Re: 2.6.16-rc1-mm4

[Keith Owens]

Chuck Ebbert (on Tue, 31 Jan 2006 21:44:20 -0500) wrote:
>In-Reply-To: <20060129144533.128af741.akpm@xxxxxxxx>
>
> $ perl scripts/reference_init.pl | grep smp_locks
> Error: ./arch/i386/kernel/alternative.o .smp_locks refers to 00000008 R_386_32          .init.text
> ...
>
>Caused by x86_smp_alternatives.patch
>
>Does this mean that the SMP lock-switching could write all over discarded
>__init code?

Looking at the patch, it builds tables that can refer to .init.text but
then it excludes table entries that do not fall between _text and
_etext.  Which makes the reference from .smp_locks to .init.text a
false positive.  Gerd, is that the way that smp_alternatives is meant
to work?  If so, I will update reference_*.pl.

+		alternatives_smp_module_add(NULL, "core kernel",
+					    __smp_locks, __smp_locks_end,
+					    _text, _etext);

-
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/