X-Authentication-Warning: delorie.com: mail set sender to geda-user-bounces using -f
Message-Id: <202102261453.11QErS3Z003361@delorie.com>
X-Recipient: geda-user AT delorie DOT com
X-Original-DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=crcomp.net;
	 s=default; h=References:Subject:From:To:Date:Sender:Reply-To:Message-ID:Cc:
	MIME-Version:Content-Type:Content-Transfer-Encoding:Content-ID:
	Content-Description:Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc
	:Resent-Message-ID:In-Reply-To:List-Id:List-Help:List-Unsubscribe:
	List-Subscribe:List-Post:List-Owner:List-Archive;
	bh=sro4XnQjFIw3PAxNJRdbPisqpWQXS/fbdHiPQv95NjY=; b=ZvDLeEQAsdNFxoIMhX0hx5kcyG
	P/JEXd8w/LJHtK8+qSDVChfq1/RxVPQjBYKvPgl3yrXoS73415Wx+gkj+Wxz1KPwJPOO3TmzYpbVw
	DMyIl32ZqPLCs7HYPgP1mSV9i3k05F7xp64Ib165tq3ZzNrp5Kq7jwD3FaDs8aTQK8khcfaFK56Gk
	hXFoj9PYLx9zZmtcVczqfQtQwCfklFKly7BQILO+HDdYXiwOJd+mxKdx0NDwb6T/6ZgOUD5TK2Oug
	c+u8+0y+Diiof49zC4s1VVcaG74oxE1gvqh2h9t23GUyuFSfq+zBUYAqIGnRixUukyxkP4VXuhIYW
	7hG7dWgQ==;
Date: 26 Feb 2021 14:53:26 UTC
To: <geda-user AT delorie DOT com>
From: "Don Kuenz (mail AT crcomp DOT net) [via geda-user AT delorie DOT com]" <geda-user AT delorie DOT com>
Subject: Re: [geda-user] FreeBSD 12.2 "make all-recursive" error
References: <41F3B7BE-7014-49F0-9A5A-5641DFFDE29D AT noqsi DOT com>
X-AntiAbuse: This header was added to track abuse, please include it with any abuse report
X-AntiAbuse: Primary Hostname - www18.qth.com
X-AntiAbuse: Original Domain - delorie.com
X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12]
X-AntiAbuse: Sender Address Domain - crcomp.net
X-Get-Message-Sender-Via: www18.qth.com: authenticated_id: mail AT crcomp DOT net
X-Authenticated-Sender: www18.qth.com: mail AT crcomp DOT net
X-Source: 
X-Source-Args: 
X-Source-Dir: 
Reply-To: geda-user AT delorie DOT com

John Doty wrote:
> Roland Lutz wrote:
>> Don Kuenz:
>>> You are correct, $< apparently doesn't work with FreeBSD 'make.'
>>> FreeBSD now builds gschem error-free when gschem/src/Makefile.am
>>> contains the lines shown below. Does the one, remaining $< near
>>> the top also need a substitution?
>>
>> That's a good question.  The build process uses $< in a number of
>> places; I wonder why it's a problem here and not everywhere else.
>> Maybe $< only works in suffix rules with FreeBSD `make'?
>>
>> If so, the $< in the `.c.x' rule should be fine.
>
> $< is classic: I used it four decades ago with AT&T Edition 7 UNIX.
> It’s part of the POSIX spec. It should work in BSD.

This lends weight to my hunch of a race condition during make. It's 
entirely plausible for a FreeBSD or Samba bug to create my proposed race 
condition.

Originally my makes took place on a client against source hosted on a 
samba server share. Once in a blue moon this samba share build process 
actually created a binary (with incorrect libguile linkage). But mostly 
the samba build stopped at random places.

Copying the source to the client creates consistent results. The build 
always stops at the same place.

Slower, more random read/write times over a networked samba share may 
occasionally exacerbate a race condition enough to manifest itself as a
binary built with the three $< in gschem/src/Makefile.am left in place
"as is." 

Danke,

-- 
Don, KB7RPU, https://www.qsl.net/kb7rpu
There was a young lady named Bright Whose speed was far faster than light;
She set out one day In a relative way And returned on the previous night.