Activity for kwikius
-
kwikius committed [8461d3]
Add clean,use canonical form for makefile filenames
-
in testing for device id, make sure tmp file is removed in all cases
-
According to the ram rom test app, the start of ram for id 0x413, "STM32F40xxx/41xxx" should be 0x20002000 not 0x20003000. in dev_table.c. Tested on stm32F405RGT6 Can test with command stm32flash -r /dev/null -S 0x20002000:4 /dev/ttyUSB0 succeeds on my part, while stm32flash -r /dev/null -S 0x20001ffc:4 /dev/ttyUSB0 fails suggesting that 0x20002000 is start of user ram in bootloader
-
App for verifying the flash and ram ranges
-
add cmdline args
-
add submodules and makefile
-
add useage
-
initial commit
-
Determined by stm32flash with read around the various extents so rom/ram just before start, start, just before end, end (Test app at the end --->) Debug file output: :~$stm32flash -r testcode.bin -S 0x8000000:4 /dev/ttyUSB0 stm32flash 0.6 http://stm32flash.sourceforge.net/ Interface serial_posix: 57600 8E1 Version : 0x31 Option 1 : 0x00 Option 2 : 0x00 Device ID : 0x0464 (STM32L412xx/422xx) - RAM : Up to 32KiB (12544b reserved by bootloader) - Flash : Up to 128KiB (size first sector: 1x2048) - Option...
-
Also tested ROM lowest rom address is 0x08'000'000 highest rom word address is 0x08'01F'FFC
-
Confirmed that the v3 Patch above is working for stmL412. Have also investigated sram ranges Lowest sram address that can be read/written by stm32flash is 0x20'003'100 highest sram address word that can be read/written by stm32flash is 0x20'007'FFC Based on that I am guessing the dev_table.c line for the L412/422 should be {0x464, "STM32L412xx/422xx" , 0x20003100, 0x20008000, 0x08000000, 0x08020000, 1, p_2k , 0x1FFF7800, 0x1FFF780F, 0x1FFF0000, 0x1FFF7000, 0}, Also tested on STM32F051 to confirm...
-
Sorry for late reply. Hope to get this done in next couple of days.
-
EDIT Ignore this . Prefer the v3 patch above.
-
EDIT Ignore this . Prefer the v3 patch above. I modified my code from github and pushed it to sourceforge repo kwikius-stm32flash on branch l412 see here https://sourceforge.net/p/kwikius-stm32flash/code/ci/l412_andy (Note: I havent used sourceforge for a long time, so not sure how it is working exactly. ) Anyway the above branch works and doesn't appear to me to take noticeably longer, since there is only one timeout at init. I should of course remove the message about finding stm32L412 etc
-
EDIT Ignore this . Prefer the v3 patch above. I modified my code from github and pushed it to sourceforge repo kwikius-stm32flash on branch l412 see here https://sourceforge.net/p/kwikius-stm32flash/code/ci/l412_andy (Note: I havent used sourceforge for a long time, so not sure how it is working exactly. ) Anyway the above branch works and doesn't appear to me to take noticeably longer, since there is only one timeout at init. I should of course remove the message about finding stm32L412 etc
-
Hi again, Tested the v3 tweak patch and it looks to be working OK :) I will try a few more tests shortly
-
initial commit
-
initial commit of README
-
Create LICENSE.md
-
Merge branch 'master' of https://github.com/kwikius/quan-trunk
-
add news re code at giyhub
-
add example to index for quan/stm32f4
-
clarified comments
-
Update README
-
remove temporary backup files
-
remove redundant macros
-
be explicit about uint32_t in template specialisation arg
-
conversion constructor takes different pi dependednt on ValueType rather than always double
-
added utility interrupt functions
-
for stm32f4 set QUAN_FLOAT_TYPE to float not double to get hardware floating point benefits
-
added comment re specialisations for zero size buffers
-
disable default copy ctor and assign function
-
cobs protocol - initial commit
-
uav rc_pulse - initial commit
-
uav serail_port_sim - initial commit
-
correct update which timers have rcr reg
-
change to conditional typedefs for timer rcr reg
-
change dumber of channels from 10 to 8. Maybe make this a templte param?
-
formatting changes
-
added missing #include <cstdint>
-
add missing include dependecy
-
initial commit
-
update fraction_of_revolution to not use assign for constexpr
-
try to sort issue with arithmetic_convert, but eventually added an overlad in arithmetic_convert.hpp
-
add overload for mathematic_angle to arithmetic. Now rads to arith works nice
-
formatting only
-
removed stuff which asked for an object. Want to derive from this without needing to have actual size objects
-
formatting only
-
added std::init list overload to pushback
-
mod so default color_ptr uses static color not null
-
removed commented out windows mfc stuff, added ll_draw_line func
-
moved stuff here out of headers
-
added and updated wxwidgets draw fns
-
lots of mods and new stuff in gx/wxwidgets
-
lots of mods n new stuff in qx/wxwidgets
-
change to use of static colours rather than craeting on fly
-
mods to serial ports to give static and dynamic versions
-
added template params for LengthType and AngleType. Should probabably rename this as gps_position
-
create a graphics_info_context, which could be not restricted to wxwidgets since it has no wxwidgets specific dependencies. clean up and rename the scaling and conversion functions
-
create a graphics_info_context, which could be not restricted to wxwidgets since it has no wxwidgets specific dependencies. clean up and rename the scaling and conversion functions
-
added abs fn to 3dvect
-
added abs fn to 2dvect
-
added peek fn to fifo
-
add return value
-
reverted some chngs to in64 for the low level calcs. needs a review
-
changed QUAN_INT32 to int64_t
-
use arithmetic_convet not explicit ctor
-
change to C++11 style ctor braces
-
change to c++11 style constructor braces
-
add licence info
-
initial commit of gpio/module_disable
-
add license info
-
gpio_pin initial commit
-
gpio - separate example from index
-
gpio - add gpio::Pin
-
separate example from index page
-
formatting only
-
filled in a bit more of perip_reg archetype
-
clarify and fix typos in the periph_reg docs
-
modifications and additions to compile for avr target
-
sorte compilation for avr
-
more efficient rotation
-
correct comment re cdeg
-
added gnu licence info
-
change include guard to refelect new directory
-
add license info
-
change include guard
-
mod include guard and add license
-
add licence info
-
redirect to archetypes version
-
update licence info
-
update licence info
-
update licence info
-
update licence info
-
update licence info
-
added char get() and init() functions
-
use quan::fifo instead of local detail fifo
-
add error if not on linux - todo write windows version
-
add cgheck on valid number of user bytes
-
use meat::integer_max metafunction instead of C constant INT32_MAX sice not supported on all targets
1 >
