From mboxrd@z Thu Jan 1 00:00:00 1970 From: skannan@codeaurora.org (Saravana Kannan) Date: Wed, 16 Jul 2014 12:42:46 -0700 Subject: [PATCH v3 1/2] cpufreq: Don't destroy/realloc policy/sysfs on hotplug/suspend In-Reply-To: References: <1405464473-3916-1-git-send-email-skannan@codeaurora.org> <1405464473-3916-2-git-send-email-skannan@codeaurora.org> <53C68C54.3010900@gmail.com> Message-ID: <53C6D5B6.808@codeaurora.org> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On 07/16/2014 08:28 AM, Viresh Kumar wrote: > On 16 July 2014 19:59, Dirk Brandewie wrote: >> stop_cpu() only needs to be called during __cpufreq_remove_dev_prepare() no >> where else. > > Oh, thanks for reminding us.. > > Look at this Saravana: > 367dc4a cpufreq: Add stop CPU callback to cpufreq_driver interface > I'll only get called at the same time as it is called today. __cpufreq_remove_dev_prepare is now renamed to __cpufreq_remove_dev. And this function is called from there. The only time stop does get called is when __cpufreq_remove_dev is called on the last CPU of a policy. So, functionally it's identical. Btw, I already added logs to all cpufreq driver ops and checked the calls come in the same order with and without my changes. -Saravana -- The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum, hosted by The Linux Foundation