From mboxrd@z Thu Jan 1 00:00:00 1970 From: Suparna Bhattacharya Subject: [PATCH] Filesystem aio rdwr patchset Date: Tue, 1 Apr 2003 21:59:57 +0530 Sender: linux-fsdevel-owner@vger.kernel.org Message-ID: <20030401215957.A1800@in.ibm.com> Reply-To: suparna@in.ibm.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-fsdevel@vger.kernel.org, linux-aio@kvack.org, linux-kernel@vger.kernel.org Return-path: To: bcrl@redhat.com, akpm@digeo.com Content-Disposition: inline List-Id: linux-fsdevel.vger.kernel.org Have taken a first pass at implementing the write path for filesystem aio. Attached as a response to this mail is the full patchset for filesystem aio (retry based model) including read and write paths. 01aioretry.patch : this is the common generic aio retry code 02aiordwr.patch : this is the filesystem read+write changes for aio using the retry model 03aiobread.patch : code for async breads which can be used by filesystems for providing async get block implementation 04ext2-aiogetblk.patch : an async get block implementation for ext2 I would really appreciate comments and review feedback from the perspective of fs developers especially on the latter 2 patches in terms of whether this seems a sound approach or if I'm missing something very crucial (which I just well might be) Is this easy to do for other filesystems as well ? Regards Suparna -- Suparna Bhattacharya (suparna@in.ibm.com) Linux Technology Center IBM Software Labs, India