On 4/6/20 1:21 PM, Daniel D. Daugherty wrote:
On 4/6/20 4:10 PM, Chris Plummer wrote:
Hello,
Please review the following:
https://bugs.openjdk.java.net/browse/JDK-8242235
http://cr.openjdk.java.net/~cjplummer/8242235/webrev.00
test/hotspot/jtreg/ProblemList.txt
No comments.
test/jdk/ProblemList.txt
No comments.
test/lib/jdk/test/lib/Platform.java
No comments.
Thumbs up.
In the bug report, you said:
although due to test renames, moves, and additions, there are actually
still about 8 tests running on Solaris, but they shouldn't be.
Perhaps those 8 tests are not affected by the bug that causes
JDK-8193639.
I don't see any recent links to JDK-8193639 that are valid failures.
There are some links to JDK-8193639 for non-Solaris platforms which means
those links are wrong, but none for Solaris (that I see).
Hi Dan,
Thanks for the review.
On closer look it's more like 3 tests since 5 of the 8 don't really
count. One was a new test I wrote that I haven't pushed yet (and didn't
problem list because I wanted to do this CR first). Two are tests that
are not run for other reasons (but are also incorrectly problem listed
for 8193639). Two are tests that only recently started running on
solaris due to splitting out into #id0 and #id1 runs, so the problem
list is not properly filtering them.
So maybe 3 tests have been running for a while without causing issues,
but I honestly don't think it's worth the effort to figure out which
tests are truly affected by JDK-8193639, and try to make the problem
list correct to allow them to run.
thanks,
Chris
Dan
The SA problem list entries due to JDK-8193639 [1] have been a burden
to maintain, and are badly out of date. Easiest solution is to remove
them and disable SA testing on Solaris, which is what all the entries
were suppose to be doing in the first place (although not always
correctly).
Once this change is pushed, I'll add a note to JDK-8193639 [1] so
anyone working on it will know to undo the change in Platform.java.
[1] https://bugs.openjdk.java.net/browse/JDK-8193639
thanks,
Chris
