We are pleased to announce the Mbed OS 5.11.2 release is now available. This is the latest patch release based on the feature set that Mbed OS 5.11 introduces.
In this release we have added target support for NUCLEO_H743ZI, GD32_E103VB and GD32_F450ZI.
We have added fixes for the following reported issues:
- 8567 Equeue unchain bugs
- 8913 Build of NUCLEO_F030R8 fails when DEVICE_SERIAL=0
- 8903 Realtek_RTL8195AM is printing a lot of traces
- 9196 ESP8266 CTS/RTS fails in stop_uart_hw_flow_ctrl()
- 9075 K82F QSPI test fails intermittently
- 9216 Defining MBED_SLEEP_TRACING_ENABLED caused error
- 9266 Astyle in travis doesn't use .astyleignore
- 9255 FLASHIAP component tests failed
- 9207 MCU_LPC546XX - breaks on master
- 9272 Block device test : skip multi thread part in case of small RAM
Support has been added for:
- PSA APIs on Pelion supported boards
- A bootloader on the DISCO_F469NI
There have also been some updates to improve the quality of the documentation.
- There was an issue with the internal KVStore, identified after the 5.11 release. The fix for this required changes which would normally not be accepted into a patch release. We have made an exception in this case due to the significant nature of the issue. Prior to the fix provided in 9156, the space left for the internal TDBStore (when working in the default internal configuration) was allocated from the sector following the end of the application until the end of flash. While this worked, it was problematic if the application would have crossed the sector boundary after an update. The fix leaves the application a couple of sectors to grow at the expense of the allocated storage (if they are available, which is the likely case). The fix is defined as a breaking change, even if not a typical one: Boards that use the default internal configuration will lose the storage content after the upgrade (as the internal storage start will move forwards). Two boards will be affected: K66F and FUTURE_SEQUANA_M0_PSA. Non default configurations won't be harmed.
- The device key in versions before 5.11.2 may have been insecure as the TRNG may have unknowingly failed. This issue has been addressed in this release by 9278 DeviceKey: [Security Fix] Generated ROT-key is still used when TRNG fails.
There are also a number of other fixes and code improvements.
For full details of this and previous releases, please visit our releases page.
You need to log in to post a discussion