Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
From:   Wang Qing <wangqing@vivo.com>
To:     Thomas Gleixner <tglx@linutronix.de>,
        Frederic Weisbecker <fweisbec@gmail.com>,
        Ingo Molnar <mingo@kernel.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Petr Mladek <pmladek@suse.com>, Wang Qing <wangqing@vivo.com>,
        Santosh Sivaraj <santosh@fossix.org>,
        "Peter Zijlstra (Intel)" <peterz@infradead.org>,
        linux-kernel@vger.kernel.org
Subject: [PATCH RFC 0/2] hrtimer: watchdog: support hrtimer suspend when CPU suspend
Date:   Tue, 22 Jun 2021 17:06:45 +0800
Message-Id: <1624352816-26450-1-git-send-email-wangqing@vivo.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Precedence: bulk

When the CPU suspend, in order to achieve a more power-saving effect,
it is hoped that the CPU sleeps as long as possible, but the timer is an 
important reason for the CPU to wake up. 

In some cases, when the CPU suspend, the timer doesn’t have to work either,
such as watchdog hrtimer. The maximum suspend time of the CPU is 4s if enable
lockup detector, which is unacceptable on some products, and in fact, watchdog
timer can suspend when the cpu suspend.

This is a patch for comments, I'm not sure if there any ill-considerations. 
If this feature is really needed, I will continue to develop.

Wang Qing (2):
  hrtimer: support hrtimer suspend when CPU suspend
  watchdog: support watchdog hrtimer suspend when CPU suspend

 include/linux/hrtimer.h  | 3 +++
 kernel/time/hrtimer.c    | 4 +++-
 kernel/time/tick-sched.c | 7 +++++++
 kernel/watchdog.c        | 3 ++-

-- 
2.7.4