Building GCC 5.1.0 for arm-eabi
These notes describe building GCC 5.1.0 for the Cortex M4 on Mac OS X.
My target board is the STM32M4F29I Discovery.
Read more Β»These notes describe building GCC 5.1.0 for the Cortex M4 on Mac OS X.
My target board is the STM32M4F29I Discovery.
Read more Β»The public version of the AdaCore Ravenscar RTS is released under the full GPL. It seemed as if it would be a good idea (and fun!) to produce an independent RTS with the GCC Runtime Library Exception(FAQ).
Read more Β»Iβve been porting my Cortex-based RTS to Arduino Due, using the J-Link JTAG interface (because who can make an RTS without a debugger?). Iβve probably forgotten my initial problems with ST-LINK (an equivalent interface included for free on the STM32F4 board, supported by OSS software), but it seemed a much less trying experience than with Segger.
There are two things with J-Link GDB Server: first, it seems like a bad idea to keep the server alive after a GDB session ends, so start it with -singlerun; and second, you absolutely must issue monitor reset before running your program. If you donβt youβll end up like I did, scratching my head for three days trying to understand why I was getting a HardFault at the SVC that FreeRTOS uses to kick tasking off.
A watchdogis used to detect when a system has gone off with the fairies; you have to reset the watchdog timer ("pat the watchdog") every so often or it takes some recovery action. In the case of the Arduino Due, with the ATSAM3X8E MCU, the recovery action is to reset the CPU.
In the ATSAM3X8E (and probably other Atmel MCUs, too), the watchdog timeout defaults to 16 seconds, and the default hardware state is that the watchdog is enabled! (the default in the Atmel Software Framework is to disable the watchdog unless you have defined CONF_BOARD_KEEP_WATCHDOG_AT_INIT
).
This is a note on overcoming a problem connecting to my STM32F429I-DISCOboard.
Read more Β»I seem to have fritzed one or two of the pins on my STM32F429I-DISC0 board (now replaced by an updated version, STM32F429I-DISC1; that last character was a 0!); after some hair-pulling it turns out that one of said pins is used by the only externally-accessible I2C peripheral on the board; so, until the replacement arrives, here are some interesting facts about the PCF8547A.
Read more Β»This is a note on building support for the MPU92509-axis chip in the AdaPilotproject (the AdaRacerhardware). It continues this report.
Read more Β»This note is about using Segger J-Link software to help debug code on a BBC micro:bit.
Read more Β»Recently, on the Gitter Ada Programming Language lobby, there have been discussions on cross-compiling for Cortex M micro-controllers, specifically the BBC micro:bit, using the Raspberry Pi.
TL;DR: it's not going to be easy, unfortunately. Maybe later (or earlier?) Raspbian releases will help.
Read more Β»This article discusses various issues rebuilding a demonstrator intended for the BBC micro:bit (version 1.3b) after an interval of several years.
Read more Β»I have written an application in Ada and I want to run it on Android (ARM devices). I have compiled a native standalone library for ARM (arm64-v8a) on a Raspberry Pi, containing the proper JNI. However, when I launch the application in Android it fails while loading the library (right on the System.loadLibrary call). Is there something theoretically wrong in proceeding this way, or should it work?
Before doing this experiment on android, I successfully loaded an Ada test library on Java using a x64 PC, so I don't think the problem is in the interface itself. I would have expected the library to be loaded without issues.
For the full code to see what I am trying to do, please, have a look at GuillermoHazebrouck/gnav on GitHub.