[tree-ssa] Computed gotos

[Zack Weinberg]

Andreas Jaeger <aj@suse.de> writes:

>>> Sorry, I probably meant "yes", as in one of these two is a suitable
>>> definition of uintptr_t on all GCC targets AFAIK:
>>>
>>> 	typedef unsigned long uintptr_t;
>>> 	typedef unsigned long long uintptr_t;
>>
>> If the first one isn't a suitable definition of uintptr_t for some
>> target, the ABI is broken.  (I know C99 allows
>> sizeof(long) < sizeof(void *). C99 is wrong.)
>
> 64-bit MS Windows uses AFAIK a P64 model where the long is 32-bit and
> pointers are 64-bit - I agree the ABI is broken indeed,

That does unfortunately mean we'll have to care eventually, but I vote
for not caring until someone does a 64-bit Windows port.

zw