From mboxrd@z Thu Jan 1 00:00:00 1970 From: James Bottomley Subject: Re: [PATCH] [9/20] Add blk_kmalloc/blk_alloc_pages Date: Mon, 17 Mar 2008 08:59:22 -0500 Message-ID: <1205762362.6767.123.camel@localhost.localdomain> References: <20080307653.720459648@firstfloor.org> <20080307175408.E290E1B41AE@basil.firstfloor.org> <1205445980.2893.85.camel@localhost.localdomain> <20080314134814.GS17940@kernel.dk> <20080314135945.GO2522@one.firstfloor.org> <20080317082711.GB17940@kernel.dk> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from accolon.hansenpartnership.com ([76.243.235.52]:43980 "EHLO accolon.hansenpartnership.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752414AbYCQN7Z (ORCPT ); Mon, 17 Mar 2008 09:59:25 -0400 In-Reply-To: <20080317082711.GB17940@kernel.dk> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Jens Axboe Cc: Andi Kleen , linux-scsi@vger.kernel.org On Mon, 2008-03-17 at 09:27 +0100, Jens Axboe wrote: > > In my original patchkit I didn't have blk_kmalloc etc. but just relied > > on the block layer bouncing everything for ISA as needed, but James > > thought it was important that SCSI LUN scan etc do not bounce for ISA. > > That is when I came up with these helpers. > > > > You think it should go back to always bouncing? That would be fine > > for me too. With that they wouldn't be needed. > > I don't see a reason for not bouncing for scanning - James? If we didn't know the allocation mask, or it was hard to get, then yes, I'd agree. However, I think that any time the kernel has the device mask available, it should actually use it for an allocation to avoid bouncing. James