From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751272AbdA1BHq (ORCPT ); Fri, 27 Jan 2017 20:07:46 -0500 Received: from mail-pg0-f48.google.com ([74.125.83.48]:33468 "EHLO mail-pg0-f48.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751124AbdA1BH3 (ORCPT ); Fri, 27 Jan 2017 20:07:29 -0500 From: John Stultz To: lkml Cc: Russell King , David Riley , Thomas Gleixner , Richard Cochran , Prarit Bhargava , Ingo Molnar , John Stultz Subject: [PATCH 2/3] delay: Add explanation of udelay() inaccuracy Date: Fri, 27 Jan 2017 16:59:30 -0800 Message-Id: <1485565171-21223-3-git-send-email-john.stultz@linaro.org> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1485565171-21223-1-git-send-email-john.stultz@linaro.org> References: <1485565171-21223-1-git-send-email-john.stultz@linaro.org> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Russell King There seems to be some misunderstanding that udelay() and friends will always guarantee the specified delay. This is a false understanding. When udelay() is based on CPU cycles, it can return early for many reasons which are detailed by Linus' reply to me in a thread in 2011: http://lists.openwall.net/linux-kernel/2011/01/12/372 However, a udelay test module was created in 2014 which allows udelay() to only be 0.5% fast, which is outside of the CPU-cycles udelay() results I measured back in 2011, which were deemed to be in the "we don't care" region. test_udelay() should be fixed to reflect the real allowable tolerance on udelay(), rather than 0.5%. Cc: David Riley Cc: Thomas Gleixner Cc: Richard Cochran Cc: Prarit Bhargava Cc: Ingo Molnar Signed-off-by: Russell King Signed-off-by: John Stultz --- include/linux/delay.h | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/include/linux/delay.h b/include/linux/delay.h index a6ecb34..2ecb3c4 100644 --- a/include/linux/delay.h +++ b/include/linux/delay.h @@ -5,6 +5,17 @@ * Copyright (C) 1993 Linus Torvalds * * Delay routines, using a pre-computed "loops_per_jiffy" value. + * + * Please note that ndelay(), udelay() and mdelay() may return early for + * several reasons: + * 1. computed loops_per_jiffy too low (due to the time taken to + * execute the timer interrupt.) + * 2. cache behaviour affecting the time it takes to execute the + * loop function. + * 3. CPU clock rate changes. + * + * Please see this thread: + * http://lists.openwall.net/linux-kernel/2011/01/09/56 */ #include -- 2.7.4