[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Pine.LNX.4.44L0.0611161654270.2460-100000@iolanthe.rowland.org>
Date:	Thu, 16 Nov 2006 16:56:35 -0500 (EST)
From:	Alan Stern <stern@...land.harvard.edu>
To:	Thomas Gleixner <tglx@...esys.com>
cc:	LKML <linux-kernel@...r.kernel.org>,
	Linus Torvalds <torvalds@...l.org>,
	john stultz <johnstul@...ibm.com>, Ingo Molnar <mingo@...e.hu>,
	David Miller <davem@...emloft.net>,
	Arjan van de Ven <arjan@...radead.org>,
	Andrew Morton <akpm@...l.org>, Andi Kleen <ak@...e.de>
Subject: Re: BUG: cpufreq notification broken
On Thu, 16 Nov 2006, Thomas Gleixner wrote:
> On Thu, 2006-11-16 at 16:26 -0500, Alan Stern wrote:
> > On Thu, 16 Nov 2006, Thomas Gleixner wrote:
> > 
> > > There is another issue with this SRCU change:
> > > 
> > > The notification comes actually after the real change, which is bad. We
> > > try to make the TSC usable by backing it with pm_timer accross such
> > > states, but this behaviour breaks the safety code.
> > 
> > I don't understand.  Sending notifications is completely separate from 
> > setting up the notifier chain's head.  The patch you mentioned didn't 
> > touch the code that sends the notifications.
> 
> Yeah, my bad. It just uses rcu based locking, but its still synchronous.
> 
> I have to dig deeper, why the change of the frequency happens _before_
> the notifier arrives.
There are supposed to be _two_ notifier calls: one before the frequency 
change and one after.  Check the callers of cpufreq_notify_transition().
Alan Stern
-
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/
Powered by blists - more mailing lists
 
