[CPUFREQ] Prevent p4-clockmod from auto-binding to the ondemand governor.
authorDave Jones <davej@redhat.com>
Thu, 5 Mar 2009 05:16:26 +0000 (00:16 -0500)
committerDave Jones <davej@redhat.com>
Thu, 5 Mar 2009 05:16:26 +0000 (00:16 -0500)
The latency of p4-clockmod sucks so hard that scaling on a regular
basis with ondemand is a really bad idea.

Signed-off-by: Matthew Garrett <mjg59@srcf.ucam.org>
Signed-off-by: Dave Jones <davej@redhat.com>
arch/x86/kernel/cpu/cpufreq/p4-clockmod.c

index 1778402305e04a20702df18258b98a0ac593bb2a..352cf9a491649a1144899d37fbf732b8b7b84c88 100644 (file)
@@ -246,7 +246,10 @@ static int cpufreq_p4_cpu_init(struct cpufreq_policy *policy)
        cpufreq_frequency_table_get_attr(p4clockmod_table, policy->cpu);
 
        /* cpuinfo and default policy values */
-       policy->cpuinfo.transition_latency = 1000000; /* assumed */
+
+       /* the transition latency is set to be 1 higher than the maximum
+        * transition latency of the ondemand governor */
+       policy->cpuinfo.transition_latency = 10000001;
        policy->cur = stock_freq;
 
        return cpufreq_frequency_table_cpuinfo(policy, &p4clockmod_table[0]);