On Sunday, March 24, 2013 08:23:27 PM Viresh Kumar wrote:
On 24 March 2013 20:07, Francesco Lavra francescolavra.fl@gmail.com wrote:
On 03/24/2013 02:48 PM, Viresh Kumar wrote:
policy->cpus contains all online cpus that have single shared clock line. And their frequencies are always updated together.
Many SMP system's cpufreq drivers take care of this in individual drivers but the best place for this code is in cpufreq core.
This patch modifies cpufreq_notify_transition() to notify frequency change for all cpus in policy->cpus and hence updates all users of this API.
One thing about this work. I compiled it for ARM and Intel. Also this stuff is tested by "Fengguang Wu" fengguang.wu@intel.com automated build system.
I am not sure if that builds all architectures or not. I tried to review my patch closely but their can be some minor mistakes.
I thought of adding this in the patch details but forgot at last.
Is their a simple way to compile stuff for all platforms? Sorry i am not aware of it :(
diff --git a/arch/blackfin/mach-common/cpufreq.c b/arch/blackfin/mach-common/cpufreq.c
ret = cpu_set_cclk(policy->cpu, freqs.new * 1000);if (ret != 0) {WARN_ONCE(ret, "cpufreq set freq failed %d\n", ret);break;This doesn't even compile, as the break statement isn't in the for_each_online_cpu() loop anymore.
I tried to review it very carefully but this situation was a bit tricky :) Thanks for trying it out.
Following should fix it for you:
commit 942ca8a6bc87e3c42beabc9102755136493e5355 Author: Viresh Kumar viresh.kumar@linaro.org Date: Sun Mar 24 20:21:43 2013 +0530
fixup! cpufreq: Notify all policy->cpus in cpufreq_notify_transition()
arch/blackfin/mach-common/cpufreq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/blackfin/mach-common/cpufreq.c b/arch/blackfin/mach-common/cpufreq.c index 4e67368..995511e80 100644 --- a/arch/blackfin/mach-common/cpufreq.c +++ b/arch/blackfin/mach-common/cpufreq.c @@ -164,7 +164,7 @@ static int bfin_target(struct cpufreq_policy *policy, ret = cpu_set_cclk(policy->cpu, freqs.new * 1000); if (ret != 0) { WARN_ONCE(ret, "cpufreq set freq failed %d\n", ret);
break;
return ret; }#endif on_each_cpu(bfin_adjust_core_timer, &index, 1);
@Rafael: Let me add fixups for now, i will send final patch later after others also review their part.
OK