README ====== This README discusses issues unique to NuttX configurations for the STMicro STM32VLDiscovery (Value Line Discovery) board. Contents ======== - Development Environment - GNU Toolchain Options - IDEs - NuttX EABI "buildroot" Toolchain - NuttX OABI "buildroot" Toolchain - NXFLAT Toolchain - LEDs - UARTs - "STMicro STM32F100RC generic" specific Configuration Options - Configurations Development Environment ======================= Either Linux or Cygwin on Windows can be used for the development environment. The source has been built only using the GNU toolchain (see below). Other toolchains will likely cause problems. GNU Toolchain Options ===================== Toolchain Configurations ------------------------ The NuttX make system has been modified to support the following different toolchain options. 1. The CodeSourcery GNU toolchain, 2. The Atollic Toolchain, 3. The devkitARM GNU toolchain, 4. Raisonance GNU toolchain, or 5. The NuttX buildroot Toolchain (see below). All testing has been conducted using the CodeSourcery toolchain for Windows. To use the Atollic, devkitARM, Raisonance GNU, or NuttX buildroot toolchain, you simply need to add one of the following configuration options to your .config (or defconfig) file: CONFIG_ARMV7M_TOOLCHAIN_CODESOURCERYW=y : CodeSourcery under Windows CONFIG_ARMV7M_TOOLCHAIN_CODESOURCERYL=y : CodeSourcery under Linux CONFIG_ARMV7M_TOOLCHAIN_ATOLLIC=y : The Atollic toolchain under Windows CONFIG_ARMV7M_TOOLCHAIN_DEVKITARM=y : devkitARM under Windows CONFIG_ARMV7M_TOOLCHAIN_RAISONANCE=y : Raisonance RIDE7 under Windows CONFIG_ARMV7M_TOOLCHAIN_BUILDROOT=y : NuttX buildroot under Linux or Cygwin (default) If you change the default toolchain, then you may also have to modify the PATH in the setenv.sh file if your make cannot find the tools. NOTE: the CodeSourcery (for Windows), Atollic, devkitARM, and Raisonance toolchains are Windows native toolchains. The CodeSourcey (for Linux) and NuttX buildroot toolchains are Cygwin and/or Linux native toolchains. There are several limitations to using a Windows based toolchain in a Cygwin environment. The three biggest are: 1. The Windows toolchain cannot follow Cygwin paths. Path conversions are performed automatically in the Cygwin makefiles using the 'cygpath' utility but you might easily find some new path problems. If so, check out 'cygpath -w' 2. Windows toolchains cannot follow Cygwin symbolic links. Many symbolic links are used in Nuttx (e.g., include/arch). The make system works around these problems for the Windows tools by copying directories instead of linking them. But this can also cause some confusion for you: For example, you may edit a file in a "linked" directory and find that your changes had no effect. That is because you are building the copy of the file in the "fake" symbolic directory. If you use a Windows toolchain, you should get in the habit of making like this: make clean_context all An alias in your .bashrc file might make that less painful. 3. Dependencies are not made when using Windows versions of the GCC. This is because the dependencies are generated using Windows pathes which do not work with the Cygwin make. MKDEP = $(TOPDIR)/tools/mknulldeps.sh The CodeSourcery Toolchain (2009q1) ----------------------------------- The CodeSourcery toolchain (2009q1) does not work with default optimization level of -Os (See Make.defs). It will work with -O0, -O1, or -O2, but not with -Os. The Atollic "Pro" and "Lite" Toolchain -------------------------------------- One problem that I had with the Atollic toolchains is that the provide a gcc.exe and g++.exe in the same bin/ file as their ARM binaries. If the Atollic bin/ path appears in your PATH variable before /usr/bin, then you will get the wrong gcc when you try to build host executables. This will cause to strange, uninterpretable errors build some host binaries in tools/ when you first make. The Atollic "Lite" Toolchain ---------------------------- The free, "Lite" version of the Atollic toolchain does not support C++ nor does it support ar, nm, objdump, or objdcopy. If you use the Atollic "Lite" toolchain, you will have to set: CONFIG_HAVE_CXX=n In order to compile successfully. Otherwise, you will get errors like: "C++ Compiler only available in TrueSTUDIO Professional" The make may then fail in some of the post link processing because of some of the other missing tools. The Make.defs file replaces the ar and nm with the default system x86 tool versions and these seem to work okay. Disable all of the following to avoid using objcopy: CONFIG_RRLOAD_BINARY=n CONFIG_INTELHEX_BINARY=n CONFIG_MOTOROLA_SREC=n CONFIG_RAW_BINARY=n devkitARM --------- The devkitARM toolchain includes a version of MSYS make. Make sure that the the paths to Cygwin's /bin and /usr/bin directories appear BEFORE the devkitARM path or will get the wrong version of make. IDEs ==== NuttX is built using command-line make. It can be used with an IDE, but some effort will be required to create the project. Makefile Build -------------- Under Eclipse, it is pretty easy to set up an "empty makefile project" and simply use the NuttX makefile to build the system. That is almost for free under Linux. Under Windows, you will need to set up the "Cygwin GCC" empty makefile project in order to work with Windows (Google for "Eclipse Cygwin" - there is a lot of help on the internet). Native Build ------------ Here are a few tips before you start that effort: 1) Select the toolchain that you will be using in your .config file 2) Start the NuttX build at least one time from the Cygwin command line before trying to create your project. This is necessary to create certain auto-generated files and directories that will be needed. 3) Set up include pathes: You will need include/, arch/arm/src/stm32, arch/arm/src/common, arch/arm/src/armv7-m, and sched/. 4) All assembly files need to have the definition option -D __ASSEMBLY__ on the command line. Startup files will probably cause you some headaches. The NuttX startup file is arch/arm/src/stm32/stm32_vectors.S. With RIDE, I have to build NuttX one time from the Cygwin command line in order to obtain the pre-built startup object needed by RIDE. NuttX EABI "buildroot" Toolchain ================================ A GNU GCC-based toolchain is assumed. The files */setenv.sh should be modified to point to the correct path to the Cortex-M3 GCC toolchain (if different from the default in your PATH variable). If you have no Cortex-M3 toolchain, one can be downloaded from the NuttX SourceForge download site (https://sourceforge.net/projects/nuttx/files/buildroot/). This GNU toolchain builds and executes in the Linux or Cygwin environment. 1. You must have already configured Nuttx in /nuttx. cd tools ./configure.sh stm32vldiscovery/ 2. Download the latest buildroot package into 3. unpack the buildroot tarball. The resulting directory may have versioning information on it like buildroot-x.y.z. If so, rename /buildroot-x.y.z to /buildroot. 4. cd /buildroot 5. cp configs/cortexm3-eabi-defconfig-4.6.3 .config 6. make oldconfig 7. make 8. Edit setenv.h, if necessary, so that the PATH variable includes the path to the newly built binaries. See the file configs/README.txt in the buildroot source tree. That has more details PLUS some special instructions that you will need to follow if you are building a Cortex-M3 toolchain for Cygwin under Windows. NOTE: Unfortunately, the 4.6.3 EABI toolchain is not compatible with the the NXFLAT tools. See the top-level TODO file (under "Binary loaders") for more information about this problem. If you plan to use NXFLAT, please do not use the GCC 4.6.3 EABI toochain; instead use the GCC 4.3.3 OABI toolchain. See instructions below. NuttX OABI "buildroot" Toolchain ================================ The older, OABI buildroot toolchain is also available. To use the OABI toolchain: 1. When building the buildroot toolchain, either (1) modify the cortexm3-eabi-defconfig-4.6.3 configuration to use EABI (using 'make menuconfig'), or (2) use an exising OABI configuration such as cortexm3-defconfig-4.3.3 2. Modify the Make.defs file to use the OABI conventions: +CROSSDEV = arm-nuttx-elf- +ARCHCPUFLAGS = -mtune=cortex-m3 -march=armv7-m -mfloat-abi=soft +NXFLATLDFLAGS2 = $(NXFLATLDFLAGS1) -T$(TOPDIR)/binfmt/libnxflat/gnu-nxflat-gotoff.ld -no-check-sections -CROSSDEV = arm-nuttx-eabi- -ARCHCPUFLAGS = -mcpu=cortex-m3 -mthumb -mfloat-abi=soft -NXFLATLDFLAGS2 = $(NXFLATLDFLAGS1) -T$(TOPDIR)/binfmt/libnxflat/gnu-nxflat-pcrel.ld -no-check-sections NXFLAT Toolchain ================ If you are *not* using the NuttX buildroot toolchain and you want to use the NXFLAT tools, then you will still have to build a portion of the buildroot tools -- just the NXFLAT tools. The buildroot with the NXFLAT tools can be downloaded from the NuttX SourceForge download site (https://sourceforge.net/projects/nuttx/files/). This GNU toolchain builds and executes in the Linux or Cygwin environment. 1. You must have already configured Nuttx in /nuttx. cd tools ./configure.sh lpcxpresso-lpc1768/ 2. Download the latest buildroot package into 3. unpack the buildroot tarball. The resulting directory may have versioning information on it like buildroot-x.y.z. If so, rename /buildroot-x.y.z to /buildroot. 4. cd /buildroot 5. cp configs/cortexm3-defconfig-nxflat .config 6. make oldconfig 7. make 8. Edit setenv.h, if necessary, so that the PATH variable includes the path to the newly builtNXFLAT binaries. LEDs ==== It is asumed that STMicro STM32F100RC generic board board has one LED on PA0. You should configure the port and pin number in configs/stm32vldiscovery/src/stm32f100rc_internal.h. This LED is not used by the board port unless CONFIG_ARCH_LEDS is defined. In that case, the usage by the board port is defined in include/board.h and src/up_leds.c. The LED is used to encode OS-related events as follows: SYMBOL Meaning LED1* green ------------------- ----------------------- ------- LED_STARTED NuttX has been started ON LED_HEAPALLOCATE Heap has been allocated ON LED_IRQSENABLED Interrupts enabled ON LED_STACKCREATED Idle stack created ON LED_INIRQ In an interrupt ON LED_SIGNAL In a signal handler ON LED_ASSERTION An assertion failed OFF LED_PANIC The system has crashed OFF So basically if the LED is off it means that there is a problem. UART ==== Default USART/UART Configuration -------------------------------- USART1 is enabled in all configurations (see */defconfig). RX and TX are configured on pins PA10 and PA9, respectively. Then connect the RX pin of your USB/Serial adapter to TX pin (PA9) and the TX pin of your adapter to RX pin (PA10) of your board besides, of course, the GND pin. "STMicro STM32F100RC generic" specific Configuration Options ============================================================ CONFIG_ARCH - Identifies the arch/ subdirectory. This should be set to: CONFIG_ARCH=arm CONFIG_ARCH_family - For use in C code: CONFIG_ARCH_ARM=y CONFIG_ARCH_architecture - For use in C code: CONFIG_ARCH_CORTEXM3=y CONFIG_ARCH_CHIP - Identifies the arch/*/chip subdirectory CONFIG_ARCH_CHIP=stm32 CONFIG_ARCH_CHIP_name - For use in C code to identify the exact chip: CONFIG_ARCH_CHIP_STM32F100RB=y CONFIG_ARCH_BOARD_STM32_CUSTOM_CLOCKCONFIG - Enables special STM32 clock configuration features. CONFIG_ARCH_BOARD_STM32_CUSTOM_CLOCKCONFIG=n CONFIG_ARCH_BOARD - Identifies the configs subdirectory and hence, the board that supports the particular chip or SoC. CONFIG_ARCH_BOARD=stm32vldiscovery CONFIG_ARCH_BOARD_name - For use in C code CONFIG_ARCH_BOARD_STM32VL_DISCOVERY=y CONFIG_ARCH_LOOPSPERMSEC - Must be calibrated for correct operation of delay loops CONFIG_ENDIAN_BIG - define if big endian (default is little endian) CONFIG_RAM_SIZE - Describes the installed DRAM (SRAM in this case): CONFIG_RAM_SIZE=24576 (24kB) CONFIG_RAM_START - The start address of installed DRAM CONFIG_RAM_START=0x20000000 CONFIG_ARCH_LEDS - Use LED to show state. Unique to boards that have LED(s) CONFIG_ARCH_INTERRUPTSTACK - This architecture supports an interrupt stack. If defined, this symbol is the size of the interrupt stack in bytes. If not defined, the user task stacks will be used during interrupt handling. CONFIG_ARCH_STACKDUMP - Do stack dumps after assertions CONFIG_ARCH_CALIBRATION - when used togeter with CONFIG_DEBUG enables some build in instrumentation that cause a 100 second delay during boot-up. This 100 second delay serves no purpose other than it allows you to calibratre CONFIG_ARCH_LOOPSPERMSEC. You simply use a stop watch to measure the 100 second delay then adjust CONFIG_ARCH_LOOPSPERMSEC until the delay actually is 100 seconds. Individual subsystems can be enabled: AHB ---- CONFIG_STM32_CRC CONFIG_STM32_DMA1 CONFIG_STM32_DMA2 APB1 ---- CONFIG_STM32_TIM2 CONFIG_STM32_TIM3 CONFIG_STM32_TIM4 CONFIG_STM32_TIM5 CONFIG_STM32_TIM6 CONFIG_STM32_TIM7 CONFIG_STM32_TIM12 CONFIG_STM32_TIM13 CONFIG_STM32_TIM14 CONFIG_STM32_RTC CONFIG_STM32_WWDG CONFIG_STM32_IWDG CONFIG_STM32_SPI2 CONFIG_STM32_SPI3 CONFIG_STM32_USART2 CONFIG_STM32_USART3 CONFIG_STM32_UART4 CONFIG_STM32_UART5 CONFIG_STM32_I2C1 CONFIG_STM32_I2C2 CONFIG_STM32_PWR -- Required for RTC CONFIG_STM32_BKP -- Required for RTC CONFIG_STM32_DAC1 CONFIG_STM32_DAC2 CONFIG_STM32_CEC APB2 ---- CONFIG_STM32_ADC1 CONFIG_STM32_TIM1 CONFIG_STM32_SPI1 CONFIG_STM32_USART1 CONFIG_STM32_TIM15 CONFIG_STM32_TIM16 CONFIG_STM32_TIM17 Timer devices may be used for different purposes. One special purpose is to generate modulated outputs for such things as motor control. If CONFIG_STM32_TIMn is defined (as above) then the following may also be defined to indicate that the timer is intended to be used for pulsed output modulation, ADC conversion, or DAC conversion. Note that ADC/DAC require two definition: Not only do you have to assign the timer (n) for used by the ADC or DAC, but then you also have to configure which ADC or DAC (m) it is assigned to. CONFIG_STM32_TIMn_PWM Reserve timer n for use by PWM, n=1,..,17 CONFIG_STM32_TIMn_ADC Reserve timer n for use by ADC, n=1,..,17 CONFIG_STM32_TIMn_ADC1 Reserve timer n to trigger ADCm, n=1,..,17 CONFIG_STM32_TIMn_DAC Reserve timer n for use by DAC, n=1,..,17 CONFIG_STM32_TIMn_DACm Reserve timer n to trigger DACm, n=1,..,17, m=1,..,2 For each timer that is enabled for PWM usage, we need the following additional configuration settings: CONFIG_STM32_TIMx_CHANNEL - Specifies the timer output channel {1,..,4} NOTE: The STM32 timers are each capable of generating different signals on each of the four channels with different duty cycles. That capability is not supported by this driver: Only one output channel per timer. JTAG Enable settings (by default full SWJ is enabled): CONFIG_STM32_JTAG_FULL_ENABLE - Enables full SWJ (JTAG-DP + SW-DP) CONFIG_STM32_JTAG_NOJNTRST_ENABLE - Enables full SWJ (JTAG-DP + SW-DP) but without JNTRST. CONFIG_STM32_JTAG_SW_ENABLE - Set JTAG-DP disabled and SW-DP enabled STMicro STM32F100RC generic specific device driver settings CONFIG_U[S]ARTn_SERIAL_CONSOLE - selects the USARTn (n=1,2,3) or UART m (m=4,5) for the console and ttys0 (default is the USART1). CONFIG_U[S]ARTn_RXBUFSIZE - Characters are buffered as received. This specific the size of the receive buffer CONFIG_U[S]ARTn_TXBUFSIZE - Characters are buffered before being sent. This specific the size of the transmit buffer CONFIG_U[S]ARTn_BAUD - The configure BAUD of the UART. Must be CONFIG_U[S]ARTn_BITS - The number of bits. Must be either 7 or 8. CONFIG_U[S]ARTn_PARTIY - 0=no parity, 1=odd parity, 2=even parity CONFIG_U[S]ARTn_2STOP - Two stop bits Configurations ============== Each STMicro STM32F100RC generic configuration is maintained in a sub-directory and can be selected as follow: cd tools ./configure.sh stm32vldiscovery/ cd - . ./setenv.sh Where is one of the following: nsh: --- Configures the NuttShell (nsh) located at apps/examples/nsh. The Configuration enables only the serial NSH interfaces. Default toolchain: CONFIG_ARMV7M_TOOLCHAIN_CODESOURCERYL=y : CodeSourcery under Linux / Mac OS X