gdb will actually accept as little "tar ext :4242" but use the full
commands to help clarify the intent for people who might not be as
familiar with the syntax.
This adds a section to the bottom of the .map file containing the
symbol->object file mapping. (the logical inverse of the existing
section at the start of the .map file that lists objects, and the
symbols that pulled them in)
Reviewed-by: Karl Palsson <karlp@tweak.net.au>
can fixes are compile tested only. I don't have those boards.
f3 i2c is tested on i2c and gives the same results
l1 low power demo seems to have been broken for some time already.
=> should be fixed, but not something I'm going to work on right now.
There was an attempt at providing support for the texane/stlink utility.
It has been broken for a long time, referring to missing scripts.
Remove it as clearly unused.
This commit does two things, it removes the "Quiet" flag
from the flash target so you can see what it is trying to
do when it trys to flash your code, and it adds a make
target print-% which can be used to print the resolved
value of make variables (also helpful in figuring out
what it is trying to do).
The "timer" example is actually "the same" for both f1 and f4. Do a
sanity sweep over the commentary, remove all vestiges that this was
cloned from a motor control example, and synchronize both examples.
Future work should extract the common portions "somewhere" but at least
make them consistent for now.
This example was functionally similar to the "other" example, but had
the actual functional mouse movement disabled. To reduce maintennance
cost with slightly divergent examples, simply drop the less useful
example.
The example is pointlessly complicated and includes lots of dead code,
but at least remove the wasteful and confusing references to the
temperature channel as pointed out on github.
Using GPIOA_IDR directly looks a bit arcane, and the example can benefit
from introducing gpio_get() and how to use it. The stm32f0-discovery
example already does it like this.
When using netcat to send commands to an already existing openocd
session, you must path full paths to the file to flash. There's no
guarantee that it is running from the same directory as the present
example
Some demos will use object files from a different directory and
those files would not be cleaned by the simple CLEAN rule. This
small change insures that those objects and their dependency files
are also cleaned.
Instead of OOCD_SERIAL triggering an attempt to use a specific ftdi based
adapter, introduce common target/interface pairs, OOCD_INTERFACE and
OOCD_TARGET, and an override via OOCD_FILE allowing full control.
Further, attempt to connect to a running openocd instance to flash if possible.
This _will not_ work well if you have multiple openocd instances open at once,
but that's a rare use case.
Examples:
OOCD_INTERFACE=stlink-v2
OOCD_TARGET=stm32f0x
--> openocd ... -f interface/stlink-v2.cfg -f target/stm32f0x.cfg)
OOCD_FILE=board/ek-lm4f120xl.cfg
--> openocd ... -f $(OOCD_FILE) (ie, the variable is passed untouched)
As part of implementing this, correct variables that were always wrong, and
update the openocd flashing commands to use current commands.
We already have "make examples" and "make lib" but make clean cleans
everything. Not super useful for end users, but rather handy for people
working on the examples.
As in the main library, instead of constructing and making our own CFLAGS
CXXFLAGS and LDFLAGS variables, use TGT_ for ones we construct, and include any
user provided variables afterwards.
Using the latest (as of today) gcc-arm-embedded toolchain caused two build
failures similar to:
error: unknown type name 'cookie_io_functions_t'
These custom IO functions are now protected by a define, so define _GNU_SOURCE
which enables all such areas. This is a libc issue.
Signed-off-by: Trevor Woerner <twoerner@gmail.com>
If the library files are missing (haven't been built) it's better to get a
failure message saying that, rather than a failed dependency with no rules to
generated it.
Before: (I'd manually removed the f4 library)
```
$ make clean all V=1
Using ../../../../../libopencm3/ path to library
rm -f *.o *.d *.elf *.bin *.hex *.srec *.list *.map
make: *** No rule to make target `adc-dac-printf.elf', needed by `elf'. Stop.
```
After:
```
$ make clean all V=1
Using ../../../../../libopencm3/ path to library
rm -f *.o *.d *.elf *.bin *.hex *.srec *.list *.map
arm-none-eabi-gcc -Os -g -Wextra -Wshadow -Wimplicit-function-declaration
-Wredundant-decls -Wmissing-prototypes -Wstrict-prototypes -fno-common
-ffunction-sections -fdata-sections -MD -Wall -Wundef
-I../../../../../libopencm3//include -DSTM32F4 -mthumb -mcpu=cortex-m4
-mfloat-abi=hard -mfpu=fpv4-sp-d16 -o adc-dac-printf.o -c adc-dac-printf.c
arm-none-eabi-gcc --static -nostartfiles -L../../../../../libopencm3//lib
-T../stm32f4-discovery.ld -Wl,-Map=adc-dac-printf.map -Wl,--gc-sections -mthumb
-mcpu=cortex-m4 -mfloat-abi=hard -mfpu=fpv4-sp-d16 adc-dac-printf.o
-lopencm3_stm32f4 -Wl,--start-group -lc -lgcc -lnosys -Wl,--end-group -o
adc-dac-printf.elf
/home/karlp/tools/gcc-arm-none-eabi-4_9-2015q2/bin/../lib/gcc/arm-none-eabi/4.9.3/../../../../arm-none-eabi/bin/ld:
cannot find -lopencm3_stm32f4
collect2: error: ld returned 1 exit status
make: *** [adc-dac-printf.elf] Error 1
```
I strongly feel that the latter is far more informative.
I'd love to meet the person who managed to use flossjtag interface
with a Stellaris board, but until we can meet, use the correct
interface, which is ti-icdi. Note that this setting is redundant,
since the board config file already specifies the correct interface;
however, OOCD_INTERFACE variable cannot be empty, by virtue of how
higher-level makefiles are written.
