Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752787AbaDVS0U (ORCPT ); Tue, 22 Apr 2014 14:26:20 -0400 Received: from mail-ee0-f51.google.com ([74.125.83.51]:47468 "EHLO mail-ee0-f51.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750906AbaDVS0T (ORCPT ); Tue, 22 Apr 2014 14:26:19 -0400 Date: Tue, 22 Apr 2014 20:26:13 +0200 From: Ingo Molnar To: Drew Richardson Cc: Ingo Molnar , Peter Zijlstra , "linux-kernel@vger.kernel.org" , Pawel Moll , Christoph Hellwig Subject: Re: [PATCH] Export sched tracepoints for use by kernel modules Message-ID: <20140422182613.GA16359@gmail.com> References: <20140422142313.GB21121@dreric01-Precision-T1600> <20140422144503.GA15892@gmail.com> <20140422145338.GA4591@dreric01-Precision-T1600> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20140422145338.GA4591@dreric01-Precision-T1600> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org * Drew Richardson wrote: > > > +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_stat_sleep); > > > +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_stat_iowait); > > > +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_stat_blocked); > > > +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_stat_runtime); > > > +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_pi_setprio); > > > +#ifdef CONFIG_DETECT_HUNG_TASK > > > +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_process_hang); > > > +#endif /* CONFIG_DETECT_HUNG_TASK */ > > > +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_move_numa); > > > +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_stick_numa); > > > +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_swap_numa); > > > > Which in-tree kernel module is using these? > > > > Thanks, > > > > Ingo > > > > Gator (http://git.linaro.org/arm/ds5/gator.git), an out of tree > kernel module, uses sched_process_fork, sched_switch and > sched_process_free. Along with exporting those tracepoints I thought > it would be helpful to export all the tracepoints from > trace/events/sched.h. So, 'Gator' seems to be a rather ugly out of tree driver, ABI and daemon slapped on top of perf events, duplicating part of perf events functionality for no apparent good technical reason, just to enable a binary-only profiler called 'ARM Streamline Analyzer'. What does the 'gator' driver offer that perf events does not offer in itself? If there's any sensible feature in there then it should be added to the mainline kernel not as an add on driver and filesystem, but as part of perf events subsystem itself. Thanks, Ingo -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/