From: Nicholas Clark <nick@ccl4.org>
Date: Fri, 8 Aug 2008 14:37:09 +0000 (+0000)
Subject: Note that I TODID most of parallel testing in the core.
X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=commitdiff_plain;h=c707cc0022231d634e4b0b1f1b9a865a02179daf;p=p5sagit%2Fp5-mst-13.2.git

Note that I TODID most of parallel testing in the core.

p4raw-id: //depot/perl@34189
---

diff --git a/pod/perltodo.pod b/pod/perltodo.pod
index 9421260..60210f9 100644
--- a/pod/perltodo.pod
+++ b/pod/perltodo.pod
@@ -54,31 +54,12 @@ checkpods and have C<make check> use podchecker.
 (This probably impacts much more than the core: also the Test::Harness
 and TAP::* modules on CPAN.)
 
-The core regression test suite is getting ever more comprehensive, which has
-the side effect that it takes longer to run. This isn't so good. Investigate
-whether it would be feasible to give the harness script the B<option> of
-running sets of tests in parallel. This would be useful for tests in
-F<t/op/*.t> and F<t/uni/*.t> and maybe some sets of tests in F<lib/>.
-
-Questions to answer
-
-=over 4
-
-=item 1
-
-How does screen layout work when you're running more than one test?
-
-=item 2
-
-How does the caller of test specify how many tests to run in parallel?
-
-=item 3
-
-How do setup/teardown tests identify themselves?
-
-=back
-
-Pugs already does parallel testing - can their approach be re-used?
+All of the tests in F<t/> can now be run in parallel, if C<$ENV{TEST_JOBS}>
+is set. However, tests within each directory in F<ext> and F<lib> are still
+run in series, with directories run in parallel. This is an adequate
+heuristic, but it might be possible to relax it further, and get more
+throughput. Specifically, it would be good to audit all of F<lib/*.t>, and
+make them use C<File::Temp>.
 
 =head2 Make Schwern poorer