README

This repository contains assorted example projects for unicore-mx.

The unicore-mx project aims to create an open-source firmware library for various ARM Cortex-M microcontrollers.

The examples are meant as starting points for different subsystems on multitude of platforms.

Feel free to add new examples and send them to us either via the mailinglist or preferably via a github pull request.

Usage

You must run "make" in the top level directory first. This builds the library and all examples. If you're simply hacking on a single example after that, you can type "make clean; make" in any of the individual project directories later.

For more verbose output, to see compiler command lines, use "make V=1" For insanity levels of verboseness, use "make V=99"

The makefiles are generally useable for your own projects with only minimal changes for the unicore-mx install path (See Reuse)

Make Flash Target

For flashing the 'miniblink' example (after you built unicore-mx and the examples by typing 'make' at the top-level directory) onto the Olimex STM32-H103 eval board (ST STM32F1 series microcontroller), you can execute:

cd examples/stm32/f1/stm32-h103/miniblink
make flash

The Makefiles of the examples are configured to use a certain OpenOCD flash programmer, you might need to change some of the variables in the Makefile if you use a different one.

The make flash target also supports a few other programmers. If you provide the Black Magic Probe serial port the target will automatically choose to program via Black Magic Probe. For example on linux you would do the following:

cd examples/stm32/f1/stm32-h103/miniblink
make flash BMP_PORT=/dev/ttyACM0

This will also work with discovery boards that got the st-link firmware replaced with the Black Magic Probe firmware.

In case you did not replace the firmware you can program using the st-flash program by invoking the stlink-flash target:

cd examples/stm32/f1/stm32vl-discovery/miniblink
make miniblink.stlink-flash

If you rather use GDB to connect to the st-util you can provide the STLINK_PORT to the flash target.

cd examples/stm32/f1/stm32vl-discovery/miniblink
make flash STLINK_PORT=:4242

Flashing Manually

You can also flash manually. Using a miriad of different tools depending on your setup. Here are a few examples.

OpenOCD

openocd -f interface/jtagkey-tiny.cfg -f target/stm32f1x.cfg
telnet localhost 4444
reset halt
flash write_image erase foobar.hex
reset

Replace the "jtagkey-tiny.cfg" with whatever JTAG device you are using, and/or replace "stm32f1x.cfg" with your respective config file. Replace "foobar.hex" with the file name of the image you want to flash.

Black Magic Probe

cd examples/stm32/f1/stm32vl-discovery/miniblink
arm-none-eabi-gdb miniblink.elf
target extended_remote /dev/ttyACM0
monitor swdp_scan
attach 1
load
run

To exit the gdb session type <Ctrl>-C and <Ctrl>-D. It is useful to add the following to the .gdbinit to make the flashing and debugging easier:

set target-async on
set confirm off
set mem inaccessible-by-default off
#set debug remote 1
tar ext /dev/ttyACM0
mon version
mon swdp_scan
att 1

Having this in your .gdbinit boils down the flashing/debugging process to:

cd examples/stm32/f1/stm32vl-discovery/miniblink
arm-none-eabi-gdb miniblink.elf
load
run

ST-Link (st-util)

This example uses the st-util by texane that you can find on GitHub.

cd examples/stm32/f1/stm32vl-discovery/miniblink
arm-none-eabi-gdb miniblink.elf
tar extended-remote :4242
load
run

Reuse

If you want to use unicore-mx in your own project, this examples repository shows the general way. (If there's interest, we can make a stub template repository)

1. Create an empty repository

   mkdir mycoolrobot && cd mycoolrobot && git init .
   

2. Add unicore-mx as a submodule

   git submodule add https://github.com/insane-adding-machines/unicore-mx
   

3. Grab a copy of the basic rules These urls grab the latest from the unicore-mx-examples repository

   wget \
     https://raw.githubusercontent.com/insane-adding-machines/unicore-mx-examples/master/examples/Makefile.rules \
     -O unicore-mx.rules.mk
   

4. Grab a copy of your target Makefile in this case, for STM32L1

   wget \
     https://raw.githubusercontent.com/insane-adding-machines/unicore-mx-examples/master/examples/stm32/l1/Makefile.include \
     -O unicore-mx.target.mk
   

5. Edit paths in unicore-mx.target.mk Edit the last line of unicore-mx.target.mk and change the include to read include ../unicore-mx.rules.mk (the amount of .. depends on where you put your project in the next step..

6. beg/borrow/steal an example project For sanity's sake, use the same target as the makefile you grabbed up above)

   cp -a \
     somewhere/unicore-mx-examples/examples/stm32/l1/stm32ldiscovery/miniblink \
     myproject
   

Add the path to UCMX_DIR, and modify the path to makefile include

diff -u
---
2014-01-24 21:10:52.687477831 +0000
+++ Makefile    2014-03-23 12:27:57.696088076 +0000
@@ -19,7 +19,8 @@

 BINARY = miniblink

+UCMX_DIR=../unicore-mx
 LDSCRIPT = $(UCMX_DIR)/lib/stm32/l1/stm32l15xxb.ld

-include ../../Makefile.include
+include ../unicore-mx.target.mk

You're done :)

You need to run "make" inside the unicore-mx directory once to build the library, then you can just run make/make clean in your project directory as often as you like.