Very Computer

by **Young4er** » Thu, 08 Nov 2001 02:41:36

Hi,

Has anyone encoutered the following error messages:

#####
drivers/block/block.o: In function `lo_send':
drivers/block/block.o(.text+0x21ab): undefined reference to
`deactivate_page'
drivers/block/block.o(.text+0x21e7): undefined reference to
`deactivate_page'
make: *** [vmlinux] Error 1
#####

when trying to compile the latest linux-2.4.14 kernel? I closer look at
the linux/drivers/block/loop.c file shows some calls to the
`deactivate_page' function but such a function is never provided. I could
compile the kernel without the "loop" device.

by **Michael Lee Yoh** » Thu, 08 Nov 2001 04:57:26

Quote:> when trying to compile the latest linux-2.4.14 kernel? I closer look at

You did a "make mrproper" first, right?

--

Software Developer, Engineering Services
Red Hat, Inc.

QUIPd 0.20: (364 of 533)
-> I hate quotations.
-> - Ralph Waldo Emerson

by **D. Stimit** » Thu, 08 Nov 2001 06:24:31

> Hi,

> Has anyone encoutered the following error messages:

> #####
> drivers/block/block.o: In function `lo_send':
> drivers/block/block.o(.text+0x21ab): undefined reference to
> `deactivate_page'
> drivers/block/block.o(.text+0x21e7): undefined reference to
> `deactivate_page'
> make: *** [vmlinux] Error 1
> #####

> when trying to compile the latest linux-2.4.14 kernel? I closer look at
> the linux/drivers/block/loop.c file shows some calls to the
> `deactivate_page' function but such a function is never provided. I could
> compile the kernel without the "loop" device.

Yes, it has flooded the kernel devel list. The function deactivate_page
is old and no longer used. You can completely remove it from the
relevant file in 2.4.14+.

by **Michael Lee Yoh** » Thu, 08 Nov 2001 06:34:34

Quote:> Yes, it has flooded the kernel devel list. The function deactivate_page
> is old and no longer used. You can completely remove it from the
> relevant file in 2.4.14+.

Does anyone else think that the easiest test for an out-the-door kernel
is to compile everything into modular form per platform? It's amazing
that this is the second time a "release" kernel has been sent out the
door unstable. What's worse is that the loopback driver (loop.c) is
quite common in the development world, and highly useful for ISO
manipulation.

Why aren't these problems being caught ahead of time? Isn't it just as
easy to compile everything in a kernel to see if each module with
compile without error?

--

Software Developer, Engineering Services
Red Hat, Inc.

QUIPd 0.20: (364 of 533)
-> I hate quotations.
-> - Ralph Waldo Emerson

by **Michael Fauro** » Thu, 08 Nov 2001 07:00:48

: #####
: drivers/block/block.o: In function `lo_send':
: drivers/block/block.o(.text+0x21ab): undefined reference to
: `deactivate_page'
: drivers/block/block.o(.text+0x21e7): undefined reference to
: `deactivate_page'
: make: *** [vmlinux] Error 1
: #####

: when trying to compile the latest linux-2.4.14 kernel?

The kernel compiled okay for me with the loop drive as a module. But it
won't load:

# 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.

by **Vladimir Annenko** » Thu, 08 Nov 2001 06:54:11

It's been confirmed. Edit the file drivers/block/loop.c and comment out
the two lines with "deactivate_page" in them.

> Hi,

> Has anyone encoutered the following error messages:

> #####
> drivers/block/block.o: In function `lo_send':
> drivers/block/block.o(.text+0x21ab): undefined reference to
> `deactivate_page'
> drivers/block/block.o(.text+0x21e7): undefined reference to
> `deactivate_page'
> make: *** [vmlinux] Error 1
> #####

> when trying to compile the latest linux-2.4.14 kernel? I closer look at
> the linux/drivers/block/loop.c file shows some calls to the
> `deactivate_page' function but such a function is never provided. I could
> compile the kernel without the "loop" device.

by **Paul Kimot** » Thu, 08 Nov 2001 08:33:15

> Does anyone else think that the easiest test for an out-the-door kernel
> is to compile everything into modular form per platform? It's amazing
> that this is the second time a "release" kernel has been sent out the
> door unstable. [...]
> Why aren't these problems being caught ahead of time? Isn't it just as
> easy to compile everything in a kernel to see if each module with
> compile without error?

Does Linus Torvalds use modules yet? He didn't used to.

--
Paul Kimoto
This message was originally posted on Usenet in plain text. Any images,
hyperlinks, or the like shown here have been added without my consent,
and may be a violation of international copyright law.

by **D. Stimit** » Thu, 08 Nov 2001 18:57:36

> > Does anyone else think that the easiest test for an out-the-door kernel
> > is to compile everything into modular form per platform? It's amazing
> > that this is the second time a "release" kernel has been sent out the
> > door unstable. [...]
> > Why aren't these problems being caught ahead of time? Isn't it just as
> > easy to compile everything in a kernel to see if each module with
> > compile without error?

> Does Linus Torvalds use modules yet? He didn't used to.

An interesting question. I've also heard that at one time he didn't use
modules, but I have no idea if that is just old rumor or if it really
has some substance. It'd be interesting to know what his views are
concerning use of modules or not.

Quote:
> --
> Paul Kimoto
> This message was originally posted on Usenet in plain text. Any images,
> hyperlinks, or the like shown here have been added without my consent,
> and may be a violation of international copyright law.

by **Michael Lee Yoh** » Fri, 09 Nov 2001 02:40:46

Quote:>>Does Linus Torvalds use modules yet? He didn't used to.

He does monitor this NG from time to time - may he'd care to enlighten us :)

Quote:> An interesting question. I've also heard that at one time he didn't use
> modules, but I have no idea if that is just old rumor or if it really
> has some substance. It'd be interesting to know what his views are
> concerning use of modules or not.

Back in the 2.2, I tended to avoid modules because of the whole module
system. Extra overhead to plugin the modules n'such. On most of my
systems nowadays, I can't afford not to use modules - the kernel would
be too big.

--

Software Developer, Engineering Services
Red Hat, Inc.

QUIPd 1.00: (201 of 533)
-> I have opinions of my own - strong opinions - but I don't always
-> agree with them.
-> - George Bush

Very Computer

linux-2.4.14 vs. drivers/block/loop.c

linux-2.4.14 vs. drivers/block/loop.c

linux-2.4.14 vs. drivers/block/loop.c

linux-2.4.14 vs. drivers/block/loop.c

linux-2.4.14 vs. drivers/block/loop.c

linux-2.4.14 vs. drivers/block/loop.c

linux-2.4.14 vs. drivers/block/loop.c

linux-2.4.14 vs. drivers/block/loop.c

linux-2.4.14 vs. drivers/block/loop.c

linux-2.4.14 vs. drivers/block/loop.c