d7d7811efab18a10e55dafd4a032dc7fa7f49994: Bug 1477512: Part 0 - Remove unused nsThread::mEventObservers array. r=erahm
Kris Maglione <maglione.k@gmail.com> - Sat, 21 Jul 2018 13:21:12 -0700 - rev 427929
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1477512: Part 0 - Remove unused nsThread::mEventObservers array. r=erahm This was moved out of nsThread in bug 1350432, but some dead code was left behind. MozReview-Commit-ID: BOhykHyIEPp
37435502b7c0301bbcf9d161a0700a92b9e0d929: Bug 1475836: Clear cached callback pref when deleting branch. r=njn
Kris Maglione <maglione.k@gmail.com> - Fri, 20 Jul 2018 11:57:37 -0700 - rev 427928
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1475836: Clear cached callback pref when deleting branch. r=njn MozReview-Commit-ID: C60kGkoFBL8
d68c6c56856cdd0306fc7c3491438d301526db51: Bug 1477254: Assert that varcache prefs match pref values at content process startup. r=njn
Kris Maglione <maglione.k@gmail.com> - Mon, 23 Jul 2018 17:32:54 -0700 - rev 427927
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1477254: Assert that varcache prefs match pref values at content process startup. r=njn In order to avoid the overhead of doing a full pref lookup for every static var cache at content process startup, we currently assume that the default value of any static varcache pref will always match the default value of its database entry (as long as the pref isn't locked). This lets us only perform lookups for preferences which have a user value, or are locked. If the default values of those preferences are changed in a bundled preference file, though, the varcache value will be correct in the parent process, but not in child processes. Since this is an easy mistake to make, we should assert that it doesn't happen. Note: This change only affects applications which use e10s. Applications like Thunderbird can still override default values of any static pref with impunity. Repacks and distributors can only do so by changing user values or locking the preference after the change (which is the standard practice for enterprise deployments). MozReview-Commit-ID: JMHQBrp9HN
aa9a8f18e98f930a3d8359565eef02f3f6efc5f9: Bug 1477579: Part 3 - Avoid duplicating static strings in category manager entries. r=froydnj
Kris Maglione <maglione.k@gmail.com> - Mon, 23 Jul 2018 17:41:06 -0700 - rev 427926
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1477579: Part 3 - Avoid duplicating static strings in category manager entries. r=froydnj Much like the component manager, many of the strings that we use for category manager entries are statically allocated. There's no need to duplicate these strings. This patch changes the category manager APIs to take nsACStrings rather than raw pointers, and to pass literal nsCStrings when we know we have a literal string to begin with. When adding the category entry, it then skips making copies of any strings with the LITERAL flag. MozReview-Commit-ID: EJEcYSdNMWs *** amend-catman
5fb0b7746a5d56563b471e3061ccca124ea45485: Bug 1477579: Part 2 - Use nsID* rather than nsID for mFactories hash keys. r=froydnj
Kris Maglione <maglione.k@gmail.com> - Sat, 21 Jul 2018 19:09:25 -0700 - rev 427925
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1477579: Part 2 - Use nsID* rather than nsID for mFactories hash keys. r=froydnj Our factory registrations already require that we store nsID pointers, which we generally handle by using pointers to static data, or arena allocating a copy of a dynamic ID. Since we already have viable pointers to these IDs, there's no reason to store an entire second copy for our hash key. We can use the pointer, instead, which saves 16 bytes per entry. MozReview-Commit-ID: 6MgKrXRSHv4
8359f8fe418419c50ab0ed93496e7445b570ba9f: Bug 1477579: Part 1 - Use literal strings for statically registered contract ID keys. r=froydnj
Kris Maglione <maglione.k@gmail.com> - Sun, 22 Jul 2018 10:54:56 -0700 - rev 427924
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1477579: Part 1 - Use literal strings for statically registered contract ID keys. r=froydnj Most of our components are static, and registered using literal C strings. For those components, we currently use a nsDependentCString as a key when creating a hash entry, which leads to an unnecessary duplication. Using literal CStrings instead avoids the duplication. MozReview-Commit-ID: 5DOUF8ZQMlh
232216b44fce618ff638e4f50a8790011fab47e5: Backed out 2 changesets (bug 1477254, bug 1475836) for bustages in builds/worker/workspace/build/src/modules/libpref/Preferences.cpp:4791:1 on a CLOSED TREE
shindli <shindli@mozilla.com> - Tue, 24 Jul 2018 03:52:40 +0300 - rev 427923
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Backed out 2 changesets (bug 1477254, bug 1475836) for bustages in builds/worker/workspace/build/src/modules/libpref/Preferences.cpp:4791:1 on a CLOSED TREE Backed out changeset 1b389b00030e (bug 1477254) Backed out changeset 3b9846062755 (bug 1475836)
9ba70b8ff0f755d5575333a71267805e0ece02f6: Bug 1473135: Stop using ReentrantMonitor for string bundle mutexes. r=gandalf
Kris Maglione <maglione.k@gmail.com> - Fri, 20 Jul 2018 17:20:00 -0700 - rev 427922
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1473135: Stop using ReentrantMonitor for string bundle mutexes. r=gandalf MozReview-Commit-ID: HUfb47Eei7P
2eaa478b89791ff48402c18a9bab985b13cc4f36: Bug 1477129: Part 3 - Re-enable e10s on FreeBSD. r=froydnj
Kris Maglione <maglione.k@gmail.com> - Thu, 19 Jul 2018 18:18:27 -0700 - rev 427921
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1477129: Part 3 - Re-enable e10s on FreeBSD. r=froydnj Backed out changeset 197fcba26a38 MozReview-Commit-ID: 4OOmP91hKXQ
aafb84f71d5c4e0b3b5742daf1c7f10a1f4a3e68: Bug 1477129: Part 2 - Fix SharedMap blob handling, and add tests. r=erahm
Kris Maglione <maglione.k@gmail.com> - Fri, 20 Jul 2018 14:44:00 -0700 - rev 427920
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1477129: Part 2 - Fix SharedMap blob handling, and add tests. r=erahm We were previously failing to send blobs to new content processes, which was a problem for those processes. But we were also attempting to extract blobs for new entries that we were serializing after we'd extracted their structured clone data, and their blob array had been thrown away (which was a problem for all processes). This patch fixes both problems. MozReview-Commit-ID: 3qbAmUTA85g
6a0d1dfe5d423d57c0de2ebf040d15c97bda5316: Bug 1477129: Part 1 - Handle BlobImpls when move constructing StructuredCloneData. r=baku
Kris Maglione <maglione.k@gmail.com> - Thu, 19 Jul 2018 18:14:13 -0700 - rev 427919
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1477129: Part 1 - Handle BlobImpls when move constructing StructuredCloneData. r=baku MozReview-Commit-ID: F1pm9qTpGCV
1b389b00030ed38cdb3543aa5d1a67795be47565: Bug 1477254: Assert that varcache prefs match pref values at content process startup. r=njn
Kris Maglione <maglione.k@gmail.com> - Mon, 23 Jul 2018 17:32:54 -0700 - rev 427918
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1477254: Assert that varcache prefs match pref values at content process startup. r=njn In order to avoid the overhead of doing a full pref lookup for every static var cache at content process startup, we currently assume that the default value of any static varcache pref will always match the default value of its database entry (as long as the pref isn't locked). This lets us only perform lookups for preferences which have a user value, or are locked. If the default values of those preferences are changed in a bundled preference file, though, the varcache value will be correct in the parent process, but not in child processes. Since this is an easy mistake to make, we should assert that it doesn't happen. Note: This change only affects applications which use e10s. Applications like Thunderbird can still override default values of any static pref with impunity. Repacks and distributors can only do so by changing user values or locking the preference after the change (which is the standard practice for enterprise deployments). MozReview-Commit-ID: JMHQBrp9HN
3b9846062755b05de6729a84aa2b0acf6ddfd0bc: Bug 1475836: Clear cached callback pref when deleting branch. r=njn
Kris Maglione <maglione.k@gmail.com> - Fri, 20 Jul 2018 11:57:37 -0700 - rev 427917
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1475836: Clear cached callback pref when deleting branch. r=njn MozReview-Commit-ID: C60kGkoFBL8
c3c7401b4277de9d310c22735d3b2695fd020490: Bug 1468053: Disable a workaround for an Apple bug described in bug 378645 involving popup windows that was fixed by Apple. r=mstange
Stephen A Pohl <spohl.mozilla.bugs@gmail.com> - Mon, 23 Jul 2018 20:20:58 -0400 - rev 427916
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1468053: Disable a workaround for an Apple bug described in bug 378645 involving popup windows that was fixed by Apple. r=mstange
53033f3d4fbd89a7010d70eace2df87d2f7ba2a1: Backout 0df7fb702b93 for slightly inaccurate commit message. r=me
Stephen A Pohl <spohl.mozilla.bugs@gmail.com> - Mon, 23 Jul 2018 20:19:19 -0400 - rev 427915
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Backout 0df7fb702b93 for slightly inaccurate commit message. r=me
0df7fb702b9318710488853822f179001e59f5b3: Bug 1468053: Disable a workaround on macOS 10.14+ for an Apple bug involving popup windows that was fixed by Apple in macOS 10.14. r=mstange
Stephen A Pohl <spohl.mozilla.bugs@gmail.com> - Mon, 23 Jul 2018 20:16:15 -0400 - rev 427914
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1468053: Disable a workaround on macOS 10.14+ for an Apple bug involving popup windows that was fixed by Apple in macOS 10.14. r=mstange
74e441c3bf95c9a34aace739f36f7314dcc6c602: Bug 1474280 - Part 2: carry timestamp when redirects; r=valentin
Liang-Heng Chen <xeonchen@gmail.com> - Mon, 23 Jul 2018 05:29:00 +0300 - rev 427913
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1474280 - Part 2: carry timestamp when redirects; r=valentin
318799faa5fe71498201a85a761c5f303447bb6f: Bug 1474280 - Part 1: Cancel timed-out trackers; r=mayhemer
Liang-Heng Chen <xeonchen@gmail.com> - Mon, 23 Jul 2018 05:28:00 +0300 - rev 427912
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1474280 - Part 1: Cancel timed-out trackers; r=mayhemer
65ea32c6a5be230bc1e7c5f5a013bfd4b68d14ef: Bug 1476124 - Implement enough ARM64 Ion code to run a simple script. r=tcampbell
Sean Stangl <sean.stangl@gmail.com> - Mon, 23 Jul 2018 15:15:00 +0300 - rev 427911
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1476124 - Implement enough ARM64 Ion code to run a simple script. r=tcampbell
16f5d9f58faab689ee88ebf568c5c504b2ac72ba: Bug 1477375 Ensure we actually disable all CacheIR Stubs when requested with --cache-ir-stubs=off. r=evilpie
Matthew Gaudet <mgaudet@mozilla.com> - Fri, 20 Jul 2018 15:47:40 -0400 - rev 427910
Push 34320 by rgurzau@mozilla.com at Tue, 24 Jul 2018 09:50:07 +0000
Bug 1477375 Ensure we actually disable all CacheIR Stubs when requested with --cache-ir-stubs=off. r=evilpie
(0) -300000 -100000 -30000 -10000 -3000 -1000 -300 -100 -50 -20 +20 +50 +100 +300 +1000 +3000 +10000 +30000 +100000 tip