988f3acb9d9809cb2600a4670998da4e2f55c5de: Remove NS_SCROLLFRAME_INVALIDATE_CONTENTS_ON_SCROLL as it doesn't appear to be needed any more
Matt Woodrow <mwoodrow@mozilla.com> - Wed, 20 Sep 2017 13:12:37 +1200 - rev 685648
Push 86010 by bmo:ethlin@mozilla.com at Wed, 25 Oct 2017 00:44:42 +0000
Remove NS_SCROLLFRAME_INVALIDATE_CONTENTS_ON_SCROLL as it doesn't appear to be needed any more
8a40c88d1757b218f163367f1c80176a3747d8ea: Move EnterPresShell calls to be closer to where we actually build the display list
Matt Woodrow <mwoodrow@mozilla.com> - Wed, 20 Sep 2017 13:12:17 +1200 - rev 685647
Push 86010 by bmo:ethlin@mozilla.com at Wed, 25 Oct 2017 00:44:42 +0000
Move EnterPresShell calls to be closer to where we actually build the display list
4c7a091d24890f5aedbb337959f4a90e8b608889: Switch to just marking display items as invalid for animation changes to avoid a stylo bug
Matt Woodrow <mwoodrow@mozilla.com> - Wed, 20 Sep 2017 13:11:52 +1200 - rev 685646
Push 86010 by bmo:ethlin@mozilla.com at Wed, 25 Oct 2017 00:44:42 +0000
Switch to just marking display items as invalid for animation changes to avoid a stylo bug
cac5687ebed2550a24361e459090d4827e041bd3: Move retained display list code into its own file
Matt Woodrow <mwoodrow@mozilla.com> - Tue, 19 Sep 2017 16:13:21 +1200 - rev 685645
Push 86010 by bmo:ethlin@mozilla.com at Wed, 25 Oct 2017 00:44:42 +0000
Move retained display list code into its own file
2b1b0c71469c97de69f60c87759e0ea15156a05a: Merge with mozilla-central
Matt Woodrow <mwoodrow@mozilla.com> - Tue, 19 Sep 2017 13:38:59 +1200 - rev 685644
Push 86010 by bmo:ethlin@mozilla.com at Wed, 25 Oct 2017 00:44:42 +0000
Merge with mozilla-central
fa6be791556eff0f906715467f33bff689d78910: Add an option to use power-of-two buckets for display list builder arena allocations
Matt Woodrow <mwoodrow@mozilla.com> - Tue, 19 Sep 2017 11:36:05 +1200 - rev 685643
Push 86010 by bmo:ethlin@mozilla.com at Wed, 25 Oct 2017 00:44:42 +0000
Add an option to use power-of-two buckets for display list builder arena allocations
43120be47129dcb53219b7476e2ed9a4c4f31267: Don't invalidate display items when we schedule a paint for APZ focus updates
Matt Woodrow <mwoodrow@mozilla.com> - Tue, 19 Sep 2017 11:35:32 +1200 - rev 685642
Push 86010 by bmo:ethlin@mozilla.com at Wed, 25 Oct 2017 00:44:42 +0000
Don't invalidate display items when we schedule a paint for APZ focus updates
224aa41383ca2b5483a5eec16823877ef4ea5fa7: Add area to the debug overlay when we rebuild the entire scrollframe
Matt Woodrow <mwoodrow@mozilla.com> - Tue, 19 Sep 2017 11:35:06 +1200 - rev 685641
Push 86010 by bmo:ethlin@mozilla.com at Wed, 25 Oct 2017 00:44:42 +0000
Add area to the debug overlay when we rebuild the entire scrollframe
3e66c3bb56881547abf5c8e0fcc4190d13fa392a: Add logging for ComputeRebuildRegion
Matt Woodrow <mwoodrow@mozilla.com> - Tue, 19 Sep 2017 11:34:31 +1200 - rev 685640
Push 86010 by bmo:ethlin@mozilla.com at Wed, 25 Oct 2017 00:44:42 +0000
Add logging for ComputeRebuildRegion
4dbe9ae4248c0f5205342494ba4031539bc1e991: Remove duplicated temporary items loop
Matt Woodrow <mwoodrow@mozilla.com> - Fri, 15 Sep 2017 15:04:32 -0400 - rev 685639
Push 86010 by bmo:ethlin@mozilla.com at Wed, 25 Oct 2017 00:44:42 +0000
Remove duplicated temporary items loop
084f83687e37a34e8186ab2b7bbb6a20a7a0eb2f: Bug 1399795 - Recompute nsDisplayTransform bounds for 3D when updating the item bounds. r=mattwoodrow
Ethan Lin <ethlin@mozilla.com> - Fri, 15 Sep 2017 15:14:33 +0800 - rev 685638
Push 86010 by bmo:ethlin@mozilla.com at Wed, 25 Oct 2017 00:44:42 +0000
Bug 1399795 - Recompute nsDisplayTransform bounds for 3D when updating the item bounds. r=mattwoodrow MozReview-Commit-ID: ZYjICI1J1i
c87a3cd41bedef3e3186e0c4a12c49827386af74: Avoid calling EndFrame() on deleted nsDisplayListBuilder
Miko Mynttinen <mikokm@gmail.com> - Thu, 14 Sep 2017 14:50:16 -0400 - rev 685637
Push 86010 by bmo:ethlin@mozilla.com at Wed, 25 Oct 2017 00:44:42 +0000
Avoid calling EndFrame() on deleted nsDisplayListBuilder
145dbc4727d3bb2d964765a1738ecdfac782a24a: Fix case where we could end up missing a call to EndFrame
Matt Woodrow <mwoodrow@mozilla.com> - Thu, 14 Sep 2017 13:15:07 -0400 - rev 685636
Push 86010 by bmo:ethlin@mozilla.com at Wed, 25 Oct 2017 00:44:42 +0000
Fix case where we could end up missing a call to EndFrame
1f3de1eeb35499aa75d48ae786945b87452b8eb3: Remove some unnecessary differences between us and m-c
Matt Woodrow <mwoodrow@mozilla.com> - Thu, 14 Sep 2017 11:03:29 -0400 - rev 685635
Push 86010 by bmo:ethlin@mozilla.com at Wed, 25 Oct 2017 00:44:42 +0000
Remove some unnecessary differences between us and m-c
10d6b57d34593860a58812fdab8ec0d4c815b114: Merge with mozilla-central
Matt Woodrow <mwoodrow@mozilla.com> - Thu, 14 Sep 2017 10:43:56 -0400 - rev 685634
Push 86010 by bmo:ethlin@mozilla.com at Wed, 25 Oct 2017 00:44:42 +0000
Merge with mozilla-central
37d8a9fe05cf7338fcb280341ffc32ea7ba44d9c: Bug 1411462 - Add test for `configure --help`; r?glandium draft
Gregory Szorc <gps@mozilla.com> - Tue, 24 Oct 2017 17:10:55 -0700 - rev 685633
Push 86009 by gszorc@mozilla.com at Wed, 25 Oct 2017 00:26:05 +0000
Bug 1411462 - Add test for `configure --help`; r?glandium We add a simple test that `configure --help` works. We do this via a cram test. I added the tests to build/tests because I'm not sure where else they should go. This test uncovers a few interesting things: 1) piping `./configure --help` to `head` directly causes a Python traceback (presumably due to the pipe disappearing once N lines have been read) 2) "checking for vcs source checkout" is printed for --help 3) It is printed twice (!!) I reckon these can be follow-up improvements. MozReview-Commit-ID: 9zQ5X8ulTkc
8d758b677b5314cc154b305653b9527fce8eedd5: Bug 1411462 - Make ./configure --help work again; r?glandium draft
Gregory Szorc <gps@mozilla.com> - Tue, 24 Oct 2017 12:56:10 -0700 - rev 685632
Push 86009 by gszorc@mozilla.com at Wed, 25 Oct 2017 00:26:05 +0000
Bug 1411462 - Make ./configure --help work again; r?glandium It stopped working due to 370a4a2a7898 (bug 1405982). MozReview-Commit-ID: 2gCmAAsE0qW
59b66eb91cbfb46b1ccdd57db53333743fa2682a: Bug 1362617 - Generalize MOZ_CHROME_MULTILOCALE to work for browser as well. r?mshal,pike draft
Zibi Braniecki <zbraniecki@mozilla.com> - Mon, 23 Oct 2017 08:04:24 -0700 - rev 685631
Push 86008 by bmo:gandalf@aviary.pl at Wed, 25 Oct 2017 00:13:45 +0000
Bug 1362617 - Generalize MOZ_CHROME_MULTILOCALE to work for browser as well. r?mshal,pike MozReview-Commit-ID: 80ldQPhNyi7
295161d8b756216e969923b713bd5862d4cdaa80: Bug 1411081 - Use opt-level=1 for rustc for default builds; r?glandium draft
Gregory Szorc <gps@mozilla.com> - Tue, 24 Oct 2017 16:48:58 -0700 - rev 685630
Push 86007 by gszorc@mozilla.com at Tue, 24 Oct 2017 23:50:58 +0000
Bug 1411081 - Use opt-level=1 for rustc for default builds; r?glandium As the data in the bug shows, the current default of opt-level=2 is several minutes slower at compiling than opt-level=1. This slows down builds significantly and the added benefits of running opt-level=2 for local development can't be justified for the common/default case. This commit changes the default for local builds from opt-level=2 to opt-level=1. --enable-release (what we use for builds shipped to users) will imply opt-level=2. --enable-optimize (the default) will use opt-level=1, and --disable-optimize will use opt-level=0. The RUSTC_OPT_LEVEL environment variable in mozconfigs can be used to set an explicit opt-level leval, regardless of what other configure options are set. This includes the other potential values, "s" and "z." MozReview-Commit-ID: 67KX5qScnFc
88a3dee5f1792dd4aa020c39970ae7ca26609031: Bug 1411081 - Derive Rust compiler flags in configure; r?glandium draft
Gregory Szorc <gps@mozilla.com> - Tue, 24 Oct 2017 14:42:01 -0700 - rev 685629
Push 86007 by gszorc@mozilla.com at Tue, 24 Oct 2017 23:50:58 +0000
Bug 1411081 - Derive Rust compiler flags in configure; r?glandium Before this commit, RUSTFLAGS was derived in rules.mk by consulting various variables set by configure. It isn't clear to me why things are implemented this way. We don't appear to have moz.build level overrides for Rust compiler flags. So there doesn't appear to be a compelling reason why we can't derive these values in configure. So, this commit ports the code for deriving default RUSTFLAGS from rules.mk to toolchain.configure. The behavior should be backwards compatible. The main difference is that we'll always pass some compiler options on the command line when we weren't before. Before, we either inherited the defaults from Cargo.toml or from rustc itself. This change could break some workflows. TBH I'm not sure what the workflow requirements are. They aren't documented in comments. Hopefully it is safe to assume that everyone uses `mach build` and the compiler settings derived from the build system. If people really need control, they can overrite the newly introduced MOZ_RUST_DEFAULT_FLAGS subst variable via an environment variable. This commit stops short of removing the definition of these per-profile options from Cargo.toml. The Cargo.toml values are overridden by the build system, so it might be worth removing them so people don't think they do anything meaningful. MozReview-Commit-ID: JhAE9Qlo8SK
(0) -300000 -100000 -30000 -10000 -3000 -1000 -300 -100 -50 -20 +20 +50 +100 +300 +1000 +3000 +10000 +30000 +100000 tip