From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Subject: Re: [PATCH 9/11] lpfc 8.1.1 : Add polled-mode support Date: Wed, 30 Nov 2005 19:18:07 +0000 Message-ID: <20051130191807.GD19164@infradead.org> References: <9BB4DECD4CFE6D43AA8EA8D768ED51C2276C5F@xbl3.emulex.com> <20051129040244.GA28049@parisc-linux.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from pentafluge.infradead.org ([213.146.154.40]:60095 "EHLO pentafluge.infradead.org") by vger.kernel.org with ESMTP id S1751545AbVK3TSJ (ORCPT ); Wed, 30 Nov 2005 14:18:09 -0500 Content-Disposition: inline In-Reply-To: <20051129040244.GA28049@parisc-linux.org> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Matthew Wilcox Cc: James.Smart@Emulex.Com, hch@infradead.org, linux-scsi@vger.kernel.org On Mon, Nov 28, 2005 at 09:02:44PM -0700, Matthew Wilcox wrote: > On Mon, Nov 28, 2005 at 04:10:37PM -0500, James.Smart@Emulex.Com wrote: > > > > > This needs some documentation for where it is useful. Do you > > > want this > > > to avoid interrupts under high loads or for something else? > > > > Yes - we've been asked to support this mode so that a heavily loaded > > system can claim back some cpu cycles spent on interrupt processing. > > (SPC benchmarks being a prime example) > > sounds like we need to implement SAPI ;-) > > [ for those with no clue what i'm talking about, see > Documentation/networking/NAPI_HOWTO.txt ] Exactly what I was thinking when writing this previous mail.