Skip to content
GitLab
Select Git revision
  1. 10 Jul, 2020 2 commits
  3. 30 Jun, 2020 1 commit
  5. 29 Jun, 2020 1 commit
    • Kevin DuBois's avatar
      SF: update VSP timebase on error condition · 241d0eed
      Kevin DuBois authored 
      When an vsync timestamp is recorded that is anonmolous to the currently
recorded timestamp ringbuffer, update the timebase for synthetic
calculations.

Test: 2 new unit tests
Test: visual spot checking interactions on pixel4 device
Test: uibench a/b anti-regression
Test: dogfood with patch based one recent rvc build.
Fixes: 159882858
Change-Id: Ie201cd593a54586d9b1f488c6d2ca44178d75cf1
      241d0eed
  7. 27 Jun, 2020 1 commit
  9. 26 Jun, 2020 9 commits
  11. 25 Jun, 2020 6 commits
  13. 24 Jun, 2020 4 commits
    • Josh Gao's avatar
      adbd_auth: return auth id when requesting a prompt. · 9e933c54
      Josh Gao authored 
      When multiple auth requests come in (e.g. if someone connects over TCP
and USB, or if we send a USB request, and then kill adb and try again),
we need to know the ID assigned to the request to disambiguate.

Bug: http://b/159061108
Test: manual
Change-Id: I68cf2335c5958decf29e58ed9132735577a7e074
      9e933c54
    • Josh Gao's avatar
      adbd_auth: improve logging. · 975cee72
      Josh Gao authored 
      Bug: http://b/159061108
Test: treehugger
Change-Id: I4aa83e71ea99752dd819a5a5a9d7a050f49c1dae
      975cee72
    • Robert Carr's avatar
      SurfaceFlinger: Force runHandleTransaction · 6417d2e5
      Robert Carr authored 
      In the linked bug, there are traces documenting cases where
the user observes jank while deferred transactions are in play.
We can see several cases where the client gets stuck for a long
time on dequeueBuffer. Around these times we can also see cases where
latchBuffer aborts early due to the !allTransactionsSignaled condition.
We expect all the transactions to be signalled at this point, because
before handlePageFlip we expect to reach handleTransactionLocked,
which should call notifyAvailableFrames, marking the sync points
available based on the head frame number. However if we study the trace
we also see there is an onMessageInvalidate, which contains
a handlePageFlip but does NOT contain the handleTransaction before it
which we would expect. However we know if onMessageInvalidate calls
handlePageFlip (via handleMessageInvalidate) then it must have called
handleMessageTransaction, as these two lines are right next to
eachother. We can thus conclude that we see the problem because
handleMessageTransaction didn't call handleTransaction due to
runHandleTransaction evaluating to false. This makes sense in the
context of recent changes. Previously deferred transactions would set
the transaction flags continuously while they were pending, so the
server would always perform transactions when waking up, and we wouldn't
get in this condition. However, setting the transaction flags
always triggers a wake up! But we don't actually want to wake up until
the frame actually arrives, so first we tried setting the transaction
flags without triggering a wake up. But we found this could prevent
other people from triggering wakeups, so a recent CL modified it to
set mForceTraversal instead. Intending to force that we call
doTransaction from handleMessageInvalidate regardless of whether a
new transaction has arrived (because we need to check if an old
transaction is now ready). However, it didn't quite work because of this
early abort due to runHandleTransaction.

Bug: 159677043
Test: Existing tests pass
Change-Id: I84f2a9cca2ebb3f59d30924efdb39c917f47111b
      6417d2e5
    • arthurhung's avatar
      Clear spam log for INPUT_FEATURE_NO_INPUT_CHANNEL window · be737674
      arthurhung authored 
      Currently, in order to detect occlusion input, all buffered layers
would update input info into InputFlinger, so the window may contain
null token, we should check if id is also identical to prevent get the
wrong result

Bug: 159349058
Test: Enable show tap, open bubble and watch logs.
Change-Id: I8412812e9b100988ab380ad64366f2548e50d9e5
      be737674
  15. 23 Jun, 2020 5 commits
    • Mady Mellor's avatar
      Revert "Clear spam log for INPUT_FEATURE_NO_INPUT_CHANNEL window" · 017bcd17
      Mady Mellor authored 
      This reverts commit 80d685ae.

Reason for revert: seeing if maybe caused regression (b/159719013)
Bug: 159719013
Bug: 159349058
Change-Id: Iac21f799822bab5c26b9753ed78617e0087810bf
      017bcd17
    • Vishnu Nair's avatar
      Call Layer::getLayerDebugInfo from the main thread · 927b3b12
      Vishnu Nair authored 
      Fixes an issue where drawing state could be accessed from a binder
thread. The function also mixed current state with drawing state
incorrectly. The function now only retrieves drawing state.

Bug: 150226608
Test: Steps in bug doesn't repro
Test: atest sffakehwc_test
Merged-In: I04daedcb9a890083cc710bab30b295e14b9872ae
Change-Id: I04daedcb9a890083cc710bab30b295e14b9872ae
      927b3b12
    • Vishnu Nair's avatar
      Call Layer::getLayerDebugInfo from the main thread · b123f47b
      Vishnu Nair authored 
      Fixes an issue where drawing state could be accessed from a binder
thread. The function also mixed current state with drawing state
incorrectly. The function now only retrieves drawing state.

Bug: 150226608
Test: Steps in bug doesn't repro
Test: atest sffakehwc_test
Merged-In: I5537c53e8214e2785473839d71fd483d1a3219b6
Change-Id: I04daedcb9a890083cc710bab30b295e14b9872ae
      b123f47b
    • arthurhung's avatar
      Clear spam log for INPUT_FEATURE_NO_INPUT_CHANNEL window · 80d685ae
      arthurhung authored 
      Currently, in order to detect occlusion input, all buffered layers
would update input info into InputFlinger, so the window may contain
null token, we should check if id is also identical to prevent get the
wrong result.

Bug: 159349058
Test: Enable show tap, open bubble and watch logs.
Change-Id: Ia1841ba51bce00fc9901fdf7324e8e31d9737082
      80d685ae
    • TreeHugger Robot's avatar
      Merge "Ensure connection is non-null" into rvc-dev · 402bbb4a
      TreeHugger Robot authored
      402bbb4a
  17. 22 Jun, 2020 8 commits
  19. 19 Jun, 2020 3 commits
    • Brian Duddie's avatar
      Skip checking the app op for step sensors for legacy apps · 457e639a
      Brian Duddie authored 
      To maintain backwards compatibility, ensure that apps with target SDK
set to P or lower get access to step count and detect sensors,
regardless of their permission status. After a change was made to the
app ops code in rvc-dev, this check resulted in compatibility breakage
for some apps built with target SDK before Q.

Fixes: 157638722
Test: confirm that a popular step counting app from the Play Store is
      able to receive step count events after the change (where it
      couldn't before)
Change-Id: Idab0ec458221524997f68311f4c6b216c969cb61
      457e639a
    • Valerie Hau's avatar
      Only force send callbacks if they are registered · 109ad715
      Valerie Hau authored 
      Some clients of ASurfaceControl may not set callbacks.  We do not want
to call finalizePendingCallbacks in this case.

Bug: 159231312
Test: build, boot, SurfaceFlinger_test, libsurfaceflinger_unittest
Change-Id: I62dc270302a60274d8d1192d0e35ce3f48721e6c
      109ad715
    • TreeHugger Robot's avatar
      Merge "Allow native process to load updatable driver." into rvc-dev · 09397229
      TreeHugger Robot authored
      09397229