From mboxrd@z Thu Jan 1 00:00:00 1970 From: Linus Torvalds Subject: Re: [GIT] Networking Date: Sun, 7 Jul 2013 14:27:54 -0700 Message-ID: References: <20130707.132133.1335860703202153192.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Cc: Andrew Morton , Network Development , Linux Kernel Mailing List To: David Miller Return-path: Received: from mail-ve0-f173.google.com ([209.85.128.173]:54365 "EHLO mail-ve0-f173.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753170Ab3GGV14 (ORCPT ); Sun, 7 Jul 2013 17:27:56 -0400 In-Reply-To: <20130707.132133.1335860703202153192.davem@davemloft.net> Sender: netdev-owner@vger.kernel.org List-ID: On Sun, Jul 7, 2013 at 1:21 PM, David Miller wrote: > > 5) arch/arm/mach-mxs/Kconfig > > Overlapping additions of select statements for SOC_IMX2. Nope. Removals. > 6) arch/arm/mach-mxs/mach-mxs.c > > In mxs_machine_init(), the of_platform_populate() call has it's final > two arguments changed to be NULL. Hmm. That's not how I read it. The next-to-last argument in changed to NULL in one branch, the other adds "parent". > 11) fs/select.c > > Overlapping additions of include directives, keep both. This LL_POLL thing is ugly as hell. Where was this discussed? Apparently on lkml without talking to people who are actually involved in select/poll. I'm not at all sure I want to pull this kind of crap without having seen the full discussion, and the performance numbers are kind of pointless without also talking about CPU utilization etc. Since when is it ok to start doing things like this without any acks from anybody else? Because quite frankly, the fs/select.c changes make me go: "No way in hell". Partly because of the idiotic and completely undescriptive naming, partly because of the disgisting calling convetions with random flags variables passed in to be changed be helper functions. So no, I'm not pulling this. Things like this need discussion and heads-up, not just random "let's add completely ugly things to core kernel functionality without even asking people". Even the commit messages in question are just PURE AND UTTER CRAP. Ask yourself honestly what the F*CK this means: net: convert lls to use time_in_range what the hell, David? What diseased mind uses idiotic short-hand like this that makes no sense? And dammit, I'd expect the VFS people to be cc'd on changes to the VFS layer functions. I wasn't, Al wasn't. So not pulled. Fix up this mess. Linus