[PATCH v10 4/6] thermal/drivers/mediatek: Add the Low Voltage Thermal Sensor driver

[Balsam CHIHI]

Hi Daniel,

On Wed, Jan 18, 2023 at 3:30 PM Daniel Lezcano
<daniel.lezcano at linaro.org> wrote:
>
>
> Hi Balsam,
>
> On 18/01/2023 14:58, Balsam CHIHI wrote:
>
> [ ... ]
>
> >> You're describing the register with nice words, but there's another way to do
> >> the same that will be even more effective.
> >>
> >> /*
> >>    * LVTS MONINT: Interrupt Monitoring register
> >>    * Each bit describes the enable status of per-sensor interrupts.
> >>    */
> >> #define LVTS_MONINT_THRES_COLD  BIT(0)  /* Cold threshold */
> >> #define LVTS_MONINT_THRES_HOT   BIT(1)  /* Hot threshold */
> >> #define LVTS_MONINT_OFFST_LOW   BIT(2)  /* Low offset */
> >> #define LVTS_MONINT_OFFST_HIGH  BIT(3)  /* High offset */
> >> #define LVTS_MONINT_OFFST_NTH   BIT(4)  /* Normal To Hot */
> >> #define EVERYTHING_ELSE ........................
> >>
> >> #define LVTS_MONINT_SNS0_MASK   GENMASK( ... )
> >> #define LVTS_MONINT_SNS1_MASK   GENMASK .....
> >>
> >> /* Find a better name for this one */
> >> #define LVTS_MONINT_EN_IRQS     ( LVTS_MONINT_THRES_COLD | LVTS_MONINT_THRES_HOT |
> >>                                    LVTS_MONINT_OFFST_LOW ..... etc etc)
> >>
> >
> > Given the complexity of the controller and the number of registers,
> > if we create a define per bits, we will end up with a huge list of
> > defines (~300).
>
> Yeah, that is too much for a little gain.
>
> However, a few can be added for the interrupt only.
>
> Instead of LVTS_MONINT_THRES ..., it could be LVTS_INT_THRES_... and
> reused for LVTS_MONINTSTS and LVTS_MONINT setup as the bits position are
> the same?
>

OK, I will add a few for the interrupts.

> [ ... ]
>
>
> >>> +static int lvts_ctrl_configure(struct device *dev, struct lvts_ctrl *lvts_ctrl)
> >>> +{
> >>> +     u32 period_unit = (118 * 1000) / (256 * 38);
> >>
> >> #define SOMETHING       118
> >> #define SOMETHING_ELSE  1000
> >> #define ....
> >>
> >> const u32 period_unit = (SOMETHING * SOMETHING_ELSE) / ....
> >>
> >
> > Constifying "u32 period_unit" generates the following compilation warning :
> > ./include/asm-generic/io.h:273:61: note: expected ‘volatile void *’
> > but argument is of type ‘const void *’
> >    273 | static inline void writel(u32 value, volatile void __iomem *addr)
> >        |                                      ~~~~~~~~~~~~~~~~~~~~~~~^~~~
>
> That is strange. period_unit is the 'value', not the 'addr'. Are you
> sure about the warning?

I double-checked and you are right.
The warning is not related to "const u32 period_unit".
It is related to the previous one, which is "const void __iomem *lvts_edata".
And this is the full log :

drivers/thermal/mediatek/lvts_thermal.c: In function ‘lvts_ctrl_calibrate’:
drivers/thermal/mediatek/lvts_thermal.c:835:47: warning: passing
argument 2 of ‘writel’ discards ‘const’ qualifier from pointer target
type [-Wdiscarded-qualifiers]
  835 |   writel(lvts_ctrl->calibration[i], lvts_edata[i]);
      |                                     ~~~~~~~~~~^~~
In file included from ./arch/arm64/include/asm/io.h:163,
                 from ./include/linux/io.h:13,
                 from ./include/linux/irq.h:20,
                 from ./include/asm-generic/hardirq.h:17,
                 from ./arch/arm64/include/asm/hardirq.h:17,
                 from ./include/linux/hardirq.h:11,
                 from ./include/linux/interrupt.h:11,
                 from drivers/thermal/mediatek/lvts_thermal.c:12:
./include/asm-generic/io.h:273:61: note: expected ‘volatile void *’
but argument is of type ‘const void *’
  273 | static inline void writel(u32 value, volatile void __iomem *addr)
      |                                      ~~~~~~~~~~~~~~~~~~~~~~~^~~~

>
>
>
>
> --
> <http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs
>
> Follow Linaro:  <http://www.facebook.com/pages/Linaro> Facebook |
> <http://twitter.com/#!/linaroorg> Twitter |
> <http://www.linaro.org/linaro-blog/> Blog
>

Best regards,
Balsam