GNU bug report logs - #46670
28.0.50; [feature/native-comp] possible miscompilation affecting lsp-mode

Previous Next

Full log

Message #79 received at 46670 <at> debbugs.gnu.org (full text, mbox):

From: Pip Cet <pipcet <at> gmail.com>
To: Andrea Corallo <akrl <at> sdf.org>
Cc: 46670 <at> debbugs.gnu.org, Mauricio Collares <mauricio <at> collares.org>
Subject: Re: bug#46670: 28.0.50; [feature/native-comp] possible miscompilation
 affecting lsp-mode
Date: Wed, 24 Feb 2021 09:28:39 +0000

[Message part 1 (text/plain, inline)]

On Wed, Feb 24, 2021 at 9:04 AM Andrea Corallo <akrl <at> sdf.org> wrote:
> Pip Cet <pipcet <at> gmail.com> writes:
>
> > On Tue, Feb 23, 2021 at 11:36 PM Andrea Corallo <akrl <at> sdf.org> wrote:
> >> Pip Cet <pipcet <at> gmail.com> writes:
> >> > Is this one of them, or am I confused?
> >>
> >> What's suspitions with that?  At present I'm admittedly quite done but
> >> it looks okay to me.
> >
> > We're emitting
> >
> > (assume ,lhs (and ,lhs ,rhs))
> >
> > even when NEGATED is t.
>
> Nope, when NEGATED is t the complete sequence we are emitting is (see
> line just following your diff hunk):
>
> (assume tmp-mvar (not rhs))

But tmp-mvar is in the same slot as RHS.

> (assume lhs (and lhs tmp-mvar))

So this is equivalent (after the next SSA rename) to

(assume lhs (and lhs rhs))

> That's indeed the reason why it's working in the 39 testcases.

No, the reason it's "working" is that we never assert our assumes.
I've got a patch here that does just that :-)

[0001-Assert-at-runtime-that-assume-s-assumptions-actually.patch (text/x-patch, attachment)]

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.