From mboxrd@z Thu Jan 1 00:00:00 1970 From: Charles Keepax Subject: Re: [alsa-devel] [PATCH v5 2/2] ASoC: samsung: Add machine driver for Exynos5433 based TM2 board Date: Fri, 2 Sep 2016 11:56:16 +0100 Message-ID: <20160902105616.GF21682@localhost.localdomain> References: <1470752514-11409-1-git-send-email-s.nawrocki@samsung.com> <1470752514-11409-2-git-send-email-s.nawrocki@samsung.com> <20160829150251.GU21682@localhost.localdomain> <65238f9c-bed0-7ad5-1961-fa2ff3848782@samsung.com> <2f5ac8bf-34da-7ce7-931c-86a8dad61dfc@samsung.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Return-path: Content-Disposition: inline In-Reply-To: <2f5ac8bf-34da-7ce7-931c-86a8dad61dfc@samsung.com> Sender: linux-samsung-soc-owner@vger.kernel.org To: Sylwester Nawrocki Cc: devicetree@vger.kernel.org, alsa-devel@alsa-project.org, linux-samsung-soc@vger.kernel.org, b.zolnierkie@samsung.com, k.kozlowski@samsung.com, broonie@kernel.org, Inha Song List-Id: devicetree@vger.kernel.org On Fri, Sep 02, 2016 at 12:49:51PM +0200, Sylwester Nawrocki wrote: > On 08/30/2016 04:57 PM, Sylwester Nawrocki wrote: > > I will use prepare/complete as with late_suspend/early_resume power > > sequences are wrong as you point out. I still need to debug why > > there is no errors reported when the SPI controller is suspended > > while there are supposed to be done SPI bus transfers. > > As it turns out the reason there is no any errors with late_suspend/ > early_resume is that the CODEC sets its regmap's "cache only" only flag > in its suspend() callback and then clears it in resume(). So any SPI > transfers which might happen when SPI controller is suspended > are prevented this way. Ah of course yeah should have realised that, but yeah you still end up with the writes not actually reaching the hardware. Thanks, Charles