linux-kernel - Re: [v2 PATCH -mm] mm: account deferred split THPs into MemAvailable

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: <20190830125304.m6aouvq5ohkerfls@box>
Date:   Fri, 30 Aug 2019 15:53:04 +0300
From:   "Kirill A. Shutemov" <kirill@...temov.name>
To:     Michal Hocko <mhocko@...nel.org>
Cc:     "Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>,
        Yang Shi <yang.shi@...ux.alibaba.com>,
        Vlastimil Babka <vbabka@...e.cz>, hannes@...xchg.org,
        rientjes@...gle.com, akpm@...ux-foundation.org, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org
Subject: Re: [v2 PATCH -mm] mm: account deferred split THPs into MemAvailable

On Wed, Aug 28, 2019 at 06:02:24PM +0200, Michal Hocko wrote:
> > > 
> > > Any idea about a bad case?
> > 
> > Not really.
> > 
> > How bad you want it to get? How many processes share the page? Access
> > pattern? Locking situation?
> 
> Let's say how hard a regular user can make this?

It bumped to ~170 ms if each THP was mapped 5 times.

Adding ptl contention (tight loop of MADV_DONTNEED) in 3 processes that
maps the page, the time to split bumped to ~740ms.

Overally, it's reasonable to take ~100ms per GB of huge pages as rule of
thumb.

-- 
 Kirill A. Shutemov