From mboxrd@z Thu Jan 1 00:00:00 1970 From: Artem Bityutskiy Subject: Re: [PATCH 2/2] ubifs: Allow O_DIRECT Date: Thu, 20 Aug 2015 14:29:42 +0300 Message-ID: <1440070182.31419.200.camel@gmail.com> References: <1440016553-26481-1-git-send-email-richard@nod.at> <1440016553-26481-2-git-send-email-richard@nod.at> Reply-To: dedekind1@gmail.com Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, Dongsheng Yang To: Richard Weinberger , linux-mtd@lists.infradead.org Return-path: In-Reply-To: <1440016553-26481-2-git-send-email-richard@nod.at> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Wed, 2015-08-19 at 22:35 +0200, Richard Weinberger wrote: > Currently UBIFS does not support direct IO, but some applications > blindly use the O_DIRECT flag. > Instead of failing upon open() we can do better and fall back > to buffered IO. > > Cc: Dongsheng Yang > Cc: dedekind1@gmail.com > Suggested-by: Dave Chinner > Signed-off-by: Richard Weinberger Richard, The idea was to explicitly reject what we do not support. Let's say I am an app which requires O_DIRECT, and which does not want to work with non-O_DIRECT. What would I do to ensure O_DIRECT? Could you please check what other file-systems which do not support O_DIRECT do in this case? Do they also fall-back to normal IO instead of explicitly failing? If yes, we can do what is considered to be the "standard" behavior. Thanks!