RESOLVED FIXED 87534
[EFL] [DRT] Reset CacheModel before running each test
https://bugs.webkit.org/show_bug.cgi?id=87534
Summary [EFL] [DRT] Reset CacheModel before running each test
Sudarsana Nagineni (babu)
Reported 2012-05-25 14:17:33 PDT
Reset CacheModel before running each test to fix some flakiness on EFL build bots. fast/dom/StyleSheet/detached-style-2.html fast/dom/StyleSheet/detached-style-pi-2.xhtml fast/dom/StyleSheet/detached-style-pi.xhtml fast/dom/StyleSheet/detached-style.html
Attachments
patch (2.47 KB, patch)
2012-05-25 15:28 PDT, Sudarsana Nagineni (babu)
no flags
Patch (3.45 KB, patch)
2012-05-29 08:40 PDT, Sudarsana Nagineni (babu)
no flags
Sudarsana Nagineni (babu)
Comment 1 2012-05-25 15:28:58 PDT
Created attachment 144157 [details] patch Reset cache model before running each test to fix some flaky tests.
Raphael Kubo da Costa (:rakuco)
Comment 2 2012-05-25 15:59:15 PDT
Nice that Mikhail is already CC'ed; https://bugs.webkit.org/show_bug.cgi?id=86840#c4 says these settings should be already being reset. Isn't that the case?
Sudarsana Nagineni (babu)
Comment 3 2012-05-26 01:05:19 PDT
They are not reset to default on the test startup. memory_cache_clear() clears the page cache and reset to the previous value. So, we need to reset the values to default to avoid any flakiness.
Mikhail Pozdnyakov
Comment 4 2012-05-26 03:35:35 PDT
(In reply to comment #2) > Nice that Mikhail is already CC'ed; https://bugs.webkit.org/show_bug.cgi?id=86840#c4 says these settings should be already being reset. Isn't that the case? My bad :( we've discussed it with Babu memory_cache_clear() is not enough actually.
Chris Dumez
Comment 5 2012-05-28 00:28:05 PDT
Comment on attachment 144157 [details] patch View in context: https://bugs.webkit.org/attachment.cgi?id=144157&action=review > Tools/DumpRenderTree/efl/DumpRenderTreeChrome.cpp:273 > + unsigned cacheTotalCapacity = 8192 * 1024; Small nit: We could mark this variable as const.
Mikhail Pozdnyakov
Comment 6 2012-05-28 00:36:36 PDT
(In reply to comment #5) > (From update of attachment 144157 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=144157&action=review > > Tools/DumpRenderTree/efl/DumpRenderTreeChrome.cpp:273 > > + unsigned cacheTotalCapacity = 8192 * 1024; > Small nit: We could mark this variable as const. And static..
Caio Marcelo de Oliveira Filho
Comment 7 2012-05-29 00:11:12 PDT
Comment on attachment 144157 [details] patch View in context: https://bugs.webkit.org/attachment.cgi?id=144157&action=review > Tools/DumpRenderTree/efl/LayoutTestControllerEfl.cpp:550 > - case 3: // WebCacheModelPrimaryWebBrowser > + case 2: // WebCacheModelPrimaryWebBrowser Nit pick: the fact that each number have a comment with a name suggests that we could have an enum, even if it needs to be synchronized. It would also have avoided the typo you're fixing now.
Sudarsana Nagineni (babu)
Comment 8 2012-05-29 08:40:44 PDT
Created attachment 144568 [details] Patch fixed review comments.
WebKit Review Bot
Comment 9 2012-06-11 13:16:47 PDT
Comment on attachment 144568 [details] Patch Clearing flags on attachment: 144568 Committed r120001: <http://trac.webkit.org/changeset/120001>
WebKit Review Bot
Comment 10 2012-06-11 13:16:52 PDT
All reviewed patches have been landed. Closing bug.
Note You need to log in before you can comment on or make changes to this bug.