From: pkk@spth.de

Am 30.11.25 um 12:28 schrieb David Brown:
>>
>> I see two implementation strategies:
>>
>> * Just ignore the values of the padding bits. You don't need to and or
>> anything after arithmetic operations. Makes arithmetic as fast as
>> possible. But you need special handling at comparisons and casts.
>>
>> * Always keep the padding bits in line with the value, i.e. and after
>> arithemetic operations for unsigned, copy value of sign bit for
>> signed. Extra effort at arithmetic operations, but no extra effort at
>> casts and comparisons.
>>
>
> That sounds about right.  It's much the same as the implementation of
> _Bool.  You either ignore the padding bits while doing the calculations
> and filter them out when they later get in the way, or you keep them
> neat and consistent (signed or unsigned extended, as appropriate) during
> calculations and it's all fine for other operations.  I have no idea
> what might be the most efficient choice overall - it could vary by
> application, but I expect implementations to have one fixed strategy.
>

_Bool is a bit different, since it promotes to int, so we don't really
have arithemetic directly on _Bool: I can definitely see an
implementation going one way for _BitInt, and the other for _Bool.

--- SoupGate-Win32 v1.05
 * Origin: you cannot sedate... all the things you hate (1:229/2)