linux-kernel - Re: [PATCH] mm/vmstat: add events for THP max_ptes

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <61c93a27.1c69fb81.627b0.8bf6@mx.google.com>
Date:   Mon, 27 Dec 2021 03:59:33 +0000
From:   CGEL <cgel.zte@...il.com>
To:     "Huang, Ying" <ying.huang@...el.com>
Cc:     akpm@...ux-foundation.org, yang.shi@...ux.alibaba.com,
        dave.hansen@...ux.intel.com, yang.yang29@....com.cn,
        minchan@...nel.org, saravanand@...com,
        linux-kernel@...r.kernel.org, linux-mm@...ck.org
Subject: Re: [PATCH] mm/vmstat: add events for THP max_ptes_* exceeds

On Mon, Dec 27, 2021 at 11:19:32AM +0800, Huang, Ying wrote:
> cgel.zte@...il.com writes:
> 
> > From: Yang Yang <yang.yang29@....com.cn>
> >
> > There are interfaces to adjust max_ptes_none, max_ptes_swap,
> > max_ptes_shared values, see
> > /sys/kernel/mm/transparent_hugepage/khugepaged/.
> >
> > But system administrator maynot know which value is the best. So
> > Add those events to support adjusting max_ptes_* to suitable values.
> >
> > For example, if default max_ptes_swap value causes too much failures,
> > and system uses zram whose IO is fast, administrator could increase
> > max_ptes_swap until THP_SCAN_EXCEED_SWAP_PTE not increase anymore.
> >
> > Signed-off-by: Yang Yang <yang.yang29@....com.cn>
> > ---
> >  include/linux/vm_event_item.h | 3 +++
> >  mm/khugepaged.c               | 7 +++++++
> >  mm/vmstat.c                   | 3 +++
> >  3 files changed, 13 insertions(+)
> >
> > diff --git a/include/linux/vm_event_item.h b/include/linux/vm_event_item.h
> > index a185cc75ff52..7b2363388bfa 100644
> > --- a/include/linux/vm_event_item.h
> > +++ b/include/linux/vm_event_item.h
> > @@ -98,6 +98,9 @@ enum vm_event_item { PGPGIN, PGPGOUT, PSWPIN, PSWPOUT,
> >  		THP_SPLIT_PAGE_FAILED,
> >  		THP_DEFERRED_SPLIT_PAGE,
> >  		THP_SPLIT_PMD,
> > +		THP_SCAN_EXCEED_NONE_PTE,
> > +		THP_SCAN_EXCEED_SWAP_PTE,
> > +		THP_SCAN_EXCEED_SHARED_PTE,
> 
> "SCAN" appears too general, how about
> 
>   THP_COLLAPSE_EXCEED_SWAP_PTE
> 
> ?
>
Thanks for your suggesting.

But I think SCAN is easy to understand, it is SCAN behaviour cause the event,
and we know this event is comes from SCAN_EXCEED_*_PTE.

> Best Regards,
> Huang, Ying
> 
> [snip]