From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752914Ab1A0PAG (ORCPT ); Thu, 27 Jan 2011 10:00:06 -0500 Received: from mailout-de.gmx.net ([213.165.64.22]:40913 "HELO mailout-de.gmx.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1752829Ab1A0PAC (ORCPT ); Thu, 27 Jan 2011 10:00:02 -0500 X-Authenticated: #911537 X-Provags-ID: V01U2FsdGVkX187wZoxD987BQ3DJGvEyjD1Hf79N45fUHeralhp+o /CLmRdlu7Pff9+ Subject: [PATCH v2 13/20] ia64: switch do_timer() to xtime_update() To: LKML From: Torben Hohn Cc: Fenghua Yu , Tony Luck , Peter Zijlstra , johnstul@us.ibm.com, hch@infradead.org, yong.zhang0@gmail.com, Thomas Gleixner Date: Thu, 27 Jan 2011 15:59:56 +0100 Message-ID: <20110127145956.23248.49107.stgit@localhost> In-Reply-To: <20110127145741.23248.68098.stgit@localhost> References: <20110127145741.23248.68098.stgit@localhost> User-Agent: StGit/0.15 MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7bit X-Y-GMX-Trusted: 0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org local_cpu_data->itm_next = new_itm; does not need to be protected by this lock. and xtime_update() takes the lock itself. Cc: Tony Luck Cc: Fenghua Yu Signed-off-by: Torben Hohn --- arch/ia64/kernel/time.c | 19 +++++-------------- arch/ia64/xen/time.c | 13 +++++-------- 2 files changed, 10 insertions(+), 22 deletions(-) diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c index 9702fa9..156ad80 100644 --- a/arch/ia64/kernel/time.c +++ b/arch/ia64/kernel/time.c @@ -190,19 +190,10 @@ timer_interrupt (int irq, void *dev_id) new_itm += local_cpu_data->itm_delta; - if (smp_processor_id() == time_keeper_id) { - /* - * Here we are in the timer irq handler. We have irqs locally - * disabled, but we don't know if the timer_bh is running on - * another CPU. We need to avoid to SMP race by acquiring the - * xtime_lock. - */ - write_seqlock(&xtime_lock); - do_timer(1); - local_cpu_data->itm_next = new_itm; - write_sequnlock(&xtime_lock); - } else - local_cpu_data->itm_next = new_itm; + if (smp_processor_id() == time_keeper_id) + xtime_update(1); + + local_cpu_data->itm_next = new_itm; if (time_after(new_itm, ia64_get_itc())) break; @@ -222,7 +213,7 @@ skip_process_time_accounting: * comfort, we increase the safety margin by * intentionally dropping the next tick(s). We do NOT * update itm.next because that would force us to call - * do_timer() which in turn would let our clock run + * xtime_update() which in turn would let our clock run * too fast (with the potentially devastating effect * of losing monotony of time). */ diff --git a/arch/ia64/xen/time.c b/arch/ia64/xen/time.c index c1c5445..1f8244a 100644 --- a/arch/ia64/xen/time.c +++ b/arch/ia64/xen/time.c @@ -139,14 +139,11 @@ consider_steal_time(unsigned long new_itm) run_posix_cpu_timers(p); delta_itm += local_cpu_data->itm_delta * (stolen + blocked); - if (cpu == time_keeper_id) { - write_seqlock(&xtime_lock); - do_timer(stolen + blocked); - local_cpu_data->itm_next = delta_itm + new_itm; - write_sequnlock(&xtime_lock); - } else { - local_cpu_data->itm_next = delta_itm + new_itm; - } + if (cpu == time_keeper_id) + xtime_update(stolen + blocked); + + local_cpu_data->itm_next = delta_itm + new_itm; + per_cpu(xen_stolen_time, cpu) += NS_PER_TICK * stolen; per_cpu(xen_blocked_time, cpu) += NS_PER_TICK * blocked; }