Re: linux-next: kbuild tree build failure

From: Sam Ravnborg
Date: Wed Oct 14 2009 - 03:45:12 EST


On Wed, Oct 14, 2009 at 12:20:24PM +1100, Stephen Rothwell wrote:
> Hi Sam,
>
> Today's linux-next build (x86_64 allmodconfig) failed like this:
>
> In file included from arch/x86/kernel/syscall_64.c:28:
> arch/x86/include/asm/unistd_64.h:665: error: array index in initializer exceeds array bounds
> arch/x86/include/asm/unistd_64.h:665: error: (near initialization for 'sys_call_table')
> arch/x86/include/asm/unistd_64.h:665: warning: excess elements in array initializer
> arch/x86/include/asm/unistd_64.h:665: warning: (near initialization for 'sys_call_table')
>
> Exposed by commit a2e2725541fad72416326798c2d7fa4dafb7d337 ("net:
> Introduce recvmmsg socket syscall"). include/asm/asm-offsets.h did not
> get updated even though arch/x86/kernel/asm-offsets.s did. Ah,
> generated/asm-offsets.h has been updated but my build tree still has the
> (generated in a previous build) include/asm/asm-offsets.h which is
> presumably found before the new arch/x86/include/asm/asm-offsets.h
> introduced by commit fe5ff47a5d79ca99ec7e3155eb19469b44905b2c ("kbuild:
> move asm-offsets.h to include/generated").
>
> Sam, can you think of a better migration path?

I once thought of forefully removing include/asm to avoid this.
Maybe we could let "make clean" remove include/asm -
I expect this would have helped in this situation?

I'm planning to redo the serie within the next couple of weeks anyway.

Sam
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/