Re: [PATCH v6 5/8] slub: only IPI CPUs that have per cpu obj to flush

From: Milton Miller
Date: Wed Jan 11 2012 - 02:05:10 EST



On Sun Jan 08 2012 about 11:28:11 EST, Gilad Ben-Yossef wrote:
> flush_all() is called for each kmem_cahce_destroy(). So every cache
> being destroyed dynamically ended up sending an IPI to each CPU in the

ends up

> system, regardless if the cache has ever been used there.
>
> For example, if you close the Infinband ipath driver char device file,
> the close file ops calls kmem_cache_destroy(). So running some
> infiniband config tool on one a single CPU dedicated to system tasks
> might interrupt the rest of the 127 CPUs I dedicated to some CPU

127 CPUs dedicated

> intensive task.

CPU intensive or latency sensitive task.


>
> I suspect there is a good chance that every line in the output of "git
> grep kmem_cache_destroy linux/ | grep '\->'" has a similar scenario.
>
> This patch attempts to rectify this issue by sending an IPI to flush
> the per cpu objects back to the free lists only to CPUs that seems to
> have such objects.

that seem to have

>
> The check which CPU to IPI is racy but we don't care since asking a
> CPU without per cpu objects to flush does no damage and as far as I
> can tell the flush_all by itself is racy against allocs on remote
> CPUs anyway, so if you meant the flush_all to be determinstic, you

required (vs meant)

> had to arrange for locking regardless.
>
> Without this patch the following artificial test case:
>
> $ cd /sys/kernel/slab
> $ for DIR in *; do cat $DIR/alloc_calls > /dev/null; done
>
> produces 166 IPIs on an cpuset isolated CPU. With it it produces none.
>
> The code path of memory allocation failure for CPUMASK_OFFSTACK=y
> config was tested using fault injection framework.
>
..
> mm/slub.c | 10 +++++++++-
> 1 files changed, 9 insertions(+), 1 deletions(-)
>
> diff --git a/mm/slub.c b/mm/slub.c
> index 09ccee8..31833d6 100644
> --- a/mm/slub.c
> +++ b/mm/slub.c
> @@ -2013,9 +2013,17 @@ static void flush_cpu_slab(void *d)
> __flush_cpu_slab(s, smp_processor_id());
> }
>
> +static int has_cpu_slab(int cpu, void *info)
> +{
> + struct kmem_cache *s = info;
> + struct kmem_cache_cpu *c = per_cpu_ptr(s->cpu_slab, cpu);
> +
> + return !!(c->page);

__flush_cpu_slab is careful to test that the the per_cpu_ptr is not
NULL before referencing the page field. free_percpu likewise ignores
NULL pointers. We need to check !!(c && c->page) here.


[change int to bool assuming you make the change in the other patch].

> +}
> +
> static void flush_all(struct kmem_cache *s)
> {
> - on_each_cpu(flush_cpu_slab, s, 1);
> + on_each_cpu_cond(has_cpu_slab, flush_cpu_slab, s, 1);
> }

milton
--
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/