From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chris Ball Subject: Re: [PATCH v6 13/14] mmc: sdhci: solve several vmmc/vqmmc regulator issues Date: Sat, 17 Nov 2012 18:02:55 -0500 Message-ID: <87r4nr7rz4.fsf@octavius.laptop.org> References: <1350471893-29633-1-git-send-email-keyuan.liu@gmail.com> <1350471893-29633-14-git-send-email-keyuan.liu@gmail.com> Mime-Version: 1.0 Content-Type: text/plain Return-path: Received: from void.printf.net ([89.145.121.20]:54562 "EHLO void.printf.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752420Ab2KQXDC (ORCPT ); Sat, 17 Nov 2012 18:03:02 -0500 In-Reply-To: <1350471893-29633-14-git-send-email-keyuan.liu@gmail.com> (Kevin Liu's message of "Wed, 17 Oct 2012 19:04:52 +0800") Sender: linux-mmc-owner@vger.kernel.org List-Id: linux-mmc@vger.kernel.org To: Kevin Liu Cc: linux-mmc@vger.kernel.org, pierre@ossman.eu, ulf.hansson@linaro.org, zgao6@marvell.com, hzhuang1@marvell.com, cxie4@marvell.com, prakity@marvell.com, kliu5@marvell.com Hi, On Wed, Oct 17 2012, Kevin Liu wrote: > From: Kevin Liu > > 1. The vmmc regulator enable in sdhci_add_host is NOT necessary since > it can be enabled during mmc_power_up by function mmc_regulator_set_ocr. > And this extra enable will make regulator_enable/regulator_disable > unbalanced. Consequently, vmmc can't be disabled during mmc_power_off. > > 2. If regulator vqmmc exist, it should be enabled regardless it support > 1.8v or not. > > 3. If regulator framework is disabled, regulator_get will return NULL. > So the CONFIG_REGULATOR judgement can be removed. Pushed to mmc-next for 3.8 without "3." above, because it causes a compile error -- now we use regulator_count_voltages() inside that block and it's not a stub function in the regulator API, so the ifdef is required. - Chris. -- Chris Ball One Laptop Per Child