We were never going to be capable of supporting every single part
variant with their own files, so stop trying. We've been supporting the
linker script generator for a long time now, so move on from these old
static files.
Breaking: if you were using one of these, and don't wish to use the
linker script generator, you should simply declare a stub linker script,
and include the generic section definitions file, and provide this in
your own application.
Old:
LDSCRIPT = $(OPENCM3_DIR)/lib/stm32/l1/stm32l15xxb.ld
New (linker script generator):
DEVICE=stm32l151cb
New (manual):
* Add file mymemorymap.ld, with contents similar too:
```
MEMORY
{
rom (rx) : ORIGIN = 0x08000000, LENGTH = 256K
ram (rwx) : ORIGIN = 0x20000000, LENGTH = 48K
}
/* Include the common ld script. */
INCLUDE cortex-m-generic.ld
```
LDSCRIPT=mymemorymap.ld
See ld/README for more information on using the linker script generator
The leading - makes it rather inconsistent with the majority of other
projects around the world. Use the form everyone else uses.
To solve this, properly pass prefix to inner makes as was always
intended.
Fixes: https://github.com/libopencm3/libopencm3/issues/1058
Instead of every "simple" target having their own duplicate file with
all the section mappings, just provide a single, simple,
"cortex-m-generic.ld" that works with our startup code and any simple
rom/ram system. This also drops the pointless copying of files all over
the place. Using -L flags properly is sufficient, and the standard file
is now in the root of the library already.
This is not strictly required, as this part is supported by the
devices.data linker script generator tool. However, as we're still in
migration for that tool, and this is the first time we're getting proper
lm3s(qemu) support, keep it for now.
Renamed every instance of variable CFLAGS in target specific Makefiles
to TGT_CFLAGS to free up CFLAGS for user defined compiler flags.
Added information in README.md about existence and usage of CFLAGS
environment variable in build process.
-ggdb3 make slightly bigger .elf files, but allows gdb to understand
macros, which libopenocm3 uses somewhat extensively. Make this the
default, and pull it up to the common base makefile, so it can be easily
substituted.
On linux, the output of CP rule was try to write to / which is - of course, forbidden for write.
This solution adds to each part of lib correct pointer to the root of lib where the libs should be written.
Bug found by Kuldeep Singh Dhaka.
We currently default to "-mfloat-abi=hard -mfpu=fpv4-sp-d16" for M4F cores, and
and variations of "-mfloat-abi=soft" for the others. Keep the M4F default, and
move others to no FP flags for consistency, but allow overriding these flags
via the FP_FLAGS environment variable.
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Added --terse and --mailback options to the make stylecheck target. It
also does continue even if it enounters a possible error.
We decided on two exceptions from the linux kernel coding standard:
- Empty wait while loops may end with ; on the same line.
- All blocks after while, if, for have to be in brackets even if they
only contain one statement. Otherwise it is easy to introduce an
error.
Checkpatch needs to be adapted to reflect those changes.
This also report bug fixes about data alignment and bss storage to EFM32 &
LM3S targets.
Note: removed chrysn copyright statement as the file is a verbatim copy of
previous files.
previously, only stm32 chips passed the information about which chip to
build on into the compiler. this information is essential to dispatch,
thus defining LPC13XX, LPC17XX, LPC43XX and LM3S in analogy to
STM32F1..4.
the cortex generic interrupts get moved to lib/cm3/vector.c, the
platorms' individual irq names, initialization and handler prototypes go
to platoform specific irq.h files.
as the vector.c file heavily depends on platoform specific headers, it
can't be built once-and-for-all in lib/cm3/, so there are inclusion
stubs in the various architecture dirs; this might be better solved with
Makefile / include path handling.
one particular file is lib/lpc43xx/vector.c; that platform's
initialization code contains an additional section to copy everything
from flash to ram (which probably performs better there). that code
still resides in the inclusion stub, and gets mashed in using defines.
would need a cleaner implementation together with the Makefile solution.
this commit contains some files of the upcoming efm32 branch, from which
it was cherry-picked.
the .bin files produced from before and after this commit only differ in
lpc43xx, where the startup sequence was subtly modified.
Adds libopencm3/cm3/assert.h header that provides assertion check macros
similar to those provided by the standard C library.
Thanks to Nicolas Schodet for help.
That way, data source address does not depend on any other unrelated change in
linker script.
This also fixes cases when .data input section is aligned on 8 bytes. The new
version does not provide any address for the output section so that it is
aligned to the strictest input section. The _data_loadaddr symbol will
always take this alignment into account.
- The library files are now being built into the lib subdirectory of the
source.
- The linker files for each library are being copied into the lib source
subdirectory.
Motivation: The relative locations of files in the source directory after make
are now the same as after make install now. This makes it easier to
reuse examples with their makefiles outside of the libopencm3
sourcecode directory.
Changed the license to LGPL3+ for some makefiles that got ommited in the
previous big patch commit.
Corrected a find and replace error in a linker file.
Karl Palsson
Karl Palsson
Daniele Lacamera
Tido Klaassen
Karl Palsson
Frantisek Burian
Alexandru Gagniuc
Piotr Esden-Tempski
Nicolas Schodet
Ken Sarkies
chrysn
Thomas W. Barr
Tomaz Solc
Gareth McMullin
Uwe Hermann
Fergus Noble