lkml.org 
[lkml]   [2017]   [Feb]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH V2 7/7] mm: add a separate RSS for MADV_FREE pages
On Fri, Feb 10, 2017 at 02:35:05PM +0100, Michal Hocko wrote:
> On Fri 03-02-17 15:33:23, Shaohua Li wrote:
> > Add a separate RSS for MADV_FREE pages. The pages are charged into
> > MM_ANONPAGES (because they are mapped anon pages) and also charged into
> > the MM_LAZYFREEPAGES. /proc/pid/statm will have an extra field to
> > display the RSS, which userspace can use to determine the RSS excluding
> > MADV_FREE pages.
> >
> > The basic idea is to increment the RSS in madvise and decrement in unmap
> > or page reclaim. There is one limitation. If a page is shared by two
> > processes, since madvise only has mm cotext of current process, it isn't
> > convenient to charge the RSS for both processes. So we don't charge the
> > RSS if the mapcount isn't 1. On the other hand, fork can make a
> > MADV_FREE page shared by two processes. To make things consistent, we
> > uncharge the RSS from the source mm in fork.
> >
> > A new flag is added to indicate if a page is accounted into the RSS. We
> > can't use SwapBacked flag to do the determination because we can't
> > guarantee the page has SwapBacked flag cleared in madvise. We are
> > reusing mappedtodisk flag which should not be set for Anon pages.
> >
> > There are a couple of other places we need to uncharge the RSS,
> > activate_page and mark_page_accessed. activate_page is used by swap,
> > where MADV_FREE pages are already not in lazyfree state before going
> > into swap. mark_page_accessed is mainly used for file pages, but there
> > are several places it's used by anonymous pages. I fixed gup, but not
> > some gpu drivers and kvm. If the drivers use MADV_FREE, we might have
> > inprecise RSS accounting.
> >
> > Please note, the accounting is never going to be precise. MADV_FREE page
> > could be written by userspace without notification to the kernel. The
> > page can't be reclaimed like other clean lazyfree pages. The page isn't
> > real lazyfree page. But since kernel isn't aware of this, the page is
> > still accounted as lazyfree, thus the accounting could be incorrect.
>
> This is all quite complex and as you say unprecise already. From the
> description it is not even clear why do we need it at all. Why is
> /proc/<pid>/smaps insufficient? I am also not fun of a new page flag -
> even though you managed to recycle an existing one which is a plus.

We have monitor app running in the system to check other apps' RSS and kill
them if RSS is abnormal. Checking /proc/pid/smaps is too complicated and slow,
don't think we can go that way. Yes, the accounting isn't precise, but should
be much better than exporting nothing to userspace.

Thanks,
Shaohua

\
 
 \ /
  Last update: 2017-02-10 19:11    [W:0.107 / U:1.772 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site