platform_hardware_interfaces/wifi/1.3/default/THREADING.README
Jong Wook Kim da830c9da7 Uprev Wifi HAL to 1.3
Move Wifi HAL from 1.2 to 1.3

Bug: 111753174
Test: Builds and wifi works
Change-Id: I6a2b203997d6328d09065f23636d4ce01420942a
2018-08-09 15:40:09 -07:00

35 lines
1.9 KiB
Text

Vendor HAL Threading Model
==========================
The vendor HAL service has two threads:
1. HIDL thread: This is the main thread which processes all the incoming HIDL
RPC's.
2. Legacy HAL event loop thread: This is the thread forked off for processing
the legacy HAL event loop (wifi_event_loop()). This thread is used to process
any asynchronous netlink events posted by the driver. Any asynchronous
callbacks passed to the legacy HAL API's are invoked on this thread.
Synchronization Concerns
========================
wifi_legacy_hal.cpp has a bunch of global "C" style functions to handle the
legacy callbacks. Each of these "C" style function invokes a corresponding
"std::function" version of the callback which does the actual processing.
The variables holding these "std::function" callbacks are reset from the HIDL
thread when they are no longer used. For example: stopGscan() will reset the
corresponding "on_gscan_*" callback variables which were set when startGscan()
was invoked. This is not thread safe since these callback variables are
accesed from the legacy hal event loop thread as well.
Synchronization Solution
========================
Adding a global lock seems to be the most trivial solution to the problem.
a) All of the asynchronous "C" style callbacks will acquire the global lock
before invoking the corresponding "std::function" callback variables.
b) All of the HIDL methods will also acquire the global lock before processing
(in hidl_return_util::validateAndCall()).
Note: It's important that we only acquire the global lock for asynchronous
callbacks, because there is no guarantee (or documentation to clarify) that the
synchronous callbacks are invoked on the same invocation thread. If that is not
the case in some implementation, we will end up deadlocking the system since the
HIDL thread would have acquired the global lock which is needed by the
synchronous callback executed on the legacy hal event loop thread.