Status: WontFix
Owner: [email protected]
Labels: Type-Bug Pri-2 OS-All Area-Misc LayoutTests

New issue 24187 by [email protected]: Multiple webkit_tests failures in  
Debug
http://code.google.com/p/chromium/issues/detail?id=24187

(migrated from http://buganizer/issue?id=1058654)

nsylvain on March 07 2008 10:32
(New)
Summary
    Multiple webkit_tests failures in Debug
Component
    Chrome > Deprecated > WebKit > Layout
Reporter
    nsylvain
CC
    chrome-bugs
Type
    Bug
Priority
    P2
Severity
    S2
In prod
    false
Notes
These failures are NOT ignored. But I modified the layout tests script
locally on the debug machine to return Success even when there are errors.
Please look at the output of the webkit_tests step to see the failures.

Current failures:

Expected to fail, but passed (1):
  LayoutTests\plugins\netscape-get-property-return-value.html

Regressions: Unexpected failures (3):
  LayoutTests\http\tests\security\window-properties-pass.html
  chrome\http\tests\security\xss-DENIED-assign-location-href-javascript.html
  pending\http\tests\security\cross-frame-access-frames.html

Regressions: Unexpected timeouts (2):
  LayoutTests\fast\dom\Window\window-appendages-cleared.html
  LayoutTests\fast\dom\Window\window-properties.html

Regressions: Unexpected crashes (11):
  LayoutTests\editing\deleting\5144139-2.html
  LayoutTests\editing\execCommand\copy-without-selection.html
  LayoutTests\fast\css\font-face-implicit-local-font.html
  LayoutTests\fast\css\font-face-unicode-range.html
  LayoutTests\fast\js\function-toString-object-literals.html
  LayoutTests\svg\custom\foreignObject-crash-on-hover.xml
  LayoutTests\svg\custom\use-elementInstance-event-target.svg
  LayoutTests\svg\custom\use-elementInstance-methods.svg
  LayoutTests\svg\custom\use-instanceRoot-modifications.svg
  LayoutTests\svg\custom\use-modify-target-container.svg
  LayoutTests\svg\custom\use-modify-target-symbol.svg
nsylvain on March 10 2008 19:22
(New)
Notes
another timeout :
LayoutTests\http\tests\navigation\lockedhistory-iframe.html

-

These 3 are not related to debug mode, they were flaky at the time I
checked the failures.

  LayoutTests\http\tests\security\window-properties-pass.html
  chrome\http\tests\security\xss-DENIED-assign-location-href-javascript.html
  pending\http\tests\security\cross-frame-access-frames.html
eseidel on March 13 2008 17:54
(New)
Hotlist
    layout_test_failures
nsylvain on March 19 2008 14:16
(New)
Notes
2 tests that are failing in debug only:

LayoutTests\fast\loader\user-style-sheet-resource-load-callbacks.html
LayoutTests\http\tests\loading\basic.html
ericroman on March 25 2008 10:51
(New)
CC
    chrome-bugs, ojan
Notes
1082769 has been marked a duplicate of this issue
jhaas on April 22 2008 12:55
(New)
Notes
I have a fix for the SVG tests out for review now.
dsiroker on April 28 2008 11:38
(Assigned)
Assignee
    jhaas
CC
    chrome-bugs, dsiroker, ojan
Priority
    P1
Hotlist
    layout_test_failures, Cr__Rel_1.0
Notes
Jonathan, would you mind investigating the other failures on this list as  
well?
jhaas on April 28 2008 16:04
(Assigned)
Notes
What I've found so far:

LayoutTests\editing\execCommand\copy-without-selection.html is a WebKit
bug, and repros in Safari. The WebKit code for handling Windows clipboards
seems a bit rickety, and is not tested very well by the layout tests.
(Several other scenarios result in the same failure.)

LayoutTests\editing\deleting\5144139-2.html is a failure in our font code.
The problem isn't in the layout test itself, it's in the CSS font-face
declarations at the top of the test. We assume that the number of
characters requested to be filled in a font glyph page is either equal to
the maximum (fixed) size of the page or twice that size. It seems that with
segmented fonts, it's possible to be passed a value less than the maximum
page size which triggers an assert in our GlyphPageTreeNodeWin.cpp. Note
that in Safari, the font is not marked as segmented for some reason.

LayoutTests\fast\js\function-toString-object-literals.html is caused by a
bug in V8's version of strtod. It dies (calls exit(), in fact) on the value
"1e-500". Got an email out to find out why.
dsiroker on April 28 2008 17:32
(Assigned)
Notes
Good stuff. For
LayoutTests\editing\execCommand\copy-without-selection.html, I would vote
to DEFER it for post-beta if our current behavior is the same as Safari 3.1
since our goal for beta is to be bug-for-bug compatible with Safari 3.1
jhaas on April 29 2008 10:53
(Assigned)
Notes
LayoutTests\fast\css\font-face-implicit-local-font.html and
LayoutTests\fast\css\font-face-unicode-range.html are caused by the same
font problem as LayoutTests\editing\deleting\5144139-2.html. Being tracked
by http://b/issue?id=1007391.

The strod problem responsible for
LayoutTests\fast\js\function-toString-object-literals.html is a bogus
assertion. V8 will be removing it, which should clear up that problem.
eseidel on April 29 2008 14:38
(Assigned)
Notes
AFAICT, all of these failures are covered by other more-specific bugs.  I
suggest we close this one for less noise in Chrome Dashboard.
jhaas on April 30 2008 14:39
(Assigned)
Notes
Many of the failures fixed, all others covered in individual bugs.
jhaas on April 30 2008 14:43
(Fixed / Fixed)
laforge on October 28 2008 12:57
(Verified / Fixed)
Verifier
    laforge
Notes
Cleanup operation, batch updating old 'FIXED' issues to 'VERIFIED' that
have not been modified since 9/2/08.
        
Status  Verified
Resolution      Fixed

Reporter        nsylvain
Assignee        jhaas
Verifier        laforge
CC      chrome-bugs, dsiroker, ojan

Type    Bug
Priority        P1
Severity        S2

Depends on      
Blocking        

Changelists     

Found in        
Targeted to     
Verified in     
In prod         false

Hotlists        
Cr__Rel_1.0
layout_test_failures
Email:  [email protected]


--
You received this message because you are listed in the owner
or CC fields of this issue, or because you starred this issue.
You may adjust your issue notification preferences at:
http://code.google.com/hosting/settings

--~--~---------~--~----~------------~-------~--~----~
Automated mail from issue updates at http://crbug.com/
Subscription options: http://groups.google.com/group/chromium-bugs
-~----------~----~----~----~------~----~------~--~---

Reply via email to