linux-kernel - Re: [PATCH] mm, memcg: Add memory.transparent_hugepage

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]

Message-ID: <20200324132350.GA7528@chrisdown.name>
Date:   Tue, 24 Mar 2020 13:23:50 +0000
From:   Chris Down <chris@...isdown.name>
To:     Hui Zhu <teawater@...il.com>
Cc:     hannes@...xchg.org, mhocko@...nel.org, vdavydov.dev@...il.com,
        akpm@...ux-foundation.org, hughd@...gle.com,
        yang.shi@...ux.alibaba.com, kirill@...temov.name,
        dan.j.williams@...el.com, aneesh.kumar@...ux.ibm.com,
        sean.j.christopherson@...el.com, thellstrom@...are.com,
        guro@...com, shakeelb@...gle.com, tj@...nel.org,
        tglx@...utronix.de, linux-kernel@...r.kernel.org,
        cgroups@...r.kernel.org, linux-mm@...ck.org,
        Hui Zhu <teawaterz@...ux.alibaba.com>
Subject: Re: [PATCH] mm, memcg: Add memory.transparent_hugepage_disabled

Hui Zhu writes:
>/sys/kernel/mm/transparent_hugepage/enabled is the only interface to
>control if the application can use THP in system level.
>Sometime, we would not want an application use THP even if
>transparent_hugepage/enabled is set to "always" or "madvise" because
>thp may need more cpu and memory resources in some cases.
>
>This commit add a new interface memory.transparent_hugepage_disabled
>in memcg.
>When it set to 1, the application inside the cgroup cannot use THP
>except dax.

I'm against this patch even in the abstract -- this adds an extremely niche use 
case to a general interface, and there are plenty of ways to already indicate 
THP preference.

Perhaps there is some new interface desirable for your use case, but adding a 
new global file to memcg certainly isn't it.