From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chris Ball Subject: Re: [PATCH 3/3] mmc: sdio: don't power up cards on system suspend Date: Fri, 17 Dec 2010 01:00:27 +0000 Message-ID: <20101217010026.GF17239@void.printf.net> References: <1290921690-6133-1-git-send-email-ohad@wizery.com> <1290921690-6133-4-git-send-email-ohad@wizery.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from void.printf.net ([89.145.121.20]:46084 "EHLO void.printf.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753542Ab0LQBA2 (ORCPT ); Thu, 16 Dec 2010 20:00:28 -0500 Content-Disposition: inline In-Reply-To: <1290921690-6133-4-git-send-email-ohad@wizery.com> Sender: linux-mmc-owner@vger.kernel.org List-Id: linux-mmc@vger.kernel.org To: Ohad Ben-Cohen Cc: linux-mmc@vger.kernel.org, Chris Ball Hi Ohad, On Sun, Nov 28, 2010 at 07:21:30AM +0200, Ohad Ben-Cohen wrote: > Initial SDIO runtime PM implementation took a conservative approach > of powering up cards (and fully reinitializing them) on system suspend, > just before the suspend handlers of the relevant drivers were executed. > > To avoid redundant power and reinitialization cycles, this patch removes > this behavior: if a card is already powered off when system suspend kicks > in, it is left at that state. > > If a card is active when a system sleep starts, everything is > straightforward and works exactly like before. But if the card was > already suspended before the sleep began, then when MMC core power it up > back on resume, its run-time PM status has to be updated to reflect the > actual post-system sleep status. > > The technique to do that is borrowed from the I2C runtime PM > implementation (for more info also see Documentation/power/runtime_pm.txt). > > Signed-off-by: Ohad Ben-Cohen Thanks, pushed to mmc-next. -- Chris Ball One Laptop Per Child