[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240220032008.GH65758@system.software.com>
Date: Tue, 20 Feb 2024 12:20:08 +0900
From: Byungchul Park <byungchul@...com>
To: Andrew Morton <akpm@...ux-foundation.org>
Cc: "Huang, Ying" <ying.huang@...el.com>, mingo@...hat.com,
peterz@...radead.org, juri.lelli@...hat.com,
vincent.guittot@...aro.org, dietmar.eggemann@....com,
rostedt@...dmis.org, bsegall@...gle.com, mgorman@...e.de,
bristot@...hat.com, vschneid@...hat.com,
linux-kernel@...r.kernel.org, linux-mm@...ck.org,
kernel_team@...ynix.com
Subject: Re: [PATCH v4] sched/numa, mm: do not try to migrate memory to
memoryless nodes
On Mon, Feb 19, 2024 at 07:05:20PM -0800, Andrew Morton wrote:
> On Tue, 20 Feb 2024 10:53:43 +0900 Byungchul Park <byungchul@...com> wrote:
>
> > > > IIUC, you will use patch as fix to the issue in
> > > >
> > > > https://lore.kernel.org/lkml/20240216111502.79759-1-byungchul@sk.com/
> > > >
> > > > If so, we need the Fixes: tag to make it land in -stable properly.
> > >
> > > Yes, this changelog is missing rather a lot of important information.
> >
> > This is not the root cause fix any more but just optimization.
>
> It would have been helpful to have told us this in the changelog :(
Sorry for that and making you guys confused.
> > That's
> > why I didn't add Fixes: tag and cc stable@...r.kernel.org in here.
> >
> > Instead, I added Fixes: tag and cc'ed stable@...r.kernel.org in the real
> > fix patch. check the following link please:
> >
> > https://lore.kernel.org/lkml/20240216111502.79759-1-byungchul@sk.com/
>
> But doesn't this patch "sched/numa, mm: do not try to migrate memory to
> memoryless nodes" also fix the bug? Do we truly need both?
Yes. The oops is gone with "sched/numa, mm: do not try to migrate memory
to memoryless nodes". However, as you know, wrongly manipulating array
index is very dangerous - what hackers are looking for. Even with security
in mind, both are necessary. Plus, no one gurantees the problematic code
is gone through with a numa node that has no managed zones in the future.
Byungchul
Powered by blists - more mailing lists