>On 3/29/2022 5:14 PM, Kant Fan wrote:
>> The member void *data in the structure devfreq can be overwrite
>> by governor_userspace.
>> Signed-off-by: Kant Fan <[email protected]>
>> ---
>> drivers/devfreq/governor_userspace.c | 6 +++++-
>> 1 file changed, 5 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/devfreq/governor_userspace.c b/drivers/devfreq/governor_userspace.c
>> index ab9db7adb3ad..dbbb448dcbcf 100644
>> --- a/drivers/devfreq/governor_userspace.c
>> +++ b/drivers/devfreq/governor_userspace.c
>> @@ -17,6 +17,7 @@
>> struct userspace_data {
>> unsigned long user_frequency;
>> bool valid;
>> + void *saved_data;
>> };
>>
>> static int devfreq_userspace_func(struct devfreq *df, unsigned long *freq)
>> @@ -91,6 +92,7 @@ static int userspace_init(struct devfreq *devfreq)
>> goto out;
>> }
>> data->valid = false;
>> + data->saved_data = devfreq->data;
>> devfreq->data = data;
>>
>> err = sysfs_create_group(&devfreq->dev.kobj, &dev_attr_group);
>> @@ -100,6 +102,8 @@ static int userspace_init(struct devfreq *devfreq)
>>
>> static void userspace_exit(struct devfreq *devfreq)
>> {
>> + struct userspace_data *data = devfreq->data;
>> + void *saved_data = data->saved_data;
>> /*
>> * Remove the sysfs entry, unless this is being called after
>> * device_del(), which should have done this already via kobject_del().
>> @@ -108,7 +112,7 @@ static void userspace_exit(struct devfreq *devfreq)
>> sysfs_remove_group(&devfreq->dev.kobj, &dev_attr_group);
>>
>> kfree(devfreq->data);
>> - devfreq->data = NULL;
>> + devfreq->data = saved_data;
>> }
>>
>> static int devfreq_userspace_handler(struct devfreq *devfreq,
>
>Dear MyungJoo, Kyungmin & Chanwoo,
>Gently ping this issue... Does this patch has a chance to be accepted?
>This seems to be a bug in devfreq userspace governor, which affects the
>switching between governors -- When switching from userspace to
>ondemand, the ondemand governor would be invalid.
>If there's any question, please let me know.
>Thank you.
Yes, indeed. This is a bug.
Actually, it appears that allocating a new memory buffer for
devfreq->data itself is a bug for a governor, this is supposed
to be allocated by a device driver. Thus, the comment of
"void *data" of "struct devfreq" should be updated:
"/* private data for governors given by device drivers */"
It'd be better to have something like, "void *internal_data"
for governors to freely handle within its context of init-exit,
which is not touched by its users (device drivers).
@Chanwoo: what's your opinion on this?
Cheers,
MyungJoo.
On 9/14/2022 5:43 PM, MyungJoo Ham wrote:
>> On 3/29/2022 5:14 PM, Kant Fan wrote:
>>> The member void *data in the structure devfreq can be overwrite
>>> by governor_userspace.
>>> Signed-off-by: Kant Fan <[email protected]>
>>> ---
>>> drivers/devfreq/governor_userspace.c | 6 +++++-
>>> 1 file changed, 5 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/drivers/devfreq/governor_userspace.c b/drivers/devfreq/governor_userspace.c
>>> index ab9db7adb3ad..dbbb448dcbcf 100644
>>> --- a/drivers/devfreq/governor_userspace.c
>>> +++ b/drivers/devfreq/governor_userspace.c
>>> @@ -17,6 +17,7 @@
>>> struct userspace_data {
>>> unsigned long user_frequency;
>>> bool valid;
>>> + void *saved_data;
>>> };
>>>
>>> static int devfreq_userspace_func(struct devfreq *df, unsigned long *freq)
>>> @@ -91,6 +92,7 @@ static int userspace_init(struct devfreq *devfreq)
>>> goto out;
>>> }
>>> data->valid = false;
>>> + data->saved_data = devfreq->data;
>>> devfreq->data = data;
>>>
>>> err = sysfs_create_group(&devfreq->dev.kobj, &dev_attr_group);
>>> @@ -100,6 +102,8 @@ static int userspace_init(struct devfreq *devfreq)
>>>
>>> static void userspace_exit(struct devfreq *devfreq)
>>> {
>>> + struct userspace_data *data = devfreq->data;
>>> + void *saved_data = data->saved_data;
>>> /*
>>> * Remove the sysfs entry, unless this is being called after
>>> * device_del(), which should have done this already via kobject_del().
>>> @@ -108,7 +112,7 @@ static void userspace_exit(struct devfreq *devfreq)
>>> sysfs_remove_group(&devfreq->dev.kobj, &dev_attr_group);
>>>
>>> kfree(devfreq->data);
>>> - devfreq->data = NULL;
>>> + devfreq->data = saved_data;
>>> }
>>>
>>> static int devfreq_userspace_handler(struct devfreq *devfreq,
>>
>> Dear MyungJoo, Kyungmin & Chanwoo,
>> Gently ping this issue... Does this patch has a chance to be accepted?
>> This seems to be a bug in devfreq userspace governor, which affects the
>> switching between governors -- When switching from userspace to
>> ondemand, the ondemand governor would be invalid.
>> If there's any question, please let me know.
>> Thank you.
>
> Yes, indeed. This is a bug.
>
> Actually, it appears that allocating a new memory buffer for
> devfreq->data itself is a bug for a governor, this is supposed
> to be allocated by a device driver. Thus, the comment of
> "void *data" of "struct devfreq" should be updated:
> "/* private data for governors given by device drivers */"
>
> It'd be better to have something like, "void *internal_data"
> for governors to freely handle within its context of init-exit,
> which is not touched by its users (device drivers).
>
> @Chanwoo: what's your opinion on this?
>
> Cheers,
> MyungJoo.
>
Hi MyungJoo,
Thanks for your suggestion. Here's the patch-v2, please have a look:
--
Subject: [PATCH] devfreq: governor: Add a private governor_data for
governors in devfreq
The member void *data in the structure devfreq can be overwrite
by governor_userspace. For example:
1. The device driver assigned the devfreq governor to simple_ondemand
by the function devfreq_add_device() and init the devfreq member
void *data to a pointer of a static structure devfreq_simple_ondemand_data
by the function devfreq_add_device().
2. The user changed the devfreq governor to userspace by the command
"echo userspace > /sys/class/devfreq/.../governor".
3. The governor userspace alloced a dynamic memory for the struct
userspace_data and assigend the member void *data of devfreq to
this memory by the function userspace_init().
4. The user changed the devfreq governor back to simple_ondemand
by the command "echo simple_ondemand > /sys/class/devfreq/.../governor".
5. The governor userspace exited and assigned the member void *data
in the structure devfreq to NULL by the function userspace_exit().
6. The governor simple_ondemand fetched the static information of
devfreq_simple_ondemand_data in the function
devfreq_simple_ondemand_func() but the member void *data of devfreq was
assigned to NULL by the function userspace_exit().
7. The information of upthreshold and downdifferential is lost
and the governor simple_ondemand can't work correctly.
The member void *data in the structure devfreq is designed for
a static pointer used in a governor and inited by the function
devfreq_add_device(). This patch add an element named governor_data
in the devfreq structure which can be used by a governor(E.g userspace)
who want to assign a private data to do some private things.
Signed-off-by: Kant Fan <[email protected]>
---
diff --git a/drivers/devfreq/governor_userspace.c
b/drivers/devfreq/governor_userspace.c
index ab9db7a..d69672c 100644
--- a/drivers/devfreq/governor_userspace.c
+++ b/drivers/devfreq/governor_userspace.c
@@ -21,7 +21,7 @@
static int devfreq_userspace_func(struct devfreq *df, unsigned long *freq)
{
- struct userspace_data *data = df->data;
+ struct userspace_data *data = df->governor_data;
if (data->valid)
*freq = data->user_frequency;
@@ -40,7 +40,7 @@
int err = 0;
mutex_lock(&devfreq->lock);
- data = devfreq->data;
+ data = devfreq->governor_data;
sscanf(buf, "%lu", &wanted);
data->user_frequency = wanted;
@@ -60,7 +60,7 @@
int err = 0;
mutex_lock(&devfreq->lock);
- data = devfreq->data;
+ data = devfreq->governor_data;
if (data->valid)
err = sprintf(buf, "%lu\n", data->user_frequency);
@@ -91,7 +91,7 @@
goto out;
}
data->valid = false;
- devfreq->data = data;
+ devfreq->governor_data = data;
err = sysfs_create_group(&devfreq->dev.kobj, &dev_attr_group);
out:
@@ -107,8 +107,8 @@
if (devfreq->dev.kobj.sd)
sysfs_remove_group(&devfreq->dev.kobj, &dev_attr_group);
- kfree(devfreq->data);
- devfreq->data = NULL;
+ kfree(devfreq->governor_data);
+ devfreq->governor_data = NULL;
}
static int devfreq_userspace_handler(struct devfreq *devfreq,
diff --git a/include/linux/devfreq.h b/include/linux/devfreq.h
index 34aab4d..d265af3 100644
--- a/include/linux/devfreq.h
+++ b/include/linux/devfreq.h
@@ -152,8 +152,8 @@
* @max_state: count of entry present in the frequency table.
* @previous_freq: previously configured frequency value.
* @last_status: devfreq user device info, performance statistics
- * @data: Private data of the governor. The devfreq framework does not
- * touch this.
+ * @data: devfreq core pass to governors, governor should not change it.
+ * @governor_data: private data for governors, devfreq core doesn't
touch it.
* @user_min_freq_req: PM QoS minimum frequency request from user (via
sysfs)
* @user_max_freq_req: PM QoS maximum frequency request from user (via
sysfs)
* @scaling_min_freq: Limit minimum frequency requested by OPP interface
@@ -193,7 +193,8 @@
unsigned long previous_freq;
struct devfreq_dev_status last_status;
- void *data; /* private data for governors */
+ void *data;
+ void *governor_data;
struct dev_pm_qos_request user_min_freq_req;
struct dev_pm_qos_request user_max_freq_req;
--
Best Regards,
Kant Fan
On 9/15/2022 3:41 PM, Kant Fan wrote:
>>>> static int devfreq_userspace_handler(struct devfreq *devfreq,
>>>
>>> Dear MyungJoo, Kyungmin & Chanwoo,
>>> Gently ping this issue... Does this patch has a chance to be accepted?
>>> This seems to be a bug in devfreq userspace governor, which affects the
>>> switching between governors -- When switching from userspace to
>>> ondemand, the ondemand governor would be invalid.
>>> If there's any question, please let me know.
>>> Thank you.
>>
>> Yes, indeed. This is a bug.
>>
>> Actually, it appears that allocating a new memory buffer for
>> devfreq->data itself is a bug for a governor, this is supposed
>> to be allocated by a device driver. Thus, the comment of
>> "void *data" of "struct devfreq" should be updated:
>> "/* private data for governors given by device drivers */"
>>
>> It'd be better to have something like, "void *internal_data"
>> for governors to freely handle within its context of init-exit,
>> which is not touched by its users (device drivers).
>>
>> @Chanwoo: what's your opinion on this?
>>
>> Cheers,
>> MyungJoo.
>>
>
> Hi MyungJoo,
> Thanks for your suggestion. Here's the patch-v2, please have a look:
>
> --
> Subject: [PATCH] devfreq: governor: Add a private governor_data for
> governors in devfreq
>
> The member void *data in the structure devfreq can be overwrite
> by governor_userspace. For example:
> 1. The device driver assigned the devfreq governor to simple_ondemand
> by the function devfreq_add_device() and init the devfreq member
> void *data to a pointer of a static structure devfreq_simple_ondemand_data
> by the function devfreq_add_device().
> 2. The user changed the devfreq governor to userspace by the command
> "echo userspace > /sys/class/devfreq/.../governor".
> 3. The governor userspace alloced a dynamic memory for the struct
> userspace_data and assigend the member void *data of devfreq to
> this memory by the function userspace_init().
> 4. The user changed the devfreq governor back to simple_ondemand
> by the command "echo simple_ondemand > /sys/class/devfreq/.../governor".
> 5. The governor userspace exited and assigned the member void *data
> in the structure devfreq to NULL by the function userspace_exit().
> 6. The governor simple_ondemand fetched the static information of
> devfreq_simple_ondemand_data in the function
> devfreq_simple_ondemand_func() but the member void *data of devfreq was
> assigned to NULL by the function userspace_exit().
> 7. The information of upthreshold and downdifferential is lost
> and the governor simple_ondemand can't work correctly.
>
> The member void *data in the structure devfreq is designed for
> a static pointer used in a governor and inited by the function
> devfreq_add_device(). This patch add an element named governor_data
> in the devfreq structure which can be used by a governor(E.g userspace)
> who want to assign a private data to do some private things.
>
> Signed-off-by: Kant Fan <[email protected]>
> ---
>
> diff --git a/drivers/devfreq/governor_userspace.c
> b/drivers/devfreq/governor_userspace.c
> index ab9db7a..d69672c 100644
> --- a/drivers/devfreq/governor_userspace.c
> +++ b/drivers/devfreq/governor_userspace.c
> @@ -21,7 +21,7 @@
>
> static int devfreq_userspace_func(struct devfreq *df, unsigned long
> *freq)
> {
> - struct userspace_data *data = df->data;
> + struct userspace_data *data = df->governor_data;
>
> if (data->valid)
> *freq = data->user_frequency;
> @@ -40,7 +40,7 @@
> int err = 0;
>
> mutex_lock(&devfreq->lock);
> - data = devfreq->data;
> + data = devfreq->governor_data;
>
> sscanf(buf, "%lu", &wanted);
> data->user_frequency = wanted;
> @@ -60,7 +60,7 @@
> int err = 0;
>
> mutex_lock(&devfreq->lock);
> - data = devfreq->data;
> + data = devfreq->governor_data;
>
> if (data->valid)
> err = sprintf(buf, "%lu\n", data->user_frequency);
> @@ -91,7 +91,7 @@
> goto out;
> }
> data->valid = false;
> - devfreq->data = data;
> + devfreq->governor_data = data;
>
> err = sysfs_create_group(&devfreq->dev.kobj, &dev_attr_group);
> out:
> @@ -107,8 +107,8 @@
> if (devfreq->dev.kobj.sd)
> sysfs_remove_group(&devfreq->dev.kobj, &dev_attr_group);
>
> - kfree(devfreq->data);
> - devfreq->data = NULL;
> + kfree(devfreq->governor_data);
> + devfreq->governor_data = NULL;
> }
>
> static int devfreq_userspace_handler(struct devfreq *devfreq,
> diff --git a/include/linux/devfreq.h b/include/linux/devfreq.h
> index 34aab4d..d265af3 100644
> --- a/include/linux/devfreq.h
> +++ b/include/linux/devfreq.h
> @@ -152,8 +152,8 @@
> * @max_state: count of entry present in the frequency table.
> * @previous_freq: previously configured frequency value.
> * @last_status: devfreq user device info, performance statistics
> - * @data: Private data of the governor. The devfreq framework does not
> - * touch this.
> + * @data: devfreq core pass to governors, governor should not change
> it.
> + * @governor_data: private data for governors, devfreq core doesn't
> touch it.
> * @user_min_freq_req: PM QoS minimum frequency request from user
> (via sysfs)
> * @user_max_freq_req: PM QoS maximum frequency request from user
> (via sysfs)
> * @scaling_min_freq: Limit minimum frequency requested by OPP
> interface
> @@ -193,7 +193,8 @@
> unsigned long previous_freq;
> struct devfreq_dev_status last_status;
>
> - void *data; /* private data for governors */
> + void *data;
> + void *governor_data;
>
> struct dev_pm_qos_request user_min_freq_req;
> struct dev_pm_qos_request user_max_freq_req;
>
Hi MyungJoo,
Sorry to disturb. Just want to say that I'm looking forward to your
advice on this patch. Thank you :>
--
Best Regards,
Kant Fan
> On 9/15/2022 3:41 PM, Kant Fan wrote:
> >
> > diff --git a/drivers/devfreq/governor_userspace.c
> > b/drivers/devfreq/governor_userspace.c
> > index ab9db7a..d69672c 100644
> > --- a/drivers/devfreq/governor_userspace.c
> > +++ b/drivers/devfreq/governor_userspace.c
> > @@ -21,7 +21,7 @@
> >
> > static int devfreq_userspace_func(struct devfreq *df, unsigned long
> > *freq)
> > {
> > - struct userspace_data *data = df->data;
> > + struct userspace_data *data = df->governor_data;
> >
> > if (data->valid)
> > *freq = data->user_frequency;
> > @@ -40,7 +40,7 @@
> > int err = 0;
> >
> > mutex_lock(&devfreq->lock);
> > - data = devfreq->data;
> > + data = devfreq->governor_data;
> >
> > sscanf(buf, "%lu", &wanted);
> > data->user_frequency = wanted;
> > @@ -60,7 +60,7 @@
> > int err = 0;
> >
> > mutex_lock(&devfreq->lock);
> > - data = devfreq->data;
> > + data = devfreq->governor_data;
> >
> > if (data->valid)
> > err = sprintf(buf, "%lu\n", data->user_frequency);
> > @@ -91,7 +91,7 @@
> > goto out;
> > }
> > data->valid = false;
> > - devfreq->data = data;
> > + devfreq->governor_data = data;
> >
> > err = sysfs_create_group(&devfreq->dev.kobj, &dev_attr_group);
> > out:
> > @@ -107,8 +107,8 @@
> > if (devfreq->dev.kobj.sd)
> > sysfs_remove_group(&devfreq->dev.kobj, &dev_attr_group);
> >
> > - kfree(devfreq->data);
> > - devfreq->data = NULL;
> > + kfree(devfreq->governor_data);
> > + devfreq->governor_data = NULL;
> > }
> >
> > static int devfreq_userspace_handler(struct devfreq *devfreq,
> > diff --git a/include/linux/devfreq.h b/include/linux/devfreq.h
> > index 34aab4d..d265af3 100644
> > --- a/include/linux/devfreq.h
> > +++ b/include/linux/devfreq.h
> > @@ -152,8 +152,8 @@
> > * @max_state: count of entry present in the frequency table.
> > * @previous_freq: previously configured frequency value.
> > * @last_status: devfreq user device info, performance statistics
> > - * @data: Private data of the governor. The devfreq framework does not
> > - * touch this.
> > + * @data: devfreq core pass to governors, governor should not change
> > it.
> > + * @governor_data: private data for governors, devfreq core doesn't
> > touch it.
> > * @user_min_freq_req: PM QoS minimum frequency request from user
> > (via sysfs)
> > * @user_max_freq_req: PM QoS maximum frequency request from user
> > (via sysfs)
> > * @scaling_min_freq: Limit minimum frequency requested by OPP
> > interface
> > @@ -193,7 +193,8 @@
> > unsigned long previous_freq;
> > struct devfreq_dev_status last_status;
> >
> > - void *data; /* private data for governors */
> > + void *data;
> > + void *governor_data;
> >
> > struct dev_pm_qos_request user_min_freq_req;
> > struct dev_pm_qos_request user_max_freq_req;
> >
>
> Hi MyungJoo,
> Sorry to disturb. Just want to say that I'm looking forward to your
> advice on this patch. Thank you :>
>
This new code looks good to me.
Anyway, Chanwoo, how do you think of this?
Acked-by: MyungJoo Ham <[email protected]>
Cheers,
MyungJoo.