Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S964866AbdHYW5M (ORCPT ); Fri, 25 Aug 2017 18:57:12 -0400 Received: from mail-pg0-f53.google.com ([74.125.83.53]:35878 "EHLO mail-pg0-f53.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S964852AbdHYW5K (ORCPT ); Fri, 25 Aug 2017 18:57:10 -0400 From: John Stultz To: lkml Cc: John Stultz , Thomas Gleixner , Ingo Molnar , Miroslav Lichvar , Richard Cochran , Prarit Bhargava , Stephen Boyd , Kevin Brodsky , Will Deacon , Daniel Mentz , Chris Wilson Subject: [RFC][PATCH] time: Fix ktime_get_raw() issues caused by incorrect base accumulation Date: Fri, 25 Aug 2017 15:57:04 -0700 Message-Id: <1503701824-1645-1-git-send-email-john.stultz@linaro.org> X-Mailer: git-send-email 2.7.4 In-Reply-To: <150366841319.27971.6041120504203143444@mail.alporthouse.com> References: <150366841319.27971.6041120504203143444@mail.alporthouse.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2595 Lines: 71 In commit fc6eead7c1e2 ("time: Clean up CLOCK_MONOTONIC_RAW time handling"), I mistakenly added the following: /* Update the monotonic raw base */ seconds = tk->raw_sec; nsec = (u32)(tk->tkr_raw.xtime_nsec >> tk->tkr_raw.shift); tk->tkr_raw.base = ns_to_ktime(seconds * NSEC_PER_SEC + nsec); Which adds the raw_sec value and the shifted down raw xtime_nsec to the base value. This is problematic as when calling ktime_get_raw(), we add the tk->tkr_raw.xtime_nsec and current offset, shift it down and add it to the raw base. This results in the shifted down tk->tkr_raw.xtime_nsec being added twice. My mistake, was that I was matching the monotonic base logic above: seconds = (u64)(tk->xtime_sec + tk->wall_to_monotonic.tv_sec); nsec = (u32) tk->wall_to_monotonic.tv_nsec; tk->tkr_mono.base = ns_to_ktime(seconds * NSEC_PER_SEC + nsec); Which adds the wall_to_monotonic.tv_nsec value, but not the tk->tkr_mono.xtime_nsec value to the base. The result of this is that ktime_get_raw() users (which are all internal users) see the raw time move faster then it should (the rate at which can vary with the current size of tkr_raw.xtime_nsec), which has resulted in at least problems with graphics rendering performance. To fix this, we simplify the tkr_raw.base accumulation to only accumulate the raw_sec portion, and do not include the tkr_raw.xtime_nsec portion, which will be added at read time. Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Miroslav Lichvar Cc: Richard Cochran Cc: Prarit Bhargava Cc: Stephen Boyd Cc: Kevin Brodsky Cc: Will Deacon Cc: Daniel Mentz Cc: Chris Wilson Reported-by: Chris Wilson Signed-off-by: John Stultz --- kernel/time/timekeeping.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index cedafa0..7e7e61c 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -637,9 +637,7 @@ static inline void tk_update_ktime_data(struct timekeeper *tk) tk->ktime_sec = seconds; /* Update the monotonic raw base */ - seconds = tk->raw_sec; - nsec = (u32)(tk->tkr_raw.xtime_nsec >> tk->tkr_raw.shift); - tk->tkr_raw.base = ns_to_ktime(seconds * NSEC_PER_SEC + nsec); + tk->tkr_raw.base = ns_to_ktime(tk->raw_sec * NSEC_PER_SEC); } /* must hold timekeeper_lock */ -- 2.7.4