From mboxrd@z Thu Jan 1 00:00:00 1970 From: Scott Branden Subject: Re: [PATCHv2 5/5] mmc: sdhci-bcm2835: add sdhci quirk SDHCI_QUIRK_MULTIBLOCK_READ_ACMD12 Date: Fri, 7 Nov 2014 10:30:45 -0800 Message-ID: <545D0FD5.1080605@broadcom.com> References: <1414651017-3545-1-git-send-email-sbranden@broadcom.com> <1414651017-3545-6-git-send-email-sbranden@broadcom.com> <5459AEEA.8050503@wwwdotorg.org> <5459CB76.3060601@broadcom.com> <545AFE2C.5050105@wwwdotorg.org> Mime-Version: 1.0 Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mail-gw1-out.broadcom.com ([216.31.210.62]:47147 "EHLO mail-gw1-out.broadcom.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752505AbaKGSbA (ORCPT ); Fri, 7 Nov 2014 13:31:00 -0500 In-Reply-To: <545AFE2C.5050105@wwwdotorg.org> Sender: linux-mmc-owner@vger.kernel.org List-Id: linux-mmc@vger.kernel.org To: Stephen Warren , Ulf Hansson , Russell King , Peter Griffin , Chris Ball , Piotr Krol Cc: linux-mmc@vger.kernel.org, linux-kernel@vger.kernel.org, Joe Perches , linux-rpi-kernel@lists.infradead.org, Ray Jui , bcm-kernel-feedback-list@broadcom.com On 14-11-05 08:50 PM, Stephen Warren wrote: > On 11/05/2014 12:02 AM, Scott Branden wrote: >> On 14-11-04 09:00 PM, Stephen Warren wrote: >>> On 10/30/2014 12:36 AM, Scott Branden wrote: >>>> SDHCI_QUIRK_MULTIBLOCK_READ_ACMD12 is missing and needed for this >>>> controller. >>> >>> This seems fine, although any explanation of why this quirk is needed >>> would be useful. >>> >> I don't know who to talk to at Arasan about this. Will try hunting >> around a little for more info as to why this is needed to have eMMC and >> SD work properly through our internal testing on other non-2835 chipset >> that shares the same SDHCI controller as 2835. > > I thought I heard that this wasn't a bug in the controller itself, but > rather an integration issue between the IP core and the register bus > it's attached to. Consequently, it may be SoC-specific or at least have > SoC-specific variations? Yes, this patch is to fix a different bug (in the IP) rather than the clock domain integration issue. >