Prev: [PATCH] percpu: remove address_space attribute if !SMP
Next: exit.c: support larger exit code
From: Andrew Benton on 6 Aug 2010 08:50
Hello World
I recently upgraded to the latest version of make, 3.82. One of the
problems I've had with it is that I can no longer recompile my kernel
with built in firmware enabled. If I have
CONFIG_FIRMWARE_IN_KERNEL=y
then the build fails like so:

IKCFG kernel/config_data.h
CC kernel/configs.o
LD kernel/built-in.o
make[2]: *** No rule to make target `/tmp/linux-build-dir/firmware/radeon/', needed by `firmware/radeon/R100_cp.bin.gen.S'. Stop.
make[1]: *** [firmware] Error 2
make: *** [sub-make] Error 2
andy:~/save/src/linux-2.6$

Please CC, I'm not subscribed

Many thanks,

Andy
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Sam Ravnborg on 6 Aug 2010 09:50
On Fri, Aug 06, 2010 at 01:45:09PM +0100, Andrew Benton wrote:
> Hello World
> I recently upgraded to the latest version of make, 3.82. One of the
> problems I've had with it is that I can no longer recompile my kernel
> with built in firmware enabled. If I have
> CONFIG_FIRMWARE_IN_KERNEL=y
> then the build fails like so:
>
> IKCFG kernel/config_data.h
> CC kernel/configs.o
> LD kernel/built-in.o
> make[2]: *** No rule to make target `/tmp/linux-build-dir/firmware/radeon/', needed by `firmware/radeon/R100_cp.bin.gen.S'. Stop.
> make[1]: *** [firmware] Error 2
> make: *** [sub-make] Error 2
> andy:~/save/src/linux-2.6$

Thanks for the report.
I saw a similar report whare make 3.82 was accused - maybe this was also from you.

Can you positively confirm that this is a make 3.82 issue - or do you just expect so?

To narrow down this a bit further can I ask you
to send the output of "make -j1 V=1" - especially
the lines before you see the bug.

[-j1 is to force you to avoid parallel makes
as this may clutter up the output]

Please also post a copy of your configuration
as this allows us to better reproduce what you see.

> Please CC, I'm not subscribed
No problem - we always do so at lkml.

Sam
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
From: Andrew Benton on 6 Aug 2010 10:10
On 06/08/10 14:46, Sam Ravnborg wrote:
> Thanks for the report.
> I saw a similar report whare make 3.82 was accused - maybe this was also from you.
>
> Can you positively confirm that this is a make 3.82 issue - or do you just expect so?
>
I'm certain that it relates to make-3.82. I've installed make-3.81 just so I can recompile my kernel. If I use:
make-3.81 O=/tmp/linux-build-dir -j4
then the build succeeds.

> To narrow down this a bit further can I ask you
> to send the output of "make -j1 V=1" - especially
> the lines before you see the bug.

make -j1 V=1 O=/tmp/linux-build-dir

.... snip ...
LD sound/usb/caiaq/built-in.o
LD sound/usb/misc/built-in.o
LD sound/usb/usx2y/built-in.o
LD sound/usb/built-in.o
CC sound/last.o
LD sound/built-in.o
MK_FW firmware/rt2561s.bin.gen.S
AS firmware/rt2561s.bin.gen.o
make[2]: *** No rule to make target `/tmp/linux-build-dir/firmware/radeon/', needed by `firmware/radeon/R700_rlc.bin.gen.S'. Stop.
make[1]: *** [firmware] Error 2
make: *** [sub-make] Error 2
andy:~$

> Please also post a copy of your configuration
> as this allows us to better reproduce what you see.
>
I've attached /proc/config.gz

Andy
 | 
Pages: 1
Prev: [PATCH] percpu: remove address_space attribute if !SMP
Next: exit.c: support larger exit code