From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755347Ab3KKT51 (ORCPT ); Mon, 11 Nov 2013 14:57:27 -0500 Received: from mx1.redhat.com ([209.132.183.28]:26948 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753476Ab3KKT5T (ORCPT ); Mon, 11 Nov 2013 14:57:19 -0500 Date: Mon, 11 Nov 2013 20:58:36 +0100 From: Oleg Nesterov To: Ingo Molnar Cc: Ingo Molnar , Srikar Dronamraju , linux-kernel@vger.kernel.org Subject: Re: [PATCH] uprobes: Cleanup !CONFIG_UPROBES decls, unexport xol_area Message-ID: <20131111195836.GA20615@redhat.com> References: <20131109190344.GA32281@redhat.com> <20131111084149.GC12405@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20131111084149.GC12405@gmail.com> User-Agent: Mutt/1.5.18 (2008-05-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 11/11, Ingo Molnar wrote: > > * Oleg Nesterov wrote: > > > +++ b/kernel/events/uprobes.c > > @@ -86,6 +86,25 @@ struct return_instance { > > }; > > > > /* > > + * On a breakpoint hit, thread contests for a slot. It frees the > > + * slot after singlestep. Currently a fixed number of slots are > > + * allocated. > > + */ > > +struct xol_area { > > So, my main complaint about the uprobes code isn't functional but > documentational, similar to what I outlined a few days ago: what this > comment does not explain is exactly what a 'XOL area' is. > > You guys are changing code that reads like gobbledygook to people reading > it for the first time. Not that I am trying to defense uprobes, but this is equally true for any piece of kernel code, at least to me ;) > It's understandable that you want to use > abbreviations and I don't object against that, but please explain key > concepts and data structures when they first come up Well, this patch only move the definition with the comments, but: > - a very good place > to do that is in places where key structures are declared. > > I didn't find any high level description of the XOL code, one which makes > clear that how we manage these out of line execution areas: I have to agree, all these comments do not really help... > The one that comes closest is: > > * This area will be used for storing instructions for execution out of line. > > ... but that is a single sentence and deep inside the XOL code already. and even this comment should be probably moved up to the "struct xol_area", > Really, please make a better job of introducing other kernel hackers to > the code you are writing ... > > Maybe even split the XOL code out into kernel/events/uprobes_xol.c or so? I do not really think a separate uprobes_xol.c makes sense. I think it would be nice to have the high-level "uprobes design" doc in uprobetracer.txt, or > That will give a natural place to explain yourselves at the beginning of > the file. or even in the beginning of uprobes.c, I agree. Don't get me wrong, I am not volunteering ;) But at least I'll try to pay more attention to the comments when I change the code next time. Oleg.