From mboxrd@z Thu Jan 1 00:00:00 1970 From: Benjamin Herrenschmidt Subject: Re: [Cbe-oss-dev] [PATCH 4/10] spidernet: zero out a pointer. Date: Wed, 23 May 2007 13:23:27 +1000 Message-ID: <1179890607.32247.896.camel@localhost.localdomain> References: <20070516215605.GA4325@austin.ibm.com> <20070516220928.GD4997@austin.ibm.com> <1179358376.4142.4.camel@concordia.ozlabs.ibm.com> <20070522233517.GU5921@austin.ibm.com> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Cc: Michael Ellerman , netdev@vger.kernel.org, Jeff Garzik , cbe-oss-dev@ozlabs.org To: Linas Vepstas Return-path: Received: from gate.crashing.org ([63.228.1.57]:43608 "EHLO gate.crashing.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755258AbXEWDXx (ORCPT ); Tue, 22 May 2007 23:23:53 -0400 In-Reply-To: <20070522233517.GU5921@austin.ibm.com> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org > Here's a delusional reply: I didn't see any point to it. > 1) a wmb would add overhead > 2) the hardware is supposed to be looking at the status flag, > anyway, and not misbehaving. > 3) there is a wmb when the descr is actually refilled in such > a way as to actually mean something to the hardware. > > All that I really acomplished here is a minor trick to > aid in debug printing when looking for something bad. And the whole thing is moot because 0 is actually a perfectly valid DMA address :-) I suspect spider will end up trying to hit some internal register or it's PCIe or whatever it has mapped at 0 internally and will blow up ... At least on spider, it's not RAM there.... Ben.