Messages in this thread | | | Date | Thu, 31 Jul 2014 13:51:37 +0200 | From | Michal Hocko <> | Subject | Re: [PATCH] swap: remove the struct cpumask has_work |
| |
On Thu 31-07-14 11:30:19, Lai Jiangshan wrote: > It is suggested that cpumask_var_t and alloc_cpumask_var() should be used > instead of struct cpumask. But I don't want to add this complicity nor > leave this unwelcome "static struct cpumask has_work;", so I just remove > it and use flush_work() to perform on all online drain_work. flush_work() > performs very quickly on initialized but unused work item, thus we don't > need the struct cpumask has_work for performance.
Why? Just because there is general recommendation for using cpumask_var_t rather than cpumask?
In this particular case cpumask shouldn't matter much as it is static. Your code will work as well, but I do not see any strong reason to change it just to get rid of cpumask which is not on stack.
> CC: akpm@linux-foundation.org > CC: Chris Metcalf <cmetcalf@tilera.com> > CC: Mel Gorman <mgorman@suse.de> > CC: Tejun Heo <tj@kernel.org> > CC: Christoph Lameter <cl@gentwo.org> > CC: Frederic Weisbecker <fweisbec@gmail.com> > Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com> > --- > mm/swap.c | 11 ++++------- > 1 files changed, 4 insertions(+), 7 deletions(-) > > diff --git a/mm/swap.c b/mm/swap.c > index 9e8e347..bb524ca 100644 > --- a/mm/swap.c > +++ b/mm/swap.c > @@ -833,27 +833,24 @@ static DEFINE_PER_CPU(struct work_struct, lru_add_drain_work); > void lru_add_drain_all(void) > { > static DEFINE_MUTEX(lock); > - static struct cpumask has_work; > int cpu; > > mutex_lock(&lock); > get_online_cpus(); > - cpumask_clear(&has_work); > > for_each_online_cpu(cpu) { > struct work_struct *work = &per_cpu(lru_add_drain_work, cpu); > > + INIT_WORK(work, lru_add_drain_per_cpu); > + > if (pagevec_count(&per_cpu(lru_add_pvec, cpu)) || > pagevec_count(&per_cpu(lru_rotate_pvecs, cpu)) || > pagevec_count(&per_cpu(lru_deactivate_pvecs, cpu)) || > - need_activate_page_drain(cpu)) { > - INIT_WORK(work, lru_add_drain_per_cpu); > + need_activate_page_drain(cpu)) > schedule_work_on(cpu, work); > - cpumask_set_cpu(cpu, &has_work); > - } > } > > - for_each_cpu(cpu, &has_work) > + for_each_online_cpu(cpu) > flush_work(&per_cpu(lru_add_drain_work, cpu)); > > put_online_cpus(); > -- > 1.7.4.4 > > -- > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > Please read the FAQ at http://www.tux.org/lkml/
-- Michal Hocko SUSE Labs
| |