From: Waiman Long <longman@redhat.com>
To: Michal Hocko <mhocko@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
Johannes Weiner <hannes@cmpxchg.org>,
Roman Gushchin <guro@fb.com>, Vlastimil Babka <vbabka@suse.cz>,
Konstantin Khlebnikov <khlebnikov@yandex-team.ru>,
Jann Horn <jannh@google.com>, Song Liu <songliubraving@fb.com>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
Rafael Aquini <aquini@redhat.com>, Mel Gorman <mgorman@suse.de>
Subject: Re: [PATCH] mm/vmstat: Reduce zone lock hold time when reading /proc/pagetypeinfo
Date: Tue, 22 Oct 2019 14:00:35 -0400 [thread overview]
Message-ID: <0b206255-5c62-18f5-d751-a5576a6c0e8f@redhat.com> (raw)
In-Reply-To: <20191022165745.GT9379@dhcp22.suse.cz>
On 10/22/19 12:57 PM, Michal Hocko wrote:
> [Cc Mel]
>
> On Tue 22-10-19 12:21:56, Waiman Long wrote:
>> The pagetypeinfo_showfree_print() function prints out the number of
>> free blocks for each of the page orders and migrate types. The current
>> code just iterates the each of the free lists to get counts. There are
>> bug reports about hard lockup panics when reading the /proc/pagetyeinfo
>> file just because it look too long to iterate all the free lists within
>> a zone while holing the zone lock with irq disabled.
>>
>> Given the fact that /proc/pagetypeinfo is readable by all, the possiblity
>> of crashing a system by the simple act of reading /proc/pagetypeinfo
>> by any user is a security problem that needs to be addressed.
> Should we make the file 0400? It is a useful thing when debugging but
> not something regular users would really need for life.
>
I am not against doing that, but it may break existing applications that
somehow need to read pagetypeinfo. That is why I didn't try to advocate
about changing protection.
>> There is a free_area structure associated with each page order. There
>> is also a nr_free count within the free_area for all the different
>> migration types combined. Tracking the number of free list entries
>> for each migration type will probably add some overhead to the fast
>> paths like moving pages from one migration type to another which may
>> not be desirable.
> Have you tried to measure that overhead?
I haven't tried to measure the performance impact yet. I did thought
about tracking nr_free for each of the migration types within a
free_area. That will require auditing the code to make sure that all the
intra-free_area migrations are properly accounted for. I can work on it
if people prefer this alternative.
>
>> we can actually skip iterating the list of one of the migration types
>> and used nr_free to compute the missing count. Since MIGRATE_MOVABLE
>> is usually the largest one on large memory systems, this is the one
>> to be skipped. Since the printing order is migration-type => order, we
>> will have to store the counts in an internal 2D array before printing
>> them out.
>>
>> Even by skipping the MIGRATE_MOVABLE pages, we may still be holding the
>> zone lock for too long blocking out other zone lock waiters from being
>> run. This can be problematic for systems with large amount of memory.
>> So a check is added to temporarily release the lock and reschedule if
>> more than 64k of list entries have been iterated for each order. With
>> a MAX_ORDER of 11, the worst case will be iterating about 700k of list
>> entries before releasing the lock.
> But you are still iterating through the whole free_list at once so if it
> gets really large then this is still possible. I think it would be
> preferable to use per migratetype nr_free if it doesn't cause any
> regressions.
>
Yes, it is still theoretically possible. I will take a further look at
having per-migrate type nr_free. BTW, there is one more place where the
free lists are being iterated with zone lock held - mark_free_pages().
Cheers,
Longman
next prev parent reply other threads:[~2019-10-22 18:00 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-22 16:21 [PATCH] mm/vmstat: Reduce zone lock hold time when reading /proc/pagetypeinfo Waiman Long
2019-10-22 16:57 ` Michal Hocko
2019-10-22 18:00 ` Waiman Long [this message]
2019-10-22 18:40 ` Waiman Long
2019-10-23 0:52 ` David Rientjes
2019-10-23 8:31 ` Mel Gorman
2019-10-23 9:04 ` Michal Hocko
2019-10-23 9:56 ` Mel Gorman
2019-10-23 10:27 ` [RFC PATCH 0/2] " Michal Hocko
2019-10-23 10:27 ` [RFC PATCH 1/2] mm, vmstat: hide /proc/pagetypeinfo from normal users Michal Hocko
2019-10-23 13:13 ` Mel Gorman
2019-10-23 13:27 ` Vlastimil Babka
2019-10-23 14:52 ` Waiman Long
2019-10-23 15:10 ` Rafael Aquini
2019-10-23 16:15 ` Vlastimil Babka
2019-10-24 19:01 ` David Rientjes
2019-10-23 10:27 ` [RFC PATCH 2/2] mm, vmstat: reduce zone->lock holding time by /proc/pagetypeinfo Michal Hocko
2019-10-23 13:15 ` Mel Gorman
2019-10-23 13:32 ` Vlastimil Babka
2019-10-23 13:37 ` Michal Hocko
2019-10-23 13:48 ` Vlastimil Babka
2019-10-23 14:31 ` Michal Hocko
2019-10-23 16:20 ` Vlastimil Babka
2019-10-23 13:46 ` Rafael Aquini
2019-10-23 14:56 ` Waiman Long
2019-10-23 15:21 ` Waiman Long
2019-10-23 16:10 ` Michal Hocko
2019-10-23 16:17 ` Waiman Long
2019-10-23 16:21 ` Waiman Long
2019-10-23 16:15 ` Vlastimil Babka
2019-10-23 16:41 ` Michal Hocko
2019-10-23 16:47 ` Waiman Long
2019-10-23 17:34 ` [PATCH 1/2] mm, vmstat: Release zone lock more frequently when reading /proc/pagetypeinfo Waiman Long
2019-10-23 18:01 ` Michal Hocko
2019-10-23 18:14 ` Waiman Long
2019-10-23 20:02 ` Michal Hocko
2019-10-23 17:34 ` [PATCH 2/2] mm, vmstat: List total free blocks for each order in /proc/pagetypeinfo Waiman Long
2019-10-23 18:02 ` Michal Hocko
2019-10-23 18:07 ` Waiman Long
2019-10-24 8:20 ` [RFC PATCH 0/2] mm/vmstat: Reduce zone lock hold time when reading /proc/pagetypeinfo Michal Hocko
2019-10-24 16:16 ` Waiman Long
2019-10-23 12:42 ` [PATCH] " Qian Cai
2019-10-23 13:25 ` Vlastimil Babka
2019-10-22 21:59 ` Andrew Morton
2019-10-23 6:15 ` Michal Hocko
2019-10-23 14:30 ` Waiman Long
2019-10-23 14:48 ` Qian Cai
2019-10-23 15:01 ` Waiman Long
2019-10-23 15:05 ` Qian Cai
2019-10-23 22:30 ` Andrew Morton
2019-10-24 5:33 ` Qian Cai
2019-10-24 7:42 ` Michal Hocko
2019-10-24 11:11 ` Qian Cai
2019-10-24 13:38 ` Michal Hocko
2019-10-24 14:55 ` Qian Cai
2019-10-24 3:33 ` Feng Tang
2019-10-24 4:34 ` Qian Cai
2019-10-24 5:34 ` Feng Tang
2019-10-24 10:51 ` Qian Cai
2019-10-25 1:38 ` Feng Tang
2019-10-23 15:03 ` Rafael Aquini
2019-10-23 15:51 ` Qian Cai
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=0b206255-5c62-18f5-d751-a5576a6c0e8f@redhat.com \
--to=longman@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=aquini@redhat.com \
--cc=gregkh@linuxfoundation.org \
--cc=guro@fb.com \
--cc=hannes@cmpxchg.org \
--cc=jannh@google.com \
--cc=khlebnikov@yandex-team.ru \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@suse.de \
--cc=mhocko@kernel.org \
--cc=songliubraving@fb.com \
--cc=vbabka@suse.cz \
/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).