From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: [PATCH 1/3] fs: allow short direct-io reads to be completed via buffered IO V2 Date: Thu, 6 May 2010 17:10:12 -0400 Message-ID: <20100506211012.GD2997@infradead.org> References: <20100506190012.GB13974@dhcp231-156.rdu.redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-btrfs@vger.kernel.org, linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org, linux-mm@kvack.org To: Josef Bacik Return-path: In-Reply-To: <20100506190012.GB13974@dhcp231-156.rdu.redhat.com> List-ID: On Thu, May 06, 2010 at 03:00:13PM -0400, Josef Bacik wrote: > V1->V2: Check to see if our current ppos is >= i_size after a short DIO read, > just in case it was actually a short read and we need to just return. > > This is similar to what already happens in the write case. If we have a short > read while doing O_DIRECT, instead of just returning, fallthrough and try to > read the rest via buffered IO. BTRFS needs this because if we encounter a > compressed or inline extent during DIO, we need to fallback on buffered. If the > extent is compressed we need to read the entire thing into memory and > de-compress it into the users pages. I have tested this with fsx and everything > works great. Thanks, This seems safe to me, but I'm a bit worried about potential breakages. Did you test this with xfsqa on xfs and ext3/4 to make sure there are no regressions? -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org