| 9e030fde05352ec4385d7baf6cc2af89e95e039c |
|
25-Jun-2017 |
Jakub Pawlowski <jpawlowski@google.com> |
Rename bt_bdaddr_t into RawAddress (3/3)
Test: compilation test
Change-Id: I4571721a0c6165a66450ee157a53d8d15bfc45d0
/system/bt/stack/gatt/gatt_int.h
|
| c2276b06572ab6fc1f900fbb1f41087e77d47e2a |
|
10-Jun-2017 |
Jakub Pawlowski <jpawlowski@google.com> |
Use bt_bdaddr_t instead of BD_ADDR
Test: compilation test, sl4a GattReadTest
Change-Id: I8d1bd6914aec55bb53495b1d0d5e3d37b86865e6
/system/bt/stack/gatt/gatt_int.h
|
| d8be0e590aa1efe1c33a8acfea0d7e14fd1a8030 |
|
09-Jun-2017 |
Jakub Pawlowski <jpawlowski@google.com> |
Get rid of GATT_TRACE_* logs in favour of base/logging.h
Change-Id: Id77d313c5bff5e611441599da351a766d0ce5916
/system/bt/stack/gatt/gatt_int.h
|
| ed55662b4885a5247bb499fa887b0e77a8cab881 |
|
03-Jun-2017 |
Jakub Pawlowski <jpawlowski@google.com> |
Simplify read/write start
When read or write is started with insuficcient encryption, and the
encryption fails, it will either return immediate error
GATT_NO_RESOURCES, or the operation will be enqueued. In case the
enqueued operation fails again on the BTM_SetEncryption, no callback
will never be called, and no operation error will be returned. The
gatt_end_operation will never get called for the CLCB, and it will never
be freed, causing memory leak.
This patch fixes this by making sure that the call to
gatt_security_check_start always consumes the CLCB - either by executing
operation, queuing it, or returning error. This also ensures that if
BTM_SetEncryption fails, it will always be handled the same way - by
calling the callback.
Test: manual
Change-Id: Ibc20c1101cc7b5b6043e75df93b23a5b24b791e9
/system/bt/stack/gatt/gatt_int.h
|
| 3dda89880b66b0f04f684d138e7c3dc9a0e61597 |
|
06-Jun-2017 |
Jakub Pawlowski <jpawlowski@google.com> |
Use std::unordered_set for keeping app_hold_link
Change-Id: Ia8a0c9a2264979c1ff89faa5ddf17a83fcb7c7ce
/system/bt/stack/gatt/gatt_int.h
|
| e4f42abab339b3329aec07502199e11d3d8b3d80 |
|
31-May-2017 |
Jakub Pawlowski <jpawlowski@google.com> |
Get rid of clcb_idx in favour of pointers
Test: compilation test
Change-Id: Ib2c4b13b85343693e6cff4a94eed49382e3aec00
/system/bt/stack/gatt/gatt_int.h
|
| d1a9313471ed5290ca560305978c3107020272c2 |
|
31-May-2017 |
Jakub Pawlowski <jpawlowski@google.com> |
use std::queue for keeping tGATT_CMD_Q
Test: compilation
Change-Id: I4df9a075d27a306067c48c652f5da1fb156c4a9b
/system/bt/stack/gatt/gatt_int.h
|
| 7cfd8391f6db2d387ea556ca2b7be0a7e4791e5d |
|
06-Jun-2017 |
Jakub Pawlowski <jpawlowski@google.com> |
Refactor tGATT_BG_CONN_DEV
* Use list for keeping the devices instead of fixes array, get rid of
the in_use field.
* Use unordered_set instead of array to keep gatt_if of apps interested
in specific device.
* the in_use flag was never cleared before, this had the potential to
fill the whole array when multiple devices were addred/removed to the
whitelist. This issue is fixed now.
Test: compilation test
Change-Id: Id87c2ae9c5b0ce0887a4fa9be50e34d646d22a13
/system/bt/stack/gatt/gatt_int.h
|
| f4c0292c63085cd4d0f5c001974b6503aeaf8ed3 |
|
30-May-2017 |
Jakub Pawlowski <jpawlowski@google.com> |
Pass GATT cb's by reference and other style fixes
Test: compilation test
Change-Id: I410d0d0a0bf49b9f3ca5e034c66fe8e82661cee7
/system/bt/stack/gatt/gatt_int.h
|
| eb3ab848f8361d54faa3a997fbecfe48af9de731 |
|
30-May-2017 |
Jakub Pawlowski <jpawlowski@google.com> |
Use std::queue for pending_enc_clcb
Test: 62188929
Change-Id: Ide07f15c5e5b6bc8b93ac350081b7bef7c7b3938
/system/bt/stack/gatt/gatt_int.h
|
| 8663a87ea030b8b7cba15772e95019e9ff4ea97f |
|
27-May-2017 |
Jakub Pawlowski <jpawlowski@google.com> |
Get rid of unused tGATT_SCCB
Test: compilation test
Change-Id: I32fee6a93442c8415f7d855c1e4802bceb6fe5ff
/system/bt/stack/gatt/gatt_int.h
|
| 6395f1547ed6179dcdeed392bd9909d18772fb25 |
|
09-May-2017 |
Jakub Pawlowski <jpawlowski@google.com> |
Improve GATT Server database handling.
- use references instead of pointers where possible. Thanks to it we can
remove some reduntant null checks and simplify the code.
- use directly allocated memory instead of own buffers. Thanks to it the
stack uses around 12kb less of memory after startup.
- use list and vector from std library, instead of some hand-written
implementations.
This patch is a prerequisite for further server refactoring, that will
further reduce the space used, make unit-testing possible, and enable
proper handling of PTS GATT tests.
Test: sl4a GattReadTest
Bug: 38225928
Change-Id: I1620be682259ccb8f0c02754806e355e3f1ad0c1
/system/bt/stack/gatt/gatt_int.h
|
| 96fb273d92a31204edcf43ab2594a19876d14056 |
|
25-Mar-2017 |
Jakub Pawlowski <jpawlowski@google.com> |
Add option to specify initial LE connection PHY (1/3)
For whitelist connections we always use all possible PHYs, for direct
connection use PHY specified by client.
Test: manual
Bug: 30622771
Change-Id: I720f134e2800dc3d282135bb7ffbe3882117c680
/system/bt/stack/gatt/gatt_int.h
|
| eafd45d08653bb1621c82a2f3cf922a43a0b1bc5 |
|
23-Mar-2017 |
Jakub Pawlowski <jpawlowski@google.com> |
Set preferred PHY and read PHY implementation (3/3)
Test: manual
Bug: 30622771
Change-Id: I4267238a0b5c7bc373ae1846ebd19a716881a4ec
/system/bt/stack/gatt/gatt_int.h
|
| ad1e23d50e431fb82ff1b8d3ce02f64ca0c5a940 |
|
10-Dec-2016 |
Jack He <siyuanh@google.com> |
Remove extern "C" from header files
Since change 290046, most files in system/bt is compiled as C++ source.
Therefore, there is no longer a need for the extern "C" wrapper around
includes that export symbols from those sources.
The following python script is applied to each file in the directory:
front = '#ifdef\s+__cplusplus\s+extern\s+"C"\s+{\s+#endif\s+'
back = '#ifdef\s+__cplusplus\s+}\s+#endif'
with open(sys.argv[1], "r") as f:
data = f.read()
data = re.sub(front, "", data)
data = re.sub(back, "", data)
print data
through a shell script:
for file in $(find . -name "*.h"); do
python remove_cpp_extern_c.py $file > tmp
cat tmp > $file
rm tmp
done
with following files not edited:
* stack/include/a2dp_*
* include/bt_trace.h
* embdrv/sbc/*
Bug: 33492510
Test: Code compilation, BtFunhausMetricsTest, BtStressTest
Change-Id: Iac21cdfb1924b50478dd0b82326e092602cbc9d4
/system/bt/stack/gatt/gatt_int.h
|
| 911d1ae03efec2d54c3b1b605589d790d1745488 |
|
29-Nov-2016 |
Myles Watson <mylesgw@google.com> |
Apply clang-format to the rest of the tree
find * | grep "\.[ch]" | xargs clang-format --style=file -i
Test: mma -j32
Change-Id: I6fcc9862bb7bc07c2a367ca58fef2b3cd27a6f05
/system/bt/stack/gatt/gatt_int.h
|
| a60c5fdaa2e22bd0a65139353b346f43ca2c6dd7 |
|
15-Nov-2016 |
Jakub Pawlowski <jpawlowski@google.com> |
Remove unused GATT_Listen
Bug: 30622771
Test: still builds
Change-Id: I4003a1241e611c9c1f2a21090ea3ba7530b0643a
/system/bt/stack/gatt/gatt_int.h
|
| 2e8c421ddce1dae511a39b8e89415e8eb1944bfd |
|
29-Oct-2016 |
Jakub Pawlowski <jpawlowski@google.com> |
Remove advertise whitelist functionality
Advertise filtering is never used. It is not finished - adding device to
whitelist can never succed. It is also making advertising API unification
hard.
Bug: 30622771
Test: no tests necessary
Change-Id: I9df9ea18b265a580c2fcdb28a6d6ce4be43f0a24
/system/bt/stack/gatt/gatt_int.h
|
| b2a292b5d8df2f359c38b0787bc01181225a9bc9 |
|
15-Oct-2016 |
Pavlin Radoslavov <pavlin@google.com> |
Renamed most C files to C++: *.c to *.cc
Also:
- Fixed C++ related compilation errors.
- Added missing 'extern "C"' guards in some of the header files.
- Added missing LOCAL_CPP_EXTENSION to Android.mk files.
- Added-back btif/src/btif_mce.cc and bta/mce/bta_mce_* to
btif/Android.mk and bta/Android.mk respectively.
- Fixed the alphabetical ordering of the *.cc files in some
of the Android.mk files.
- Added missing Copyright header to "osi/include/list.h"
- Updated "osi/src/wakelock.cc" to use C++ std::string
instead of dynamic allocation of C-style strings.
Test: code compilation, unit tests, and A2DP streaming
Change-Id: Ia2f7215ed9df32775c701b68fc86b09875b942c7
/system/bt/stack/gatt/gatt_int.h
|
| 37c1b5f8ccf4fc6d5132e07f98f0f40f65d5bfaf |
|
15-Oct-2016 |
Pavlin Radoslavov <pavlin@google.com> |
Removed unused *_DYNAMIC_MEMORY conditional statements
Test: compilation from the top-level directory
Change-Id: I181e6ae2e71c232ca0dc733dabda5cb0fc2672d8
/system/bt/stack/gatt/gatt_int.h
|
| 109891d07f7946c69070a01e0ebef56ee3eb7707 |
|
18-Jul-2016 |
Jakub Pawlowski <jpawlowski@google.com> |
Make gatt_attr use opportunistic connection
This patch creates new connection mode - opportunistic connection. When
such connection is made, no call to gatt_update_app_hold_link_status
will be made when it's started.
This means that connecting and disconnecting in this mode won't trigger
disconnect timer. When other, app creates regular connection and then
disconnects, the physical connection might get disconnected.
Opportunistic connection is used only for code setting CCC right now.
Bug: 30186455
Change-Id: Ia5be7682b6c3dcb4993060f818dee603aef3e19e
/system/bt/stack/gatt/gatt_int.h
|
| d19e0785e662e640191a075eda07acce61c2aeda |
|
15-Jul-2016 |
Marie Janssen <jamuraa@google.com> |
Use standard types, consistent ifdef style everywhere
Remove the typedefs in stack/include/bt_types.h
Use standard types everywhere.
Use standard style for #if statements:
- #if (VAR_NAME == TRUE)
- #if (VAR_NAME1 == TRUE && VAR_NAME2 == TRUE)
Use __func__ instead of __FUNCTION__
Fix some debug statements to use __func__
Update script to be less disruptive to aligned assignment blocks.
Change-Id: I8f8f068e6c26ce74fd3b3707e1e31fd0b919cdd0
/system/bt/stack/gatt/gatt_int.h
|
| a641b6fa2a25e1b5382945d13c4fa49d36084a78 |
|
26-Mar-2016 |
Jakub Pawlowski <jpawlowski@google.com> |
GATT Server refactoring (4/4)
Right now in order to add gatt service, we need to add it piece by
piece - each characteristic and descriptor separately. This causes lots
of concurrency errors in higher layers. Instead accept whole service
definition and start whole service at once.
Also, use handles to identify GATT server attributes. It's already done
on clinet side.
Bug: 27999121
Change-Id: I42c796809423d116fe6755ef8b1c40da270427a5
/system/bt/stack/gatt/gatt_int.h
|
| c88b20a599a52228c7b29d03751b7f13868d417a |
|
22-Jun-2016 |
Jakub Pawlowski <jpawlowski@google.com> |
GATT: use one type for all server attributes
Right now we have three types, for attributes depending on how long
the uuid is. Instead, use one type.
Change-Id: I33bb4f3c57b7b1516eda871a77b71661b7ba39ce
/system/bt/stack/gatt/gatt_int.h
|
| 3d6accfcc0bdc5c7a8713320c9bb069cbf253348 |
|
11-May-2016 |
Bryce Lee <brycelee@google.com> |
Move extern "C" to top of headers and remove tBTA_HF_CLIENT_HDR.
Previously, extern "C" was placed in various locations within the header
files, sometimes below actual struct declarations. Doing so would lead to
alignment issues between C and C++ code.
tBTA_HF_CLIENT_HDR was removed from bta_hf_client_api since it was
extraneous and empty - leading to alignment issues.
Change-Id: Icdd338f1affe4c3a70c7bbd716249be6a16443d2
/system/bt/stack/gatt/gatt_int.h
|
| 78bcff79e1b1f0efce436b33bdd6da88745bfc8a |
|
05-Dec-2015 |
Pavlin Radoslavov <pavlin@google.com> |
Refactor the Bluetooth timers
* Updated the alarm API:
- Existing API alarm_new() is modified to take an alarm name
as an argument.
- New API alarm_new_periodic() is used to create a periodic
alarm.
- Added new API alarm_is_scheduled() to test whether an alarm is
scheduled.
- Existing API alarm_set_periodic() is removed: a periodic
alarm is created by alarm_new_periodic().
- Added new API alarm_set_on_queue() to set an alarm whose
callback is executed on a specific queue.
- Added new API alarm_register_processing_queue() and
alarm_unregister_processing_queue() to register/unregister
a queue and the corresponding thread for alarm processing.
- Added corresponding unit tests.
* Updated the alarm internals:
- Added alarm_info_t for collecting alarm-related information
and statistics.
- Collect and store alarm-related statistics into alarm_info_t
per alarm.
- Include the alarm-related statistics and info into the native
dumpsys output for Bluetooth.
- Once an alarm expires, the alarm execution is scheduled for
processing on another internal alarm-specific thread, not
on the thread that is maintaining the alarms.
- Implemented callback execution ordering guarantee among
timers on the same thread with exactly same timeout values.
* Refactor some of the usage of alarm_set() and simplify the
code by using alarm_set_on_queue() instead.
* Removed the non_repeating timers wrapper, and use directly
the alarm mechanism / API.
* Refactored all timer_entry_t timers and replaced them with alarm_t
timers:
- Replaced the btu_start_timer() / btu_stop_timer() /
btu_start_quick_timer() / btu_stop_quick_timer() /
btu_oneshot_alarm() mechanism with alarm_set_on_queue() and
alarm_cancel()
- Removed the whole mechanism around the BTU_TTYPE_* timers.
* Fixed a bug when processing the GATT indication confirmation timer
expiration (timer tGATT_TCB.conf_timer: b/26610829).
* Renamed and/or split misc. timeout functions, fields, and timers
* Renamed time-related constants and changed the values from seconds
to milliseconds
* Replaced timer tAVDT_CCB.timer_entry with three mutually exclusive timers:
idle_ccb_timer, ret_ccb_timer, rsp_ccb_timer
The reason we are using three timers, is because in the original code
function avdt_ccb_chk_timer() used the timer type in its logic: it
would stop the timer only if the type is "idle".
* Removed btm_ble_timeout() and replaced it with multiple timeout
callback functions (per timer)
* Fix the actual value of the global constant BT_1SEC_TIMEOUT and
rename it to BT_1SEC_TIMEOUT_MS
* Removed btu_cb and associated timers and events, because they are
never used.
* Removed unused timers, functions, struct and declarations that are
not used / needed.
Bug: 26611369
Bug: 26610829
Change-Id: I812c8c31710a5daefc58b01fcf35c353768f390f
/system/bt/stack/gatt/gatt_int.h
|
| 9f1c9ad85759c7663e3bbc494f8e75b2f107da67 |
|
28-Sep-2015 |
Pavlin Radoslavov <pavlin@google.com> |
GKI cleanup - Moved struct TIMER_LIST_ENT to OSI
* Moved struct TIMER_LIST_ENT to file osi/include/non_repeating_timer.h
and renamed it to timer_entry_t
NOTE: This is a short-term solution. timer_entry_t should be
removed, and its usage everywhere should be replaced by
struct non_repeating_timer_t .
* Renamed TIMER_CBACK to timer_callback_t
* Renamed TIMER_PARAM_TYPE to timer_param_t
Change-Id: I9ca830718bf900195f9c0a513a97f6995322693b
/system/bt/stack/gatt/gatt_int.h
|
| 1a3844f933bd63c8a381371dabfb35c6a0249e3e |
|
25-Sep-2015 |
Pavlin Radoslavov <pavlin@google.com> |
GKI cleanup - Replaced usage of GKI queue with OSI fixed_queue
* Added new functions to OSI:
- fixed_queue_init()
- fixed_queue_length()
- fixed_queue_try_remove_from_queue()
- fixed_queue_try_peek_last()
* Renamed fixed_queue_try_peek() to fixed_queue_try_peek_first()
* Replaced usage of GKI queue functions with OSI fixed_queue functions:
- GKI_init_q() -> fixed_queue_new(SIZE_MAX)
NOTE: unlike GKI_init_q(), fixed_queue_new() allocates memory /
state that needs to be released by calling fixed_queue_free()
- GKI_enqueue() -> fixed_queue_enqueue()
- GKI_dequeue() -> fixed_queue_try_dequeue()
NOTE: fixed_queue_try_dequeue() is non-blocking
- GKI_queue_length() -> fixed_queue_length()
- GKI_queue_is_empty() -> fixed_queue_is_empty()
- GKI_getfirst() -> fixed_queue_try_peek_first()
- GKI_getlast() -> fixed_queue_try_peek_last()
- GKI_remove_from_queue() -> fixed_queue_try_remove_from_queue()
- Queue elements iteration.
In the fixed_queue implementation we have to use the underlying
list_t mechanism to iterate over the elements.
OLD:
p = GKI_getfirst(queue);
...
while ((p = GKI_getnext(p) != NULL) {
...
}
NEW:
list_t *list = fixed_queue_get_list(queue);
for (const list_node_t *node = list_begin(list);
node != list_end(list); node = list_next(node)) {
p = list_node(node);
}
* Remove initialization of the GKI module, because it is not needed
anymore
* Removed unused files in GKI:
gki/common/gki_common.h
gki/ulinux/gki_int.h
gki/ulinux/gki_ulinux.c
Change-Id: I3ff9464db75252d6faf7476a9ca67c88e535c51c
/system/bt/stack/gatt/gatt_int.h
|
| 444a8da807abaf5f9e813ce70c56a79160495fb3 |
|
06-Mar-2015 |
Satya Calloji <satyac@broadcom.com> |
LE Privacy 1.2 and LE secure connections
Bug: 19816438
Original author: Chaojing Sun <cjsun@broadcom.com>
Change-Id: I5951f4d4e038f8348a62aa6d19b2111bae0b3ecc
/system/bt/stack/gatt/gatt_int.h
|
| 24933b5b1a06274c47133debac5251a97128a267 |
|
25-Sep-2014 |
June R. Tate-Gans <jtgans@google.com> |
Removing unnecessary BTAPI #defines.
/system/bt/stack/gatt/gatt_int.h
|
| 49ada1eb9382ce02a0d81b1a7c3065cccf38d09e |
|
21-Feb-2014 |
Chris Manton <cmanton@google.com> |
Release GATT resources upon teardown
/system/bt/stack/gatt/gatt_int.h
|
| adc9f28ad418356cb81640059b59eee4d862e6b4 |
|
06-Dec-2014 |
Chaojing Sun <cjsun@broadcom.com> |
LE: Register for service change indication
Some bonded remote devices require the client to register for service
change indication and actually set the client configuration descriptor
before service change indications can be removed.
This change add an additional step after device bonding has occured to
register for service change indication if possible.
Bug: 18173911
Change-Id: I25386faec0d58834ee2b0a9d1db2d2e052311264
/system/bt/stack/gatt/gatt_int.h
|
| 17b04bd498405f2bb109a85562ebbdcb6bb06e95 |
|
28-Mar-2014 |
Andre Eisenbach <andre@broadcom.com> |
LE: Add notification sent and congestion callbacks (2/4)
This change introduces two new callbacks for applications to better
handle LE notification flow control and transport congestion. The
notification callback is invoked when the remote platform confirms an
indication or when a local notification has been passed to the
controller. No new notifications should be sent until a callback is
received.
Congestion callbacks are triggered when a GATT operation cannot be sent
to the local Bluetooth controller. Repeatedly calling
writeCharacteristic() for example will eventually trigger a congestion
callback. Applications cannot send additional data until a further
callback is received, indicating that the congestion has cleared up.
Also fixes a memory leak issue that can trigger GKI exception for
certain GATT operations.
Change-Id: Ib470c15ca1f577573a632e9afe4cb9e86680fd8c
/system/bt/stack/gatt/gatt_int.h
|
| 7fa4fba6f59f97df00aff07dbe8fb21b114b3c2c |
|
17-Apr-2014 |
Ganesh Ganapathi Batta <ganeshg@broadcom.com> |
Merge BT 4.1 features
The features include:
- LE Peripheral Mode
- Link Layer topology (LE Central & Peripheral Concurrency)
- Dual Mode Topology (Ability to choose LE transport when connecting with
other Dual Mode devices)
- Fast advertising Interval
- Limited Discovery Time Changes
- GAP Authentication and Lost Bond
- Dual Mode Addressing
- Common Profile and Service Error Code
- 32 bit UUIDs
Change-Id: Ic6701da4cf6aaa390ff2c8816b43157f36b7fb42
Conflicts:
stack/btu/btu_hcif.c
/system/bt/stack/gatt/gatt_int.h
|
| 5fba60a8761b28adec2704dcbf69376ebde4f19b |
|
06-Feb-2014 |
Zhihai Xu <zhihaixu@google.com> |
resend discovery primary service ATT request with timeout
This is to work around the problem with iphone.
The iphone didn't respond with our first ATT request
(read By group type Request for for primary service).
the clockwork/phone host keep waiting for ATT response until disconnection
from iphone due to timeout.
The workaround is to
resend discovery primary service before disconnection from remote device
when response timeout happen.
This workaround will be better for us to interop with remote BLE device.
bug:12895830
Change-Id: I236af8eca9790f2dae7098061c74cec55348ca6d
/system/bt/stack/gatt/gatt_int.h
|
| ead3cde4bac0c3e32cd31f149093f004eef8ceeb |
|
06-Feb-2013 |
Ganesh Ganapathi Batta <ganeshg@broadcom.com> |
Initial version of BLE support for Bluedroid
Change-Id: I9825a5cef9be2559c34c2a529b211b7d471147cf
/system/bt/stack/gatt/gatt_int.h
|
| 5738f83aeb59361a0a2eda2460113f6dc9194271 |
|
13-Dec-2012 |
The Android Open Source Project <initial-contribution@android.com> |
Snapshot cdeccf6fdd8c2d494ea2867cb37a025bf8879baf
Change-Id: Ia2de32ccb97a9641462c72363b0a8c4288f4f36d
/system/bt/stack/gatt/gatt_int.h
|