[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20140131221118.GE2502@saruman.home>
Date: Fri, 31 Jan 2014 16:11:19 -0600
From: Felipe Balbi <balbi@...com>
To: Alan Stern <stern@...land.harvard.edu>
CC: Felipe Balbi <balbi@...com>,
Santosh Shilimkar <santosh.shilimkar@...com>,
Grygorii Strashko <grygorii.strashko@...com>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
<ivan.khoronzhuk@...com>, <linux-usb@...r.kernel.org>,
<linux-arm-kernel@...ts.infradead.org>,
<linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] usb: dwc3: keystone: switch to use runtime pm
On Fri, Jan 31, 2014 at 04:13:19PM -0500, Alan Stern wrote:
> On Fri, 31 Jan 2014, Felipe Balbi wrote:
>
> > probe()
> > {
> > ...
> >
> > clk_get(dev, "fck");
> > clk_prepare(clk);
> > clk_enable(clk);
> > pm_runtime_set_active(dev);
> > pm_runtime_enable(dev);
> > pm_runtime_get_sync(dev);
> > ...
> > }
>
> > note that because of pm_runtime_set_active() that first
> > pm_runtime_get_sync() in probe() will simply increase the reference
> > counter without calling my ->runtime_resume() callback, which is exactly
> > what we want, as that would completely avoid situations of bad context
> > being restored because of that initial pm_runtime_get_sync().
>
> Very minor note... A slightly better way to do the same thing is:
>
> pm_runtime_set_active(dev);
> pm_runtime_get_noresume(dev);
> pm_runtime_enable(dev);
>
> The get_noresume says that you want to increment the usage counter
> without performing any callbacks, and doing it before the
> pm_runtime_enable avoids any window during which a runtime suspend
> might somehow occur.
aha, that's perfect :-) Thanks Alan.
--
balbi
Download attachment "signature.asc" of type "application/pgp-signature" (820 bytes)
Powered by blists - more mailing lists