From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pavel Machek Subject: Re: [PATCH] input: CPU frequency booster Date: Mon, 12 Jan 2015 14:32:48 +0100 Message-ID: <20150112133248.GC15838@amd> References: <1326977615-4387-1-git-send-email-amiettinen@nvidia.com> <201201192353.00761.trenn@suse.de> Mime-Version: 1.0 Return-path: Content-Disposition: inline In-Reply-To: <201201192353.00761.trenn@suse.de> Sender: cpufreq-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Thomas Renninger Cc: Antti P Miettinen , davej@redhat.com, rjw@sisk.pl, len.brown@intel.com, khilman@ti.com, j-pihet@ti.com, markgross@thegnar.org, cpufreq@vger.kernel.org, linux-pm@lists.linux-foundation.org Hi! > I can't see why pm_qos is needed at all, better use cpufreq > providing interface. > PPC (BIOS request to cut max_frequency) in > drivers/acpi/processor_perflib.c is a nice example > > I'd also put this into drivers/cpufreq/cpufreq_input_boost.c. > > Find below a re-written version of the booster. > Instead of pm_qos, it simply uses cpufreq interface: > /sys/devices/system/cpu/cpufreq/input_boost_freq > /sys/devices/system/cpu/cpufreq/input_boost_time > > I started with global cpufreq variables, they could also > be declared per cpu like scaling_min_freq or others > (at least input_boost_freq, input_boost_time should probably > stay global). That would be: > /sys/devices/system/cpu/cpuX/cpufreq/input_boost_freq Going through old mails. This did not seem to go anywhere? Is it still required? I don't see it in kernel. For some reason, it looks like it has potential to help Nokia N900 a lot... Pavel -- (english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html