In the current code, if we fail during a frequency transition we simply send the POSTCHANGE notification with old frequency. This isn't enough.
One of the core user of these notifications is the code responsible for keeping loops_per_jiffy aligned with frequency change. And mostly it is written as:
if ((val == CPUFREQ_PRECHANGE && freq->old < freq->new) || (val == CPUFREQ_POSTCHANGE && freq->old > freq->new)) { update-loops-per-jiffy... }
So, suppose we are changing to a higher frequency and failed during transition, then following will happen: - CPUFREQ_PRECHANGE notification with freq-new > freq-old - CPUFREQ_POSTCHANGE notification with freq-new == freq-old
The first one will update loops_per_jiffy and second one will do nothing. Even if we send the 2nd notification by exchanging values of freq-new and old, some users of these notifications might get unstable.
This can be fixed by simply calling cpufreq_notify_post_transition() with error code and this routine will take care of sending notifications in correct order.
Also it returns a proper error message in case frequency transition failed.
Signed-off-by: Viresh Kumar viresh.kumar@linaro.org --- drivers/cpufreq/unicore2-cpufreq.c | 16 +++++++--------- 1 file changed, 7 insertions(+), 9 deletions(-)
diff --git a/drivers/cpufreq/unicore2-cpufreq.c b/drivers/cpufreq/unicore2-cpufreq.c index 951152f..99d280d6 100644 --- a/drivers/cpufreq/unicore2-cpufreq.c +++ b/drivers/cpufreq/unicore2-cpufreq.c @@ -38,19 +38,17 @@ static int ucv2_target(struct cpufreq_policy *policy, unsigned int target_freq, unsigned int relation) { - unsigned int cur = policy->cur; struct cpufreq_freqs freqs; + int ret;
- cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE); - - if (!clk_set_rate(policy->clk, target_freq * 1000)) { - freqs.old = cur; - freqs.new = target_freq; - } + freqs.old = policy->cur; + freqs.new = target_freq;
- cpufreq_notify_transition(policy, &freqs, CPUFREQ_POSTCHANGE); + cpufreq_notify_transition(policy, &freqs, CPUFREQ_PRECHANGE); + ret = clk_set_rate(policy->clk, target_freq * 1000); + cpufreq_notify_post_transition(policy, &freqs, ret);
- return 0; + return ret; }
static int __init ucv2_cpu_init(struct cpufreq_policy *policy)