From mboxrd@z Thu Jan 1 00:00:00 1970 From: David Miller Subject: Re: [net-next 00/36] bnx2x patch series Date: Wed, 12 Aug 2009 14:50:19 -0700 (PDT) Message-ID: <20090812.145019.231249971.davem@davemloft.net> References: <1250101162.27379.151.camel@lb-tlvb-eilong> Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, yanivr@broadcom.com, vladz@broadcom.com, gertner@broadcom.com, benli@broadcom.com To: eilong@broadcom.com Return-path: Received: from 74-93-104-97-Washington.hfc.comcastbusiness.net ([74.93.104.97]:49513 "EHLO sunset.davemloft.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751685AbZHLVuJ (ORCPT ); Wed, 12 Aug 2009 17:50:09 -0400 In-Reply-To: <1250101162.27379.151.camel@lb-tlvb-eilong> Sender: netdev-owner@vger.kernel.org List-ID: Come on... 36 patches? :-/ Please trickle changes in, don't send patch bombs. I think I've told you this not once, but several times. But I keep seeing these sizable patch sets. I frankly don't care that it might not mesh well with how you code up and validate changes internally, because it absolutely does NOT work well for how bugs really get found and fixed upstream. If you trickle changes in, the guilty change is obvious to spot and it gets found before you do more development that depends upon that buggy change. Whereas if you patch bomb, someone has to do a lot of work and bisecting to nail down the bad change. And the bug might be so fundamental in a patch that it also invalidates all the followon work you did. I really don't want to apply any of this stuff, to be honest with you. I simply don't. I keep giving guidelines and they keep getting ignored.