bug#46843: closed (28.0.50; [native-comp] (lambda (x) (if (equal x b) (eq x b) (error ""))) miscompiled)

[GNU bug Tracking System]

Your message dated Mon, 01 Mar 2021 17:26:29 +0000
with message-id <xjfh7lubxa2.fsf@sdf.org>
and subject line Re: bug#46843: 28.0.50; [native-comp] (lambda (x) (if (equal x 
b) (eq x b) (error ""))) miscompiled
has caused the debbugs.gnu.org bug report #46843,
regarding 28.0.50; [native-comp] (lambda (x) (if (equal x b) (eq x b) (error 
""))) miscompiled
to be marked as done.

(If you believe you have received this mail in error, please contact
help-debbugs@gnu.org.)


-- 
46843: http://debbugs.gnu.org/cgi/bugreport.cgi?bug=46843
GNU Bug Tracking System
Contact help-debbugs@gnu.org with problems
> --- Begin Message ---
>
>
>
> Subject: 
>
> 28.0.50; [native-comp] (lambda (x) (if (equal x b) (eq x b) (error ""))) miscompiled
>
>
>
>
> Date: 
>
> Mon, 1 Mar 2021 06:51:30 +0000
>
>
>
>
> In *scratch*, evaluate
>
> (funcall
>  (let* ((lexical-binding t)
>         (a (list 'a))
>         (b (cons a a)))
>    (native-compile
>     `(lambda (x) (if (equal x ',b) (eq x ',b) (error "")))))
>  '((a) a))
>
> Expected result:
>
> nil
>
> Actual result:
>
> t
>
>
>
> --- End Message ---
> --- Begin Message ---
>
>
>
> Subject: 
>
> Re: bug#46843: 28.0.50; [native-comp] (lambda (x) (if (equal x b) (eq x b) (error ""))) miscompiled
>
>
>
>
> Date: 
>
> Mon, 01 Mar 2021 17:26:29 +0000
>
>
>
>
> User-agent: 
>
> Gnus/5.13 (Gnus v5.13) Emacs/28.0.50 (gnu/linux)
>
>
>
>
> Andrea Corallo via "Bug reports for GNU Emacs, the Swiss army knife of
> text editors" <bug-gnu-emacs@gnu.org> writes:
>
> > Pip Cet <pipcet@gmail.com> writes:
> >
> >> In *scratch*, evaluate
> >>
> >> (funcall
> >>  (let* ((lexical-binding t)
> >>         (a (list 'a))
> >>         (b (cons a a)))
> >>    (native-compile
> >>     `(lambda (x) (if (equal x ',b) (eq x ',b) (error "")))))
> >>  '((a) a))
> >>
> >> Expected result:
> >>
> >> nil
> >>
> >> Actual result:
> >>
> >> t
> >
> > Thanks for reporting this, will come-up with a fix.
>
> Right 3d014e1bf4 is fixing the wrong assumtion that `equal' `eql' can be
> treated as `eq' in propagation as every immediate in the compilation
> unit is hash consed.  Indeed this assumtion is wrong for more than one
> reason.
>
> Now under `equal' `eql' we propagate only fixnums and symbols values,
> all the other kind of values are promoted to their respective types and
> propagated as such.
>
> The patch adds some tests for that too.
>
> As the two reproducer are working here I'm closing this.
>
> Thanks for reporting.
>
>   Andrea
>
>
> --- End Message ---