From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mike Christie Subject: Re: [PATCH] rm alloc_page from dm-emc take 2 Date: Mon, 11 Apr 2005 04:48:12 -0700 Message-ID: <425A63FC.1060808@us.ibm.com> References: <425A6145.7010302@cs.wisc.edu> Reply-To: device-mapper development Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <425A6145.7010302@cs.wisc.edu> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: dm-devel-bounces@redhat.com Errors-To: dm-devel-bounces@redhat.com To: device-mapper development Cc: Lars Marowsky-Bree , Alasdair G Kergon List-Id: dm-devel.ids Mike Christie wrote: > I am not sure if the first mail got sent out. Here is > the correct mail with the patch. > > Hey Lars, > > Here is what I meant. A Page for just something around 8-16 bytes > seems wasteful. Plus there is no reserve. > > If you have to end up setting more bits maybe it is better to > preallocate a command per device at create time, or maybe since > there are four different combos I should have done it with this > patch? > > This was made against 2.6.12-rc2. Only compile tested. > > Mike > > + if (bio_add_page(bio, virt_to_page(data), data_size, > + offset_in_page(data)) != data_size) { My concern though is that I do no think I can do this. I mean we cannot start the IO from the middle of a page.