From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752552Ab2AVX6c (ORCPT ); Sun, 22 Jan 2012 18:58:32 -0500 Received: from ozlabs.org ([203.10.76.45]:34299 "EHLO ozlabs.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752499Ab2AVX6b (ORCPT ); Sun, 22 Jan 2012 18:58:31 -0500 From: Rusty Russell To: Steven Rostedt , Ingo Molnar Cc: LKML , Andrew Morton , Frederic Weisbecker Subject: Re: [PATCH][GIT PULL][v3.3] tracing: Add header wrappers event_headers_start.h and event_headers_end.h In-Reply-To: <1326909412.17534.91.camel@gandalf.stny.rr.com> References: <1326754637.7642.177.camel@gandalf.stny.rr.com> <20120117095418.GG10397@elte.hu> <1326807145.17534.26.camel@gandalf.stny.rr.com> <20120118120711.GB14863@elte.hu> <1326909412.17534.91.camel@gandalf.stny.rr.com> User-Agent: Notmuch/0.6.1-1 (http://notmuchmail.org) Emacs/23.3.1 (i686-pc-linux-gnu) Date: Mon, 23 Jan 2012 09:29:27 +1030 Message-ID: <877h0jgx80.fsf@rustcorp.com.au> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 18 Jan 2012 12:56:52 -0500, Steven Rostedt wrote: > On Wed, 2012-01-18 at 13:07 +0100, Ingo Molnar wrote: > > > I'd argue that __raise_softirq_irqoff() should not be inline - > > that would solve a whole host of issues. Event tracing is > > enabled in most distros so there's no real overhead argument to > > be made here either - so it's probably a bit faster in fact to > > have this uninlined. What do you think? > > Sure, I have no problem moving that out of the header. I don't think > raising a softirq is in that critical a path that it can't be a function > call. > > There's only one trace/events header left which is in module.h. Perhaps > we can move __module_get() and try_module_get() out of the header. We > could just move the "if" part out. Agreed. Since GCC should be able to eliminate that branch in almost all cases, since it's usually a literal NULL or address of a (non-weak) symbol. Be interesting to see the before/after sizes with this out-of-line. Thanks, Rusty.