It has previously been increased for a specific test, but we have found
another test that needs it increased. Rather than increasing it for
individual tests, increase it for all tests.
--HG--
extra : rebase_source : f76c2a43bdcfa97a86fb5d63ff9708e6215c2678
extra : source : f9a184b70afcf294f7eb84897448fb3db03f681a
This avoids excessive amounts of test data accumulating which can slow tests down.
As part of this change, the pref for enabling the logging was made 'Live'.
--HG--
extra : rebase_source : e1d0547e5d3125e9db44bb47ddfff3b4f8d50e99
extra : source : 73fb13743a74a5a920d099fda70d9b1a1fa3a6ed
Kyle assures me it's safe to call Send__delete__ with references to 'this' on
the stack.
--HG--
extra : commitid : I7Wj1a8Hmb9
extra : rebase_source : 8cf8851a7f4400fb1120179423ec4cf710d75765
It has previously been increased for a specific test, but we have found
another test that needs it increased. Rather than increasing it for
individual tests, increase it for all tests.
--HG--
extra : rebase_source : 1e32b0c851bb138d1ed247fea72312c09380266c
extra : source : fb4d8f6db088a2da12d01403280b959d176fee0a
This avoids excessive amounts of test data accumulating which can slow tests down.
As part of this change, the pref for enabling the logging was made 'Live'.
--HG--
extra : rebase_source : 0abbf10b3e3b3a67bbb810d4554a7b55bfc6fd66
extra : source : d7bac1264671c0a10da034d6d9b08f01235ebdb3
We don't want to split native NSEvents into two PanGestureInput events.
--HG--
extra : commitid : 7UnFAreSEcF
extra : rebase_source : 70121ddcf52f00547e70fcbea16975224835f890
extra : histedit_source : 0b8840795e6ab1cc808d75aa042c030b70f09f6f
Now that PanGestureInput events are only processed during a PanGestureBlock, we can get rid of the mPanGestureState workaround.
--HG--
extra : commitid : 4KYXEbxJ5zZ
extra : rebase_source : 9ee0b85d3bec2554871f621270998b2e2d86a3ce
extra : histedit_source : e7c56cca500680c95744110ccaeefa635f7f1613
This will make bug 1156606 unnecessary.
--HG--
extra : commitid : K69ATp0TzlZ
extra : rebase_source : 9fd2b710623f6f54fe707a5ab8fb3c4b923348f3
extra : histedit_source : f86517d5d01ef4a708dcfa5529bed165d9016b10
PanGestureInput events need to be sent to the pre-scroll position, just like wheel events.
--HG--
extra : commitid : CjK4ecz5AHl
extra : rebase_source : 438bc15164d10ddbc18e01435289859dafcf8db4
extra : histedit_source : 88a04fa28a5ceebebf48ed900d05eb9416b00b40
This was discussed in bug 1107716.
--HG--
extra : commitid : 8EvQL9qrm3X
extra : rebase_source : 3785c8ac8e7e90f171cf9091f6fc50d76d34f703
extra : histedit_source : ffc8c6cd57f535b03621e149bb300e60683d1612
When scrolling an inactive subframe, the target APZC confirmation needs to be sent along with the layers transaction. If a new wheel event comes in while that target APZC notification is in-flight, the subframe will already have a display port, so the confirmation will be sent in a more direct fashion and can arrive at the APZ controller thread before the new APZC for the scroll frame has been created. Then the current input block will have a null target APZC and no scrolling will happen until a new input block is created. (For wheel transactions, a null target APZC ends the transaction, so the next event will create a new input block.)
--HG--
extra : commitid : Ij9XfKKZAqe
extra : rebase_source : 66196ed4e3e98159f7b17f7e8d90c5f2985425e6
extra : histedit_source : c028ec2f7817bb8d9396849495bfeaa17cbe7064
Lee Salzman
Botond Ballo
Kearwood (Kip) Gilbert
Benoit Girard
JerryShih
Robert O'Callahan
Phil Ringnalda
Andrew Comminos
Kartikaya Gupta
Markus Stange
Masayuki Nakano
Ryan VanderMeulen
Mason Chang
Morris Tseng
Victor Porof
Nathan Froyd
CJKu
Wes Kocher
Jamie Nicol