Re: [PATCH 2/2] irq: only update affinity in chip set_affinity()

[Ingo Molnar <mingo@elte.hu>]

* Yinghai Lu <yhlu.kernel@gmail.com> wrote:

> On Wed, Apr 8, 2009 at 5:54 AM, Ingo Molnar <mingo@elte.hu> wrote:
> >
> > * Yinghai Lu <yinghai@kernel.org> wrote:
> >
> >> Impact: keep affinity consistent
> >>
> >> irq_set_affinity() and move_masked_irq() try to assign affinity
> >> before calling chip set_affinity(). some archs are assigning again
> >> in set_affinity again.
> >>
> >> something like:
> >> cpumask_cpy(desc->affinity, mask);
> >> desc->chip->set_affinity(mask);
> >>
> >> in the failing path, affinity should not be touched.
> >>
> >> also set_extra_move_desc() ( called by set_affinity) will rely on
> >> the old affinity to decide if need to move irq_desc to different
> >> node when logical flat apic mode is used.
> >>
> >> So try remove those assignment, and make some missed arch to
> >> assign affinity in their set_affinity.
> >>
> >> Signed-off-by: Yinghai Lu <yinghai@kernel.org>
> >>
> >> --
> >>  arch/alpha/kernel/sys_dp264.c         |    6 ++++--
> >>  arch/alpha/kernel/sys_titan.c         |    3 ++-
> >>  arch/arm/common/gic.c                 |    1 +
> >>  arch/cris/arch-v32/kernel/irq.c       |    1 +
> >>  arch/ia64/kernel/iosapic.c            |    3 +++
> >>  arch/ia64/sn/kernel/irq.c             |    3 +++
> >>  arch/mips/cavium-octeon/octeon-irq.c  |    6 ++++++
> >>  arch/mips/sibyte/bcm1480/irq.c        |    2 ++
> >>  arch/mips/sibyte/sb1250/irq.c         |    2 ++
> >>  arch/powerpc/platforms/pseries/xics.c |    5 +++++
> >>  arch/powerpc/sysdev/mpic.c            |    2 ++
> >>  arch/sparc/kernel/irq_64.c            |    7 +++++++
> >>  drivers/xen/events.c                  |    2 ++
> >>  kernel/irq/manage.c                   |    6 ++----
> >>  kernel/irq/migration.c                |    8 +++-----
> >>  15 files changed, 45 insertions(+), 12 deletions(-)
> >
> > Hm, this spreads a lot of instances of identical lines:
> >
> >   cpumask_copy(irq_desc[irq].affinity, mask_val);
> >
> > all around architectures. How is that an improvement?
> >
> 
> in failing path in set_affinity, for example it can not get vector 
> in specified cpu, then affinity should not be changed.

isnt the right solution then to propagate the failure code back to 
the generic code?

Preferably via a new callback, and the patches only touching the 
core code plus maybe x86, so that other architectures can be 
converted/fixed more gradually.

	Ingo