[Webkit-unassigned] [Bug 167622] New: [GTK] Performance test IndexedDB/index-get.html timeouts on the GTK+ perf bot.

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Mon Jan 30 17:41:11 PST 2017


            Bug ID: 167622
           Summary: [GTK] Performance test IndexedDB/index-get.html
                    timeouts on the GTK+ perf bot.
    Classification: Unclassified
           Product: WebKit
           Version: WebKit Nightly Build
          Hardware: Unspecified
                OS: Unspecified
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: Tools / Tests
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: clopez at igalia.com
                CC: achristensen at apple.com, beidson at apple.com,
                    bugs-noreply at webkitgtk.org, lforschler at apple.com

The test IndexedDB/index-get.html timeouts on the GTK+ perf bot since it was added on r209765 <http://trac.webkit.org/r209765>.

The test however works fine on my development machine.

The only difference I can think of that may explain this is that the GTK+ perf bot has 8GB of RAM, my dev machine 16GB.

However, there are no signs of the machine running out of memory or the Linux Kernel OOM killer being triggered.

It seems that the test is deadlocking. As soon as it is started on the GTK+ perf bot, the Web and Database process stop using CPU time and remains in that state forever.

I have attached a GDB process to the Web and Database process and dumped the stack trace. I'm attaching the traces here.

I also tried this patch on the GTK+ perf bot. With it the test works as expected (no longer timeouts):

diff --git a/PerformanceTests/IndexedDB/index-get.html b/PerformanceTests/IndexedDB/index-get.html
index 89e52673a85..e10aa888fd3 100644
--- a/PerformanceTests/IndexedDB/index-get.html
+++ b/PerformanceTests/IndexedDB/index-get.html
@@ -36,7 +36,7 @@ function nextStep()

-var numberOfEntries = 5000;
+var numberOfEntries = 2000;

 function *runIteration()

You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.webkit.org/pipermail/webkit-unassigned/attachments/20170131/7ff91364/attachment.html>

More information about the webkit-unassigned mailing list