<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta content="text/html;charset=ISO-8859-1" http-equiv="Content-Type">
</head>
<body bgcolor="#ffffff" text="#000000">
Martin,<br>
<br>
It would be wrong to *just* mark the broken tests as @run othervm,
because then we would have no obvious way to distinguish between those
tests which are intentionally othervm, and those tests which should be
samevm but which are broken and have to be othervm "for now". <br>
<br>
It would be better to also mark such tests in some other way as well,
perhaps using a keyword, so that such tests can be easily identified
and tracked.<br>
<br>
-- Jon<br>
<br>
Martin Buchholz wrote:
<blockquote
cite="mid:1ccfd1c10911031942l2628ab63k4f377812936ed2bd@mail.gmail.com"
type="cite">One of the things you've done is to create "test sets".<br>
I did something like that in my "jtr" script (ask Tim if you can't find
it).<br>
For each logical component, it's non-trivial to find all the tests<br>
for that (e.g. how to test string handling). <br>
This is especially true for shared directories like java/lang<br>
and java/util. I don't think there's much value in creating a special<br>
target for testing java/util, since<br>
1) jtreg can do that easily already<br>
2) this doesn't match the job of any particular technology developer<br>
<br>
What would be really useful is to try to fix all the tests<br>
that fail under -samevm (perhaps by marking them<br>
@run othervm)<br>
<br>
Martin<br>
<br>
<div class="gmail_quote">On Tue, Nov 3, 2009 at 18:03, Kelly O'Hair <span
dir="ltr"><<a moz-do-not-send="true"
href="mailto:Kelly.Ohair@sun.com">Kelly.Ohair@sun.com</a>></span>
wrote:<br>
<blockquote class="gmail_quote"
style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><br>
I need some formal reviewers on these jdk/test/Makefile changes.<br>
<br>
More polish and changes may need to be done later, but there<br>
is value in what I have now, and I need lots of help to improve<br>
things (and fix some of our testcases).<br>
<br>
Here is the bugid and webrev:<br>
<br>
6888927: Fix jdk jtreg tests to indicate which ones need othervm, allow
for use of samevm option<br>
<a moz-do-not-send="true"
href="http://cr.openjdk.java.net/%7Eohair/openjdk7/jdk7-build-samevm-6888927/webrev/"
target="_blank">http://cr.openjdk.java.net/~ohair/openjdk7/jdk7-build-samevm-6888927/webrev/</a><br>
<br>
The goal was to see if we could easily run most of the regression<br>
tests in the jdk/test directory, in minimum time, and without so<br>
much noise (e.g. do not run unstable or problematic tests).<br>
This could then be used as a benchmark to validate some quality measure<br>
of the jdk7 that was built.<br>
<br>
The primary changes include:<br>
* New jdk/test/Makefile targets: jdk_all jdk_lang jdk_util jdk_nio ...<br>
* Addition of a jdk/test/ProblemList file to be used by
jdk/test/Makefile<br>
<br>
I tried to group all the tests (by directory names) into jdk_* sets,
then<br>
tried to balance them as to how long each set ran and if they could be
run<br>
with the jtreg -samevm option. The balancing was tricky, and will need
more<br>
polish, along with the virtual frame buffer idea. I also had to give up<br>
on the awt, rmi, and swing tests until the Xvfb issues are figured out.<br>
<br>
I chose to not actually modify the tests themselves if they needed
fixing,<br>
it was just too overwhelming. So developers and teams may want to
browse the<br>
ProblemList for their favorite tests (which aren't my favorites ;^) and<br>
consider what they might want to do.<br>
<br>
With this new jdk/test/Makefile, anybody can:<br>
<br>
cd jdk/test<br>
gnumake [PRODUCT_HOME=${YOUR_JDK7_HOME}] [JT_HOME=${YOUR_JTREG_HOME}]
jdk_all<br>
<br>
And run about 3,000+ tests that SHOULD PASS in roughly 2 hours if your<br>
machine isn't too slow and old. I was able to run it in 90 minutes on a<br>
monster OpenSolaris AMD machine which had 16Gb RAM and 16 2.2Ghz cpus.<br>
<br>
Then I added "-j 4" to the gnumake command line and it only took 30mins!<br>
So the various jdk_* make targets can be run in parallel.<br>
<br>
Please let me know what you think.<br>
<br>
-kto<br>
<br>
<br>
</blockquote>
</div>
<br>
</blockquote>
<br>
</body>
</html>