Skip to contentSkip to conversation

fix bug 376 - tiny extra echo in Makefile

PATCH In review Muli Ben-Yehuda 3 participants 4 messages Feb 20, 2003 Kbuild

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 fix bug 376 - tiny extra echo in Makefile

http://bugme.osdl.org/show_bug.cgi?id=376

It's not really a bug, which means this isn't really a fix, but it does make things more consistent. Please consider for inclusion or let me know you aren't interested so I can close the bug. Thanks.

# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or higher. # This patch includes the following deltas:
# ChangeSet 1.1004 -> 1.1005
# Makefile 1.362 -> 1.363
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 03/02/20 [email protected] 1.1005
# don't print the echo command when printing 'Generating build number' # --------------------------------------------
#

Makefile +1 -1
diff -Nru a/Makefile b/Makefile
@@ -325,7 +325,7 @@
 define rule_vmlinux__
 	set -e
 	$(if $(filter .tmp_kallsyms%,$^),,
-	  echo '  Generating build number'
+	  @echo '  Generating build number'
 	  . $(src)/scripts/mkversion > .tmp_version
 	  mv -f .tmp_version .version
 	  $(Q)$(MAKE) $(build)=init
Signature · 3 lines
Muli Ben-Yehuda
http://www.mulix.org
http://syscalltrack.sf.net
#1 lore raw
Quoted 21 lines · Muli Ben-Yehuda

It's not really a bug, which means this isn't really a fix, but it does make things more consistent. Please consider for inclusion or let me know you aren't interested so I can close the bug. Thanks.

# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or higher. # This patch includes the following deltas:
# ChangeSet 1.1004 -> 1.1005
# Makefile 1.362 -> 1.363
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 03/02/20 [email protected] 1.1005
# don't print the echo command when printing 'Generating build number' # --------------------------------------------
#
diff -Nru a/Makefile b/Makefile

Makefile line 331

Quoted 8 lines
Makefile +1 -1
@@ -325,7 +325,7 @@
 define rule_vmlinux__
 	set -e
 	$(if $(filter .tmp_kallsyms%,$^),,
-	  echo '  Generating build number'
+	  @echo '  Generating build number'
 	  . $(src)/scripts/mkversion > .tmp_version
 	  mv -f .tmp_version .version
 	  $(Q)$(MAKE) $(build)=init

The thing is, I cannot reproduce it here.

Also, the entire rule_vmlinux__ is invoked with a '@' in front, so nothing there should be echoed. However, for the original submitter that obviously doesn't work, the ". $(src)..." line is echoed to. What version of "make" is used?

--Kai

#2 lore raw
Quoted 16 lines · Kai Germaschewski

Makefile line 331

Quoted 11 lines · Muli Ben-Yehuda
Makefile +1 -1
@@ -325,7 +325,7 @@
 define rule_vmlinux__
 	set -e
 	$(if $(filter .tmp_kallsyms%,$^),,
-	  echo '  Generating build number'
+	  @echo '  Generating build number'
 	  . $(src)/scripts/mkversion > .tmp_version
 	  mv -f .tmp_version .version
 	  $(Q)$(MAKE) $(build)=init

The thing is, I cannot reproduce it here.

Quoted 4 lines

Also, the entire rule_vmlinux__ is invoked with a '@' in front, so nothing there should be echoed. However, for the original submitter that obviously doesn't work, the ". $(src)..." line is echoed to. What version of "make" is used?

Burton, which make version are you using? on which distro?

FWIW, I can't reproduce it here on RH 7.3 with make 3.79.1. Trying on debian unstable now.

Signature · 2 lines
Muli Ben-Yehuda
http://www.mulix.org
#3 lore raw
In reply to Kai Germaschewski
Quoted 7 lines · Kai Germaschewski

Makefile line 327

Quoted 7 lines

diff -Nru a/Makefile b/Makefile

Makefile +0 -0
@@ -325,7 +325,7 @@
 define rule_vmlinux__
 	set -e
 	$(if $(filter .tmp_kallsyms%,$^),,

While on the rule_vmlinux__ subject someone claimed that "set -e" does not make sense because it will be invoked in its private copy of the shell. Should the whole rule not be escaped "\", so it is executed in one go by make?
Thats how other users of "set -e" works.

Sam