Re: [PATCH][RT][RFC] irq_work: Have non HARD_IRQ irq work just run from ticks

[Steven Rostedt]

On Tue, 23 Jun 2015 20:45:35 +0200
Sebastian Andrzej Siewior <bigeasy@xxxxxxxxxxxxx> wrote:

> On 06/23/2015 06:30 PM, Steven Rostedt wrote:
> > I'd be sleeping better at night with a third queue. I'll write up a
> > patch and post that as an RFC as well. This will at a minimum keep with
> > the paradigm of mainline linux.
> 
> We had three queues. We are down to two and I think this is a good
> thing. We only deal with FULL_NOHZ in hardirq context (looking at
> v4.0-RT). Before that (FULL_NOHZ) everything was handled in softirq and
> nobody was concerned (saw a problem).
> 
> It would be good if we would address a regression or a bug or something
> like that.

The irq_work users may need to be audited. But I understand that it may
not be a regression. I'm still worried that something will come along
that will be affected by this with strange side effects.


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