Re: bug in simplify_shallow

On 2023-08-27 17:01, Ruud H.G. van Tol wrote:
> On 2023-08-27 16:13, Karim Belabas wrote:
>> * Ruud H.G. van Tol [2023-08-27 15:42]:
>>> On 2023-08-27 14:58, Bill Allombert wrote:
>>>> [...]to be honest I did not remember that listput returned x and it is
>>>> not explicitly documented
>>> ? ??listput
>>> listput( list,x,{n}):
>>> [...] The function returns the inserted element.
>> What "the inserted element" means is ambiguous in the context of PARI's
>> memory model. What the implementation does is return a direct pointer to
>> the new list entry, that inserted element being a clone of the input x.
>>
>> What Bill and I suggested was to return "a copy of the inserted element"
>> instead. I don't really see a scenario where this would break
>> compatibility ... except this would make insertion about twice slower,
>> even in cases where the returned value is ignored.
>
> Yes, so I wondered if the "void calling context" is decidable at 
> compile-time,
> such that any new overhead can be avoided where feasible.

Alternatively, postpone the free's until end-of-scope.

-- Ruud