GNU bug report logs - #41077
[feature/native-comp] Segfaults when compiling ELC+ELN

Previous Next

Full log

Message #37 received at 41077-done <at> debbugs.gnu.org (full text, mbox):

From: Andrea Corallo <akrl <at> sdf.org>
To: Kévin Le Gouguec <kevin.legouguec <at> gmail.com>
Cc: 41077-done <at> debbugs.gnu.org
Subject: Re: bug#41077: [feature/native-comp] virtual memory exhausted
Date: Sun, 10 May 2020 22:17:56 +0000

Kévin Le Gouguec <kevin.legouguec <at> gmail.com> writes:

> Andrea Corallo <akrl <at> sdf.org> writes:
>
>>> As soon as I've figured out how to use the elisp-benchmarks package,
>>> I'll post some figures; is there a specific place (bug number,
>>> emacs-devel thread) where you usually collect such feedback?
>>
>> Cool, we have no bug number for that.  I guess you can use the "On elisp
>> running native" thread or as you feel.
>
> Mmm, I've installed the elisp-benchmarks package, and I've run
>
>     $SOURCEDIR/src/emacs -Q -batch -l ./elisp-benchmarks.el -f elisp-benchmarks-run
>
> from ~/.emacs.d/elpa/elisp-benchmarks-1.1, once with SOURCEDIR pointing
> to the master worktree and once again with SOURCEDIR pointing to the
> feature/native-comp worktree, and I get more or less the exact same
> performance[1].
>
> I've run the feature/native-comp emacs under strace and AFAICT the
> native libraries get loaded[2]; I suspect I'm making some very basic
> mistake but I can't put my finger on it.  Is there anything else I
> should be looking for to assert that things work as intended?
>
> Thank you for your patience.

Hi Kévin,

No issue, I think you need to hack a little bit elisp-benchmarks because
by default the package manager just byte compile the files during
installation and elisp-benchmarks does not support officialy this
feature branch.

That would translate into somenthing like changing the call to
`byte-compile-file' into `native-compile' and running it with recompile
t.

Other option maybe quicker: you can manually native compile the
benchmarcks and the native emacs instance should load them.

You will see that when a .eln is loaded the message is a little
different to indicate that.

Probably what you have observed tracing was the activity of all the
other .elns I guess.

Best

  Andrea

-- 
akrl <at> sdf.org

Previous Next

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