[permalink] [raw]

Subject: Re: [PATCH v2 1/5] hwmon: (max31790) Rework to use regmap

Hello.

I'm sending a new version of my patch on max31790. This new version
fixes the cache issue and actually makes it work by setting
.cache_type. You were right about the "yes/no" ranges, so I flipped
those.

By the way, it seems that the reason your reply got lost is because of
weird addresses in the "Cc:" email field, they end with "cesnet.cz",
so it could be that I'm sending email incorrectly. Let me know if I'm
doing something wrong.

Thanks,
Václav

čt 22. 4. 2021 v 3:31 odesílatel Guenter Roeck <[email protected]> napsal:
>
> On 4/12/21 7:59 PM, Václav Kubernát wrote:
> > Converting the driver to use regmap makes it more generic. It also makes
> > it a lot easier to debug through debugfs.
> >
> > Signed-off-by: Václav Kubernát <[email protected]>
> > ---
> > drivers/hwmon/Kconfig | 1 +
> > drivers/hwmon/max31790.c | 254 ++++++++++++++++++++-------------------
> > 2 files changed, 133 insertions(+), 122 deletions(-)
> >
> > diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig
> > index 1ecf697d8d99..9f11d036c316 100644
> > --- a/drivers/hwmon/Kconfig
> > +++ b/drivers/hwmon/Kconfig
> > @@ -1095,6 +1095,7 @@ config SENSORS_MAX6697
> > config SENSORS_MAX31790
> > tristate "Maxim MAX31790 sensor chip"
> > depends on I2C
> > + select REGMAP_I2C
> > help
> > If you say yes here you get support for 6-Channel PWM-Output
> > Fan RPM Controller.
> > diff --git a/drivers/hwmon/max31790.c b/drivers/hwmon/max31790.c
> > index 2c6b333a28e9..e3765ce4444a 100644
> > --- a/drivers/hwmon/max31790.c
> > +++ b/drivers/hwmon/max31790.c
> > @@ -12,6 +12,7 @@
> > #include <linux/init.h>
> > #include <linux/jiffies.h>
> > #include <linux/module.h>
> > +#include <linux/regmap.h>
> > #include <linux/slab.h>
> >
> > /* MAX31790 registers */
> > @@ -46,92 +47,53 @@
> >
> > #define NR_CHANNEL 6
> >
> > +#define MAX31790_REG_USER_BYTE_67 0x67
> > +
> > +#define BULK_TO_U16(msb, lsb) (((msb) << 8) + (lsb))
> > +#define U16_MSB(num) (((num) & 0xFF00) >> 8)
> > +#define U16_LSB(num) ((num) & 0x00FF)
> > +
> > +static const struct regmap_range max31790_ro_range = {
> > + .range_min = MAX31790_REG_TACH_COUNT(0),
> > + .range_max = MAX31790_REG_PWMOUT(0) - 1,
> > +};
> > +
> > +static const struct regmap_access_table max31790_wr_table = {
> > + .no_ranges = &max31790_ro_range,
> > + .n_no_ranges = 1,
> > +};
> > +
> > +static const struct regmap_range max31790_volatile_ranges[] = {
> > + regmap_reg_range(MAX31790_REG_TACH_COUNT(0), MAX31790_REG_TACH_COUNT(12)),
> > + regmap_reg_range(MAX31790_REG_FAN_FAULT_STATUS2, MAX31790_REG_FAN_FAULT_STATUS1),
> > +};
> > +
> > +static const struct regmap_access_table max31790_volatile_table = {
> > + .no_ranges = max31790_volatile_ranges,
> > + .n_no_ranges = 2,
> > + .n_yes_ranges = 0
> > +};
>
> Looks like my reply to this got lost. Other regmap code suggests that
> volatile register ranges are identified with yes_ranges, not with no_ranges.
> "no" seems to mean "not volatile". Please verify and confirm if the
> above code does what you want it to do.
>
> Thanks,
> Guenter

2021-04-26 14:19:23

by Guenter Roeck

[permalink] [raw]

Subject: Re: [PATCH v2 1/5] hwmon: (max31790) Rework to use regmap

On Mon, Apr 26, 2021 at 02:46:27PM +0200, V?clav Kubern?t wrote:
> Hello.
>
> I'm sending a new version of my patch on max31790. This new version
> fixes the cache issue and actually makes it work by setting
> .cache_type. You were right about the "yes/no" ranges, so I flipped
> those.
>
> By the way, it seems that the reason your reply got lost is because of
> weird addresses in the "Cc:" email field, they end with "cesnet.cz",
> so it could be that I'm sending email incorrectly. Let me know if I'm
> doing something wrong.
>

Yes, the To: field of your series is either empty (for the first patch
of the series), or it is something like:
To: unlisted-recipients: no To-header on input <;

Also, you send your follow-up series as response of the previous series
which doesn't follow the guidance for submitting patches and may result
in the entire series getting lost.

Guenter

2021-04-26 14:31:13

by Václav Kubernát

[permalink] [raw]

Subject: Re: [PATCH v2 1/5] hwmon: (max31790) Rework to use regmap

po 26. 4. 2021 v 16:18 odesílatel Guenter Roeck <[email protected]> napsal:
>
> On Mon, Apr 26, 2021 at 02:46:27PM +0200, Václav Kubernát wrote:
> > Hello.
> >
> > I'm sending a new version of my patch on max31790. This new version
> > fixes the cache issue and actually makes it work by setting
> > .cache_type. You were right about the "yes/no" ranges, so I flipped
> > those.
> >
> > By the way, it seems that the reason your reply got lost is because of
> > weird addresses in the "Cc:" email field, they end with "cesnet.cz",
> > so it could be that I'm sending email incorrectly. Let me know if I'm
> > doing something wrong.
> >
>
> Yes, the To: field of your series is either empty (for the first patch
> of the series), or it is something like:
> To: unlisted-recipients: no To-header on input <;
>
> Also, you send your follow-up series as response of the previous series
> which doesn't follow the guidance for submitting patches and may result
> in the entire series getting lost.
>

Sorry, I will fix my email-sending procedure. Should I resend the
patch series without the In-Reply-To field?

Václav

> Guenter

2021-04-26 14:38:02

by Guenter Roeck

[permalink] [raw]

Subject: Re: [PATCH v2 1/5] hwmon: (max31790) Rework to use regmap

On Mon, Apr 26, 2021 at 04:29:55PM +0200, V?clav Kubern?t wrote:
> po 26. 4. 2021 v 16:18 odes?latel Guenter Roeck <[email protected]> napsal:
> >
> > On Mon, Apr 26, 2021 at 02:46:27PM +0200, V?clav Kubern?t wrote:
> > > Hello.
> > >
> > > I'm sending a new version of my patch on max31790. This new version
> > > fixes the cache issue and actually makes it work by setting
> > > .cache_type. You were right about the "yes/no" ranges, so I flipped
> > > those.
> > >
> > > By the way, it seems that the reason your reply got lost is because of
> > > weird addresses in the "Cc:" email field, they end with "cesnet.cz",
> > > so it could be that I'm sending email incorrectly. Let me know if I'm
> > > doing something wrong.
> > >
> >
> > Yes, the To: field of your series is either empty (for the first patch
> > of the series), or it is something like:
> > To: unlisted-recipients: no To-header on input <;
> >
> > Also, you send your follow-up series as response of the previous series
> > which doesn't follow the guidance for submitting patches and may result
> > in the entire series getting lost.
> >
>
> Sorry, I will fix my email-sending procedure. Should I resend the
> patch series without the In-Reply-To field?
>
No, just keep it in mind for next time.

Thanks,
Guenter