(In reply to Michael Catanzaro from comment #76)
> We wound up not reverting the patches.
> 
> aarch64 is still broken (bug #209670), but ppc64le is now fine.
> 
> Solving bug #209670 is basically impossible. We are being expected to
> provide a value at compile time that cannot be known until runtime. Very
> frustrating.
> 
> Oh, and I discovered people are actually using huge pages, so WebKit is
> guaranteed to be broken for them no matter what. *shrug!*

One way around this is to disable this mitigation for those platforms
that cannot be statically configured: basically, on those platforms, at
runtime, check if page size is within acceptable limits (and whatever
other criteria you need), and if not, disable the mitigation and don’t
do the page freeze.

On platforms that can guarantee their page sizes, this option to disable
the mitigation should be #if’d out.

-- 
You received this bug notification because you are a member of Desktop
Packages, which is subscribed to webkit2gtk in Ubuntu.
https://bugs.launchpad.net/bugs/1868108

Title:
  [autopkgtest][focal] ruby-gnome/3.4.1-2build1
  class:TestWebKit2GtkWebView failure

Status in Webkit:
  Fix Released
Status in webkit2gtk package in Ubuntu:
  Fix Released
Status in webkit2gtk source package in Focal:
  Fix Released

Bug description:
  ruby-gnome/3.4.1-2* package fails with:

  
--------------------------------------------------------------------------------
  2020-03-19T13:29:50+00:00: Running test for 
/tmp/autopkgtest.2iRDtE/build.GH8/src/webkit2-gtk
  
--------------------------------------------------------------------------------
  Loaded suite test
  Started
  (run-test.rb:9142): dbind-WARNING **: 13:29:51.769: AT-SPI: Error retrieving 
accessibility bus address: org.freedesktop.DBus.Error.ServiceUnknown: The name 
org.a11y.Bus was not provided by any .service files

  ..[2020-03-19 13:29:52] INFO  WEBrick 1.6.0
  [2020-03-19 13:29:52] INFO  ruby 2.7.0 (2019-12-25) [powerpc64le-linux-gnu]
  [2020-03-19 13:29:52] INFO  WEBrick::HTTPServer#start: pid=9142 port=37651

  (WebKitWebProcess:9153): dbind-WARNING **: 13:29:52.607: AT-SPI: Error 
retrieving accessibility bus address: 
org.freedesktop.DBus.Error.ServiceUnknown: The name org.a11y.Bus was not 
provided by any .service files
  F
  
===============================================================================
  Failure: test: #load_uri(TestWebKit2GtkWebView::#load_uri):
    <true> expected but was
    <false>
  
/tmp/autopkgtest.2iRDtE/build.GH8/src/webkit2-gtk/test/test-webkit2-gtk-web-view.rb:106:in
 `block (2 levels) in <class:TestWebKit2GtkWebView>'
       103:       @view.load_uri(http_url)
       104:       loop.run
       105: 
    => 106:       assert_true(loaded)
       107:     end
       108:   end
       109: end
  
===============================================================================[2020-03-19
 13:30:22] INFO  going to shutdown ...
  [2020-03-19 13:30:22] INFO  WEBrick::HTTPServer#start done.

  ......
  Finished in 30.95292437 seconds.
  
-------------------------------------------------------------------------------
  9 tests, 9 assertions, 1 failures, 0 errors, 0 pendings, 0 omissions, 0 
notifications
  88.8889% passed
  
-------------------------------------------------------------------------------
  0.29 tests/s, 0.29 assertions/s
  Failed to run test: webkit2-gtk
  
--------------------------------------------------------------------------------
  Failed targets: webkit2-gtk
  Gdk-Message: 13:30:22.622: WebKitWebProcess: Fatal IO error 2 (No such file 
or directory) on X server :99.

  autopkgtest [13:30:22]: test run-test: -----------------------]
  run-test             FAIL non-zero exit status 1
  autopkgtest [13:30:23]: test run-test:  - - - - - - - - - - results - - - - - 
- - - - -
  autopkgtest [13:30:23]: @@@@@@@@@@@@@@@@@@@@ summary
  run-test             FAIL non-zero exit status 1
  Exit request sent.

  When running in ppc64el. This is likely due a timeout in glib main
  loop for that architecture but can't know for sure. I'm opening this
  bug so I can blacklist that particular test.

To manage notifications about this bug go to:
https://bugs.launchpad.net/webkit/+bug/1868108/+subscriptions

-- 
Mailing list: https://launchpad.net/~desktop-packages
Post to     : [email protected]
Unsubscribe : https://launchpad.net/~desktop-packages
More help   : https://help.launchpad.net/ListHelp

Reply via email to