Commit graph

1112 commits

Author SHA1 Message Date
Markus Stange
901b7c53e8 Bug 1236040 - When updating the scroll position, only schedule image visibility checks if the display port has changed. r=tn
--HG--
extra : commitid : Hir2YBOuC1J
extra : rebase_source : 6c8d9df502979615e9f72a47b035b15d17f9c518
2015-12-31 00:04:18 +01:00
David Anderson
74d965910c Fix crash trying to allocate textures for oversize scrollbars. (bug 1214261, r=mstange) 2015-12-17 11:13:45 -08:00
Timothy Nikkel
4ec53b3bfa Bug 1234725. Don't change the dirty rect to the display port when building display lists for documents. r=mstange
Instead do it when we first encounter the root scroll frame.

Doing this goes back to bug 635053, where we did it because fixed position items weren't getting included. However in bug 974643 we learned that this was wrong. Displayports aren't relevant to fixed pos content, displayports are only relevant to scrolled content. And we set the dirty rects of fixed pos content specially. The only other thing that should be affected is scrollbars, and we already carefully set their dirty rects too.
2016-01-09 00:25:58 -06:00
Timothy Nikkel
6a24899c5f Bug 1194851. Change the GetDisplayPort API to take an optional RelativeTo parameter, instead of having that expressed in the function name. r=botond 2016-01-07 18:27:49 -06:00
Timothy Nikkel
4e74899f7a Bug 1194851. Rename GetDisplayPortRelativeToScrollPort back to GetDisplayPort. r=botond 2016-01-07 18:27:49 -06:00
Timothy Nikkel
3c7c901321 Bug 1194851. Split GetDisplayPort callers into GetDisplayPortRelativeToScrollPort and GetDisplayPortRelativeToScrollFrame. r=botond
Temporarily rename GetDisplayPort to GetDisplayPortRelativeToScrollPort for the duration of this patchset.

This means that every caller of GetDisplayPort is guaranteed to be touched by this patchset (assuming it compiles), and thus each call site can be checked in review to make sure it is relative to the correct coordinate system.
2016-01-07 18:27:49 -06:00
Timothy Nikkel
f06c43de07 Bug 1194851. Always store the displayport base relative to the scrollport (not the scroll frame). r=botond
This also has the side effect of making the computation that takes place in GetDisplayPortFromMarginsData happen relative to the scrollport. As opposed to some things relative to the scroll port, and some things relative to the scroll frame.
2016-01-07 18:27:49 -06:00
Timothy Nikkel
f42493831a Bug 1194851. Change GetOrMaybeCreateDisplayPort into just MaybeCreateDisplayPort, and make it only return a bool, and not the displayport rect. r=botond
This makes the one caller that needs the displayport rect have to ask for it seperately.

The reason for this is later in the patch series we need to add "RelativeToScrollFrame/Port" to all displayport getters, but there is no semantically good way to do that to the name GetOrMaybeCreateDisplayPort.
2016-01-07 18:27:48 -06:00
Botond Ballo
b9aee9ebb5 Bug 1194851 - Add helper functions HasDisplayPort() and HasCriticalDisplayPort(). r=kats
This makes it clearer whether we're just checking if a displayport exists,
or we're actually consuming its value.

--HG--
extra : source : 70bb222e860669a6cf6e803dbd57f088ca4fbc04
2015-12-16 17:22:23 -05:00
Markus Stange
b22db3eab2 Bug 1147673 - Make display items know about their scroll clips. r=tn, r=roc
--HG--
extra : commitid : 1j8QG2b0rIG
extra : rebase_source : 027065697e7e0ef445d8bcf7a81cf0270dcf7b8b
2015-12-22 16:54:19 +01:00
Markus Stange
30bf0a5627 Bug 1147673 - Use ancestor clip for root scrollable framemetrics clips. r=tn
--HG--
extra : commitid : Do9D8ZPMoLv
extra : rebase_source : 558a5311c4f2bceaf301a6c0c5e1233effdc2bd4
2015-11-05 17:54:44 +01:00
Kartikaya Gupta
4a4a0bcf45 Bug 1234242 - Keep a RefPtr to the refresh driver to unregister the ScrollEvent, rather than relying on a chain of pointers remaining non-null. r=mstange a=KWierso
--HG--
extra : commitid : JouJUWdzJl6
2015-12-21 14:40:20 -05:00
Markus Stange
8d6e827631 Bug 1209970 - Fire scroll events early in the refresh tick. r=mats
With APZ we want to be firing scroll events to content more consistently, so
we tie them to the refresh driver tick rather than firing them on paint or
haphazardly on the next spin of the event loop.

Patch by Markus Stange, test fixes by Kartikaya Gupta

--HG--
extra : commitid : 7nnkRC8afAJ
2015-12-17 17:19:30 -05:00
Avi Halachmi
9b56ccfb99 Bug 1228147: part 2: Add telemetry probe for synchronous scroll. r=kats 2015-12-16 20:06:40 +02:00
Wes Kocher
0f6a2a587a Backed out 2 changesets (bug 1228147) for build failures in histogram.h on a CLOSED TREE
Backed out changeset 35b4ba91e093 (bug 1228147)
Backed out changeset 8a2f45c8d25d (bug 1228147)

--HG--
extra : commitid : Atv8AVyyd9h
2015-12-16 09:27:37 -08:00
Avi Halachmi
7762c6b58a Bug 1228147: part 2: Add telemetry probe for synchronous scroll. r=kats 2015-12-16 17:18:45 +02:00
Kartikaya Gupta
edc45db293 Bug 1232048 - Fix breakage in scroll snapping when snapping twice in a row to the same thing with non-smooth-scrolls in between. r=botond
--HG--
extra : commitid : HOoM79IZd2u
2015-12-16 10:42:49 -05:00
Carsten "Tomcat" Book
66e5f0c4ab Backed out changeset 04b102bec363 (bug 1147673) 2015-12-16 11:52:39 +01:00
Carsten "Tomcat" Book
27bbe6ec93 Backed out changeset 68b33692bed3 (bug 1147673) 2015-12-16 11:52:37 +01:00
Markus Stange
bf668586d7 Bug 1147673 - Make display items know about their scroll clips. r=tn, r=roc
--HG--
extra : commitid : 7KbOQKKSVq5
extra : rebase_source : d23daf86ed8d375bd5ab3ed7dfe3cff80bc0ff4c
extra : histedit_source : cce4ded91f0672142796a852d20d052e7b6247ce
2015-12-12 10:58:37 -05:00
Markus Stange
9079fbe047 Bug 1147673 - Use ancestor clip for root scrollable framemetrics clips. r=tn
--HG--
extra : commitid : 8i9bmvj9S6Y
extra : rebase_source : 6316afb1c7967fd944281f0686b2c6ba7379606f
extra : histedit_source : 1289a445613a59b213b79396a60076ebc4e83beb
2015-11-05 17:54:44 +01:00
Kartikaya Gupta
0dcf99f8a4 Bug 1229052 - Log a warning when we detect a scroll-linked effect based on the scroll event. r=roc 2015-12-14 14:47:24 -05:00
Kartikaya Gupta
6b0c32286a Back out 5 csets from bug 1229052 on a CLOSED TREE for failing to fix the nonunified build error even after many attempts. r=backout 2015-12-14 15:52:31 -05:00
Kartikaya Gupta
fcb193563c Bug 1229052 - Log a warning when we detect a scroll-linked effect based on the scroll event. r=roc 2015-12-14 14:47:24 -05:00
Kartikaya Gupta
6c7c3b6adf Bug 1228407 - Avoid a scenario where content floods APZ with smooth-scroll requests and blocks it from syncing a new scroll offset back to the main thread. r=botond
--HG--
extra : commitid : 6YhRu80eZzC
2015-12-09 10:13:34 -05:00
Kartikaya Gupta
9491a614c4 Back out three patches from bug 1228407 on a CLOSED TREE for causing various mochitest failures. r=backout
--HG--
extra : commitid : 7gaWjh5Noye
2015-12-08 21:00:30 -05:00
Kartikaya Gupta
56e9023bca Bug 1228407 - Avoid a scenario where content floods APZ with smooth-scroll requests and blocks it from syncing a new scroll offset back to the main thread. r=botond
--HG--
extra : commitid : LdMBKbQQw7V
2015-12-08 14:56:26 -05:00
Kartikaya Gupta
0f7fe3417c Bug 1156392 - Remove an ifdef condition that we don't need any more. r=tn 2015-12-01 22:21:42 -05:00
Kartikaya Gupta
fe65090c97 Bug 1228357 - When saving/restoring a scrollframe's state, only save/restore resolution for the root scrollframe. r=roc 2015-11-26 21:25:53 -05:00
Ting-Yu Lin
79afc81f6e Bug 1226875 - Remove nsIFrame::GetFirstChild(). r=mats
--HG--
extra : commitid : J8NTL74JosY
extra : rebase_source : 50a196905ac06d09fea17bb1aa17efe821cc7cbb
2015-11-26 17:12:39 +08:00
Nicholas Nethercote
52b3d826da Bug 1225007 (part 1, attempt 3) - Use LayoutDevicePixel more in Cocoa widget code. r=kats.
--HG--
extra : rebase_source : e4b3bb36c18822557f7724a58979eddd9d8cac62
2015-11-19 14:10:38 +11:00
Kartikaya Gupta
c460ebbd06 Bug 1226872 - Stop keeping a copy of the resolution in the root scrollframe. r=roc
The only reason we had this in the scrollframe at all was so that it could be
saved/restored as part of the frame state when leaving a page and then going
back to it. However we can accomplish this by just reading/writing the resolution
from/to the presshell instead, so there's no need to keep a second copy of it.

--HG--
extra : commitid : J4QBfG2GGjn
2015-11-23 09:14:15 -05:00
Phil Ringnalda
87802cdbad Backed out 2 changesets (bug 1225007) for frequent OS X "Shouldn't return empty rect" assertion failures, a=backout
Backed out changeset c5b352c74b35 (bug 1225007)
Backed out changeset 8a513c70ce7c (bug 1225007)
2015-11-19 20:55:26 -08:00
Nicholas Nethercote
ce671f3678 Bug 1225007 (part 1, attempt 2) - Use LayoutDevicePixel more in Cocoa widget code. r=kats.
--HG--
extra : rebase_source : ea132cb4a6eda3553682114f3a1a280acfaf0aa3
2015-11-19 14:10:38 +11:00
Nicholas Nethercote
26b0d282a5 Backout 52ba24123046 and 66768d6a50fa (bug 1225007) for causing lots of test assertions on Mac. CLOSED TREE 2015-11-17 17:02:39 -08:00
Nicholas Nethercote
eb12a384c0 Bug 1225007 (part 1) - Use LayoutDevicePixel more in Cocoa widget code. r=kats.
--HG--
extra : rebase_source : ad3466d05e476ebf65affdde3fca52494e1327cb
2015-11-15 21:37:52 -08:00
Matt Brubeck
871cea5bf5 Bug 1081272 - Include position and transform when computing header/footer size [r=tn] 2015-11-12 16:55:02 -08:00
Matt Brubeck
3777952460 Bug 1224307 - Fix page scrolling for non-full-width fixed headers/footers [r=tn] 2015-11-12 16:55:00 -08:00
Phil Ringnalda
78c184fb46 Back out changeset cd0be67d2344 (bug 1224307) for test_page_scroll_with_fixed_pos.html bustage
CLOSED TREE
2015-11-12 20:18:38 -08:00
Phil Ringnalda
08a38cf3b8 Back out changeset 13335ded96b9 (bug 1081272) for test_page_scroll_with_fixed_pos.html bustage
CLOSED TREE
2015-11-12 20:18:03 -08:00
Matt Brubeck
ae413e1b82 Bug 1081272 - Include position and transform when computing header/footer size [r=tn] 2015-11-12 16:55:02 -08:00
Matt Brubeck
c1f53f3bdd Bug 1224307 - Fix page scrolling for non-full-width fixed headers/footers [r=tn] 2015-11-12 16:55:00 -08:00
Mats Palmgren
eb154663cc Bug 1092626 - Don't try to restore the scroll position if the frame is still dirty or has dirty children. r=roc 2015-11-11 11:33:34 +01:00
Timothy Nikkel
cf2a0eeaa4 Bug 1220114. Part 2. If a scroll frame is forced to layerize because of an active descendant scroll frame then set a displayport on the ancestor scroll frame so that next paint we don't have to force layerization after the fact and we can do a fully proper paint. r=mstange 2015-11-06 17:32:03 -06:00
Timothy Nikkel
0570e20dc3 Bug 1220114. Part 1. Change ScrollFrameHelper::DecideScrollableLayer to recompute the current animated geometry root any time mWillBuildScrollableLayer changes in addition to when usingDisplayPort changes. r=mstange
Changes in either one can result in AGRs being created or removed (although nothing removes them at this point in time).
2015-11-06 17:32:03 -06:00
Timothy Nikkel
ae9c7ca837 Bug 1208780. Set a zero-margin displayport on all scrollable ancestors of frames with displayports if they don't already have a displayport. r=botond 2015-11-05 18:57:17 -06:00
Markus Stange
ef4655be32 Bug 1201330 - Keep scroll handler induced layer activity active until the scroll frame becomes inactive. r=roc
--HG--
extra : commitid : K9ntCNQz1Fx
extra : rebase_source : ed8a567f22626d7074bbc58eb5e61baa251fe2c3
2015-10-30 16:28:53 +01:00
Birunthan Mohanathas
9985829ecc Bug 1219392 - Capitalize mozilla::unused to avoid conflicts. r=froydnj 2015-11-02 07:53:26 +02:00
Timothy Nikkel
ab02b09009 Bug 1220020. Recompute the current animated geometry root when a child scroll frame forces us to be layerized, and therefore turns us into an animated geometry root. r=mstange 2015-11-01 11:45:40 -06:00
Carsten "Tomcat" Book
2fe6ba353f Backed out changeset d5d68c0258f3 (bug 1201330)
--HG--
extra : rebase_source : 2b98ebd0dd0ccd247b50e4b45210137622ef4441
2015-10-30 14:44:06 +01:00