Fwd: Re: 2.4.19-rc1-jam2 (-rc1aa2)

Fwd: Re: 2.4.19-rc1-jam2 (-rc1aa2)

[Dieter Nützel]

Hello Chris,

symbol clash between latest -AA (-jam2, 00-aa-rc1aa2) kernel and your 
data-logging stuff ;-(

Regards,
	Dieter

----------  Forwarded Message  ----------

Subject: Re: 2.4.19-rc1-jam2 (-rc1aa2)
Date: Thu, 11 Jul 2002 02:26:03 +0200
From: Dieter Nützel <Dieter.Nuetzel@hamburg.de>
To: "J.A. Magallon" <jamagallon@able.es>
Cc: Andrea Arcangeli <andrea@suse.de>

On Thursday 11 July 2002 02:01, J.A. Magallon wrote:
> On 2002.07.11 Dieter Nützel wrote:
> >fs/fs.o(__ksymtab+0x10): multiple definition of `__ksymtab_balance_dirty'
> >kernel/kernel.o(__ksymtab+0x870): first defined here
> >fs/fs.o(.kstrtab+0x47): multiple definition of `__kstrtab_balance_dirty'
> >kernel/kernel.o(.kstrtab+0x23c7): first defined here

[-]

> I think your gcc is making one other definition of balance_dirty from this:
>
> kernel/ksyms.c:EXPORT_SYMBOL(balance_dirty);
>
> gcc, binutils version ?

Reading specs from /usr/lib/gcc-lib/i486-suse-linux/2.95.3/specs
gcc version 2.95.3 20010315 (SuSE)

GNU ld version 2.11.90.0.29 (with BFD 2.11.90.0.29)

> Are you building with someting like -fnocommon ?

Not that I know.

Same flags as with 2.4.19rc1aa1.

I use latest ReiserFS stuff, but same code works fine with 2.4.19rc1aa1.
Even Page Coloring works.
Only new stuff is -aa2 (XFS, not enabled, I dropped it before 'cause I do not
use it) and jam2.

SunWave1 src/linux# grep -r balance_dirty kernel fs
kernel/ksyms.c:EXPORT_SYMBOL(balance_dirty);
Übereinstimmungen in Binärdatei kernel/ksyms.o.
Übereinstimmungen in Binärdatei kernel/kernel.o.
fs/xfs/pagebuf/page_buf_io.c:   balance_dirty();
fs/xfs/pagebuf/page_buf_io.c:   int need_balance_dirty = 0;
fs/xfs/pagebuf/page_buf_io.c:           need_balance_dirty = 1;
fs/xfs/pagebuf/page_buf_io.c:   if (need_balance_dirty) {
fs/xfs/pagebuf/page_buf_io.c:                   balance_dirty();
Übereinstimmungen in Binärdatei fs/fs.o.
Übereinstimmungen in Binärdatei fs/reiserfs/reiserfs.o.
fs/reiserfs/inode.c:    int need_balance_dirty = 0 ;
fs/reiserfs/inode.c:                need_balance_dirty = 1;
fs/reiserfs/inode.c:    if (need_balance_dirty) {
fs/reiserfs/inode.c:        balance_dirty() ;
fs/reiserfs/inode.c:    int partial = 0, need_balance_dirty = 0;
fs/reiserfs/inode.c:            need_balance_dirty = 1;
fs/reiserfs/inode.c:    if (need_balance_dirty)
fs/reiserfs/inode.c:    balance_dirty();
Übereinstimmungen in Binärdatei fs/reiserfs/inode.o.
fs/reiserfs/do_balan.c:     tb->need_balance_dirty = 1;
fs/reiserfs/do_balan.c:    tb->need_balance_dirty = 0;
fs/buffer.c:static int balance_dirty_state(void)
fs/buffer.c:void balance_dirty(void)
fs/buffer.c:    int state = balance_dirty_state();
fs/buffer.c:EXPORT_SYMBOL(balance_dirty);
fs/buffer.c:/* atomic version, the user must call balance_dirty() by hand
fs/buffer.c:            balance_dirty();
fs/buffer.c:    int partial = 0, need_balance_dirty = 0;
fs/buffer.c:                            need_balance_dirty = 1;
fs/buffer.c:    if (need_balance_dirty)
fs/buffer.c:            balance_dirty();
fs/buffer.c:    if (balance_dirty_state() >= 0)
Übereinstimmungen in Binärdatei fs/buffer.o.

Thanks and good night.

-Dieter

-------------------------------------------------------

-- 
Dieter Nützel
Graduate Student, Computer Science

University of Hamburg
Department of Computer Science
@home: Dieter.Nuetzel@hamburg.de

Re: Fwd: Re: 2.4.19-rc1-jam2 (-rc1aa2)

[Chris Mason]

On Wed, 2002-07-10 at 20:35, Dieter Nützel wrote:
> Hello Chris,
> 
> symbol clash between latest -AA (-jam2, 00-aa-rc1aa2) kernel and your 
> data-logging stuff ;-(

You should be able to drop the EXPORT_SYMBOL(balance_dirty); from my
patch.

-chris

Re: Re: Fwd: Re: 2.4.19-rc1-jam2 (-rc1aa2)

[Manuel Krause]

On 07/11/2002 03:26 AM, Chris Mason wrote:
> On Wed, 2002-07-10 at 20:35, Dieter Nützel wrote:
> 
>>Hello Chris,
>>
>>symbol clash between latest -AA (-jam2, 00-aa-rc1aa2) kernel and your 
>>data-logging stuff ;-(
> 
> 
> You should be able to drop the EXPORT_SYMBOL(balance_dirty); from my
> patch.
> 
> -chris
> 

;-)

Are your patches going to follow 2.4.19/2.4.20-preX+namesys - once 
2.4.19/2.4.20-pre is finally out

(...me, just praying for this circumstance, also for namesys' hidden 
patches on this next kernel!...)  ?

:-))

I have no problems with: 2.4.19-rc1 + reiserfs-patches[01,02,03,05] + 
all your patches (the data-logging one is now -11, checked regression 
from -13) + rml-preempt-kernel.


Manuel

Re: Re: Fwd: Re: 2.4.19-rc1-jam2 (-rc1aa2)

[Oleg Drokin]

Hello!

On Thu, Jul 11, 2002 at 04:23:01AM +0200, Manuel Krause wrote:

> Are your patches going to follow 2.4.19/2.4.20-preX+namesys - once 
> 2.4.19/2.4.20-pre is finally out

We hope to get these in time for 2.4.20.

> (...me, just praying for this circumstance, also for namesys' hidden 
> patches on this next kernel!...)  ?

And these too.

Bye,
    Oleg

Re: Re: Fwd: Re: 2.4.19-rc1-jam2 (-rc1aa2)

[Chris Mason]

On Thu, 2002-07-11 at 03:22, Oleg Drokin wrote:
> Hello!
> 
> On Thu, Jul 11, 2002 at 04:23:01AM +0200, Manuel Krause wrote:
> 
> > Are your patches going to follow 2.4.19/2.4.20-preX+namesys - once 
> > 2.4.19/2.4.20-pre is finally out
> 
> We hope to get these in time for 2.4.20.

The data logging code will follow the pure kernel, and will include
whatever other bits I need (as separate patches) to make things work. 
Once the big chunk of namesys stuff gets merged, I'll rework my patches
on top of them.

-chris