OSDN Git Service
Sal Savage [Fri, 19 Jun 2020 23:35:51 +0000 (23:35 +0000)]
Merge "Don't query for metadata on playing status" into rvc-dev
Sal Savage [Thu, 4 Jun 2020 20:35:39 +0000 (13:35 -0700)]
Don't query for metadata on playing status
These removed lines of code were for a problem that no longer seems to
exist with other phones. It was a band-aid at best, having us pull
metadaa on 'playing status' when a phone really should have just been
sending the update to us anyway. Pulling metadata here causes a bad user
experience with cover artwork where the new metadata is interpreted as a
track change in Java land and cover artwork can be redownloaded. This
happens primarily with non-database aware players after a BIP reconnect
where we're guarenteed nothing about the handle we receive.
Tag: #compatibility
Bug:
158232844
Test: atest BluetoothInstrumentationTests
Merged-In: Idb88e7c1e1553319cb26eba2c18e3cd2bbd552cc
Change-Id: Idb88e7c1e1553319cb26eba2c18e3cd2bbd552cc
Sanjay Ramankandath [Wed, 3 Jun 2020 22:24:06 +0000 (22:24 +0000)]
RootCanal: Use a state machine in ScriptedBeacon
Add indication for ble add playback completion
bt HCI: Make sure the device numbers match
Clean up static variables in scripted beacon
Clean up logs in scripted beacon
Test: boot cuttlefish and check logcat
Tag: #gd-refactor
Bug:
154491371
Bug:
158019041
(cherry picked from commit
79f168b7f3b03722cf2d0bb5ee97258824dc3832)
(cherry picked from commit
1c66bdaa38c46e859b66f8cc055993c5f0e7f5a1)
(cherry picked from commit
82656e9682e9de7596b7327618cf680936f34589)
Change-Id: I7f54ecef4ba08acc33b9d773b02b8428ca91a415
(cherry picked from commit
53c68dba78cc8f746c93e3db062ef358a976f0bf)
Merged-In: I7f54ecef4ba08acc33b9d773b02b8428ca91a415
Venkata Jagadeesh Garaga [Fri, 17 Apr 2020 10:36:17 +0000 (16:06 +0530)]
GAP: Use proper page scan rep mode for connection
Issue: Page time out seen when user initiated pairing from LE scan response
Steps to reproduce
==================
Steps:
1. Turn on BT
2. inquiry any dual mode device
3. Initiate pairing (only LE scan response received but br/edr
inquiry response not received)
Actual Result:
Remote name request or create connection got page timeout some times
Expected result:
Remote name request and create connection should be successful
Root Cause:
If user finds only LE scan response for dual mode devices, then stack
treats it as dual mode based on br/edr not supported bit. If pairing
initiated from LE scan response, then stack initiates RNR and ACL connection
on br/edr transport with wrong page scan response from LE inquiry database,
it leads to page time out.
Fix:
Use page scan response from inquiry data base if br/edr inquiry response
received, else use default page scan response R1 in RNR and create connection.
Test: Follow steps above
Bug:
154286616
Change-Id: I281ebc6d0e67d88c30328cb67df7f28793f354c1
Myles Watson [Fri, 29 May 2020 19:17:43 +0000 (12:17 -0700)]
RootCanal: Use GD defaults to define OS_ANDROID
When OS_ANDROID is defined logs go to logcat.
When OS_ANDROID is not defined logs to stderr.
Bug:
157706301
Test: cert/run --host
Tag: #gd-refactor
Change-Id: I3e735a4716fbc35a3d66e03909fdbf905436d285
Merged-In: I3e735a4716fbc35a3d66e03909fdbf905436d285
TreeHugger Robot [Fri, 29 May 2020 16:09:13 +0000 (16:09 +0000)]
Merge "Enable bitpool sanity checks" into rvc-dev
Greg Kaiser [Tue, 26 May 2020 12:42:36 +0000 (05:42 -0700)]
btif_config: Fix pointer to freed memory
We change the scope of our 'string' declaration so the contents
are still in scope when we use a pointer to them. Previously,
the contents were out of scope when we referenced them through
a pointer.
Bug:
148758680
Test: TreeHugger
Change-Id: Ib0aa689bf4547d079b77965d958cc382ba0276e6
Joseph Pirozzo [Mon, 25 May 2020 17:36:26 +0000 (10:36 -0700)]
Enable bitpool sanity checks
Enable bitpool sanity checks to run all the time, not just in debug
mode.
Tag: #security
Test: sbcdecoder_fuzzer
Bug:
146398979
Change-Id: Iff58305cd18de35e37290f0c09fba01ee14e787a
weichinweng [Mon, 27 Apr 2020 03:52:15 +0000 (11:52 +0800)]
NIAP: Add a map to store the LTTKM
When do pairing, the set_encrypt_key_or_remove_key will be later than
get_key due to they are in different thread. Use a map to store them and
will check whether the key exist in map. If no, will get the key from
the JAVA bluetoothKeystore.
Bug:
148758680
Test: m
Change-Id: I044b11a231c6f65978019ce6092657c1bdd814d1
Merged-In: I044b11a231c6f65978019ce6092657c1bdd814d1
weichinweng [Thu, 23 Apr 2020 09:33:02 +0000 (17:33 +0800)]
NIAP: Use bluetooth keystore to encrypt/decrypt LTTKM.
Bug:
148758680
Test: m
Change-Id: Ic1e4ddd186d900c5058f40e8eeef500ce23087b7
Merged-In: Ic1e4ddd186d900c5058f40e8eeef500ce23087b7
weichinweng [Tue, 21 Apr 2020 01:42:27 +0000 (09:42 +0800)]
NIAP: implement bluetooth keystore interface.(2/2)
Bug:
148758680
Test: m
Change-Id: I89f782039123d1deac173d13670de4d9dae5252a
Merged-In: I89f782039123d1deac173d13670de4d9dae5252a
weichinweng [Tue, 21 Apr 2020 02:26:26 +0000 (10:26 +0800)]
NIAP: Add new argument to determine config checksum check result.(2/4)
Bug:
148758680
Test: m
Change-Id: I07eb90a144e172212c7a1a55ef200111efd297e0
Merged-In: I07eb90a144e172212c7a1a55ef200111efd297e0
weichinweng [Mon, 20 Apr 2020 09:48:43 +0000 (17:48 +0800)]
NIAP: Add keystore.proto for java layer usage.
Add keystore.proto for java layer usage.
Bug:
148758680
Test: m
Change-Id: I8fecbcdc46eb12336c9fd1fa1e2166df80d80f19
Merged-In: I8fecbcdc46eb12336c9fd1fa1e2166df80d80f19
TreeHugger Robot [Thu, 14 May 2020 09:50:48 +0000 (09:50 +0000)]
Merge "BluetoothMetrics: Fix LRU eviction callback bug" into rvc-dev
Chen Chen [Wed, 13 May 2020 22:35:42 +0000 (15:35 -0700)]
BluetoothMetrics: Fix LRU eviction callback bug
Test: atest bluetooth_test_common
Bug:
156504089
Tag: #stability
Change-Id: I594070a0dc1c33bd8f971bb8a7752fd317872a55
Merged-In: I594070a0dc1c33bd8f971bb8a7752fd317872a55
(cherry picked from commit
184414edd1a0851892865241d9287ca0d808b54e)
Sanjay Ramankandath [Fri, 24 Apr 2020 14:40:59 +0000 (14:40 +0000)]
Support for reading playback file in scripted beacon
Bug:
154491371
Tag: #feature
Test: Manual test on cuttlefish
Test: gd/cert/run --host
Change-Id: Ie0ce31796a0c28826002b0677d27e5db075dde6d
Myles Watson [Mon, 20 Apr 2020 22:57:07 +0000 (15:57 -0700)]
RootCanal: Add a scripted beacon for testing
Bug:
154491371
Tag: #feature
Test: Scan for devices
Test: gd/cert/run --host
Change-Id: I3354d621d4253d051eb7b814de8d7889dcf99f64
Sal Savage [Thu, 30 Apr 2020 22:10:08 +0000 (15:10 -0700)]
Allow Java-land to request current track metadata through JNI
AVRCP Targets are not supposed to give image handles over with metadata
until an AVRCP BIP client connects from a particular controller. Because
this connection can happen any time and there's nothing forcing a target
to send us a track changed notification to inform us of the new handle,
the best thing we can do to work with all devices is to request it when
we connect on BIP. Otherwise, we risk a race condition between the BIP
connection and any amount of track changed notifications that might
happen organically. Our JNI doesn't current support requesting current
track metadata.
This change adds/implements a get_current_track_metadata() method to the
JNI. This, paired with a change in /packages/apps/Bluetooth/, fixes the
above mentioned race condition and keeps us from missing out on possible
cover art image handles.
Bug:
152655644
Test: Build, flash, test with AVRCP target cover art devices
Change-Id: I9d1f28068b4fded61d5336da8d864fbd9ef34f98
Merged-In: I9d1f28068b4fded61d5336da8d864fbd9ef34f98
TreeHugger Robot [Wed, 6 May 2020 17:22:06 +0000 (17:22 +0000)]
Merge "Check search state when cancel read remote name procedure" into rvc-dev
Cheney Ni [Tue, 5 May 2020 07:17:29 +0000 (15:17 +0800)]
service: A2DP source callback to query the preferred codec
In order to apply user's preferred codec while reconnecting, we need a
callback method to talk to Bluetooth stack whether the mandatory codec
has higher priority than others. This change implements a dummy method
in the fluoride service, so is working with the stack.
Bug:
134131114
Bug:
147572898
Test: bluetoothtbd_test
Change-Id: Ib99cf12eba85450526e22723d1509405469c2ebc
Merged-In: Ib99cf12eba85450526e22723d1509405469c2ebc
(cherry picked from commit
3254942410f182533675645392d7f72d14bf7543)
Cheney Ni [Tue, 14 Apr 2020 06:25:20 +0000 (14:25 +0800)]
A2DP: Don't select disabled optional codecs when local SRC reconnecting
When local SRC is establishing a connection, the Bluetooth stack follows
those default priorities to choose a codec, and A2dpService needs to
judge again after this connection up event. If a user prefers the
mandatory codec, the peer state is OPEN -> RECONFIG -> OPEN which
is redundant but lousy. Since codecs preference is saved at the
upper-layer, this change queries the mandatory codec, and raises
the priority for more preferred codec while reconnecting from local.
Bug:
134131114
Bug:
147572898
Test: Local A2DP initializes connections
Change-Id: I110ffa463dd4c9ec62e18e1843b5aebc84d73093
Merged-In: I110ffa463dd4c9ec62e18e1843b5aebc84d73093
(cherry picked from commit
2fda2f0e415b2da9e7d5201bf8089492db88e73a)
Jooyung Han [Thu, 16 Apr 2020 09:48:33 +0000 (18:48 +0900)]
Set min_sdk_version to be part of mainline modules
Modules contributing mainline modules (APK/APEX) should set
min_sdk_version as well as apex_available.
For now setting min_sdk_version doesn't change build outputs.
But build-time checks will be added soon.
Bug:
152655956
Test: m
Merged-In: Ifb9e0d4cd57a0e8212341cf493cd9fb20d0e264c
Change-Id: Ifb9e0d4cd57a0e8212341cf493cd9fb20d0e264c
(cherry picked from commit
da09207b693706b0ed5cc2fef65401dbe972925e)
TreeHugger Robot [Wed, 29 Apr 2020 02:08:38 +0000 (02:08 +0000)]
Merge "Revert "Add min_sdk_version:R to updatable apexes"" into rvc-dev
cheng_cai [Tue, 14 Apr 2020 06:19:16 +0000 (14:19 +0800)]
Check search state when cancel read remote name procedure
Application may call cancel_discovery when the stack is doing
create_bond. The bonding process fails, because cancel_discovery
causes to cancel remote name request procedure.
Bug:
154074383
Test: CtsVerifier->Bluetooth Test->Bluetooth HID Host
Change-Id: Icadf9abe342fffdf06b9623fb5c7e6de05a337bc
(cherry picked from commit
e3f9fe7425705dc819a638c158ad9f87cb2bd1eb)
Jooyung Han [Tue, 28 Apr 2020 18:07:40 +0000 (03:07 +0900)]
Revert "Add min_sdk_version:R to updatable apexes"
This reverts commit
971de2aa7e2227092051b32b3a805de3020ab74c.
Reason for revert: the BT apex is not yet updatable
Bug:
152655956
Test: m nothing
Change-Id: I91c516d4ce950ff2f70a1536152bac0cf878e919
Change-Id: I784c3ea6c6ebc95c6fce70dac76b90afc7d0bff6
Grzegorz Kołodziejczyk [Tue, 28 Apr 2020 10:26:43 +0000 (10:26 +0000)]
Update call of HAL 1.1 bluetooth HCI initialization
Updates initialization call if HAL for bluetooth is 1.1
Bug:
144413056
Change-Id: I7d8ac8af74303d15f83c3dbaea784ec1ad7d54fa
TreeHugger Robot [Thu, 23 Apr 2020 08:42:17 +0000 (08:42 +0000)]
Merge "NIAP: remove keystore related code for now." into rvc-dev
TreeHugger Robot [Wed, 22 Apr 2020 23:52:11 +0000 (23:52 +0000)]
Merge "Fix bugprone-use-after-move warnings" into rvc-dev
weichinweng [Wed, 8 Apr 2020 08:41:19 +0000 (16:41 +0800)]
Into the BG connection when the HA disconnects from remote side
In some cases, it will add the random address into the BG connection
after the HA device is connected via a random address. Then the
ble_bgconn will add the public address into the controller white list if
the random address has the public address. Finally, it will get a second
connection handle with the public address on the same HearingAid device.
The patch changes the behavior that when the HearingAid device
disconnects from the remote side, add the device into the BG connection.
Bug:
152517031
Test: 1.run unit test.
2.Forget/Repair HearingAid device from Bluetooth UI.
3.Disconnect/Reconnect HearingAid device from Bluetooth UI.
4.HearingAid device power off/on, check them can reconnect.
Change-Id: Id491faeaddaa32bbea0c88c7f0563d299f7f33f8
Merged-In: Id491faeaddaa32bbea0c88c7f0563d299f7f33f8
Chih-Hung Hsieh [Thu, 5 Mar 2020 23:17:42 +0000 (15:17 -0800)]
Fix bugprone-use-after-move warnings
Bug:
150783499
Test: TreeHugger
Change-Id: I004a4084f2d7ab1a1f2908e930a17527d3b0d0bf
Merged-In: I004a4084f2d7ab1a1f2908e930a17527d3b0d0bf
KimJeongYeon [Tue, 10 Mar 2020 04:00:40 +0000 (13:00 +0900)]
btsnoop: Fix btsnoop_net runs unexpected
By default, btsnoop_net was disabled using "BT_NET_DEBUG" macro.
But, it enabled actually and manages socket connection for sending
btsnoop logs. The reason was, TRUE/FALSE macro were not defined.
This fix including "bt_types.h" to work macro properly.
Bug:
154647071
Test: Turn on btsnoop logging and observe btsnoop_net runs or not.
Signed-off-by: KimJeongYeon <jeongyeon.kim@samsung.com>
Change-Id: I595d06510170d27fbf81068fa7ce8769eabd0992
Merged-In: I595d06510170d27fbf81068fa7ce8769eabd0992
(cherry picked from commit
bc6072c798a597624bda3e999c3432d384d851a0)
TreeHugger Robot [Mon, 20 Apr 2020 14:13:41 +0000 (14:13 +0000)]
Merge "Send the only song in queue as now playing song when get item attributes of now playing" into rvc-dev
Cheney Ni [Sat, 11 Apr 2020 21:19:31 +0000 (05:19 +0800)]
A2DP: Fix integer sanitizer in SBC encoder
There were three potential integer overflow within SBC software encoder:
* embdrv/sbc/encoder/srce/sbc_packing.c:144:38: runtime error: unsigned
integer overflow:
4294967231 + 8192 cannot be represented in type
'unsigned int'
* embdrv/sbc/encoder/srce/sbc_packing.c:147:9: runtime error: signed
integer overflow: 37932 * 65535 cannot be represented in type 'int'
* embdrv/sbc/encoder/srce/sbc_packing.c:147:9: runtime error: signed
integer overflow:
178177545 +
2146959360 cannot be represented in type
'int'
They were caught by the integer sanitizer, and
1. (*ps32SbPtr >> 2) is either greater than 0xFF00,0000 or less than
0x007F,FFFF, so just cast to a signed integer explicitly.
2. Positive integer between 0x8000,0000 ~ 0xFFFF,FFFF can't be
represented in type 'int', but is still feasible in 32-bits.
3. s32OutLow is the lower byte of a 64 bits integer, but can't have the
carry values which is only for the higher byte.
This change gives the compiler a signed 64-bits variable, and trusts it
to do better optimization at multiplication.
Bug:
153402404
Test: make sure there are no integer sanitization errors.
Change-Id: I5046a42f9927c1aa7c25da2828c4f921ba7a5021
Merged-In: I5046a42f9927c1aa7c25da2828c4f921ba7a5021
(cherry picked from commit
a42db783434da238e4daade95ce2adb1bca0f138)
Chris Manton [Tue, 10 Mar 2020 15:59:02 +0000 (08:59 -0700)]
DO NOT MERGE Check a2dp packet length is zero
already in master
Bug:
142546668
Test: net_test_stack_a2dp_native
Change-Id: I105b445293c02cb4f37c759fd5b05758fd4e3646
Merged-In: I105b445293c02cb4f37c759fd5b05758fd4e3646
(cherry picked from commit
96392b0f2cfb2adc72cc7cad0d74dec8f4041582)
weichinweng [Fri, 17 Apr 2020 10:27:44 +0000 (18:27 +0800)]
NIAP: remove keystore related code for now.
Since we need to use keystore function on JAVA layer. Remove keystore
related code for now.
Bug:
148758680
Test: m
Change-Id: I642078f11cdb858a739abe4ad69c36060ad0f99c
Merged-In: I642078f11cdb858a739abe4ad69c36060ad0f99c
Jiyong Park [Fri, 17 Apr 2020 03:55:59 +0000 (03:55 +0000)]
Revert "Mark updatable APEXes"
This reverts commit
f3d123910b4c899b3aed8d82272fa5d24b6dc426.
Reason for revert: the BT apex is not yet updatable
Bug:
149805758
Test: m
Change-Id: I92cf6ac12f989726a797fec6a0c0f121346c188e
Ted Wang [Thu, 26 Mar 2020 10:11:26 +0000 (18:11 +0800)]
Send the only song in queue as now playing song when get item attributes
of now playing
Bug:
152007084
Test: Manully
Change-Id: I1d31612f20fe21e8364c578718ca282eeb75eac0
Merged-In: I1d31612f20fe21e8364c578718ca282eeb75eac0
Joseph Pirozzo [Tue, 14 Apr 2020 22:58:56 +0000 (22:58 +0000)]
Merge "Fix potential OOB vulnerability when SBC framedata is prefetched" into rvc-dev
Sal Savage [Tue, 14 Apr 2020 19:22:21 +0000 (19:22 +0000)]
Merge "AVCRP Controller support Get Item Attributes" into rvc-dev
Cheney Ni [Wed, 19 Feb 2020 09:41:59 +0000 (17:41 +0800)]
A2DP: Wait for previous AVDTP instance to be clean up before re-enabling
Because more connections spend more time to disconnect, the stack may
still be busy at disabling AVDTP, if we re-enable the A2dpService
quickly. This change will defer BTA_AV_API_ENABLE and BTA_AV_API_REG
events if the stack is still disabling the previous instance, so gives
the stack half more seconds to handle its previous task.
Bug:
135655859
Test: manually
Change-Id: I8e700ce195aa634819ea1ad8cf5892cc35669e10
Merged-In: I8e700ce195aa634819ea1ad8cf5892cc35669e10
(cherry picked from commit
21fe257db23887b3000a6c8d38ee339fcccbdc2b)
Cheney Ni [Wed, 18 Mar 2020 08:37:29 +0000 (16:37 +0800)]
A2DP: Workaround to clean up previous AVDTP instance while re-enabling
If the user disables and re-enables A2dpService quickly, the stack has
chances to be still busy at dropping those out-of-date links, and this
is highly dependent on the handset. If the handset responds more slowly, it
spends more time to disable, but causes the new AVDTP instance won't
work. This change checks the previous disabling flag, and cleans up the
old instance up before re-enabling.
Bug:
135655859
Test: manually
Change-Id: I625cfffab6d4df327fb8159aa4b759a09d90e076
Merged-In: I625cfffab6d4df327fb8159aa4b759a09d90e076
(cherry picked from commit
07689d65f4c6fbc8b20dd0890657936fc6d6cb7b)
Cheney Ni [Fri, 27 Mar 2020 03:54:28 +0000 (11:54 +0800)]
A2DP: Be more careful in allocating a BtifAvPeer instance
While a new peer is coming either by in-coming or out-going connection,
the stack first tries to find or create an associated BtifAvPeer object
for it, but uses an unknown BTA handle if BTA_AV is still registering.
This handle will be updated after BTA_AV registered, but causes
a mismatch in the handle usage between BTIF and BTA layers since the
register process just finished. This change disallows to use an unknown
BTA handle, so it prevents such abnormal cases. Besides, it adds more
log messages about BTA_AV registering, and also makes sure the disabling
flag won't be activated after disabled.
Bug:
135655859
Bug:
152597903
Test: Switch BT state quickly between STATE_BLE_ON and STATE_ON
Change-Id: I9df3d64f301dffbecdeaf3de18dd455be1c63ce2
Merged-In: I9df3d64f301dffbecdeaf3de18dd455be1c63ce2
(cherry picked from commit
714e639cb7291ecfb9feb12acad9bc7c8f455ba0)
Sal Savage [Tue, 3 Mar 2020 22:43:06 +0000 (14:43 -0800)]
AVCRP Controller support Get Item Attributes
Add support for the GetItemAttributes command when the browsing chanel
is connected.
Bug:
149411071
Test: Play a song from a paired phone compare metadata and HCI logs.
Change-Id: I0ac3c3bb7c9d1c3c72f0e2bad13d276c836ecf6f
Merged-In: I0ac3c3bb7c9d1c3c72f0e2bad13d276c836ecf6f
Joseph Pirozzo [Tue, 31 Mar 2020 20:57:21 +0000 (13:57 -0700)]
Fix potential OOB vulnerability when SBC framedata is prefetched
Macros were filling a value buffer when reading bit stream data in
advance of any fetch being called, updated to only fetch data into
buffer when needed.
Bug:
110107252
Bug:
146032016
Bug:
146398979
Test: sbcdecoder_fuzzer
Change-Id: Ibb7793c9a015eaddcc8faa6ccfd8ec998d03ed86
Sterling Huber [Thu, 9 Apr 2020 17:39:35 +0000 (17:39 +0000)]
[automerger skipped] Merge "Check a2dp packet length is zero" into qt-dev am:
c909de62bf -s ours am:
db1917d8ae
Change-Id: Id1c8c3155d96ee1e050dd1d308ebeca838a22c8e
Sterling Huber [Thu, 9 Apr 2020 17:17:37 +0000 (17:17 +0000)]
[automerger skipped] Merge "Check a2dp packet length is zero" into qt-dev am:
c909de62bf -s ours
am skip reason: skipped by user cmanton
Change-Id: I8ff4f933b88c68473b378d0894b89ae8b10431fd
TreeHugger Robot [Wed, 8 Apr 2020 17:00:45 +0000 (17:00 +0000)]
Merge "btm: fixing oob write in multi-adv SetData." into qt-qpr1-dev am:
aeeb427c94
Change-Id: Ibb621c8984997088b6ec96cc61fc65733e73c673
TreeHugger Robot [Wed, 8 Apr 2020 16:40:55 +0000 (16:40 +0000)]
Merge "btm: fixing oob write in multi-adv SetData." into qt-qpr1-dev
Sterling Huber [Wed, 8 Apr 2020 00:13:33 +0000 (00:13 +0000)]
Merge "Check a2dp packet length is zero" into qt-dev
Jakub Pawlowski [Wed, 8 Apr 2020 00:10:53 +0000 (00:10 +0000)]
[automerger skipped] Fix potential stack overflow caused by integer overflow am:
1570b62c88 am:
5b4bf745e9 am:
6360e91bb3 am:
3cd74938d8 -s ours am:
db0c360dce -s ours am:
ca7a1eabe0 -s ours
am skip reason: Change-Id I0655b0b62301f78cd8705cc7b0e4fc11522f00ca with SHA-1
ec0d507ed2 is in history
Change-Id: Id31d6d296c0c76921d0b581362e7e1e0a75f7f48
Jakub Pawlowski [Tue, 7 Apr 2020 23:50:35 +0000 (23:50 +0000)]
[automerger skipped] Fix potential stack overflow caused by integer overflow am:
1570b62c88 am:
5b4bf745e9 am:
6360e91bb3 am:
3cd74938d8 -s ours am:
db0c360dce -s ours
am skip reason: Change-Id I0655b0b62301f78cd8705cc7b0e4fc11522f00ca with SHA-1
ec0d507ed2 is in history
Change-Id: I2748027eb3a50e6ed5130be43861fa69d794bee8
Jakub Pawlowski [Tue, 7 Apr 2020 23:38:36 +0000 (23:38 +0000)]
[automerger skipped] Fix potential stack overflow caused by integer overflow am:
1570b62c88 am:
5b4bf745e9 am:
6360e91bb3 am:
3cd74938d8 -s ours
am skip reason: Change-Id I0655b0b62301f78cd8705cc7b0e4fc11522f00ca with SHA-1
ec0d507ed2 is in history
Change-Id: Ife4a1d737088dbf51ac29f0a6411d9468b456593
Jakub Pawlowski [Tue, 7 Apr 2020 23:24:21 +0000 (23:24 +0000)]
Fix potential stack overflow caused by integer overflow am:
1570b62c88 am:
5b4bf745e9 am:
6360e91bb3
Change-Id: I1e23e75fd05e41b9cd31bb93b945135bef21c04f
Jakub Pawlowski [Tue, 7 Apr 2020 23:09:11 +0000 (23:09 +0000)]
Fix potential stack overflow caused by integer overflow am:
1570b62c88 am:
5b4bf745e9
Change-Id: I91417fff40a89042993a9fcbfaa6b7f45c162d89
Jakub Pawlowski [Tue, 7 Apr 2020 22:51:21 +0000 (22:51 +0000)]
Fix potential stack overflow caused by integer overflow am:
1570b62c88
Change-Id: Iaf75d5b63297d50e3115422f15fa8511133ef45a
Jakub Pawlowski [Tue, 7 Apr 2020 22:43:47 +0000 (22:43 +0000)]
[automerger skipped] Fix potential stack overflow caused by integer overflow am:
ec0d507ed2 am:
8ccc0c8d5f -s ours
am skip reason: Change-Id I0655b0b62301f78cd8705cc7b0e4fc11522f00ca with SHA-1
4c0c93eb61 is in history
Change-Id: I2d54dc3becde918d809bfc2f970bc192bb562b1b
Jakub Pawlowski [Tue, 7 Apr 2020 22:31:00 +0000 (22:31 +0000)]
Fix potential stack overflow caused by integer overflow am:
ec0d507ed2
Change-Id: Ia8ae7d5d144e3a5e94728dc236d9e7af6ddd3dd2
Chris Manton [Tue, 10 Mar 2020 15:59:02 +0000 (08:59 -0700)]
Check a2dp packet length is zero
Bug:
142546668
Test: net_test_stack_a2dp_native
Change-Id: I105b445293c02cb4f37c759fd5b05758fd4e3646
Merged-In: I105b445293c02cb4f37c759fd5b05758fd4e3646
Joseph Pirozzo [Tue, 7 Apr 2020 19:32:41 +0000 (19:32 +0000)]
Merge "Remove BTA_HF_CLIENT_SCO_LISTEN_E fallthrough" into rvc-dev
Jakub Pawlowski [Mon, 16 Mar 2020 11:09:15 +0000 (12:09 +0100)]
Fix potential stack overflow caused by integer overflow
Bug:
151155194
Change-Id: I0655b0b62301f78cd8705cc7b0e4fc11522f00ca
Joseph Pirozzo [Tue, 7 Apr 2020 16:00:03 +0000 (09:00 -0700)]
Remove BTA_HF_CLIENT_SCO_LISTEN_E fallthrough
Remove unintended fallthrough.
Bug:
143901894
Test: connect SCO from both HF and AG.
Change-Id: Ifa5363479d4c844ed0b6a8585c25147b23156599
Merged-In: Ifa5363479d4c844ed0b6a8585c25147b23156599
TreeHugger Robot [Mon, 6 Apr 2020 14:49:41 +0000 (14:49 +0000)]
Merge "A2DP: Do the connection task in the stack main thread" into rvc-dev
Jooyung Han [Sun, 5 Apr 2020 00:14:36 +0000 (00:14 +0000)]
Merge "Add min_sdk_version:R to updatable apexes" into rvc-dev
TreeHugger Robot [Fri, 3 Apr 2020 19:22:52 +0000 (19:22 +0000)]
Merge "BluetoothMetrics: Add SDP DI manufacturers info to Bluetooth config; Add SDP ID info and LMP info to Bluetooth log" into rvc-dev
Cheney Ni [Tue, 31 Mar 2020 10:38:49 +0000 (18:38 +0800)]
A2DP: Do the connection task in the stack main thread
There is a profile queue to handle connection tasks sequentially, and
invokes the connect API in JNI thread while the previous task finishes.
However, since BtifAvStateMachine is running in the stack main thread,
it brings a race condition about the BtifAVPeer object between the JNI
and stack main threads. This change does A2DP connection task in the
main thread, so has no such race conditions.
Bug:
152595833
Test: make sure the opening event is in main thread manually
Change-Id: Ib40dc7c5fa3e839a894d6f7947ac8266eacd2022
Merged-In: Ib40dc7c5fa3e839a894d6f7947ac8266eacd2022
(cherry picked from commit
367c497abbc0ff5551a1b3210db35ad6540e2dfe)
Rahul Sabnis [Fri, 3 Apr 2020 00:02:36 +0000 (17:02 -0700)]
Add rahulsabnis@ as an owner of bluetooth code to allow for owner
code reviews
Bug:
153115116
Test: Manual
Merged-In: I8153358384bbf18102f84f544551d358d95b45e7
Change-Id: I8153358384bbf18102f84f544551d358d95b45e7
Chen Chen [Wed, 1 Apr 2020 02:12:39 +0000 (19:12 -0700)]
BluetoothMetrics: Add SDP DI manufacturers info to Bluetooth config;
Add SDP ID info and LMP info to Bluetooth log
Bug:
142563377
Test: manually check Bluetooth config file
Change-Id: I4a7c348ac570a730cccc31e51bdc9669847f96a5
Jooyung Han [Mon, 30 Mar 2020 04:27:34 +0000 (13:27 +0900)]
Add min_sdk_version:R to updatable apexes
APEXes introduced in R need to set min_sdk_version to ensure that they
are built against correct version(30 or R) of stubs (libc/liblog/...).
Bug:
152655956
Test: /vendor/google/build/build_mainline_modules.sh
Change-Id: I055f46f574c7349f46b8c1cb6bb39a515ae69967
Ugo Yu [Tue, 31 Mar 2020 04:41:43 +0000 (04:41 +0000)]
Merge "Replacing updateBleAppCount to enableBle and disableBle" into rvc-dev
TreeHugger Robot [Mon, 30 Mar 2020 14:17:34 +0000 (14:17 +0000)]
Merge "Retry service discovery for Avrcp if due to sdp connection failed" into rvc-dev
Ugo Yu [Tue, 17 Mar 2020 11:52:41 +0000 (19:52 +0800)]
Replacing updateBleAppCount to enableBle and disableBle
Bug:
139454316
Test: enableBle() -> disableBle() and check Bluetooth is OFF.
Change-Id: Ifa0b26af7b5c1eef65658948cffe6e409a787344
Chen Chen [Sat, 28 Mar 2020 00:54:26 +0000 (00:54 +0000)]
Merge "Bluetooth Metric Id: Add metric id allocation to Bluetooth metric calls" into rvc-dev
TreeHugger Robot [Fri, 27 Mar 2020 21:06:22 +0000 (21:06 +0000)]
Merge "BluetoothAudioHAL: Have full initialization for audio hardware objects" into rvc-dev
weichinweng [Thu, 12 Mar 2020 06:47:27 +0000 (14:47 +0800)]
HFP: Cleanup bt_hf_callbacks on the jni thread.
For the race condition, the bt_hf_callbacks will be set null on main
thread when doing btif_hf_upstreams_evt on jni thread. Then it will
trigger crash due to btif_hf_upstreams_evt calling a null point.
The patch is cleanup bt_hf_callbacks on the jni thread to avoid the race
condition.
Bug:
140781191
Test: manual, unit test, auto script stress test.
Change-Id: I67400749120aebe13fa3e791f624c8e595d63379
Merged-In: I67400749120aebe13fa3e791f624c8e595d63379
Chris Manton [Fri, 27 Mar 2020 00:49:20 +0000 (00:49 +0000)]
[automerger skipped] Merge "DO NOT MERGE Fix potential overflow in btif_rc" into qt-qpr1-dev am:
6a91b4d117 -s ours
am skip reason: subject contains skip directive
Change-Id: Ieff23e6c5693306d2c8b9402ed7788808f812ac3
Chris Manton [Fri, 27 Mar 2020 00:36:19 +0000 (00:36 +0000)]
Merge "DO NOT MERGE Fix potential overflow in btif_rc" into qt-qpr1-dev
TreeHugger Robot [Thu, 26 Mar 2020 21:58:22 +0000 (21:58 +0000)]
Merge "Fix potential OOB vulnerability when an HCI event is received" into qt-qpr1-dev am:
ee0e886a89
Change-Id: I18a2a32e6f34602f76d3e018a29e9bd24694c7c5
Kelly Rossmoyer [Thu, 26 Mar 2020 21:58:08 +0000 (21:58 +0000)]
[automerger skipped] Revert "DO NOT MERGE Ensure hci command status event has sufficient packet length" am:
006e8e481c -s ours
am skip reason: subject contains skip directive
Change-Id: Idb50aa70e42f832df92cdaaaf705a6cec38f67dd
TreeHugger Robot [Thu, 26 Mar 2020 21:45:36 +0000 (21:45 +0000)]
Merge "Fix potential OOB vulnerability when an HCI event is received" into qt-qpr1-dev
Chris Manton [Wed, 4 Mar 2020 04:04:51 +0000 (20:04 -0800)]
DO NOT MERGE Fix potential overflow in btif_rc
Bug:
142878416
Test: net_test_btif_rc
Change-Id: Ia263bd5c863644f2adde759a103d79b812a9a5de
(cherry picked from commit
e5dd9c735a30188286c7c9b88c247e4f629bdd54)
Kelly Rossmoyer [Thu, 26 Mar 2020 20:57:42 +0000 (20:57 +0000)]
Revert "DO NOT MERGE Ensure hci command status event has sufficient packet length"
This reverts commit
179e1ee138a8811deffd49eae77f85246e5092a9.
Reason for revert: build breakage (b/
152543627)
Change-Id: I437c99613732a5feb4ecbd082598f2efa5853c1e
TreeHugger Robot [Thu, 26 Mar 2020 19:43:52 +0000 (19:43 +0000)]
[automerger skipped] Merge "DO NOT MERGE Ensure hci command status event has sufficient packet length" into qt-qpr1-dev am:
466138189c -s ours
am skip reason: subject contains skip directive
Change-Id: I3e56d0584e861ec214834a7568cf48c8b2d65f19
TreeHugger Robot [Thu, 26 Mar 2020 19:27:55 +0000 (19:27 +0000)]
Merge "DO NOT MERGE Ensure hci command status event has sufficient packet length" into qt-qpr1-dev
Ted Wang [Tue, 17 Mar 2020 12:35:02 +0000 (20:35 +0800)]
Retry service discovery for Avrcp if due to sdp connection failed
Some remote devices are not allow multiple l2cap connection with same
psm, so service discovery for Avrcp might failed due to this reason,
causing absolute volume not enable on this device.
Retry one time if the reason of service discovery fail reason is sdp
connection failed.
Bug:
150354122
Test: Manually Test
Merged-In: I3bb511434788211950884f72ff0a1256533aaa05
Change-Id: I3bb511434788211950884f72ff0a1256533aaa05
Cheney Ni [Wed, 25 Mar 2020 10:37:33 +0000 (18:37 +0800)]
BluetoothAudioHAL: Have full initialization for audio hardware objects
Audio frameworks usually check whether a module has specific features by
particular function pointers, and must be null if unsupported. If a
module is not fully initialized, some pointers would be undefined, but
causes abnormal crashes when accessed by audio frameworks.
Bug:
152379666
Test: manually
Change-Id: Id6a79b651f2acbd35bd509ba36a80adea5c13e05
Merged-In: Id6a79b651f2acbd35bd509ba36a80adea5c13e05
(cherry picked from commit
7397c51ac706d3577b277d7b5bcdc78c78fcc6ac)
Cheney Ni [Mon, 23 Mar 2020 14:40:39 +0000 (22:40 +0800)]
A2DP: Remember every peer's AVDTP delay reporting
Almost every headset has its own delay reporting. This change remembers
those particular values individually, and applies when the peer is the
active device.
Bug:
152276107
Test: manually
Change-Id: I634dfd18a034fe4020f48139d7e0f14a1309098a
Merged-In: I634dfd18a034fe4020f48139d7e0f14a1309098a
(cherry picked from commit
0bf47fa2cf71188728aac2d232e2f02e8661d92c)
Chen Chen [Tue, 24 Mar 2020 21:52:52 +0000 (14:52 -0700)]
BluetoothMetrics: Return true if SaveDevice is called twice. Add more
logging
Test: atest BluetoothMetricIdAllocatorTest
Bug:
152251473
Change-Id: I3d5ee8d086a4e589e84a766d3d0923d3ae2278ba
TreeHugger Robot [Mon, 23 Mar 2020 05:53:51 +0000 (05:53 +0000)]
[automerger skipped] Merge "AVDTP: Prevent OOB read when parsing rejected response" into qt-qpr1-dev am:
f021675ca6 -s ours
am skip reason: Change-Id Iddb887c79bd8a2caa2ae5f21af15219807f9dd63 with SHA-1
c49665a9f9 is in history
Change-Id: I68f5adcc7dc7f005c3d484ae68472635cbda02b1
TreeHugger Robot [Mon, 23 Mar 2020 05:36:25 +0000 (05:36 +0000)]
Merge "AVDTP: Prevent OOB read when parsing rejected response" into qt-qpr1-dev
Rahul Sabnis [Fri, 20 Mar 2020 02:40:40 +0000 (19:40 -0700)]
Re-add binder method for getPriority in IBluetoothA2dp.aidl as well as
getPriority and setPriority in IBluetoothHeadset.aidl
Bug:
148689314
Test: Manual
Change-Id: Id9efe237b24dbf24e91491826d34f61cbb761991
Chen Chen [Sat, 14 Mar 2020 00:00:24 +0000 (17:00 -0700)]
Bluetooth Metrics: Init and Close MetricIdAllocator and use its APIs
Test: Manually pair device and cat config file
Bug:
142563377
Change-Id: I4492dc9b4c9d810e7919e97ef36ea151b76b2dbc
Merged-In: I4492dc9b4c9d810e7919e97ef36ea151b76b2dbc
(cherry picked from commit
f6a356ead8e711e3b94e0e915a5b94cfac978ffa)
Joseph Pirozzo [Fri, 20 Mar 2020 17:28:47 +0000 (17:28 +0000)]
[automerger skipped] Merge changes I685873b0,Iac3078fa into qt-qpr1-dev am:
2c4f9ef514 -s ours
am skip reason: Change-Id I685873b0c4c74ddb4e273e2a38307ec2af0bbd13 with SHA-1
daa34793f2 is in history
Change-Id: Ia10cdd7d6309e64bae3d59de1a77620eccb6d86e
Joseph Pirozzo [Fri, 20 Mar 2020 17:19:37 +0000 (17:19 +0000)]
Merge changes I685873b0,Iac3078fa into qt-qpr1-dev
* changes:
HF_Client: Send BTA_HF_CLIENT_RFC_CLOSE_EVT when client_cb == NULL
HF_Client: Free the RFC if the handle doesn't match
Jakub Pawlowski [Fri, 20 Mar 2020 14:24:00 +0000 (15:24 +0100)]
Fix potential stack overflow caused by integer overflow
Bug:
151155194
Merged-In: I0655b0b62301f78cd8705cc7b0e4fc11522f00ca
Change-Id: I0655b0b62301f78cd8705cc7b0e4fc11522f00ca
Jakub Pawlowski [Mon, 16 Mar 2020 11:09:15 +0000 (12:09 +0100)]
Fix potential stack overflow caused by integer overflow
Bug:
151155194
Merged-In: I0655b0b62301f78cd8705cc7b0e4fc11522f00ca
Change-Id: I0655b0b62301f78cd8705cc7b0e4fc11522f00ca
TreeHugger Robot [Fri, 20 Mar 2020 03:44:15 +0000 (03:44 +0000)]
[automerger skipped] Merge "Fix bluetooth can't turn off during network reset (3/3)" into qt-qpr1-dev am:
60af186b8e -s ours
am skip reason: Change-Id Ic36ffd63c376f84e3c9f0388820da86f63465c8f with SHA-1
9adf47f6c8 is in history
Change-Id: I20a62a958b5d6ce65712af48eb31fe7d0bf489c3
TreeHugger Robot [Fri, 20 Mar 2020 03:43:16 +0000 (03:43 +0000)]
Merge "avrc: Validating msg size before accessing fields" into qt-qpr1-dev am:
49437a62bd
Change-Id: I6ff012808218c6026584d4dc9217b99d8c6a3169
TreeHugger Robot [Fri, 20 Mar 2020 03:33:52 +0000 (03:33 +0000)]
Merge "Fix bluetooth can't turn off during network reset (3/3)" into qt-qpr1-dev
TreeHugger Robot [Fri, 20 Mar 2020 03:29:10 +0000 (03:29 +0000)]
Merge "avrc: Validating msg size before accessing fields" into qt-qpr1-dev
TreeHugger Robot [Fri, 20 Mar 2020 03:03:23 +0000 (03:03 +0000)]
Merge "Fix potential OOB vulnerability when an HCI event is received" into rvc-dev