Re: [PATCH] thermal/intel: Fix intel_tcc_get_temp() to support negative CPU temperature

[Rafael J. Wysocki]

On Tue, Feb 6, 2024 at 4:53 PM Stanislaw Gruszka
<stanislaw.gruszka@xxxxxxxxxxxxxxx> wrote:
>
> On Tue, Feb 06, 2024 at 09:54:09AM +0800, Zhang Rui wrote:
> > CPU temperature can be negative in some cases. Thus the negative CPU
> > temperature should not be considered as a failure.
> >
> > Fix intel_tcc_get_temp() and its users to support negative CPU
> > temperature.
> >
> > Fixes: a3c1f066e1c5 ("thermal/intel: Introduce Intel TCC library")
> > Signed-off-by: Zhang Rui <rui.zhang@xxxxxxxxx>
> Reviewed-by: Stanislaw Gruszka <stanislaw.gruszka@xxxxxxxxxxxxxxx>
>
> > ---
> >  .../intel/int340x_thermal/processor_thermal_device.c |  8 ++++----
> >  drivers/thermal/intel/intel_tcc.c                    | 12 ++++++------
> >  drivers/thermal/intel/x86_pkg_temp_thermal.c         |  8 ++++----
> >  include/linux/intel_tcc.h                            |  2 +-
> >  4 files changed, 15 insertions(+), 15 deletions(-)
> >
> > diff --git a/drivers/thermal/intel/int340x_thermal/processor_thermal_device.c b/drivers/thermal/intel/int340x_thermal/processor_thermal_device.c
> > index 649f67fdf345..d75fae7b7ed2 100644
> > --- a/drivers/thermal/intel/int340x_thermal/processor_thermal_device.c
> > +++ b/drivers/thermal/intel/int340x_thermal/processor_thermal_device.c
> > @@ -176,14 +176,14 @@ static int proc_thermal_get_zone_temp(struct thermal_zone_device *zone,
> >                                        int *temp)
> >  {
> >       int cpu;
> > -     int curr_temp;
> > +     int curr_temp, ret;
> >
> >       *temp = 0;
> >
> >       for_each_online_cpu(cpu) {
> > -             curr_temp = intel_tcc_get_temp(cpu, false);
> > -             if (curr_temp < 0)
> > -                     return curr_temp;
> > +             ret = intel_tcc_get_temp(cpu, &curr_temp, false);
> > +             if (ret < 0)
> > +                     return ret;
> >               if (!*temp || curr_temp > *temp)
> >                       *temp = curr_temp;
> >       }
> > diff --git a/drivers/thermal/intel/intel_tcc.c b/drivers/thermal/intel/intel_tcc.c
> > index 2e5c741c41ca..5e8b7f34b395 100644
> > --- a/drivers/thermal/intel/intel_tcc.c
> > +++ b/drivers/thermal/intel/intel_tcc.c
> > @@ -103,18 +103,19 @@ EXPORT_SYMBOL_NS_GPL(intel_tcc_set_offset, INTEL_TCC);
> >  /**
> >   * intel_tcc_get_temp() - returns the current temperature
> >   * @cpu: cpu that the MSR should be run on, nagative value means any cpu.
> > + * @temp: pointer to the memory for saving cpu temperature.
> >   * @pkg: true: Package Thermal Sensor. false: Core Thermal Sensor.
> >   *
> >   * Get the current temperature returned by the CPU core/package level
> >   * thermal sensor, in degrees C.
> >   *
> > - * Return: Temperature in degrees C on success, negative error code otherwise.
> > + * Return: 0 on success, negative error code otherwise.
> >   */
> > -int intel_tcc_get_temp(int cpu, bool pkg)
> > +int intel_tcc_get_temp(int cpu, int *temp, bool pkg)
> >  {
> >       u32 low, high;
> >       u32 msr = pkg ? MSR_IA32_PACKAGE_THERM_STATUS : MSR_IA32_THERM_STATUS;
> > -     int tjmax, temp, err;
> > +     int tjmax, err;
> >
> >       tjmax = intel_tcc_get_tjmax(cpu);
> >       if (tjmax < 0)
> > @@ -131,9 +132,8 @@ int intel_tcc_get_temp(int cpu, bool pkg)
> >       if (!(low & BIT(31)))
> >               return -ENODATA;
> >
> > -     temp = tjmax - ((low >> 16) & 0x7f);
> > +     *temp = tjmax - ((low >> 16) & 0x7f);
> >
> > -     /* Do not allow negative CPU temperature */
> > -     return temp >= 0 ? temp : -ENODATA;
> > +     return 0;
> >  }
> >  EXPORT_SYMBOL_NS_GPL(intel_tcc_get_temp, INTEL_TCC);
> > diff --git a/drivers/thermal/intel/x86_pkg_temp_thermal.c b/drivers/thermal/intel/x86_pkg_temp_thermal.c
> > index 11a7f8108bbb..61c3d450ee60 100644
> > --- a/drivers/thermal/intel/x86_pkg_temp_thermal.c
> > +++ b/drivers/thermal/intel/x86_pkg_temp_thermal.c
> > @@ -108,11 +108,11 @@ static struct zone_device *pkg_temp_thermal_get_dev(unsigned int cpu)
> >  static int sys_get_curr_temp(struct thermal_zone_device *tzd, int *temp)
> >  {
> >       struct zone_device *zonedev = thermal_zone_device_priv(tzd);
> > -     int val;
> > +     int val, ret;
> >
> > -     val = intel_tcc_get_temp(zonedev->cpu, true);
> > -     if (val < 0)
> > -             return val;
> > +     ret = intel_tcc_get_temp(zonedev->cpu, &val, true);
> > +     if (ret < 0)
> > +             return ret;
> >
> >       *temp = val * 1000;
> >       pr_debug("sys_get_curr_temp %d\n", *temp);
> > diff --git a/include/linux/intel_tcc.h b/include/linux/intel_tcc.h
> > index f422612c28d6..8ff8eabb4a98 100644
> > --- a/include/linux/intel_tcc.h
> > +++ b/include/linux/intel_tcc.h
> > @@ -13,6 +13,6 @@
> >  int intel_tcc_get_tjmax(int cpu);
> >  int intel_tcc_get_offset(int cpu);
> >  int intel_tcc_set_offset(int cpu, int offset);
> > -int intel_tcc_get_temp(int cpu, bool pkg);
> > +int intel_tcc_get_temp(int cpu, int *temp, bool pkg);
> >
> >  #endif /* __INTEL_TCC_H__ */
> > --

Applied as 6.9 material, thanks!