From mboxrd@z Thu Jan 1 00:00:00 1970 From: Benjamin LaHaise Subject: Re: [RFC] iovec in ->aio_read/->aio_write Date: Tue, 15 Oct 2002 15:34:27 -0400 Sender: owner-linux-aio@kvack.org Message-ID: <20021015153427.A16156@redhat.com> References: <20021015223315.A21139@sgi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: To: Christoph Hellwig , akpm@digeo.com, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, linux-aio@kvack.org Content-Disposition: inline In-Reply-To: <20021015223315.A21139@sgi.com>; from hch@sgi.com on Tue, Oct 15, 2002 at 10:33:15PM -0400 List-Id: linux-fsdevel.vger.kernel.org On Tue, Oct 15, 2002 at 10:33:15PM -0400, Christoph Hellwig wrote: > Proposed next steps: Convert over all readv/writev users > to aio_read/aio_write and remove the methods. Implement > aio_read/aio_write in all filesystems using the generic > pagecache code and kill the "normal" generic_file_read > and generic_file_write. > > Comments? Please not right now? ;-) At least introduce it as aio_readv/aio_writev as currently the way we deal with iovecs uglifies a lot of code, with no benefit for the vast majority of applications. As for killing generic_file_read/write and using the aio counterparts, that is the plan pending a bit more testing of things. I want to get there step by step without breaking everything along the way. -ben -- To unsubscribe, send a message with 'unsubscribe linux-aio' in the body to majordomo@kvack.org. For more info on Linux AIO, see: http://www.kvack.org/aio/