linux-kernel - Re: mm: deadlock between get_online_cpus/pcpu

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: <20170208140332.syic3peyfavd3kl6@techsingularity.net>
Date:   Wed, 8 Feb 2017 14:03:32 +0000
From:   Mel Gorman <mgorman@...hsingularity.net>
To:     Thomas Gleixner <tglx@...utronix.de>
Cc:     Michal Hocko <mhocko@...nel.org>, Christoph Lameter <cl@...ux.com>,
        Vlastimil Babka <vbabka@...e.cz>,
        Dmitry Vyukov <dvyukov@...gle.com>, Tejun Heo <tj@...nel.org>,
        "linux-mm@...ck.org" <linux-mm@...ck.org>,
        LKML <linux-kernel@...r.kernel.org>,
        Ingo Molnar <mingo@...nel.org>,
        Peter Zijlstra <peterz@...radead.org>,
        syzkaller <syzkaller@...glegroups.com>,
        Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: mm: deadlock between get_online_cpus/pcpu_alloc

On Wed, Feb 08, 2017 at 02:23:19PM +0100, Thomas Gleixner wrote:
> On Wed, 8 Feb 2017, Mel Gorman wrote:
> > It may be worth noting that patches in Andrew's tree no longer disable
> > interrupts in the per-cpu allocator and now per-cpu draining will
> > be from workqueue context. The reasoning was due to the overhead of
> > the page allocator with figures included. Interrupts will bypass the
> > per-cpu allocator and use the irq-safe zone->lock to allocate from
> > the core.  It'll collide with the RT patch. Primary patch of interest is
> > http://www.ozlabs.org/~akpm/mmots/broken-out/mm-page_alloc-only-use-per-cpu-allocator-for-irq-safe-requests.patch
> 
> Yeah, we'll sort that out once it hits Linus tree and we move RT forward.
> Though I have once complaint right away:
> 
> +	preempt_enable_no_resched();
> 
> This is a nono, even in mainline. You effectively disable a preemption
> point.
> 

This came up during review on whether it should or shouldn't be a preemption
point. Initially it was preempt_enable() but a preemption point didn't
exist before, the reviewer pushed for it and as it was the allocator fast
path that was unlikely to need a reschedule or preempt, I made the change.

I can alter it before it hits mainline if you say RT is going to have an
issue with it.

> > The draining from workqueue context may be a problem for RT but one
> > option would be to move the drain to only drain for high-order pages
> > after direct reclaim combined with only draining for order-0 if
> > __alloc_pages_may_oom is about to be called.
> 
> Why would the draining from workqueue context be an issue on RT?
> 

It probably isn't. The latency of the operation is likely longer than an IPI
was but given the context it occurs in, I severely doubted it mattered. I
couldn't think of a reason why it would matter to RT but there was no harm
double checking.

-- 
Mel Gorman
SUSE Labs