This test is a new test added in r608486.
I'm not reverting this, since this test mostly passes within the retries in
the single run of layout tests. I'm assigning mythria@ as the author of this
test.
The failure diff looks like the following:
--- /b/s/w/ioCkh2GG/layout-test-results/virtual/site-isolated-code-cache/http/tests/devtools/isolated-code-cache/same-origin-test-expected.txt
+++ /b/s/w/ioCkh2GG/layout-test-results/virtual/site-isolated-code-cache/http/tests/devtools/isolated-code-cache/same-origin-test-actual.txt
@@ -84,7 +84,7 @@
columnNumber : 0
consumedCacheSize : <number>
lineNumber : 0
- notStreamedReason : "already used streamed data"
+ notStreamedReason : "script too small"
streamed : <boolean>
url : .../devtools/resources/v8-cache-script.js
}
Thanks for this bug. I think this is happening because the ScriptResource that as fetched earlier could be in the memory cache. If it is in the memory cache we have "already used streamed data" or if it was evicted then we get "script too small". I am trying to force a GC so we get "script too small" always. I will upload a cl for that.
Is there a way I can check if it is still flaky or not? I mean run a try job or something, since I can't reproduce this locally. When I look at the bot it is mostly green, but may be because with num_tries = 3, one of them passes.
Comment 1 by yutak@chromium.org
, Nov 19Labels: -Pri-1 Pri-2
Owner: mythria@chromium.org