[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <HE1PR05MB162584392E880F4306659593AAA40@HE1PR05MB1625.eurprd05.prod.outlook.com>
Date: Tue, 23 Feb 2016 10:58:29 +0000
From: Noam Camus <noamca@...lanox.com>
To: Peter Zijlstra <peterz@...radead.org>,
Vineet Gupta <Vineet.Gupta1@...opsys.com>
CC: "linux-arch@...r.kernel.org" <linux-arch@...r.kernel.org>,
Marc Zyngier <marc.zyngier@....com>,
Frederic Weisbecker <fweisbec@...il.com>,
lkml <linux-kernel@...r.kernel.org>,
arcml <linux-snps-arc@...ts.infradead.org>
Subject: RE: Interesting csd deadlock on ARC
>From: Peter Zijlstra [mailto:peterz@...radead.org]
>Sent: Tuesday, February 23, 2016 12:40 PM
>The only requirement for irq_work is that it runs after the NMI completes and runs from regular IRQ context. >There are no strict interrupt priority requirements, only that it happens.
We here already use self IPI and irq_work on ARC (with no NMI concerns).
Please see patch at:
https://github.com/EZchip/linux/commit/e42556738e610295f917c79dae166373cd11de88
Noam
Powered by blists - more mailing lists