From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Madhusudhan Chikkature Rajashekar" Subject: MMC/SD cards hotplug scenario Date: Wed, 21 May 2008 11:42:04 +0530 Message-ID: <004e01c8bb09$989fac00$538918ac@ent.ti.com> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Return-path: Received: from calf.ext.ti.com ([198.47.26.144]:36121 "EHLO calf.ext.ti.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756289AbYEUGMq (ORCPT ); Wed, 21 May 2008 02:12:46 -0400 Sender: linux-omap-owner@vger.kernel.org List-Id: linux-omap@vger.kernel.org To: 'Pierre Ossman' Cc: linux-arm-kernel@lists.arm.linux.org.uk, linux-omap@vger.kernel.org Hi Pierre/All, I have a query regarding one of the MMC/SD cards hotplug use case scenario and its expected behavior. In the case where the MMC/SD card is removed physically from the slot during an active data transfer happening on the card, I notice lot of I/O errors being generated. The numbers of I/O errors seem to depend on the size of the file being copied(the errors semm to be a mixture of FS + MMC core errors). After the end of the I/O errors I can umount the partition that was mounted and I reinsert the card. It seem not to work very well consistently. I mean some times the card removal/insertion does seem to work that well. Whereas card insertion/removal works perfectly fine if tested when no data transfer is happening. Does MMC/SD core need to handle some things to support this scenario? Does MMC/SD core handle this scenario gracefully? Has anyone tried something similar before? As far as I understand this type of scenario can not be handled well in the controller drivers alone. Am I missing something here? Regards, Madhu