* Changed data type of temperature
* Changed data type in EddystoneTLM class and example
* Revert "Changed data type in EddystoneTLM class and example"
This reverts commit 1f3a9414ee13341f80bd6d51cdbd57254aadfe97.
* Draft of Eddystone TLM example
* Adds MACROs to convert beacon temperature
2 Macros
EDDYSTONE_TEMP_U16_TO_FLOAT(tempU16) - takes the TLM BigEndian 8.8 fixed point representation and returns its float value
EDDYSTONE_TEMP_FLOAT_TO_U16(tempFloat) - takes a float (temperature) and returns its BigEndian 8.8 fixed point representation
* Fixed temp
* Changed to conform with PR comments
* Fixed comment on closing bracket
* Prints negative TEMP big endian as just 2 bytes
* Extacts correct Eddyston Service Data
* Fixes BLEEddystoneTLM::toString() negative temp
* Fixes URL field length
* Fixes Eddystone URL decoding
* Fixes MSB for iBeacon UUID
iBeacons use big endian BLE fields.
* Fix to detect iBeacon that also has Service UUID
This fix makes the BLE_iBeacon.ino to work correctly with the BLE_Beacon_Scanner.ino example
---------
Co-authored-by: Rodrigo Garcia <rodrigo.garcia@espressif.com>
* Remove inline code comments for the struct description
* Renamed ESPNow examples
* Changed example from C++ class to C struct
* Updated MultiHomedServers
* Updated BT Serial examples
* Added short overview of exaples - missing Enterprise
* Added draft of Enterprise header note
* Fixed note in README
* Fixed compilation error
* Changed FunctionalInterrupt example
* Fixes
* Returned FunctionalInterrupts to original state + added modified version as new example
* Added methods + example to retrive local MAC for BT
* Added .skip files in the new example folder
* Fixed typos and formatting + added doxygen comments
* changed std::string to String
* another std::string -> String
* Changed std::string to String
* chaged string type in example
* Moved and renamed example ESP32/FreeRTOS to MultiThreading/BasicMultiThreading
* Added dummy files
* Modified original example
* Fixed BasicMultiThreading.ino
* Added Example demonstrating use of queues
* Extended info in BasicMultiThreading
* Renamed Queues to singular Queue
* Added Mutex example
* Added Semaphore example
* Moved info from example to README
* Moved doc from Mutex to README
* Added Queue README
* Removed unecesary text
* Fixed grammar
* Increased stack size for Sempahore example
* Added headers into .ino files
* Added word Example at the end of title in README
* removed unused line
* Added forgotten README
* Modified BasicMultiThreading example
* Added missing S3 entry in README
* moved location
[ESP32: SDMMCFS::begin hardcodes the usage of slot 1, only check if the pins match slot 1 pins.]
setPins() was testing pins D1, D2 and D3 all against D1 ... fine in 1 pin mode when all are -1 not so much if you're trying to get 4 pin mode working.
I now see this function doesn't really do anything on the ESP32...accept now correctly checks that you are trying to use the slot 1 pins.
Co-authored-by: Jan Procházka <90197375+P-R-O-C-H-Y@users.noreply.github.com>
* Update WiFiClient.cpp
This change will allow specifying connect timeouts below 1 second. Without this change, if connect timeouts under 1 second are given, the connect defaults to 0ms and fails.
This will also allow timeouts in fractions of seconds, e.g. 1500ms. Without this change, connect timeouts are truncated to full second increments.
* Make parameter timeout_ms clear
* Change connection timeout_ms name for clarity
---------
Co-authored-by: Rodrigo Garcia <rodrigo.garcia@espressif.com>
* Fixes BLE data printing
BLE data has no '\0' terminator, therefore it can't be printed as a regular C string.
This fix just prints the BLE data based on its length.
* Simplify printing to a single call
* Modified WiFiClient example to use thingspeak instead of non-functionig sparkfun
* Moved instructions to README
* Fixed spelling
* Added link to S3 datasheet
---------
Co-authored-by: Jan Procházka <90197375+P-R-O-C-H-Y@users.noreply.github.com>
* Fixed issue where esp32 won't reconnect to WiFi AP if the AP was restarted
* Added the autoReconnect condition and removed the reconnect reason
* Chenges on the WiFi reconnection and added new example
* Added new line at the end
* Changes reverted with some minor changes. New example remained
* Changes reverted and sync
Co-authored-by: pedro.minatel <pedro.minatel@espressif.com>
* ESP Insights: Added library support
* ESP Insights: Added Examples
* ESP Insights: Added custom partitions file
* ESP Insights: Added API documentation.
* Added recipe and script to create Insights package
* Updated ESP Insights examples.
* Changed Insights Firmware package output directory
* Changed license to include SPDX license
* Fix Insights package for Windows
* Updated .exe of insights script
* Added coredump partition to all schemes
* Updated header files of Insights diagnostics
* hotfix: Added elf-sha256-offset flag in elf2bin
builder
* Update API to be more Arduino-like and partitions offsets
Co-authored-by: Me No Dev <me-no-dev@users.noreply.github.com>
* adds extended HardwareSerial examples
* Adds new example with Serial RxTimeout
* adds and improves Serial onReceive expamples
* adjust includes CMake - UART example
* adjust includes CMake - UART example
* fixes CMake and CI
* adds ESP/Serial to CMakeList
* adds ESP/Serial to CMakeList
* fixes demo include
* fixes BREAK demo
* fixes onReceive demo
* Changes FIFO Full criteria
Changed the "1-by-1" Serial only when baud rate is 57600 or lower.
* example code replacement
* replaces functions in hal
* Update `WiFiScan` example to show more useful data
* Add channel to output.
* Add type of encryption to output.
* Format output as table.
* Update outdated example description.
* Add `scanDelete()` as hint to free memory.
* added 32 chars SSID + readme edit
Co-authored-by: Jan Procházka <90197375+P-R-O-C-H-Y@users.noreply.github.com>
* replaced original iBeacon example with a known-good one
* addressed VojtechBartoska's comment
* incorporated P-R-O-C-H-Y's change
Co-authored-by: Jan Procházka <90197375+P-R-O-C-H-Y@users.noreply.github.com>
* refactor(toString method): Add RSSI / serviceData
These are key characteristics of the BLE advertising packet. Really useful, adding to this method allows data to be easily obtained and passed over TCP/IP for processing.
* fix: remove line 565 bug
Currently, if a second connection occur in Bluetooth, this second connection is automatically disconnected (line 284).
This disconnection trigger the ESP_SPP_CLOSE_EVT, which clear the bit SPP_CONNECTED.
But the first connection remain active, and this flag shouldn't be cleared.
This fix the issue, by clearing the flag only if the last connection is closed
Co-authored-by: Rodrigo Garcia <rodrigo.garcia@espressif.com>
This commit addes an example showing how to receive messages using the TWAI
driver interface and a CAN transceiver. Specifically, the example demonstrates:
- How to configure and install the TWAI drvier
- How to poll for TWAI events (i.e., alerts) using twai_read_alerts()
- How to handle the various events (such as TWAI_ALERT_RX_DATA)
Closes https://github.com/espressif/arduino-esp32/pull/7430
Co-authored-by: Stephan Martin <designer2k2@gmail.com>
* Added enableScenes API
* Added enableScenes API documentation
* Added enableScenes API to example
Co-authored-by: Jan Procházka <90197375+P-R-O-C-H-Y@users.noreply.github.com>
Fixed#7406 . The "reason2str" macro in WiFiGeneric.cpp tries to read memory from index "-1" in "system_event_reasons" array when handling STA_DISCONNECTED event with reason 0. Dealing with reason 0 as a reason 1 (WIFI_REASON_UNSPECIFIED) will solve the problem (the reason for this event to arrive with reason 0 is unknown). #7406
The original code assumes 100Hz FreeRTOS tick rate and just supplies vTaskDelay with the assumed number of ticks required for the wanted delay. This patch simply fixes it to use portTICK_PERIOD_MS, thereby working correctly regardless of what tick rate FreeRTOS has been configured to run at.
The HID semaphore allows USBHID::SendReport() to wait for the completion of
report sending.
With a zero timeout xSemaphoreTake() after calling tud_hid_n_report(),
occasionally, the following would happening:
1. USBHID::SendReport() would send a report by calling tud_hid_n_report().
2. The send would complete and (presumably on another thread)
tud_hid_report_complete_cb() would be called and it would xSemaphoreGive()
the semaphore.
3. In USBHID::SendReport(), the zero timeout xSemaphoreTake(sem, 0) would
succeed, taking the semaphore.
4. On the next line, xSemaphoreTake(sem, timeout_ms ...) would timeout
because the semaphore was already taken by the previous line of code.
The result would be waiting timeout_ms for no reason.
The purpose of the zero timeout xSemaphoreTake() is to clear the semaphore in
case a previous SendReport() timed out waiting for the semaphore. In that case,
tud_hid_report_complete_cb() may be called after the timeout, giving the
semaphore. Then the next SendReport() would start with the semaphore given,
which isn't desired if we want to call xSemaphoreTake(sem, timeout_ms ...) on
it.
There have also been other cases where tud_hid_report_complete_cb() is called
an extra time, causing the same situation.
The fix is to move the zero timeout xSemaphoreTake() before the call to
tud_hid_n_report(). This eliminates the race between the zero timeout
xSemaphoreTake() and tud_hid_report_complete_cb() in the common case when no
timeout occurs.
There is still a possible race condition between the zero timeout
xSemaphoreTake() and tud_hid_report_complete_cb() in the case of a timeout,
but that should be rarer.
Arduino-esp32 2.0.4 was released with a version of TinyUSB hid_device.h
that uses uint16_t for the last argument:
https://github.com/espressif/arduino-esp32/blob/2.0.4/tools/sdk/esp32s2/include/arduino_tinyusb/tinyusb/src/class/hid/hid_device.h
TU_ATTR_WEAK void tud_hid_report_complete_cb(uint8_t instance, uint8_t const* report, uint16_t len);
But USBHID implements this callback with uint8_t:
https://github.com/espressif/arduino-esp32/blob/2.0.4/libraries/USB/src/USBHID.cpp
void tud_hid_report_complete_cb(uint8_t instance, uint8_t const* report, uint8_t len){
if (tinyusb_hid_device_input_sem) {
xSemaphoreGive(tinyusb_hid_device_input_sem);
}
}
The result is that when USBHIDKeyboard sends a report to the host, it
times out, waiting 100 ms for the callback to be called. It does this
once for pressing the key and once for releasing the key, so
100 ms * 2 = 200 ms.
The latest version of hid_device.h reverts the last argument to uint8_t:
860b104691/tools/sdk/esp32s2/include/arduino_tinyusb/tinyusb/src/class/hid/hid_device.h
TU_ATTR_WEAK void tud_hid_report_complete_cb(uint8_t instance, uint8_t const* report, /*uint16_t*/ uint8_t len );
But these commits suggest that the last argument will eventually be
changed to uint16_t:
556b5d5044
change report len in hid API from uint8_t to uint16_t
since HS interrupt endpoint can be up to 1024, 8-bit is not enough.
affected APIs are:
- tud_hid_n_report() / tud_hid_report()
- tud_hid_report_complete_cb()
b495d6f8ec
temporarily revert len back to uint8_t in tud_hid_report_complete_cb() for up coming release
To prevent this from becoming broken again, in preparation for the change
to uint16_t, make USBHID resilient to any type for the last argument for
tud_hid_report_complete_cb() by using some C++ template metaprogramming,
adapted from https://stackoverflow.com/a/22632571.
Co-authored-by: Rodrigo Garcia <rodrigo.garcia@espressif.com>
* Removed option to change CHANNEL
* Revert "Removed option to change CHANNEL"
This reverts commit b2ec27de1d2d4703a15f6397550a46896413828a.
* Fixed the ability to change CHANNEL
* WiFi scan only on selected channel
otherwise the compiler will complain about unused variable event
Co-authored-by: Rodrigo Garcia <rodrigo.garcia@espressif.com>
Co-authored-by: Jan Procházka <90197375+P-R-O-C-H-Y@users.noreply.github.com>
* Improve RGB LED Driver
Replaces the use of the `LED_BUILTIN` variable by creating a new variable called `RGB_BUILTIN`. On boards with both a regular LED and RGB LED, this change provides functionality to control either LED.
The `LED_BRIGHTNESS` variable is changed to `RGB_BRIGHTNESS`, which aligns more closely with the `RGB_BUILTIN` variable name.
`BOARD_HAS_NEOPIXEL` is no longer necessary; it is replaced by `RGB_BUILTIN`.
* Update BlinkRGB example
Update example code for changes with the RGB driver:
- Replace `LED_BUILTIN` and `BOARD_HAS_NEOPIXEL` with `RGB_BUILTIN`
- Replace `LED_BRIGHTNESS` with `RGB_BRIGHTNESS`
* Update board variants
Update board variants for changes with the RGB driver:
- Remove `BOARD_HAS_NEOPIXEL`
- Define `RGB_BUILTIN` pin
- Replace `LED_BRIGHTNESS` with `RGB_BRIGHTNESS` to align with `RGB_BUILTIN` name
Co-authored-by: Rodrigo Garcia <rodrigo.garcia@espressif.com>
Co-authored-by: Vojtěch Bartoška <76958047+VojtechBartoska@users.noreply.github.com>
With recent compiler update, the following warning is generated:
packages/esp32/hardware/esp32/2.0.4/libraries/LittleFS/src/LittleFS.cpp:85:5: warning: missing initializer for member 'esp_vfs_littlefs_conf_t::dont_mount' [-Wmissing-field-initializers]
This commit adds the missing identifier.
Co-authored-by: Jan Procházka <90197375+P-R-O-C-H-Y@users.noreply.github.com>
* Added definitions for various parameters and UI types.
* Overload for `const char*` added.
* Function `addValidStrList` added.
This is needed for mode parameters (see example RMakerCustomAirCooler.ino).
* Example of a custom device that uses toggle, mode and range parameters.
* Revert: Added definitions for various parameters and UI types.
* Fixed declaration for addValidStrList
* Fixed missing gpio definition for ESP32C3 target
Co-authored-by: Me No Dev <me-no-dev@users.noreply.github.com>
Description of Change
Original code does not work (crashing) - related issue #6832 - now fixed.
Extended description to be more helpful.
Added options to modify the setup at the top of the code via constants.
Added option do plot with Arduino Serial plotter.
Even if the crashing was solved alone the ledc PWM would not output any signal.
Tests scenarios
Tested on ESP32 with oscilloscope and signal generator.
Related links
Closing #6832
Fixed issue with ssl_client->socket being closed without being consistently set to -1. Changed close to lwip_close to be consistent with lwip_socket, resolved issue with long term assert after 4+ days (assert failed: lock_release_generic locks.c:186).
Co-authored-by: Me No Dev <me-no-dev@users.noreply.github.com>
Description of Change
This PR fixes an issue related to BLE Server and Descriptors.
The issue:
If the BLE Server code changes its own descriptors, it is not reflected in the GATTS database.
BLE2902 CCCD also didn't reflect any changes to the GATTS database.
Because of this issue, the client could never read the real Descriptor values from the remote Server.
Tests scenarios
Tested with ESP32.
Related links
Fixes#6863Fixes#6868
* Initial implementation of RGB driver via digitalWrite
* Moved constants to pins_arduino.h
* Changed pin definition + added example
* Wrapped BlinkRGB in #ifdef BOARD_HAS_NEOPIXEL
* Removed forgotten log from example
* Moved RGBLedWrite to new file esp32-hal-rgb-led and created pinMode in variatn.cpp
* Updated example - lowered single channel brightness to LED_BRIGHTNESS
* Changed function name from RGBLedWrite to neopixelWrite + code polishing
* Moved pinSetup portion related to RGB back to common file
- Created new temporary variable, stopping confusion with the reuse of "error". As Wire.requestFrom() doesn't return an error.
- Added a cast to help clarify when and why bytes are being read
When building the core with `-Wextra` a few locations have `case:`
fall throughs or skipped field initializers.
Add proper comments for B64 cases to avoid GCC warnings
Initialized unused fields in Tone and WiFiGeneric to avoid missing
field initializer warnings.
````
2022-05-19T17:40:42.2280300Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/libb64/cdecode.c: In function 'base64_decode_block_signed':
2022-05-19T17:40:42.2282122Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/libb64/cdecode.c:42:23: warning: this statement may fall through [-Wimplicit-fallthrough=]
2022-05-19T17:40:42.2283247Z *plainchar = (fragment & 0x03f) << 2;
2022-05-19T17:40:42.2284240Z ~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~
2022-05-19T17:40:42.2285087Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/libb64/cdecode.c:43:7: note: here
2022-05-19T17:40:42.2285435Z case step_b:
2022-05-19T17:40:42.2285691Z ^~~~
2022-05-19T17:40:42.2286515Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/libb64/cdecode.c:53:23: warning: this statement may fall through [-Wimplicit-fallthrough=]
2022-05-19T17:40:42.2286932Z *plainchar = (fragment & 0x00f) << 4;
2022-05-19T17:40:42.2287219Z ~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~
2022-05-19T17:40:42.2287609Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/libb64/cdecode.c:54:7: note: here
2022-05-19T17:40:42.2287909Z case step_c:
2022-05-19T17:40:42.2288200Z ^~~~
2022-05-19T17:40:42.2288972Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/libb64/cdecode.c:64:23: warning: this statement may fall through [-Wimplicit-fallthrough=]
2022-05-19T17:40:42.2289491Z *plainchar = (fragment & 0x003) << 6;
2022-05-19T17:40:42.2289745Z ~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~
2022-05-19T17:40:42.2290162Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/libb64/cdecode.c:65:7: note: here
2022-05-19T17:40:42.2290509Z case step_d:
2022-05-19T17:40:42.2290714Z ^~~~
2022-05-19T17:40:42.2482744Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/libb64/cencode.c: In function 'base64_encode_block':
2022-05-19T17:40:42.2484713Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/libb64/cencode.c:46:20: warning: this statement may fall through [-Wimplicit-fallthrough=]
2022-05-19T17:40:42.2485415Z result = (fragment & 0x003) << 4;
2022-05-19T17:40:42.2486713Z ~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~
2022-05-19T17:40:42.2487696Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/libb64/cencode.c:47:9: note: here
2022-05-19T17:40:42.2488519Z case step_B:
2022-05-19T17:40:42.2489175Z ^~~~
2022-05-19T17:40:42.2492458Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/libb64/cencode.c:56:20: warning: this statement may fall through [-Wimplicit-fallthrough=]
2022-05-19T17:40:42.2493351Z result = (fragment & 0x00f) << 2;
2022-05-19T17:40:42.2494227Z ~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~
2022-05-19T17:40:42.2496324Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/libb64/cencode.c:57:9: note: here
2022-05-19T17:40:42.2496937Z case step_C:
2022-05-19T17:40:42.2497261Z ^~~~
2022-05-19T17:40:44.6354962Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/Tone.cpp: In function 'void setToneChannel(uint8_t)':
2022-05-19T17:40:44.6356417Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/Tone.cpp:99:5: warning: missing initializer for member 'tone_msg_t::pin' [-Wmissing-field-initializers]
2022-05-19T17:40:44.6357120Z };
2022-05-19T17:40:44.6358732Z ^
2022-05-19T17:40:44.6364470Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/Tone.cpp:99:5: warning: missing initializer for member 'tone_msg_t::frequency' [-Wmissing-field-initializers]
2022-05-19T17:40:44.6367914Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/Tone.cpp:99:5: warning: missing initializer for member 'tone_msg_t::duration' [-Wmissing-field-initializers]
2022-05-19T17:40:44.6372875Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/Tone.cpp: In function 'void noTone(uint8_t)':
2022-05-19T17:40:44.6373943Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/Tone.cpp:110:5: warning: missing initializer for member 'tone_msg_t::frequency' [-Wmissing-field-initializers]
2022-05-19T17:40:44.6375154Z };
2022-05-19T17:40:44.6375825Z ^
2022-05-19T17:40:44.6379852Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/Tone.cpp:110:5: warning: missing initializer for member 'tone_msg_t::duration' [-Wmissing-field-initializers]
2022-05-19T17:40:44.6383291Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/Tone.cpp:110:5: warning: missing initializer for member 'tone_msg_t::channel' [-Wmissing-field-initializers]
2022-05-19T17:40:44.6388688Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/Tone.cpp: In function 'void tone(uint8_t, unsigned int, long unsigned int)':
2022-05-19T17:40:44.6389829Z /home/runner/arduino_ide/hardware/espressif/esp32/cores/esp32/Tone.cpp:128:5: warning: missing initializer for member 'tone_msg_t::channel' [-Wmissing-field-initializers]
2022-05-19T17:40:44.6390677Z };
2022-05-19T17:40:44.6391420Z ^
2022-05-19T17:42:00.6768353Z /home/runner/arduino_ide/hardware/espressif/esp32/libraries/WiFi/src/WiFiGeneric.cpp: In static member function 'static bool WiFiGenericClass::setDualAntennaConfig(uint8_t, uint8_t, wifi_rx_ant_t, wifi_tx_ant_t)':
2022-05-19T17:42:00.6769293Z /home/runner/arduino_ide/hardware/espressif/esp32/libraries/WiFi/src/WiFiGeneric.cpp:1333:5: warning: missing initializer for member 'wifi_ant_config_t::rx_ant_default' [-Wmissing-field-initializers]
2022-05-19T17:42:00.6769658Z };
2022-05-19T17:42:00.6769824Z ^
````
add a comment to make it clear, since platformIO still creating the `spiffs.bin` for filesystem image. The file name `spiffs.bin` is so much CONFUSING! it makes the user fill the LittleFS creating is fail.
Better allocate the buffer for f_mkfs from the heap otherwise the stack requirement of this library is huge due to a work buffer allocated for f_mkfs in sdcard_mount. The work buffer is only needed if argument format_if_empty is set true (which is by default false).
This change is quite important if you plan to use this library in a task. as now it increased the tasks stacks size by 4k, even this memory is never used if users are not aware of the large stack requirement during init this library may have other variables on stack that would have written memory outsides its range which can cause various side effects.
Co-authored-by: Jan Procházka <90197375+P-R-O-C-H-Y@users.noreply.github.com>
The internal _isRemoteAddressSet variable is set to true
when calling connect() functions. If connecting fails _isRemoteAddressSet
needs to be re-set to false, otherwise other functions, such as
discover() will fail without clear error messages.
Summary
Added compiler.warning_flags to all chips in platform.txt to reflect users setting of warning level output during compilation (set up in Arduino IDE preferences)
Impact
When a warning is set to none the compilation will no longer display warnings
Related links
Solves issue #6118
Description of Change
Fixes IPAddress INADDR_NONE declaration when using Arduino WiFi or ETH.
This symbol was defined as 0xffffffff by lwip /inet.h, making it impossible to use INADDR_NONE correctly.
This PR only works when <wifi-provisioning/wifi_config.h> has a modification to include <lwip/ip4_addr.h> instead of <lwip/inet.h>. This will be done directly to the sdk folder in the github structure and it has been fixed in IDF by a separated Merge Request. This will be reflected in the future, for good.
Tests scenarios
This PR was tested with all Arduino WiFi examples, including AsyncUDP. Also with ETH examples.
It was also tested for #6610 test cases.
Testing done for ESP32, ESP32-S2, ESP32-C3 and ESP32-S3.
Related links
fixes#6610fixes#6247fixes#4732
Added options to WiFi STA for connect:
- setMinSecurity: default WIFI_AUTH_WPA2_PSK
- setScanMethod: default WIFI_FAST_SCAN
- setSortMethod: default WIFI_CONNECT_AP_BY_SIGNAL (required all channels scan method)
Added parameters for SSID and BSSID to WiFi.scanNetworks()
Fixes: #6485
Fixes: Any issue about WiFi connecting slower now than in 1.0.x
Private Key. Issue presented during any subsequent invocation of loadCACert, loadCertificate, and
loadPrivateKey, respectively, after the first invocation.
* Revert "Edited VFSFileImpl::read to use both read/fread (#6456)"
This reverts commit 7b89b39e10.
* Added default file buffer size + function to change it by user
Also support ESP_IDF_VERSION_MAJOR > 3
Fixes#6458
Co-authored-by: Rodrigo Garcia <rodrigo.garcia@espressif.com>
Co-authored-by: Me No Dev <me-no-dev@users.noreply.github.com>
* Fixes USB CDC setRxBufferSize(), begin(), _onRX()
* Fixes SetRxBufferSize(0) with end()
* Fixes reset when 2x call to end()
* Adds RX_OVERFLOW_EVENT and Queue Copy in setBufferSize
* changed event name to ARDUINO_USB_CDC_RX_OVERFLOW_EVENT
Summary
The examples demonstrate how to create a WPA2 Enterprise connection, but it requires using various direct esp_idf functions. This patch is intended to create another overloaded version of the WiFi.begin() function that allows a user to create a WPA2 Enterprise connection in much the same way as different kinds of connections.
My only question for the core maintainers is whether I should leave those #ifdef's in there. I added them so that it was easy to disable all the code I added via defines from my platformio.ini file, but they technically aren't necessary.
Impact
This should make it easier for novice users to create WPA2 Enterprise connections. For my university, I didn't need a root certificate or the client certificate or client key, so I haven't been able to debug those scenarios, but I built the begin functions to allow any one of those to be used, if needed.
I can confirm that eduroam-style WPA2 Enterprise networks that only require authentication with a username and password works as expected.
In my project I'm getting the error
```
In file included from lib/Discovery/Discovery.cpp:2:
C:/Users/David/.platformio/packages/framework-arduinoespressif32/libraries/AsyncUDP/src/AsyncUDP.h:47:1: error: expected class-name before '{' token
```
Adding a reference to Stream.h fixes it.
* Added dual antenna for WiFi (based on the ESP32-WROOM-DA module)
* Fixed build error
* Fixed indentation and renamed function to setDualAntennaConfig
* Added the RX and TX selection modes as configuration
* Mode code optimization
Summary
Touch Sensor refactoring to be based on IDF 4.4.
Adds support to ESP32S2 and future ESP32S3.
Adds some new APIs:
For all chips:
void touchAttachInterruptArg(uint8_t pin, void (*userFunc)(void*), void *arg, uint32_t threshold);
This function allows the user to add and pass a void* parameter to the ISR user callback.
void touchDetachInterrupt(uint8_t pin);
This function detaches ISR call back for the touch pad pin.
Only ESP32 chip
void touchInterruptSetThresholdDirection(bool mustbeLower);
This function allows the user to set if the ISR callback will be activated when the touch sensor readings are lower or higher than the threshold defined. See example TouchButton.ino.
Only ESP32-S2 and ESP32-S3 chips
bool touchInterruptGetLastStatus(uint8_t pin);
This function reports if the touch pad pin is touched or untouched. It can be used with ISR to identify when it is touched and untouched (released). See example TouchButtonV2.ino.
Impact
None. ll original APIs and examples now run on ESP32 and ESP32-S2.
Related links
Fix#6095Fix#6034Fix#5799Fix#5745Fix#5527Fix#5493Fix#4321Fix#4044Fix#2722Fix#2625
CaptivePortal.ino example did randomly crash for me ... so I start investigate ;-)
Decoding stack results
0x4016faea: WiFiUDP::write(unsigned char const*, unsigned int) at C:\Users\knoeb\AppData\Local\Arduino15\packages\esp32\hardware\esp32\1.0.5\libraries\WiFi\src\WiFiUdp.cpp line 201
0x400d4a4a: DNSServer::replyWithIP() at C:\Users\knoeb\AppData\Local\Temp\arduino_build_486825\sketch\src\DNSServer\DNSServer.cpp line 187
0x400d4d01: DNSServer::processNextRequest() at C:\Users\knoeb\AppData\Local\Temp\arduino_build_486825\sketch\src\DNSServer\DNSServer.cpp line 117
0x400d3e81: loop() at D:\Drive\Dokumente\HTL_Lehrer\2021_22\Projekte\Stromzaehler_Patrick\arduino/arduino.ino line 1078
0x400dd545: loopTask(void*) at C:\Users\knoeb\AppData\Local\Arduino15\packages\esp32\hardware\esp32\1.0.5\cores\esp32\main.cpp line 37
0x4008a0de: vPortTaskWrapper at /home/runner/work/esp32-arduino-lib-builder/esp32-arduino-lib-builder/esp-idf/components/freertos/port.c line 143
I found with wireshark a possibility to trigger the crash on demand is:
nslookup 3.1o1osr0092ons87rp375p1pq8q066o8p56or1sqsps6rs17r4384q9748qr1r52.699p1r741q737393648s29917o45p16q50rn517rnsp73pp68p1q259s92693qp.s607408539s0p06p7559os0899866344r7qq7rpns960o9576q65.r5n94r5so9784pq1.i.03.s.sophosxl.net
The problem was that QNameLength is a signed byte and therefore its not possible to count up to 255. Additionally we need 256 bytes for the QName string to accommodate for the zero termination.
With LittleFS the `fs.exists(path)` returns true also on folders. A `isDirectory()` call is required to set _isFile to false on directories.
This enables serving all files from a folder like : `server->serveStatic("/", LittleFS, "/", cacheHeader.c_str());
File f = fs.open(path);
_isFile = (f && (! f.isDirectory()));
Reliability fix for autoReconnect when assoc_fail, autoReconnect did not work before for these failures.
Changes behavior of WIFI_REASON_ASSOC_FAIL event when autoReconnect is set, removes WIFI_REASON_ASSOC_FAIL/WL_CONNECT_FAILED so retry waitforconnectresult loop stays active for the retry, was not working before.
* change parameter to signed int
As of wrong paramater, the following problem existed, that will be fixed now with this change.
BTScanResultsSet.cpp:67:8: warning: comparison of unsigned expression < 0 is always false [-Wtype-limits]
if (i < 0)
* Change parameter and variable to int
As of wrong paramater, the following problem existed, that will be fixed now with this change.
BTScanResultsSet.cpp:67:8: warning: comparison of unsigned expression < 0 is always false [-Wtype-limits]
if (i < 0)
* Revert "Examples update, add a note for configTime() that only one ntp server is supported by lwip",
fixed in espressif/esp32-arduino-lib-builder#51
This reverts commit 6b1020967a171c549b3d956825fd0d395de9cce0.
* SimpleTime: add NTPoDHCP option and TimeZone env variable
Summary
This PR is refactoring of SigmaDelta HAL in order to use IDF instead of current Register manipulation approach.
Impact
Change in API:
uint32_t sigmaDeltaSetup(uint8_t channel, uint32_t freq);
changed to -->
uint32_t sigmaDeltaSetup(uint8_t pin, uint8_t channel, uint32_t freq);
void sigmaDeltaAttachPin(uint8_t pin); removed, no longer needed. Pin is attached in sigmaDeltaSetup()
Summary
RMT HAL refactoring based on IDF.
Impact
Improves RMT by adding IDF v4.4 support.
Receiving RMT can handle any size of data.
rmtInit() has a new parameter - RxBufferSize - to hold any number of data when receiving RMT.
rmtWrite() has a new parameter - wait_tx_done - to block writing until sending all data.
Related links
fix#5905
23 Braille Display Page (0x41)
Braille display allow visually impaired computer users to read out text using raised pins. The pins are electro-mechanically
activated. These devices also have support for controls that help navigate the computer screen. Typically, braille displays
interface with software known as a screen reader in order to perform this navigation.
## Summary
Arduino ```setup()``` and ```loop()``` run under a Task with a fixed Stack size of 8KB.
Users may want to change this size.
This PR adds this possibility by just adding a line of code, as for example:
``` dart
ESP_LOOP_TASK_STACK_SIZE(16384);
void setup() {
}
void loop() {
}
```
## Impact
None. It adds a new functionality to ESP32 Arduino.
If ```ESP_LOOP_TASK_STACK_SIZE(newSize);``` is not declared/used, it will compile the sketch with the default stack size of 8KB.
## Related links
fix#6010https://github.com/espressif/arduino-esp32/issues/6010#issuecomment-992701658
Thanks @igrr for the suggestion!
* Refactoring function headers
Changing the header so the main TwoWire::requestFrom() definition uses a `size_t` instead of a `uint8_t`, removing the 255 bytes limit on I2C requests.
Co-authored-by: Flaviu Tamas <me@flaviutamas.com>
Co-authored-by: Me No Dev <me-no-dev@users.noreply.github.com>
* using ksz8081 only from ESP-IDF 4.4 onwards
The previous assertion only considerate the existance of ESP-IDF 4.3, but with the ESP-IDF 4.3.1 release this assertion would generate errors. Now only includes from ESP-IDF 4.4 onwards.
Co-authored-by: Me No Dev <me-no-dev@users.noreply.github.com>
Current implementation clears _headers when request was sent. If the
user added custom request headers, they will be lost and the redirected
request will not contain them. This commit changes the scope of cleanup
so that the headers survive redirects but don't survive connection
reuse.
1.0.6 changed scanning method to always scan all available channels during connect. This results in results in connect taking about ~3 seconds instead of ~1. This patch changes the behavior to use WIFI_FAST_SCAN if client used a specific channel.
Summary
Only affects ETH (not WiFi)
This PR solves #5733 by allowing the DHCP IDF server to complete its tasks and only then it sets a static IP.
This also solves another related failure reported in this issue (#5733 (comment)).
fix#5733
Impact
Adds a delay in order to wait for DHCP to actually terminate before setting the static IP configuration in ETH.
When using the FTM examples I got an error in the FTM_Initiator.ino:
*"wifi:channel=0 is invalid"*
I solved it for myself by simply passing the channel argument to be 1 in *WiFi.initiateFTM*. However, a better fix would be to directly change the default channel here.
* examples: WiFi and Ethernet examples organization
* examples: Change on CI scripts to include new folder
* Library name reverted to WiFi
* examples: WiFi and Ethernet examples organization in a new folder
* examples: Eth files moved from WiFi to Ethernet lib and CMakeLists updated
The client always appends "Basic" to the authorization header, however there are other auth schemes that can be used: https://developer.mozilla.org/en-US/docs/Web/HTTP/Authentication
For example "Bearer" when using OAuth.
This PR adds a `setAuthorizationType` method to the HTTPClient which allows this scheme to be configured by the caller. Authorization type is set to "Basic" by default so this will have no impact on existing usecases.
This adds a function to WiFiClientSecure to set the ALPN protocol.
This is required for an MQTT client to connect to AWS IoT when using an AWS Custom Authorizer, as described here.
Example code snippet:
...
WiFiClientSecure wiFiClient;
// ALPN protocol, needed with AWS custom authorizer
const char *aws_protos[] = {"mqtt", NULL};
void setup() {
wiFiClient.setCACert(AWSCAPEM);
wiFiClient.setAlpnProtocols(aws_protos);
}
...
Summary
Related to the issue #5773 and #2280.
_user_defined_size is removed from EEPROMClass because it is redundant in the current code
EEPROMClass::length() returns _size that is the true available size of EEPROM
Impact
_user_defined_size is removed from EEPROMClass
EEPROMClass::length() returns _size that is the true available size of EEPROM
Summary
Modifies WiFi lib to allow dynamic buffer allocation along with SPIRAM MALLOC enabled
This gives more heap space to the users
Related PR in Arduino Lib Builder: espressif/esp32-arduino-lib-builder#47
Impact
WiFi will work the same as it was in version 1.0.6, restoring free heap.
close#5630close#5474close#5699close#5697
* SPI.h add new call to return a SS pin number used.
As code example states, the SS pin needs to be explicitly set for output for SPI to work, but the pin number have to be coded in addition to the SPI logic in the library, which means this duplicates code. It is much better to just be able to get the pin number from library itself.
* Update SPI_Multiple_Buses.ino to use new pinSS method
Simplify the example case, to show usage of pinSS method. This also simplifies the example, removing duplicated code.
Summary
New feature: Add the ability to get the peer certificate of an SSL connectio. This is useful for IoT when the root/cert trust chain has a shorter lifecylce than the device itself. Includes example code.
It adds two methods to the WiFiClientSecure client:
bool getFingerprintSHA256( uint8_t fingerprint_remote_sha256[32]) -- return true and the fingerprint (i.e. the SHA256 of the raw x509 as a DER - identical to what you see in for example your webbrowser). Or false on error.
const mbedtls_x509_crt* getPeerCertificate(); -- return the actual X509 struct or NULL on error.
Impact
No impact; backwards compatible (only adds to the API)
When compiling with verbose logging, the build would error with a message saying `len` is not defined in `tud_vendor_rx_cb()`. This change fixes the error.
* SD.open() new feature for creating all folders in path
This PR adds to the SD.open() function option to create all folders to the file.
SD.open(const char* path, const char* mode, const bool create)
Default value of create is false.
When true folders are created.
From issue #5019
* Update vfs_api.cpp
memccpy -> memcpy
* SD_MMC deinit fix
Wrong deinit function was called.
* Fix build compilation due to changes in the HW_TIMER's structs
* Fix compilation warnings and errors with USB
* Update USBCDC.cpp
* Update CMakeLists.txt
* Update HWCDC.cpp
* Implement Thread-Safe I2C based on ESP-IDF API
* Update esp32-hal.h
* use proper types for size and timeout
* Allow disabling of the HAL locks
* Limit frequency settings to prevent Interrupt WDT
* SD.open() new feature for creating all folders in path
This PR adds to the SD.open() function option to create all folders to the file.
SD.open(const char* path, const char* mode, const bool create)
Default value of create is false.
When true folders are created.
From issue #5019
* Update vfs_api.cpp
memccpy -> memcpy
* File f = open() edit
added false for create
* SDMMC frequency selection based on board type
On Olimex ESP32 EVB I/O operations with SD card can cause error when LAN is used in same time.
Problem is disappearing if SD MMC frequency lower down from SDMMC_FREQ_HIGHSPEED to SDMMC_FREQ_DEFAULT.
No problem if WiFi used instead LAN.
* Code rewritten according to https://github.com/espressif/arduino-esp32/pull/5688#pullrequestreview-759359645
esp_eth_phy_new_ksz8081 was added in IDF after version 4.3 and generates a compilation error with IDF-4.3. This change will only use esp_eth_phy_new_ksz8081 for IDF versions after 4.3
Fixes: https://github.com/espressif/arduino-esp32/issues/5573
To reproduce:
1. Run any sample code that starts a BLE server, and does not call `setCallbacks`.
2. Connect to the device using the "LightBlue" app on iOS.
3. Observe crash shown in the issue linked above.
* Fix issue #5507 "Constructor WebServer::WebServer(IPAddress addr, int port) produces an unexpected result"
"The class Webserver is declared with two explicit constructors, one with signature:
WebServer::WebServer(IPAddress addr, int port)
Using this results in a server listening on the port number obtained by converting the value of the IPAddress addr argument (in host byte order) to a uint32_t and then to a uint16_t, which is manifestly not the result that would be expected.
...
As for a fix, we can assume from these results that this constructor is not being used and therefore could simply be deleted."
* Issue 5507
Reverse changes in commit bee1e7088cef913391155f096b42cd4bb89c5c6f after discussion.
Alternative fix to be done.
* Fix issue #5507 "Constructor WebServer::WebServer(IPAddress addr, int port) produces an unexpected result"
This change adds support for multi-homed servers to libraries/WiFi. It was assumed to be there already by libraries/WebServer, but was not.
This led to unexpected results when the IP address-specific constructor of class WebServer was used (see issue 5507).
This change was tested using three concurrent instances of WebServer, one bound to the WiFi station address, one bound to the WiFi soft AP address,
and one bound to INADDR_ANY. See libraries/WebServer/examples/MultiHomedServers for the test method.
* Fix issue #5507 "Constructor WebServer::WebServer(IPAddress addr, int port) produces an unexpected result" (cont.)
This fixes what I think might be the cause of CI failures on GitHub for the previous commit, namely the absence of an include file
in examples/MultiHomedServers.
* Fix issue #5507 "Constructor WebServer::WebServer(IPAddress addr, int port) produces an unexpected result" (cont.)
Change port numbers in examples/MultiHomedServers per pull-request comment from me-no-dev ...
"for this test to be valid, both servers should be on the same port. That is how you can make sure that the functionality works."
"Using a Core Debug Level of Verbose and the WebServer serveStatic() function with the default value of nullptr for its cache_header argument, results in a LoadProhibited exception in _svfprintf_r().
This is because serveStatic() calls log_v() with cache_header corresponding to a "%s" in its format but without checking that cache_header is not nullptr, and then logv() (indirectly) calls _svfprintf_r().
On the other hand, with a Core Debug Level other than Verbose, this does not occur."
Changed serveStatic() to the check value of cache_header and if it is nullptr, instead pass an empty string to log_v().
* Add initial support for USB MSC
* Add Firmware Upload/Download With MSC
Current running firmware is available as file inside the MSC Disk. To update the firmware on the ESP, just copy a regular firmware bin into the drive
* Support overwriting of the firmware file
Overwriting a file is done totally differently on MacOS, Windows and Linux. This change supports it on all of them.
* Allow CDC, FirmwareMSC and DFU to be enabled on boot
* Add example ESP32-S2 USB-ONLY board
* Various device code optimizations
Added `end()` methods to MSC classes
Made begin() methods safe to be called multiple times
Optimized CDC class
* Fix CDC Connect/Disconnect detection in Arduino IDE on Windows
* Rework cdc_write
* Update ESP32-S2 board configs
## The problem
WiFiClient's connect method variant where no timeout is passed can block esp32 MCU and may then cause watchdog to kick in and reset the device. This behavior is different from that, what is in arduino-esp8266 core.
## Summary
Some cross-esp libraries (working both on esp32 and 8266), like PubSubClient simply call connect method on WiFiClient, to get connected to remote server. However, connect behavior varies betwen esp arduino 8266 and esp arduino 32 cores. This pull request tries introduce same behavior - to make connect method non-blocking on esp32, like it is with 8266 arduino core.
## Proposed solution
Introduce default fixed timeout that can be changed by #define - by default set to 3 seconds.
### Affected components:
WiFiClient
### Affected methods:
```c++
int connect(IPAddress ip, uint16_t port);
int connect(const char *host, uint16_t port);
```
### Impact
May impact projects or libraries using connect method variant without specified timeout, where:
- remote is located far away or
- connection is heavily limited, or
- remote is slow, when it comes to accept the connection
In [HTTPClient](https://github.com/espressif/arduino-esp32/tree/master/libraries/HTTPClient), if the `Transfer-Encoding` header is set to `identity`, an error (Transfer-Encoding not supported) will occur.
HTTPClient will consider the request as `identity` if the `Transfer-Encoding` header is not set. But it is also defined a response with `identity` explicitly set in the `Transfer-Encoding` header (ref:[MDN Web Docs](https://developer.mozilla.org/en-US/docs/Web/HTTP/Headers/Transfer-Encoding)).
This pull request will allow the request to be processed normally even when `identity` is explicitly set.
lwip lib bundled with esp32 Arduino supports only one ntp server. Any additional servers set are just silently ignored.
This default is different from esp8266 Arduino core and very confusing. Most of the examples provided uses 3 different ntp servers for redundancy while only the first one is used actually.
Addressing issue #4964
Closes#5398
Using the same non-blocking socket connect pattern for respecting connection timeout, copied from WiFiClient::connect.
WiFiClient::connect uses lwip_connect_r, whereas start_ssl_client uses lwip_connect. I haven't found what is the difference between them. I tested both, both work ok, so I kept lwip_connect.
* Fix compile archive arguments for the new toolchain
* Add menu to S2 for picking through which port to upload
Internal USB CDC requires to reset and wait for the new port (because persistence is not yet stable)
* USB CDC should also be started in main
* Fix URL and USB version for WebUSB
* Update vendor callback API
* Update CDC::write to use TX_DONE semaphore
* Update USB_Serial example
A typographical error is corrected in the 2nd line of the comment.
From = " This is un example howto use Touch Intrrerupts "
To = " This is an example of how to use Touch Intrrerupts "
* Update toolchain
* Update package_esp32_index.template.json
* add optional component dependencies after Kconfig options are known (#5404)
Until this commit, Kconfig options (e.g. CONFIG_TINYUSB_ENABLED) were
used in conditions preceding idf_component_register to determine which
components need to be added to `arduino` component requirements.
However the Kconfig options aren't known at the early expansion stage,
when the component CMakeLists.txt files are expanded the first time
and requirements are evaluated. So all the conditions evaluated as if
the options were not set.
This commit changes the logic to only add these components as
dependencies when the Kconfig options are known. Dependencies become
"weak", which means that if one of the components isn't included into
the build for some reason, it is not added as a dependency.
This may happen, for example, if the component is not present in the
`components` directory or is excluded by setting `COMPONENTS` variable
in the project CMakeLists.txt file.
This also ensures that if the component is not present, it will not be
added as a dependency, and this will allow the build to proceed.
Follow-up to https://github.com/espressif/arduino-esp32/pull/5391.
Closes https://github.com/espressif/arduino-esp32/issues/5319.
* IDF master d93887f9f
* PlatformIO updates for CI (#5387)
* Update PlatformIO CI build script
- Switch to the latest toolchains 8.4.0 for ESP32, ESP32S2, ESP32C3
- Use PlatformIO from master branch for better robustness
* Update package.json for PlatformIO
Co-authored-by: Ivan Grokhotkov <ivan@espressif.com>
Co-authored-by: Valerii Koval <valeros@users.noreply.github.com>
Commit d15e1b0e [1] in the 2.0.0 pull request #4996 introduced a filter
to connect only to WiFi networks with a RSSI of -75 or better. This
results in strage behaviour, as a scan still shows networks, that can't
be connected to, even though with older versions, connecting to those
networks was possible and the connection (albeit slow) was stable.
Remove the RSSI filter for now by setting the threshold to -127, i.e.
the lowest possible value. Maybe in the future the filter threshold
could be exposed to users, to allow filtering out nearly unreachable
networks.
[1] github.com/espressif/arduino-esp32/pull/4996/commits/d15e1b0
When iterating through APlist from first to the last element, a corrected password for already known SSID is not used. Therefore, I propose to iterate from the back of the list. With iterating from the back through the vector, an already known SSID with a corrected password is used instead.
Proposed fix to #4627 as a remediation to PR #3995#3995 introduced that a device detected on BLE would not be entered in vector if a callback has been defined.
By doing so, it was not possible anymore to have a counter in a call back (AND device in vector) to limit the number of detected device in a single scan, which could crash ESP32 as a result.
* add onMtuChanged to BLEServerCallbacks
Add method onMtuChanged to the BLEServerCallbacks so the application can be notified if the MTU size changes.
* Correct missing semicolon and misspelled method.
Correct missing semicolon and misspelled method.
