From mboxrd@z Thu Jan 1 00:00:00 1970 From: broonie@opensource.wolfsonmicro.com (Mark Brown) Date: Tue, 5 Jun 2012 11:20:40 +0100 Subject: [PATCH v2 REBASE] ASoC: fsl_ssi: convert to use devm_clk_get In-Reply-To: <20120605101558.GJ8262@n2100.arm.linux.org.uk> References: <1338774174-3960-1-git-send-email-richard.zhao@freescale.com> <20120605064714.GL30400@pengutronix.de> <20120605100316.GG23408@opensource.wolfsonmicro.com> <20120605100851.GI8262@n2100.arm.linux.org.uk> <20120605101413.GJ23408@opensource.wolfsonmicro.com> <20120605101558.GJ8262@n2100.arm.linux.org.uk> Message-ID: <20120605102040.GL23408@opensource.wolfsonmicro.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Tue, Jun 05, 2012 at 11:15:59AM +0100, Russell King - ARM Linux wrote: > On Tue, Jun 05, 2012 at 11:14:13AM +0100, Mark Brown wrote: > > That'd work too, of course (providing everyone has been implementing > > clk_put()). Though getting everyone on the same page with the clock API > > seems much more worthwhile if we can do it. > Why wouldn't clk_put() be implemented? Drivers call it just like clk_get(). Lazy people who can't be bothered, especially on platforms that do things like just return a pointer to a statically allocated clock or never really expect their drivers to be unloaded. Hopefully everyone filled in the blank but I've just got a general pessimism about relying on things like that. > The devm*() stuff should just be a layer on top of clk_get()/clk_put() and > nothing more. It is, yes. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 836 bytes Desc: Digital signature URL: