linux-kernel - Re: [PATCH 2/3] cpuidle: coupled: abort idle if pokes are pending

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: <1377508221.16229.34.camel@jlo-ubuntu-64.nvidia.com>
Date:	Mon, 26 Aug 2013 17:10:21 +0800
From:	Joseph Lo <josephl@...dia.com>
To:	Colin Cross <ccross@...roid.com>
CC:	"linux-pm@...r.kernel.org" <linux-pm@...r.kernel.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Neil Zhang <zhangwm@...vell.com>,
	"linux-tegra@...r.kernel.org" <linux-tegra@...r.kernel.org>,
	"stable@...r.kernel.org" <stable@...r.kernel.org>,
	"Rafael J. Wysocki" <rjw@...k.pl>,
	Daniel Lezcano <daniel.lezcano@...aro.org>
Subject: Re: [PATCH 2/3] cpuidle: coupled: abort idle if pokes are pending

On Sat, 2013-08-24 at 03:45 +0800, Colin Cross wrote:
> Joseph Lo <josephl@...dia.com> reported a lockup on Tegra3 caused
> by a race condition in coupled cpuidle.  When two or more cpus
Actually this issue can be reproduced on both Tegra20/30 platforms. And
I suggest using Tegra20 to replace Tegra3 here, we only apply coupled
CPU idle function on Tegra20 in the mainline right now.
> enter idle at the same time, the first cpus to arrive may go to the
> ready loop without processing pending pokes from the last cpu to
> arrive.
> 
> This patch adds a check for pending pokes once all cpus have been
> synchronized in the ready loop and resets the coupled state and
> retries if any cpus failed to handle their pending poke.
> 
> Retrying on all cpus may trigger the same issue again, so this patch
> also adds a check to ensure that each cpu has received at least one
> poke between when it enters the waiting loop and when it moves on to
> the ready loop.
> 
> Reported-by: Joseph Lo <josephl@...dia.com>
> CC: stable@...r.kernel.org
> Signed-off-by: Colin Cross <ccross@...roid.com>
> ---
>  drivers/cpuidle/coupled.c | 107 +++++++++++++++++++++++++++++++++++-----------
>  1 file changed, 82 insertions(+), 25 deletions(-)
> 
[snip]
> +/*
> + * The cpuidle_coupled_poke_pending mask is used to ensure that each cpu has
s/cpuidle_coupled_poke_pending/cpuidle_coupled_poked/? :)
> + * been poked once to minimize entering the ready loop with a poke pending,
> + * which would require aborting and retrying.
> + */
> +static cpumask_t cpuidle_coupled_poked;
> 
I fixed this issue by checking if there is a pending SGI, then abort the
coupled state on Tegra20. It still can be reproduced easily if I remove
the checking code. So I tested the case with this patch, the result is
good. This patch can fix the issue indeed.

I also tested with the other two patches. It didn't cause any
regression.

So this series:
Tested-by: Joseph Lo <josephl@...dia.com>


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/