[gentoo-catalyst] Problems to compile klibc(udev) with genkernel 3.2.1

[gentoo-catalyst] Problems to compile klibc(udev) with genkernel 3.2.1

["Andreas Rückert"]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="us-ascii", Size: 3587 bytes --]

Hi!

I'm stuck again. I guess I tried more than 20 times to run stage 2 
with a current portage snapshot. No luck so far. I tried at least
5 kernel versions, 2 processor types and lots of spec file variants.

Using udev, klibc fails with 2 missing header files:
=================================
* udev: >> Compiling...
COMMAND: make j1  KERNEL_DIR=/usr/src/linux USE_KLIBC=true USE_LOG=false
DEBUG=false udevdir=/dev all etc/udev/udev.conf
In file included from ../linux/include/linux/compiler.h:44,
                 from ../linux/include/linux/stddef.h:4,
                 from ../linux/include/linux/posix_types.h:4,
                 from ../include/sys/types.h:16,
                 from ../include/unistd.h:11,
                 from ../include/stdio.h:11,
                 from vsnprintf.c:13:
../linux/include/linux/compiler-gcc3.h:22:1: warning: "__attribute_pure__"
redefined
In file included from ../include/unistd.h:9,
                 from ../include/stdio.h:11,
                 from vsnprintf.c:13:
../include/klibc/compiler.h:77:1: warning: this is the location of the
previous definition
In file included from ../linux/include/linux/compiler.h:44,
                 from ../linux/include/linux/stddef.h:4,
                 from ../linux/include/linux/posix_types.h:4,
                 from ../include/sys/types.h:16,
                 from ../include/unistd.h:11,
                 from ../include/stdio.h:11,
                 from vsnprintf.c:13:
../linux/include/linux/compiler-gcc3.h:23:1: warning: "__attribute_const__"
redefined
In file included from ../include/unistd.h:9,
                 from ../include/stdio.h:11,
                 from vsnprintf.c:13:
../include/klibc/compiler.h:60:1: warning: this is the location of the
previous definition
In file included from ../include/sys/types.h:16,
                 from ../include/unistd.h:11,
                 from ../include/stdio.h:11,
                 from vsnprintf.c:13:
../linux/include/linux/posix_types.h:47:29: asm/posix_types.h: No such file
or directory
In file included from ../include/unistd.h:11,
                 from ../include/stdio.h:11,
                 from vsnprintf.c:13:
../include/sys/types.h:18:23: asm/types.h: No such file or directory
In file included from ../include/unistd.h:11,
                 from ../include/stdio.h:11,
                 from vsnprintf.c:13:
../include/sys/types.h:25: error: parse error before "ino_t"
../include/sys/types.h:25: warning: type defaults to `int' in declaration of
`ino_t'
../include/sys/types.h:25: warning: data definition has no type or storage
class
../include/sys/types.h:26: error: parse error before "mode_t"
../include/sys/types.h:26: warning: type defaults to `int' in declaration of
`mode_t'
../include/sys/types.h:26: warning: data definition has no type or storage
class
../include/sys/types.h:27: error: parse error before "nlink_t"
<snipped many more errors>
=================================
I guess the problem here is, that some directory is missing in the 
include directory list, and so the compiler cannot find types.h and 
posix_types.h. 
I searched google and the gentoo forums with no luck so far.

The other variant would be to use devfs again. Genkernel then fails,
because it could not copy the image for the minimum kernel package
(no clue why yet).

Thanks for any hint,
Andreas

-- 
5 GB Mailbox, 50 FreeSMS http://www.gmx.net/de/go/promail
+++ GMX - die erste Adresse für Mail, Message, More +++
-- 
gentoo-catalyst@gentoo.org mailing list

[gentoo-catalyst] Re: Problems to compile klibc(udev) with genkernel 3.2.1

[Christophe PEREZ]

Le Wed, 13 Jul 2005 07:52:28 +0200, Andreas Rückert a écrit :

> Thanks for any hint,

The problem may be the same I had.
You compile your stage2 (livecd-stage2 ?) with a snapshot and a
(livecd-?)stage1 that don't correspond.

I had to build : snapshot, stage1, stage2, stage3, livecd-stage1,
and finaly livecd-stage2.
Very long but worked fine.

My 2 cents...

PS : I try to answer just devels are sleeping at this time ;-)

-- 
Christophe PEREZ
--
gentoo-catalyst@gentoo.org mailing list

Re: [gentoo-catalyst] Problems to compile klibc(udev) with genkernel 3.2.1

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 637 bytes --]

On Wed, 2005-07-13 at 07:52 +0200, "Andreas Rückert" wrote:
> Hi!
> 
> I'm stuck again. I guess I tried more than 20 times to run stage 2 
> with a current portage snapshot. No luck so far. I tried at least
> 5 kernel versions, 2 processor types and lots of spec file variants.

livecd-stage2, right?

File a bug against genkernel.

It seems to be working fine for us on the release building, however.
Perhaps there is something in your spec file causing this issue?

Also, try upgrading to genkernel 3.2.2 first.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead/QA Manager
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-catalyst] Re: Problems to compile klibc(udev) with genkernel 3.2.1

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 559 bytes --]

On Wed, 2005-07-13 at 02:35 -0400, Christophe PEREZ wrote:
> I had to build : snapshot, stage1, stage2, stage3, livecd-stage1,
> and finaly livecd-stage2.

You can skip building the initial stages and just use one that it more
up-to-date.  I usually use either a 2005.0 stage3 or a pre-release
2005.1 stage3.

Of course, at the moment I am using my own stage3 tarballs, but that's
just because we're in the process of working on the release.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead/QA Manager
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-catalyst] Re: Problems to compile klibc(udev) with genkernel 3.2.1

[Andreas Rueckert]

Hi!

On Wednesday 13 July 2005 16:36, Chris Gianelloni wrote:
> On Wed, 2005-07-13 at 02:35 -0400, Christophe PEREZ wrote:
> > I had to build : snapshot, stage1, stage2, stage3, livecd-stage1,
> > and finaly livecd-stage2.
>
> You can skip building the initial stages and just use one that it more
> up-to-date.  I usually use either a 2005.0 stage3 or a pre-release
> 2005.1 stage3.

That's what I currently use:
source_subpath: default/stage3-x86-2005.0

> Of course, at the moment I am using my own stage3 tarballs, but that's
> just because we're in the process of working on the release.

I'll try Christophe's suggestion and rebuild everything from scratch. Maybe 
it'll help...

Ciao,
Andreas
-- 
gentoo-catalyst@gentoo.org mailing list

[gentoo-catalyst] Re: Re: Problems to compile klibc(udev) with genkernel 3.2.1

[Christophe PEREZ]

Le Wed, 13 Jul 2005 19:02:20 +0200, Andreas Rueckert a écrit :

> I'll try Christophe's suggestion and rebuild everything from scratch. Maybe 
> it'll help...

Always try Chris suggestion before, never mine ! ;-)
-> FAQ

-- 
Christophe PEREZ
--
gentoo-catalyst@gentoo.org mailing list

Re: [gentoo-catalyst] Re: Problems to compile klibc(udev) with genkernel 3.2.1

[Andreas Rueckert]

Hi!

On Wednesday 13 July 2005 08:35, Christophe PEREZ wrote:
> Le Wed, 13 Jul 2005 07:52:28 +0200, Andreas Rückert a écrit :
> > Thanks for any hint,
>
> The problem may be the same I had.
> You compile your stage2 (livecd-stage2 ?) with a snapshot and a
> (livecd-?)stage1 that don't correspond.

I think it's both x86 2005.0

> I had to build : snapshot, stage1, stage2, stage3, livecd-stage1,
> and finaly livecd-stage2.
> Very long but worked fine.

I'll give it a try. Time is not that much of an issue, since I bought a 
dedicated machine running nothing but catalyst. And it's running for 2 weeks 
now, but without much success... :-(

> My 2 cents...
>
> PS : I try to answer just devels are sleeping at this time ;-)

Thanks a lot! I pretty much gave up on sleeping...looking after this machine 
at 2am and then 7am the next time to make sure that the build process was not 
interrupted again...

Ciao,
Andreas

-- 
gentoo-catalyst@gentoo.org mailing list

Re: [gentoo-catalyst] Re: Re: Problems to compile klibc(udev) with genkernel 3.2.1

[Andreas Rueckert]

Hi!

On Wednesday 13 July 2005 19:07, Christophe PEREZ wrote:
> Le Wed, 13 Jul 2005 19:02:20 +0200, Andreas Rueckert a écrit :
> > I'll try Christophe's suggestion and rebuild everything from scratch.
> > Maybe it'll help...
>
> Always try Chris suggestion before, never mine ! ;-)
> -> FAQ

I know. But your suggestion came first, and I still think it's a good idea in 
general to have a complete generation pipeline for our livecd, so we have 
more control over the entire build process. I already modified our makefile 
to include the first 3 stages before the livecd stuff is started.

Ciao and thanks again,
Andreas

-- 
gentoo-catalyst@gentoo.org mailing list

Re: [gentoo-catalyst] Problems to compile klibc(udev) with genkernel 3.2.1

[Andreas Rueckert]

Hi!

On Wednesday 13 July 2005 16:32, Chris Gianelloni wrote:
> On Wed, 2005-07-13 at 07:52 +0200, "Andreas Rückert" wrote:
> > Hi!
> >
> > I'm stuck again. I guess I tried more than 20 times to run stage 2
> > with a current portage snapshot. No luck so far. I tried at least
> > 5 kernel versions, 2 processor types and lots of spec file variants.
>
> livecd-stage2, right?

Yup!

> File a bug against genkernel.

Ok, but let me complete the current building process, so I'm sure, that it's 
not my seed stage, that is causing the problem.

> It seems to be working fine for us on the release building, however.
> Perhaps there is something in your spec file causing this issue?
>
> Also, try upgrading to genkernel 3.2.2 first.

That's my next step. Updating catalyst and genkernel, then rebuilding all the 
stages and if it doesn't work, file a bug.

The problem is just, that the release cycles of genkernel and catalyst have 
become so short, that my build processes takes longer. So when my catalyst 
calls are completed and I get an error message in livecd-stage2, there's 
already a new release to try... :-)
Maybe I should invest in something like a 4-way Opteron, so I could keep up 
with your pace... :-)

Ciao,
Andreas

-- 
gentoo-catalyst@gentoo.org mailing list

Re: [gentoo-catalyst] Problems to compile klibc(udev) with genkernel 3.2.1

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 1501 bytes --]

On Thu, 2005-07-14 at 11:25 +0200, Andreas Rueckert wrote:
> The problem is just, that the release cycles of genkernel and catalyst have 
> become so short, that my build processes takes longer. So when my catalyst 
> calls are completed and I get an error message in livecd-stage2, there's 
> already a new release to try... :-)

Yeah.  When we near a release, both catalyst and genkernel start dumping
out new versions as we tweak things and fix bugs.  Things will slow down
in another month or so with catalyst, at least.  I honestly don't expect
any more than one more release of catalyst 1.x any time soon.

After the release, we will be shifting our focus on catalyst 2.0 and
things will get shaken up a little bit, as catalyst 2 is "mostly
compatible" with catalyst 1.x spec files.  Don't worry, we'll update the
templates that ship with the package, so you guys won't be on your own.
We also plan on updating the online documentation to be accurate for
both catalyst 1.x and catalyst 2.x.

> Maybe I should invest in something like a 4-way Opteron, so I could keep up 
> with your pace... :-)

Hey now, I'm only using a dual Opteron.  :P

The main reason for the rapid changes is we have a fairly large number
of people (all of Release Engineering) doing builds at the moment since
we're nearing a release.  That's a lot of people squashing and finding
bugs.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead/QA Manager
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]