From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ira Snyder Subject: Re: [PATCH RFC v5] net: add PCINet driver Date: Thu, 15 Jan 2009 15:31:31 -0800 Message-ID: <20090115233131.GE9091@ovro.caltech.edu> References: <20090107195052.GA24981@ovro.caltech.edu> <200901152222.55006.arnd@arndb.de> <20090115214056.GC9091@ovro.caltech.edu> <200901152353.25052.arnd@arndb.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linuxppc-dev@ozlabs.org, Jan-Bernd Themann , netdev@vger.kernel.org, Rusty Russell , linux-kernel@vger.kernel.org, shemminger@vyatta.com, David Miller To: Arnd Bergmann Return-path: Received: from ovro.ovro.caltech.edu ([192.100.16.2]:53134 "EHLO ovro.ovro.caltech.edu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S935062AbZAOXbd (ORCPT ); Thu, 15 Jan 2009 18:31:33 -0500 Content-Disposition: inline In-Reply-To: <200901152353.25052.arnd@arndb.de> Sender: netdev-owner@vger.kernel.org List-ID: On Thu, Jan 15, 2009 at 11:53:24PM +0100, Arnd Bergmann wrote: > On Thursday 15 January 2009, Ira Snyder wrote: > > I do have mailboxes (two inbound, two outbound) which can generate > > interrupts, as well as doorbell registers (one inbound, one outbound). > > The doorbell register's bits are "write 1 to clear", and can only be > > cleared by the opposite side. > > > > Ok, in this case, you should be able to use one of these to implement > the DMA complete interrupt, but adding an extra DMA transfer to the > mailbox after each transfer that you want to get notified for. Yep. I wouldn't have come up with that idea on my own, but it seems obvious now :) It seems a little strange to keep the DMA descriptors host memory (the DMA controller will need to issue a PCI read to get them) but I guess that is how real network cards work. Thanks, Ira