d87e751f71
Added directories with autogenerated files to the ignore list.
2012-11-07 16:42:11 -08:00
971a295d43
Adjusted path to nvic.h.
2012-11-07 16:28:29 -08:00
d417666095
Eliminate redundant gpio code from f2/f4/l1
...
Implemented as per exti2
2012-11-07 21:58:16 +00:00
db5d19b4d4
Update documentation to reflect reality
2012-11-07 21:54:18 +00:00
bffaa72f7e
Remove debug code.
2012-11-07 21:54:18 +00:00
ce8f47e7df
Enable nvic and exti support for L1
...
And include an example that uses it.
2012-11-07 21:54:18 +00:00
b860504fed
Add L1 to nvic, now that it's all generated
2012-11-07 21:50:27 +00:00
25acaa5878
Add usart support for L1.
...
Only tested with basic tx blocking, ie, the same example code as on
F2/F4, but the description of the block is almost identical.
2012-11-07 21:50:27 +00:00
2e5cfa39a5
Ignore generated files
2012-11-07 21:50:27 +00:00
4941286454
STM32L1 support, rebased onto upstream generalizations branch.
...
Working example again.
2012-11-07 21:50:27 +00:00
e4f84278f2
Add most of the rcc functions.
...
(Add the forgotten gpio.c file from before)
2012-11-07 21:50:27 +00:00
9aed64a19d
Finish RCC definitions => Working example!
2012-11-07 21:50:27 +00:00
8318384cf1
More progress towards L1 support.
...
Believe gpio is complete, but untested without finishing at least the
RCC defines.
RCC defines are a work in progress
2012-11-07 21:50:27 +00:00
2011941b55
Vectors and Memory Map for STM32L1 series
2012-11-07 21:50:27 +00:00
f1f1aa84f3
Use make flags to help control where the library is installed
...
Where the library is installed has quite an affect on what -L and -I
flags are used. If you install into the toolchain, you don't want
to use them at all, but if you install out of tree (/opt/mycm3 for
example) you need to specify the -L and -I flags.
Update the documentation and the example makefiles to support this
2012-11-07 21:50:27 +00:00
507a1742dc
Merge pull request #57 F2/F4 timer merging action
...
Merge remote-tracking branch 'tdaede/f4-timer'
2012-10-31 13:47:54 -07:00
76c7357c00
Fixed header clean target find call for Darwin BSD find too.
2012-10-31 13:45:58 -07:00
8091f10576
Merge pull request #56 Allow make to build libs/examples in parallel.
...
Merge remote-tracking branch 'karlp/pr_parallel_makes'
2012-10-31 13:43:44 -07:00
08129a5644
Fixed stm32f4-discovery/usb_cdcacm example for deduplication patches.
2012-10-31 13:41:24 -07:00
def9b9f381
Added new f4 adc module to the library makefile.
2012-10-31 13:39:24 -07:00
f5df25d6de
Merge pull request #55 STM32F4: New adc module
...
Merge remote-tracking branch 'ksarkies/adc-f4'
2012-10-31 13:38:51 -07:00
60aa32e2b9
Merge pull request #53 CAN fixes
...
Merge remote-tracking branch 'tdaede/master'
2012-10-31 13:37:19 -07:00
68e6818761
STM32F4 New: headers and source code for DMA
...
This has been tested with some dac, timer and (as yet uncommitted) adc examples.
2012-10-31 13:36:07 -07:00
5a6b1a1934
STM32F1 Addition to DMA to add interrupt flag read and clear.
...
Additional macros and defines added to header for the purpose.
Also disable of increment modes (seem to have been omitted originally).
2012-10-31 13:35:59 -07:00
6b2ca89f48
Merge pull request #22 fixing usb and f4 alignment issues
...
Merge remote-tracking branch 'jdk/master'
Conflicts:
lib/stm32/f1/Makefile
lib/stm32/f4/Makefile
2012-10-31 13:34:18 -07:00
016c74af25
F2 timers are the same as F4 timers, so make the same changes to them
2012-10-31 15:34:10 -05:00
fd3ee124f8
Fixed find call for Darwin BSD find.
2012-10-31 13:29:38 -07:00
9468933735
Add back timer_set_option for STM32F4, and document.
2012-10-31 00:42:58 -05:00
a56764667c
Update comments and copy author from F4 version.
2012-10-31 00:11:59 -05:00
d7296f176a
Move timer.c to be common to all stm32 series
2012-10-30 16:52:54 -05:00
90cdddd96b
Allow make to build libs/examples in parallel.
...
You cannot issue make inside a for loop if you want to let it run in
parallel. Performance increases seen:
10:03 < zyp> I tested make all -j8 without your change, it takes 8.7s
10:03 < zyp> so on my cpu, your change gives >2x speedup
My own cpu gives more modest speed increases, of only about 20%.
2012-10-30 10:03:12 +00:00
0bd415365d
STM32F4: New adc module
...
Note adc.h was provided by Matthew Lai but was never merged.
adc.c is new.
2012-10-30 10:22:52 +10:30
a1f4c29872
Merge branch 'efm32'
...
this adds support for energy micro's efm32 tiny gecko, gecko, leopard gecko and
giant gecko series.
also, the experimental libopencmsis is included in this branch, as it is
required for full operation of the free energymicro standard library
2012-10-29 14:44:37 +01:00
a01e5c201b
Merge branch 'generalizations'
...
this merges common c and header files of different architectures, adds a
dispatch mechanism and yaml descriptions of interrupt handlers from which the
whole interrupt table setup c code is generated.
2012-10-29 14:34:42 +01:00
0fd7cc3559
Fix CAN standard ID mask to be 11 bits.
2012-10-28 02:56:22 -05:00
f385341495
Release the correct fifo in can_receive()
2012-10-28 02:53:53 -05:00
1bb0e9b970
Add CAN support for F2 and F4 series.
...
All stm32 series currently released share the same bxCAN hardware, so share can.c among all the families.
2012-10-28 02:50:46 -05:00
f594af86c8
fixes for generated header cleanup
...
* make target phony
* don't complain about missing files
2012-10-23 01:59:41 +02:00
efce73f815
make `make clean` work again
...
the irq2nvic_h generates files in a way make doesn't understand, so it
now also provides an option to clean up again
2012-10-23 01:38:27 +02:00
2ad04777bf
only generate irq headers on demand
...
as header file generation is not directly controlled by make (which, by
the way, makes the generatedheaders target phony), the script has to
take care of not needlessly generating files itself lest to have make
rebuild everything everytime
2012-10-23 01:31:30 +02:00
e55a9e3875
Missing EXTI ports and flag numbers for F4
...
(Also missing for forthcoming L1)
2012-10-22 22:37:44 +00:00
0a1cf977eb
Added irq.yaml to lpc_13xx and lpc_17xx families
2012-10-21 08:54:40 +02:00
1e88ad186f
remove common modules from efm32*'s explicit build list
...
this is 75c216582
2012-10-19 19:30:25 +02:00
a433217c5d
Merge branch 'generalizations' into efm32
2012-10-19 19:29:32 +02:00
f62eee7644
remove nvic.o from stm32/f[24]'s explicit list
...
this was missed out in 75c216582
2012-10-19 19:29:06 +02:00
14a5fc78cf
completing dispatch
...
the LM3S irq list was previously unused as it was missing in the
dispatch files; now it got added. (before HEAD^, it wouldn't have made
any difference because the discriminating constant wasn't defined
anyway)
also, this enhances the warning messages
2012-10-19 19:10:42 +02:00
99d7b21032
define platform specific constants for all chips
...
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.
2012-10-19 19:10:42 +02:00
75c2165827
build common .o files everywhere (fixes issue #29 )
...
vector.o, nvic.o, scb.o and assert.o are available on every platform,
but at least some of them differ between the implementations. they
already got built explicityly on some platforms; now adding them to the
common Makefile.include.
2012-10-19 18:56:39 +02:00
1210738845
add generic OBJS to efm32 libraries
...
better fixes would be adding them to the generic (included) Makefile or
building the really device independent stuff in a _generic library
2012-10-19 17:03:23 +02:00
cece71ba2f
install *all* efm32 linker scripts
...
this will install the per-family linker scripts twice (doesn't hurt,
gets overwritten), but more importantly intalls the per-chip scripts
too.
2012-10-19 16:13:34 +02:00
Piotr Esden-Tempski