emacs-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Emacs port to gcc -fcheck-pointer-bounds


From: Eli Zaretskii
Subject: Re: Emacs port to gcc -fcheck-pointer-bounds
Date: Mon, 11 Dec 2017 17:26:41 +0200

> Cc: address@hidden, address@hidden
> From: Paul Eggert <address@hidden>
> Date: Sun, 10 Dec 2017 23:54:59 -0800
> 
> Eli Zaretskii wrote:
> > But if we install this patch regardless of -fcheck-pointer-bounds,
> > then the --with-wide-int build will cease working, won't it?
> 
> No, --with-wide-int still works fine, because it causes 
> LISP_WORDS_ARE_POINTERS 
> to be false, so src/lisp.h falls back on using EMACS_INT for the basic Lisp 
> type, just as before.

So we will have two different data types implementing a Lisp object,
one --with-wide-int and another without it?  That sounds suboptimal to
me.

Why not reserve the new representation to the -fcheck-pointer-bounds
builds only?  Such builds will anyway be used only by Emacs developers
(not even by all of them), and for those who do a different
representation won't be a significant obstacle.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]