All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sven Luther <sven.luther@wanadoo.fr>
To: Sven Luther <sven.luther@wanadoo.fr>
Cc: linuxppc-dev@ozlabs.org
Subject: Re: ARCH=powerpc kexec seems to be broken ...
Date: Wed, 28 Dec 2005 17:51:36 +0100	[thread overview]
Message-ID: <20051228165136.GA19559@localhost.localdomain> (raw)
In-Reply-To: <20051228163829.GA19292@localhost.localdomain>

On Wed, Dec 28, 2005 at 05:38:29PM +0100, Sven Luther wrote:
> On Wed, Dec 28, 2005 at 11:48:13AM +0100, Sven Luther wrote:
> > Hi, ...
> > 
> > I have for some time tried to build 2.6.15-rc debian kernels with
> > ARCH=powerpc, and never quite succeeded. Thanks to Johannes Berg
> > for helping investigating the issue.
> > 
> > The problem seems to be that my .config enables KEXEC, and SWSUSP,
> > which has nasty problems with ARCH=powerpc. In a 32bit config, i get :
> > 
> > 00:03 < svenl> arch/ppc/kernel/built-in.o: In function `swsusp_arch_resume': multiple definition of `machine_shutdown'
> > 00:03 < svenl> arch/powerpc/kernel/built-in.o: first defined here
> > 
> > Which seems to indicate that machine_shutdown is built twice, once in the
> > arch/ppc stuff and a second time in the arch/powerpc stuff.
> > 
> > I have a similar problem for the 64bit case, but have not yet reached the
> > point of the build where this happens, so i can't paste the error message
> > here, but it was of the same kind as above if i remember well.
> > 
> > Johannes said :
> > 
> > 11:23 < johill> you enabled kexec, right?
> > 11:24 < svenl> yeah, i did.
> > 11:24 < svenl> is it broken for ARCH=powerpc right now ?
> > 11:24 < johill> seems so
> > 11:28 < johill> yeah. I don't think I can fix it, know too little about kexec
> > 11:30 < johill> machine_shutdown should at least be moved from
> > arch/ppc/kernel/machine_kexec.c to some file that is ARCH=ppc specific
> > 11:31 < svenl> johill: as i understand this, i think it is already in some
> > ARCH=ppc specific file, which is the cause of this trouble.
> > 11:32 < johill> the way I interpret it is that the machine_kexec.c file is
> > compiled for ARCH=powerpc too since it contains the 32bit specific
> > implementation of kexec
> > 11:32 < johill> but it contains a non-static machine_shutdown() function which
> > ARCH=powerpc already moved to some non-32bit specific file
> > 
> > So, it would be neat to either fix it, or at least mark the option as broken
> > on 64bit powerpc, or something such.
> 
> And here is the 64bit error message : 
> 
>   __restore_cpu_setup': multiple definition of `machine_shutdown'
>   arch/powerpc/kernel/built-in.o: first defined here
> 
> I will disable KEXEC and see how it goes.

Oh well, sorry for the trouble, this seams due to having CONFIG_64bit set but
not CONFIG_PPC64, which is kind of not going to work.

Friendly,

Sven Luther

      reply	other threads:[~2005-12-28 16:52 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-12-28 10:48 ARCH=powerpc kexec seems to be broken Sven Luther
2005-12-28 16:38 ` Sven Luther
2005-12-28 16:51   ` Sven Luther [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20051228165136.GA19559@localhost.localdomain \
    --to=sven.luther@wanadoo.fr \
    --cc=linuxppc-dev@ozlabs.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.