[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20160915154701.GN27974@sirena.org.uk>
Date: Thu, 15 Sep 2016 16:47:01 +0100
From: Mark Brown <broonie@...nel.org>
To: Viresh Kumar <viresh.kumar@...aro.org>
Cc: Lee Jones <lee.jones@...aro.org>, linaro-kernel@...ts.linaro.org,
Krzysztof Kozłowski <k.kozlowski@...sung.com>,
patches@...nsource.wolfsonmicro.com, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/2] mfd: wm8994-core: Don't use managed regulator bulk
get API
On Thu, Sep 15, 2016 at 08:38:44PM +0530, Viresh Kumar wrote:
> On 15-09-16, 15:47, Mark Brown wrote:
> > On Thu, Sep 15, 2016 at 04:47:01PM +0530, Viresh Kumar wrote:
> > > Fix the kernel warnings and crashes by moving away from managed
> > > regulator bulk get API by using regulator_bulk_get() and explicitly
> > > calling regulator_put() for all the supplies in exit paths.
> > Moving away from regulator bulk get to regulator bulk get?
> Moving away from "managed" regulator bulk get to regulator bulk get?
> Isn't this how we call the devm_* APIs as?
That's really unclearly worded (especially since one uses the function
name and the other doesn't).
> > > - ret = devm_regulator_bulk_get(wm8994->dev, wm8994->num_supplies,
> > > + ret = regulator_bulk_get(wm8994->dev, wm8994->num_supplies,
> > > wm8994->supplies);
> > > +err_regulator_put:
> > > + for (i = wm8994->num_supplies - 1; i >= 0; i--)
> > > + regulator_put(wm8994->supplies[i].consumer);
> > Why are you unwinding regulator_bulk_get() with an open coded
> > regulator_bulk_put()?
> Because there is no regulator_bulk_put(). Should I add it?
I mean regulator_bulk_free() there.
Download attachment "signature.asc" of type "application/pgp-signature" (474 bytes)
Powered by blists - more mailing lists