Re: feature/native-comp 5bc0855 2/2: Don't treat '=' as simple equality emitting constraints (bug#46812)

[Andrea Corallo]

Andrea Corallo via "Emacs development discussions."
<emacs-devel@gnu.org> writes:

> Pip Cet <pipcet@gmail.com> writes:
>
>> On Tue, Mar 2, 2021 at 5:14 PM Pip Cet <pipcet@gmail.com> wrote:
>>> Don't forget this one either (currently miscompiled to have subr-type
>>> ((t) nil)):
>>>
>>> (lambda (x)
>>>   (unless (= x 0.0) (error ""))
>>>   (unless (eql x -0.0) (error ""))
>>>   x)
>>
>> That one's still broken.
>>
>> (lambda (x) (if (eql x 1.0) (error "")) (unless (floatp x) (error "")) x)
>>
>> is also mis-analyzed to have subr-type (function (t) nil). However,
>> the compiled code appears to work correctly in both cases. Do you care
>> about such bugs?
>
> Indeed I care, but I'm not a full time Emacs developer, I just had no
> time to even look into this as ATM I'm busy on other fronts (including
> what pays my bills).  Will come back on this as soon as I can.

The two reported miscompilations should be fixed now as of c60f2f458a,
please have a look.

Still have to look into the missed optimization for the null returning
function, will do.

Thanks

  Andrea