From mboxrd@z Thu Jan 1 00:00:00 1970 From: Konstantin Dorfman Subject: Re: [PATCH v1] mmc: fix async request mechanism for sequential read scenarios Date: Sun, 28 Oct 2012 14:52:51 +0200 Message-ID: <508D2AA3.8080902@codeaurora.org> References: <5088206C.7080101@stericsson.com> <50893E88.9000908@codeaurora.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from wolverine02.qualcomm.com ([199.106.114.251]:59426 "EHLO wolverine02.qualcomm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752209Ab2J1Mwy (ORCPT ); Sun, 28 Oct 2012 08:52:54 -0400 In-Reply-To: Sender: linux-mmc-owner@vger.kernel.org List-Id: linux-mmc@vger.kernel.org To: Venkatraman S Cc: =?ISO-8859-1?Q?Per_F=F6rlin?= , Per Forlin , "cjb@laptop.org" , "linux-mmc@vger.kernel.org" Hello Venkatraman, You are very welcome with any ideas, On 10/26/2012 01:55 PM, Venkatraman S wrote: > > Being nitpicky, I think it contradicts with the commit log that you have > for the patch.. > > When the block layer notifies the MMC layer on a new request, we check > for the above case where MMC layer is waiting on a previous request > completion and the current request is NULL. > > Can you be more specific? What is contradiction? Can you suggest better patch description? What I'm trying to say here: the patch will modify block layer notification (mmc_request() callback) on new incoming request in such way, that when MMC layer is blocked by waiting for completion of previous request (in core/core.c) - it will unblock MMC layer, rollback in mmc_queue_thread code up to blk_fetch() point. This will permit to the MMC layer start processing of newly notified request immediately (and not after completion of currently running request). Does it make sense? -- Konstantin Dorfman, QUALCOMM ISRAEL, on behalf of Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation