[PATCH 3/3] cpufreq: Add a generic cpufreq-cpu0 driver
From: Shawn Guo <hidden>
Date: 2012-07-21 06:38:13
Also in:
linux-devicetree
Possibly related (same subject, not in this thread)
- 2012-07-31 · Re: [PATCH 3/3] cpufreq: Add a generic cpufreq-cpu0 driver · Mark Brown <hidden>
- 2012-07-31 · Re: [PATCH 3/3] cpufreq: Add a generic cpufreq-cpu0 driver · Shawn Guo <hidden>
- 2012-07-30 · Re: [PATCH 3/3] cpufreq: Add a generic cpufreq-cpu0 driver · Mark Brown <hidden>
- 2012-07-30 · Re: [PATCH 3/3] cpufreq: Add a generic cpufreq-cpu0 driver · Mark Brown <hidden>
- 2012-07-30 · Re: [PATCH 3/3] cpufreq: Add a generic cpufreq-cpu0 driver · Shawn Guo <hidden>
On Sat, Jul 21, 2012 at 10:34:29AM +0530, Shilimkar, Santosh wrote:
On Fri, Jul 20, 2012 at 9:20 PM, Turquette, Mike [off-list ref] wrote:quoted
How about cpufreq-single-thread.c and CONFIG_CPUFREQ_SINGLE_THREAD? That makes sense for both UP and SMP.Indeed. This sounds more appropriate and also reflects what actually happens with a UP or shared clock SMP case.
While I agree with this observation, the suggested naming does not reflect the rationale of CPU0 though, which is really important for driver to work, and is exactly the thing I like about *_CPU0 naming. The driver needs the following stuff around CPU0 to be functional. - Device tree node /cpus/cpu at 0 - clk lookup with dev_id being "cpu0" - regulator with id being "cpu0" I think the *_CPU0 naming does a better job on emphasising those in the first place. -- Regards, Shawn