Received-SPF: pass (google.com: domain of linux-kernel+bounces-142454-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
From: Jani Nikula <jani.nikula@intel.com>
To: Steven Price <steven.price@arm.com>, Andrew Morton
 <akpm@linux-foundation.org>, David Laight <David.Laight@ACULAB.COM>
Cc: "dri-devel@lists.freedesktop.org" <dri-devel@lists.freedesktop.org>,
 "intel-gfx@lists.freedesktop.org" <intel-gfx@lists.freedesktop.org>,
 "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>, David Gow
 <davidgow@google.com>, Christian =?utf-8?Q?K=C3=B6nig?=
 <christian.koenig@amd.com>
Subject: Re: [PATCH 0/4] log2: make is_power_of_2() more generic
In-Reply-To: <8905800b-a977-e821-01ea-a43333f46904@arm.com>
Organization: Intel Finland Oy - BIC 0357606-4 - Westendinkatu 7, 02160 Espoo
References: <20230330104243.2120761-1-jani.nikula@intel.com>
 <20230330125041.83b0f39fa3a4ec1a42dfd95f@linux-foundation.org>
 <549987e4967d45159573901d330c96a0@AcuMS.aculab.com>
 <20230330151846.fdbc8edbfbaa6eaddb056dc7@linux-foundation.org>
 <87edp52ufk.fsf@intel.com> <8905800b-a977-e821-01ea-a43333f46904@arm.com>
Date: Fri, 12 Apr 2024 13:01:36 +0300
Message-ID: <87jzl2527z.fsf@intel.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain

On Wed, 05 Apr 2023, Steven Price <steven.price@arm.com> wrote:
> On 31/03/2023 09:31, Jani Nikula wrote:
>> On Thu, 30 Mar 2023, Andrew Morton <akpm@linux-foundation.org> wrote:
>>> On Thu, 30 Mar 2023 21:53:03 +0000 David Laight <David.Laight@ACULAB.COM> wrote:
>>>
>>>>> But wouldn't all these issues be addressed by simply doing
>>>>>
>>>>> #define is_power_of_2(n) (n != 0 && ((n & (n - 1)) == 0))
>>>>>
>>>>> ?
>>>>>
>>>>> (With suitable tweaks to avoid evaluating `n' more than once)
>>>>
>>>> I think you need to use the 'horrid tricks' from min() to get
>>>> a constant expression from constant inputs.
>>>
>>> This
>>>
>>> --- a/include/linux/log2.h~a
>>> +++ a/include/linux/log2.h
>>> @@ -41,11 +41,11 @@ int __ilog2_u64(u64 n)
>>>   * *not* considered a power of two.
>>>   * Return: true if @n is a power of 2, otherwise false.
>>>   */
>>> -static inline __attribute__((const))
>>> -bool is_power_of_2(unsigned long n)
>>> -{
>>> -	return (n != 0 && ((n & (n - 1)) == 0));
>>> -}
>>> +#define is_power_of_2(_n)				\
>>> +	({						\
>>> +		typeof(_n) n = (_n);			\
>>> +		n != 0 && ((n & (n - 1)) == 0);		\
>>> +	})
>>>  
>>>  /**
>>>   * __roundup_pow_of_two() - round up to nearest power of two
>>> _
>>>
>>> worked for me in a simple test.
>>>
>>> --- a/fs/open.c~b
>>> +++ a/fs/open.c
>>> @@ -1564,3 +1564,10 @@ int stream_open(struct inode *inode, str
>>>  }
>>>  
>>>  EXPORT_SYMBOL(stream_open);
>>> +
>>> +#include <linux/log2.h>
>>> +
>>> +int foo(void)
>>> +{
>>> +	return is_power_of_2(43);
>>> +}
>>> _
>>>
>>>
>>> foo:
>>> # fs/open.c:1573: }
>>> 	xorl	%eax, %eax	#
>>> 	ret	
>>>
>>>
>>> Is there some more tricky situation where it breaks?
>> 
>> It doesn't work with BUILD_BUG_ON_ZERO().
>
> Like most programming problems, you just need another layer of
> indirection! The below works for me in all the cases I could think of
> (including __uint128_t).
>
>
> #define __IS_POWER_OF_2(n) (n != 0 && ((n & (n - 1)) == 0))
>
> #define _IS_POWER_OF_2(n, unique_n)				\
> 	({							\
> 		typeof(n) unique_n = (n);			\
> 		__IS_POWER_OF_2(unique_n);			\
> 	})
>
> #define is_power_of_2(n)					\
> 	__builtin_choose_expr(__is_constexpr((n)),		\
> 			      __IS_POWER_OF_2((n)),		\
> 			      _IS_POWER_OF_2(n, __UNIQUE_ID(_n)))
>
>
> Although Jani's original might be easier to understand.

I dropped the ball since I couldn't make heads or tails what I should be
doing. And a year has passed. I'll note that the kernel has a number of
helpers for "is power of 2" for u64 and for constant expressions,
outside of log2.h.

I tried to make is_power_of_2() work for all the cases. Would it be more
palatable if I just added all the variants separately to log2.h?

- Leave is_power_of_2() as is
- Add is_power_of_2_u64() for 32-bit build compatible 64-bit checks
- Add IS_POWER_OF_2() macro for constant expressions

Please just tell me what to do and I'll do it.

BR,
Jani.


-- 
Jani Nikula, Intel