Fix dependency tree so that lwip doesn't depend on any specific network
interface component.
Network interface drivers shall depend on esp_netif.
esp_netif shall depend on lwip (but not on any specific interface
driver) -- it optionally depends on vfs and esp_eth (need ethernet
header for L2/bridge mode)
* esp-netif to optionally depend on esp-eth (only for l2tap config)
* esp_eth.h now includes the original ethernet header and the
ethernet-netif glue layer
* Updated examples and test to explicitely use esp-eth dependency if
needed
Some components were including esp_timer.h without declaring a
dependency on esp_timer component. This used to work due to a
transitive public dependency on esp_timer from freertos component.
Add explicit dependencies where needed.
Also some source files were using esp_timer functions without
including the header file. This used to work because esp_timer.h was
included from freertos port header file. This commit adds esp_timer.h
includes where needed.
This commit removes the usage of all legacy FreeRTOS data types that
are exposed via configENABLE_BACKWARD_COMPATIBILITY. Legacy types can
still be used by enabling CONFIG_FREERTOS_ENABLE_BACKWARD_COMPATIBILITY.
* add support for KSZ8001, KSZ8021, KSZ8031, KSZ8051 and KSZ8061
* remove duplicate code
* simplify architecture to make the code base extensible (for future work)
The ethernet\iperf example has a bug, once bandwidth is more than 80 Mbits/s the average counter shows wrong total transmitted data value. Issue occur due to overflow of the uint32_t total_len variable.
* created new average formula to avoid the overflow
* updated interval formula according to new variables
* renamed variable according to its purpose
Ethernet driver events properly bounded with ESP NETIF actions to support multiple Ethernet modules used at a time.
Components using Ethernet updated to conform with new API.
Closes https://github.com/espressif/esp-idf/issues/7318
The code checked CONFIG_ETH_USE_SPI_ETHERNET (which is usually set),
but CONFIG_EXAMPLE_ETH_SPI_xxx_GPIO options are only defined if
CONFIG_EXAMPLE_USE_SPI_ETHERNET is set. Fix the ifdef accordingly.
Regression from aea901f0.
Added the RMT/addr LED blink to the example and more detailed README.md
Moved component/led_strip to common_components
Added missing README file to common_components/led_strip
README file update
Makefile and Kconfig fixed for led_strip component
Fixing end-of-line on main/blink.c
Component updated to handle multiple instances
Added note on the RMT channel number (ESP32 and ESP32-S2)
Removed components folder from rmt/led_strip example and README updated
Changed the led_strip_denit function and added ESP32-C3 RMT info on channel configuration
Updates on README, Kconfig default settings and configure_led() function added
1. Add Example for DPP Enrollee
2. Use DPP Supplicant API's to setup connection
3. Add support for multiple channels in Bootstrapping
4. Add Unity testcase for testing Offchannel operations
Closes https://github.com/espressif/esp-idf/issues/5654
- Newlib uses significantly more stack space when printing to an unbuffered stream
- For examples tests, disabling buffering on stdout is not really required
This issue was found during one of the OTA example test failure, root cause
being stack overflow in `esp_event` task.
OpenCores Ethernet MAC has a relatively simple interface, and is
already supported in QEMU. This makes it a good candidate for enabling
network support when running IDF apps in QEMU, compared to the
relatively more complex task of writing a QEMU model of ESP32 EMAC.
This driver is written with QEMU in mind: it does not implement or
handle things that aren't implemented or handled in the QEMU model:
error flags, error interrupts. The transmit part of the driver also
assumes that the TX operation is done immediately when the TX
descriptor is written (which is the case with QEMU), hence waiting for
the TX operation to complete is not necessary.
For simplicity, the driver assumes that the peripheral register
occupy the same memory range as the ESP32 EMAC registers, and the
same interrupt source number is used.
This fixes the issue that if Wi-Fi is stopped from a shutdown handler,
the code in connect.c tries to reconnect, and fails because Wi-Fi is
already stopped.
Also make the error check in connect.c less strict.
