From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jason Gunthorpe Subject: Re: IB: increase DMA max_segment_size on Mellanox hardware Date: Mon, 17 Jan 2011 22:22:29 -0700 Message-ID: <20110118052229.GC2691@obsidianresearch.com> References: <1295230184.2574.26.camel@obelisk.thedillows.org> <4D342510.80701@voltaire.com> <1295318783.3051.81.camel@obelisk.thedillows.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <1295318783.3051.81.camel-1q1vX8mYZiGLUyTwlgNVppKKF0rrzTr+@public.gmane.org> Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: David Dillow Cc: Or Gerlitz , linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Roland Dreier , Eli Cohen List-Id: linux-rdma@vger.kernel.org On Mon, Jan 17, 2011 at 09:46:23PM -0500, David Dillow wrote: > To be sure, this is not going to be a huge efficiency gain, but it does > reduce CPU usage and raises the largest possible IO size with the > default srp_sg_tablesize of 12 from 768 KB to some 24 GB. Not that we'd > likely need that large of a request or be able to allocate 12 2GB > contiguous memory areas, of course. AH! So this is why.. I went to some trouble to add huge page support to some code expecting to get bigger/more consisent SRP IOs and didn't quite get that result :) Jason -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html