Re: make check error in native compiler resulting from commit 1cd188799f86bcb13ad76e82e3436b1b7e9f9e9f on 2021-12-30.

[Alan Mackenzie]

Hello, Andrea.

On Fri, Feb 04, 2022 at 14:58:47 +0000, Andrea Corallo wrote:
> Andrea Corallo <akrl@sdf.org> writes:

> > Alan Mackenzie <acm@muc.de> writes:
> >> In that patch, I made the following change so that a comparison
> >> involving Qnil would not go through the (newly) expensive emit_EQ:

> >> diff --git a/lisp/emacs-lisp/comp.el b/lisp/emacs-lisp/comp.el
> >> index 0a10505257..8581fe8066 100644
> >> --- a/lisp/emacs-lisp/comp.el
> >> +++ b/lisp/emacs-lisp/comp.el
> >> @@ -1829,9 +1829,7 @@ comp-limplify-lap-inst
> >>        (byte-listp auto)
> >>        (byte-eq auto)
> >>        (byte-memq auto)
> >> -      (byte-not
> >> -       (comp-emit-set-call (comp-call 'eq (comp-slot-n (comp-sp))
> >> -                                      (make-comp-mvar :constant nil))))
> >> +      (byte-not null)
> >>        (byte-car auto)
> >>        (byte-cdr auto)
> >>        (byte-cons auto)

> Yeah, I think reverting this hunk and keeping the `eq' explicit in
> LIMPLE is the right thing to do, it solves this specific issue and we
> get back null and not inlined for free.

> The explicit `eq' gets already nicely handled by the logic you have
> added in comp.c:2244 so we emit libgccjit IR using emit_BASE_EQ in this
> case.

> I've pushed dcf30f14f9 to do that as it works here.

OK, thank you!

> Thanks

>   Andrea

-- 
Alan Mackenzie (Nuremberg, Germany).