OSDN Git Service
Winson Chung [Thu, 11 Nov 2021 19:58:50 +0000 (19:58 +0000)]
[automerger skipped] Add mechanism for a task's windows to be trusted overlays (SF) am:
7605fb4273 am:
cac2aa7517 -s ours am:
10b892cda2 -s ours am:
d3962fdc82 -s ours
am skip reason: Merged-In Id92ccb087bd0d8dbaeeef3ba50b67fe015e53db8 with SHA-1
7605fb4273 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975348
Change-Id: I45c767eb9cce3b9f94fb88a7b48c231e3a501a20
Chris Ye [Thu, 11 Nov 2021 19:58:49 +0000 (19:58 +0000)]
[automerger skipped] Change InputWindowInfo::isTrustedOverlay() to be permission and flag based. am:
39bc6117dd am:
020dd044a6 -s ours am:
5d53e37fe0 -s ours am:
6baa59fcbf -s ours
am skip reason: Merged-In I58cf9f38c496e0ae8b2193dca45c0805e831bc9e with SHA-1
39bc6117dd is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975347
Change-Id: I1d1a87381dff68329f1d768f4a7274f1510a6201
Winson Chung [Thu, 11 Nov 2021 19:40:52 +0000 (19:40 +0000)]
[automerger skipped] Add mechanism for a task's windows to be trusted overlays (SF) am:
7605fb4273 am:
cac2aa7517 -s ours am:
10b892cda2 -s ours
am skip reason: Merged-In Id92ccb087bd0d8dbaeeef3ba50b67fe015e53db8 with SHA-1
7605fb4273 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975348
Change-Id: Ie083bcf7b9c50a6f43c5937896cf19c47a5ca6b2
Chris Ye [Thu, 11 Nov 2021 19:40:51 +0000 (19:40 +0000)]
[automerger skipped] Change InputWindowInfo::isTrustedOverlay() to be permission and flag based. am:
39bc6117dd am:
020dd044a6 -s ours am:
5d53e37fe0 -s ours
am skip reason: Merged-In I58cf9f38c496e0ae8b2193dca45c0805e831bc9e with SHA-1
39bc6117dd is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975347
Change-Id: I7fd0f5006851287307c7e1059ad5e2893c5c19dc
Winson Chung [Thu, 11 Nov 2021 19:21:50 +0000 (19:21 +0000)]
[automerger skipped] Add mechanism for a task's windows to be trusted overlays (SF) am:
7605fb4273 am:
cac2aa7517 -s ours
am skip reason: Merged-In Id92ccb087bd0d8dbaeeef3ba50b67fe015e53db8 with SHA-1
7605fb4273 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975348
Change-Id: Iddcca623eda377c55b3f2106b5122be428936f1f
Chris Ye [Thu, 11 Nov 2021 19:21:47 +0000 (19:21 +0000)]
[automerger skipped] Change InputWindowInfo::isTrustedOverlay() to be permission and flag based. am:
39bc6117dd am:
020dd044a6 -s ours
am skip reason: Merged-In I58cf9f38c496e0ae8b2193dca45c0805e831bc9e with SHA-1
39bc6117dd is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975347
Change-Id: Iac9f5d211d7ab954ab94d8824917b621044f3d54
Siarhei Vishniakou [Wed, 10 Nov 2021 04:07:40 +0000 (04:07 +0000)]
[automerger skipped] Check if the window is partially obscured for slippery enters am:
6e689ffe3f am:
224d576c0b am:
e26b6061f8 -s ours am:
eb59e4756d -s ours am:
8c83865531 -s ours am:
4986d714bf -s ours am:
67efbd2880 -s ours
am skip reason: Merged-In I93d9681479f41244ffed4b1f88cceb69be71adf2 with SHA-1
d8c6ef2138 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16185286
Change-Id: Iff4bd828b9e97162721b59a8cedb4cea693b35ef
Siarhei Vishniakou [Wed, 10 Nov 2021 03:51:02 +0000 (03:51 +0000)]
[automerger skipped] Check if the window is partially obscured for slippery enters am:
6e689ffe3f am:
224d576c0b am:
e26b6061f8 -s ours am:
eb59e4756d -s ours am:
8c83865531 -s ours am:
4986d714bf -s ours
am skip reason: Merged-In I93d9681479f41244ffed4b1f88cceb69be71adf2 with SHA-1
d8c6ef2138 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16185286
Change-Id: I00e050a7cc2d62aebb628cb2daa19596c41d6171
Siarhei Vishniakou [Wed, 10 Nov 2021 03:31:46 +0000 (03:31 +0000)]
[automerger skipped] Check if the window is partially obscured for slippery enters am:
6e689ffe3f am:
224d576c0b am:
e26b6061f8 -s ours am:
eb59e4756d -s ours am:
8c83865531 -s ours
am skip reason: Merged-In I93d9681479f41244ffed4b1f88cceb69be71adf2 with SHA-1
d8c6ef2138 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16185286
Change-Id: Ie6d4c961e9f95da4f0acbe6cd4665e1fc757f0b7
Siarhei Vishniakou [Wed, 10 Nov 2021 03:19:04 +0000 (03:19 +0000)]
[automerger skipped] Check if the window is partially obscured for slippery enters am:
6e689ffe3f am:
224d576c0b am:
e26b6061f8 -s ours am:
eb59e4756d -s ours
am skip reason: Merged-In I93d9681479f41244ffed4b1f88cceb69be71adf2 with SHA-1
d8c6ef2138 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16185286
Change-Id: Ic4e82083fcc577c808292d6ae76c5d4e072e9c89
Siarhei Vishniakou [Wed, 10 Nov 2021 03:06:07 +0000 (03:06 +0000)]
[automerger skipped] Check if the window is partially obscured for slippery enters am:
6e689ffe3f am:
224d576c0b am:
e26b6061f8 -s ours
am skip reason: Merged-In I93d9681479f41244ffed4b1f88cceb69be71adf2 with SHA-1
d8c6ef2138 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16185286
Change-Id: I3fa807e2a5e61dfad45b5e27e3116e9e64338c8b
Siarhei Vishniakou [Wed, 10 Nov 2021 02:55:28 +0000 (02:55 +0000)]
Check if the window is partially obscured for slippery enters am:
6e689ffe3f am:
224d576c0b
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16185286
Change-Id: I49fb56b83a8799ac6a380b269bca06adc5289aba
Siarhei Vishniakou [Wed, 10 Nov 2021 02:44:01 +0000 (02:44 +0000)]
Check if the window is partially obscured for slippery enters am:
6e689ffe3f
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16185286
Change-Id: I6645db77a493d55aacc8f645db625ea5f89a8d36
TreeHugger Robot [Wed, 10 Nov 2021 01:26:42 +0000 (01:26 +0000)]
[automerger skipped] Merge "Check if the window is partially obscured for slippery enters" into rvc-dev am:
b27497c4bf am:
1158690bb7 -s ours am:
56b929e3e7 -s ours am:
17ec686b0c -s ours
am skip reason: Merged-In I93d9681479f41244ffed4b1f88cceb69be71adf2 with SHA-1
870ececa8d is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16194632
Change-Id: If140b949b86057c71f482a04ace76f70e1cebe6a
TreeHugger Robot [Wed, 10 Nov 2021 01:06:24 +0000 (01:06 +0000)]
[automerger skipped] Merge "Check if the window is partially obscured for slippery enters" into rvc-dev am:
b27497c4bf am:
1158690bb7 -s ours am:
56b929e3e7 -s ours
am skip reason: Merged-In I93d9681479f41244ffed4b1f88cceb69be71adf2 with SHA-1
870ececa8d is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16194632
Change-Id: Iff92d8b7d7cae9f6d4c4527f2d65f89742fef26b
Winson Chung [Wed, 10 Nov 2021 00:44:07 +0000 (00:44 +0000)]
Merge changes from topic "trusted_overlay_backport" into rvc-dev am:
294e7b87e3 am:
e93bd77429 am:
f13c62852e am:
7c2f24b284
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975349
Change-Id: I28025893f506121b09cd83d0fec20ad77d423c18
Robin Lee [Wed, 10 Nov 2021 00:44:06 +0000 (00:44 +0000)]
[automerger skipped] DO NOT MERGE Initialize DrawingState::trustedOverlay to false in constructor am:
41f48c7b10 am:
cf8e45638d -s ours am:
6b44d942a9 -s ours am:
4d6cce1f13 -s ours
am skip reason: subject contains skip directive
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975349
Change-Id: I1d7202453bc6a95b29be0658935a37e291b258c6
Winson Chung [Wed, 10 Nov 2021 00:44:06 +0000 (00:44 +0000)]
[automerger skipped] Add mechanism for a task's windows to be trusted overlays (SF) am:
7605fb4273 am:
4df1658eda -s ours am:
ede8f4d536 -s ours am:
c4b83bacd6 -s ours
am skip reason: Merged-In Id92ccb087bd0d8dbaeeef3ba50b67fe015e53db8 with SHA-1
a30f7c9974 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975348
Change-Id: I85e1f2d7c00ad4bbf9bdb0b5e8d271697de1d118
Chris Ye [Wed, 10 Nov 2021 00:44:05 +0000 (00:44 +0000)]
[automerger skipped] Change InputWindowInfo::isTrustedOverlay() to be permission and flag based. am:
39bc6117dd am:
8bec11e493 -s ours am:
139309f468 -s ours am:
aeb77c0bee -s ours
am skip reason: Merged-In I58cf9f38c496e0ae8b2193dca45c0805e831bc9e with SHA-1
fcdff3e735 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975347
Change-Id: If1fa8889d6d3de49830dd28eff61b270ec4e3c5c
Steven Moreland [Wed, 10 Nov 2021 00:43:41 +0000 (00:43 +0000)]
[automerger skipped] avoid extra release of unowned objects in Parcel error path am:
d668098e47 am:
41db1b0e7c -s ours am:
c4fe8c2478 -s ours am:
5c32ca5dfc -s ours
am skip reason: Merged-In I34411675687cb3d18bffa082984ebdf308e1c1a6 with SHA-1
d668098e47 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16177016
Change-Id: I89938a9f28647c2436e7e397c45841c04fcec4a0
Steven Moreland [Wed, 10 Nov 2021 00:43:14 +0000 (00:43 +0000)]
[automerger skipped] avoid extra release of unowned objects in Parcel error path am:
d668098e47 am:
d471beffd1 -s ours am:
57380b88ba -s ours am:
fb6c7fc94f -s ours
am skip reason: Merged-In I34411675687cb3d18bffa082984ebdf308e1c1a6 with SHA-1
7c8497e012 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16177016
Change-Id: Ic99a1fd0117a376cbd1be87ec622807819027a5d
TreeHugger Robot [Wed, 10 Nov 2021 00:41:53 +0000 (00:41 +0000)]
[automerger skipped] Merge "Check if the window is partially obscured for slippery enters" into rvc-dev am:
b27497c4bf am:
1158690bb7 -s ours
am skip reason: Merged-In I93d9681479f41244ffed4b1f88cceb69be71adf2 with SHA-1
870ececa8d is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16194632
Change-Id: I58801fbec843c7c1e09fa9f0a51eed9c8225c839
Winson Chung [Wed, 10 Nov 2021 00:19:22 +0000 (00:19 +0000)]
Merge changes from topic "trusted_overlay_backport" into rvc-dev am:
294e7b87e3 am:
e93bd77429 am:
f13c62852e
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975349
Change-Id: Iff1f45324e088239c71593777e4d521ea6f4937a
Robin Lee [Wed, 10 Nov 2021 00:19:22 +0000 (00:19 +0000)]
[automerger skipped] DO NOT MERGE Initialize DrawingState::trustedOverlay to false in constructor am:
41f48c7b10 am:
cf8e45638d -s ours am:
6b44d942a9 -s ours
am skip reason: subject contains skip directive
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975349
Change-Id: I01a1a627578e8c95f6ffe8da07cc04a04129a46e
Winson Chung [Wed, 10 Nov 2021 00:19:21 +0000 (00:19 +0000)]
[automerger skipped] Add mechanism for a task's windows to be trusted overlays (SF) am:
7605fb4273 am:
4df1658eda -s ours am:
ede8f4d536 -s ours
am skip reason: Merged-In Id92ccb087bd0d8dbaeeef3ba50b67fe015e53db8 with SHA-1
a30f7c9974 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975348
Change-Id: I6f9bb604284a68e4feeefca0e384b2f6f600d850
Chris Ye [Wed, 10 Nov 2021 00:19:20 +0000 (00:19 +0000)]
[automerger skipped] Change InputWindowInfo::isTrustedOverlay() to be permission and flag based. am:
39bc6117dd am:
8bec11e493 -s ours am:
139309f468 -s ours
am skip reason: Merged-In I58cf9f38c496e0ae8b2193dca45c0805e831bc9e with SHA-1
fcdff3e735 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975347
Change-Id: I1b2cf100831a8535b497ae4fa685231ba9876b1b
Steven Moreland [Wed, 10 Nov 2021 00:18:58 +0000 (00:18 +0000)]
[automerger skipped] avoid extra release of unowned objects in Parcel error path am:
d668098e47 am:
41db1b0e7c -s ours am:
c4fe8c2478 -s ours
am skip reason: Merged-In I34411675687cb3d18bffa082984ebdf308e1c1a6 with SHA-1
d668098e47 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16177016
Change-Id: I904d09f651f803ac1839cbab27b9a91936e8e133
Steven Moreland [Wed, 10 Nov 2021 00:18:41 +0000 (00:18 +0000)]
[automerger skipped] avoid extra release of unowned objects in Parcel error path am:
d668098e47 am:
d471beffd1 -s ours am:
57380b88ba -s ours
am skip reason: Merged-In I34411675687cb3d18bffa082984ebdf308e1c1a6 with SHA-1
7c8497e012 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16177016
Change-Id: I8fd02cbeeae4a7e90fce6f3f718ad641bc4ce875
TreeHugger Robot [Wed, 10 Nov 2021 00:18:34 +0000 (00:18 +0000)]
Merge "Check if the window is partially obscured for slippery enters" into rvc-dev am:
b27497c4bf
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16194632
Change-Id: I597107cf0a85b6dbc73dcc4092d3c077d678e661
TreeHugger Robot [Tue, 9 Nov 2021 23:53:32 +0000 (23:53 +0000)]
Merge "Check if the window is partially obscured for slippery enters" into rvc-dev
Winson Chung [Tue, 9 Nov 2021 23:42:36 +0000 (23:42 +0000)]
Merge changes from topic "trusted_overlay_backport" into rvc-dev am:
294e7b87e3 am:
e93bd77429
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975349
Change-Id: I48c83881683865b1b9d1de1d4a63531a84112cf6
Robin Lee [Tue, 9 Nov 2021 23:42:34 +0000 (23:42 +0000)]
[automerger skipped] DO NOT MERGE Initialize DrawingState::trustedOverlay to false in constructor am:
41f48c7b10 am:
cf8e45638d -s ours
am skip reason: subject contains skip directive
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975349
Change-Id: I5359f45577d476b0fe0d14367b046bee00e64f7e
Winson Chung [Tue, 9 Nov 2021 23:42:31 +0000 (23:42 +0000)]
[automerger skipped] Add mechanism for a task's windows to be trusted overlays (SF) am:
7605fb4273 am:
4df1658eda -s ours
am skip reason: Merged-In Id92ccb087bd0d8dbaeeef3ba50b67fe015e53db8 with SHA-1
a30f7c9974 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975348
Change-Id: Ifa92f0a463c513896f4ffa417a65002d89847be8
Chris Ye [Tue, 9 Nov 2021 23:42:30 +0000 (23:42 +0000)]
[automerger skipped] Change InputWindowInfo::isTrustedOverlay() to be permission and flag based. am:
39bc6117dd am:
8bec11e493 -s ours
am skip reason: Merged-In I58cf9f38c496e0ae8b2193dca45c0805e831bc9e with SHA-1
fcdff3e735 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975347
Change-Id: I19e8bd2cdbacdd8f15c400e047f744927934b52e
Steven Moreland [Tue, 9 Nov 2021 23:41:50 +0000 (23:41 +0000)]
[automerger skipped] avoid extra release of unowned objects in Parcel error path am:
d668098e47 am:
41db1b0e7c -s ours
am skip reason: Merged-In I34411675687cb3d18bffa082984ebdf308e1c1a6 with SHA-1
d668098e47 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16177016
Change-Id: I4eb1f071d3e5556e0a34a3b64c7e7b11f6a391bc
Steven Moreland [Tue, 9 Nov 2021 23:40:51 +0000 (23:40 +0000)]
[automerger skipped] avoid extra release of unowned objects in Parcel error path am:
d668098e47 am:
d471beffd1 -s ours
am skip reason: Merged-In I34411675687cb3d18bffa082984ebdf308e1c1a6 with SHA-1
7c8497e012 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16177016
Change-Id: I3c25c7887fefde636fc6b2e29c2baca0ef85d52d
Winson Chung [Tue, 9 Nov 2021 23:21:01 +0000 (23:21 +0000)]
Merge changes from topic "trusted_overlay_backport" into rvc-dev am:
294e7b87e3
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975349
Change-Id: I82ff44bf8945c25f93934ce0d66a9d6df79c39c0
Robin Lee [Tue, 9 Nov 2021 23:21:00 +0000 (23:21 +0000)]
DO NOT MERGE Initialize DrawingState::trustedOverlay to false in constructor am:
41f48c7b10
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975349
Change-Id: I0912c6d109ad74b0525cc5505a77177ed1f96e4a
Winson Chung [Tue, 9 Nov 2021 23:20:59 +0000 (23:20 +0000)]
Add mechanism for a task's windows to be trusted overlays (SF) am:
7605fb4273
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975348
Change-Id: I4058c131ac922d0a13482465c08dab0f94d3cab1
Chris Ye [Tue, 9 Nov 2021 23:20:58 +0000 (23:20 +0000)]
Change InputWindowInfo::isTrustedOverlay() to be permission and flag based. am:
39bc6117dd
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975347
Change-Id: I16c075529730d5d7aab08fa0779dac436aa956e4
Winson Chung [Tue, 9 Nov 2021 23:18:51 +0000 (23:18 +0000)]
Add mechanism for a task's windows to be trusted overlays (SF) am:
7605fb4273
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975348
Change-Id: I23c2b8453f21a1d05b533a95bc82601d013a7319
Chris Ye [Tue, 9 Nov 2021 23:18:50 +0000 (23:18 +0000)]
Change InputWindowInfo::isTrustedOverlay() to be permission and flag based. am:
39bc6117dd
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15975347
Change-Id: I4bcdda3e303a6842d5cdfd3227b298e4e5c6429e
Steven Moreland [Tue, 9 Nov 2021 23:18:19 +0000 (23:18 +0000)]
avoid extra release of unowned objects in Parcel error path am:
d668098e47
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16177016
Change-Id: I339bd999acca04d1a5252a1d3237200a1bbf0088
Steven Moreland [Tue, 9 Nov 2021 23:15:53 +0000 (23:15 +0000)]
avoid extra release of unowned objects in Parcel error path am:
d668098e47
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16177016
Change-Id: Ieee2d8a542b2ebe0f73c179dc7912bf02255892c
Winson Chung [Tue, 9 Nov 2021 23:03:14 +0000 (23:03 +0000)]
Merge changes from topic "trusted_overlay_backport" into rvc-dev
* changes:
DO NOT MERGE Initialize DrawingState::trustedOverlay to false in constructor
Add mechanism for a task's windows to be trusted overlays (SF)
Change InputWindowInfo::isTrustedOverlay() to be permission and flag based.
Steven Moreland [Tue, 9 Nov 2021 22:37:48 +0000 (22:37 +0000)]
avoid extra release of unowned objects in Parcel error path am:
7c8497e012 am:
0d3d30f369
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16177015
Change-Id: I43bc55cd7ff9f645d0ab6fe84f18aea272a9b213
Siarhei Vishniakou [Wed, 9 Dec 2020 18:07:46 +0000 (08:07 -1000)]
Check if the window is partially obscured for slippery enters
Currently, we only check whether a window is partially obscured during
the initial tap down. However, there is another use case: slippery
enter.
During a slippery enter, the touch down is generated into the
slipped-into window, and touch cancel is generated for the slipped-from
window. The window receiving the slippery enter does not need to have
any flags.
Until we figure out whether we can restrict the usage of this flag to
system components, add this check as an intermediate fix.
Bug:
157929241
Test: atest FlagSlipperyTest
Test: atest inputflinger_tests
Change-Id: I93d9681479f41244ffed4b1f88cceb69be71adf2
Merged-In: I93d9681479f41244ffed4b1f88cceb69be71adf2
Steven Moreland [Tue, 9 Nov 2021 22:15:21 +0000 (22:15 +0000)]
avoid extra release of unowned objects in Parcel error path am:
7c8497e012
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16177015
Change-Id: Ifcbf7e9fbfd7807fde9b8dc5fd79317a7a45c65f
Robin Lee [Fri, 10 Sep 2021 12:03:42 +0000 (12:03 +0000)]
DO NOT MERGE Initialize DrawingState::trustedOverlay to false in constructor
To avoid it being initialised to true randomly.
Bug:
199483370
Bug:
196389741
Change-Id: I75be2b1d305e22f8a71532b9f5b8ea6c469baaaa
Merged-In: I75be2b1d305e22f8a71532b9f5b8ea6c469baaaa
Winson Chung [Tue, 29 Jun 2021 22:42:56 +0000 (15:42 -0700)]
Add mechanism for a task's windows to be trusted overlays (SF)
- Add a layer state to indicate that this layer and its children
in the hierarchy are trusted. This can only be set by callers
holding ACCESS_SURFACE_FLINGER, and will be used for the PIP
task layer to indicate that activities in PIP are trusted (as
they are controlled only by the user and SystemUI)
Bug:
191529039
Bug:
196389741
Test: TBD
Change-Id: Id92ccb087bd0d8dbaeeef3ba50b67fe015e53db8
Merged-In: Id92ccb087bd0d8dbaeeef3ba50b67fe015e53db8
Chris Ye [Sun, 10 May 2020 22:16:04 +0000 (15:16 -0700)]
Change InputWindowInfo::isTrustedOverlay() to be permission and flag based.
Add private flag to WindowManager.LayoutParams. If the flag is set,
check if caller has INTERNAL_SYSTEM_WINDOW permission.
Bug:
155781676
Bug:
196389741
Test: atest WindowManagerServiceTests
Change-Id: I58cf9f38c496e0ae8b2193dca45c0805e831bc9e
Merged-In: I58cf9f38c496e0ae8b2193dca45c0805e831bc9e
Rob Carr [Fri, 5 Nov 2021 17:01:07 +0000 (17:01 +0000)]
Merge "Change log to match with function it's coming from" into sc-v2-dev
Siarhei Vishniakou [Wed, 9 Dec 2020 18:07:46 +0000 (08:07 -1000)]
Check if the window is partially obscured for slippery enters
Currently, we only check whether a window is partially obscured during
the initial tap down. However, there is another use case: slippery
enter.
During a slippery enter, the touch down is generated into the
slipped-into window, and touch cancel is generated for the slipped-from
window. The window receiving the slippery enter does not need to have
any flags.
Until we figure out whether we can restrict the usage of this flag to
system components, add this check as an intermediate fix.
Bug:
157929241
Test: atest FlagSlipperyTest
Test: atest inputflinger_tests
Change-Id: I93d9681479f41244ffed4b1f88cceb69be71adf2
Merged-In: I93d9681479f41244ffed4b1f88cceb69be71adf2
(cherry picked from commit
870ececa8d5dfb293e671c716f98ccddae24147f)
chaviw [Thu, 14 Oct 2021 16:57:22 +0000 (11:57 -0500)]
Use bufferItem's fence when calling release
When we acquire and release immediately, we want to use the bufferItem's
release fence instead of NO_FENCE.
Also add an error log if there's a failure in acquireBuffer, which
really should never happen.
Test: BLASTBufferQueueTest
Bug:
200285149
Change-Id: I5e869b0f66f37b15b9317985b30f539dbfea831c
(cherry picked from commit
6ebdf5f0f7ca5354a1a9a0e6419275c7b873849d)
chaviw [Mon, 1 Nov 2021 14:50:57 +0000 (09:50 -0500)]
Change log to match with function it's coming from
Added the function name in the error log when SurfaceControl can't be
found in the callbacks.
Test: Log
Bug:
200285149
Change-Id: I6c721a699c25c8659f5aa5e703de134c8c0a31b7
(cherry picked from commit
768bfa07e598608f135140f1001bf09c7de35b1e)
Rob Carr [Wed, 3 Nov 2021 21:51:50 +0000 (21:51 +0000)]
Merge "DO NOT MERGE: Move blast sync handling to BBQ" into sc-v2-dev
Chavi Weingarten [Wed, 3 Nov 2021 18:25:13 +0000 (18:25 +0000)]
Merge "Allow offscreen mirrored layers to be captured." into sc-v2-dev
chaviw [Wed, 6 Oct 2021 16:53:40 +0000 (11:53 -0500)]
DO NOT MERGE: Move blast sync handling to BBQ
Add logic in BBQ so it can handle waiting the transaction callback vs a
sync transaction request. The following behavior will occur
1. If a nextTransaction (sync) was set, we will wait until the
transaction callback for that frame before continuing to acquire
new frames. Once the transaction callback for the sync transaction is
invoked, BBQ will flush the shadow queue. It will try to process
as many frames as it can that were queued up during the time BBQ was
blocked from processing.
2. If BBQ is waiting on a sync transaction callback and then another
sync transaction is requested afterwards, BBQ will allow it to acquire
a buffer instead of just adding to the shadow queue. It will acquire
the new frame in the new sync transaction and allow the caller that
requested the sync to apply it. At this point, it's up to the callers
to ensure they apply the two sync transactions in order to ensure
frames are applied in order.
3. Similar to 2, but if there are queue requests in between the two
sync requests that aren't trying to be synced. When the second sync
frame is getting acquired, BBQ will acquire and release any frames
that were requested in between. This is so we don't skip or
have to wait in the first sync transaction callback.
Test: BLASTBufferQueueTest
Bug:
200285149
Change-Id: I8da8de1a3fe2a44ca2199ff92cfd4b60c7f01183
(cherry picked from commit
d7deef7278f934a1750738b600c11c1771ae7ac6)
Vishnu Nair [Wed, 3 Nov 2021 16:06:41 +0000 (16:06 +0000)]
Merge "CompositionEngine: Update blur state correctly" into sc-v2-dev
Vishnu Nair [Tue, 2 Nov 2021 22:12:08 +0000 (15:12 -0700)]
CompositionEngine: Update blur state correctly
Previously when blur state (blur radius or blur region) was
updated, we would recompute visible regions. This is not
necessary and triggers expensive visible region checks that
may contribute to sf missing the frame deadline.
Fix this by only triggering a visible region change if we
start or stop drawing blur. In addition, when visible region
does not change, update the blur parameters so Composition
Engine picks up the most up-to-date blur parameters.
Test: atest libcompositionengine_test
Fixes:
202589767
Change-Id: I442f190d92d6b911bb6bd9b4a440e0dcdbd7fb2e
Steven Moreland [Tue, 2 Nov 2021 01:17:23 +0000 (18:17 -0700)]
avoid extra release of unowned objects in Parcel error path
Another bug due to a huge amount of complexity in the Parcel
implementation.
Bug:
203847542
Test: added testcase fails on device w/o Parcel.cpp fix, and it passes
on a device with the fix
Merged-In: I34411675687cb3d18bffa082984ebdf308e1c1a6
Change-Id: I34411675687cb3d18bffa082984ebdf308e1c1a6
(cherry picked from commit
04390376b043bf6a15ff2943a9ed63d9d8173842)
Steven Moreland [Tue, 2 Nov 2021 01:17:23 +0000 (18:17 -0700)]
avoid extra release of unowned objects in Parcel error path
Another bug due to a huge amount of complexity in the Parcel
implementation.
Bug:
203847542
Test: added testcase fails on device w/o Parcel.cpp fix, and it passes
on a device with the fix
Merged-In: I34411675687cb3d18bffa082984ebdf308e1c1a6
Change-Id: I34411675687cb3d18bffa082984ebdf308e1c1a6
(cherry picked from commit
04390376b043bf6a15ff2943a9ed63d9d8173842)
chaviw [Wed, 27 Oct 2021 16:11:24 +0000 (11:11 -0500)]
Allow offscreen mirrored layers to be captured.
If an offscreen layer is valid and has content, we should be able to
take a screen capture of it.
If we're mirroring content, but it's not on screen, we need
to ensure updateMirrorInfo is called to get the updated hierarchy. This
will allow the mirror content to get properly screenshot
Test: screencapture offscreen
Test: MirrorLayerTest
Test: ScreenCaptureTest
Change-Id: I31f806eb616e2d6f800da6328c9878a3e47d6a14
Merged-In: I31f806eb616e2d6f800da6328c9878a3e47d6a14
Bug:
188222480
Dmytro Chystiakov [Thu, 21 Oct 2021 21:57:27 +0000 (14:57 -0700)]
Increase core count in atrace.rc
Increase core count in atrace.rc to support future devices with
more cores
Bug: b/
203651019
Test: Build
Change-Id: I4c7b0ba6a4883cc21f45428368b60759c54ef985
Merged-In: I4c7b0ba6a4883cc21f45428368b60759c54ef985
(cherry picked from commit
2895eeb632400c708ae0a01596ff2eda9c3f6de2)
Prabir Pradhan [Tue, 19 Oct 2021 18:24:45 +0000 (11:24 -0700)]
MotionEvent: Avoid clipping tranformed orientation angle values
A recent refactor (ag/
14556109) exposed an issue in the existing
orientation angle transformation logic where the output was clipped
incorrectly. There is no need to clip the ouput of atan2f because its
output is in the range [-pi, pi], which conforms to the MotionEvent's
orientation API.
Test: manual with stylus and test app
Bug:
202534592
Merged-In: I55df7470049922d0f579c2c1921dcacd0757ece1
Change-Id: I55df7470049922d0f579c2c1921dcacd0757ece1
Ady Abraham [Fri, 15 Oct 2021 18:15:40 +0000 (18:15 +0000)]
Merge changes I25b3f807,Iaa13a92d into sc-v2-dev
* changes:
SF: acquire/release screen for active display only
SF: add HWVsync state to dumpsys
Chris Kuiper [Thu, 14 Oct 2021 23:39:23 +0000 (23:39 +0000)]
Merge changes Ib7a08010,Idd72b82e into sc-qpr1-dev am:
0cb6215dd6
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
16041353
Change-Id: I4ebed08aced85bc804a6dda3e30446e49bdb0699
TreeHugger Robot [Thu, 14 Oct 2021 23:39:19 +0000 (23:39 +0000)]
Merge "binderRpcTest: disable" into sc-qpr1-dev am:
e0b134b18d
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15900892
Change-Id: Ie4d211c30a1e578ee8b6e72d5742596fb51b4d66
Chris Kuiper [Thu, 14 Oct 2021 23:29:58 +0000 (23:29 +0000)]
Merge changes Ib7a08010,Idd72b82e into sc-qpr1-dev
* changes:
sensorservice: Track Proximity state change when app goes idle.
sensorservice: Refactor tracking logic for the Proximity sensor.
TreeHugger Robot [Thu, 14 Oct 2021 23:17:51 +0000 (23:17 +0000)]
Merge "binderRpcTest: disable" into sc-qpr1-dev
Chris Kuiper [Tue, 12 Oct 2021 23:36:43 +0000 (16:36 -0700)]
sensorservice: Track Proximity state change when app goes idle.
This tracks the Proximity state change when an app subscribed to the
Proximity sensor goes idle and its connection is paused.
Bug: b/
202328704
Tests: See b/
202328704#comment9 for tests done.
Change-Id: Ib7a08010d903406b67940de56a49e9efb200ad68
Chris Kuiper [Tue, 12 Oct 2021 23:30:01 +0000 (16:30 -0700)]
sensorservice: Refactor tracking logic for the Proximity sensor.
This simplifies the code that tracks if the proximity sensor is
active by removing all counting of clients and simply checking the
sensor's isSensorActive() method and reporting changes to it.
Bug: b/
202328704
Tests: See b/
202328704#comment9 for tests done.
Change-Id: Idd72b82ed0fd5ee089e795ab4b233241350979c2
Ady Abraham [Thu, 14 Oct 2021 00:04:27 +0000 (17:04 -0700)]
SF: acquire/release screen for active display only
SurfaceFlinger assumes that there is at most a single internal display
powered on at a given time, and mark it as the active display. However,
in order to be robust against rare race conditions where displays might
be on together for a short period of time, we add a check to make sure
that we tell the scheduler that the screen was acquired/released only
for the active display.
Bug:
201605862
Test: SF unit tests
Change-Id: I25b3f807d9f5d93ae88ac8a6026cee76cb69f493
Ady Abraham [Wed, 13 Oct 2021 23:59:49 +0000 (16:59 -0700)]
SF: add HWVsync state to dumpsys
Log debugging information about whether HWVsync callbacks
where enabled with HWC.
Bug:
201605862
Test: adb logcat -s SurfaceFlinger
Change-Id: Iaa13a92d37fcfea1328a6e91b17372c3737ffff6
Vishnu Nair [Thu, 7 Oct 2021 19:17:57 +0000 (12:17 -0700)]
SurfaceFlinger: Don't render rounded corners without a valid crop
A behaviour difference introduced by I6975332392756d3d96fed1d6f81245a9c7bf0a19
This will cause layers to not render if there is a corner radius
left behind.
Test: SurfaceFlinger_test
Test: repro steps in bug
Bug:
200781179
Change-Id: Ie2548fc37bb7676c10be69130d04f93e7323e79b
Vishnu Nair [Mon, 4 Oct 2021 22:03:18 +0000 (22:03 +0000)]
Merge "BlastBufferQueue: Fix async worker deadlock" into sc-v2-dev
Vishnu Nair [Fri, 1 Oct 2021 22:32:33 +0000 (15:32 -0700)]
BlastBufferQueue: Fix async worker deadlock
The async onBufferReleased callback can trigger another
onBufferReleased which will end up deadlocking the
async worker thread. Fix this by executing the callbacks
outside the lock.
Test: atest android.media.cts.MediaSyncTest#testPlaybackRateDouble --rerun-util-failure 100
Fixes:
201604213
Change-Id: I40d163c3644c6a0128936cf41e8bf8969766d9da
Wiwit Rifa'i [Thu, 30 Sep 2021 23:59:58 +0000 (23:59 +0000)]
Merge changes from topic "idle_timer_per_display" into sc-v2-dev
* changes:
SF: update kernel idle timer when active display changes
SF: query RefreshRateConfigs for kernel idle timer support
SF: configure idle timer based on display id
SF: move idle timer to refresh rate configs
Vishnu Nair [Thu, 30 Sep 2021 00:33:10 +0000 (17:33 -0700)]
BlastBufferQueue: Fix scaling when buffer scaling mode changes
The current logic updates the target size of the buffer when a buffer
arrives with FREEZE scaling mode (target size == buffer size) or when
the target size is set and the previous buffer has a scaling mode
!FREEZE. This is because if the previous scaling mode is freeze, we
do not want to change the target size otherwise we will break the
contract with the client and scale the currently presented buffer.
In the case of YoutubeMusic on TV, the client sets a target size,
then submits a buffer with scaling mode FREEZE, then changes
the target size but submits a buffer with scaling mode
SCALE_TO_WINDOW. The new target size never gets sent to
SurfaceFlinger. To fix this we check if the target size has been
updated when acquiring a buffer and update the target size.
Test: atest BLASTBufferQueueTest SurfaceViewSyncTest
Bug:
196339769
Change-Id: I6db3411b64552e5a4416d60420c92e698beda311
Anthony Stange [Wed, 29 Sep 2021 15:47:53 +0000 (15:47 +0000)]
Block dynamic connection event if sensor not found
A bug in the Sensors HIDL spec marks onDynamicSensorsConnected as oneway
which means that HALs are not blocked on the framework processing that
event. This means the onDynamicSensorsConnected invocation and dynamic
meta event can be received out of order. Blocking processing of the
event gives time for onDynamicSensorsConnected to be invoked so that the
event can be correctly processed.
Bug:
201529167
Test: Connect dynamic sensor and validate that the HAL can deliver the
event and method invocation without any sleeps on the HAL side.
Change-Id: Ie1cad22f26de60a7cecb3c6048280c4919165dbb
Ady Abraham [Tue, 28 Sep 2021 21:08:54 +0000 (14:08 -0700)]
SF: update kernel idle timer when active display changes
The refresh rate policy for a display might change when it is not
the active display, and when this display becomes active, we need
to update the kernel timer to reflect the updated policy.
Bug:
188838426
Test: check graphics.display.kernel_idle_timer.enabled is updated
according to the most updated policy when switching active
display.
Change-Id: I94ee9a019e4f38bf1bd28202144dbc30af2f9261
Ady Abraham [Sat, 25 Sep 2021 00:04:36 +0000 (17:04 -0700)]
SF: query RefreshRateConfigs for kernel idle timer support
Use the per-display idle timer configuration instead of the global one.
Bug:
188838426
Test: check graphics.display.kernel_idle_timer.enabled can only be
toggled when the active display is a display that supports kernel
idle timer.
Change-Id: I4b2258723ed52eb6e94daaff04f65b80d9d0c9d5
Ady Abraham [Sat, 4 Sep 2021 01:05:48 +0000 (18:05 -0700)]
SF: configure idle timer based on display id
Use the DisplayId when reading the timer configuration from system
properties.
Bug:
188838426
Test: check idle timer can be set separately for each display based
on display id
Change-Id: I9051374a78d99676ace7f1ded6accc8a50b17c79
Ady Abraham [Sat, 4 Sep 2021 00:32:34 +0000 (17:32 -0700)]
SF: move idle timer to refresh rate configs
Move Idle timer management to RefreshRateConfigs to be able to maintain
separate timer for each display.
Bug:
188838426
Test: check idle timer is still working for both kernel idle timer and
SF idle timer
Change-Id: If0bd6c9a1ba7e204160e722edc6aedaaeef84997
TreeHugger Robot [Mon, 27 Sep 2021 15:58:39 +0000 (15:58 +0000)]
Merge "Repeat proxy limit callbacks if we keep going above the limit." into sc-qpr1-dev am:
41b552872d
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15892490
Change-Id: I282d445b9bbfa32019711af9a341fa179a9405a8
Prabir Pradhan [Mon, 27 Sep 2021 15:46:26 +0000 (15:46 +0000)]
[automerger skipped] Use sequence numbers to synchronize enabling Pointer Capture (1/2) am:
f192a1085a -s ours
am skip reason: Merged-In I6ae9c5498dc2f783b4c7211fa3665d42e29d2919 with SHA-1
ac483a63a0 is already in history
Original change: https://googleplex-android-review.googlesource.com/c/platform/frameworks/native/+/
15871351
Change-Id: I9955ebac0c8e01b2da2bb2b9beb0608b625a7d1a
TreeHugger Robot [Fri, 24 Sep 2021 15:27:43 +0000 (15:27 +0000)]
Merge "Repeat proxy limit callbacks if we keep going above the limit." into sc-qpr1-dev
Martijn Coenen [Fri, 3 Sep 2021 16:06:24 +0000 (18:06 +0200)]
Repeat proxy limit callbacks if we keep going above the limit.
When we hit the high watermark for number of live proxies in a process,
we call a callback with the offending UID. ActivityManagerService will
then kill all processes belonging to this UID. The proxy limit callback
is only "re-armed" when the number of proxies for that UID drops below a
certain value again. However, because the proxies beloging to the now
dead processes will only be removed when the GC next runs, and restarted
processes may start adding new proxies in the mean time, we may actually
never drop below the lower threshold again, and never fire the callback
again, despite an ever-increasing amount of proxies.
To prevent this, repeatedly fire the limit callback every time that we
go another # of high watermark proxies over the limit. This, in
combination with a GC in the framework, should make sure that we
regularly kill and clean up these processes.
Bug:
198340142
Test: Manual
Change-Id: I8d877d409863fdae9cffd43709b647d0c1edb440
Merged-In: I8d877d409863fdae9cffd43709b647d0c1edb440
Steven Moreland [Thu, 23 Sep 2021 23:49:35 +0000 (16:49 -0700)]
binderRpcTest: disable
This test is disabled in S, and some of the fixes required for old
kernels aren't present here, but new systems are pulling it in.
Fixes:
200742446
Merged-In: Ieaeee990843aafdf6baedadd2f26177d53feeb96
Change-Id: I7d50218c7a645196a9f92f9b4f0a2cd7c447eef5
Vishnu Nair [Thu, 23 Sep 2021 19:14:57 +0000 (19:14 +0000)]
Merge "SF: Temporarily expose primary display id" into sc-v2-dev
Vishnu Nair [Tue, 14 Sep 2021 01:40:17 +0000 (18:40 -0700)]
SurfaceFlinger: Implement drop input modes
ALL: If this mode is set on a layer, set the DROP_INPUT
feature flag on the layer and its children if it has a valid input
channel. This will ensure these layers will not be able to receive any
input.
OBSCURED: If this mode is set, set the DROP_INPUT
feature flag on the layer and its children if they have a valid input
channel and they are considered occluded. This can happen if the layer
has a alpha set by it's parent OR if its buffer has been cropped by its
parent. Otherwise, the input feature flag DROP_INPUT_IF_OBSCURED flag
will be set so inputflinger can decide to drop input based on
occlusion.
Test: atest libgui_test InputDispatcherDropInputFeatureTest
Bug:
197364677
Merged-In: Ibce11ee7df5b5c8c226ebfab29574a99cd656f6d
Change-Id: Ibce11ee7df5b5c8c226ebfab29574a99cd656f6d
TreeHugger Robot [Wed, 22 Sep 2021 14:25:35 +0000 (14:25 +0000)]
Merge "libui: Correct the stream format back to decimal in bufferDumpHelper" into sc-v2-dev
Yichi Chen [Mon, 13 Sep 2021 12:00:52 +0000 (20:00 +0800)]
libui: Correct the stream format back to decimal in bufferDumpHelper
The stream format was changed to hexadecimal unexpectedly when the
dataspace info was added. The patch corrects the stream format back to
decimal after the dataspace is printed.
Bug:
199037164
Test: dumpsys SurfaceFlinger
Change-Id: Ibb1324ac34d22cae80114bb2bbe96c82f47f41e6
TreeHugger Robot [Tue, 21 Sep 2021 15:37:30 +0000 (15:37 +0000)]
Merge "SurfaceControl: Add setDropInputMode api" into sc-v2-dev
TreeHugger Robot [Tue, 21 Sep 2021 15:15:45 +0000 (15:15 +0000)]
Merge "SurfaceFlinger: Handle multiple rounded corner settings" into sc-v2-dev
Vishnu Nair [Sat, 18 Sep 2021 02:51:45 +0000 (19:51 -0700)]
SF: Temporarily expose primary display id
WM needs to know the address of the primary display in order to
consume ro.surface_flinger.primary_display_orientation correctly.
This path will be removed in the future once we support per
display install orientations.
Bug:
196167822
Test: check initial buffer transforms on displays with a different
install orientation
Merged-In: I2b0a92196a8a1f05f7d99b531775b9979f07af98
Change-Id: I2b0a92196a8a1f05f7d99b531775b9979f07af98
Vishnu Nair [Thu, 16 Sep 2021 23:24:47 +0000 (16:24 -0700)]
SurfaceFlinger: Handle multiple rounded corner settings
Current implementation gives precedence to the parent layers rounded
corner settings. This causes surprising artifacts when rounded
corners are applied at different parts of the hierarchy. For example,
the pip window's rounded corner will disappear during animation if
the animation leash also applies a rounded corner.
Ideal fix is to support overlapping corner radius settings. But for
most cases, we can fix this with a simple heuristic. A layer will
use its own rounded corner settings. If the layer does not have any
rounded corner settings, then its closest parent rounded corner
settings will be used. A parent can override the child's rounded
corner settings if its crop is entirely inside the child's
rounded corner crop.
What this fixes:
Scenarios where both child and parent set a rounded corner radius
will draw the rounded corner on the child correctly. Example,
pip with onehanded mode.
What doesn't change:
Layers will a single rounded corner radius its hierarchy remains
the same.
What's still broken:
If a child's rounded corner intersects with the parent's rounded
corner, then the child will draw over the parent's rounded
corner region. Example pip with onehanded mode and the pip
window is dragged to the top right corner.
Test: SurfaceFlinger_test
Fixes:
191993356
Change-Id: I6975332392756d3d96fed1d6f81245a9c7bf0a19
Vishnu Nair [Thu, 2 Sep 2021 16:52:48 +0000 (09:52 -0700)]
SurfaceControl: Add setDropInputMode api
Introduces an API to drop input events on this SurfaceControl. This
policy will be inherited by its children. The caller must hold the
ACCESS_SURFACE_FLINGER permission.
Options include:
ALL: SurfaceControl and its children will not receive any
input regardless of whether it has a valid input channel.
OBSCURED: SurfaceControl and its children will not
receive any input if the layer is obscured, cropped by its parent or
translucent.
These policies are used to enable features that allow for a less trusted
interaction model between apps. See the bug for more details.
Test: atest libgui_test InputDispatcherDropInputFeatureTest
Bug:
197364677
Merged-In: I443741d5ab51a45d37fb865f11c433c436d96c1e
Change-Id: I443741d5ab51a45d37fb865f11c433c436d96c1e
Prabir Pradhan [Fri, 6 Aug 2021 14:01:18 +0000 (14:01 +0000)]
Use sequence numbers to synchronize enabling Pointer Capture (1/2)
InputReader only processes configuration change from its main thread.
This means that if there is more than one Pointer Capture change
request when the thread is busy or sleeping, it will only process the
latest one. To ensure requests to enable Pointer Capture are synchronized
with Dispatcher, we must use sequence numbers.
Requests to enable Pointer Capture have a sequence number. Requests to
disable Pointer Capture do not have a value.
Bug:
195312888
Test: atest inputflinger_tests
Test: manual with GeforceNow app, see bug.
Merged-In: I6ae9c5498dc2f783b4c7211fa3665d42e29d2919
Change-Id: I6ae9c5498dc2f783b4c7211fa3665d42e29d2919
Vishnu Nair [Fri, 17 Sep 2021 15:09:24 +0000 (15:09 +0000)]
Merge "InputFlinger: Add DROP_INPUT and DROP_INPUT_IF_OBSCURED feature flags" into sc-v2-dev