FAQ Search Today's Posts Mark Forums Read
» Video Reviews

» Linux Archive

Linux-archive is a website aiming to archive linux email lists and to make them easily accessible for linux users/developers.


» Sponsor

» Partners

» Sponsor

Go Back   Linux Archive > Debian > Debian Kernel

 
 
LinkBack Thread Tools
 
Old 09-11-2008, 12:56 PM
Stephen R Marenka
 
Default 2.6.26-5 vs m68k atari abi

It looks like the CONFIG_LBD=n change caused an abi change on m68k
atari. (Seems obvious now, but I didn't run into this problem with
the build test.)

So now the kernel ftbfs on m68k atari with the following.

| Removed symbols:
| gfs2_mount_lockproto module: fs/gfs2/gfs2, version: 0x914aceea, export: EXPORT_SYMBOL_GPL
| gfs2_register_lockproto module: fs/gfs2/gfs2, version: 0x1929be86, export: EXPORT_SYMBOL_GPL
| gfs2_unmount_lockproto module: fs/gfs2/gfs2, version: 0xd1baae74, export: EXPORT_SYMBOL_GPL
| gfs2_unregister_lockproto module: fs/gfs2/gfs2, version: 0x24aada0c, export: EXPORT_SYMBOL_GPL
| gfs2_withdraw_lockproto module: fs/gfs2/gfs2, version: 0x69b66631, export: EXPORT_SYMBOL_GPL

How do I fix this?

Thanks,

Stephen

--
Stephen R. Marenka If life's not fun, you're not doing it right!
<stephen@marenka.net>
 
Old 09-11-2008, 01:45 PM
maximilian attems
 
Default 2.6.26-5 vs m68k atari abi

On Thu, Sep 11, 2008 at 07:56:21AM -0500, Stephen R Marenka wrote:
> It looks like the CONFIG_LBD=n change caused an abi change on m68k
> atari. (Seems obvious now, but I didn't run into this problem with
> the build test.)
>
> So now the kernel ftbfs on m68k atari with the following.
>
> | Removed symbols:
> | gfs2_mount_lockproto module: fs/gfs2/gfs2, version: 0x914aceea, export: EXPORT_SYMBOL_GPL
> | gfs2_register_lockproto module: fs/gfs2/gfs2, version: 0x1929be86, export: EXPORT_SYMBOL_GPL
> | gfs2_unmount_lockproto module: fs/gfs2/gfs2, version: 0xd1baae74, export: EXPORT_SYMBOL_GPL
> | gfs2_unregister_lockproto module: fs/gfs2/gfs2, version: 0x24aada0c, export: EXPORT_SYMBOL_GPL
> | gfs2_withdraw_lockproto module: fs/gfs2/gfs2, version: 0x69b66631, export: EXPORT_SYMBOL_GPL
>
> How do I fix this?

add an ignore gfs2_* pattern as i doubt that this in high use on m68k.

--
maks


--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 09-11-2008, 06:02 PM
Stephen R Marenka
 
Default 2.6.26-5 vs m68k atari abi

On Thu, Sep 11, 2008 at 03:45:13PM +0200, maximilian attems wrote:
> On Thu, Sep 11, 2008 at 07:56:21AM -0500, Stephen R Marenka wrote:
> > It looks like the CONFIG_LBD=n change caused an abi change on m68k
> > atari. (Seems obvious now, but I didn't run into this problem with
> > the build test.)

> > How do I fix this?
>
> add an ignore gfs2_* pattern as i doubt that this in high use on m68k.

I agree, where does this pattern go?

Thanks,

Stephen

--
Stephen R. Marenka If life's not fun, you're not doing it right!
<stephen@marenka.net>
 
Old 09-11-2008, 06:36 PM
dann frazier
 
Default 2.6.26-5 vs m68k atari abi

On Thu, Sep 11, 2008 at 01:02:01PM -0500, Stephen R Marenka wrote:
> On Thu, Sep 11, 2008 at 03:45:13PM +0200, maximilian attems wrote:
> > On Thu, Sep 11, 2008 at 07:56:21AM -0500, Stephen R Marenka wrote:
> > > It looks like the CONFIG_LBD=n change caused an abi change on m68k
> > > atari. (Seems obvious now, but I didn't run into this problem with
> > > the build test.)
>
> > > How do I fix this?
> >
> > add an ignore gfs2_* pattern as i doubt that this in high use on m68k.
>
> I agree, where does this pattern go?

see config/hppa/defines for an example

--
dann frazier


--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 09-12-2008, 03:48 PM
Stephen R Marenka
 
Default 2.6.26-5 vs m68k atari abi

On Thu, Sep 11, 2008 at 12:36:17PM -0600, dann frazier wrote:
> On Thu, Sep 11, 2008 at 01:02:01PM -0500, Stephen R Marenka wrote:
> > On Thu, Sep 11, 2008 at 03:45:13PM +0200, maximilian attems wrote:
> > > On Thu, Sep 11, 2008 at 07:56:21AM -0500, Stephen R Marenka wrote:
> > > > It looks like the CONFIG_LBD=n change caused an abi change on m68k
> > > > atari. (Seems obvious now, but I didn't run into this problem with
> > > > the build test.)
> >
> > > > How do I fix this?
> > >
> > > add an ignore gfs2_* pattern as i doubt that this in high use on m68k.
> >
> > I agree, where does this pattern go?
>
> see config/hppa/defines for an example

Thanks, now I understand.

Unfortunately, it looks like a number of other symbols were changed too.
I don't suppose I'm allowed to ignore_changes: *?

So how do I proceed? Bump the abi or ignore the differences?

This kernel is not widely distributed yet. Not making the change
causes a double fault in the general use case of the atari kernel
under aranym with debian-installer. (Pretty much the only thing keeping
me from running daily regression tests with d-i.)

Thanks,

Stephen

--
Stephen R. Marenka If life's not fun, you're not doing it right!
<stephen@marenka.net>
 
Old 09-16-2008, 01:04 PM
Stephen R Marenka
 
Default 2.6.26-5 vs m68k atari abi

On Fri, Sep 12, 2008 at 10:48:53AM -0500, Stephen R Marenka wrote:
> On Thu, Sep 11, 2008 at 12:36:17PM -0600, dann frazier wrote:
> > On Thu, Sep 11, 2008 at 01:02:01PM -0500, Stephen R Marenka wrote:
> > > On Thu, Sep 11, 2008 at 03:45:13PM +0200, maximilian attems wrote:
> > > > On Thu, Sep 11, 2008 at 07:56:21AM -0500, Stephen R Marenka wrote:
> > > > > It looks like the CONFIG_LBD=n change caused an abi change on m68k
> > > > > atari. (Seems obvious now, but I didn't run into this problem with
> > > > > the build test.)
> > >
> > > > > How do I fix this?
> > > >
> > > > add an ignore gfs2_* pattern as i doubt that this in high use on m68k.
> > >
> > > I agree, where does this pattern go?
> >
> > see config/hppa/defines for an example
>
> Thanks, now I understand.
>
> Unfortunately, it looks like a number of other symbols were changed too.
> I don't suppose I'm allowed to ignore_changes: *?
>
> So how do I proceed? Bump the abi or ignore the differences?
>
> This kernel is not widely distributed yet. Not making the change
> causes a double fault in the general use case of the atari kernel
> under aranym with debian-installer. (Pretty much the only thing keeping
> me from running daily regression tests with d-i.)

Could someone give me a pointer on how to update the abi for m68k? So
far the only thing I've managed to get to work is "ignore_changes: *",
which I'm guessing someone will fuss about if I commit.

Thanks,

Stephen

--
Stephen R. Marenka If life's not fun, you're not doing it right!
<stephen@marenka.net>
 
Old 09-16-2008, 01:14 PM
maximilian attems
 
Default 2.6.26-5 vs m68k atari abi

On Tue, Sep 16, 2008 at 08:04:21AM -0500, Stephen R Marenka wrote:
> On Fri, Sep 12, 2008 at 10:48:53AM -0500, Stephen R Marenka wrote:
> > On Thu, Sep 11, 2008 at 12:36:17PM -0600, dann frazier wrote:
> > > On Thu, Sep 11, 2008 at 01:02:01PM -0500, Stephen R Marenka wrote:
> > > > On Thu, Sep 11, 2008 at 03:45:13PM +0200, maximilian attems wrote:
> > > > > On Thu, Sep 11, 2008 at 07:56:21AM -0500, Stephen R Marenka wrote:
> > > > > > It looks like the CONFIG_LBD=n change caused an abi change on m68k
> > > > > > atari. (Seems obvious now, but I didn't run into this problem with
> > > > > > the build test.)
> > > >
> > > > > > How do I fix this?
> > > > >
> > > > > add an ignore gfs2_* pattern as i doubt that this in high use on m68k.
> > > >
> > > > I agree, where does this pattern go?
> > >
> > > see config/hppa/defines for an example
> >
> > Thanks, now I understand.
> >
> > Unfortunately, it looks like a number of other symbols were changed too.
> > I don't suppose I'm allowed to ignore_changes: *?
> >
> > So how do I proceed? Bump the abi or ignore the differences?
> >
> > This kernel is not widely distributed yet. Not making the change
> > causes a double fault in the general use case of the atari kernel
> > under aranym with debian-installer. (Pretty much the only thing keeping
> > me from running daily regression tests with d-i.)
>
> Could someone give me a pointer on how to update the abi for m68k? So
> far the only thing I've managed to get to work is "ignore_changes: *",
> which I'm guessing someone will fuss about if I commit.
>
> Thanks,
>
> Stephen

no you are the maintainer, go for it.

--
maks


--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 09-16-2008, 01:34 PM
Bastian Blank
 
Default 2.6.26-5 vs m68k atari abi

On Tue, Sep 16, 2008 at 08:04:21AM -0500, Stephen R Marenka wrote:
> Could someone give me a pointer on how to update the abi for m68k?

debian/bin/abiupdate.py

> So
> far the only thing I've managed to get to work is "ignore_changes: *",
> which I'm guessing someone will fuss about if I commit.

What did you try? But as the change is in a core structure, you need to
use the hammer. (And you have to handle the fallout.)

Bastian

--
If I can have honesty, it's easier to overlook mistakes.
-- Kirk, "Space Seed", stardate 3141.9


--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
 
Old 09-17-2008, 02:57 PM
Stephen R Marenka
 
Default 2.6.26-5 vs m68k atari abi

On Tue, Sep 16, 2008 at 03:34:09PM +0200, Bastian Blank wrote:
> On Tue, Sep 16, 2008 at 08:04:21AM -0500, Stephen R Marenka wrote:
> > Could someone give me a pointer on how to update the abi for m68k?
>
> debian/bin/abiupdate.py
>
> > So
> > far the only thing I've managed to get to work is "ignore_changes: *",
> > which I'm guessing someone will fuss about if I commit.
>
> What did you try? But as the change is in a core structure, you need to
> use the hammer. (And you have to handle the fallout.)

So please correct me where I misunderstand.

* Build -5 with ignore_changes: *.
* Use abiupdate.py to update the stored abi for -6.
* -6 shouldn't need abiupdate.py.

What other fallout should I expect?

Thanks for your patience,

Stephen

--
Stephen R. Marenka If life's not fun, you're not doing it right!
<stephen@marenka.net>
 

Thread Tools




All times are GMT. The time now is 06:47 AM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.
Copyright 2007 - 2008, www.linux-archive.org