On Wed 21-03-18 01:35:05, Andrey Ryabinin wrote:
> On 03/21/2018 01:15 AM, David Rientjes wrote:
> > On Wed, 21 Mar 2018, Andrey Ryabinin wrote:
> >
> >>>>> It would probably be best to limit the
> >>>>> nr_pages to the amount that needs to be reclaimed, though, rather than
> >>>>> over reclaiming.
> >>>>
> >>>> How do you achieve that? The charging path is not synchornized with the
> >>>> shrinking one at all.
> >>>>
> >>>
> >>> The point is to get a better guess at how many pages, up to
> >>> SWAP_CLUSTER_MAX, that need to be reclaimed instead of 1.
> >>>
> >>>>> If you wanted to be invasive, you could change page_counter_limit() to
> >>>>> return the count - limit, fix up the callers that look for -EBUSY, and
> >>>>> then use max(val, SWAP_CLUSTER_MAX) as your nr_pages.
> >>>>
> >>>> I am not sure I understand
> >>>>
> >>>
> >>> Have page_counter_limit() return the number of pages over limit, i.e.
> >>> count - limit, since it compares the two anyway. Fix up existing callers
> >>> and then clamp that value to SWAP_CLUSTER_MAX in
> >>> mem_cgroup_resize_limit(). It's a more accurate guess than either 1 or
> >>> 1024.
> >>>
> >>
> >> JFYI, it's never 1, it's always SWAP_CLUSTER_MAX.
> >> See try_to_free_mem_cgroup_pages():
> >> ....
> >> struct scan_control sc = {
> >> .nr_to_reclaim = max(nr_pages, SWAP_CLUSTER_MAX),
> >>
> >
> > Is SWAP_CLUSTER_MAX the best answer if I'm lowering the limit by 1GB?
> >
>
> Absolutely not. I completely on your side here.
> I've tried to fix this recently - http://lkml.kernel.org/r/20180119132544.19569-2-aryabinin@xxxxxxxxxxxxx
> I guess that Andrew decided to not take my patch, because Michal wasn't
> happy about it (see mail archives if you want more details).
I was unhappy about the explanation and justification of the patch. It
is still not clear to me why try_to_free_mem_cgroup_pages with a single
target should be slower than multiple calls of this function with
smaller batches when the real reclaim is still SWAP_CLUSTER_MAX batched.
There is also a theoretical risk of over reclaim. Especially with large
targets.
--
Michal Hocko
SUSE Labs
--
To unsubscribe from this list: send the line "unsubscribe cgroups" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
