: *** [sub_dirs] Error 2
make[1]: Leaving directory `/usr/src/linux-2.0.34/fs'
make: *** [linuxsubdirs] Error 2



--  
Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED]  /dev/null

Re: problem compiling kernel 2.0.34 under Debian 1.3.1

[Shaleh]

 gcc: Internal compiler error: program cc1 got fatal signal 11

sig11 (signal 11) is often a sign of a hardware problem.  Either you
machine is over/under clocked, over heating, has a memory glitch or
something.  Sig 11 can also be one of the problems that appears and then
never re-appears.


--  
Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED]  /dev/null

Re: problem compiling kernel 2.0.34 under Debian 1.3.1

[Brandon Mitchell]

On Mon, 6 Jul 1998, Shaleh wrote:

  gcc: Internal compiler error: program cc1 got fatal signal 11
 
 sig11 (signal 11) is often a sign of a hardware problem.  Either you
 machine is over/under clocked, over heating, has a memory glitch or
 something.  Sig 11 can also be one of the problems that appears and then
 never re-appears.

There is a program for checking your memory (sorry, forgot the name, my
memory is going bad :-). The sig 11 homepage may also be helpful:

http://www.bitwizard.nl/sig11/

HTH,
Brandon

--+--
Brandon Mitchell [EMAIL PROTECTED] | Debian Testing Group Status
PGP Key:   finger -l [EMAIL PROTECTED] |  http://bhmit1.home.ml.org/deb/
Dijkstra probably hates me (Linus Torvalds, in kernel/sched.c)


--  
Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED]  /dev/null

Re: problem compiling kernel 2.0.34 under Debian 1.3.1

[servis]

*-Patrick Olson ( 6 Jul)
| 
| I've been running Debian 1.3.1 with kernel 2.0.29
| 
| I decided to upgrade to kernel 2.0.34 but it fails during make zImage with
| an error message.  Can anyone help?
| 
| Here's the error message and a few of the lines before it:
| 
| gcc -D__KERNEL__ -I/usr/src/linux-2.0.34/include -Wall -Wstrict-prototypes
| -O2 -fomit-frame-pointer -fno-strength-reduce -pipe -m486 -malign-loops=2
| -malign-jumps=2 -malign-functions=2 -DCPU=586  -c -o file.o file.c
| gcc: Internal compiler error: program cc1 got fatal signal 11
| cpp: output pipe has been closed
| make[3]: *** [file.o] Error 1
| make[3]: Leaving directory `/usr/src/linux-2.0.34/fs/ext2'
| make[2]: *** [first_rule] Error 2
| make[2]: Leaving directory `/usr/src/linux-2.0.34/fs/ext2'
| make[1]: *** [sub_dirs] Error 2
| make[1]: Leaving directory `/usr/src/linux-2.0.34/fs'
| make: *** [linuxsubdirs] Error 2
| 

Check out, http://www.bitwizard.nl/sig11/.  Your problem may be found
through the means mentioned on that page.  Usually a hardware problem.

-- 
Brian 
-- 
Mechanical Engineering  [EMAIL PROTECTED]
Purdue University   http://www.ecn.purdue.edu/~servis


--  
Unsubscribe?  mail -s unsubscribe [EMAIL PROTECTED]  /dev/null

Re: problem compiling kernel with sound

[Shaleh]

 #error You will need to configure the sound driver with
 CONFIG_AUDIO option.

I had to manually edit the option file in the kernel directory (Forgot
its real name and am at work right now).  Everytime I tried to specifiy
an IRQ it told me I typed an invalid number -- even the default 7.


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to
[EMAIL PROTECTED] . 
Trouble?  e-mail to [EMAIL PROTECTED] .

Re: problem compiling kernel with sound

[Oliver Elphick]

In message [EMAIL PROTECTED], writes:
  Has anyone had any luck with compiling a kernel on Debian? I can compile
  fine just as long as I don't try to compile with sound. NOTE: I did
  use the make-kpkg -revision custom.1.0 kernel_image option. This is the
  error I get (The same thing happens with 2.0.30 also):
  
  
  gcc -D__KERNEL__ -I/usr/src/kernel-source-2.0.29/include -Wall
  -Wstrict-prototypes -O2 -fomit-frame-pointer -fno-strength-reduce -pipe
  -m386 -DCPU=386 -DMODULE -DMODVERSIONS -include
  /usr/src/kernel-source-2.0.29/include/linux/modversions.h  -c -o
  sb_common.o sb_common.c
  sb_common.c:21: #error You will need to configure the sound driver with
  CONFIG_AUDIO option.
  
I suggest that you use `make xconfig' or `make menuconfig' to set up
your kernel configuration.  That should leave the configuration in
a consistent state.

After `make xconfig' do `make depend' and `make clean' before doing
the `make' to compile the source.

-- 
Oliver Elphick[EMAIL PROTECTED]
Isle of Wight  http://lfix.co.uk/oliver

PGP key from public servers; key ID 32B8FAA1

 Make it idiot-proof, and someone will breed a better idiot.



--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to
[EMAIL PROTECTED] . Trouble? 
e-mail to [EMAIL PROTECTED] .

problem compiling kernel with sound

[Ricardo Muggli]

Has anyone had any luck with compiling a kernel on Debian? I can compile
fine just as long as I don't try to compile with sound. NOTE: I did
use the make-kpkg -revision custom.1.0 kernel_image option. This is the
error I get (The same thing happens with 2.0.30 also):


gcc -D__KERNEL__ -I/usr/src/kernel-source-2.0.29/include -Wall
-Wstrict-prototypes -O2 -fomit-frame-pointer -fno-strength-reduce -pipe
-m386 -DCPU=386 -DMODULE -DMODVERSIONS -include
/usr/src/kernel-source-2.0.29/include/linux/modversions.h  -c -o
sb_common.o sb_common.c
sb_common.c:21: #error You will need to configure the sound driver with
CONFIG_AUDIO option.
sb_common.c:266: warning: `sb16_set_mpu_port' defined but not used
make[4]: *** [sb_common.o] Error 1
make[4]: Leaving directory `/usr/src/kernel-source-2.0.29/drivers/sound'
make[3]: *** [modules] Error 2
make[3]: Leaving directory `/usr/src/kernel-source-2.0.29/drivers'
make[2]: *** [modules] Error 2
make[2]: Leaving directory `/usr/src/kernel-source-2.0.29'
make[1]: *** [build] Error 2
make[1]: Leaving directory `/usr/src/kernel-source-2.0.29'
make: *** [stamp-image] Error 2


Thanks for any help. I've had great responses from this list before and
have learned a LOT from reading all of the other posts. thanks

- ricardo



--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to
[EMAIL PROTECTED] . Trouble? 
e-mail to [EMAIL PROTECTED] .

Re: problem compiling kernel with sound

[kestrel]

On Sun, 14 Sep 1997, Oliver Elphick wrote:

 In message [EMAIL PROTECTED], writes:
   Has anyone had any luck with compiling a kernel on Debian? I can compile
 [snip]
  
 I suggest that you use `make xconfig' or `make menuconfig' to set up
 your kernel configuration.  That should leave the configuration in
 a consistent state.
 
 After `make xconfig' do `make depend' and `make clean' before doing
 the `make' to compile the source.
 
A note about make menuconfig, if you have a SoundBlaster and don't use the
defult settings you run into some problems unless you run the old configure
script, the dialog boxes won't let you alter the defaults for some reason.

G'razel the shifty kitty
([EMAIL PROTECTED])
http://www.aye.net/~kestrel
found on Tapestries FurryMUCK


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to
[EMAIL PROTECTED] . 
Trouble?  e-mail to [EMAIL PROTECTED] .

Problem compiling kernel

[Seth Rutenberg]

I am having difficulty compiling kernel 2.0.27 2.
So I tried a compile without any of the kernel features for the sake of
diagnosing the problem, and the problem persisted.
It comes at the end when compiling is finished and make calls the ld
command:

ld: warning: cannot find entry symbol _start; defaulting to 000fffe0
nm vmlinux | grep -v '\(compiled\)\|\(\.o$\)\|\( a \)' | sort 
System.map
make[1]: Entering directory
`/usr/src/kernel-source-2.0.27-2/arch/i386/boot'
make[2]: Entering directory
`/usr/src/kernel-source-2.0.27-2/arch/i386/boot/compressed'
./xtract /usr/src/kernel-source-2.0.27-2/vmlinux | gzip -9 | ./piggyback
 piggy.o
Non-GCC header of 'system'
Compressed size 20.
ld -qmagic -Ttext 0xfe0 -o vmlinux head.o misc.o piggy.o
ld: warning: cannot find entry symbol _start; defaulting to 0fe0
misc.o: In function `fill_inbuf':
misc.o(.text+0x1ebc): undefined reference to `input_data'
misc.o(.text+0x1ec1): undefined reference to `input_len'
misc.o(.text+0x1ed7): undefined reference to `input_data'
make[2]: *** [vmlinux] Error 1
make[2]: Leaving directory
`/usr/src/kernel-source-2.0.27-2/arch/i386/boot/compressed'
make[1]: *** [compressed/vmlinux] Error 2
make[1]: Leaving directory
`/usr/src/kernel-source-2.0.27-2/arch/i386/boot'
make: *** [zImage] Error 2

Can anyone please help with this?


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to
[EMAIL PROTECTED] . 
Trouble?  e-mail to [EMAIL PROTECTED] .

problem compiling kernel using make-kpkg

[Lawrence Chim]

I have problem to compile the Linux kernel 2.0.28 using make-kpkg,
here is a part of the screen dump, it didn't happen when I compiled
2.0.26

MIDI interface support (CONFIG_MIDI) [Y/n/?] 
FM synthesizer (YM3812/OPL-3) support (CONFIG_YM3812) [Y/n/?] 
I/O base for SB Check from manual of the card (SBC_BASE) [220] 

  Sorry, no help available for this option yet.

I/O base for SB Check from manual of the card (SBC_BASE) [220] 

  Sorry, no help available for this option yet.


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to
[EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]

Re: problem compiling kernel using make-kpkg

[Jon Rabone]

 I have problem to compile the Linux kernel 2.0.28 using make-kpkg,
 here is a part of the screen dump, it didn't happen when I compiled
 2.0.26

This is because the behaviour of GNU expr regexps in shellutils 1.14 has
changed to be more POSIX-like, AND 2.0.28 has started using a different
method to configure the sound driver which uses expr to parse the base
address / irq.

Fix: Edit scripts/Configure, find the lines using expr in the functions
hex and int, put a backslash in front of the + or the ? you find in the
expr line. 

Jon.


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to
[EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]

Re: problem compiling kernel using make-kpkg

[Brian Mays]

Lawrence Chim [EMAIL PROTECTED] writes:

 I have problem to compile the Linux kernel 2.0.28 using make-kpkg,
 here is a part of the screen dump, it didn't happen when I compiled
 2.0.26
 
 MIDI interface support (CONFIG_MIDI) [Y/n/?] 
 FM synthesizer (YM3812/OPL-3) support (CONFIG_YM3812) [Y/n/?] 
 I/O base for SB Check from manual of the card (SBC_BASE) [220] 
 
   Sorry, no help available for this option yet.
 
 I/O base for SB Check from manual of the card (SBC_BASE) [220] 
 
   Sorry, no help available for this option yet.

This problem is caused by an error in the expr program in the
shellutils package.  I know that this problem no longer exists on my
system where I have installed shellutils v1.14-4 and libc5 v5.4.17-1.
If you want a quick way to fix this problem you can make the following
changes to the linux/scripts/Configure file.  Change the line that reads

if expr $ans : '0$\|-?[1-9][0-9]*$'  /dev/null; then

to

if expr $ans : '0$\|-\?[1-9][0-9]*$'  /dev/null; then

Also change the line that reads

if expr $ans : '[0-9a-fA-F]+$'  /dev/null; then

to

if expr $ans : '[0-9a-fA-F]\+$'  /dev/null; then

-- 
Brian


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word unsubscribe to
[EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED]