OSDN Git Service
Treehugger Robot [Wed, 10 Jul 2019 00:30:41 +0000 (00:30 +0000)]
Merge "PDL: Add variable-size CustomField(s)"
Treehugger Robot [Wed, 10 Jul 2019 00:30:13 +0000 (00:30 +0000)]
Merge "PDL: Add ParentDef for packets"
Treehugger Robot [Wed, 10 Jul 2019 00:26:13 +0000 (00:26 +0000)]
Merge "PDL: Move GenInclude and GenUsing to CustomDef"
Treehugger Robot [Wed, 10 Jul 2019 00:05:59 +0000 (00:05 +0000)]
Merge "Rootcanal: Implement reject connection request"
Andre Eisenbach [Mon, 8 Jul 2019 21:49:06 +0000 (14:49 -0700)]
Switch Nintendo workaround to match by name
Bug:
136478524
Test: manual
Change-Id: Ib60cd5adfc4db3ba93fe5ed8c15353225b6cf165
Myles Watson [Tue, 2 Jul 2019 19:02:47 +0000 (12:02 -0700)]
PDL: Add variable-size CustomField(s)
Test: bluetooth_packet_parser_test
Change-Id: I88e3dd2ab4804e8b5a0692d005969640cc110b99
Myles Watson [Tue, 9 Jul 2019 21:09:42 +0000 (14:09 -0700)]
PDL: Add ParentDef for packets
Factor out logic for holding fields in preparation for
structs.
Test: bluetooth_packet_parser_test
Change-Id: Ic9eb6d5387efe5be439a0d6b222c557b05bd3fa9
Myles Watson [Tue, 2 Jul 2019 17:14:47 +0000 (10:14 -0700)]
PDL: Move GenInclude and GenUsing to CustomDef
Test: bluetooth_packet_parser_test
Change-Id: I9ebe10f9d537a254ddd21d62d2c1a23f38e00d0e
Hansong Zhang [Mon, 8 Jul 2019 22:32:26 +0000 (15:32 -0700)]
Rootcanal: Implement reject connection request
Test: manual
Change-Id: I5413cf83f031b6eb2b83b943c6157f845aa40b2d
Andre Eisenbach [Mon, 8 Jul 2019 20:04:39 +0000 (20:04 +0000)]
Merge "Match Nintento Pro Controller by name for interop"
Andre Eisenbach [Mon, 1 Jul 2019 17:48:10 +0000 (10:48 -0700)]
Match Nintento Pro Controller by name for interop
The Nintento Pro Controller uses an unknown number of
BDA ranges (at least 3 found internally). Switching
to name matching...
Bug:
136478524
Test: manual
Change-Id: Ie2dfb700c9550af0c314759ec496a04a8e1aa062
Treehugger Robot [Mon, 8 Jul 2019 16:46:45 +0000 (16:46 +0000)]
Merge "GD:Wait channel ready after stop/start server"
Chienyuan [Fri, 5 Jul 2019 10:18:56 +0000 (18:18 +0800)]
GD:Wait channel ready after stop/start server
When running multiple tests, the second test will always fail due to
client send rpc request before channel ready. Add wait_channel_ready
function to GdDeviceBase and use it to check channel ready when each
testcase start.
Bug:
136577370
Test: run run_cert.sh
Change-Id: I210dcd69d6e40a315710ba2b8e9e8819e462b42b
Treehugger Robot [Thu, 4 Jul 2019 08:00:23 +0000 (08:00 +0000)]
Merge "Hearing Aid: set BLE phy to 2M if local and peer indicate PHY 2M support"
weichinweng [Wed, 3 Jul 2019 09:52:19 +0000 (17:52 +0800)]
Hearing Aid: set BLE phy to 2M if local and peer indicate PHY 2M support
Set the BLE phy to 2M if the BT controller and peer Hearing Aid device
indicate phy 2M support, otherwise, use default phy channel.
Bug:
135769508
Test: 1.Forget/Repair Hearing Aid device check whether phy is expected.
2.Disconnect/Reconnect Hearing Aid device check whether phy is expected.
3.HearingAid device power off/on, check whether phy is expected.
Change-Id: Ib410a8885f3a68c361bb97d5c9a3833f2a1be528
Myles Watson [Wed, 5 Jun 2019 22:35:52 +0000 (15:35 -0700)]
PDL: Add missing quote in matching constraint error
Test: misspell a constraint and compile a .pdl file
Change-Id: I86aa9a8a4270d4c5fd922c53112ffdf8fc0f86fe
Myles Watson [Fri, 21 Jun 2019 13:04:08 +0000 (06:04 -0700)]
RootCanal: Always add default devices
Bug:
129546852
Test: android.bluetooth.cts.BluetoothLeScanTest#testBasicBleScan
Change-Id: I9912dc7a98155972483e1e1dca812ebbae806bdd
Myles Watson [Wed, 26 Jun 2019 17:27:19 +0000 (10:27 -0700)]
RootCanal: Make beacons less verbose
Test: scan for Bluetooth devices, watch logcat
Change-Id: I971a7e56359adc39d006a1b10b5333c2f31ebaff
Hansong Zhang [Fri, 21 Jun 2019 20:53:33 +0000 (13:53 -0700)]
Rootcanal: Fix check in RejectSlaveConnection
Test: manual
Change-Id: Ic37c123f5326e2cf5675f2eeb1734eb9514d86bc
Venkata Jagadeesh Garaga [Thu, 18 Apr 2019 11:43:49 +0000 (17:13 +0530)]
GAP: Correct the continuous pkt length in l2cap
L2cap continuous pkt length wrongly calculated in
reassembly logic when remote sends more data
than expected.
Wrong pkt length leading to memory corruption
Hence the Correct the continuous pkt length in
l2cap reassembly logic.
Change-Id: I758d9e31465b99e436b9b1841320000f08186c97
CRs-Fixed:
2434229
Bug:
135239489
Hansong Zhang [Wed, 19 Jun 2019 23:53:15 +0000 (16:53 -0700)]
Rootcanal: detect HCI socket close
Test: manual
Change-Id: I609f917061e5d83e011edfeb61de281ff45b13e4
Chris Manton [Thu, 20 Jun 2019 18:22:04 +0000 (18:22 +0000)]
Merge "Extend setup acts script with more helpful information on failure"
Hansong Zhang [Wed, 19 Jun 2019 18:21:37 +0000 (11:21 -0700)]
HCI layer: Gracefully handle multiple incoming ACL
Test: bluetooth_test_gd
Change-Id: Ieeaf9ba516f998d8c0bf13c5032cf231361e091d
Hansong Zhang [Wed, 19 Jun 2019 21:53:16 +0000 (14:53 -0700)]
OS Queue: Add Enqueue Buffer
Enqueue Buffer is a common interface to handle RegisterEnqueue() and
UnregisterEnqueue() according to current buffer size
Test: bluetooth_test_gd
Change-Id: I5b3ef749cd09f866a6c676af403ab6fa39d34701
Hansong Zhang [Tue, 18 Jun 2019 23:04:01 +0000 (16:04 -0700)]
HCI Controller: Read BD_ADDR
Stack needs to know BD_ADDR, when it reports its own address to facade
Test: bluetooth_test_gd
Change-Id: I77754bb89f3638a6b15a3037d0427754b439cdcf
Chris Manton [Wed, 19 Jun 2019 17:54:04 +0000 (10:54 -0700)]
Extend setup acts script with more helpful information on failure
Test: source ./setup_acts.sh
Change-Id: If04190a79a175d5f6f11c168286d6a48a64be9ea
Hansong Zhang [Tue, 7 May 2019 00:54:51 +0000 (17:54 -0700)]
HCI: ACL Manager
ACL manager is the interface for user layer (L2CAP) to manage connection
and send/receive data packets
Test: atest --host bluetooth_test_gd
Change-Id: I43aa5aabd2474aca5e0b4bae06a6cbfae56fd156
Treehugger Robot [Tue, 18 Jun 2019 19:44:27 +0000 (19:44 +0000)]
Merge "Limit sniff mode for Nintendo Pro Controller to 11.25ms"
Andre Eisenbach [Thu, 13 Jun 2019 18:50:57 +0000 (11:50 -0700)]
Limit sniff mode for Nintendo Pro Controller to 11.25ms
Bug:
133410149
Test: manual
Change-Id: I4bffb759323bedd36e754cf758ac1d61cffc886e
Jakub Pawlowski [Tue, 18 Jun 2019 12:50:07 +0000 (14:50 +0200)]
Fix coverage build
Bug:
135496567
Change-Id: I0de650ce2987926098577e7ab690c0bd9409e39a
Hansong Zhang [Fri, 14 Jun 2019 01:31:40 +0000 (18:31 -0700)]
Stack manager: Clear handler during ShutDown()
* Clear StackManager.handler_ before it is deleted
* Use local variable promise in StartUp and ShutDown
* Add unit test
Test: bluetooth_test_gd
Change-Id: I0a6de64c5f159fdf4c97c123aa3fa2f37b2b7e34
Hansong Zhang [Sat, 15 Jun 2019 15:12:55 +0000 (08:12 -0700)]
Improve blocking queue
Instead of take_for(), add a helper function to wait until the queue is
non-empty, so we don't require T to have a empty constructor.
Test: bluetooth_test_gd
Change-Id: Ia30a8149646fc75ecf19778888636c4129dbf100
Myles Watson [Fri, 17 May 2019 21:41:00 +0000 (14:41 -0700)]
PDL: Make GetPayload() const
Test: bluetooth_packet_parser_test
Change-Id: Icd2265863c694c0f794211e161bd50b6a20c25a5
Myles Watson [Tue, 11 Jun 2019 20:46:15 +0000 (13:46 -0700)]
HAL: Add unregisterIncomingPacketCallback
Test: bluetooth_test_gd
Change-Id: I17f8227c351451b4249c2a005ea4428ace734dc9
Myles Watson [Mon, 3 Jun 2019 23:33:59 +0000 (16:33 -0700)]
HCI: Add a controller for properties and credits
Test: bluetooth_test_gd --gtest_filter=*ControllerT*
Change-Id: Icd7c58e2e8314f6a75874ea7a98a1e23e8307ab6
Treehugger Robot [Thu, 13 Jun 2019 20:28:27 +0000 (20:28 +0000)]
Merge "HCI: HCI command credit and timeout handling"
Jakub Pawlowski [Thu, 13 Jun 2019 14:31:53 +0000 (16:31 +0200)]
Fix for BluetoothGattCallback.onConnectionStateChange being called more than once
This patch fixes the issue, where the onConenctionStateCallback is being
called twice when device is connected. This happens only if device was
connected before, and the physical connection was kept alive.
This issue was introduced during white list refactor.
Test: GattConnectTest:test_gatt_connect_in_quick_succession
Bug:
117182283
Change-Id: Id94db9429e31224ef35eb2c99c0f9a65e7c27d32
Myles Watson [Wed, 12 Jun 2019 21:57:46 +0000 (14:57 -0700)]
HCI: HCI command credit and timeout handling
Test: bluetooth_gd_test
Change-Id: Iff11fd2e21f71c23d716f7ca5f920490dd7c79f3
Hansong Zhang [Tue, 11 Jun 2019 23:02:35 +0000 (16:02 -0700)]
HCI layer: Synchronize API with handler
In HciLayer::impl, synchronize command_queue_, event_handlers_ with
handler within HciLayer module.
Use member function send_acl and send_sco instead of callback function.
Test: atest --host bluetooth_test_gd
Change-Id: Ibd541578828896fe9a5ccb42be7987a0fb556f6c
Hansong Zhang [Fri, 7 Jun 2019 04:44:55 +0000 (21:44 -0700)]
Replace std::function with base::Closure
std::function doesn't support non-copyable arguments. Use base::Closure
and base::OnceClosure from libchrome instead.
Test: atest --host bluetooth_test_gd
Change-Id: Ic18bbc000730e8b0d7acff0097870bd258ee6150
Myles Watson [Fri, 7 Jun 2019 15:57:16 +0000 (08:57 -0700)]
OS: Add ordering guarantees for Handler.Clear()
Assert that Clear() is called once before ~Handler()
Test: bluetooth_test_gd
Change-Id: I40f89d13a05deab5f0d7ab1596a9101b097ef52a
Myles Watson [Wed, 12 Jun 2019 17:19:17 +0000 (10:19 -0700)]
OS: Add Reactor::WaitForUnregisteredReactable()
Test: bluetooth_test_gd --gtest_filter=*ReactorTest*
Change-Id: I9328d6fcd40941b25d252abc563e2a933ec06c25
weichinweng [Tue, 11 Jun 2019 14:02:46 +0000 (22:02 +0800)]
Load gatt_database if it is empty in bta_gattc_process_indicate
If Service Changed indication is received but gatt_database is empty,
will try to load database from storage.
Bug:
131814072
Bug:
134994428
Test: 1.CTS Bluetooth LE Secure Test. 2.Unit test.
3. Pair BLE device via APP. 4. Pair device via Bluetooth Setting
Change-Id: Id3f3f0f33d5e0469a6f2ac6612fa9f5ea0b24142
Jakub Pawlowski [Tue, 11 Jun 2019 18:18:48 +0000 (18:18 +0000)]
Merge "Revert "Open connection if it doesn't exsist in bta_gattc_process_indicate""
Jakub Pawlowski [Tue, 11 Jun 2019 11:30:04 +0000 (11:30 +0000)]
Revert "Open connection if it doesn't exsist in bta_gattc_process_indicate"
This reverts commit
ffc95919835bbf89d635516955bff1b3b8a17641.
Reason for revert: breaks HA bonding from within app.
Bug:
134942525
Change-Id: I824391b1c84b535d7f251e176a5169eb79921c7f
Myles Watson [Fri, 7 Jun 2019 15:57:16 +0000 (08:57 -0700)]
OS: Update Handler unittest
Test: bluetooth_test_gd
Change-Id: I14bb8c8852d8749e9c531ac18f18a39c92dae922
Myles Watson [Mon, 10 Jun 2019 20:10:52 +0000 (13:10 -0700)]
OS: Don't hold the lock while the callback runs
Test: bluetooth_test_gd \
--gtest_filter=*unregister_from_different_thread*
Change-Id: Icd820cd1a68a6ec26f0284b5c2bd7ba673bebb0c
The Android Open Source Project [Mon, 10 Jun 2019 16:32:28 +0000 (09:32 -0700)]
DO NOT MERGE - Merge pie-platform-release (PPRL.190605.003) into master
Bug:
134605042
Change-Id: Icf8a142122c66c40c9e906380778d1d0237ba4be
TreeHugger Robot [Fri, 7 Jun 2019 07:06:20 +0000 (07:06 +0000)]
Merge "DO NOT MERGE - Merge pi-platform-release (PPRL.190605.003) into stage-aosp-master" into stage-aosp-master
Xin Li [Wed, 5 Jun 2019 22:41:48 +0000 (15:41 -0700)]
DO NOT MERGE - Merge pi-platform-release (PPRL.190605.003) into stage-aosp-master
Bug:
134605042
Change-Id: I5be87422d75fb89d35135a5e087920fae1f6b2d6
Myles Watson [Wed, 5 Jun 2019 23:41:55 +0000 (16:41 -0700)]
Merge "OS: Create Alarms on a Handler's thread"
am:
e1c1b2d8f6
Change-Id: Ie4238a454195704381a14337cf6c62ba340569f4
Treehugger Robot [Wed, 5 Jun 2019 23:15:28 +0000 (23:15 +0000)]
Merge "OS: Create Alarms on a Handler's thread"
Myles Watson [Mon, 3 Jun 2019 16:25:28 +0000 (09:25 -0700)]
OS: Create Alarms on a Handler's thread
Test: bluetooth_test_gd --gtest_filter=*Alarm*
Change-Id: I017a5f74f1d72468ca487536c88a0cf3eb48ddea
Ugo Yu [Wed, 5 Jun 2019 17:30:25 +0000 (10:30 -0700)]
Rename and reset the proirity of HID host thread
am:
4e25ef13b2
Change-Id: If957c7e793d4ae2462d9b1c5a57e17a91224d02a
Ugo Yu [Wed, 5 Jun 2019 12:08:29 +0000 (20:08 +0800)]
Rename and reset the proirity of HID host thread
* Rename the HH event polling thread to bt_hh_thread
* HH event polling thread is created by bt_main_thread
with RT priority. Lower the thread priority since the
tasks in this thread is not timing critical
Bug:
131787936
Test: Connect to a HID device
Run "adb shell ps -A -T -Z -O rtprio"
Change-Id: Ibd983e96bc339393842d4c2e2aa226534095e225
Myles Watson [Wed, 5 Jun 2019 02:41:15 +0000 (19:41 -0700)]
Use Start for testing and add a Stop timeout
am:
879fbfbed8
Change-Id: I3ae3d628144a8aadd91f86814513f28413d4d60e
android-build-team Robot [Wed, 5 Jun 2019 01:45:50 +0000 (01:45 +0000)]
Snap for
5524043 from
d0b9f6155d878426fa948b28b51b4b4c5989201e to pi-platform-release
Change-Id: Id0f4c260caf8b1b91cca9955c2a57e5c194574ba
Jakub Pawlowski [Wed, 5 Jun 2019 01:02:17 +0000 (18:02 -0700)]
Hearing Aid: properly track audio Suspend/Resume
am:
054762bcfe
Change-Id: I3aa6513dab1773d521933bc69fb1eb731d6835f1
Myles Watson [Tue, 7 May 2019 00:54:51 +0000 (17:54 -0700)]
Use Start for testing and add a Stop timeout
Test: atest --host bluetooth_test_gd
Change-Id: Ib966f46a35572dfbe05a10e3d44c4b4503d616a3
Jakub Pawlowski [Tue, 4 Jun 2019 21:23:16 +0000 (23:23 +0200)]
Hearing Aid: properly track audio Suspend/Resume
When audio is suspended or resumed, we are notified by OnAudioSuspend or
OnAudioResume callback. Unfortunately, when we are the one triggering
the HearingAidAudioSource::Stop(), we would not receive the suspend
callback. This is causing HA to believe audio is being streamed when
it's not.
This patch fixes that by setting audio_running to false when we make
call to Stop()
Bug:
132793574
Test: Play audio, restart Hearing Aid, verify START command is not being
sent
Change-Id: I3c4551114ac275ea8c72f64a8d036373429acd6d
Myles Watson [Mon, 3 Jun 2019 21:55:31 +0000 (14:55 -0700)]
OS: Add death tests for en/de-queue registration
am:
b06e0b59cb
Change-Id: Ifcbaec5cd56ec89e1a4a965e8b53b38a085f51bf
Myles Watson [Mon, 3 Jun 2019 21:34:58 +0000 (14:34 -0700)]
Revert "OS: Construct Handler using Reactor"
am:
dde08cdba3
Change-Id: I866c437c67e72b0b8f0372f3aa02a1f463f143c8
Myles Watson [Mon, 3 Jun 2019 16:13:21 +0000 (09:13 -0700)]
OS: Add death tests for en/de-queue registration
Test: bluetooth_gd_test
Change-Id: Iadf126018a2c919fa10096506009674e7a2cd73c
Myles Watson [Mon, 3 Jun 2019 16:06:26 +0000 (09:06 -0700)]
Revert "OS: Construct Handler using Reactor"
Reactors are an implementation-specific detail.
This reverts commit
0a2f52e692bb65cea7ecff7ce1b1e839b786d5c7.
Test: bluetooth_gd_test
Change-Id: If74a9e0f32671ba96a723e51eee495f2aebf1e7c
Sumit Deshmukh [Mon, 3 Jun 2019 07:42:51 +0000 (00:42 -0700)]
Merge "BLE: [IOT] Initiate disconnection when encryption fails during pairing"
am:
d4ce321925
Change-Id: Ifd3589b0b55ba1501e84251351f41b56886b32dc
Treehugger Robot [Mon, 3 Jun 2019 07:37:32 +0000 (07:37 +0000)]
Merge "BLE: [IOT] Initiate disconnection when encryption fails during pairing"
Jakub Pawlowski [Fri, 31 May 2019 23:35:11 +0000 (16:35 -0700)]
Merge "GD SMP ECC: Add const qualifier to ECC_PointMult"
am:
f3136f33b4
Change-Id: I781d4aedae80ae00cd4e3493c8c7fc795f74fc15
Treehugger Robot [Fri, 31 May 2019 23:05:56 +0000 (23:05 +0000)]
Merge "GD SMP ECC: Add const qualifier to ECC_PointMult"
Jakub Pawlowski [Fri, 31 May 2019 20:13:52 +0000 (13:13 -0700)]
GD SMP: use fixed size arrays in packets
am:
4ea3700df7
Change-Id: Idbd285f34f15fc9436739d8efd4fd6aeeb0a4d15
Jakub Pawlowski [Fri, 31 May 2019 12:16:35 +0000 (14:16 +0200)]
GD SMP: use fixed size arrays in packets
Test: compilation
Bug:
126463795
Change-Id: Iad1fd5004efab67294c6926475625e4c4b35a38c
Jakub Pawlowski [Fri, 31 May 2019 15:56:55 +0000 (08:56 -0700)]
PDL: fixed size type arrays
am:
a1a27f8f4c
Change-Id: I179ac2bd7dca5ccf27049a1c89a7e8d765dcc669
Jakub Pawlowski [Fri, 31 May 2019 12:07:44 +0000 (14:07 +0200)]
PDL: fixed size type arrays
Currently attempt to add fixed size array is causing segfault. This
patch fixes that. It also adds a test to verify those arrays are
properly encoded and decoded.
Test: bluetooth_packet_parser_test
Change-Id: I09c7e5595a7d1819a43e9203846bb487fb23aca3
Hansong Zhang [Thu, 30 May 2019 23:51:39 +0000 (16:51 -0700)]
HCI Layer: Register CommandComplete and CommandStatus on Start()
am:
712e9e2f79
Change-Id: Id706b779f96c5e9ddfc7f994e9e4d83c65898b99
Hansong Zhang [Thu, 30 May 2019 00:52:24 +0000 (17:52 -0700)]
HCI Layer: Register CommandComplete and CommandStatus on Start()
Module.GetHandler() works only after a Module is started. Move
RegisterEventHandler() for CommandComplete and CommandStatus to Start().
Test: atest bluetooth_test_gd --host
Change-Id: I7331c8ddb8d61d697aa4480e7c591c3817d21afa
Jakub Pawlowski [Thu, 30 May 2019 17:57:36 +0000 (10:57 -0700)]
Merge "PDL: Add Arrays"
am:
7d7dd39e90
Change-Id: I56e5fdfba619fb7ea9cede16320c1be360b80133
Jakub Pawlowski [Thu, 30 May 2019 17:45:17 +0000 (17:45 +0000)]
Merge "PDL: Add Arrays"
Cheney Ni [Thu, 30 May 2019 03:16:45 +0000 (20:16 -0700)]
Merge "Make sure the Hearing Aid source is ready before resuming or suspending calls"
am:
f58ec178f5
Change-Id: Ied52e1d024172e4cb369b0af3641f04e294c195c
Treehugger Robot [Thu, 30 May 2019 02:58:08 +0000 (02:58 +0000)]
Merge "Make sure the Hearing Aid source is ready before resuming or suspending calls"
Myles Watson [Wed, 6 Feb 2019 22:02:36 +0000 (14:02 -0800)]
PDL: Add Arrays
Test: bluetooth_packet_parser_test
Change-Id: If7f057a16ab40066a53d2c66980c5579a480650f
weichinweng [Wed, 29 May 2019 10:46:42 +0000 (03:46 -0700)]
Open connection if it doesn't exsist in bta_gattc_process_indicate
am:
ffc9591983
Change-Id: I1aab49dcdb7c2212271b6c0b8006b0edfdbb6712
weichinweng [Wed, 29 May 2019 06:27:15 +0000 (14:27 +0800)]
Open connection if it doesn't exsist in bta_gattc_process_indicate
If Service Changed indication is received before the connection is open
, send the BTA_GATTC_INT_CONN_EVT to open it before processing the
indication.
Bug:
131814072
Test: CTS Bluetooth LE Secure Test. Unit test.
Change-Id: I1b3415ef6aafb52a6ac229409514a85f35de5842
Myles Watson [Tue, 28 May 2019 20:46:45 +0000 (13:46 -0700)]
Merge "OS: Hold the mutex when decreasing the semaphore"
am:
056e104366
Change-Id: I008b66416a6f1236a005ff1148e7cc57a14e2868
Myles Watson [Tue, 28 May 2019 20:46:26 +0000 (13:46 -0700)]
Merge "OS: Protect invalidation_list_ with the mutex"
am:
0f9422e984
Change-Id: Ib22eaeb3df605985fb3739c476ccdfc717bd8137
Treehugger Robot [Tue, 28 May 2019 20:31:20 +0000 (20:31 +0000)]
Merge "OS: Hold the mutex when decreasing the semaphore"
Treehugger Robot [Tue, 28 May 2019 20:29:21 +0000 (20:29 +0000)]
Merge "OS: Protect invalidation_list_ with the mutex"
Cheney Ni [Fri, 24 May 2019 06:59:43 +0000 (14:59 +0800)]
Make sure the Hearing Aid source is ready before resuming or suspending calls
Those calls to resume / suspend streaming are from the audio's IPC
threads like HwBinder or UIPC, and calls to start / end session are from
the stack main thread. There is timing that those threads may run in
parallel, and trys to change the streaming state when the Hearing Aid
source is not ready. This change checks the source state before resuming
or suspending streaming, and breaks the deadlock between IPC threads and
the stack main thread.
Bug:
132860565
Test: 1. Connected to a Hearing Aids and disable the Bluetooth
2. Switch the active device manually
Change-Id: I6ae11faab3d530c65f55ffa9b8a9f3cae2bde5f4
Colin Cross [Tue, 28 May 2019 17:58:24 +0000 (10:58 -0700)]
Merge changes I42c47d98,If38ad50d
am:
558c5114b6
Change-Id: I0cd8aba87330c1b8dd77eeefb83542e15d5b76e3
Jakub Pawlowski [Mon, 27 May 2019 19:40:13 +0000 (21:40 +0200)]
GD SMP ECC: Add const qualifier to ECC_PointMult
Second parameter passed to ECC_PointMult is either remote device public
key, or the constant from the BT spec. Both of them should never be modified.
Assigning the "z" value inside ECC_PointMult had no effect on the rest
of computation.
Test: SmpEccValidationTest
Bug:
126463795
Change-Id: I0530d26e0e8841cb8b263ea948ae7a19850f0670
Treehugger Robot [Tue, 28 May 2019 17:52:14 +0000 (17:52 +0000)]
Merge changes I42c47d98,If38ad50d
* changes:
Fix mac build
bpfmt -w $(find . -name Android.bp)
Jakub Pawlowski [Tue, 28 May 2019 17:20:12 +0000 (10:20 -0700)]
Merge "Add missing header in gd/packet/parser/checksum_type_checker.h"
am:
7b4e117cd7
Change-Id: Iaa4dd3b3a6da0812a0e0486dc2286c85794ba4ef
Myles Watson [Tue, 28 May 2019 16:43:55 +0000 (09:43 -0700)]
OS: Hold the mutex when decreasing the semaphore
Test: bluetooth_test_gd # Run in a loop
Change-Id: I72e9f28c417986f63390cbbbe6467f64df6e9237
Treehugger Robot [Tue, 28 May 2019 17:02:33 +0000 (17:02 +0000)]
Merge "Add missing header in gd/packet/parser/checksum_type_checker.h"
Myles Watson [Fri, 24 May 2019 21:23:41 +0000 (14:23 -0700)]
OS: Protect invalidation_list_ with the mutex
Test: AclManagerTest.startup_teardown
Change-Id: I5a9467691ffd1bb6bc2bb9d1080dce5f5bc655fa
Jakub Pawlowski [Tue, 28 May 2019 16:18:58 +0000 (09:18 -0700)]
GD ECC: add const qualifiers
am:
c49f35a9e7
Change-Id: I957679ee7cc5e3eb3f8407d6bd608066bac1695d
Jakub Pawlowski [Tue, 28 May 2019 15:55:47 +0000 (17:55 +0200)]
Add missing header in gd/packet/parser/checksum_type_checker.h
Test: compilation
Change-Id: I7a72cba471db5e59c8a6fa2ee1ba03cc258719ef
Jakub Pawlowski [Mon, 27 May 2019 18:25:49 +0000 (20:25 +0200)]
GD ECC: add const qualifiers
Test: SmpEccValidationTest
Bug:
126463795
Change-Id: I2e8b6027f959a181655286fb061bbb30b91e57f8
Ted Wang [Mon, 27 May 2019 02:01:26 +0000 (19:01 -0700)]
Merge "Disable absolute volume control for AirPods 2"
am:
d05a79af66
Change-Id: If83644e1903e4e2b8046ff8f21d04131ce9780f6
Ted Wang [Mon, 27 May 2019 01:50:04 +0000 (01:50 +0000)]
Merge "Disable absolute volume control for AirPods 2"
Jakub Pawlowski [Sat, 25 May 2019 13:21:24 +0000 (06:21 -0700)]
GD SMP ECC improvements part 1
am:
93ecb665a5
Change-Id: If3afe3f844e8d6134fba4af4105cb5386244923d
Jakub Pawlowski [Sat, 25 May 2019 09:06:40 +0000 (02:06 -0700)]
Merge "Eliptic Curve Cryptography for GD"
am:
e5656f79b2
Change-Id: I80ae2e38c782c96b37cc9603cd5eb7f4da71a239