From owner-freebsd-hackers@FreeBSD.ORG Wed Dec 15 01:37:42 2004 Return-Path: Delivered-To: freebsd-hackers@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8757416A4CF for ; Wed, 15 Dec 2004 01:37:42 +0000 (GMT) Received: from rproxy.gmail.com (rproxy.gmail.com [64.233.170.194]) by mx1.FreeBSD.org (Postfix) with ESMTP id 06FCD43D49 for ; Wed, 15 Dec 2004 01:37:42 +0000 (GMT) (envelope-from sferris@gmail.com) Received: by rproxy.gmail.com with SMTP id a36so1355157rnf for ; Tue, 14 Dec 2004 17:37:41 -0800 (PST) DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:reply-to:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:references; b=eNd+1Gs/mCibGXHCMwX8L1Yvt9fXhyoubQ4lTOh7pDchs8tEbW06kGnFDG0XeENUW8DQiS0OCTvbEGSDoC/F8XInWdgblz+wsYBRwwJGdeRURci9xYmbIgpZQY2qJHMw8VzSDj+tANh+7Wdlwm0S4/i7W5BouTh5Tn79W14gdKI= Received: by 10.38.126.79 with SMTP id y79mr2265103rnc; Tue, 14 Dec 2004 17:37:41 -0800 (PST) Received: by 10.38.83.5 with HTTP; Tue, 14 Dec 2004 17:37:41 -0800 (PST) Message-ID: <1eea89cd04121417376b64086e@mail.gmail.com> Date: Tue, 14 Dec 2004 19:37:41 -0600 From: "Scott M. Ferris" To: Kip Macy In-Reply-To: Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit References: <1eea89cd041214114766fd34dc@mail.gmail.com> cc: hackers@freebsd.org cc: Peter Blok Subject: Re: My project wish-list for the next 12 months X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: "Scott M. Ferris" List-Id: Technical Discussions relating to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 15 Dec 2004 01:37:42 -0000 On Tue, 14 Dec 2004 17:05:25 -0800 (PST), Kip Macy wrote: > > Hardware-based iSCSI HBAs solve this by having their own memory and > > TCP stack separate from the OS. Software-only iSCSI initiators such > > as linux-iscsi usually just hope it doesn't happen, and that's why I > > don't usually recommend software-only iSCSI initiators to anyone. > > How is that any better than having the SW iscsi initiator pre-allocate a pool of > memory for its use at startup? You could preallocate, but the hard part would be getting the OS network stack to allocate from the iSCSI driver's preallocated memory pool, instead of calling the usual kernel memory allocator (or more likely as a fallback if the normal kernel allocation fails). How do you tell the TCP stack and NIC drivers to allocate control blocks, mbufs, and anything else they might need from the preallocated iSCSI pool? It's not a problem in theory, but in practice that's not something you usually see in the API for the network stack. -- Scott M. Ferris