| Message ID | 1538469324-8028-1-git-send-email-cv-dong@jinso.co.jp (mailing list archive) |
|---|---|
| State | Superseded |
| Delegated to: | Geert Uytterhoeven |
| Headers | show |
| Series | hwmon: Fix the 'No sensors found' error after replacing the parameter NULL by the actual device | expand |
On 10/02/2018 01:35 AM, Cao Van Dong wrote: > In __hwmon_device_register() function of hwmon.c, we have assigned 'dev' directly to 'hdev->parent'. > Formerly, when registering the hwmon device, we pass NULL as the device. This is not affected. > Recently, the developer has replaced the parameter NULL as the device by the actual device. > This causes the "No sensors found" error. This patch is to fix this error. > > This patch is based on the v4.19-rc3 tag. > NACK. This is wrong. The passed device is the hwmon driver's parent device. Using that device's parent would be wrong. Indeed, that device could be a platform device with no parent. The problem must be fixed in the calling code. Guenter > --- > drivers/hwmon/hwmon.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/hwmon/hwmon.c b/drivers/hwmon/hwmon.c > index 33d5128..7c8d3cb 100644 > --- a/drivers/hwmon/hwmon.c > +++ b/drivers/hwmon/hwmon.c > @@ -610,7 +610,7 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata, > > hwdev->name = name; > hdev->class = &hwmon_class; > - hdev->parent = dev; > + hdev->parent = dev->parent; > hdev->of_node = dev ? dev->of_node : NULL; > hwdev->chip = chip; > dev_set_drvdata(hdev, drvdata); >
On 10/02/2018 02:06 AM, Cao Van Dong wrote: > Dear Geert-san, > > Thanks for your comment! > >>> In __hwmon_device_register() function of hwmon.c, we have assigned 'dev' directly to 'hdev->parent'. >>> Formerly, when registering the hwmon device, we pass NULL as the device. This is not affected. >>> Recently, the developer has replaced the parameter NULL as the device by the actual device. >> Do you know the commit ID of this recent change? >> It's useful for review of your change, and to know to which versions your >> patch should be backported. > > The commit ID of this recent change is"f6b6b52 thermal_hwmon: Pass the originating device down to hwmon_device_register_with_info". > If thermal doesn't want hwmon to use the thermal zone device, it should pass the parent of that device as parameter. >>> This causes the "No sensors found" error. This patch is to fix this error. >> On which platform do you see this failure? >> Thanks again! > > This error is found on Gen2 Lager board v4.19-rc2 by RVC team. > Some more details about the error would be useful. Guenter > > Best regards, > Dong/Jinso > >
CC Marc, Eduardo On Tue, Oct 2, 2018 at 2:50 PM Guenter Roeck <linux@roeck-us.net> wrote: > On 10/02/2018 01:35 AM, Cao Van Dong wrote: > > In __hwmon_device_register() function of hwmon.c, we have assigned 'dev' directly to 'hdev->parent'. > > Formerly, when registering the hwmon device, we pass NULL as the device. This is not affected. > > Recently, the developer has replaced the parameter NULL as the device by the actual device. > > This causes the "No sensors found" error. This patch is to fix this error. > > > > This patch is based on the v4.19-rc3 tag. > > > > NACK. > > This is wrong. The passed device is the hwmon driver's parent device. Using that device's > parent would be wrong. Indeed, that device could be a platform device with no parent. > The problem must be fixed in the calling code. > > Guenter > > > --- > > drivers/hwmon/hwmon.c | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > diff --git a/drivers/hwmon/hwmon.c b/drivers/hwmon/hwmon.c > > index 33d5128..7c8d3cb 100644 > > --- a/drivers/hwmon/hwmon.c > > +++ b/drivers/hwmon/hwmon.c > > @@ -610,7 +610,7 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata, > > > > hwdev->name = name; > > hdev->class = &hwmon_class; > > - hdev->parent = dev; > > + hdev->parent = dev->parent; > > hdev->of_node = dev ? dev->of_node : NULL; > > hwdev->chip = chip; > > dev_set_drvdata(hdev, drvdata);
CC Marc, Eduardo On Tue, Oct 2, 2018 at 2:56 PM Guenter Roeck <linux@roeck-us.net> wrote: > > On 10/02/2018 02:06 AM, Cao Van Dong wrote: > > Dear Geert-san, > > > > Thanks for your comment! > > > >>> In __hwmon_device_register() function of hwmon.c, we have assigned 'dev' directly to 'hdev->parent'. > >>> Formerly, when registering the hwmon device, we pass NULL as the device. This is not affected. > >>> Recently, the developer has replaced the parameter NULL as the device by the actual device. > >> Do you know the commit ID of this recent change? > >> It's useful for review of your change, and to know to which versions your > >> patch should be backported. > > > > The commit ID of this recent change is"f6b6b52 thermal_hwmon: Pass the originating device down to hwmon_device_register_with_info". > > > If thermal doesn't want hwmon to use the thermal zone device, it should pass > the parent of that device as parameter. > > >>> This causes the "No sensors found" error. This patch is to fix this error. > >> On which platform do you see this failure? > >> Thanks again! > > > > This error is found on Gen2 Lager board v4.19-rc2 by RVC team. > > > > Some more details about the error would be useful. Gr{oetje,eeting}s, Geert
Dear Guente-san, Thanks for your comment! >>>> In __hwmon_device_register() function of hwmon.c, we have assigned >>>> 'dev' directly to 'hdev->parent'. >>>> Formerly, when registering the hwmon device, we pass NULL as the >>>> device. This is not affected. >>>> Recently, the developer has replaced the parameter NULL as the >>>> device by the actual device. >>> Do you know the commit ID of this recent change? >>> It's useful for review of your change, and to know to which versions >>> your >>> patch should be backported. >> >> The commit ID of this recent change is"f6b6b52 thermal_hwmon: Pass >> the originating device down to hwmon_device_register_with_info". >> > If thermal doesn't want hwmon to use the thermal zone device, it > should pass > the parent of that device as parameter. I will see the feedback from Marc-san. > >>>> This causes the "No sensors found" error. This patch is to fix this >>>> error. >>> On which platform do you see this failure? >>> Thanks again! >> >> This error is found on Gen2 Lager board v4.19-rc2 by RVC team. >> > > Some more details about the error would be useful. Below are some information about confirm the error: - Before the commit "f6b6b52ef7a54160c0a4d2b56a58f84a2e5b33c5" (pass NULL as the device): "root@linaro-nano:~# sensors cpu_thermal-virtual-0 Adapter: Virtual device temp1: +30.0�°C (crit = +95.0�°C) root@linaro-nano:~# " - After the commit "f6b6b52ef7a54160c0a4d2b56a58f84a2e5b33c5" (pass the actual device): "root@linaro-nano:~# sensors No sensors found! Make sure you loaded all the kernel drivers you need. Try sensors-detect to find out which these are. root@linaro-nano:~#" - At my patch (change dev to dev->parent): "root@linaro-nano:~# sensors cpu_thermal-virtual-0 Adapter: Virtual device temp1: +30.0�°C (crit = +95.0�°C) root@linaro-nano:~#" Regards, Dong/Jinso
Dear Guenter-san, Thanks for your comment! >> In __hwmon_device_register() function of hwmon.c, we have assigned >> 'dev' directly to 'hdev->parent'. >> Formerly, when registering the hwmon device, we pass NULL as the >> device. This is not affected. >> Recently, the developer has replaced the parameter NULL as the device >> by the actual device. >> This causes the "No sensors found" error. This patch is to fix this >> error. >> >> This patch is based on the v4.19-rc3 tag. >> > > NACK. > > This is wrong. The passed device is the hwmon driver's parent device. > Using that device's > parent would be wrong. Indeed, that device could be a platform device > with no parent. Maybe I did not know all problem. If I'm wrong, sorry for the inconvenience. > The problem must be fixed in the calling code. > I will see the feedback from Marc-san. Regards, Dong/Jinso >> --- >> drivers/hwmon/hwmon.c | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> diff --git a/drivers/hwmon/hwmon.c b/drivers/hwmon/hwmon.c >> index 33d5128..7c8d3cb 100644 >> --- a/drivers/hwmon/hwmon.c >> +++ b/drivers/hwmon/hwmon.c >> @@ -610,7 +610,7 @@ __hwmon_device_register(struct device *dev, const >> char *name, void *drvdata, >> hwdev->name = name; >> hdev->class = &hwmon_class; >> - hdev->parent = dev; >> + hdev->parent = dev->parent; >> hdev->of_node = dev ? dev->of_node : NULL; >> hwdev->chip = chip; >> dev_set_drvdata(hdev, drvdata);
On 10/02/2018 06:33 PM, Cao Van Dong wrote: > Dear Guenter-san, > > Thanks for your comment! > >>> In __hwmon_device_register() function of hwmon.c, we have assigned 'dev' directly to 'hdev->parent'. >>> Formerly, when registering the hwmon device, we pass NULL as the device. This is not affected. >>> Recently, the developer has replaced the parameter NULL as the device by the actual device. >>> This causes the "No sensors found" error. This patch is to fix this error. >>> >>> This patch is based on the v4.19-rc3 tag. >>> >> >> NACK. >> >> This is wrong. The passed device is the hwmon driver's parent device. Using that device's >> parent would be wrong. Indeed, that device could be a platform device with no parent. > Maybe I did not know all problem. If I'm wrong, sorry for the inconvenience. > >> The problem must be fixed in the calling code. >> > I will see the feedback from Marc-san. > The "No sensors found" message is from the "sensors" application. The device type associated with the passed device is unknown to libsensors (it is parsed from the parent device subsystem name - what is that, anyway ?). This could be addressed in libsensors, or the thermal subsystem could attach itself to a known subsystem, or the thermal subsystem must pass a pointer the parent device. Again, trying to address the problem in the hwmon core is simply wrong. Guenter > > Regards, > Dong/Jinso > >>> --- >>> drivers/hwmon/hwmon.c | 2 +- >>> 1 file changed, 1 insertion(+), 1 deletion(-) >>> >>> diff --git a/drivers/hwmon/hwmon.c b/drivers/hwmon/hwmon.c >>> index 33d5128..7c8d3cb 100644 >>> --- a/drivers/hwmon/hwmon.c >>> +++ b/drivers/hwmon/hwmon.c >>> @@ -610,7 +610,7 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata, >>> hwdev->name = name; >>> hdev->class = &hwmon_class; >>> - hdev->parent = dev; >>> + hdev->parent = dev->parent; >>> hdev->of_node = dev ? dev->of_node : NULL; >>> hwdev->chip = chip; >>> dev_set_drvdata(hdev, drvdata); >
Dear Guenter-san, Thanks for your comment! >>>> In __hwmon_device_register() function of hwmon.c, we have assigned >>>> 'dev' directly to 'hdev->parent'. >>>> Formerly, when registering the hwmon device, we pass NULL as the >>>> device. This is not affected. >>>> Recently, the developer has replaced the parameter NULL as the >>>> device by the actual device. >>>> This causes the "No sensors found" error. This patch is to fix this >>>> error. >>>> >>>> This patch is based on the v4.19-rc3 tag. >>>> >>> >>> NACK. >>> >>> This is wrong. The passed device is the hwmon driver's parent >>> device. Using that device's >>> parent would be wrong. Indeed, that device could be a platform >>> device with no parent. >> Maybe I did not know all problem. If I'm wrong, sorry for the >> inconvenience. >> >>> The problem must be fixed in the calling code. >>> >> I will see the feedback from Marc-san. >> > > The "No sensors found" message is from the "sensors" application. > The device type associated with the passed device is unknown to > libsensors (it is parsed from the parent device subsystem name - > what is that, anyway ?). This could be addressed in libsensors, > or the thermal subsystem could attach itself to a known subsystem, > or the thermal subsystem must pass a pointer the parent device. > > Again, trying to address the problem in the hwmon core is simply wrong. > I will create other patch(v2) to fix for this error from the calling function. Regards, Dong/Jinso >>>> --- >>>> drivers/hwmon/hwmon.c | 2 +- >>>> 1 file changed, 1 insertion(+), 1 deletion(-) >>>> >>>> diff --git a/drivers/hwmon/hwmon.c b/drivers/hwmon/hwmon.c >>>> index 33d5128..7c8d3cb 100644 >>>> --- a/drivers/hwmon/hwmon.c >>>> +++ b/drivers/hwmon/hwmon.c >>>> @@ -610,7 +610,7 @@ __hwmon_device_register(struct device *dev, >>>> const char *name, void *drvdata, >>>> hwdev->name = name; >>>> hdev->class = &hwmon_class; >>>> - hdev->parent = dev; >>>> + hdev->parent = dev->parent; >>>> hdev->of_node = dev ? dev->of_node : NULL; >>>> hwdev->chip = chip; >>>> dev_set_drvdata(hdev, drvdata);
diff --git a/drivers/hwmon/hwmon.c b/drivers/hwmon/hwmon.c index 33d5128..7c8d3cb 100644 --- a/drivers/hwmon/hwmon.c +++ b/drivers/hwmon/hwmon.c @@ -610,7 +610,7 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata, hwdev->name = name; hdev->class = &hwmon_class; - hdev->parent = dev; + hdev->parent = dev->parent; hdev->of_node = dev ? dev->of_node : NULL; hwdev->chip = chip; dev_set_drvdata(hdev, drvdata);