Bug 247033 - [testcase] [hang] Generated content and display table-types maximize CPU %tage activity and hang application
Summary: [testcase] [hang] Generated content and display table-types maximize CPU %tag...
Status: CONFIRMED
Alias: None
Product: konqueror
Classification: Applications
Component: khtml (show other bugs)
Version: 4.10.2
Platform: Ubuntu Linux
: NOR grave
Target Milestone: ---
Assignee: Konqueror Developers
URL: http://test.csswg.org/suites/css2.1/l...
Keywords: reproducible, testcase
Depends on:
Blocks:
 
Reported: 2010-08-08 06:30 UTC by Gérard Talbot (no longer involved)
Modified: 2013-05-21 09:42 UTC (History)
2 users (show)

See Also:
Latest Commit:
Version Fixed In:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Gérard Talbot (no longer involved) 2010-08-08 06:30:11 UTC
Version:           4.4.0 (using KDE 4.4.5) 
OS:                Linux

Trying to load a specially crafted webpage with many :before and many :after involving css declarations of type display: table-* is impossible: the Konqueror application just hangs (CPU activity is maximized and remain maximized).

Reproducible: Always

Steps to Reproduce:
Go to
http://the-dees.webs.com/iepp1/02-gc-table-types.html

Actual Results:  
- Konqueror 4.4.5 is unable to load the webpage
- Matthew Dawson's plasmoid load system viewer 0.6 will clearly indicate a huge increase of CPU activity

Expected Results:  
Webpage should load.

- Firefox 3.6.8, Opera 10.60 and Chrome 5.0.375.125 load the webpage without problems
- If required, I could create a reduced testcase out of the original webpage
Comment 1 Gérard Talbot (no longer involved) 2010-09-25 08:32:43 UTC
The following 2 testcases create a hang situation: in fact, Konqueror 4.4.5 refuses to load the pages of the 2 testcases and then the CPU %tage of activity increases dramatically and remain high, maintain high.

http://www.gtalbot.org/BrowserBugsSection/css21testsuite/after-content-display-011.html

http://www.gtalbot.org/BrowserBugsSection/css21testsuite/before-content-display-011.html

Both testcases have been submitted already to the CSS 2.1 test suite (RC1 build 20100917):
http://test.csswg.org/suites/css2.1/20100917/html4/after-content-display-011.htm

http://test.csswg.org/suites/css2.1/20100917/html4/before-content-display-011.htm

Relevant code involved in after-content-display-011.html:

  <style type="text/css">
  div:after
  {
  content: "Filler text ";
  display: table-row;
  }
  </style>

  <p>Test passes if there are two lines of "Filler text" below.</p>

  <div>Filler text</div>

regards, Gérard
Comment 2 Gérard Talbot (no longer involved) 2011-01-10 02:57:30 UTC
http://test.csswg.org/suites/css2.1/20101210/html4/before-after-display-types-001.htm (RC4)

and

http://test.csswg.org/suites/css2.1/20101210/html4/before-after-table-parts-001.htm (RC4)

will also create an infinite reflow loop causing CPU activity maintained steadily very high (and a hang bug: application becomes unstable, unresponsive, prone to crash).

Konqueror 4.5.5, Qt 4.7.0, Linux 2.6.35-24-generic-pae, i686 (32bits) here.
Comment 3 Gérard Talbot (no longer involved) 2011-04-29 04:02:35 UTC
I still got an unresponsive application after trying to load

http://test.csswg.org/suites/css2.1/20110323/html4/after-content-display-011.htm

Konqueror 4.6.2, Qt 4.7.2, Linux 2.6.38-8-generic-pae, i686 (32bits), (Kubuntu 11.04 Natty Natty Narwhal) here.

Links to testcases must use .../20110323/...

Eg.
http://test.csswg.org/suites/css2.1/20100917/html4/before-content-display-011.htm
now becomes
http://test.csswg.org/suites/css2.1/20110323/html4/before-content-display-011.htm

regards, Gérard
Comment 4 Gérard Talbot (no longer involved) 2012-03-22 23:38:35 UTC
Konqueror 4.8.1 with KHTML rendering engine

Component: khtml

Setting severity to GRAVE

CONFIRMING



KDE Platform Version: 4.8.1
Konqueror version: 4.8.1 (with KHTML rendering engine)
Qt Version: 4.8.0
Operating System: Linux 3.0.0-17-generic-pae i686 (32bits)
Distribution: Kubuntu 11.10
here.

regards, Gérard
Comment 5 Rolf Eike Beer 2012-07-31 20:12:09 UTC
Bug is still there in SC 4.8.4. Another observation: memory consumption slowly increases, something like 1MB every few seconds on my Core2Quad.