Address several ESP32 network driver issues #1137
Conversation
| #ifdef CONFIG_AVM_ENABLE_NETWORK_PORT_DRIVER | ||
|
|
||
| REGISTER_PORT_DRIVER(network, network_driver_init, NULL, network_driver_create_port) | ||
| REGISTER_PORT_DRIVER(network, network_driver_init, network_driver_stop, network_driver_create_port) |
There was a problem hiding this comment.
network_driver_stop will not be called when calling network:stop/0. It gets called during tests (when GlobalContext is destoryed and recreated).
So this change sounds correct, but it doesn't do what you are saying.
UncleGrumpy
force-pushed
the
esp32_network_issues
branch
4 times, most recently
from
908bd03
to
40b8959
Compare
|
Thank you for the explanation of the These problems have all been fixed, and the driver shuts down and exits cleanly. There is still one ESP-IDF message (in some builds, depending on the IDF version) in the console logs that is being reported as an error:
I looked into this and it is actually a known ESP-IDF issue, that will be changed to an information message in a future update. See espressif/esp-idf#12473. |
UncleGrumpy
changed the title
UncleGrumpy
force-pushed
the
esp32_network_issues
branch
from
40b8959
to
8eb624e
Compare
UncleGrumpy
force-pushed
the
esp32_network_issues
branch
from
8eb624e
to
a222bd4
Compare
Removes several possible uses of free() on memory that had been previously free'd. This would happen under specific error conditions in the `start_network` function in network_driver.c that `network:start/1` uses to configure and start the network. Signed-off-by: Winford <winford@object.stream>
…SP32 Adds a destroy callback to the ESP32 network driver to completely stop the driver and free all network resources when network:stop/0 is used. Previosly the driver was not being stopped internally and resources were not freed when the gen_server was stopped, causing instability, and possible crashes when event callbacks were triggered, but there was no process alive to handle them. Closes atomvm#643 Signed-off-by: Winford <winford@object.stream>
Adds an event handler for `event 21` the `WIFI_EVENT_STA_BEACON_TIMEOUT` event
and an option to add an Erlang callback handler for the event. The event will
be logged with an info level message that includes a suggestion about the two
most likely causes, poor rssi and network congestion. A callback config option
`{beacon_timeout, fun()}` may be added to the `sta` config.
Closes atomvm#1100
Signed-off-by: Winford <winford@object.stream>
UncleGrumpy
force-pushed
the
esp32_network_issues
branch
from
a222bd4
to
dfa2f88
Compare
These changes close the following issues for the ESP32 network driver:
network:stop/0#643] When network:stop/0 is used the driver is now completely stopped and all resources are freed.These changes are made under both the "Apache 2.0" and the "GNU Lesser General
Public License 2.1 or later" license terms (dual license).
SPDX-License-Identifier: Apache-2.0 OR LGPL-2.1-or-later