From: Yang Shi <yang.shi@linux.alibaba.com>
To: "Huang, Ying" <ying.huang@intel.com>
Cc: hannes@cmpxchg.org, mhocko@suse.com, mgorman@techsingularity.net,
kirill.shutemov@linux.intel.com, josef@toxicpanda.com,
hughd@google.com, shakeelb@google.com, hdanton@sina.com,
akpm@linux-foundation.org, linux-mm@kvack.org,
linux-kernel@vger.kernel.org
Subject: Re: [RESEND v5 PATCH 2/2] mm: vmscan: correct some vmscan counters for THP swapout
Date: Mon, 27 May 2019 10:59:55 +0800 [thread overview]
Message-ID: <e63dd512-c448-5bec-7461-32cd4e5c8cac@linux.alibaba.com> (raw)
In-Reply-To: <a32dbca4-6239-828b-9f81-f24d582ddd75@linux.alibaba.com>
On 5/27/19 10:47 AM, Yang Shi wrote:
>
>
> On 5/27/19 10:11 AM, Huang, Ying wrote:
>> Yang Shi <yang.shi@linux.alibaba.com> writes:
>>
>>> Since commit bd4c82c22c36 ("mm, THP, swap: delay splitting THP after
>>> swapped out"), THP can be swapped out in a whole. But, nr_reclaimed
>>> and some other vm counters still get inc'ed by one even though a whole
>>> THP (512 pages) gets swapped out.
>>>
>>> This doesn't make too much sense to memory reclaim. For example,
>>> direct
>>> reclaim may just need reclaim SWAP_CLUSTER_MAX pages, reclaiming one
>>> THP
>>> could fulfill it. But, if nr_reclaimed is not increased correctly,
>>> direct reclaim may just waste time to reclaim more pages,
>>> SWAP_CLUSTER_MAX * 512 pages in worst case.
>>>
>>> And, it may cause pgsteal_{kswapd|direct} is greater than
>>> pgscan_{kswapd|direct}, like the below:
>>>
>>> pgsteal_kswapd 122933
>>> pgsteal_direct 26600225
>>> pgscan_kswapd 174153
>>> pgscan_direct 14678312
>>>
>>> nr_reclaimed and nr_scanned must be fixed in parallel otherwise it
>>> would
>>> break some page reclaim logic, e.g.
>>>
>>> vmpressure: this looks at the scanned/reclaimed ratio so it won't
>>> change semantics as long as scanned & reclaimed are fixed in parallel.
>>>
>>> compaction/reclaim: compaction wants a certain number of physical pages
>>> freed up before going back to compacting.
>>>
>>> kswapd priority raising: kswapd raises priority if we scan fewer pages
>>> than the reclaim target (which itself is obviously expressed in order-0
>>> pages). As a result, kswapd can falsely raise its aggressiveness even
>>> when it's making great progress.
>>>
>>> Other than nr_scanned and nr_reclaimed, some other counters, e.g.
>>> pgactivate, nr_skipped, nr_ref_keep and nr_unmap_fail need to be fixed
>>> too since they are user visible via cgroup, /proc/vmstat or trace
>>> points, otherwise they would be underreported.
>>>
>>> When isolating pages from LRUs, nr_taken has been accounted in base
>>> page, but nr_scanned and nr_skipped are still accounted in THP. It
>>> doesn't make too much sense too since this may cause trace point
>>> underreport the numbers as well.
>>>
>>> So accounting those counters in base page instead of accounting THP as
>>> one page.
>>>
>>> nr_dirty, nr_unqueued_dirty, nr_congested and nr_writeback are used by
>>> file cache, so they are not impacted by THP swap.
>>>
>>> This change may result in lower steal/scan ratio in some cases since
>>> THP may get split during page reclaim, then a part of tail pages get
>>> reclaimed instead of the whole 512 pages, but nr_scanned is accounted
>>> by 512, particularly for direct reclaim. But, this should be not a
>>> significant issue.
>>>
>>> Cc: "Huang, Ying" <ying.huang@intel.com>
>>> Cc: Johannes Weiner <hannes@cmpxchg.org>
>>> Cc: Michal Hocko <mhocko@suse.com>
>>> Cc: Mel Gorman <mgorman@techsingularity.net>
>>> Cc: "Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>
>>> Cc: Hugh Dickins <hughd@google.com>
>>> Cc: Shakeel Butt <shakeelb@google.com>
>>> Signed-off-by: Yang Shi <yang.shi@linux.alibaba.com>
>>> ---
>>> v5: Fixed sc->nr_scanned double accounting per Huang Ying
>>> Added some comments to address the concern about premature OOM
>>> per Hillf Danton
>>> v4: Fixed the comments from Johannes and Huang Ying
>>> v3: Removed Shakeel's Reviewed-by since the patch has been changed
>>> significantly
>>> Switched back to use compound_order per Matthew
>>> Fixed more counters per Johannes
>>> v2: Added Shakeel's Reviewed-by
>>> Use hpage_nr_pages instead of compound_order per Huang Ying and
>>> William Kucharski
>>>
>>> mm/vmscan.c | 42 +++++++++++++++++++++++++++++++-----------
>>> 1 file changed, 31 insertions(+), 11 deletions(-)
>>>
>>> diff --git a/mm/vmscan.c b/mm/vmscan.c
>>> index b65bc50..f4f4d57 100644
>>> --- a/mm/vmscan.c
>>> +++ b/mm/vmscan.c
>>> @@ -1118,6 +1118,7 @@ static unsigned long shrink_page_list(struct
>>> list_head *page_list,
>>> int may_enter_fs;
>>> enum page_references references = PAGEREF_RECLAIM_CLEAN;
>>> bool dirty, writeback;
>>> + unsigned int nr_pages;
>>> cond_resched();
>>> @@ -1129,6 +1130,13 @@ static unsigned long
>>> shrink_page_list(struct list_head *page_list,
>>> VM_BUG_ON_PAGE(PageActive(page), page);
>>> + nr_pages = 1 << compound_order(page);
>>> +
>>> + /*
>>> + * Accounted one page for THP for now. If THP gets swapped
>>> + * out in a whole, will account all tail pages later to
>>> + * avoid accounting tail pages twice.
>>> + */
>>> sc->nr_scanned++;
>>> if (unlikely(!page_evictable(page)))
>>> @@ -1250,7 +1258,7 @@ static unsigned long shrink_page_list(struct
>>> list_head *page_list,
>>> case PAGEREF_ACTIVATE:
>>> goto activate_locked;
>>> case PAGEREF_KEEP:
>>> - stat->nr_ref_keep++;
>>> + stat->nr_ref_keep += nr_pages;
>>> goto keep_locked;
>>> case PAGEREF_RECLAIM:
>>> case PAGEREF_RECLAIM_CLEAN:
>> If the "Accessed bit" of a THP is set in the page table that maps it, it
>> will go PAGEREF_ACTIVATE path here. And the sc->nr_scanned should
>> increase 512 instead of 1. Otherwise sc->nr_activate may be larger than
>> sc->nr_scanned.
>
> Yes, it looks so. It seems the easiest way is to add "nr_pages - 1" in
> activate_locked label if the page is still a THP.
>
> If we add all tail pages at the very beginning, then we have to minus
> tail pages when THP gets split, there are a few places do this.
Reiterating all the goto, it seems minus tail pages may be easier since
there are a lot places which goto activate_locked, keep_locked and keep
after THP has been added into swap cache. We can't tell if the tail
pages are accounted or not unless we introduce a new flag, otherwise
adding tails pages in those labels may account tail pages twice.
>
>>
>> Best Regards,
>> Huang, Ying
>
prev parent reply other threads:[~2019-05-27 3:00 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-05-27 1:57 [RESEND v5 PATCH 1/2] mm: vmscan: remove double slab pressure by inc'ing sc->nr_scanned Yang Shi
2019-05-27 1:57 ` [RESEND v5 PATCH 2/2] mm: vmscan: correct some vmscan counters for THP swapout Yang Shi
2019-05-27 2:11 ` Huang, Ying
2019-05-27 2:47 ` Yang Shi
2019-05-27 2:58 ` Huang, Ying
2019-05-27 3:15 ` Yang Shi
2019-05-27 2:59 ` Yang Shi [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=e63dd512-c448-5bec-7461-32cd4e5c8cac@linux.alibaba.com \
--to=yang.shi@linux.alibaba.com \
--cc=akpm@linux-foundation.org \
--cc=hannes@cmpxchg.org \
--cc=hdanton@sina.com \
--cc=hughd@google.com \
--cc=josef@toxicpanda.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@techsingularity.net \
--cc=mhocko@suse.com \
--cc=shakeelb@google.com \
--cc=ying.huang@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).