GNU bug report logs - #21066
Proceeding Order in AM generated Makefile

Previous Next

To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 21066 in the body.
You can then email your comments to 21066 AT debbugs.gnu.org in the normal way.

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

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

From: Dilyan Palauzov <dilyan.palauzov <at> aegee.org>
To: automake-bug <at> gnu.org
Subject: Proceeding Order in AM generated Makefile
Date: Wed, 15 Jul 2015 17:10:40 +0200

Hello,

please read

http://lists.gnu.org/archive/html/bug-make/2015-07/msg00054.html 
http://lists.gnu.org/archive/html/bug-make/2015-07/msg00055.html

and tell me, if this is make or Automake problem.

If it is Automake problem, were you aware of it and do you have an idea 
how to solve it?

The concrete case is building Cyrus Imapd, which uses a single Makefile 
for every .c file, library, executable (except the perl stuff).

Thanks
  Dilian

Message #8 received at 21066 <at> debbugs.gnu.org (full text, mbox):

From: Dilyan Palauzov <dilyan.palauzov <at> aegee.org>
To: 21066 <at> debbugs.gnu.org
Subject: I use Automake 1.15
Date: Wed, 15 Jul 2015 17:16:25 +0200

I use Automake 1.15.

Message #11 received at 21066 <at> debbugs.gnu.org (full text, mbox):

From: Hans-Bernhard Bröker <HBBroeker <at> t-online.de>
To: Dilyan Palauzov <dilyan.palauzov <at> aegee.org>, 21066 <at> debbugs.gnu.org
Subject: Re: bug#21066: Proceeding Order in AM generated Makefile
Date: Wed, 15 Jul 2015 22:03:06 +0200

Am 15.07.2015 um 17:10 schrieb Dilyan Palauzov:
> http://lists.gnu.org/archive/html/bug-make/2015-07/msg00054.html
> http://lists.gnu.org/archive/html/bug-make/2015-07/msg00055.html
>
> and tell me, if this is make or Automake problem.

You've been told in the answer you already got that this is not, 
actually, a "problem".  You opted not to discuss that answer any 
further, and just to echo the question elsewhere instead.

It's not make's job, much less automake's, to guess how long individual 
production rules might take.  The tree of dependencies is walked, that's it.

Message #14 received at 21066 <at> debbugs.gnu.org (full text, mbox):

From: Mathieu Lirzin <mthl <at> gnu.org>
To: Hans-Bernhard Bröker <HBBroeker <at> t-online.de>
Cc: 21066 <at> debbugs.gnu.org, Dilyan Palauzov <dilyan.palauzov <at> aegee.org>
Subject: Re: bug#21066: Proceeding Order in AM generated Makefile
Date: Sun, 16 Jul 2017 15:18:40 +0200

Hello,

Hans-Bernhard Bröker <HBBroeker <at> t-online.de> writes:

> Am 15.07.2015 um 17:10 schrieb Dilyan Palauzov:
>> http://lists.gnu.org/archive/html/bug-make/2015-07/msg00054.html
>> http://lists.gnu.org/archive/html/bug-make/2015-07/msg00055.html
>>
>> and tell me, if this is make or Automake problem.
>
> You've been told in the answer you already got that this is not,
> actually, a "problem".  You opted not to discuss that answer any
> further, and just to echo the question elsewhere instead.
>
> It's not make's job, much less automake's, to guess how long
> individual production rules might take.  The tree of dependencies is
> walked, that's it.

IIUC The observed problem is some lack of parallelism when running 'make
-jX'.  It is likely that Make derives the dependency graph between
targets solely from their prerequisites.  So any lack of parallelism
seems related to some unnecessary prerequisites added in the provided
rules.

I am closing this bug since this seems like a bug in the Makefile you
have written not in Automake.  Feel free to reopen it if you can provide
a minimal example demonstrating that the Make rules generated by
Automake could have more parallelism.

Thanks.

-- 
Mathieu Lirzin
GPG: F2A3 8D7E EB2B 6640 5761  070D 0ADE E100 9460 4D37

Previous Next

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