linux-kernel - Re: [PATCH v2 2/2] mm: initialize deferred pages with interrupts enabled

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:   Thu, 2 Apr 2020 19:16:31 +0200
From:   Michal Hocko <mhocko@...nel.org>
To:     Pavel Tatashin <pasha.tatashin@...een.com>
Cc:     David Hildenbrand <david@...hat.com>,
        LKML <linux-kernel@...r.kernel.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        linux-mm <linux-mm@...ck.org>,
        Dan Williams <dan.j.williams@...el.com>,
        Shile Zhang <shile.zhang@...ux.alibaba.com>,
        Daniel Jordan <daniel.m.jordan@...cle.com>,
        Kirill Tkhai <ktkhai@...tuozzo.com>,
        James Morris <jmorris@...ei.org>,
        Sasha Levin <sashal@...nel.org>,
        Vlastimil Babka <vbabka@...e.cz>
Subject: Re: [PATCH v2 2/2] mm: initialize deferred pages with interrupts
 enabled

On Thu 02-04-20 11:13:49, Pavel Tatashin wrote:
> > > I do wonder if this change is strictly required in this patch (IOW, if
> > > we could keep calling touch_nmi_watchdog() also without holding a spinlock)
> >
> > Exactly. I would go with your patch on top.
> >
> > > Anyhow, it's the right thing to do.
> 
> Michal,
> 
> The reason I changed it here is because in the original patch that
> this patch fixes we changed cond_sched() to touch_nmi_watchdog():
> $ git show 3a2d7fa8a3d5 | grep -E '(nmi|sched)'
> - cond_resched();
> + touch_nmi_watchdog();
> - cond_resched();
> + touch_nmi_watchdog();
> 
> So, should I move it to a separate patch or is it OK to keep it here?

Having in a separate patch would be better IMO. If for nothing else, the
RCU stall would be easier to see.
-- 
Michal Hocko
SUSE Labs