Re: [PATCH 01/11] ARM: topology: mark init_cpu_topology as __init

[Arnd Bergmann]

On Wednesday 08 August 2012, Stephen Boyd wrote:
> 
> On 08/08/12 07:47, Arnd Bergmann wrote:
> > The init_cpu_topology function can be put into the __init section
> > and discarded after boot, because it is only called from
> > smp_prepare_cpus, which is also marked __init. This was reported
> > by gcc after Vincent Guittot added the parse_dt_topology function
> > in 339ca09d7ada "ARM: 7463/1: topology: Update cpu_power according to DT
> > information".
> >
> > Without this patch, building kzm9g_defconfig results in:
> >
> > WARNING: vmlinux.o(.text+0xb5a0): Section mismatch in reference from the function init_cpu_topology() to the function .init.text:parse_dt_topology()
> > The function init_cpu_topology() references
> > the function __init parse_dt_topology().
> > This is often because init_cpu_topology lacks a __init
> > annotation or the annotation of parse_dt_topology is wrong.
> >
> > Signed-off-by: Arnd Bergmann <arnd@xxxxxxxx>
> > Cc: Vincent Guittot <vincent.guittot@xxxxxxxxxx>
> > Cc: Namhyung Kim <namhyung@xxxxxxxxxx>
> > Cc: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>
> > Cc: Russell King <rmk+kernel@xxxxxxxxxxxxxxxx>
> > ---
> 
> This is already in the patch tracker
> 
> http://www.arm.linux.org.uk/developer/patches/viewpatch.php?id=7482/1

Ok, dropped here.

> I also notice that store_cpu_topology() should be marked __cpuinit and
> update_siblings_mask, middle_capacity, cpu_capacity, and
> update_cpu_power should be static. We should probably also rename
> update_cpu_power() so as to not conflict with the one in kernel/sched/fair.c

Yes, all good points. 

	Arnd
--
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/