From mboxrd@z Thu Jan 1 00:00:00 1970 Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759201Ab1CaTPK (ORCPT ); Thu, 31 Mar 2011 15:15:10 -0400 Received: from usul.saidi.cx ([204.11.33.34]:56764 "EHLO usul.overt.org" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1759164Ab1CaTPH (ORCPT ); Thu, 31 Mar 2011 15:15:07 -0400 X-Greylist: delayed 1624 seconds by postgrey-1.27 at vger.kernel.org; Thu, 31 Mar 2011 15:15:07 EDT To: Chris Ball Subject: Re: SDHCI: Ricoh [1180:e823] unable to mount MMC cards. X-PHP-Originating-Script: 501:func.inc MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Date: Thu, 31 Mar 2011 11:47:46 -0700 From: Philip Langdale Cc: Manoj , , , Maxim Levitsky In-Reply-To: References: <4D93AD5C.1070308@canonical.com> <4D93BDA9.1020107@canonical.com> <4D948BC4.6020904@canonical.com> <4D949298.4000701@canonical.com> <4D94A510.4040509@canonical.com> <4D94B552.8020108@canonical.com> <4D94C0E5.6070801@canonical.com> Message-ID: User-Agent: RoundCube Webmail/0.4.2 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 31 Mar 2011 14:33:19 -0400, Chris Ball wrote: > Hi Manoj, > > On Thu, Mar 31 2011, Manoj wrote: >> Card did not mount. New dmesg is attached. > > Okay. There's nothing in the log about Ricoh, which probably means > that > it bailed out early in the function. You could add a printk to the > top > of ricoh_mmc_fixup_rl5c476() if you want to be really sure that the > code > there is getting a chance to run. > > I think the likely conclusion is that: > > (a) The Ricoh controller has a proprietary MMC function that is > stealing > MMC cards away from the SDHCI controller even though the SDHCI > could > read them if it had a chance to. > > (b) We don't know how to turn the MMC controller off, because our old > method for doing so isn't working on this new 0xe823 model. > > I'm adding a few people who've touched this code to CC: in case they > have any ideas on working out how to disable the MMC function. If > you > have contact with Ricoh directly or through a vendor, please ask them > for the PCI configuration writes needed to disable the MMC function > on > this model so that you can read MMC cards with the SD interface > instead. For what it's worth, I previously investigated a Dell e6400 with an 0x822 and an 0x823 and even without the MMC function disabled, the SD function would get the SD card. The machine I currently have access to just has the 0x822 function so I doubt it can be used to prove anything. I was under the impression they'd changed the priority of the controllers (and probably requiring the MMC driver to set a flag to make it steal cards) --phil