Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753618AbbFDNaq (ORCPT ); Thu, 4 Jun 2015 09:30:46 -0400 Received: from casper.infradead.org ([85.118.1.10]:45146 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753529AbbFDNaD (ORCPT ); Thu, 4 Jun 2015 09:30:03 -0400 Date: Thu, 4 Jun 2015 15:29:55 +0200 From: Peter Zijlstra To: Imre Palik Cc: Thomas Gleixner , "H. Peter Anvin" , x86@kernel.org, Paul Mackerras , Ingo Molnar , Arnaldo Carvalho de Melo , linux-kernel@vger.kernel.org, "Palik, Imre" , Anthony Liguori Subject: Re: [RFC PATCH] perf: honoring cpuid for number of fixed counters Message-ID: <20150604132955.GJ3644@twins.programming.kicks-ass.net> References: <1433318628-6330-1-git-send-email-imrep.amz@gmail.com> <20150603083615.GZ3644@twins.programming.kicks-ass.net> <557029DC.6080201@gmail.com> <20150604114917.GI3644@twins.programming.kicks-ass.net> <557044ED.8060508@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <557044ED.8060508@gmail.com> User-Agent: Mutt/1.5.21 (2012-12-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2035 Lines: 55 On Thu, Jun 04, 2015 at 02:30:37PM +0200, Imre Palik wrote: > The trouble is that the number of fixed counters is not taken into > account when scheduling the events, and the cpu model based event > constraints will favour fixed counters. So perf tries to use them. Ah! so that is what your hunk below does. Tricky, and without comment that. --- diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 87848eb..eaa0b5e 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -647,6 +647,8 @@ static void perf_sched_init(struct perf_sched *sched, struct perf_event **events sched->state.event = idx; /* start with min weight */ sched->state.weight = wmin; sched->state.unassigned = num; + sched->state.used[0] = + ~0UL << (INTEL_PMC_IDX_FIXED + x86_pmu.num_counters_fixed); } static void perf_sched_save_state(struct perf_sched *sched) --- Please change the FIXED_EVENT constraints init instead; that way validate_event() will actually work too, otherwise it thinks it can schedule the fixed function only events. That is, change the below loop from intel_pmu_init(): if (x86_pmu.event_constraints) { /* * event on fixed counter2 (REF_CYCLES) only works on this * counter, so do not extend mask to generic counters */ for_each_event_constraint(c, x86_pmu.event_constraints) { if (c->cmask != FIXED_EVENT_FLAGS || c->idxmsk64 == INTEL_PMC_MSK_FIXED_REF_CYCLES) { continue; } c->idxmsk64 |= (1ULL << x86_pmu.num_counters) - 1; c->weight += x86_pmu.num_counters; } } To clear all counters that are not in fact present, that way we keep the event constraints correct instead of working around invalid constraints. -- 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/