GNU bug report logs - #80065
[PATCH 1/1] Have default 'compile-command' run make in parallel

Previous Next

To reply to this bug, email your comments to 80065 AT debbugs.gnu.org.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Philip Kaludercic <philipk <at> posteo.net>
To: bug-gnu-emacs <at> gnu.org
Subject: [PATCH 1/1] Have default 'compile-command' run make in parallel
Date: Tue, 23 Dec 2025 18:25:20 +0000

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

I have been using this in my personal configuration for a few years now,
and I think it might be nice to consider this upstream as well.  To my
knowledge, all real-world "make" implementations support the "-j" flag.

[0001-Have-default-'compile-command'-run-make-in-paralle.patch (text/x-patch, attachment)]

Message #8 received at 80065 <at> debbugs.gnu.org (full text, mbox):

From: Stéphane Marks <shipmints <at> gmail.com>
To: Philip Kaludercic <philipk <at> posteo.net>
Cc: 80065 <at> debbugs.gnu.org
Subject: Re: bug#80065: [PATCH 1/1] Have default 'compile-command' run make in
 parallel
Date: Tue, 23 Dec 2025 13:33:55 -0500

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

To be nice to unsuspecting users, this might default to (/ (num-processors)
2) or -2.  I have 16 cores and limit to 10 or 12 usually.

On Tue, Dec 23, 2025 at 1:26 PM Philip Kaludercic <philipk <at> posteo.net>
wrote:

>
> I have been using this in my personal configuration for a few years now,
> and I think it might be nice to consider this upstream as well.  To my
> knowledge, all real-world "make" implementations support the "-j" flag.
>
>

[Message part 2 (text/html, inline)]

Message #11 received at 80065 <at> debbugs.gnu.org (full text, mbox):

From: Philip Kaludercic <philipk <at> posteo.net>
To: Stéphane Marks <shipmints <at> gmail.com>
Cc: 80065 <at> debbugs.gnu.org
Subject: Re: bug#80065: [PATCH 1/1] Have default 'compile-command' run make
 in parallel
Date: Tue, 23 Dec 2025 19:00:33 +0000

Stéphane Marks <shipmints <at> gmail.com> writes:

> To be nice to unsuspecting users, this might default to (/ (num-processors)
> 2) or -2.  I have 16 cores and limit to 10 or 12 usually.

Right, I would suggest something like (ceiling (num-processors) 2) to
avoid edge-cases on single-core and dual-core systems:

(ceiling 1 2) ;=> 1 (#o1, #x1, ?\C-a)
(ceiling 2 2) ;=> 1 (#o1, #x1, ?\C-a)
(ceiling 3 2) ;=> 2 (#o2, #x2, ?\C-b)
(ceiling 4 2) ;=> 2 (#o2, #x2, ?\C-b)
...

> On Tue, Dec 23, 2025 at 1:26 PM Philip Kaludercic <philipk <at> posteo.net>
> wrote:
>
>>
>> I have been using this in my personal configuration for a few years now,
>> and I think it might be nice to consider this upstream as well.  To my
>> knowledge, all real-world "make" implementations support the "-j" flag.
>>
>>

Message #14 received at 80065 <at> debbugs.gnu.org (full text, mbox):

From: Stéphane Marks <shipmints <at> gmail.com>
To: Philip Kaludercic <philipk <at> posteo.net>
Cc: 80065 <at> debbugs.gnu.org
Subject: Re: bug#80065: [PATCH 1/1] Have default 'compile-command' run make in
 parallel
Date: Tue, 23 Dec 2025 14:12:45 -0500

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

On Tue, Dec 23, 2025 at 2:00 PM Philip Kaludercic <philipk <at> posteo.net>
wrote:

> Stéphane Marks <shipmints <at> gmail.com> writes:
>
> > To be nice to unsuspecting users, this might default to (/
> (num-processors)
> > 2) or -2.  I have 16 cores and limit to 10 or 12 usually.
>
> Right, I would suggest something like (ceiling (num-processors) 2) to
> avoid edge-cases on single-core and dual-core systems:
>
> (ceiling 1 2) ;=> 1 (#o1, #x1, ?\C-a)
> (ceiling 2 2) ;=> 1 (#o1, #x1, ?\C-a)
> (ceiling 3 2) ;=> 2 (#o2, #x2, ?\C-b)
> (ceiling 4 2) ;=> 2 (#o2, #x2, ?\C-b)
> ...
>
> > On Tue, Dec 23, 2025 at 1:26 PM Philip Kaludercic <philipk <at> posteo.net>
> > wrote:
> >
> >>
> >> I have been using this in my personal configuration for a few years now,
> >> and I think it might be nice to consider this upstream as well.  To my
> >> knowledge, all real-world "make" implementations support the "-j" flag.
>

Might want to make compile-num-processors a defcustom or defvar so one can
override in their .dir-locals.el for a project or specific directory?  That
could be an integer or a function, I guess.

[Message part 2 (text/html, inline)]

Message #17 received at 80065 <at> debbugs.gnu.org (full text, mbox):

From: Philip Kaludercic <philipk <at> posteo.net>
To: Stéphane Marks <shipmints <at> gmail.com>
Cc: 80065 <at> debbugs.gnu.org
Subject: Re: bug#80065: [PATCH 1/1] Have default 'compile-command' run make
 in parallel
Date: Tue, 23 Dec 2025 20:25:43 +0000

Stéphane Marks <shipmints <at> gmail.com> writes:

> On Tue, Dec 23, 2025 at 2:00 PM Philip Kaludercic <philipk <at> posteo.net>
> wrote:
>
>> Stéphane Marks <shipmints <at> gmail.com> writes:
>>
>> > To be nice to unsuspecting users, this might default to (/
>> (num-processors)
>> > 2) or -2.  I have 16 cores and limit to 10 or 12 usually.
>>
>> Right, I would suggest something like (ceiling (num-processors) 2) to
>> avoid edge-cases on single-core and dual-core systems:
>>
>> (ceiling 1 2) ;=> 1 (#o1, #x1, ?\C-a)
>> (ceiling 2 2) ;=> 1 (#o1, #x1, ?\C-a)
>> (ceiling 3 2) ;=> 2 (#o2, #x2, ?\C-b)
>> (ceiling 4 2) ;=> 2 (#o2, #x2, ?\C-b)
>> ...
>>
>> > On Tue, Dec 23, 2025 at 1:26 PM Philip Kaludercic <philipk <at> posteo.net>
>> > wrote:
>> >
>> >>
>> >> I have been using this in my personal configuration for a few years now,
>> >> and I think it might be nice to consider this upstream as well.  To my
>> >> knowledge, all real-world "make" implementations support the "-j" flag.
>>
>
> Might want to make compile-num-processors a defcustom or defvar so one can
> override in their .dir-locals.el for a project or specific directory?  That
> could be an integer or a function, I guess.

I figure that the complications of having the default value of one user
option depend on that of another is not worth the fuss, especially when
we consider that `compile-command' is already local safe.

Previous Next

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