If the esp32 is reset during a i2c read cycle the slave device may be in control of the SDA line.
If the SDA line is held low, the esp32 cannot issue a START or STOP to recover the bus.
The previous code did not correctly configure the SCL output pin, and it cycled SCL 9 times with SDA Low. Since the slave device was in a READ cycle, it just continued outputting the bits of the current byte. When the ACK/NAK bit space occurred, The low output value of SDA was interpreted as ACK so the slave device continued with the next byte. It never terminated the READ cycle.
This new code will correctly recover from an interrupted READ
* ReSTART fix, Sequencing fix
pr #1665 introduce a problem with ReSTART, when solving this problem I found an interaction between the TxFifo refill, RxFifo empty and CMD[] fill. during certain sequences a dataqueue command would be skipped, this skipping resulted in a mismatch between the contents of the TxFifo and the i2c command sequence. The problem manifested as an ACK error.
In addition to this required bug fix I propose:
* `Wire.begin()` be changed from a `void` to a `bool` this will allow the reset functionality of `Wire.begin()` to be reported. Currently `Wire.begin()` attempts to reset the i2c Peripheral, but cannot report success/failure.
* `Wire.busy()` be added. this `bool` function returns the hardware status of the bus. This status can be use in multi-master environments for application level interleaving of commands, also in single master environment, it can be used to detect a 'hung' bus. With the functional change to `Wire.begin()` this allows app level recover of a hung bus.
* `Wire.lastError()` value updated for all errors, previously when interleaving `Wire.endTransmission(false)` and `Wire.readTransmission(false)`, the 128 byte `Wire.write()` buffer was exhausted without generating and error(very exotic). I discovered this error when I created a sequence of directed reads to a EEPROM. Each directed read used 2 bytes of the 128 byte `write()` buffer, so after 64 consecutive ReSTART writes with ReSTART reads, `Wire()` had no room to record the directed address bytes. It generated just a NAK check without setting the EEPROMs internal register address. The succeeding ReSTART read succeeded at incorrect address.
* Changes to the HAL layer:
** added `i2cGetStatus()` which returns the i2c peripheral status word, used to detect bus_busy currently
** added `i2cDebug()` programmatic control of debug buffer output
** changed `i2cAddQueue()` to allow data_only queue element this will allow a i2c transaction to use multiple data pointers.
** removed direct access to DumpInts(), DumpI2c() from app, use i2cDebug() to set trigger points
*
* Update esp32-hal-i2c.c
* Update Wire.cpp
* ReSTART, Sequencing
pr #1665 introduce a problem with ReSTART, when solving this problem I found an interaction between the TxFifo refill, RxFifo empty and CMD[] fill. during certain sequences a dataqueue command would be skipped, this skipping resulted in a mismatch between the contents of the TxFifo and the i2c command sequence. The problem manifested as an ACK error.
In addition to this required bug fix I propose:
* `Wire.begin()` be changed from a `void` to a `bool` this will allow the reset functionality of `Wire.begin()` to be reported. Currently `Wire.begin()` attempts to reset the i2c Peripheral, but cannot report success/failure.
* `Wire.busy()` be added. this `bool` function returns the hardware status of the bus. This status can be use in multi-master environments for application level interleaving of commands, also in single master environment, it can be used to detect a 'hung' bus. With the functional change to `Wire.begin()` this allows app level recover of a hung bus.
* `Wire.lastError()` value updated for all errors, previously when interleaving `Wire.endTransmission(false)` and `Wire.readTransmission(false)`, the 128 byte `Wire.write()` buffer was exhausted without generating and error(very exotic). I discovered this error when I created a sequence of directed reads to a EEPROM. Each directed read used 2 bytes of the 128 byte `write()` buffer, so after 64 consecutive ReSTART writes with ReSTART reads, `Wire()` had no room to record the directed address bytes. It generated just a NAK check without setting the EEPROMs internal register address. The succeeding ReSTART read succeeded at incorrect address.
* Changes to the HAL layer:
** added `i2cGetStatus()` which returns the i2c peripheral status word, used to detect bus_busy currently
** added `i2cDebug()` programmatic control of debug buffer output
** changed `i2cAddQueue()` to allow data_only queue element this will allow a i2c transaction to use multiple data pointers.
** removed direct access to DumpInts(), DumpI2c() from app, use i2cDebug() to set trigger points
*
* Forgot DebugFlags Return
@andriyadi found this, total brain fade on my part.
This version no longer needs an interrupt for each byte transferred. It only needs interrupts for START, STOP, FIFO empty/Full or error conditions. This dramatically reduces the interrupt overhead. I think the prior version was causing an interrupt overload condition where the ISR was not able to process every interrupt as they happened.
the 'eject' ERROR is and indication of an interrupt triggering without an source. I am working to eliminate these serviceable interrupt. This update increase stability on a HelTek Wifi Lora 32 board. with a SSD1306 OLED. This update fixes a glaring error in the interrupt allocation code, the Interrupt mask was wrong. I also dynamically adjust the FiFo thresholds based on Bus clockrate. The change to FiFo thresholds has reduced the number for 'eject' events. I also change 'eject' from and ERROR to DEBUG. An 'eject' event does not compromise i2c transmissions. It happens after a transaction has completed.
Chuck.
If Core Debug Level is at DEBUG, a confusing debug message will be emitted if the I2C transaction takes longer complete than the calculated minimum time. This original debug message was just to prove that this new i2c code could correctly handle SCL stretching or interrupt latency issues. This delay is not a problem, or an error. Usually it is caused by a higher priory interrupt starving the i2c ISR. Usually WiFi is the culprit. As long of this delay is within the configured timeout (by default 50ms, or can be set with Wire.setTimeOut(milliseconds);) no problem will occur and the transaction will successfully complete.
Chuck.
* fix sdmmc config
* Fix warnings in EEPROM
from @Curclamas
* remove leftover TAG in EEPROM
* Initial add of @stickbreaker i2c
* Add log_n
* fix warnings when log is off
* i2c code clean up and reorganization
* add flags to interrupt allocator
* fix sdmmc config
* Fix warnings in EEPROM
from @Curclamas
* remove leftover TAG in EEPROM
* fix errors with latest IDF
* fix debug optimization (#1365)
incorrect optimization for debugging tick markers.
* Fix some missing BT header
* Change BTSerial log calls
* Update BLE lib
* Arduino-ESP32 release management scripted (#1515)
* Calculate an absolute path for a custom partitions table (#1452)
* * Arduino-ESP32 release management scripted
(ready-to-merge)
* * secure env for espressif/arduino-esp32
* * build tests enabled
* gitter webhook enabled
* * gitter room link fixed
* better comment
* * filepaths fixed
* BT Serial adjustments
* * don't run sketch builds & tests for tagged builds
* Return false from WiFi.hostByName() if hostname is not resolved
* Free BT Memory when BT is not used
* WIFI_MODE_NULL is not supported anymore
* Select some key examples to build with PlatformIO to save some time
* Update BLE lib
* Fixed BLE lib
* Major WiFi overhaul
- auto reconnect on connection loss now works
- moved to event groups
- some code clean up and procedure optimizations
- new methods to get a more elaborate system ststus
* Add cmake tests to travis
* Add initial AsyncUDP
* Add NetBIOS lib and fix CMake includes
* Add Initial WebServer
* Fix WebServer and examples
* travis not quiting on build fail
* Try different travis build
* Update IDF to aaf1239
* Fix WPS Example
* fix script permission and add some fail tests to sketch builder
* Add missing space in WiFiClient::write(Stream &stream)
This change uses the `_RST` bit of the i2c peripheral to force the hardware into power on reset state. The prior code assumed the peripheral was already in a reset, it just cleared the reset bit.
Chuck.
The existing code did not follow protocol with 10bit addressed devices. Per _Philps/NXP Semiconductors UM10204 I2C-bus specification and user manual Rev. 6 4April2014_ pg.15 3.1.11 10-bit addressing:
~The first seven bits of the first byte are the combination of 1111 0xx of which the last two bits (xx) are the two Most-Significant Bits (MSB) of the 10-bit address; the eighth bit of the first byte is the R/!W! bit the determines the direction of the message~
The i2cWrite() function was returning to the app before the i2c transaction had completed. This caused the next Wire() call to return a I2C_ERROR_BUSY.
* Addition of a i2cReset method and timeout handling for the case where the i2c hardware FSM (state machine) gets stuck in a busy state.
* Use newly added i2cReset function within the wire library.
IDF Options:
- Autostart Arduino (implements app_main)
- Disable HAL locks
- Set HAL debug level
- Auto-connect STA if configured (else will connect after WiFi.begin())
