diff mbox series

[v3,06/12] minmax: Introduce {min,max}_array()

Message ID 20230612122926.107333-7-herve.codina@bootlin.com
State New
Headers show
Series Add support for IIO devices in ASoC | expand

Commit Message

Herve Codina June 12, 2023, 12:29 p.m. UTC
Introduce min_array() (resp max_array()) in order to get the
minimal (resp maximum) of values present in an array.

Signed-off-by: Herve Codina <herve.codina@bootlin.com>
---
 include/linux/minmax.h | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

Comments

Herve Codina June 13, 2023, 8 a.m. UTC | #1
Hi Andy,

On Mon, 12 Jun 2023 17:10:40 +0300
Andy Shevchenko <andy.shevchenko@gmail.com> wrote:

> On Mon, Jun 12, 2023 at 3:30 PM Herve Codina <herve.codina@bootlin.com> wrote:
> >
> > Introduce min_array() (resp max_array()) in order to get the
> > minimal (resp maximum) of values present in an array.  
> 
> Some comments below, after addressing them,
> Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>
> 
> > Signed-off-by: Herve Codina <herve.codina@bootlin.com>
> > ---
> >  include/linux/minmax.h | 26 ++++++++++++++++++++++++++
> >  1 file changed, 26 insertions(+)
> >
> > diff --git a/include/linux/minmax.h b/include/linux/minmax.h
> > index 396df1121bff..37a211f22404 100644
> > --- a/include/linux/minmax.h
> > +++ b/include/linux/minmax.h
> > @@ -133,6 +133,32 @@
> >   */
> >  #define max_t(type, x, y)      __careful_cmp((type)(x), (type)(y), >)
> >
> > +#define __minmax_array(op, array, len) ({                      \  
> 
> Maybe it's my MUA, maybe the code contains spaces, can you switch to
> TABs if it's the case?
> 
> > +       typeof(array) __array = (array);                        \  
> 
> We have __must_be_array()

Using __must_be_array() will lead to some failure.
Indeed, we can have:
  --- 8< ---
  int *buff
  ...
  min = min_array(buff, nb_item);
  --- 8< ---

In this case, __must_be_array() will report that buff is not an array.

To avoid any confusion, what do you think if I renamed {min,max}_array()
to {min,max}_buffer() and replace __array by __buff and use *(__buff + xxx)
instead of array[xxx] in the macro.

This will lead to:
--- 8< ---
#define __minmax_buffer(op, buff, len) ({			\
	typeof(buff) __buff = (buff);			\
	typeof(len) __len = (len);				\
	typeof(*buff + 0) __element = *(__buff + --__len);	\
	while (__len--)						\
		__element = op(__element, *(__buff + __len]));	\
	__element; })

#define min_buffer(buffer, len) __minmax_array(min, buffer, len)
#define max_buffer(buffer, len) __minmax_array(max, buffer, len)
--- 8< ---

Regards,
Hervé

> 
> You will need to fix the inclusions in minmax.h at the same time, it needs
> linux/build_bug.h (which includes compiler.h needed for __UNIQUE_ID()
> and for the above mentioned one).
> 
> > +       typeof(len) __len = (len);                              \
> > +       typeof(*__array + 0) __element = __array[--__len];      \  
> 
> After above, this can be written as __array[0].
> 
> > +       while (__len--)                                         \
> > +               __element = op(__element, __array[__len]);      \
> > +       __element; })
> > +
> > +/**
> > + * min_array - return minimum of values present in an array
> > + * @array: array
> > + * @len: array length
> > + *
> > + * Note that @len must not be zero (empty array).
> > + */
> > +#define min_array(array, len) __minmax_array(min, array, len)
> > +
> > +/**
> > + * max_array - return maximum of values present in an array
> > + * @array: array
> > + * @len: array length
> > + *
> > + * Note that @len must not be zero (empty array).
> > + */
> > +#define max_array(array, len) __minmax_array(max, array, len)
> > +
> >  /**
> >   * clamp_t - return a value clamped to a given range using a given type
> >   * @type: the type of variable to use  
>
Andy Shevchenko June 13, 2023, 5:08 p.m. UTC | #2
On Tue, Jun 13, 2023 at 11:00 AM Herve Codina <herve.codina@bootlin.com> wrote:
> On Mon, 12 Jun 2023 17:10:40 +0300
> Andy Shevchenko <andy.shevchenko@gmail.com> wrote:
> > On Mon, Jun 12, 2023 at 3:30 PM Herve Codina <herve.codina@bootlin.com> wrote:
> > >
> > > Introduce min_array() (resp max_array()) in order to get the
> > > minimal (resp maximum) of values present in an array.
> >
> > Some comments below, after addressing them,
> > Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>

...

> > > +       typeof(array) __array = (array);                        \
> >
> > We have __must_be_array()
>
> Using __must_be_array() will lead to some failure.
> Indeed, we can have:
>   --- 8< ---
>   int *buff
>   ...
>   min = min_array(buff, nb_item);
>   --- 8< ---
>
> In this case, __must_be_array() will report that buff is not an array.

Oh, I missed that.

> To avoid any confusion, what do you think if I renamed {min,max}_array()
> to {min,max}_buffer() and replace __array by __buff and use *(__buff + xxx)
> instead of array[xxx] in the macro.

But functionally it's still against an array.

I would stick with "array" in the name, but add a comment why
__must_be_array() is not used. If we need a stricter variant, we may
add a new wrapper with that check. That said, I think we can use
__array[0] and similar indexed accesses.


--
With Best Regards,
Andy Shevchenko
Herve Codina June 13, 2023, 5:34 p.m. UTC | #3
On Tue, 13 Jun 2023 20:08:08 +0300
Andy Shevchenko <andy.shevchenko@gmail.com> wrote:

> On Tue, Jun 13, 2023 at 11:00 AM Herve Codina <herve.codina@bootlin.com> wrote:
> > On Mon, 12 Jun 2023 17:10:40 +0300
> > Andy Shevchenko <andy.shevchenko@gmail.com> wrote:  
> > > On Mon, Jun 12, 2023 at 3:30 PM Herve Codina <herve.codina@bootlin.com> wrote:  
> > > >
> > > > Introduce min_array() (resp max_array()) in order to get the
> > > > minimal (resp maximum) of values present in an array.  
> > >
> > > Some comments below, after addressing them,
> > > Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com>  
> 
> ...
> 
> > > > +       typeof(array) __array = (array);                        \  
> > >
> > > We have __must_be_array()  
> >
> > Using __must_be_array() will lead to some failure.
> > Indeed, we can have:
> >   --- 8< ---
> >   int *buff
> >   ...
> >   min = min_array(buff, nb_item);
> >   --- 8< ---
> >
> > In this case, __must_be_array() will report that buff is not an array.  
> 
> Oh, I missed that.
> 
> > To avoid any confusion, what do you think if I renamed {min,max}_array()
> > to {min,max}_buffer() and replace __array by __buff and use *(__buff + xxx)
> > instead of array[xxx] in the macro.  
> 
> But functionally it's still against an array.
> 
> I would stick with "array" in the name, but add a comment why
> __must_be_array() is not used. If we need a stricter variant, we may
> add a new wrapper with that check. That said, I think we can use
> __array[0] and similar indexed accesses.
> 

Right, I will provide an updated version on the next iteration.

Thanks for your feedback.
Hervé
diff mbox series

Patch

diff --git a/include/linux/minmax.h b/include/linux/minmax.h
index 396df1121bff..37a211f22404 100644
--- a/include/linux/minmax.h
+++ b/include/linux/minmax.h
@@ -133,6 +133,32 @@ 
  */
 #define max_t(type, x, y)	__careful_cmp((type)(x), (type)(y), >)
 
+#define __minmax_array(op, array, len) ({			\
+	typeof(array) __array = (array);			\
+	typeof(len) __len = (len);				\
+	typeof(*__array + 0) __element = __array[--__len];	\
+	while (__len--)						\
+		__element = op(__element, __array[__len]);	\
+	__element; })
+
+/**
+ * min_array - return minimum of values present in an array
+ * @array: array
+ * @len: array length
+ *
+ * Note that @len must not be zero (empty array).
+ */
+#define min_array(array, len) __minmax_array(min, array, len)
+
+/**
+ * max_array - return maximum of values present in an array
+ * @array: array
+ * @len: array length
+ *
+ * Note that @len must not be zero (empty array).
+ */
+#define max_array(array, len) __minmax_array(max, array, len)
+
 /**
  * clamp_t - return a value clamped to a given range using a given type
  * @type: the type of variable to use