Commit Graph

37670 Commits

Author SHA1 Message Date
Benjamin Smedberg
3534762440 Fix typo in error-handling case that we rarely hit, r=trivial 2010-01-26 17:46:47 -05:00
Jim Mathies
ed02ce9b5d Bug 541362 - [OOPP] hang closing tab with Yahoo web messenger loaded. r=bent. 2010-01-26 15:40:43 -06:00
Jim Mathies
ed4ba1d2fa Bug 540052 - [OOPP] Test failure in test_plugin_focus.html. r=bent. 2010-01-26 15:40:15 -06:00
Chris Jones
7df4c01b33 Yet more temporary debug spew for diagnosing mochitest hang. 2010-01-26 14:57:16 -06:00
Benjamin Smedberg
aebc8a52c7 Bug 535564 yet another attempt - Don't inherit the leak log handle from parent to child processes: Java spawns a process which holds this handle alive beyond our shutdown, causing our test harnesses to be unable to delete it, r?jgriffin 2010-01-26 15:50:53 -05:00
Chris Jones
abf4355d51 Temporary mochitest debugging: eliminate race condition between parent exit and child minidump request. 2010-01-26 09:40:42 -08:00
Benjamin Smedberg
cdf47037d7 Bug 540604 - silverlight doesn't work with OOPP because it doesn't set the stype outparam for NPP_NewStream, and relies on us initializing the value with NP_NORMAL, r=cjones 2010-01-26 12:05:54 -05:00
Benjamin Smedberg
c2eaa59e85 More testing from bug 535564 - Move the process-log removal after we kill ssltunnel: I have a hunch that the ssltunnel process is inheriting the file handle from python. 2010-01-26 12:50:48 -05:00
Benjamin Smedberg
fff188953b Bug 535564 - automation.py: WindowsError: [Error 13] The process cannot access the file because it is being used by another process. Don't keep the PID log open all the time, but only open it when it is needed. r?griffin
--HG--
extra : rebase_source : 1072e20feb1db8a98f99de5617e416ca23efb511
2010-01-26 09:53:32 -05:00
Chris Jones
382bd97b0b More temporary mochitest hang diagnostics: kill the plugin process with SIGSEGV if it doesn't shut down within 30 seconds. 2010-01-25 18:49:02 -06:00
Karl Tomlinson
7a9997e1bf b=540910 restore flushing of Xlib requests from the parent to the server before testing from the OOP plugin 2010-01-26 13:48:16 +13:00
Benjamin Smedberg
6ff062a712 Rename Crash() to IntentionalCrash() so that automation can grep for it. Make it non-static, and add trailing instructions to it so that it isn't inlined nor tail-called away in crash reports.
--HG--
extra : rebase_source : c74c5bf161508fbc46478bba81d6bd791091485b
2010-01-25 17:26:23 -05:00
Chris Jones
e03c1f3854 More temporary debug spew to sort linux mochitest hang. 2010-01-25 14:10:12 -06:00
Benjamin Smedberg
10c46ca526 Printf-debugging to help figure out bug 535564 2010-01-25 14:33:50 -05:00
Benjamin Smedberg
003e478253 Re-enable the timer/asynccall tests now that the ChildNPObject leaks are fixed 2010-01-25 14:15:13 -05:00
Benjamin Smedberg
0ba2d6e348 Merge mozilla-central with electrolysis. 2010-01-25 11:57:19 -05:00
Benjamin Smedberg
2b12cc735a Merge electrolysis tip with intermediate merge point. 2010-01-25 11:57:00 -05:00
Benjamin Smedberg
837889eb6b Merge part of mozilla-central to electrolysis after the pushes from 21-Jan. 2010-01-25 11:56:06 -05:00
Boris Zbarsky
e3646615a4 Bug 538063. Fix the assert condition to not trigger for <area>s. r=tn 2010-01-25 11:03:48 -05:00
Boris Zbarsky
0b39e83022 Bug 541382. Buttons with non-visible overflow should clip their contents. r=roc 2010-01-25 11:03:43 -05:00
Alexander Surkov
f19fe9bb77 Bug 541108 - nsAccessible::DoCommand shouldn't return value, r=davidb 2010-01-25 23:09:25 +08:00
Alexander Surkov
ac429f5b6b Bug 523785 - Use NS_DispatchToMainThread instead of nsITimer, r=ginn, davidb, sr=smaug 2010-01-25 23:08:08 +08:00
Chris Jones
bd0804d7d6 More debug spew for sorting mochitest-plain hang. 2010-01-22 18:03:38 -06:00
Chris Jones
db57f114be Temporary spew part 2: convert fprintf(stderr into printf( so that child output shows up in the log. 2010-01-22 15:31:56 -06:00
Chris Jones
7b53db6a22 Temporary debug spew for diagnosing Linux mochitest hang. 2010-01-22 12:53:18 -06:00
Benjamin Smedberg
9c1e8d42c6 Bug 540967 - "missing output line for total leaks" because the parent process isn't waiting for the child process to clean up/write leak logs, r=cjones
--HG--
extra : transplant_source : EW%5C-%60M%BCabd%A0%DE%F8%1EG%DF%03%14%B3%24
2010-01-22 10:21:33 -05:00
Benjamin Smedberg
ad605908e9 Bug 540967 - "missing output line for total leaks" because the parent process isn't waiting for the child process to clean up/write leak logs, r=cjones 2010-01-22 10:21:33 -05:00
Benjamin Smedberg
25d837343c Bug 540868 part 3: when deallocating NPObjects in leakcount builds, let the leak monitor know that we cleaned them up correctly, r=bent
--HG--
extra : transplant_source : H%AC%29%DF8y%E8f%F1%2B%3B%5C%A3Z%10%D1R%7E%CC%10
2010-01-22 10:06:30 -05:00
Benjamin Smedberg
8adba8c2a2 Bug 540868 part 2: keep all NPObjects (not just those with actors) in mObjectMap, and invalidate/deallocate them when we destroy the associated instance, r=bent
--HG--
extra : transplant_source : %96H%CA%BF%2BX%18s%C8%1B%2Bh%8F%92%05%F0%9F/N%01
2010-01-22 10:06:13 -05:00
Benjamin Smedberg
6f7867ca0f Bug 540858 part I: separate the refcount logging of NPObject (which works for both plugin and proxy NPObject) from ctor/dtor logging of ChildNPObject (proxy only), r=bent
--HG--
extra : transplant_source : %16%16%92s%8C%B1%AC%8AT%23%DF%DD%A0%B7RH%84C%A7%3A
2010-01-22 10:05:29 -05:00
Karl Tomlinson
a2b3065284 b=540910 use timeouts in plugin_clipping_helper2 so that the X server processes Xlib requests from the parent before testing requests from the OOP plugin 2010-01-22 23:41:07 +13:00
Benjamin Smedberg
32dd918b29 Bug 540868 part 3: when deallocating NPObjects in leakcount builds, let the leak monitor know that we cleaned them up correctly, r=bent 2010-01-22 10:06:30 -05:00
Benjamin Smedberg
5ff7ecd852 Bug 540868 part 2: keep all NPObjects (not just those with actors) in mObjectMap, and invalidate/deallocate them when we destroy the associated instance, r=bent 2010-01-22 10:06:13 -05:00
Benjamin Smedberg
1fc18fe407 Bug 540858 part I: separate the refcount logging of NPObject (which works for both plugin and proxy NPObject) from ctor/dtor logging of ChildNPObject (proxy only), r=bent 2010-01-22 10:05:29 -05:00
Chris Jones
6345658b6f Bug 521929, part 3: IPDL/C++ test.
--HG--
extra : transplant_source : %9E%82%3E%20E%81J%97r%CD%05%93%FC%E0%9C%877P%DF%EE
2010-01-21 20:04:11 -06:00
Chris Jones
73a1d2211e Bug 521929, part 3: IPDL/C++ test. 2010-01-21 20:04:11 -06:00
Chris Jones
2649cb4bf4 Bug 521929, part 2: Save racy RPC replies onto a special stack until they're the reply to the right out-call. r=bent
--HG--
extra : transplant_source : %95R%85%B4%AD%0F%3D%9B%A5%18n%9B%94%BF%DA%9A%1BE%40%AC
2010-01-21 20:04:10 -06:00
Chris Jones
b79dc7dd3a Bug 521929, part 2: Save racy RPC replies onto a special stack until they're the reply to the right out-call. r=bent 2010-01-21 20:04:10 -06:00
Chris Jones
bcc3aca59d Bug 521929, part 1: Add a "seqno" field to synchronous messages. r=bent
--HG--
extra : transplant_source : %D0%5D%CF%0DUv%AC%AD%09%F98%C4%14%E2%F8%14%2CUZ%97
2010-01-21 20:04:09 -06:00
Robert Longson
c1b79c43cb Bug 539576 - SVGTransform matrix changes still not live in all circumstances (with bustage fix). r=jwatt 2010-01-25 14:36:09 +00:00
Robert Longson
b5611e5ba7 Backout Bug 539576 due to test failures 2010-01-25 13:44:55 +00:00
Robert Longson
1034acc582 Bug 539576 - SVGTransform matrix changes still not live in all circumstances. r=jwatt 2010-01-25 12:26:15 +00:00
Olli Pettay
afce7361da Bug 539060 - Crash [@ do_QueryFrame::operator<nsIPageSequenceFrame> nsIPageSequenceFrame*() ] in print preview , r=roc
--HG--
extra : rebase_source : 246c540277da95dbc670bcfc899dc9f9796f603e
2010-01-25 13:15:25 +02:00
Chris Jones
5f451f4f10 Bug 521929, part 1: Add a "seqno" field to synchronous messages. r=bent 2010-01-21 20:04:09 -06:00
Benjamin Smedberg
c68bd0290e Merge backout of the printfs to diagnose bug 541155 2010-01-21 15:33:20 -05:00
Benjamin Smedberg
3fd32375e2 Backed out changeset df78efe07b18, debugging only printfs now that 541155 is diagnosed. 2010-01-21 15:32:58 -05:00
Karl Tomlinson
4388dea01e b=540910 flush Xlib requests from the parent to the server before testing from the OOP plugin, and increase the timeout so that GTK's idle timer gets run for XEmbed plug sizing 2010-01-21 23:32:17 +13:00
Jonathan Griffin
34998937f9 Test for bug 539755, NPN_GetAuthenticationInfo 2010-01-21 14:35:03 -05:00
Benjamin Smedberg
68f46ef6d9 Stupid logic error from bug 539755 2010-01-21 14:35:00 -05:00
Benjamin Smedberg
abe8c90cf2 Bug 541155 - test_loadflags.html doesn't reset prefs properly (manifests as a failure in modules/plugin/test/test_cookies.html), r=dwitte 2010-01-21 14:11:49 -05:00