Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753130AbcKTN0u (ORCPT ); Sun, 20 Nov 2016 08:26:50 -0500 Received: from Galois.linutronix.de ([146.0.238.70]:54710 "EHLO Galois.linutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751870AbcKTN0t (ORCPT ); Sun, 20 Nov 2016 08:26:49 -0500 Date: Sun, 20 Nov 2016 14:24:05 +0100 (CET) From: Thomas Gleixner To: Joel Fernandes cc: linux-kernel@vger.kernel.org, John Stultz , Steven Rostedt , "Rafael J . Wysocki" Subject: Re: [RFC] timekeeping: Use cached readouts for monotonic and raw clocks in suspend In-Reply-To: <1479591671-13397-1-git-send-email-joelaf@google.com> Message-ID: References: <1479591671-13397-1-git-send-email-joelaf@google.com> User-Agent: Alpine 2.20 (DEB 67 2015-01-07) 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 Content-Length: 412 Lines: 12 On Sat, 19 Nov 2016, Joel Fernandes wrote: > I am planning to add boot clock as a trace clock that can account suspend time > during tracing, however ktime_get_with_offset throws a warning as the > clocksource is attempted to be accessed in suspend. ktime_get_with_offset() cannot be used as trace clock at all because it can life lock in NMI context. That's why we have ktime_get_mono_fast(). Thanks, tglx