[PATCH] bitops: simplify get_count_order_long()

[PATCH] bitops: simplify get_count_order_long()

[Wei Yang]

These two cases could be unified into one.

Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
---
 include/linux/bitops.h | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/include/linux/bitops.h b/include/linux/bitops.h
index 9acf654f0b19..5b5609e81a84 100644
--- a/include/linux/bitops.h
+++ b/include/linux/bitops.h
@@ -206,10 +206,7 @@ static inline int get_count_order_long(unsigned long l)
 {
 	if (l == 0UL)
 		return -1;
-	else if (l & (l - 1UL))
-		return (int)fls_long(l);
-	else
-		return (int)fls_long(l) - 1;
+	return (int)fls_long(--l);
 }
 
 /**
-- 
2.23.0

Re: [PATCH] bitops: simplify get_count_order_long()

[Andy Shevchenko]

On Sun, May 24, 2020 at 12:35:51PM +0000, Wei Yang wrote:
> These two cases could be unified into one.

Care to provide a test case which supports your change?

> Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
> ---
>  include/linux/bitops.h | 5 +----
>  1 file changed, 1 insertion(+), 4 deletions(-)
> 
> diff --git a/include/linux/bitops.h b/include/linux/bitops.h
> index 9acf654f0b19..5b5609e81a84 100644
> --- a/include/linux/bitops.h
> +++ b/include/linux/bitops.h
> @@ -206,10 +206,7 @@ static inline int get_count_order_long(unsigned long l)
>  {
>  	if (l == 0UL)
>  		return -1;
> -	else if (l & (l - 1UL))
> -		return (int)fls_long(l);
> -	else
> -		return (int)fls_long(l) - 1;
> +	return (int)fls_long(--l);
>  }

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH] bitops: simplify get_count_order_long()

[Wei Yang]

On Mon, May 25, 2020 at 12:14:58PM +0300, Andy Shevchenko wrote:
>On Sun, May 24, 2020 at 12:35:51PM +0000, Wei Yang wrote:
>> These two cases could be unified into one.
>
>Care to provide a test case which supports your change?
>

Hmm.. where should I put the test? tools/testing/selftests/ ?

>> Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
>> ---
>>  include/linux/bitops.h | 5 +----
>>  1 file changed, 1 insertion(+), 4 deletions(-)
>> 
>> diff --git a/include/linux/bitops.h b/include/linux/bitops.h
>> index 9acf654f0b19..5b5609e81a84 100644
>> --- a/include/linux/bitops.h
>> +++ b/include/linux/bitops.h
>> @@ -206,10 +206,7 @@ static inline int get_count_order_long(unsigned long l)
>>  {
>>  	if (l == 0UL)
>>  		return -1;
>> -	else if (l & (l - 1UL))
>> -		return (int)fls_long(l);
>> -	else
>> -		return (int)fls_long(l) - 1;
>> +	return (int)fls_long(--l);
>>  }
>
>-- 
>With Best Regards,
>Andy Shevchenko
>

-- 
Wei Yang
Help you, Help me

Re: [PATCH] bitops: simplify get_count_order_long()

[Andy Shevchenko]

On Mon, May 25, 2020 at 02:43:12PM +0000, Wei Yang wrote:
> On Mon, May 25, 2020 at 12:14:58PM +0300, Andy Shevchenko wrote:
> >On Sun, May 24, 2020 at 12:35:51PM +0000, Wei Yang wrote:
> >> These two cases could be unified into one.
> >
> >Care to provide a test case which supports your change?
> >
> 
> Hmm.. where should I put the test? tools/testing/selftests/ ?

I guess into test_bitops.c [1]? I though it eventually should make kernel, but I don't see it.

Andrew, can you apply that or do you need Jesse to resend?

[1]: https://lore.kernel.org/lkml/20200310221747.2848474-2-jesse.brandeburg@intel.com/

> 
> >> Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
> >> ---
> >>  include/linux/bitops.h | 5 +----
> >>  1 file changed, 1 insertion(+), 4 deletions(-)
> >> 
> >> diff --git a/include/linux/bitops.h b/include/linux/bitops.h
> >> index 9acf654f0b19..5b5609e81a84 100644
> >> --- a/include/linux/bitops.h
> >> +++ b/include/linux/bitops.h
> >> @@ -206,10 +206,7 @@ static inline int get_count_order_long(unsigned long l)
> >>  {
> >>  	if (l == 0UL)
> >>  		return -1;
> >> -	else if (l & (l - 1UL))
> >> -		return (int)fls_long(l);
> >> -	else
> >> -		return (int)fls_long(l) - 1;
> >> +	return (int)fls_long(--l);
> >>  }
> >
> >-- 
> >With Best Regards,
> >Andy Shevchenko
> >
> 
> -- 
> Wei Yang
> Help you, Help me

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH] bitops: simplify get_count_order_long()

[Andrew Morton]

On Mon, 25 May 2020 18:32:16 +0300 Andy Shevchenko <andriy.shevchenko@linux.intel.com> wrote:

> On Mon, May 25, 2020 at 02:43:12PM +0000, Wei Yang wrote:
> > On Mon, May 25, 2020 at 12:14:58PM +0300, Andy Shevchenko wrote:
> > >On Sun, May 24, 2020 at 12:35:51PM +0000, Wei Yang wrote:
> > >> These two cases could be unified into one.
> > >
> > >Care to provide a test case which supports your change?

I hurt my brain convincing myself, so I got practical:


int fls(unsigned int x)
{
	return x ? sizeof(x) * 8 - __builtin_clz(x) : 0;
}

static int get_count_order(unsigned l)
{
	if (l == 0)
		return -1;
	else if (l & (l - 1UL))
		return fls(l);
	else
		return fls(l) - 1;
}										

static int get_count_order2(unsigned long l)
{
	if (l == 0)
		return -1;
	return fls(--l);
}

main()
{
	unsigned i;

	for (i = 1; i < 64; i++) {
		printf("%d %d\n", get_count_order(i),
			get_count_order2(i));
	}
}


> > 
> > Hmm.. where should I put the test? tools/testing/selftests/ ?
> 
> I guess into test_bitops.c [1]? I though it eventually should make kernel, but I don't see it.
> 
> Andrew, can you apply that or do you need Jesse to resend?
> 

Got it.

Re: [PATCH] bitops: simplify get_count_order_long()

[Wei Yang]

On Mon, May 25, 2020 at 01:41:10PM -0700, Andrew Morton wrote:
>On Mon, 25 May 2020 18:32:16 +0300 Andy Shevchenko <andriy.shevchenko@linux.intel.com> wrote:
>
>> On Mon, May 25, 2020 at 02:43:12PM +0000, Wei Yang wrote:
>> > On Mon, May 25, 2020 at 12:14:58PM +0300, Andy Shevchenko wrote:
>> > >On Sun, May 24, 2020 at 12:35:51PM +0000, Wei Yang wrote:
>> > >> These two cases could be unified into one.
>> > >
>> > >Care to provide a test case which supports your change?
>
>I hurt my brain convincing myself, so I got practical:
>
>
>int fls(unsigned int x)
>{
>	return x ? sizeof(x) * 8 - __builtin_clz(x) : 0;
>}
>
>static int get_count_order(unsigned l)
>{
>	if (l == 0)
>		return -1;
>	else if (l & (l - 1UL))
>		return fls(l);
>	else
>		return fls(l) - 1;
>}										
>
>static int get_count_order2(unsigned long l)
>{
>	if (l == 0)
>		return -1;
>	return fls(--l);
>}
>
>main()
>{
>	unsigned i;
>
>	for (i = 1; i < 64; i++) {
>		printf("%d %d\n", get_count_order(i),
>			get_count_order2(i));
>	}
>}
>
>
>> > 
>> > Hmm.. where should I put the test? tools/testing/selftests/ ?
>> 
>> I guess into test_bitops.c [1]? I though it eventually should make kernel, but I don't see it.
>> 
>> Andrew, can you apply that or do you need Jesse to resend?
>> 
>
>Got it.

I see the patch just merged, so I suppose to add the above test code into that
one?

-- 
Wei Yang
Help you, Help me

Re: [PATCH] bitops: simplify get_count_order_long()

[Andrew Morton]

On Mon, 25 May 2020 21:57:41 +0000 Wei Yang <richard.weiyang@gmail.com> wrote:

> I see the patch just merged, so I suppose to add the above test code into that
> one?

Well, that's not really test code.

But yes, something which tests both the 32-bit and 64-bit functions would be
nice, sometime.

Re: [PATCH] bitops: simplify get_count_order_long()

[Wei Yang]

On Mon, May 25, 2020 at 03:31:46PM -0700, Andrew Morton wrote:
>On Mon, 25 May 2020 21:57:41 +0000 Wei Yang <richard.weiyang@gmail.com> wrote:
>
>> I see the patch just merged, so I suppose to add the above test code into that
>> one?
>
>Well, that's not really test code.
>
>But yes, something which tests both the 32-bit and 64-bit functions would be
>nice, sometime.

Mimic the test_bitops.c, I wrote a test like this:

/* a tiny module only meant to test get_count_order/long */
unsigned int order_comb[][2] = {
	{0x00000003,  2},
	{0x00000004,  2},
	{0x00001fff, 13},
	{0x00002000, 13},
	{0x50000000, 32},
	{0x80000000, 32},
};

static int __init test_getorder_startup(void)
{
	int i;

	for (i = 0; i < ARRAY_SIZE(order_comb); i++) {
		if (order_comb[i][1] != get_count_order(order_comb[i][0]))
			pr_warn("get_count_order wrong for %lx\n",
					order_comb[i][0]);
	}

	return 0;
}

Since I don't get a way to iterate all the possibilities, some random
combination is chosen. Is this one looks good?

-- 
Wei Yang
Help you, Help me

Re: [PATCH] bitops: simplify get_count_order_long()

[Andrew Morton]

On Wed, 27 May 2020 22:45:42 +0000 Wei Yang <richard.weiyang@gmail.com> wrote:

> /* a tiny module only meant to test get_count_order/long */
> unsigned int order_comb[][2] = {
> 	{0x00000003,  2},
> 	{0x00000004,  2},
> 	{0x00001fff, 13},
> 	{0x00002000, 13},
> 	{0x50000000, 32},
> 	{0x80000000, 32},
> };
> 
> static int __init test_getorder_startup(void)
> {
> 	int i;
> 
> 	for (i = 0; i < ARRAY_SIZE(order_comb); i++) {
> 		if (order_comb[i][1] != get_count_order(order_comb[i][0]))
> 			pr_warn("get_count_order wrong for %lx\n",
> 					order_comb[i][0]);
> 	}
> 
> 	return 0;
> }
> 
> Since I don't get a way to iterate all the possibilities, some random
> combination is chosen. Is this one looks good?

Looks good.

You might want to add a less-negative number as well?  0x80030000. 
Something that won't turn positive if it has 1 subtracted from it.

Re: [PATCH] bitops: simplify get_count_order_long()

[Wei Yang]

On Wed, May 27, 2020 at 04:05:08PM -0700, Andrew Morton wrote:
>On Wed, 27 May 2020 22:45:42 +0000 Wei Yang <richard.weiyang@gmail.com> wrote:
>
>> /* a tiny module only meant to test get_count_order/long */
>> unsigned int order_comb[][2] = {
>> 	{0x00000003,  2},
>> 	{0x00000004,  2},
>> 	{0x00001fff, 13},
>> 	{0x00002000, 13},
>> 	{0x50000000, 32},
>> 	{0x80000000, 32},
>> };
>> 
>> static int __init test_getorder_startup(void)
>> {
>> 	int i;
>> 
>> 	for (i = 0; i < ARRAY_SIZE(order_comb); i++) {
>> 		if (order_comb[i][1] != get_count_order(order_comb[i][0]))
>> 			pr_warn("get_count_order wrong for %lx\n",
>> 					order_comb[i][0]);
>> 	}
>> 
>> 	return 0;
>> }
>> 
>> Since I don't get a way to iterate all the possibilities, some random
>> combination is chosen. Is this one looks good?
>
>Looks good.
>
>You might want to add a less-negative number as well?  0x80030000. 
>Something that won't turn positive if it has 1 subtracted from it.

Thanks, this is a good suggestion.

-- 
Wei Yang
Help you, Help me