From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sebastian Andrzej Siewior Subject: Re: [PATCH RT v2] timer: Raise softirq if there's irq_work Date: Fri, 24 Jan 2014 21:20:39 +0100 Message-ID: <20140124202039.GA32662@linutronix.de> References: <20140124145157.1f30d447@gandalf.local.home> <20140124150933.22d8c20a@gandalf.local.home> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Cc: LKML , linux-rt-users , Thomas Gleixner , Clark Williams , "Luis Claudio R. Goncalves" , John Kacur , Mike Galbraith , Joakim Hernberg , Joe Korty , Muli Baron To: Steven Rostedt Return-path: Content-Disposition: inline In-Reply-To: <20140124150933.22d8c20a@gandalf.local.home> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-rt-users.vger.kernel.org * Steven Rostedt | 2014-01-24 15:09:33 [-0500]: >[ Talking with Sebastian on IRC, it seems that doing the irq_work_run() > from the interrupt in -rt is a bad thing. Here we simply raise the > softirq if there's irq work to do. This too boots on my i7 ] It is okay in general because most of the users should not run in bare interrupt context. The only exception here is the nohz_full_kick_work thing. >After trying hard to figure out why my i7 box was locking up with the >new active_timers code, that does not run the timer softirq if there >are no active timers, I took an extra look at the softirq handler and >noticed that it doesn't just run timer softirqs, it also runs irq work. > >This was the bug that was locking up the system. It wasn't missing a >timer, it was missing irq work. By always doing the irq work callbacks, >the system boots fine. > >No need to check for defined(CONFIG_IRQ_WORK). When that's not set the >"irq_work_needs_cpu()" is a static inline that returns false. > >Signed-off-by: Steven Rostedt Thank you Steven, this makes sense. Sebastian