c9731c5f17
This is a huge patch that should be a global NOP. For unix type platforms it enables -Wconversion to issue warnings (-> error) for all automatic casts that seem less than ideal but are normally concealed by the toolchain. This is things like passing an int to a size_t argument. Once enabled, I went through all args on my default build (which build most things) and tried to make the removed default cast explicit. With that approach it neither change nor bloat the code, since it compiles to whatever it was doing before, just with the casts made explicit... in a few cases I changed some length args from int to size_t but largely left the causes alone. From now on, new code that is relying on less than ideal casting will complain and nudge me to improve it by warnings. |
||
|---|---|---|
| .. | ||
| async-dns | ||
| dhcpclient | ||
| ntpclient | ||
| smd | ||
| CMakeLists.txt | ||
| README.md | ||
| system.c | ||
LWS System Helpers
Lws now has a little collection of helper utilities for common network-based functions necessary for normal device operation, eg, async DNS, ntpclient (necessary for tls validation), and DHCP client.
Conventions
If any system helper is enabled for build, lws creates an additional vhost
"system" at Context Creation time. Wsi that are created for the system
features are bound to this. In the context object, this is available as
.vhost_system.
Attaching to an existing context from other threads
To simplify the case different pieces of code want to attach to a single lws_context at runtime, from different thread contexts, lws_system has an api via an lws_system operation function pointer where the other threads can use platform-specific locking to request callbacks to their own code from the lws event loop thread context safely.
For convenience, the callback can be delayed until the system has entered or passed a specified system state, eg, LWS_SYSTATE_OPERATIONAL so the code will only get called back after the network, ntpclient and auth have been done. Additionally an opaque pointer can be passed to the callback when it is called from the lws event loop context.
Implementing the system-specific locking
lws_system_ops_t struct has a member .attach
int (*attach)(struct lws_context *context, int tsi, lws_attach_cb_t *cb,
lws_system_states_t state, void *opaque,
struct lws_attach_item **get);
This should be defined in user code as setting locking, then passing the arguments through to a non-threadsafe helper
int
__lws_system_attach(struct lws_context *context, int tsi, lws_attach_cb_t *cb,
lws_system_states_t state, void *opaque,
struct lws_attach_item **get);
that does the actual attach work. When it returns, the locking should be unlocked and the return passed back.
Attaching the callback request
User code should call the lws_system_ops_t .attach function like
lws_system_get_ops(context)->attach(...);
The callback function which will be called from the lws event loop context should look like this
void my_callback(struct lws_context *context, int tsi, void *opaque);
with the callback function name passed into the (*attach)() call above. When the callback happens, the opaque user pointer set at the (*attach)() call is passed back to it as an argument.