[PATCH] powerpc: define __LITTLE_ENDIAN and __BIG_ENDIAN for math-emu

[PATCH] powerpc: define __LITTLE_ENDIAN and __BIG_ENDIAN for math-emu

[Mingcong Bai]

Similar to commit b929926f01f2 ("sh: define __BIG_ENDIAN for math-emu"),
define __LITTLE_ENDIAN and __BIG_ENDIAN as 0 to mitigate build-time
warnings:

  ./include/math-emu/double.h:59:21: error: ‘__BIG_ENDIAN’ is not defined, evaluates to ‘0’ [-Werror=undef]
     59 | #if __BYTE_ORDER == __BIG_ENDIAN
        |

Cc: stable@vger.kernel.org
Fixes: 13da9e200fe4 ("Revert "endian: #define __BYTE_ORDER"")
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202507301656.7FEX6J5W-lkp@intel.com/
Signed-off-by: Mingcong Bai <jeffbai@aosc.io>
---
 arch/powerpc/include/asm/sfp-machine.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/powerpc/include/asm/sfp-machine.h b/arch/powerpc/include/asm/sfp-machine.h
index 8b957aabb826d..db8525605c026 100644
--- a/arch/powerpc/include/asm/sfp-machine.h
+++ b/arch/powerpc/include/asm/sfp-machine.h
@@ -319,10 +319,12 @@
 #define abort()								\
 	return 0
 
-#ifdef __BIG_ENDIAN
+#ifdef __BIG_ENDIAN__
 #define __BYTE_ORDER __BIG_ENDIAN
+#define __LITTLE_ENDIAN 0
 #else
 #define __BYTE_ORDER __LITTLE_ENDIAN
+#define __BIG_ENDIAN 0
 #endif
 
 /* Exception flags. */
-- 
2.52.0

Re: [PATCH] powerpc: define __LITTLE_ENDIAN and __BIG_ENDIAN for math-emu

[David Laight]

On Sun, 17 May 2026 12:14:21 +0800
Mingcong Bai <jeffbai@aosc.io> wrote:

> Similar to commit b929926f01f2 ("sh: define __BIG_ENDIAN for math-emu"),
> define __LITTLE_ENDIAN and __BIG_ENDIAN as 0 to mitigate build-time
> warnings:
> 
>   ./include/math-emu/double.h:59:21: error: ‘__BIG_ENDIAN’ is not defined, evaluates to ‘0’ [-Werror=undef]
>      59 | #if __BYTE_ORDER == __BIG_ENDIAN
>         |
> 
> Cc: stable@vger.kernel.org
> Fixes: 13da9e200fe4 ("Revert "endian: #define __BYTE_ORDER"")
> Reported-by: kernel test robot <lkp@intel.com>
> Closes: https://lore.kernel.org/oe-kbuild-all/202507301656.7FEX6J5W-lkp@intel.com/
> Signed-off-by: Mingcong Bai <jeffbai@aosc.io>
> ---
>  arch/powerpc/include/asm/sfp-machine.h | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/powerpc/include/asm/sfp-machine.h b/arch/powerpc/include/asm/sfp-machine.h
> index 8b957aabb826d..db8525605c026 100644
> --- a/arch/powerpc/include/asm/sfp-machine.h
> +++ b/arch/powerpc/include/asm/sfp-machine.h
> @@ -319,10 +319,12 @@
>  #define abort()								\
>  	return 0
>  
> -#ifdef __BIG_ENDIAN
> +#ifdef __BIG_ENDIAN__
>  #define __BYTE_ORDER __BIG_ENDIAN
> +#define __LITTLE_ENDIAN 0
>  #else
>  #define __BYTE_ORDER __LITTLE_ENDIAN
> +#define __BIG_ENDIAN 0
>  #endif

I thought the expected/correct value for __BYTE_ORDER__ was either 1234 or 4321.
(apart from pdp11's 2143).

-- David

>  
>  /* Exception flags. */

Re: [PATCH] powerpc: define __LITTLE_ENDIAN and __BIG_ENDIAN for math-emu

[Xi Ruoyao]

On Sun, 2026-05-17 at 14:54 +0100, David Laight wrote:
> On Sun, 17 May 2026 12:14:21 +0800
> Mingcong Bai <jeffbai@aosc.io> wrote:
> 
> > Similar to commit b929926f01f2 ("sh: define __BIG_ENDIAN for math-emu"),
> > define __LITTLE_ENDIAN and __BIG_ENDIAN as 0 to mitigate build-time
> > warnings:
> > 
> >   ./include/math-emu/double.h:59:21: error: ‘__BIG_ENDIAN’ is not defined, evaluates to ‘0’ [-Werror=undef]
> >      59 | #if __BYTE_ORDER == __BIG_ENDIAN
> >         |
> > 
> > Cc: stable@vger.kernel.org
> > Fixes: 13da9e200fe4 ("Revert "endian: #define __BYTE_ORDER"")
> > Reported-by: kernel test robot <lkp@intel.com>
> > Closes: https://lore.kernel.org/oe-kbuild-all/202507301656.7FEX6J5W-lkp@intel.com/
> > Signed-off-by: Mingcong Bai <jeffbai@aosc.io>
> > ---
> >  arch/powerpc/include/asm/sfp-machine.h | 4 +++-
> >  1 file changed, 3 insertions(+), 1 deletion(-)
> > 
> > diff --git a/arch/powerpc/include/asm/sfp-machine.h b/arch/powerpc/include/asm/sfp-machine.h
> > index 8b957aabb826d..db8525605c026 100644
> > --- a/arch/powerpc/include/asm/sfp-machine.h
> > +++ b/arch/powerpc/include/asm/sfp-machine.h
> > @@ -319,10 +319,12 @@
> >  #define abort()								\
> >  	return 0
> >  
> > -#ifdef __BIG_ENDIAN
> > +#ifdef __BIG_ENDIAN__
> >  #define __BYTE_ORDER __BIG_ENDIAN
> > +#define __LITTLE_ENDIAN 0
> >  #else
> >  #define __BYTE_ORDER __LITTLE_ENDIAN
> > +#define __BIG_ENDIAN 0
> >  #endif
> 
> I thought the expected/correct value for __BYTE_ORDER__ was either 1234 or 4321.
> (apart from pdp11's 2143).

Should we just do

#define __BYTE_ORDER __BYTE_ORDER__
#define __LITTLE_ENDIAN __ORDER_LITTLE_ENDIAN__
#define __BIG_ENDIAN __ORDER_BIG_ENDIAN__

then?  __BYTE_ORDER__ etc. are available since gcc 4.6 and now we
requires gcc >= 8 to build the kernel.


-- 
Xi Ruoyao <xry111@xry111.site>

Re: [PATCH] powerpc: define __LITTLE_ENDIAN and __BIG_ENDIAN for math-emu

[Christophe Leroy (CS GROUP)]

Le 17/05/2026 à 15:54, David Laight a écrit :
> On Sun, 17 May 2026 12:14:21 +0800
> Mingcong Bai <jeffbai@aosc.io> wrote:
> 
>> Similar to commit b929926f01f2 ("sh: define __BIG_ENDIAN for math-emu"),
>> define __LITTLE_ENDIAN and __BIG_ENDIAN as 0 to mitigate build-time
>> warnings:
>>
>>    ./include/math-emu/double.h:59:21: error: ‘__BIG_ENDIAN’ is not defined, evaluates to ‘0’ [-Werror=undef]
>>       59 | #if __BYTE_ORDER == __BIG_ENDIAN
>>          |
>>
>> Cc: stable@vger.kernel.org
>> Fixes: 13da9e200fe4 ("Revert "endian: #define __BYTE_ORDER"")
>> Reported-by: kernel test robot <lkp@intel.com>
>> Closes: https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore.kernel.org%2Foe-kbuild-all%2F202507301656.7FEX6J5W-lkp%40intel.com%2F&data=05%7C02%7Cchristophe.leroy%40csgroup.eu%7C08977974fb1c495e9bd508deb41bd275%7C8b87af7d86474dc78df45f69a2011bb5%7C0%7C0%7C639146228768693730%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=4qGulR%2BL7i7inksEbEH9jNGZS8HG80uvm3I9IyYzZww%3D&reserved=0
>> Signed-off-by: Mingcong Bai <jeffbai@aosc.io>
>> ---
>>   arch/powerpc/include/asm/sfp-machine.h | 4 +++-
>>   1 file changed, 3 insertions(+), 1 deletion(-)
>>
>> diff --git a/arch/powerpc/include/asm/sfp-machine.h b/arch/powerpc/include/asm/sfp-machine.h
>> index 8b957aabb826d..db8525605c026 100644
>> --- a/arch/powerpc/include/asm/sfp-machine.h
>> +++ b/arch/powerpc/include/asm/sfp-machine.h
>> @@ -319,10 +319,12 @@
>>   #define abort()								\
>>   	return 0
>>   
>> -#ifdef __BIG_ENDIAN
>> +#ifdef __BIG_ENDIAN__
>>   #define __BYTE_ORDER __BIG_ENDIAN
>> +#define __LITTLE_ENDIAN 0
>>   #else
>>   #define __BYTE_ORDER __LITTLE_ENDIAN
>> +#define __BIG_ENDIAN 0
>>   #endif
> 
> I thought the expected/correct value for __BYTE_ORDER__ was either 1234 or 4321.
> (apart from pdp11's 2143).

That's the case, in include/linux/kconfig.h we have:

#ifdef CONFIG_CPU_BIG_ENDIAN
#define __BIG_ENDIAN 4321
#else
#define __LITTLE_ENDIAN 1234
#endif

But as far as I understand the problem is that math-emu expects 
__BIG_ENDIAN to be defined at all time as it has tests like:

#if __BYTE_ORDER == __BIG_ENDIAN

Christophe

Re: [PATCH] powerpc: define __LITTLE_ENDIAN and __BIG_ENDIAN for math-emu

[David Laight]

On Wed, 20 May 2026 15:14:40 +0200
"Christophe Leroy (CS GROUP)" <chleroy@kernel.org> wrote:

> Le 17/05/2026 à 15:54, David Laight a écrit :
> > On Sun, 17 May 2026 12:14:21 +0800
> > Mingcong Bai <jeffbai@aosc.io> wrote:
> >   
> >> Similar to commit b929926f01f2 ("sh: define __BIG_ENDIAN for math-emu"),
> >> define __LITTLE_ENDIAN and __BIG_ENDIAN as 0 to mitigate build-time
> >> warnings:
> >>
> >>    ./include/math-emu/double.h:59:21: error: ‘__BIG_ENDIAN’ is not defined, evaluates to ‘0’ [-Werror=undef]
> >>       59 | #if __BYTE_ORDER == __BIG_ENDIAN
> >>          |
> >>
> >> Cc: stable@vger.kernel.org
> >> Fixes: 13da9e200fe4 ("Revert "endian: #define __BYTE_ORDER"")
> >> Reported-by: kernel test robot <lkp@intel.com>
> >> Closes: https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore.kernel.org%2Foe-kbuild-all%2F202507301656.7FEX6J5W-lkp%40intel.com%2F&data=05%7C02%7Cchristophe.leroy%40csgroup.eu%7C08977974fb1c495e9bd508deb41bd275%7C8b87af7d86474dc78df45f69a2011bb5%7C0%7C0%7C639146228768693730%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=4qGulR%2BL7i7inksEbEH9jNGZS8HG80uvm3I9IyYzZww%3D&reserved=0
> >> Signed-off-by: Mingcong Bai <jeffbai@aosc.io>
> >> ---
> >>   arch/powerpc/include/asm/sfp-machine.h | 4 +++-
> >>   1 file changed, 3 insertions(+), 1 deletion(-)
> >>
> >> diff --git a/arch/powerpc/include/asm/sfp-machine.h b/arch/powerpc/include/asm/sfp-machine.h
> >> index 8b957aabb826d..db8525605c026 100644
> >> --- a/arch/powerpc/include/asm/sfp-machine.h
> >> +++ b/arch/powerpc/include/asm/sfp-machine.h
> >> @@ -319,10 +319,12 @@
> >>   #define abort()								\
> >>   	return 0
> >>   
> >> -#ifdef __BIG_ENDIAN
> >> +#ifdef __BIG_ENDIAN__
> >>   #define __BYTE_ORDER __BIG_ENDIAN
> >> +#define __LITTLE_ENDIAN 0
> >>   #else
> >>   #define __BYTE_ORDER __LITTLE_ENDIAN
> >> +#define __BIG_ENDIAN 0
> >>   #endif  
> > 
> > I thought the expected/correct value for __BYTE_ORDER__ was either 1234 or 4321.
> > (apart from pdp11's 2143).  
> 
> That's the case, in include/linux/kconfig.h we have:
> 
> #ifdef CONFIG_CPU_BIG_ENDIAN
> #define __BIG_ENDIAN 4321
> #else
> #define __LITTLE_ENDIAN 1234
> #endif
> 
> But as far as I understand the problem is that math-emu expects 
> __BIG_ENDIAN to be defined at all time as it has tests like:
> 
> #if __BYTE_ORDER == __BIG_ENDIAN

The gcc docs have (https://gcc.gnu.org/onlinedocs/cpp/Common-Predefined-Macros.html):
     
__BYTE_ORDER__
__ORDER_LITTLE_ENDIAN__
__ORDER_BIG_ENDIAN__
__ORDER_PDP_ENDIAN__

    __BYTE_ORDER__ is defined to one of the values __ORDER_LITTLE_ENDIAN__, __ORDER_BIG_ENDIAN__, or __ORDER_PDP_ENDIAN__ to reflect the layout of multi-byte and multi-word quantities in memory. If __BYTE_ORDER__ is equal to __ORDER_LITTLE_ENDIAN__ or __ORDER_BIG_ENDIAN__, then multi-byte and multi-word quantities are laid out identically: the byte (word) at the lowest address is the least significant or most significant byte (word) of the quantity, respectively. If __BYTE_ORDER__ is equal to __ORDER_PDP_ENDIAN__, then bytes in 16-bit words are laid out in a little-endian fashion, whereas the 16-bit subwords of a 32-bit quantity are laid out in big-endian fashion.

    You should use these macros for testing like this:

    /* Test for a little-endian machine */
    #if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__

The doc doesn't mention the value, but __ORDER_BIG_ENDIAN__ is 4321 (decimal).

So the math-emu code is neither following gcc's rules or the kernel ones.

Your change will break anything that currently does:
#ifdef __BIG_ENDIAN

Any change would have to be limited to code that is implementing math-emu.

-- David

> 
> Christophe
> 

Re: [PATCH] powerpc: define __LITTLE_ENDIAN and __BIG_ENDIAN for math-emu

[Christophe Leroy (CS GROUP)]

Le 20/05/2026 à 20:43, David Laight a écrit :
> On Wed, 20 May 2026 15:14:40 +0200
> "Christophe Leroy (CS GROUP)" <chleroy@kernel.org> wrote:
> 
>> Le 17/05/2026 à 15:54, David Laight a écrit :
>>> On Sun, 17 May 2026 12:14:21 +0800
>>> Mingcong Bai <jeffbai@aosc.io> wrote:
>>>    
>>>> Similar to commit b929926f01f2 ("sh: define __BIG_ENDIAN for math-emu"),
>>>> define __LITTLE_ENDIAN and __BIG_ENDIAN as 0 to mitigate build-time
>>>> warnings:
>>>>
>>>>     ./include/math-emu/double.h:59:21: error: ‘__BIG_ENDIAN’ is not defined, evaluates to ‘0’ [-Werror=undef]
>>>>        59 | #if __BYTE_ORDER == __BIG_ENDIAN
>>>>           |
>>>>
>>>> Cc: stable@vger.kernel.org
>>>> Fixes: 13da9e200fe4 ("Revert "endian: #define __BYTE_ORDER"")
>>>> Reported-by: kernel test robot <lkp@intel.com>
>>>> Closes: https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore.kernel.org%2Foe-kbuild-all%2F202507301656.7FEX6J5W-lkp%40intel.com%2F&data=05%7C02%7Cchristophe.leroy2%40cs-soprasteria.com%7C3ed26b8c3d6449fdc29608deb69fac66%7C8b87af7d86474dc78df45f69a2011bb5%7C0%7C0%7C639148994069314641%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=FfFnXxMPXXxoOMM8fYU4df5gMjk3B2dPgQsjwUagaNA%3D&reserved=0
>>>> Signed-off-by: Mingcong Bai <jeffbai@aosc.io>
>>>> ---
>>>>    arch/powerpc/include/asm/sfp-machine.h | 4 +++-
>>>>    1 file changed, 3 insertions(+), 1 deletion(-)
>>>>
>>>> diff --git a/arch/powerpc/include/asm/sfp-machine.h b/arch/powerpc/include/asm/sfp-machine.h
>>>> index 8b957aabb826d..db8525605c026 100644
>>>> --- a/arch/powerpc/include/asm/sfp-machine.h
>>>> +++ b/arch/powerpc/include/asm/sfp-machine.h
>>>> @@ -319,10 +319,12 @@
>>>>    #define abort()								\
>>>>    	return 0
>>>>    
>>>> -#ifdef __BIG_ENDIAN
>>>> +#ifdef __BIG_ENDIAN__
>>>>    #define __BYTE_ORDER __BIG_ENDIAN
>>>> +#define __LITTLE_ENDIAN 0
>>>>    #else
>>>>    #define __BYTE_ORDER __LITTLE_ENDIAN
>>>> +#define __BIG_ENDIAN 0
>>>>    #endif
>>>
>>> I thought the expected/correct value for __BYTE_ORDER__ was either 1234 or 4321.
>>> (apart from pdp11's 2143).
>>
>> That's the case, in include/linux/kconfig.h we have:
>>
>> #ifdef CONFIG_CPU_BIG_ENDIAN
>> #define __BIG_ENDIAN 4321
>> #else
>> #define __LITTLE_ENDIAN 1234
>> #endif
>>
>> But as far as I understand the problem is that math-emu expects
>> __BIG_ENDIAN to be defined at all time as it has tests like:
>>
>> #if __BYTE_ORDER == __BIG_ENDIAN
> 
> The gcc docs have (https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgcc.gnu.org%2Fonlinedocs%2Fcpp%2FCommon-Predefined-Macros.html&data=05%7C02%7Cchristophe.leroy2%40cs-soprasteria.com%7C3ed26b8c3d6449fdc29608deb69fac66%7C8b87af7d86474dc78df45f69a2011bb5%7C0%7C0%7C639148994069350793%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=D0BlZT73XnqHXHN2ukPFFUQw5lCCwaKfkmp6vMHz0Gk%3D&reserved=0):
>       
> __BYTE_ORDER__
> __ORDER_LITTLE_ENDIAN__
> __ORDER_BIG_ENDIAN__
> __ORDER_PDP_ENDIAN__
> 
>      __BYTE_ORDER__ is defined to one of the values __ORDER_LITTLE_ENDIAN__, __ORDER_BIG_ENDIAN__, or __ORDER_PDP_ENDIAN__ to reflect the layout of multi-byte and multi-word quantities in memory. If __BYTE_ORDER__ is equal to __ORDER_LITTLE_ENDIAN__ or __ORDER_BIG_ENDIAN__, then multi-byte and multi-word quantities are laid out identically: the byte (word) at the lowest address is the least significant or most significant byte (word) of the quantity, respectively. If __BYTE_ORDER__ is equal to __ORDER_PDP_ENDIAN__, then bytes in 16-bit words are laid out in a little-endian fashion, whereas the 16-bit subwords of a 32-bit quantity are laid out in big-endian fashion.
> 
>      You should use these macros for testing like this:
> 
>      /* Test for a little-endian machine */
>      #if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
> 
> The doc doesn't mention the value, but __ORDER_BIG_ENDIAN__ is 4321 (decimal).
> 
> So the math-emu code is neither following gcc's rules or the kernel ones.
> 
> Your change will break anything that currently does:
> #ifdef __BIG_ENDIAN
> 
> Any change would have to be limited to code that is implementing math-emu.

asm/sfp-machine.h is only included by math-emu it seems, so the change 
should be safe.

Apparently math-emu predates git history, not sure where it comes from.

Christophe