Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757665AbbGQMpt (ORCPT ); Fri, 17 Jul 2015 08:45:49 -0400 Received: from casper.infradead.org ([85.118.1.10]:49249 "EHLO casper.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757520AbbGQMps (ORCPT ); Fri, 17 Jul 2015 08:45:48 -0400 Date: Fri, 17 Jul 2015 14:45:38 +0200 From: Peter Zijlstra To: "Wangnan (F)" Cc: kaixu xia , ast@plumgrid.com, davem@davemloft.net, acme@kernel.org, mingo@redhat.com, masami.hiramatsu.pt@hitachi.com, jolsa@kernel.org, linux-kernel@vger.kernel.org, pi3orama@163.com, hekuang@huawei.com Subject: Re: [RFC PATCH 5/6] bpf: Implement function bpf_read_pmu() that get the selected hardware PMU conuter Message-ID: <20150717124538.GI19282@twins.programming.kicks-ass.net> References: <1437129816-13176-6-git-send-email-xiakaixu@huawei.com> <20150717110541.GJ12596@twins.programming.kicks-ass.net> <55A8E703.70306@huawei.com> <20150717113924.GD19282@twins.programming.kicks-ass.net> <55A8EABE.1060308@huawei.com> <20150717115505.GF19282@twins.programming.kicks-ass.net> <20150717115615.GA18673@twins.programming.kicks-ass.net> <55A8EE83.3000708@huawei.com> <20150717121836.GH19282@twins.programming.kicks-ass.net> <55A8F4BF.3020902@huawei.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <55A8F4BF.3020902@huawei.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: 1079 Lines: 28 On Fri, Jul 17, 2015 at 08:27:43PM +0800, Wangnan (F) wrote: > I think we can check the limitation in BPF program. You typically do not want to rely on your program for correctness. > What about this: > > event must on current CPU or must be on current process. If not, > bpf_read_pmu() should simply return an error. OK, that's workable. That enforces the constraints outside of the program itself. > With current design it is easy to implement, and users can still control > it through bpf map. > > But what if we really want cross-cpu PMU accessing? Impossible? Under the assumption that the eBPF program is called from tracing, and therefore from any context (task, softirq, irq and nmi), yes impossible. You cannot do (synchronous) IPIs from either IRQ context or with IRQs disabled. And both are valid trace contexts. -- 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/