b7e5005679
[native] Check for hardware version by pull-up and -down. Print HW Version in mon command.
...
Until now the native hardware was pulling PB5-7 down and checking if
they were asserted high. BMPMV2b is pulling the pins down instead of
high. The hardware version routine now determines the hardware version
based on the fact if a pin is asserted at all. This means that if a pin
is left floating, the version number bit will be 0, and if the pin is
asserted either high or low the bit will be set to 1. While we were
already at it the "monitor version" command in GDB will now also print
the hardware version number.
2016-02-14 22:49:29 -08:00
9c846d5029
Add nRF52 support.
2016-02-10 20:23:43 -07:00
d5e263122f
Increase JTAG_MAX_DEVS to 32.
...
Fixes #112
2016-02-02 10:04:57 +13:00
7d957cbfdf
Update nRF51 IDs, add nRF51422 and xxAB/xxAC devices.
...
There were a lot of nRF51 device ids missing. I wrote a script to pull nRF51 IDs from the openocd repository, which seems to be more up-to-date. The script's output goes in nrf51.c.
2015-12-17 17:50:41 -07:00
c9d3cf71dd
gdb_main.c: fix buffer overflow on large reads
...
When gdb issues a `m xx,200` command, the probe should respond with a
packet of size 2*0x200=1024 which is the size of the packet buffer.
However, the `hexify()` procedures writes 1025 bytes in the buffer.
During my tests, it caused the probe to hang when issuing a `dump`
command. Presumably by overwritting the `cur_target` variable.
2015-12-14 11:28:28 +01:00
2d7b24c566
efm32: inhibit srst
...
EFM32s don't seem to like to be completely resetted while being debugged
(load and break were broken)
2015-12-14 11:28:28 +01:00
1f62fa4909
CRC32: improved computation time
...
Reading one byte at a time was so slow that it caused the gdb client to
timeout.
2015-12-14 11:28:28 +01:00
61c6c767b0
Merge pull request #107 from richardeoin/efm32
...
[efm32] Use SW-DP IDCODE register for identification
2015-12-13 14:59:49 +13:00
544e1080a3
[efm32] Use SW-DP IDCODE register for identification
...
Instead of the Device ID register which isn't constant for EFM32 parts
2015-12-01 21:41:08 +00:00
e6d388d057
Merge pull request #106 from richardeoin/efm32
...
Support for Silicon Labs EFM32/EZR32
2015-11-17 07:59:35 -08:00
fa7cccff23
[efm32] Formatting fixes
2015-11-14 14:56:03 +00:00
b678b83062
[efm32] oops, fixed ezr32lg probe message
2015-11-14 14:42:09 +00:00
fe3874bc82
[efm32] add extra details to probe message
2015-11-14 14:41:48 +00:00
091adda0b4
Merge branch 'umetronics-master'
2015-11-11 12:41:57 -08:00
600f0d27d9
Use sniprintf on newlib platforms.
2015-11-11 12:41:13 -08:00
56b10a1255
Merge branch 'f205' of https://github.com/schodet/blackmagic into schodet-f205
...
Conflicts:
src/stm32f4.c
2015-11-11 12:27:27 -08:00
07f7571dd1
Add basic support for EFM32/EZR32 chips
...
Tested with EZR32LG230
2015-10-23 19:58:44 +01:00
3f9ea8250a
Defined snprintf to sniprintf on launchpad-icdi
2015-10-22 19:33:45 +03:00
2365bbe010
TI TM4C/LM4F cdc-acm fixes
2015-09-30 15:48:01 +03:00
7e5a664df9
Merge pull request #98 from UweBonnes/stm32l4
...
Add STM32L4 handling. No option write handling yet.
2015-09-01 11:01:36 +12:00
6ec35bc9ff
Merge pull request #97 from UweBonnes/stm32f7
...
Stm32f7
2015-09-01 10:58:57 +12:00
3aa333ee1c
Merge pull request #95 from bvanheu/feature/nrf51_new_idcode
...
Add 0x0072 as a valid & supported idcode for NRF51
2015-09-01 10:30:55 +12:00
f7492d93be
Cortex-M: Detect and handle flash patch revision.
...
According to ARM v7-M Architecture Reference Manual
ARM DDI 0403E.b (ID120114)
2015-08-22 13:01:24 +02:00
51e2adc6d9
Add STM32F7.
2015-08-22 13:01:24 +02:00
2336a313e4
Add STM32L4 handling. No option write handling yet.
2015-08-22 12:59:56 +02:00
97cb75145d
Add support for STM32F2xx family
...
The F405 rev A uses a wrong CPUID, this patch use the core identifier to make
the distinction.
Tested with:
- F205
- F405 rev Z
- F429
Unfortunately, I do not have any F405 rev A.
2015-08-17 14:35:11 +02:00
ef574b72b1
Add support for STM32F446x family.
...
Tested with a custom hardware.
2015-07-31 09:05:27 -07:00
420d2c41c7
Add 0x0072 as a valid & supported idcode for NRF51
2015-07-31 00:00:00 -04:00
beacf9c85c
Refactor stm32f1_probe
2015-07-30 07:58:55 +01:00
101821ae31
[Bugfix] stm32f1_probe would always return true, breaking support for all other targets
...
The intention of `if (t->driver)` conditional was to test if any of the cases in the preceeding switch/case were met. However t->driver was previously set to point to a default value in cortexm.c:200 and therefore this would always evaluate to true. I've replaced the broken if statement with a duplicate of the switch/case run above.
It looks like this was introduced in 09544bc710#92 ) but 492d6c9cf8
2015-07-12 17:32:48 +01:00
a7f6753153
STM32f0: F07 and F09 have 2kiByte blocksize.
2015-06-22 16:44:19 +02:00
09544bc710
src/stm32f1.c: More verbose FO detection message.
2015-06-22 16:44:08 +02:00
4af5c03d75
lmi: Use registers for stub parameters.
2015-04-21 20:05:41 +12:00
ca17de6243
lpc11xx: Correctly detect LPC8xx devices.
2015-04-11 17:23:01 -07:00
9009ed6581
cortexm: Add target option to inhibit assersion of SRST.
2015-04-11 16:08:59 -07:00
24122aa318
lpc43xx: add chipid for LPC43S37.
2015-04-11 16:08:25 -07:00
09b781f1c1
target: Remove old flash interface.
2015-04-11 16:05:41 -07:00
3d8b34f180
lpc: split out common code and rewrite to use new interface.
2015-04-11 16:05:40 -07:00
cd5d569d38
lpc: Reduce differences between lpc11xx and lpc43xx code.
2015-04-11 16:05:40 -07:00
54eb3a719f
samd: use new flash interface.
2015-04-11 16:05:40 -07:00
d340053078
nrf51: use new flash interface.
2015-04-11 16:05:40 -07:00
1541f1c7a8
kinetis: use new flash interface.
2015-04-11 16:05:40 -07:00
9e09ae2e1e
lmi: use new flash interface.
2015-04-11 16:05:40 -07:00
622497f7e2
stm32f1: use new flash interface.
2015-04-11 16:05:40 -07:00
7cd3432994
stm32f4: Add second bank for F42x/F43x.
2015-04-11 16:05:40 -07:00
1efad1a03a
stm32f4: use new flash interface.
2015-04-11 16:05:40 -07:00
fc2f266a13
sam3x: Update to use new buffered flash interface.
2015-04-11 16:05:40 -07:00
21c209fd8f
sam3x: simplify probe function.
2015-04-11 16:05:40 -07:00
3ed4207e8a
stm32l0: Update to use new buffered flash writes. Remove old stubs.
2015-04-11 16:05:40 -07:00
45328ea124
Add buffering support for flash drivers.
...
Some devices can get a significant boost in performance by writing to
flash memories one page at a time. Generic function to do this are
provided at the target layer and may be used by flash drivers.
2015-04-11 16:05:40 -07:00
36f749fed9
Fix flash buffer alignment in target layer.
2015-04-03 21:21:30 -07:00
7202db5860
Add new functions to wrap flash driver erase/write/done operations.
2015-04-03 21:20:01 -07:00
691d21989a
Add function to add simple flash driver to target.
...
Clean up ram/flash/memory map on target destruction.
2015-04-03 21:18:16 -07:00
0fc635b3f8
Add functions for dynamically generating the XML memory map.
2015-04-03 21:18:16 -07:00
482070c91b
Merge branch 'v1.6-branch'
...
Conflicts:
src/Makefile
src/platforms/stlink/platform.h
src/platforms/swlink/platform.h
2015-04-03 21:02:01 -07:00
97824b7f02
Use git describe for version in generated header file.
2015-04-03 20:21:33 -07:00
865beaba98
Suppress warning about longjmp clobbering local variable.
2015-04-02 21:06:18 -07:00
bc89217e3f
sam3x: add SAM3U support
2015-04-01 00:04:52 +02:00
21d464293c
gdb: don't access target fields directly.
2015-03-29 18:06:13 -07:00
c2cde32716
Improve cortexm_read/write_regs. Use more sensible return types.
2015-03-29 17:37:03 -07:00
274b818517
Make cortexm_pc_write static.
2015-03-29 16:48:51 -07:00
779dcf0f69
stm32l0: use cortexm_run_stub.
2015-03-29 16:47:55 -07:00
012922ed00
nrf51: use cortexm_run_stub.
2015-03-29 16:31:38 -07:00
810269dce1
nrf51: Use generated stub.
2015-03-29 16:28:18 -07:00
cac286d398
lmi: Use generated stub with cortexm_run_stub.
2015-03-29 16:20:57 -07:00
2215e33bcf
lmi: Clean up magic numbers, remove adiv5 links, and doc ref.
2015-03-29 15:58:18 -07:00
31fae032f7
cortexm: Separate stub loading from running.
2015-03-29 15:20:00 -07:00
9f271d5cd7
Consistently use 'target *t' for target var.
2015-03-28 20:47:17 -07:00
1e54139f4a
target: Remove pc_read/pc_write methods.
2015-03-28 20:15:33 -07:00
4313b64bbe
Add a flash_done target method.
2015-03-26 20:47:57 -07:00
f74def3552
Remove old STM32L1 driver.
2015-03-26 20:47:57 -07:00
d9af3cca6c
Remove unfinished ARM7TDMI driver.
2015-03-26 20:47:57 -07:00
64c79e1d03
Consistently name target var 't' in cortexm.
2015-03-26 20:47:57 -07:00
bda4279028
Send error reply and abandon target list on unhandled timeout.
2015-03-26 20:42:47 -07:00
68f54a3545
Remove old platform specific error handling macros.
2015-03-22 14:06:56 -07:00
5ab8564ff6
Clean up handling of lost targets using new exceptions mechanism.
2015-03-22 14:05:12 -07:00
588bad34ba
Build with -Os for swlink.
2015-03-22 12:26:45 -07:00
9a8dbdeff7
Fix errors when building for non-native platforms.
2015-03-22 12:26:45 -07:00
83b83ca48f
Use controlled timeout on SW/JTAG DP transactions.
2015-03-22 12:26:45 -07:00
d0a03f55a6
Handle timeout exceptions during scans and report to the user.
2015-03-22 12:26:45 -07:00
d6225eec76
Raise timeout exception when target is in WFI.
...
Ignore the exception when polling for halt, and report the exception
to the user if halting the target fails.
Remove old allow_timeout flag in DP struct that's no longer needed.
2015-03-22 12:26:45 -07:00
fa046601a5
Add exception handling mechanism and raise exception on low-level comms failure.
2015-03-22 12:26:45 -07:00
a497127a8d
Move MIN macro to general.h
2015-03-22 12:23:03 -07:00
acfd258aa7
Rewrite STM32F4 stub in C and use new interface.
2015-03-19 21:49:09 -07:00
8ddb186b35
Allow stub to return an error code.
2015-03-19 21:49:09 -07:00
437aedda11
Rewrote stm32f1 stub in C and call with parameters in registers.
2015-03-19 07:48:57 -07:00
c2462a6788
Add cortexm generic stub call, and use in stm32f1 driver.
2015-03-19 07:48:47 -07:00
e380ced517
Change STM32F1 driver to use generated stub.
2015-03-19 07:43:36 -07:00
1de685198c
Flatten samd_probe.
2015-03-17 21:58:36 -07:00
5c337c9aa7
Whitespace.
2015-03-17 21:58:36 -07:00
1366c32f89
samd: Simplified samd_flash_write further by always using manual writes.
...
Tested on all three devices as listed at the top of samd.c
2015-03-17 21:58:36 -07:00
0bf7778759
samd: Remove low level ADIv5 calls an favour of target_mem_write.
2015-03-17 21:58:36 -07:00
018d9cce80
Check vasprintf return value.
2015-03-17 21:19:02 -07:00
f5f87bf70f
Build for stlink with -Os.
2015-03-15 20:39:19 -07:00
c3f798438a
Remove abstraction of adiv5_dp_write. Both implementations were identical.
2015-03-15 17:12:31 -07:00
2bf54f9a72
Replace adiv5_ap_mem* functions with inline wrappers to target mem*.
2015-03-15 16:02:09 -07:00
ee3af96a73
Consolidate target_mem_read* and target_mem_write* methods.
...
There are now only mem_read and mem_write, that must
handle all alignments. These methods return void, errors must be
checked with target_check_error.
2015-03-15 14:31:57 -07:00
2e785e56fa
adiv5: Encode APnDP into register definition.
...
Clean up magic numbers in adiv5 calls.
Removed old adiv5_dp_write_ap and adiv5_dp_read_ap.
2015-03-14 18:03:04 -07:00
6f5b1873d7
Use size_t for sizes in flash functions. Fix all sign compare warnings.
2015-03-14 15:18:22 -07:00
f5c856af3b
Fix pointer sign warnings and remove -Wno-pointer-sign.
2015-03-13 20:35:39 -07:00
f52a51403f
stm32l0: cleanup includes.
2015-03-11 20:47:51 -07:00
1196b03d66
stm32l0: Fix pointer/integer casting warnings.
2015-03-11 20:47:51 -07:00
3d28c1418c
Cleaned up printf/scanf for C99 integer types.
2015-03-11 20:47:42 -07:00
4baa8aba4f
cdcacm: Fix DCD for second ACM interface.
2015-03-11 20:13:38 -07:00
7c1c19881f
Reassert DCD after DTR is written. Fixes #41 .
2015-03-11 05:45:43 -07:00
bf1cb71eb7
Revisions on Gareth's comments.
...
o Implemented byte writes to EEPROM now that the emulator has a
byte-wide target write.
o Added comment describing the reason that mass erase doesn't work.
o Removed all unused code.
o Changed to Linux kernel indent style.
o Changed to Linux kernel function to parenthesis style.
o Stub generation doesn't use Perl, switched to sed. Also, only
including the instructions instead of the source and the instructions.
o Handling unaligned destination writes.
2015-03-08 16:25:22 -07:00
e0a8ce5a88
STM32L0x target support.
...
Target support for stm32l0's and stm32l1's including option bytes and
data EEPROM. This module will superceed the previous stm32l1 driver.
o Program flash write and erase.
o Options modification and interpretive status.
o Stubs for program flash writes and erases. Stubs are modestly
faster than non-stub version. The stm32l0 will not execute stubs
when the MCU has crashed. A monitor option may be used to force
non-stub flash writes.
o Stubs generated from C++ code and converted to arrays of half-words.
o Writes to data EEPROM supoprted when loading segments.
o EEPROM data monitor command to write words.
o Stubs supported on stm32l1.
2015-03-08 16:17:34 -07:00
0d99e22a26
Quieten build system.
2015-03-07 11:03:50 -08:00
1eac78b1c8
Move targets to make bin and hex files to toplevel makefile.
2015-03-07 11:03:50 -08:00
c8f469c868
SAM4S: Erase flash in 8K instead of 16K chunks.
2015-03-05 10:46:22 -05:00
49390fee2a
Fix Windows DFU instructions in the Hydrabus readme.
2015-03-04 20:22:19 -08:00
9f0c29d329
Add support for HydraBus (tested with SWD with STM32F405 and SWD+JTAG with LPC4330 M0/M4)
2015-03-04 20:22:00 -08:00
c731c6ece3
Make F4 conditionals check STM32F4, not F4DISCOVERY.
2015-03-03 21:39:29 -08:00
2e37db2270
Error on any failure on 'make all_platforms'
2015-03-02 23:14:43 -08:00
87b8f330f0
Add a makefile target to build all platforms.
2015-03-02 22:20:25 -08:00
3e466f2d23
Factor out timing routines common to all STM32 targets.
2015-03-02 21:59:04 -08:00
b07ffffcee
Factor out common cdcacm.c from stm32 and tm4c.
...
Cleaned up interface to enter bootloader and read serial number.
2015-03-02 10:16:33 -08:00
4d4813de87
Clean up includes everywhere.
...
All source files include general.h first and before anything else.
This inlcludes platform.h and platform_support.h
No header file needs to include to include any of these, but should include
any others needed for it's own declarations.
2015-03-01 22:16:04 -08:00
9d80641261
Move common platform include to a common directory.
2015-03-01 20:15:31 -08:00
476f83b69a
Move common USB stuff out of platform header files.
2015-02-28 22:48:26 -08:00
5eff0ab5d6
Factor out common GPIO code in STM32 platforms.
2015-02-28 22:09:50 -08:00
48fb7ec662
Add missing includes to fix warnings.
2015-02-28 22:06:10 -08:00
75001a4421
Factor Morse code functions out of platform code.
2015-02-28 22:05:10 -08:00
5d136398e1
Use sensible return convention for platform_target_get_power()
2015-02-28 21:23:24 -08:00
1b2cd547bc
Merge pull request #72 from dlaw/master
...
Add support for SAM4 erase command
2015-01-26 10:59:32 +13:00
1552218e8f
Merge pull request #71 from maknoll/master
...
SAM3S support
2015-01-26 10:56:52 +13:00
b771d0dcd6
Add support for SAM4 erase command
2015-01-23 16:36:12 -05:00
a7283196d1
SAM3S support
2015-01-22 16:09:01 +01:00
ea78e2d737
Complete refactor of samd20 -> samd
2015-01-18 22:40:02 +00:00
565795e168
Added support for SAMD10/11/21, and list of tested SAM D devices
...
This allows blackmagic to be used with the new Arduino Zero board via
the unpopulated SWD header on the east end of the board.
2015-01-18 22:35:59 +00:00
3726061773
Support for the NVMCTRL Security Bit (PROT=1)
2015-01-18 20:46:26 +00:00
e75ea64a01
Renamed nvmc control a register to match datasheet
2015-01-18 14:11:58 +00:00
1c1312b467
Added better feedback for samd erase failures
2015-01-18 13:22:41 +00:00
a3ab9b24d1
Avoid repetition of cortexm code in stmd20 driver.
2015-01-18 12:16:17 +13:00
f1981925b3
Merge branch 'samd20-revb-fix' of https://github.com/richardeoin/blackmagic into richardeoin-samd20-revb-fix
2015-01-18 11:03:10 +13:00
cad30cfa39
Fixed run command on SAM D20 old revision B silicon.
...
This appears to be related to Atmel Errata Reference 12015 (see §
35.4.1 DSU in the datasheet) but only applies to revision B silicon -
not revision C. While this is an old silicion revision (latest is
revision E as of 17 Jan 2015) I found one on a development board.
2015-01-17 18:58:00 +00:00
9072faf6ca
Fixed spaces / tabs from last commit
2015-01-17 18:35:36 +00:00
c12002b636
Check target_attach succeeded in vRun command, and return error if not
2015-01-17 18:30:58 +00:00
abcc04d4f3
Merge pull request #63 from beewoolie/static-stubs
...
Fix to move stubs into rodata.
2015-01-17 13:18:20 +13:00
985627af7e
SAM4S support
2015-01-12 14:53:58 -05:00
da28970c3e
Fix to move stubs into rodata.
...
o Stubs were declared as globals which means they will be initialized
data. No need for this. Now, stubs are static const and are only
present in the program image.
2015-01-09 18:18:10 -08:00
9c5ffd61f7
First cut at Freescale Kinetis support.
2015-01-06 22:26:00 +13:00
d162e75ceb
stm32f4.c: Add STM32F411 ID.
2014-12-23 21:32:21 +13:00
fdd7799304
Merge pull request #60 from themadinventor/lp-icdi-clean
...
Added (preliminary) support for TI TM4C123 Launchpad
2014-12-23 21:31:43 +13:00
346258934b
Added support for TI TM4C123 Launchpad
2014-11-23 22:59:16 +01:00
9434299613
Add support for halfword memory accesses.
2014-10-21 00:12:50 +02:00
fd9eef821e
Revert USB double buffered OUT handling on STM32F1.
...
This introduced a bug where the endpoint can get stuck, forever sending NAK.
2014-10-16 13:50:56 -07:00
f85a8c4ebb
Fix build for f4discovery by defining platform_hwversion().
2014-10-16 13:50:56 -07:00
a5549d630c
Whitespace cleanup.
2014-10-16 13:50:56 -07:00
60626e5bbe
Merge pull request #57 from aibara/dev/lpc43xx_wdt
...
LPC43XX: Attempt to handle case of running WDT.
2014-09-17 10:14:21 +12:00
Piotr Esden-Tempski
Mark Rages
Gareth McMullin
Mark Rages
Laurent Bonnans
Clement Burin des Roziers
Richard Meadows
Antti Nykänen
Uwe Bonnes
Nicolas Schodet
Benjamin Vanheuverzwijn
Florian Larysch
Marc Singer
David Lawrence
Benjamin Vernoux
Martin Knoll
Fredrik Ahlberg
Vegard Storheil Eriksen