summaryrefslogtreecommitdiff
path: root/Makefile.micro
diff options
context:
space:
mode:
authorNicholas Clark <nick@ccl4.org>2011-03-22 16:55:18 +0000
committerNicholas Clark <nick@ccl4.org>2011-03-28 10:21:48 +0100
commit1295f46c24c81f9da55f6c9e00e77ddb9cc49e34 (patch)
treee5f03f62cf0982b55b676404d7f5693a8ef95081 /Makefile.micro
parent61ce2c64c2aeaee1e05dafa775a57c3292b5f189 (diff)
downloadperl-1295f46c24c81f9da55f6c9e00e77ddb9cc49e34.tar.gz
Add very simple microtest for microperl
For now, all the tests we run don't use Config, so don't try to build it first. [Hence this is about half of the original commit by David Leadbeater, which did build Config, from config.sh] There's an unresolved problem with building Config - building Config.pm etc from config.sh won't reflect the capabilities of the microperl being tested, so that argues for the right solution to be adapting configpm to be able to use uconfig.sh instead. But doing that means that when building and (re)testing "regular" perl, we could end up with a build tree with lib/Config.pm "up to date", as far as make is concerned, but totally the wrong contents in reality. Tests will fail, heads will be scratched, time will be wasted, etc.
Diffstat (limited to 'Makefile.micro')
-rw-r--r--Makefile.micro4
1 files changed, 4 insertions, 0 deletions
diff --git a/Makefile.micro b/Makefile.micro
index 3836118f1c..e1aae13ab2 100644
--- a/Makefile.micro
+++ b/Makefile.micro
@@ -183,4 +183,8 @@ generate_uudmap$(_O): generate_uudmap.c
generate_uudmap$(_X): generate_uudmap$(_O)
$(LD) -o generate_uudmap $(LDFLAGS) generate_uudmap$(_O) $(LIBS)
+microtest: microperl
+ - cd t && (rm -f perl; ln -s ../microperl perl) \
+ && ./perl TEST base/*.t cmd/*.t
+
# That's it, folks!