Unresolved symbols

Unresolved symbols

[David Kesselring]

I've been unabled to track down these errors. I think it's because I don't
understand how some of the linux h files are used by an independently
compiled kernel module. Why is "extern __inline__" used in a file like
atomic.h.
If any of you have any pointers on the following errors, I'd appreciate
your comments.
/lib/modules/2.4.22/kernel/net/pcifvnet.o: unresolved symbol
atomic_sub_return
/lib/modules/2.4.22/kernel/net/pcifvnet.o: unresolved symbol __udelay
/lib/modules/2.4.22/kernel/net/pcifvnet.o: unresolved symbol atomic_add
/lib/modules/2.4.22/kernel/net/pcifvnet.o: unresolved symbol strcpy
/lib/modules/2.4.22/kernel/net/pcifvnet.o: unresolved symbol atomic_sub

Thanks,

David Kesselring
Atmel MMC
dkesselr@mmc.atmel.com
919-462-6587

Re: Unresolved symbols

[David Daney]

David Kesselring wrote:

>I've been unabled to track down these errors. I think it's because I don't
>understand how some of the linux h files are used by an independently
>compiled kernel module. Why is "extern __inline__" used in a file like
>atomic.h.
>If any of you have any pointers on the following errors, I'd appreciate
>your comments.
>/lib/modules/2.4.22/kernel/net/pcifvnet.o: unresolved symbol
>atomic_sub_return
>/lib/modules/2.4.22/kernel/net/pcifvnet.o: unresolved symbol __udelay
>/lib/modules/2.4.22/kernel/net/pcifvnet.o: unresolved symbol atomic_add
>/lib/modules/2.4.22/kernel/net/pcifvnet.o: unresolved symbol strcpy
>/lib/modules/2.4.22/kernel/net/pcifvnet.o: unresolved symbol atomic_sub
>  
>
Often if you don't compile with optimization turned on you get things 
like this.  You have to compile with optimization set high enough to get 
inlining enabled.

David Daney.

Re: Unresolved symbols

[Jan-Benedict Glaw]

[-- Attachment #1: Type: text/plain, Size: 1525 bytes --]

On Tue, 2003-10-28 10:31:59 -0800, David Daney <ddaney@avtrex.com>
wrote in message <3F9EB61F.8010906@avtrex.com>:
> David Kesselring wrote:
> 
> >I've been unabled to track down these errors. I think it's because I don't
> >understand how some of the linux h files are used by an independently
> >compiled kernel module. Why is "extern __inline__" used in a file like
> >atomic.h.
> >If any of you have any pointers on the following errors, I'd appreciate
> >your comments.
> >/lib/modules/2.4.22/kernel/net/pcifvnet.o: unresolved symbol
> >atomic_sub_return
> >/lib/modules/2.4.22/kernel/net/pcifvnet.o: unresolved symbol __udelay
> >/lib/modules/2.4.22/kernel/net/pcifvnet.o: unresolved symbol atomic_add
> >/lib/modules/2.4.22/kernel/net/pcifvnet.o: unresolved symbol strcpy
> >/lib/modules/2.4.22/kernel/net/pcifvnet.o: unresolved symbol atomic_sub
> Often if you don't compile with optimization turned on you get things 
> like this.  You have to compile with optimization set high enough to get 
> inlining enabled.

Right. Once again, if you compile modules outside the kernel, please:

	Use exactly the GCC parameters that are used by in-kernel-tree
			compiled modules!

MfG, JBG

-- 
   Jan-Benedict Glaw       jbglaw@lug-owl.de    . +49-172-7608481
   "Eine Freie Meinung in  einem Freien Kopf    | Gegen Zensur | Gegen Krieg
    fuer einen Freien Staat voll Freier Bürger" | im Internet! |   im Irak!
   ret = do_actions((curr | FREE_SPEECH) & ~(NEW_COPYRIGHT_LAW | DRM | TCPA));

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Unresolved symbols

[Geert Uytterhoeven]

On Tue, 28 Oct 2003, David Kesselring wrote:
> I've been unabled to track down these errors. I think it's because I don't
> understand how some of the linux h files are used by an independently
> compiled kernel module. Why is "extern __inline__" used in a file like
> atomic.h.

`extern inline' is wrong, and should be replaced by `static inline', which is
work-in-progress.

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Re: Unresolved symbols

[Ralf Baechle]

On Tue, Oct 28, 2003 at 09:36:05PM +0100, Geert Uytterhoeven wrote:
> Date: Tue, 28 Oct 2003 21:36:05 +0100 (MET)
> From: Geert Uytterhoeven <geert@linux-m68k.org>
> To: David Kesselring <dkesselr@mmc.atmel.com>
> cc: Linux/MIPS Development <linux-mips@linux-mips.org>
> Subject: Re: Unresolved symbols
> Content-Type: TEXT/PLAIN; charset=US-ASCII
> 
> On Tue, 28 Oct 2003, David Kesselring wrote:
> > I've been unabled to track down these errors. I think it's because I don't
> > understand how some of the linux h files are used by an independently
> > compiled kernel module. Why is "extern __inline__" used in a file like
> > atomic.h.
> 
> `extern inline' is wrong, and should be replaced by `static inline', which is
> work-in-progress.

I've replaced all extern inline in cvs 2.4 / 2.6.

  Ralf