GitLab

if an LE device is found during an LE scan, the BT host will try to fetch its name (creating an LE
connection) even though the advertising report received indicates that it is a non-connectable
advertising.
This is pointless and should not happen.

Change-Id: I246e4795b4f4e33f3202908ebff3724e8e89523c

Change-Id: I482f1096e87966f33ba8584ea60188abab3bcc42

If the string argument of send_at_cmd is a null pointer,
then the function should not crash: indeed this is a valid
parameters which can be used to send an AT cmd which doesn't
require any extra parameter.

Test: trivial fix. Existing unit tests still pass.
Change-Id: I3d83eebed660d703cd5a93a19dc73f4f354ab7bf
Signed-off-by: Sebastien Griffoul <sebastien.griffoul@intel.com>

Currently, only C++ code in the service/ folder can use libbase logging.
This patch makes sure that logging can be used and is properly
configured for usage when the stack is running as part of Bluetooth.apk.

Bug: 31806042
Change-Id: I1f8be79ba9999b53ece0b5217a893b4bd20ecafc

Currently after 100 devices are added to the device security database,
any subsequent LE connection to a newly discovered LE device address
will fail as entries in the security record database are not reused.

This patch removes a device record if the device itself is removed and
also ensures that the oldest device security record is deleted if the
limit is reached to ensure a new record can be allocated.

Bug: 31625900
Test: SL4A multi-device connection test + regression
Change-Id: I22f6c82c64a9a9bfb2a16d79182903e5aa011355
(cherry picked from commit 013c32bf)

Build-only change to support the multilib case where the second arch
is translated (and thus bluetooth library is needed).

"True" multilib case is explicitly not supported.

Test: Build system refactoring CL. Existing unit tests still pass.

BUG=31422117

Change-Id: I09f239d39f5dbe0848a89367327db1ea1074ca39

Right now, data types, control blocks, and functions used in the BTM
layer are defined in the same header files. This means that if someone
wants to write a test that uses those data types, they must also define
all control blocks, or compile the whole module.

This patch separates the data types from other definitions. Thanks to
it, we will be able to write unit tests, once other dependencies get
separated.

Change-Id: Ibc089e273cc37642fbb8672964b266c20f8d825d

Right now, LE multi advertising related code uses it's own
implementation of queue on top of static array. Use std::queue, and
base::Callback instead.

Changing function pointers to base::Callback instances is required for
further refactoring, which will turn this code into class.

Bug: 30622771
Test: sl4a BleAdvertiseApiTest ConcurrentBleAdvertisingTest
Change-Id: I839e036c6edb4e19a17def4d68d351296468f88c

In case the server cache does not exist for a specific
connection, the API used to copy the server DB into the
DB structure needs to return immediately. This change
makes sure of that and prevents dereferencing of a null
block causing the runtime error.

Change-Id: Iec3040a1280ef9d80b1b9c76eca8071dff499411

This patch converts btm_ble_multi_adv.c file to .cc. It also separate
header definitions from rest of BTM layer. It is another step towards
making BLE advertising-related code testable.

Bug: 30622771
Change-Id: I89965e10fbf773e48176ee19ef613dd0df2fb951

The value of codec_type can be extracted from the codec_info,
so it is not needed to pass it around as an argument, or keep
it as an field.

Also:
* Added new APIs and the corresponding unit tests:
  - A2D_GetPacketTimestamp()
  - A2D_BuildCodecHeader() - replacement for bta_av_sbc_bld_hdr()
  - A2D_CodecName()
* Replaced tBTA_AV_SEP.codec_type with tBTA_AV_SEP.codec_info

Bug: 30958229
Test: A2DP tested manually. Unit tests included in the CL.
Change-Id: I3bfd4750e8639d77a9938ab3c274c9d6ea859fb6

Bug: 31466840
Change-Id: Ibd8f8b85eb59be8bfbb8a7c83b5935802624a748

If enable the "connection parameters update" during GATT discovery,
it may cause some device connection failure due to connection timeout.
Since the the "connection parameter update" is enabled again when GATT
discovery completed, remove the enable of "connection parameter update"
after SMP complete to fix the issue.

This issue was already fixed in commit 2d41fe1c, but then it was
re-introduced in commit 444a8da8, which also removed comment explaining
the issue, added back in this patch.

Bug: 29060797
Change-Id: I42b0b36056821c30d887484e22bfcbd04ea7ca03

Bug: 31533166
Change-Id: Ia04e9c51280a3e379c0670b39a84eb5a53c5332b

Motorola Roadster is advertizing Absolute Volume but it has a
buggy implementation.

Bug: 29740052
Change-Id: I18c1a2f03617669138be2e7e1d9c7bb391cd1129

* changes:
  Use only PCM as the media feeding format
  Localize the usage of tA2D_SBC_CIE only within the SBC codec

* Removed field tA2D_AV_MEDIA_FEEDINGS.format, because we always
  use/assume the feeding format is PCM.
* Removed tA2D_AV_MEDIA_FEED_CFG_PCM, tA2D_AV_MEDIA_FEED_CFG,
  tA2D_AV_CODEC_PCM, and tA2D_AV_CODEC_NONE, because they are not
  needed/used anymore.
* Removed feeding mode:
  - tBTIF_MEDIA_INIT_AUDIO_FEEDING.feeding_mode
  - tBTIF_AV_FEEDING_MODE, BTIF_AV_FEEDING_ASYNCHRONOUS,
    BTIF_AV_FEEDING_SYNCHRONOUS
  - tBTIF_MEDIA_CB.feeding_mode
* Added new APIs and the corresponding unit tests:
  - A2D_CodecConfigMatchesCapabilities()
  - A2D_BuildSinkConfig()
  - A2D_CodecEquals()
  - A2D_CodecRequiresReconfig()

Also:
 * Removed header file btif/include/btif_av_api.h because it is not
   needed anymore.
 * Removed functions or typedefs that are unused or replaced:
   - bta_av_co_audio_codec_match() -> replaced by
     A2D_CodecConfigMatchesCapabilities()
   - bta_av_co_audio_codec_cfg_matches_caps()
   - bta_av_co_audio_peer_reset_config()
   - bta_av_co_audio_discard_config()
   - bta_av_co_get_codec_info()
   - bta_av_co_peer_cp_supported()
   - tA2D_AV_CODEC_ID
   - tBTIF_AV_CODEC_INFO
 * Increased the sampling_freq storage from uint16_t to uint32_t
 * Decreased the number of channels n_channels storage from uint16_t
   to uint8_t
 * Removed unneessary global function declarations inside btif_av_co.h
 * Changed the order of the A2D_BuildSrc2SinkConfig() arguments, so
   the order is more intuitive.
 * Misc other cleanup:
   - file bta_av_co.cc: code simplification, renamed snk -> sink,
     Removed unnecessary cast-away of a const argument: bta_av_co_cp_scmst
     Renamed/updated bta_av_co_audio_peer_supports_codec() ->
             bta_av_co_find_peer_sink_supports_codec()
     Renamed/updated bta_av_co_audio_peer_src_supports_codec() ->
             bta_av_co_find_peer_src_supports_codec()

Bug: 30958229
Test: A2DP tested manually. Unit tests included in the CL.
Change-Id: I60981c62b1a88491b5232c4f6515933c4261fac9

Bug: 31549206
Change-Id: I10504d73a962dde583ddb5771cb0a0fe6ca6d580

Test: make test-vendor

Change-Id: I6eefd385ecb992f854544ca92c5a8559b6f281f0

* Added new A2DP API and use it as appropriate
 - A2D_IsValidCodec()
 - A2D_CodecTypeEquals()
 - A2D_GetMinBitpool()
 - A2D_GetMaxBitpool()
* Removed A2D_GetDefaultConfigSbc(), because it is not needed anymore.
* Removed bta_av_co_audio_get_sbc_config() and replaced it
  with code that is not SBC codec-specific:
  bta_av_co_audio_encoder_init() and bta_av_co_audio_encoder_update()
* Removed bta_av_co_get_remote_bitpool_pref() because it is
  not used/needed anymore.
* Moved the following declarations the a2d_sbc.h header file to
  the a2d_sbc.c codec-specific implementation, because they don't need
  to be exposed anymore.
  - struct tA2D_SBC_CIE
  - function A2D_BldSbcInfo()
  - function A2D_ParsSbcInfo()
* Added corresponding unit tests for the new API

Bug: 30958229
Test: A2DP tested manually. Unit tests included in the CL.
Change-Id: Ib242956c303f0ca3c5f70ebc00416a7a45e70228

Bug: 31274906
Change-Id: Ifb4ee2a5f1ebcc14e2e1b60caa2fe6536d20946b

LE related code should not use BTM_ACL_IS_CONNECTED, because it assumes
classic transport.

In this particular case, connection created event, which was result of
advertising, caused cancellation of "LE Create Connection".
Attempt to send "LE Create Connection Cancel" was checking if connection
is already established, but was failing to do so becuase
BTM_ACL_IS_CONNECTED was used instead of btm_bda_to_acl with proper
transport. This caused just created connection to be dropped.

Bug: 31442085
Change-Id: Ia345a3d00e8567160f4abb2a31c4130ee69d67a3

Change-Id: I3750edd23be239d1b19aa1b7d949cfa3d074cb2f

Accessing p_lcb must be guarded with NULL check.

Bug: 31442085
Change-Id: I0f6e3f8e0247c5e8946c4d098f8a3ba442f5fb28

Cancelling discovery was not handled properly in
btif_dm_search_services_evt and was causing crashes due to assert in
default case.

Bug: 31442085
Change-Id: If9a2c203730c9b06df5ff8d8e251f0c35addcb82

Add a pointer to the C++ style guide.

Add a HOWTO for those who have outstanding patches that
they'd like to reformat.

Change-Id: Ia749c6498424cfbdf0bb759a1736978f91a34289

FOR_EACH_* macros are resolved to for loop. Using std::move inside it
will cause already moved object to be used when more than one observer
is registered. This patch fixes that.

This issue was already fixed in commit e84c4bdb, but was reintroduced in
a641b6fa.

Change-Id: I14b72371d369e4bdcaf3a6e30045bf8a1e8c3e1b

If the received HCI type is unknown, then log an event and abort.
The most likely reason for that to happen is if the UART stream
is corrupted. We cannot recover from that, and there is not much
else we can do.

Also, fixed a bug in an HCI-related unit test that was exposed
by the above change.

Bug: 31432127
Change-Id: Ia888c485f177af4962268bf8f593b27fd7a4b080