Re: bin/60016 (make(1) tests failing since sjg MaybeSubMake changes)

["Robert Elz via gnats" <gnats-admin%NetBSD.org@localhost>]

The following reply was made to PR bin/60016; it has been noted by GNATS.

From: Robert Elz <kre%munnari.OZ.AU@localhost>
To: "Simon J. Gerraty" <sjg%juniper.net@localhost>
Cc: Martin Husemann <martin%duskware.de@localhost>, gnats-bugs%netbsd.org@localhost,
        netbsd-bugs%netbsd.org@localhost, campbell+netbsd%mumble.net@localhost, rillig%netbsd.org@localhost
Subject: Re: bin/60016 (make(1) tests failing since sjg MaybeSubMake changes)
Date: Tue, 10 Mar 2026 06:48:09 +0700

     Date:        Mon, 9 Mar 2026 14:39:02 -0700
     From:        "Simon J. Gerraty" <sjg%juniper.net@localhost>
     Message-ID:  <34773.1773092342%kaos.jnpr.net@localhost>
 
   | That test failed because .make was missing.
 
 Are you sure?
 
 When make does:
 
 	.make:
 		 @ln -sf ${MAKE} ${.TARGET}
 
 (or the newer version with ${_make}) what is the value of ${MAKE} ?
 
 sh (and just about everything else) will say:
 
 	sh: ./.make: not found
 
 or something like it (perhaps "No such file or directory") if there is
 no file named ".make" in "." (the normal, common, case) or if .make is
 a symlink, and the target of the symlink doesn't exist (or sometimes other
 errors based upon the lookup of the target).
 
 For "sh" "not found" is used, as usually commands don't have paths
 included, and the error is the result of failing to find the command
 in $PATH (but not here).
 
 If ${MAKE} isn't the full pathname to the make executable, but just "make"
 say, then "ln -s make .make; sh ./.make" is going to say "not found",
 as ATF runs its tests starting in an empty directory, and nothing I see
 for that test is creating anything called "make".
 
 kre