LeafLabs Garden » Forum: The Lounge

LeafLabs Garden » Forum: The Lounge - Recent Topics http://forums.leaflabs.com/forum.php?id=9 A place to share, learn, and grow... en-US Fri, 22 Jan 2016 00:01:16 +0000 http://bbpress.org/?v=1.0.2 <![CDATA[Search]]> q http://forums.leaflabs.com/search.php gbulmer on "New 'Arduino-Nano' size STM32 Nucleo mbed's" http://forums.leaflabs.com/topic.php?id=74453#post-105784 Mon, 19 Oct 2015 19:22:44 +0000 gbulmer 105784@http://forums.leaflabs.com/ There are some interesting new ST-developed STM32 Nucleos, in a smaller, 'Arduino Nano' form-factor: <a href="http://www.st.com/web/catalog/tools/FM116/CL1620/SC959/SS1532/LN1847/PF262544?s_searchtype=partnumber" rel="nofollow">http://www.st.com/web/catalog/tools/FM116/CL1620/SC959/SS1532/LN1847/PF262544?s_searchtype=partnumber</a> There are three so far, including an STM32F303 They are available for £7.40 from UK Farnell: <a href="http://uk.farnell.com/stmicroelectronics/nucleo-f303k8/dev-board-stm32f303k8-nucleo-32/dp/2500224" rel="nofollow">http://uk.farnell.com/stmicroelectronics/nucleo-f303k8/dev-board-stm32f303k8-nucleo-32/dp/2500224</a> Also Mouser and Digikey are listing it. AFAICT it is an mbed, and so should support the 'flash drive upload' feature. rogerclark on "New forum for Arduino on STM32" http://forums.leaflabs.com/topic.php?id=74431#post-105723 Mon, 27 Apr 2015 17:28:02 +0000 rogerclark 105723@http://forums.leaflabs.com/ Guys, By popular demand I have created a new forum for anyone interested in STM32 on Arduino, especially the Maple and Maple mini, but also for generic STM32F103 boards Its early days, but the new forum is at <a href="http://www.stm32duino.com/" rel="nofollow">http://www.stm32duino.com/</a> So please take a look, or post if you have a question Cheers Roger Clark bigplik on "any OLED display working with maple?" http://forums.leaflabs.com/topic.php?id=74422#post-105692 Wed, 25 Mar 2015 07:53:16 +0000 bigplik 105692@http://forums.leaflabs.com/ hi, is any OLED display working nicely with mapleIDE or let to use arduino libraries like ssd1306 or u8glib? dbolgheroni on "Edit Post" http://forums.leaflabs.com/topic.php?id=74410#post-105624 Wed, 14 Jan 2015 17:30:48 +0000 dbolgheroni 105624@http://forums.leaflabs.com/ Is it possible to edit a post? Can't find an 'edit' link. Thank you. gruvin on "Added EEPROM user-mdoule and now my rules.mk is being ignored?" http://forums.leaflabs.com/topic.php?id=74276#post-105245 Wed, 19 Feb 2014 22:26:32 +0000 gruvin 105245@http://forums.leaflabs.com/ Hi I'm perplexed. :-/ I added a module "EEPROM" (files found from another forum contributor, based on ST AN2594) to my command line toolchain <code>~/libmaple/libraries/</code> folder. I had to create a <code>rules.mk</code> file. This is what I have for that ... <pre><code>~/libmaple/libraries/EEPROM/rules.mk: --- # Standard things sp := $(sp).x dirstack_$(sp) := $(d) d := $(dir) BUILDDIRS += $(BUILD_PATH)/$(d) # Local flags CFLAGS_$(d) := $(WIRISH_INCLUDES) $(LIBMAPLE_INCLUDES) CXXFLAGS_$(d) := $(WIRISH_INCLUDES) $(LIBMAPLE_INCLUDES) # Local rules and targets cSRCS_$(d) := flash_stm32.c cppSRCS_$(d) := EEPROM.cpp cFILES_$(d) := $(cSRCS_$(d):%=$(d)/%) cppFILES_$(d) := $(cppSRCS_$(d):%=$(d)/%) OBJS_$(d) := $(cFILES_$(d):%.c=$(BUILD_PATH)/%.o) \ $(cppFILES_$(d):%.cpp=$(BUILD_PATH)/%.o) DEPS_$(d) := $(OBJS_$(d):%.o=%.d) $(OBJS_$(d)): TGT_CFLAGS := $(CFLAGS_$(d)) $(OBJS_$(d)): TGT_CXXFLAGS := $(CXXFLAGS_$(d)) TGT_BIN += $(OBJS_$(d)) # Standard things -include $(DEPS_$(d)) d := $(dirstack_$(sp)) sp := $(basename $(sp))</code></pre> Now when I compile, it appears that my main project <code>rules.mk</code> file is being ignored -- or at least the <code>cppSRCS</code> part ... <pre><code>make ... [CXX] /Users/bryan/libmaple/libraries/FreeRTOS/MapleFreeRTOS.cpp [CC] /Users/bryan/libmaple/libraries/EEPROM/flash_stm32.c [CXX] /Users/bryan/libmaple/libraries/EEPROM/EEPROM.cpp [CXX] main.cpp [LD] maple_mini.elf</code></pre> Notice that there is no longer a <code>real-main.cpp</code>, as there normally is! Huh?! When I remove the new EEPROM library from <code>~/libmaple/Makefile</code> ... <pre><code># User modules ###USER_MODULES := $(SRCROOT)/libraries/EEPROM <--- commented out # User modules: ifneq ($(USER_MODULES),) LIBMAPLE_MODULES += $(USER_MODULES) endif</code></pre> ... my <code>real-main.cpp</code> gets compiled again and all is well -- except I don't have my new EEPROM library available, of course. - - - I've spent well over an hour trying to figure out what I've done wrong. I just can't see it for the looking. :-/ Any ideas? Thanks! P.S: When I aded this same new EEPROM module to the IDE, by placing it in <code>~/Documents/MapleIDE/libraries</code>, all worked as expected. That is, the Maple IDE compiler worked. But then, it doesn't use a <code>real-main.cpp</code> anyway. Hmmm. {shrug} bnewbold on "New independent libmaple fork/project: http://rambutan.cc" http://forums.leaflabs.com/topic.php?id=74366#post-105502 Fri, 05 Sep 2014 23:46:18 +0000 bnewbold 105502@http://forums.leaflabs.com/ Hello folks! I've started sketching out an independent fork/continuation of libmaple and supporting tools, which i'm calling "rambutan". The premise is to drop all-in-one IDE support and focus on the core C library for peripheral and family completeness within the STM32 M3/M4 line. Being hardware-neutral and making development on custom boards easy is also a priority. This isn't an official leaflabs project. I'm also now not a leaflabs employee, just a friend and fellow traveler. But it does have their blessing and may get some patches and support from individual developers in the future. The current plan is to do a "v0.1" of librambutan that would largely backwards compatible with libmaple (master branch), including the build system, to make it easier to accept outstanding libmaple ports and patches. "v1.0" development could include refactoring of symbols and filenames, breaking compatibility. The project website (<a href="http://rambutan.cc" rel="nofollow">http://rambutan.cc</a>) links to forums, code, and documentation. There are probably some rough edges. I will be in transit (moving to Seattle! Wo!) for the next two weeks, but will try to keep up with discussion. Tooling, policy, and documentation will probably take priority over development or merging code until early October. Comments, advice, cynicism, and patches are all welcomed! If this was to become the collective continuation of the leaflabs forum community as well i'd be really happy, though that may be presumptuous to propose. paulj on "newlib and libmaple" http://forums.leaflabs.com/topic.php?id=74363#post-105474 Tue, 29 Jul 2014 14:57:55 +0000 paulj 105474@http://forums.leaflabs.com/ I am running Gentoo Linux, and I have the arm-none-eabi toolchain already installed. gcc is at version 4.7.2, and newlib at version 2.0.0. When compiling libmaple, I get the following error: <pre><code>build/./wirish/syscalls.o: In function _exit': /home/paul/maple/libmaple/./wirish/syscalls.c:60: multiple definition of</code></pre> _sbrk' /usr/lib/gcc/arm-none-eabi/4.7.3/../../../../arm-none-eabi/lib/libc.a(lib_a-syscalls.o):/var/tmp/portage/cross-arm-none-eabi/newlib-2.0.0/work/newlib-2.0.0/newlib/libc/sys/arm/syscalls.c:482: first defined here /usr/lib/gcc/arm-none-eabi/4.7.3/../../../../arm-none-eabi/lib/libc.a(lib_a-syscalls.o): In function `_sbrk': syscalls.c:(.text+0x6a8): undefined reference to `end' collect2: error: ld returned 1 exit status make: *** [build/maple.elf] Error 1 ` This appears to be a clash between the newlib library and the libmaple library. So my question is this: what do I need to do to stop the compiler linking to the newlib code (assuming it is not necessary?). I compile following the instructions in the "Command Line Tools" section of the wiki. Alternatively, do I have to do something else, such as loading an earlier version of newlib, or compiling it with some options set when making the toolchain? Many thanks for your help! Rgds., Paul gbulmer on "Community design - New improved Orone-mini: DIY, USB, power and PCB" http://forums.leaflabs.com/topic.php?id=74227#post-105128 Sat, 04 Jan 2014 23:13:43 +0000 gbulmer 105128@http://forums.leaflabs.com/ Folks, I've released a new design, <a href="https://github.com/gbulmer/openstm32hw/tree/master/Orone-mini-v0/Orone-mini-S8J-v0r001">Orone-mini-S8J</a> at github. It is significantly improved over the earlier Orone-mini-S8H. Several people who built the earlier Orone-mini designs fed back on their experiences. All identified the difficulty of assembling the board using the small NUF2042XV6 USB EMI/ESD. So I have adopted the much larger STF202-22. It is the same part as used on Siy's designs. However, that circuit leaves the USB +5 power line (Vusb) unprotected. So Orone-mini-S8J has a separate discrete Transient Voltage Suppression (TVS) diode protecting the Vusb connection. Ventosus ran some <a href="http://forums.leaflabs.com/topic.php?id=10410&page=7#post-25373">tests to measure the ADC noise performance</a> of Orone-mini vs Maple. The test results indicated Orone-mini was worse than Maple analogue-power, but better than Maple digital power. As that generation of Orone-mini only connected the digital power supply to av+, and is only a two layer PCB, it was very encouraging. I've changed the design, so that Orone-mini-S8J now provides av+ from its analogue-power supply. Note: the analogue supply is only useful to provide external power from av+ when Orone-mini is connected to, about, 9V or less. Several Maple-mini's had suffered voltage regulator failure, which was the reason the earlier Orone-mini's had av+ connected to Orone-mini's higher capacity digital power supply. Just like Maple mini, the small analogue-PSU voltage regulator used for the analogue power supply is limited by its power dissipation. This should not effect Orone-mini when it's run from 5V, by powering it over USB The board has been extensively reworked to improve the integrity of the ground plane and power distribution (i.e. fewer big holes :-). Note:These changes necessitated changes to the pin-out of the extra signals and voltages that Orone-mini provides, but which are not available on Maple-mini. Hence Orone-mini-S8J is not pin identical to Orone-mini-S8H. However Orone-mini-S8J is pin identical to Maple-mini. The silk screen has been slightly updated to highlight the Serial Wire Debug (SWD) pins, making it easier to use with a JTAG-SWD debugger, for example an ST-Link/V2 or the even lower cost ST-Link/V2 on STM32FxDiscovery boards (under $10). These debuggers support hardware break-points, data watch points, single-stepping, etc. using gdb. Orone-mini is fully pin-compatible, and software compatible with Maple-mini if you assemble it with an STM32F103CB. Orone-mini can also be assembled using an STM32F303CC, doubling its Flash, and more than doubling SRAM, It could be used with Ventosus' initiated libmaple port for STM32F303. This would have similar performance to an Arduino Due for a similar cost to an eBay clone of an Arduino UNO. An STM32F303Cx has more, and better peripherals than an STM32F103CB. For example: - 4 ADCs vs 2 ADCs, and ADC's are 5msps vs 1Msps (potentially 20Msps vs 2Msps) - Dual DAC vs none - 3 SPI vs 2 - 2 advanced timers vs 1 (useful for e.g. brushless motor control) - 2 I2S vs none - 17 channel capacitive touch sensing vs none - 7 analogue comparators vs none - 4 programmable gain op amps vs none etc. AFAIK, an STM32F303 has some much better peripherals than an Arduino Due's SAM3X8E, for example better ADC and analogue signal processing. Further, STM2F303 has 5V tolerant pins, though Due's SAM3X8E has more I/O. Orone-mini-S8J retains Orone-mini's benefits, described at github. Summary of changes: - Larger, easier to assemble USB circuit, without sacrificing any ESD protection - av+ from analogue power supply (aim is less electrical noise, and better ADC) - reworked ground plane and crystal oscillator for better ground and lower noise This is an initial release; I have not yet updated the Bill of Materials. Please feedback any comments on this thread. Please do not use the design yet. It is not adequately quality assured. (Full disclosure: I am not a member of LeafLabs staff.) camguy on "STM32F429 EvoPrimer Features DMA2D Graphics Accelerator" http://forums.leaflabs.com/topic.php?id=74051#post-104768 Sat, 12 Oct 2013 07:46:13 +0000 camguy 104768@http://forums.leaflabs.com/ Hi All, Just FYI... there's a new STM32F429 EvoPrimer development board available now. The demo application features what ST call the Chrom-ART Graphics Accelerator a DMA2D implementation for accelerated graphics handling. It facilitates bitmap encoding/decoding, management of multiple objects on the display, layering of objects and blending to include management of transparency. There's a video of the demo on YouTube: <a href>http://www.youtube.com/watch?v=TJWeVvicvZk&hd=1</a> And the sample apps are available at stm32circle.com (Direct link: <a href>http://www.stm32circle.com/resources/stm3242Iprimer.php</a>) anthrolume on "HOWTO: drive WS2812 controller-on-LED strips" http://forums.leaflabs.com/topic.php?id=74191#post-105034 Sat, 07 Dec 2013 21:22:26 +0000 anthrolume 105034@http://forums.leaflabs.com/ Hi gang, I just did some work getting a Maple Mini to drive a strip of 120 WS28128-based LEDs and I thought I'd share the core code because it was pretty tedious. I looked at some code provided by Adafruit to come up with this implementation, including borrowing an inline-assembly delay function from them. The delay values were determined empirically using time cursors on a storage oscilloscope. This code uses three loops because I keep my frame buffers in RGB byte order, and the WS2812 wants data in GRB order. This method is part of one of my classes. The only external thing you'll need is a constant called kNumLEDs that says how many RGB-tuples are pointed to by the rgb argument (i.e. rgb is a pointer to an array of kNumLEDs * 3 bytes). Hope this helps someone out. -Bryan <a href="http://anthrolume.com" rel="nofollow">http://anthrolume.com</a> <pre><code>// These values were determined empirically #define kSpinCountZeroBitHigh 4 // these give 330ns high and 800ns low #define kSpinCountZeroBitLow 9 #define kSpinCountOneBitHigh 9 // these give 750 high and 360 low #define kSpinCountOneBitLow 4 // WS2812 chip - NeoPixel, etc. void Strand::_present(byte* rgb /* = NULL */) { if (rgb == NULL) rgb = getWriteBuffer(); // Be sure we leave 50+ microseconds for data latch while((micros() - _endTime) < 50L); volatile uint32 *setResetRegister = &(PIN_MAP[_pin].gpio_device->regs->BSRR); uint32 highValue = (1 << PIN_MAP[_pin].gpio_bit); uint32 lowValue = (1 << PIN_MAP[_pin].gpio_bit + 16); byte *end = rgb + 3 * kNumLEDs; byte r, g, b, mask; // Turn off interrupts until we're done with our loop noInterrupts(); while (rgb < end) { byte r = *rgb++; byte g = *rgb++; byte b = *rgb++; // Our frame buffers are RGB, but WS2812s want colors in GRB order, so we // do that reording here. // Emit green for (mask = 0x80; mask; mask >>= 1) { *setResetRegister = highValue; if (g & mask) { spinDelay(kSpinCountOneBitHigh); *setResetRegister = lowValue; spinDelay(kSpinCountOneBitLow); } else { spinDelay(kSpinCountZeroBitHigh); *setResetRegister = lowValue; spinDelay(kSpinCountZeroBitLow); } } // Emit red for (mask = 0x80; mask; mask >>= 1) { *setResetRegister = highValue; if (r & mask) { spinDelay(kSpinCountOneBitHigh); *setResetRegister = lowValue; spinDelay(kSpinCountOneBitLow); } else { spinDelay(kSpinCountZeroBitHigh); *setResetRegister = lowValue; spinDelay(kSpinCountZeroBitLow); } } // Emit blue for (mask = 0x80; mask; mask >>= 1) { *setResetRegister = highValue; if (b & mask) { spinDelay(kSpinCountOneBitHigh); *setResetRegister = lowValue; spinDelay(kSpinCountOneBitLow); } else { spinDelay(kSpinCountZeroBitHigh); *setResetRegister = lowValue; spinDelay(kSpinCountZeroBitLow); } } } interrupts(); _endTime = micros(); }</code></pre> breadbread1984 on "cant run freertos demo" http://forums.leaflabs.com/topic.php?id=74266#post-105199 Mon, 27 Jan 2014 07:25:09 +0000 breadbread1984 105199@http://forums.leaflabs.com/ I try to run a freertos demo on maple. After I upload the firmware, the /dev/ttyACM0 disappear as soon as the maple completes booting. I think there should be some problem with the firmware. There is the source of the firmware. Could someone find the problem of the source for me. thanks in advance. <pre><code>#include "wirish.h" #include "FreeRTOS.h" #include "task.h" #define PWM_PIN 2 void setup() { /* Set up the LED to blink */ pinMode(BOARD_LED_PIN, OUTPUT); /* Turn on PWM on pin PWM_PIN */ pinMode(PWM_PIN, PWM); pwmWrite(PWM_PIN, 0x8000); /* Send a message out USART2 */ Serial2.begin(9600); } /* Multi-task example */ void vTaskCodePing( void * pvParameters ) { while(1) { SerialUSB.print("Ping\n"); delay(900); } } void vTaskCodePang( void * pvParameters ) { while(1) { SerialUSB.print("PANG\n"); delay(900); } } #define STACK_SIZE 200 /* Please Do Not Remove & Edit Following Code */ int main(void) { setup(); /* Create two tasks Ping and Pang */ xTaskCreate(vTaskCodePing, (const signed char*)"Ping", STACK_SIZE, NULL, 1, NULL); xTaskCreate(vTaskCodePang, (const signed char*)"Pang", STACK_SIZE, NULL, 1, NULL); vTaskStartScheduler(); return 0; }</code></pre> cbrunschen on "ARMstrap - "ARMing the Next-Generation Engineer"" http://forums.leaflabs.com/topic.php?id=74053#post-104780 Mon, 14 Oct 2013 12:22:01 +0000 cbrunschen 104780@http://forums.leaflabs.com/ I came across this: <a href="http://armstrap.org/" rel="nofollow">http://armstrap.org/</a> <a href="http://www.adafruit.com/blog/2013/10/14/armstrap-arming-the-next-generation-engineer/" rel="nofollow">http://www.adafruit.com/blog/2013/10/14/armstrap-arming-the-next-generation-engineer/</a> // Christian gbulmer on "Smallest USB serial code?" http://forums.leaflabs.com/topic.php?id=74185#post-105018 Fri, 06 Dec 2013 17:42:14 +0000 gbulmer 105018@http://forums.leaflabs.com/ I want to make a little gizmo, with no more than a couple of KBytes of specific code. I'd like it to be a USB serial device. I'd like to squish everything into the lowest-cost 16 kB Flash STM32F103. Has anyone experience with, or done any experiments to find the smallest practical size for USB serial device code on an STM32F103? Anonymous on "Where can I get my hands on a Maple Native?" http://forums.leaflabs.com/topic.php?id=74067#post- Thu, 21 Jan 2016 19:01:16 +0000 Anonymous @http://forums.leaflabs.com/ BlackBelt2025 on "Where can I get my hands on a Maple Native?" http://forums.leaflabs.com/topic.php?id=74067#post-104796 Sat, 19 Oct 2013 21:35:18 +0000 BlackBelt2025 104796@http://forums.leaflabs.com/ I am working on a research project where we are using a Maple-powered embedded system. However, we have an issue where the Maple does not have DAC. I did notice that the Maple Native does have its own DAC, which would be very valuable to this project? Is there any way I can get my hands on a Maple Native?