From mboxrd@z Thu Jan 1 00:00:00 1970 From: Theodore Ts'o Subject: Re: Unable to randomly boot rootfs (EXT3 or EXT4) from SD: MMC error -110 (TIMEDOUT) Date: Wed, 28 Aug 2013 13:56:22 -0400 Message-ID: <20130828175622.GA27079@thunk.org> References: <20130827220143.GC9182@thunk.org> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-mmc-owner@vger.kernel.org To: Kevyn-Alexandre =?iso-8859-1?Q?Par=E9?= Cc: linux-omap@vger.kernel.org, linux-mmc@vger.kernel.org, linux-ext4@vger.kernel.org, arnd@linaro.org List-Id: linux-omap@vger.kernel.org On Wed, Aug 28, 2013 at 11:22:58AM -0400, Kevyn-Alexandre Par=E9 wrote: > We try multiple of our system ( Different micro SD from SanDisk + > Overo with OMAP3) and they are all doing this with the SanDisk 64GB, > 32GB : > http://www.sandisk.com/products/memory-cards/microsd/ultra-class10-fo= r-android/?capacity=3D64GB >=20 > The problem seem to be easier to reproduce when we just flash the > micro SD (partition, format & copy rootfs) at it first boot. > Also, when we let the system down for a couple of days and then > sometimes we need to reboot the system multiple times before it fall > back alright. >=20 > We found a old patch that was simply increasing the timeout of the ho= st: > http://cgit.openembedded.org/cgit.cgi/openembedded/tree/recipes/linux= /linux-omap-2.6.39/sakoman/0024-omap-mmc-Adjust-dto-to-eliminate-timeou= t-errors.patch?h=3Dorg.openembedded.dev >=20 > By settings the timeout to zero I reproduce kind of the same timeout > error at every boot. That was expected since I don't let enough time > to the controller to respond... >=20 > Could the mix of that MMC controller and these card is not optimal? I= s > the driver fully supporting the UHS-I speed grade? I've cc'ed Arnd Bergman who works at Linaro; I suspect that if he doesn't know the answer off the top of his head, he can probably direct you to someone who will. - Ted