2020-02-26 21:40:09

by Rafael J. Wysocki

Subject: [PATCH] cpufreq: Fix policy initialization for internal governor drivers

From: Rafael J. Wysocki <[email protected]>

Before commit 1e4f63aecb53 ("cpufreq: Avoid creating excessively
large stack frames") the initial value of the policy field in struct
cpufreq_policy set by the driver's ->init() callback was implicitly
passed from cpufreq_init_policy() to cpufreq_set_policy() if the
default governor was neither "performance" nor "powersave". After
that commit, however, cpufreq_init_policy() must take that case into
consideration explicitly and handle it as appropriate, so make that
happen.

Fixes: 1e4f63aecb53 ("cpufreq: Avoid creating excessively large stack frames")
Link: https://lore.kernel.org/linux-pm/[email protected]/
Reported-by: Artem Bityutskiy <[email protected]>
Cc: 5.4+ <[email protected]> # 5.4+
Signed-off-by: Rafael J. Wysocki <[email protected]>
---

Note that I'm going to fast-track this patch for 5.6-rc4 which means that
it will go into my linux-next branch as soon as it shows up in the lists.

Thanks and sorry for the breakage!

---
drivers/cpufreq/cpufreq.c | 14 ++++++++++++--
1 file changed, 12 insertions(+), 2 deletions(-)

Index: linux-pm/drivers/cpufreq/cpufreq.c
===================================================================
--- linux-pm.orig/drivers/cpufreq/cpufreq.c
+++ linux-pm/drivers/cpufreq/cpufreq.c
@@ -1076,9 +1076,17 @@ static int cpufreq_init_policy(struct cp
pol = policy->last_policy;
} else if (def_gov) {
pol = cpufreq_parse_policy(def_gov->name);
- } else {
- return -ENODATA;
+ /*
+ * In case the default governor is neiter "performance"
+ * nor "powersave", fall back to the initial policy
+ * value set by the driver.
+ */
+ if (pol == CPUFREQ_POLICY_UNKNOWN)
+ pol = policy->policy;
}
+ if (pol != CPUFREQ_POLICY_PERFORMANCE &&
+ pol != CPUFREQ_POLICY_POWERSAVE)
+ return -ENODATA;
}

return cpufreq_set_policy(policy, gov, pol);

2020-02-27 05:19:51

by Viresh Kumar

Subject: Re: [PATCH] cpufreq: Fix policy initialization for internal governor drivers

On 26-02-20, 22:39, Rafael J. Wysocki wrote:
> From: Rafael J. Wysocki <[email protected]>
>
> Before commit 1e4f63aecb53 ("cpufreq: Avoid creating excessively
> large stack frames") the initial value of the policy field in struct
> cpufreq_policy set by the driver's ->init() callback was implicitly
> passed from cpufreq_init_policy() to cpufreq_set_policy() if the
> default governor was neither "performance" nor "powersave". After
> that commit, however, cpufreq_init_policy() must take that case into
> consideration explicitly and handle it as appropriate, so make that
> happen.
>
> Fixes: 1e4f63aecb53 ("cpufreq: Avoid creating excessively large stack frames")
> Link: https://lore.kernel.org/linux-pm/[email protected]/
> Reported-by: Artem Bityutskiy <[email protected]>
> Cc: 5.4+ <[email protected]> # 5.4+
> Signed-off-by: Rafael J. Wysocki <[email protected]>
> ---
>
> Note that I'm going to fast-track this patch for 5.6-rc4 which means that
> it will go into my linux-next branch as soon as it shows up in the lists.
>
> Thanks and sorry for the breakage!
>
> ---
> drivers/cpufreq/cpufreq.c | 14 ++++++++++++--
> 1 file changed, 12 insertions(+), 2 deletions(-)
>
> Index: linux-pm/drivers/cpufreq/cpufreq.c
> ===================================================================
> --- linux-pm.orig/drivers/cpufreq/cpufreq.c
> +++ linux-pm/drivers/cpufreq/cpufreq.c
> @@ -1076,9 +1076,17 @@ static int cpufreq_init_policy(struct cp
> pol = policy->last_policy;
> } else if (def_gov) {
> pol = cpufreq_parse_policy(def_gov->name);
> - } else {
> - return -ENODATA;
> + /*
> + * In case the default governor is neiter "performance"
> + * nor "powersave", fall back to the initial policy
> + * value set by the driver.
> + */
> + if (pol == CPUFREQ_POLICY_UNKNOWN)
> + pol = policy->policy;
> }
> + if (pol != CPUFREQ_POLICY_PERFORMANCE &&
> + pol != CPUFREQ_POLICY_POWERSAVE)
> + return -ENODATA;
> }
>
> return cpufreq_set_policy(policy, gov, pol);

Acked-by: Viresh Kumar <[email protected]>

--
viresh

2020-02-27 08:29:38

by Artem Bityutskiy

Subject: Re: [PATCH] cpufreq: Fix policy initialization for internal governor drivers

On Wed, 2020-02-26 at 22:39 +0100, Rafael J. Wysocki wrote:
> From: Rafael J. Wysocki <[email protected]>
>
> Before commit 1e4f63aecb53 ("cpufreq: Avoid creating excessively
> large stack frames") the initial value of the policy field in struct
> cpufreq_policy set by the driver's ->init() callback was implicitly
> passed from cpufreq_init_policy() to cpufreq_set_policy() if the
> default governor was neither "performance" nor "powersave". After
> that commit, however, cpufreq_init_policy() must take that case into
> consideration explicitly and handle it as appropriate, so make that
> happen.
>
> Fixes: 1e4f63aecb53 ("cpufreq: Avoid creating excessively large stack frames")
> Link: https://lore.kernel.org/linux-pm/[email protected]/
> Reported-by: Artem Bityutskiy <[email protected]>
> Cc: 5.4+ <[email protected]> # 5.4+
> Signed-off-by: Rafael J. Wysocki <[email protected]>

Many thanks, fixes the issue.

Artem.