Home Original page

[PATCH 2/3] gas/gen-sframe: avoid gcc extension using __VA_ARGS__

Jan Beulich jbeulich@suse.com
Mon Dec 8 07:27:53 GMT 2025 
On 05.12.2025 16:05, Jens Remus wrote:
> On 12/5/2025 2:32 PM, Jan Beulich wrote:
>> We shouldn't be using extensions when we don't have a suitable fallback in
>> place. Introducing a format-argument-less counterpart macro would feel a
>> little odd here. Instead make the sole use site have a (fake) argument
>> (the non-translatable part of the string).
>>
>> --- a/gas/dw2gencfi.c
>> +++ b/gas/dw2gencfi.c
>> @@ -2622,7 +2622,7 @@ cfi_finish (void)
>>  #endif
>>  	/* Avoid erroring with DEFAULT_SFRAME for non-default options, like
>>  	   -32 on x86_64.  */
>> -	sframe_as_bad (_(".sframe not supported for target"));
>> +	sframe_as_bad (_("%s not supported for target"), ".sframe");
> 
> Couldn't this change be omitted, when changing the definition of
> sframe_as_bad as follows?
> 
>>      }
>>  
>>    if ((all_cfi_sections & CFI_EMIT_debug_frame) != 0)
>> --- a/gas/gen-sframe.h
>> +++ b/gas/gen-sframe.h
>> @@ -28,7 +28,7 @@
>>  #define sframe_as_bad(format, ...) \
> 
> #define sframe_as_bad(...) \
> 
>>    do {					       \
>>      if (flag_gen_sframe == GEN_SFRAME_ENABLED) \
>> -      as_bad (format, ##__VA_ARGS__);	       \
>> +      as_bad (format, __VA_ARGS__);            \
> 
>       as_bad (__VA_ARGS__); \
> 
>>    } while (0)
>>  
>>  #define SFRAME_FRE_ELEM_LOC_REG		0

Yes, but see what you and I think about the similar aspect in objcopy.c. One
of the problems with this form is that the macro then also permits for no
arguments at all. Sure, the compiler will flag a missing argument to as_bad()
then, but still.

Jan
More information about the Binutils mailing list