I can reproduce the behavior with today's git code (modulo the one unpushed
change of Dmitry's that has my +1,
https://savannah.gnu.org/bugs/index.php?64259). Rereading
https://www.gnu.org/software/make/manual/html_node/Options_002fRecursion.html,
I think you've got a case. My two cents is hardly worth bothering the list
with, but, reading:
* unlike other environment variables, MAKEFLAGS specified in the
environment take precedence over MAKEFLAGS specified in the makefile
I was moved to try adding "override " to the first line of Makefile. Then the
behavior is the same with latest make and the 4.2.1 that I happen to have in
/usr/bin.
________________________________
From: bug-make-bounces+martin.dorey=hds....@gnu.org
<bug-make-bounces+martin.dorey=hds....@gnu.org> on behalf of Ouellette, Paul
<p-ouelle...@ti.com>
Sent: Thursday, January 25, 2024 07:28
To: bug-make@gnu.org <bug-make@gnu.org>
Subject: Behavior change related to MAKEFLAGS and recursive make in version
4.4.1
***** EXTERNAL EMAIL *****
Hello,
Consider the following makefiles:
$ cat Makefile
MAKEFLAGS+=VAR=foo
all:
$(info make VAR=$(VAR))
@echo "env VAR=$$VAR"
$(MAKE) -C lib
$ cat lib/Makefile
all:
@echo "env VAR=$$VAR"
$(info make VAR=$(VAR))
With GNU Make 4.4 and older, the output is:
$ make --no-print-directory
make VAR=foo
env VAR=foo
make -C lib
make VAR=foo
env VAR=foo
With GNU Make 4.4.1, the output is:
$ make --no-print-directory
make VAR=foo
env VAR=
make -C lib
make VAR=
env VAR=
I have a makefile that depends on VAR being set in the sub-make as in Make 4.4.
Was I depending on a bug or undefined behavior or is this a bug in Make 4.4.1?
