GNU bug report logs - #50934
28.0.60; paren.el should be preloaded

Previous Next

Full log

View this message in rfc822 format

From: Andrea Corallo <akrl <at> sdf.org>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 50934 <at> debbugs.gnu.org
Subject: bug#50934: 28.0.60; paren.el should be preloaded
Date: Fri, 01 Oct 2021 13:35:57 +0000

Eli Zaretskii <eliz <at> gnu.org> writes:

>> From: Andrea Corallo <akrl <at> sdf.org>
>> Cc: 50934 <at> debbugs.gnu.org
>> Date: Fri, 01 Oct 2021 13:11:33 +0000
>> 
>> Eli Zaretskii <eliz <at> gnu.org> writes:
>> 
>> >> Date: Fri, 01 Oct 2021 13:49:31 +0300
>> >> From: Eli Zaretskii <eliz <at> gnu.org>
>> >> 
>> >> Since show-paren-mode is now ON by default, whenever Emacs starts it
>> >> now loads paren.elc? during startup.  So I think we should now preload
>> >> paren.el.
>> >
>> > One consequence of paren.el's not being preloaded is that in a
>> > native-compilation build, Emacs begins compiling paren.el as soon as
>> > it starts, and that causes it also to compile about a dozen other
>> > files that comp.el requires.  Should we perhaps native-compile all the
>> > prerequisites of comp.el as part of COMPILE_FIRST when we build Emacs?
>> 
>> I think we should keep an eye on the build time impact of this.  If I'm
>> not mistaken native compiling COMPILE_FIRST targets is slower than
>> afterward cause comp.el and friends are running interpreted.
>
> The native-compilation build is already painfully slow, and most of it
> isn't spent in COMPILE_FIRST.  We could, of course, add those files to
> shortlisp instead, if you think this is better.

I suspect shortlisp might be a better option (the number of threads of
the machine here comes into play as well), but I guess a measure is the
only way to get an answer.  IOW just wanted to raise the warning.

Regards

  Andrea

Previous Next

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