From mboxrd@z Thu Jan 1 00:00:00 1970 From: arnd@arndb.de (Arnd Bergmann) Date: Fri, 11 Feb 2011 16:25:13 +0100 Subject: MMC quirks relating to performance/lifetime. In-Reply-To: References: <201102111551.15508.arnd@arndb.de> Message-ID: <201102111625.14226.arnd@arndb.de> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Friday 11 February 2011, Lei Wen wrote: > > Several orders of magnitude. It is very easy to get a card that can write > > 12 MB/s into a case where it writes no more than 30 KB/s, doing only > > things that happen frequently with ext3. > > > > Maybe we could get that case into mmc_test code, so that we could track > that in latter whether it already be fixed or not? Or in other word, to prove > the firmware in sd card is stupid or not. :) There are many kinds of stupid, and a lot of cards are. I've actually had excellent success with simply measuring from user space, which is much easier than in mmc_test. Unfortunately, you have to write to the card to do that, which may destroy the data even if you write the same data that is already on it. See https://wiki.linaro.org/WorkingGroups/KernelConsolidation/Projects/FlashCardSurvey for most of my results. I'm about to write up a better paper with all the measurements, and will make my tools available soon. Arnd