On 16.06.20 19:03, Dan Williams wrote:
> On Tue, Jun 16, 2020 at 10:00 AM Dan Williams <dan.j.williams@xxxxxxxxx> wrote:
>>
>> On Tue, Jun 16, 2020 at 5:51 AM Michal Hocko <mhocko@xxxxxxxxxx> wrote:
>>>
>>> On Tue 16-06-20 13:52:12, David Hildenbrand wrote:
>>>> Commit e900a918b098 ("mm: shuffle initial free memory to improve
>>>> memory-side-cache utilization") introduced shuffling of free pages
>>>> during system boot and whenever we online memory blocks.
>>>>
>>>> However, whenever we online memory blocks, all pages that will be
>>>> exposed to the buddy end up getting freed via __free_one_page(). In the
>>>> general case, we free these pages in MAX_ORDER - 1 chunks, which
>>>> corresponds to the shuffle order.
>>>>
>>>> Inside __free_one_page(), we will already shuffle the newly onlined pages
>>>> using "to_tail = shuffle_pick_tail();". Drop explicit zone shuffling on
>>>> memory hotplug.
>
> This was already explained in the initial patch submission. The
> shuffle_pick_tail() shuffling at run time is only sufficient for
> maintaining the shuffle. It's not sufficient for effectively
> randomizing the free list.
Thanks for pointing that out. Something like that should definitely
belong into the code in form of a comment. I'll think of something.
Tow things:
1. The "shuffle when onlining" is really sub-optimal. Assume you hotplug
a 64GB DIMM. With 128MB memory blocks you get 512 zone shuffles. I guess
something better would have been to schedule a reshuffle some ms in the
future.
2. You'll run into the same issue whenever you free a bigger,
consecutive range. Like free_contig_rang()'ing one or more multiple
gigantic pages. We might want to reshuffle (or schedule a reshuffle)
here as well.
--
Thanks,
David / dhildenb
