[PATCH] aarch64: fix build with old glibc

Jan Beulich jbeulich@suse.com
Fri Jun 28 12:08:21 GMT 2024


On 28.06.2024 13:58, Richard Earnshaw (lists) wrote:
> On 28/06/2024 12:51, Jan Beulich wrote:
>> On 28.06.2024 13:29, Richard Earnshaw (lists) wrote:
>>> On 28/06/2024 08:07, Jan Beulich wrote:
>>>> As was pointed out several times before, old glibc declares index(),
>>>> resulting in warnings from -Wshadow, in turn failing the build due to
>>>> -Werror.
>>>> ---
>>>> Is the intermediate local variable needed here at all? Can't we pass the
>>>> address of operand->imm.value into parse_index_expression()?
>>>>
>>>> --- a/gas/config/tc-aarch64.c
>>>> +++ b/gas/config/tc-aarch64.c
>>>> @@ -3943,10 +3943,10 @@ parse_shifter_zt0_with_bit_index (char *
>>>>        return true;
>>>>      }
>>>>  
>>>> -  int64_t index;
>>>> -  if (!parse_index_expression (str, &index))
>>>> +  int64_t idx;
>>>> +  if (!parse_index_expression (str, &idx))
>>>>        return false;
>>>> -  operand->imm.value = index;
>>>> +  operand->imm.value = idx;
>>>>  
>>>>    if (!skip_past_comma (str))
>>>>        return true;
>>>
>>> OK, but I think this counts as obvious :)
>>
>> Thanks, and yes, I wouldn't have waited for too long before putting this in.
>> Would be nice though to have an answer to the question raised, as that might
>> then (or might not) call for a further simplified v2.
> 
> SOrry I missed that bit :)
> 
> I'm afraid I don't immediately know the answer to that as I wasn't the original author of that code.

Let me direct the question there then: Saurabh?

Jan


More information about the Binutils mailing list