On Fri, Jan 24, 2014 at 03:35:42PM -0500, Steven Rostedt wrote:

On Fri, 24 Jan 2014 21:20:39 +0100
Sebastian Andrzej Siewior [off-list ref] wrote:

quoted

* Steven Rostedt | 2014-01-24 15:09:33 [-0500]:

quoted

[ 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.

I know we discussed this on IRC, but I wanted to publicly state that
the missing irq work callback was the RCU's rsp_wakeup() function.

Failing to invoke rsp_wakeup() when it was needed could potentially
stop RCU grace periods from happening, so having rsp_wakeup() happen
when it is needed is pretty important...

But I would guess that you knew that already.  ;-)

							Thanx, Paul