linux-kernel - Re: [RFC PATCH 2/2] mm/vmscan.c: Prevent allocating shrinker

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

Date:   Mon, 6 Dec 2021 15:15:01 +0100
From:   Michal Hocko <mhocko@...e.com>
To:     David Hildenbrand <david@...hat.com>
Cc:     Nico Pache <npache@...hat.com>, linux-kernel@...r.kernel.org,
        linux-mm@...ck.org, akpm@...ux-foundation.org, shakeelb@...gle.com,
        ktkhai@...tuozzo.com, shy828301@...il.com, guro@...com,
        vbabka@...e.cz, vdavydov.dev@...il.com, raquini@...hat.com
Subject: Re: [RFC PATCH 2/2] mm/vmscan.c: Prevent allocating shrinker_info on
 offlined nodes

On Mon 06-12-21 14:47:13, David Hildenbrand wrote:
> On 06.12.21 14:06, Michal Hocko wrote:
[...]
> > This is certainly possible. But it a) adds a branch to the hotpath and
> > b) it doesn't solve any other potential dereference of garbage node.
> 
> I don't think a) is  a problem but it's easy to measure.

Let me just clarify on this one. This single particular branch will be
indeed hard to match to any performance changes. On a single call of the
allocator it is nothing. But it is a condition that will be executed by
each caller while it won't ever trigger in most systems and workloads.

In other words it will cause a lot of wasted cpu cycles long term. A
more expensive one time initialization is worth in order to have
allocator fast path lighter IMHO.
-- 
Michal Hocko
SUSE Labs