* Re: [PATCH] cpufreq: Fix policy->freq_table is NULL in __cpufreq_driver_target()
2023-03-29 13:36 [PATCH] cpufreq: Fix policy->freq_table is NULL in __cpufreq_driver_target() Yajun Deng
@ 2023-03-29 14:21 ` Rafael J. Wysocki
2023-03-30 1:39 ` Yajun Deng
` (2 subsequent siblings)
3 siblings, 0 replies; 7+ messages in thread
From: Rafael J. Wysocki @ 2023-03-29 14:21 UTC (permalink / raw)
To: Yajun Deng; +Cc: rafael, viresh.kumar, linux-pm, linux-kernel
On Wed, Mar 29, 2023 at 3:36 PM Yajun Deng <yajun.deng@linux.dev> wrote:
>
> __resolve_freq() may be return target_freq if policy->freq_table is
> NULL. In this case, it should return -EINVAL before __target_index().
Even so, __target_index() itself doesn't dereference freq_table
AFAICS, so arguably the driver should be prepared to deal with a NULL
freq_table which comes from it after all.
Or, if you want to argue that drivers providing ->target_index() must
also provide freq_table that is not NULL, a check to that effect needs
to be done at the driver registration time IMO.
> Signed-off-by: Yajun Deng <yajun.deng@linux.dev>
> ---
> drivers/cpufreq/cpufreq.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
> index c0e5be0fe2d6..308a3df1a940 100644
> --- a/drivers/cpufreq/cpufreq.c
> +++ b/drivers/cpufreq/cpufreq.c
> @@ -2299,7 +2299,7 @@ int __cpufreq_driver_target(struct cpufreq_policy *policy,
> return cpufreq_driver->target(policy, target_freq, relation);
> }
>
> - if (!cpufreq_driver->target_index)
> + if (!cpufreq_driver->target_index || !policy->freq_table)
> return -EINVAL;
>
> return __target_index(policy, policy->cached_resolved_idx);
> --
^ permalink raw reply [flat|nested] 7+ messages in thread* Re: [PATCH] cpufreq: Fix policy->freq_table is NULL in __cpufreq_driver_target()
2023-03-29 13:36 [PATCH] cpufreq: Fix policy->freq_table is NULL in __cpufreq_driver_target() Yajun Deng
2023-03-29 14:21 ` Rafael J. Wysocki
@ 2023-03-30 1:39 ` Yajun Deng
2023-03-30 3:57 ` Viresh Kumar
2023-04-03 4:11 ` Viresh Kumar
2023-04-04 3:08 ` Yajun Deng
3 siblings, 1 reply; 7+ messages in thread
From: Yajun Deng @ 2023-03-30 1:39 UTC (permalink / raw)
To: Rafael J. Wysocki; +Cc: viresh.kumar, linux-pm, linux-kernel
March 29, 2023 10:21 PM, "Rafael J. Wysocki" <rafael@kernel.org> wrote:
> On Wed, Mar 29, 2023 at 3:36 PM Yajun Deng <yajun.deng@linux.dev> wrote:
>
>> __resolve_freq() may be return target_freq if policy->freq_table is
>> NULL. In this case, it should return -EINVAL before __target_index().
>
> Even so, __target_index() itself doesn't dereference freq_table
> AFAICS, so arguably the driver should be prepared to deal with a NULL
> freq_table which comes from it after all.
>
But there is a statement 'unsigned int newfreq = policy->freq_table[index].frequency;'
in __target_index(), if driver doesn't provide freq_table, __target_index()
will fault before the driver itself.
> Or, if you want to argue that drivers providing ->target_index() must
> also provide freq_table that is not NULL, a check to that effect needs
> to be done at the driver registration time IMO.
>
>> Signed-off-by: Yajun Deng <yajun.deng@linux.dev>
>> ---
>> drivers/cpufreq/cpufreq.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
>> index c0e5be0fe2d6..308a3df1a940 100644
>> --- a/drivers/cpufreq/cpufreq.c
>> +++ b/drivers/cpufreq/cpufreq.c
>> @@ -2299,7 +2299,7 @@ int __cpufreq_driver_target(struct cpufreq_policy *policy,
>> return cpufreq_driver->target(policy, target_freq, relation);
>> }
>>
>> - if (!cpufreq_driver->target_index)
>> + if (!cpufreq_driver->target_index || !policy->freq_table)
>> return -EINVAL;
>>
>> return __target_index(policy, policy->cached_resolved_idx);
>> --
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] cpufreq: Fix policy->freq_table is NULL in __cpufreq_driver_target()
2023-03-30 1:39 ` Yajun Deng
@ 2023-03-30 3:57 ` Viresh Kumar
2023-03-30 10:10 ` Rafael J. Wysocki
0 siblings, 1 reply; 7+ messages in thread
From: Viresh Kumar @ 2023-03-30 3:57 UTC (permalink / raw)
To: Yajun Deng; +Cc: Rafael J. Wysocki, linux-pm, linux-kernel
On 30-03-23, 01:39, Yajun Deng wrote:
> March 29, 2023 10:21 PM, "Rafael J. Wysocki" <rafael@kernel.org> wrote:
>
> > On Wed, Mar 29, 2023 at 3:36 PM Yajun Deng <yajun.deng@linux.dev> wrote:
> >
> >> __resolve_freq() may be return target_freq if policy->freq_table is
> >> NULL. In this case, it should return -EINVAL before __target_index().
> >
> > Even so, __target_index() itself doesn't dereference freq_table
> > AFAICS, so arguably the driver should be prepared to deal with a NULL
> > freq_table which comes from it after all.
> >
>
> But there is a statement 'unsigned int newfreq = policy->freq_table[index].frequency;'
> in __target_index(), if driver doesn't provide freq_table, __target_index()
> will fault before the driver itself.
Driver must provide a freq table here.
--
viresh
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] cpufreq: Fix policy->freq_table is NULL in __cpufreq_driver_target()
2023-03-30 3:57 ` Viresh Kumar
@ 2023-03-30 10:10 ` Rafael J. Wysocki
0 siblings, 0 replies; 7+ messages in thread
From: Rafael J. Wysocki @ 2023-03-30 10:10 UTC (permalink / raw)
To: Viresh Kumar; +Cc: Yajun Deng, Rafael J. Wysocki, linux-pm, linux-kernel
On Thu, Mar 30, 2023 at 5:57 AM Viresh Kumar <viresh.kumar@linaro.org> wrote:
>
> On 30-03-23, 01:39, Yajun Deng wrote:
> > March 29, 2023 10:21 PM, "Rafael J. Wysocki" <rafael@kernel.org> wrote:
> >
> > > On Wed, Mar 29, 2023 at 3:36 PM Yajun Deng <yajun.deng@linux.dev> wrote:
> > >
> > >> __resolve_freq() may be return target_freq if policy->freq_table is
> > >> NULL. In this case, it should return -EINVAL before __target_index().
> > >
> > > Even so, __target_index() itself doesn't dereference freq_table
> > > AFAICS, so arguably the driver should be prepared to deal with a NULL
> > > freq_table which comes from it after all.
> > >
> >
> > But there is a statement 'unsigned int newfreq = policy->freq_table[index].frequency;'
> > in __target_index(), if driver doesn't provide freq_table, __target_index()
> > will fault before the driver itself.
>
> Driver must provide a freq table here.
OK, so let's do the check when the driver gets registered.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] cpufreq: Fix policy->freq_table is NULL in __cpufreq_driver_target()
2023-03-29 13:36 [PATCH] cpufreq: Fix policy->freq_table is NULL in __cpufreq_driver_target() Yajun Deng
2023-03-29 14:21 ` Rafael J. Wysocki
2023-03-30 1:39 ` Yajun Deng
@ 2023-04-03 4:11 ` Viresh Kumar
2023-04-04 3:08 ` Yajun Deng
3 siblings, 0 replies; 7+ messages in thread
From: Viresh Kumar @ 2023-04-03 4:11 UTC (permalink / raw)
To: Yajun Deng; +Cc: rafael, linux-pm, linux-kernel
On 29-03-23, 21:36, Yajun Deng wrote:
> __resolve_freq() may be return target_freq if policy->freq_table is
> NULL. In this case, it should return -EINVAL before __target_index().
>
> Signed-off-by: Yajun Deng <yajun.deng@linux.dev>
> ---
> drivers/cpufreq/cpufreq.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
> index c0e5be0fe2d6..308a3df1a940 100644
> --- a/drivers/cpufreq/cpufreq.c
> +++ b/drivers/cpufreq/cpufreq.c
> @@ -2299,7 +2299,7 @@ int __cpufreq_driver_target(struct cpufreq_policy *policy,
> return cpufreq_driver->target(policy, target_freq, relation);
> }
>
> - if (!cpufreq_driver->target_index)
> + if (!cpufreq_driver->target_index || !policy->freq_table)
> return -EINVAL;
Hi,
I have sent an alternate patch [1] for this, please try it.
--
viresh
[1] https://lore.kernel.org/all/53d4ed4e5b18a59a48790434f8146fb207e11c49.1680494945.git.viresh.kumar@linaro.org/
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH] cpufreq: Fix policy->freq_table is NULL in __cpufreq_driver_target()
2023-03-29 13:36 [PATCH] cpufreq: Fix policy->freq_table is NULL in __cpufreq_driver_target() Yajun Deng
` (2 preceding siblings ...)
2023-04-03 4:11 ` Viresh Kumar
@ 2023-04-04 3:08 ` Yajun Deng
3 siblings, 0 replies; 7+ messages in thread
From: Yajun Deng @ 2023-04-04 3:08 UTC (permalink / raw)
To: Viresh Kumar; +Cc: rafael, linux-pm, linux-kernel
April 3, 2023 12:11 PM, "Viresh Kumar" <viresh.kumar@linaro.org> wrote:
> On 29-03-23, 21:36, Yajun Deng wrote:
>
>> __resolve_freq() may be return target_freq if policy->freq_table is
>> NULL. In this case, it should return -EINVAL before __target_index().
>>
>> Signed-off-by: Yajun Deng <yajun.deng@linux.dev>
>> ---
>> drivers/cpufreq/cpufreq.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
>> index c0e5be0fe2d6..308a3df1a940 100644
>> --- a/drivers/cpufreq/cpufreq.c
>> +++ b/drivers/cpufreq/cpufreq.c
>> @@ -2299,7 +2299,7 @@ int __cpufreq_driver_target(struct cpufreq_policy *policy,
>> return cpufreq_driver->target(policy, target_freq, relation);
>> }
>>
>> - if (!cpufreq_driver->target_index)
>> + if (!cpufreq_driver->target_index || !policy->freq_table)
>> return -EINVAL;
>
> Hi,
>
> I have sent an alternate patch [1] for this, please try it.
>
Thanks, v2 is fine.
> --
> viresh
>
> [1]
> https://lore.kernel.org/all/53d4ed4e5b18a59a48790434f8146fb207e11c49.1680494945.git.viresh.kumar@lin
> ro.org
^ permalink raw reply [flat|nested] 7+ messages in thread