mbox series

[0/4] Add regmap_field helpers for simple bit operations

Message ID 180e702a15f.e737e37e45859.3135149506136486394@zohomail.com
Headers show
Series Add regmap_field helpers for simple bit operations | expand

Message

Li Chen May 21, 2022, 2:25 p.m. UTC
From: Li Chen <lchen@ambarella.com>

This series proposes to add simple bit operations for setting, clearing
and testing specific bits with regmap_field.

Li Chen (4):
  regmap: provide regmap_field helpers for simple bit operations
  ASoC: sunxi: Use {regmap/regmap_field}_{set/clear}_bits helpers
  pinctrl: bcm: Use regmap_field_{set/clear}_bits helpers
  pinctrl: st: Switch to use regmap_field_test_bits

 drivers/base/regmap/regmap.c          | 22 ++++++++
 drivers/pinctrl/bcm/pinctrl-bcm6358.c |  2 +-
 drivers/pinctrl/pinctrl-st.c          | 23 ++++----
 include/linux/regmap.h                | 37 +++++++++++++
 sound/soc/sunxi/sun4i-codec.c         | 78 +++++++++++----------------
 5 files changed, 99 insertions(+), 63 deletions(-)

Comments

Samuel Holland May 23, 2022, 3:09 a.m. UTC | #1
Hi,

On 5/22/22 9:23 PM, Li Chen wrote:
> From: Li Chen <lchen@ambarella.com>
> 
> We have set/clear/test operations for regmap, but not for regmap_field yet.
> So let's introduce regmap_field helpers too.
> 
> In many instances regmap_field_update_bits() is used for simple bit setting
> and clearing. In these cases the last argument is redundant and we can
> hide it with a static inline function.
> 
> This adds three new helpers for simple bit operations: set_bits,
> clear_bits and test_bits (the last one defined as a regular function).
> 
> Signed-off-by: Li Chen <lchen@ambarella.com>
> ---
>  drivers/base/regmap/regmap.c | 22 +++++++++++++++++++++
>  include/linux/regmap.h       | 37 ++++++++++++++++++++++++++++++++++++
>  2 files changed, 59 insertions(+)
> 
> diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c
> index 5e12f7cb5147..a37d6041b7bd 100644
> --- a/drivers/base/regmap/regmap.c
> +++ b/drivers/base/regmap/regmap.c
> @@ -2208,6 +2208,28 @@ int regmap_field_update_bits_base(struct regmap_field *field,
>  }
>  EXPORT_SYMBOL_GPL(regmap_field_update_bits_base);
>  
> +/**
> + * regmap_field_test_bits() - Check if all specified bits are set in a
> + *                            register field.
> + *
> + * @field: Register field to operate on
> + * @bits: Bits to test
> + *
> + * Returns -1 if the underlying regmap_field_read() fails, 0 if at least one of the
> + * tested bits is not set and 1 if all tested bits are set.
> + */
> +int regmap_field_test_bits(struct regmap_field *field, unsigned int bits)
> +{
> +	unsigned int val, ret;
> +
> +	ret = regmap_field_read(field, &val);
> +	if (ret)
> +		return ret;
> +
> +	return (val & bits) == bits;
> +}
> +EXPORT_SYMBOL_GPL(regmap_field_test_bits);
> +
>  /**
>   * regmap_fields_update_bits_base() - Perform a read/modify/write cycle a
>   *                                    register field with port ID
> diff --git a/include/linux/regmap.h b/include/linux/regmap.h
> index de81a94d7b30..10b410734d9e 100644
> --- a/include/linux/regmap.h
> +++ b/include/linux/regmap.h
> @@ -1324,6 +1324,22 @@ static inline int regmap_field_update_bits(struct regmap_field *field,
>  					     NULL, false, false);
>  }
>  
> +static inline int regmap_field_set_bits(struct regmap_field *field,
> +					unsigned int bits)
> +{
> +	return regmap_field_update_bits_base(field, bits, 0, NULL, false,
> +					     false);
> +}
> +
> +static inline int regmap_field_clear_bits(struct regmap_field *field,
> +					  unsigned int bits)
> +{
> +	return regmap_field_update_bits_base(field, bits, bits, NULL, false,
> +					     false);

The contents of these two functions are swapped when compared to their names.

Regards,
Samuel

> +}
> +
> +int regmap_field_test_bits(struct regmap_field *field, unsigned int bits);
> +
>  static inline int
>  regmap_field_force_update_bits(struct regmap_field *field,
>  			       unsigned int mask, unsigned int val)
> @@ -1757,6 +1773,27 @@ regmap_field_force_update_bits(struct regmap_field *field,
>  	return -EINVAL;
>  }
>  
> +static inline int regmap_field_set_bits(struct regmap_field *field,
> +					unsigned int bits)
> +{
> +	WARN_ONCE(1, "regmap API is disabled");
> +	return -EINVAL;
> +}
> +
> +static inline int regmap_field_clear_bits(struct regmap_field *field,
> +					  unsigned int bits)
> +{
> +	WARN_ONCE(1, "regmap API is disabled");
> +	return -EINVAL;
> +}
> +
> +static inline int regmap_field_test_bits(struct regmap_field *field,
> +					 unsigned int bits)
> +{
> +	WARN_ONCE(1, "regmap API is disabled");
> +	return -EINVAL;
> +}
> +
>  static inline int regmap_fields_write(struct regmap_field *field,
>  				      unsigned int id, unsigned int val)
>  {
>
Li Chen May 23, 2022, 3:30 a.m. UTC | #2
Hi Samuel,

 ---- On Sun, 22 May 2022 20:09:55 -0700 Samuel Holland <samuel@sholland.org> wrote ----
 > Hi,
 > 
 > On 5/22/22 9:23 PM, Li Chen wrote:
 > > From: Li Chen <lchen@ambarella.com>
 > > 
 > > We have set/clear/test operations for regmap, but not for regmap_field yet.
 > > So let's introduce regmap_field helpers too.
 > > 
 > > In many instances regmap_field_update_bits() is used for simple bit setting
 > > and clearing. In these cases the last argument is redundant and we can
 > > hide it with a static inline function.
 > > 
 > > This adds three new helpers for simple bit operations: set_bits,
 > > clear_bits and test_bits (the last one defined as a regular function).
 > > 
 > > Signed-off-by: Li Chen <lchen@ambarella.com>
 > > ---
 > >  drivers/base/regmap/regmap.c | 22 +++++++++++++++++++++
 > >  include/linux/regmap.h       | 37 ++++++++++++++++++++++++++++++++++++
 > >  2 files changed, 59 insertions(+)
 > > 
 > > diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c
 > > index 5e12f7cb5147..a37d6041b7bd 100644
 > > --- a/drivers/base/regmap/regmap.c
 > > +++ b/drivers/base/regmap/regmap.c
 > > @@ -2208,6 +2208,28 @@ int regmap_field_update_bits_base(struct regmap_field *field,
 > >  }
 > >  EXPORT_SYMBOL_GPL(regmap_field_update_bits_base);
 > >  
 > > +/**
 > > + * regmap_field_test_bits() - Check if all specified bits are set in a
 > > + *                            register field.
 > > + *
 > > + * @field: Register field to operate on
 > > + * @bits: Bits to test
 > > + *
 > > + * Returns -1 if the underlying regmap_field_read() fails, 0 if at least one of the
 > > + * tested bits is not set and 1 if all tested bits are set.
 > > + */
 > > +int regmap_field_test_bits(struct regmap_field *field, unsigned int bits)
 > > +{
 > > +    unsigned int val, ret;
 > > +
 > > +    ret = regmap_field_read(field, &val);
 > > +    if (ret)
 > > +        return ret;
 > > +
 > > +    return (val & bits) == bits;
 > > +}
 > > +EXPORT_SYMBOL_GPL(regmap_field_test_bits);
 > > +
 > >  /**
 > >   * regmap_fields_update_bits_base() - Perform a read/modify/write cycle a
 > >   *                                    register field with port ID
 > > diff --git a/include/linux/regmap.h b/include/linux/regmap.h
 > > index de81a94d7b30..10b410734d9e 100644
 > > --- a/include/linux/regmap.h
 > > +++ b/include/linux/regmap.h
 > > @@ -1324,6 +1324,22 @@ static inline int regmap_field_update_bits(struct regmap_field *field,
 > >                           NULL, false, false);
 > >  }
 > >  
 > > +static inline int regmap_field_set_bits(struct regmap_field *field,
 > > +                    unsigned int bits)
 > > +{
 > > +    return regmap_field_update_bits_base(field, bits, 0, NULL, false,
 > > +                         false);
 > > +}
 > > +
 > > +static inline int regmap_field_clear_bits(struct regmap_field *field,
 > > +                      unsigned int bits)
 > > +{
 > > +    return regmap_field_update_bits_base(field, bits, bits, NULL, false,
 > > +                         false);
 > 
 > The contents of these two functions are swapped when compared to their names.

Thanks for spotting this out!
Fixed in v3.

Regards,
Li
 > 
 > Regards,
 > Samuel
 > 
 > > +}
 > > +
 > > +int regmap_field_test_bits(struct regmap_field *field, unsigned int bits);
 > > +
 > >  static inline int
 > >  regmap_field_force_update_bits(struct regmap_field *field,
 > >                     unsigned int mask, unsigned int val)
 > > @@ -1757,6 +1773,27 @@ regmap_field_force_update_bits(struct regmap_field *field,
 > >      return -EINVAL;
 > >  }
 > >  
 > > +static inline int regmap_field_set_bits(struct regmap_field *field,
 > > +                    unsigned int bits)
 > > +{
 > > +    WARN_ONCE(1, "regmap API is disabled");
 > > +    return -EINVAL;
 > > +}
 > > +
 > > +static inline int regmap_field_clear_bits(struct regmap_field *field,
 > > +                      unsigned int bits)
 > > +{
 > > +    WARN_ONCE(1, "regmap API is disabled");
 > > +    return -EINVAL;
 > > +}
 > > +
 > > +static inline int regmap_field_test_bits(struct regmap_field *field,
 > > +                     unsigned int bits)
 > > +{
 > > +    WARN_ONCE(1, "regmap API is disabled");
 > > +    return -EINVAL;
 > > +}
 > > +
 > >  static inline int regmap_fields_write(struct regmap_field *field,
 > >                        unsigned int id, unsigned int val)
 > >  {
 > > 
 > 
 >
Mark Brown May 23, 2022, 11:39 a.m. UTC | #3
On Sun, May 22, 2022 at 07:22:37PM -0700, Li Chen wrote:
> From: Li Chen <lchen@ambarella.com>
> 
> This series proposes to add simple bit operations for setting, clearing
> and testing specific bits with regmap_field.

Please don't send new patches in reply to old patches or serieses, this
makes it harder for both people and tools to understand what is going
on - it can bury things in mailboxes and make it difficult to keep track
of what current patches are, both for the new patches and the old ones.
Li Chen May 23, 2022, 1:19 p.m. UTC | #4
Hi Mark

 ---- On Mon, 23 May 2022 04:39:46 -0700 Mark Brown <broonie@kernel.org> wrote ----
 > On Sun, May 22, 2022 at 07:22:37PM -0700, Li Chen wrote:
 > > From: Li Chen <lchen@ambarella.com>
 > > 
 > > This series proposes to add simple bit operations for setting, clearing
 > > and testing specific bits with regmap_field.
 > 
 > Please don't send new patches in reply to old patches or serieses, this
 > makes it harder for both people and tools to understand what is going
 > on - it can bury things in mailboxes and make it difficult to keep track
 > of what current patches are, both for the new patches and the old ones.
 > 

Thanks for letting me know, I won't do this again.

Regards,
Li