[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ZPWH53NUqTIoDcMV@dhcp22.suse.cz>
Date: Mon, 4 Sep 2023 09:31:51 +0200
From: Michal Hocko <mhocko@...e.com>
To: Zhaoyang Huang <huangzhaoyang@...il.com>
Cc: Matthew Wilcox <willy@...radead.org>, glider@...gle.com,
"zhaoyang.huang" <zhaoyang.huang@...soc.com>,
Andrew Morton <akpm@...ux-foundation.org>, linux-mm@...ck.org,
linux-kernel@...r.kernel.org, ke.wang@...soc.com
Subject: Re: [PATCH] mm: make __GFP_SKIP_ZERO visible to skip zero operation
On Fri 01-09-23 18:29:07, Zhaoyang Huang wrote:
> loop alex
>
> On Thu, Aug 31, 2023 at 8:16 PM Matthew Wilcox <willy@...radead.org> wrote:
> >
> > On Thu, Aug 31, 2023 at 06:52:52PM +0800, zhaoyang.huang wrote:
> > > From: Zhaoyang Huang <zhaoyang.huang@...soc.com>
> > >
> > > There is no explicit gfp flags to let the allocation skip zero
> > > operation when CONFIG_INIT_ON_ALLOC_DEFAULT_ON=y. I would like to make
> > > __GFP_SKIP_ZERO be visible even if kasan is not configured.
> >
> > This bypasses a security feature so you're going to have to do a little
> > better than "I want it".
> Thanks for pointing this out. What I want to do is to give the user a
> way to exempt some types of pages from being zeroed, which could help
> on performance issues.
Could you be more explicit about those users? Your patch doesn't add
any.
> Could we have the most safety concern admin
> use INIT_ON_FREE while the less concerned use INIT_ON_ALLOC &
> __GFP_SKIP_ZERO as a light version method?
Are you suggesting the __GFP_SKIP_ZERO would be ignored in any setups
except for init_on_alloc?
--
Michal Hocko
SUSE Labs
Powered by blists - more mailing lists