Re: [RFC PATCH] cpufreq/hotplug: Fix cpu-hotplug cpufreq race conditions

[Viresh Kumar]

On 02-06-15, 11:01, Preeti U Murthy wrote:
> How will a policy lock help here at all, when cpus from multiple
> policies are calling into __cpufreq_governor() ? How will a policy lock
> serialize their entry into cpufreq_governor_dbs() ?

So different policies don't really depend on each other. The only
thing common to them are the governor's sysfs files (only if
governor-per-policy isn't set, i.e. in your case). Those sysfs files
and their kernel counterpart variables aren't touched unless all the
policies have EXITED. All these START/STOP calls touch only the data
relevant to those policies only.

In case of per-policy  governors, even those sysfs files are separate
for each policy.

And so a policy lock should be sufficient, rest should be handled
within the governors with locks or whatever.

> > These band-aid wouldn't take us anywhere.
> 
> Why do you say that the approach mentioned in this patch is a bandaid ?
> The patch ensures that there are no interruptions in a logical sequence
> of calls into cpufreq_governor_dbs(), as it should be.

Because this happened as we are forced to drop the policy-locks.
That's the real problem. This whole thing should be performed under
locks, instead of setting variables to mark governor busy under locks.

-- 
viresh
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/