linux-kernel - Re: [PATCH v2] mm: add lazyfree folio to lru tail

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 for Android: free password hash cracker in your pocket

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

Message-ID: <CAGsJ_4wUrm1Q7Oxb5BHC2ypyf4wAH+UO9KYhCUwsEbhkcc7QGg@mail.gmail.com>
Date: Tue, 27 Aug 2024 14:18:18 +1200
From: Barry Song <21cnbao@...il.com>
To: Hailong Liu <hailong.liu@...o.com>
Cc: Lokesh Gidra <lokeshgidra@...gle.com>, Suren Baghdasaryan <surenb@...gle.com>, 
	Nicolas Geoffray <ngeoffray@...gle.com>, Michal Hocko <mhocko@...e.com>, gaoxu <gaoxu2@...or.com>, 
	Andrew Morton <akpm@...ux-foundation.org>, "linux-mm@...ck.org" <linux-mm@...ck.org>, 
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>, Shaohua Li <shli@...com>, 
	yipengxiang <yipengxiang@...or.com>, fengbaopeng <fengbaopeng@...or.com>, 
	Kalesh Singh <kaleshsingh@...gle.com>
Subject: Re: [PATCH v2] mm: add lazyfree folio to lru tail

On Tue, Aug 27, 2024 at 2:13 PM Hailong Liu <hailong.liu@...o.com> wrote:
>
> On Mon, 26. Aug 09:37, Lokesh Gidra wrote:
> >
> > IMHO, within LRU_INACTIVE_FILE, MADV_FREE'ed pages should be
> > prioritized for reclamation over file ones.
>
>
> > >
> > > Adding Lokesh.
> > > Lokesh, could you please comment on the reasoning behind the above
> > > mentioned change?
> >
> > Adding Nicolas as well, in case he wants to add something.
> IMHO, lruvec_add_folio is enough. if lruvec_add_folio_tail why not use
> MADV_DONTNEED instead? In MM the reclaim policy prefer to reclaim file cache, if
> MADV_FREE'd pages directly add to the tail, they might be reclaimed instantly.
> Also the benefit of workingset_refault_file cannot be convinced for me.

My understanding is that MADV_DONTNEED will immediately free the memory,
whereas MADV_FREE will release memory only under memory pressure. If
memory pressure is low, the anonymous memory may still be gotten back
without causing page faults. This might be what Lokesh is aiming to achieve.

>
> So we should know the reasons and the benefits of the changes. page faults or ?
> > >
> > > >
> > > > >
> > > > > --
> > > > > Michal Hocko
> > > > > SUSE Labs
> > > > >
> > > >

Thanks
Barry