2.4.14: drivers/block/block.o: undefined reference to `deactivate_page'

2.4.14: drivers/block/block.o: undefined reference to `deactivate_page'

Post by Jonathan Kame » Thu, 08 Nov 2001 01:26:22



2.4.14 won't compile for me.  It appears that the function
deactivate_page was removed or renamed in 2.4.14, but a reference to it
wasn't removed from drivers/block/loop.c.
 
 
 

2.4.14: drivers/block/block.o: undefined reference to `deactivate_page'

Post by bdonla » Thu, 08 Nov 2001 04:45:14



> 2.4.14 won't compile for me.  It appears that the function
> deactivate_page was removed or renamed in 2.4.14, but a reference to it
> wasn't removed from drivers/block/loop.c.

Go to drivers/block/loop.c
Delete the two deactivate_page(page); lines ( or comment them out ).

--
make me
---
make: *** No rule to make target `me'.  Stop.

 
 
 

2.4.14: drivers/block/block.o: undefined reference to `deactivate_page'

Post by Michael Fauro » Thu, 08 Nov 2001 07:00:41


: 2.4.14 won't compile for me.  It appears that the function
: deactivate_page was removed or renamed in 2.4.14, but a reference to it
: wasn't removed from drivers/block/loop.c.

2.4.14 built okay for me, but I find that I can no longer mount a loop
filesystem any longer.  If I try to load the loop module manually, it
produces the following message:

# insmod loop
Using /lib/modules/2.4.14-1/kernel/drivers/block/loop.o
/lib/modules/2.4.14-1/kernel/drivers/block/loop.o: unresolved symbol deactivate_page

--
------------------------------------------------------------------------------
 Michael | mfaurot  | The groundhog is like most other prophets; it delivers
 Faurot  | atww.org | its message and then disappears.

 
 
 

2.4.14: drivers/block/block.o: undefined reference to `deactivate_page'

Post by Gene Hesket » Thu, 08 Nov 2001 10:42:06


Gene Heskett sends Greetings to Jonathan Kamens;

Quote:>2.4.14 won't compile for me.  It appears that the function
>deactivate_page was removed or renamed in 2.4.14, but a reference to it
>wasn't removed from drivers/block/loop.c.

Thats a bit odd Jonathon, but lenmme see what ythe bottom of my last
attempt says.

I get a failure at the "ld this and that >vmlinux" stage, and its
reported as being in drivers/block/block.o, but thats the same variable
name as is being reported here, deactivate_page.

Cheers, Gene
--
  Gene Heskett, CET, UHK       |Amiga A2k Zeus040, 70MB ram, 31 gigs

             email gene underscore heskett at iolinc dot net
#Amiga based X10 home automation program EZHome, see at:#
               <http://www.iolinc.net/gene_heskett>
This messages reply content, but not any previously quoted material,
is ? 2001 by Gene Heskett, all rights reserved.
--

 
 
 

2.4.14: drivers/block/block.o: undefined reference to `deactivate_page'

Post by Kasper Dupon » Thu, 08 Nov 2001 14:34:25




> > 2.4.14 won't compile for me.  It appears that the function
> > deactivate_page was removed or renamed in 2.4.14, but a reference to it
> > wasn't removed from drivers/block/loop.c.

> Go to drivers/block/loop.c
> Delete the two deactivate_page(page); lines ( or comment them out ).

How can we be sure that removing the calls will not break anything?

--
Kasper Dupont

 
 
 

2.4.14: drivers/block/block.o: undefined reference to `deactivate_page'

Post by Gene Hesket » Thu, 08 Nov 2001 23:05:14


Gene Heskett sends Greetings to Kasper Dupont;



>> > 2.4.14 won't compile for me.  It appears that the function
>> > deactivate_page was removed or renamed in 2.4.14, but a reference to it
>> > wasn't removed from drivers/block/loop.c.

>> Go to drivers/block/loop.c
>> Delete the two deactivate_page(page); lines ( or comment them out ).
>How can we be sure that removing the calls will not break anything?

I don't think we can Kasper.  The differences are in <linux/swap.h>,
huge amounts of rewrite done.  I looked at the possibility of
transplanting a few lines from the 2.4.13-ac6 version of that file but
so many names were changed I'd have no idea where to start.

Temporarily, I suspect it will build if in the 'make xconfig' one were to
turn off the loopback device.  I intend to try that tonight as I just
had 2.4.13-ac6 totally loose track of its cache after I had mounted a
previously unmounted reiserfs partition.  I noticed that the MEM bar of
xosview was full width of its window, and a free reported I had the
magic 4gigs number for a cache buffer.  Obviously, reboot, and again after I
turn the tmpfs stuff on, which accidently got cleared in that config.

Cheers, Gene
--
  Gene Heskett, CET, UHK       |Amiga A2k Zeus040, 70MB ram, 31 gigs

             email gene underscore heskett at iolinc dot net
#Amiga based X10 home automation program EZHome, see at:#
               <http://www.iolinc.net/gene_heskett>
This messages reply content, but not any previously quoted material,
is ? 2001 by Gene Heskett, all rights reserved.
--

 
 
 

2.4.14: drivers/block/block.o: undefined reference to `deactivate_page'

Post by bill davids » Fri, 09 Nov 2001 01:41:20




| : 2.4.14 won't compile for me.  It appears that the function
| : deactivate_page was removed or renamed in 2.4.14, but a reference to it
| : wasn't removed from drivers/block/loop.c.
|
| 2.4.14 built okay for me, but I find that I can no longer mount a loop
| filesystem any longer.  If I try to load the loop module manually, it
| produces the following message:
|
| # insmod loop
| Using /lib/modules/2.4.14-1/kernel/drivers/block/loop.o
| /lib/modules/2.4.14-1/kernel/drivers/block/loop.o: unresolved symbol deactivate_page

  I believe that if you compile the driver in the kernel you get this at
build time. Same problem, just bites you a bit later. I would have
expected depmod to notice this, but I don't really have a list of what
it finds and doesn't. Or perhaps it did and you didn't notice when you
installed the modules?

--

  His first management concern is not solving the problem, but covering
his ass. If he lived in the middle ages he'd wear his codpiece backward.

 
 
 

2.4.14: drivers/block/block.o: undefined reference to `deactivate_page'

Post by Ed Clar » Sat, 10 Nov 2001 19:50:39



Quote:>Temporarily, I suspect it will build if in the 'make xconfig' one were to
>turn off the loopback device.  I intend to try that tonight as I just
>had 2.4.13-ac6 totally loose track of its cache after I had mounted a
>previously unmounted reiserfs partition.  I noticed that the MEM bar of
>xosview was full width of its window, and a free reported I had the
>magic 4gigs number for a cache buffer.  Obviously, reboot, and again after I
>turn the tmpfs stuff on, which accidently got cleared in that config.

How do you run mkinitrd if you don't have a loop device?  Isn't 2.4.x
supposed to be a stable release tree?  What's going on with major
rewrite action in a stable release?
 
 
 

2.4.14: drivers/block/block.o: undefined reference to `deactivate_page'

Post by Bernd Selbac » Sat, 10 Nov 2001 20:05:31


<snip>
Quote:> >> Go to drivers/block/loop.c
> >> Delete the two deactivate_page(page); lines ( or comment them out ).

> >How can we be sure that removing the calls will not break anything?

<snip>

I think we don't have to remove anything, the changes are in
linux/mm/swap.c. The
function activate_page is still there, but the reverse function
deactivate_page was
removed from this source. The code for activate_page hasn't change, so
when the
original code fro deactivate_page is inserted in swap.c (i've copied
this piece of code
from the 2.4.13 source-tree), the compiler returns no errors and also
loopback mounts are working.

Bernd

 
 
 

2.4.14: drivers/block/block.o: undefined reference to `deactivate_page'

Post by Gene Hesket » Sun, 11 Nov 2001 11:16:27


Gene Heskett sends Greetings to Ed Clarke;


>>Temporarily, I suspect it will build if in the 'make xconfig' one were to
>>turn off the loopback device.  I intend to try that tonight as I just
>>had 2.4.13-ac6 totally loose track of its cache after I had mounted a
>>previously unmounted reiserfs partition.  I noticed that the MEM bar of
>>xosview was full width of its window, and a free reported I had the
>>magic 4gigs number for a cache buffer.  Obviously, reboot, and again after I
>>turn the tmpfs stuff on, which accidently got cleared in that config.
>How do you run mkinitrd if you don't have a loop device?  Isn't 2.4.x
>supposed to be a stable release tree?  What's going on with major
>rewrite action in a stable release?

Well, for what its worth, this is almost the only thing fixed in
2.4.15-pre1, which I put in yesterday morning with zip effort, and my
loopback.device is once again among the living.

As to what happened to 2.4.13-ac6 that caused it to suddenly think I
had -4.127etc gigs of memory, I have no idea.

Cheers, Gene
--
  Gene Heskett, CET, UHK       |Amiga A2k Zeus040, 70MB ram, 31 gigs

             email gene underscore heskett at iolinc dot net
#Amiga based X10 home automation program EZHome, see at:#
               <http://www.iolinc.net/gene_heskett>
This messages reply content, but not any previously quoted material,
is ? 2001 by Gene Heskett, all rights reserved.
--

 
 
 

2.4.14: drivers/block/block.o: undefined reference to `deactivate_page'

Post by Young4er » Thu, 15 Nov 2001 03:09:18


On Friday 09 November 2001 06:05 am, `Bernd Selbach' of


> <snip>
>> >> Go to drivers/block/loop.c
>> >> Delete the two deactivate_page(page); lines ( or comment
>> >> them out ).

>> >How can we be sure that removing the calls will not break
>> >anything?

> <snip>

> I think we don't have to remove anything, the changes are in
> linux/mm/swap.c. The
> function activate_page is still there, but the reverse function
> deactivate_page was
> removed from this source. The code for activate_page hasn't
> change, so when the
> original code fro deactivate_page is inserted in swap.c (i've
> copied this piece of code
> from the 2.4.13 source-tree), the compiler returns no errors
> and also loopback mounts are working.

> Bernd

I just removed the deactive_page call and now my computer is
running perfectly.  I could mount any ISO on any /dev/loopX
device without any problem.