New MINOR release 04.03.00

Changes since 04.02.00:

- fix: wrong comment
- feat: temperature safety check (#12442)
- chore: bump FreeRTOS (11.0.1) and LWIP (2.2.0) versions

New MINOR release 04.02.00

Changes since 04.01.01:

- feat: self-centering FB limit around current SP (#11727)
- fix: log in late setup phase and before fatalErrors
- feat: Gen.3 power-on self test (#11192)
- fix: adapt for actual Gen.3 carrier specs
- chore: bump to FreeRTOS V10.6.1
- chore: formatting CMakeLists.txt and increase minimum version
- fix: compilation fails for debug builds
- chore: code formatting

New PATCH release 04.01.01.

Changes since 04.01.00:

- fix watchdog error message at every boot
- update libmodbus to v3.1.10

New MINOR release 04.01.00.

Changes since 04.00.00:

- fix: do not toggle LD_SHDN (keep always enabled)
- fix mis-interpretation of INT_PD signal
- Add support for Gen.2 Mezzanines in Gen.2 Carriers (#10310)

- update FreeRTOS kernel and libmodbus
- fix warnings and add some comments
- update MCUXpresso Config Tools to v12.1
- use local mirrors for submodules

New BACKPORT release 03.03.01

Changes since 04.00.00:

- revert change of IRQ frequency introduced with 03.03.00

New MINOR release 03.03.00

Changes since 03.02.02:

- reduce IRQ load by changing power feedback loop frequency from 100 kHz
  to 1 kHz. NOTE: This alters the step size used for ramping up/down the
  current also by a factor 100!
- attempt to fix DHCP issues: disable fallback to hard-coded static IP
  and instead reboot the LDD if DHCP did not succeed within a certain
  timeout.

New MAJOR release 04.00.00

Changes since 03.02.02:

- Full reimplementation based on FreeRTOS and lwIP

New PATCH release 03.02.02

Changes since 03.02.01:

- Revert "comment out some debug printfs spamming the console"
  The removal of the printfs was causing spurious timeouts in the communication on LDD8 boxes. Those timeouts happen
  quite frequently, which makes it very hard to interact with the LDD through the control system!

New PATCH release 03.02.01

Changes since 03.02.00:

- fix not protecting HW watchdog setup and feeding sequences (should fix #7217)
- print watchdog reset count at firmware start
- comment out some debug printfs spamming the console

New MINOR release 03.02.00

Changes since 03.01.01:

- add interlock detection and latching (see #6449)
- add watchdogs to critical task running the interlock detection (both hardware and software watchdog)

- make CAN communication less verbose on serial console

- add clang-format file (from MSK project-template)
- fix update scripts not working with spaces in file name
- add multi remote update script

New PATCH release 03.01.01

Changes since 03.01.00:

fix remote FW update for CAN slave

The problem was caused by too wide use of the CAN semaphore. Instead of holding the semaphore for the entire process we need to obtain and release it for each operation on the CAN bus. It is not 100% clear why this is the case. Also it is not clear whether this has side effects, since other communication on the CAN bus might take place in between.

New MINOR release 03.01.00

Changes since 03.00.06:

Register interface change (backwards compatible). See version 1.7 of register address space definition document on redmine.

- added register for readback of actual current setpoint
- introduce flag to identify untagged/inofficial builds
- fixed a wrong fall-through case, was affecting the CPU ID readback.

New PATCH release 03.00.06

Changes since 03.00.05:

- Removed one Compliler warning.
- Added all functions for reading carrier device ID.
- Started I2C mux preparation
- made some parts of the CAN communication theoretically more reliably, unfortunately this does not help with the issues with concurrent bus access
- added missing wait/post on the CAN semaphore when sending normal (i.e. non-firmware upgrade) frame on master for a CAN slave

- added information how to enable automatic download before debugging
- added instructions how to setup a debugger in QtCreator
- Fixed cmake not properly being configured for a cross platform target. It was producing proper output files but preventing the use of a debugger.

New PATCH release 03.00.05

Changes since 03.00.04:

- Fixed byte order when reading the FIR parameter registers (writing was correct).

JTAG programming scripts:

- added new script to update LDD firmware via JTAG with openocd
- renamed the jtag update script which uses the JLink GDB server for better clarity
- changed JLink GDB server script to use JTAG instead of SWD to be compatible with the LDD-8 boxes

New PATCH release 03.00.04

Changes since 03.00.03:

- The bit shift for the supply voltage and current read out has now been corrected. The mistake was that according to the (very confusing) datasheet the data width for current measurements is 10+1 and for voltage measurements is 10 bits. Accordingly the bit shifts shall be 5 resp. 6 bits.

New PATCH release 03.00.03

Changes since 03.00.02:

- fixed some comminucation issues which lead to only 8 of the 9 ports being open at a time

New PATCH release 03.00.02

Changes since 03.00.01:

- fixed incorrect bit shifts in supply voltage and current monitor. The shift now is according to the data sheet of the PAC1720.

New PATCH release 03.00.01

Changes since 03.00.00:

- Fixed possible blocking communication tasks if many connections are opened at the same time (e.g. when using the LDDMonitor tool).

New MAJOR release 03.00.00

Changes since 02.05.00:

- force current set point to 0 if LD power is switched off - otherwise there might be some lower current generated even when switched off...
- allow connection to master LDD via id 0 and 8 (i.e. ports 4876 and 4884)

- switched build environment to cmake to be independent of Kinetis Design Studio (which still can be used, see README file)

- some code cleaning: fixed warnings, possibly undefined values and wrong data types. Behavioural change cannot be excluded.
- reformatted the source code for readability

(no annotation given in original tag)

This tag was renamed from 2.5.0 to 02.05.00