Hello all,      So, this time there is a query on implementing IRDA communication by bit banging the GPIOs. So, I come up with an experiment to do so. Before that, why bit banging when KE02 supports IR communication right away by using FTM at TX and ACMP at RX? something similar to this Implementing infrared functions on UART0 with FRDM-KE02Z platform.        The answer lies in the waveform attached. In the code from the above discussion, the PWM modulation is performed on FTM channel for transmission, while the communication that customer required was different and the waveform required is the one attached.                                                                                           Note that the IRDA transmitter on FRDM-KE02Z is connected to UART0. But, in the attached code, UART0 is not configured for IRDA_TX. while ACMP is used for RX.      This post is intended to be a reference to those who want to implement IRDA with 3/16th bit width by bit banging.      Here is the schematic snapshot of the FRDM-KE02.   There are 2 sets of codes attached. 1. FRDM_KE02_singleboard_IRDA_BITBANGING_TX_RX - IRDA on a single board. 2. FRDM_KE02_interboard_IRDA_BITBANGING_TX_RX - IRDA between 2 boards.   To run a single board demo, the orientation of IRDA Transmitter and IRDA receiver is already taken care as both are on the same board. But, to run the second demo, the orientation is supposed to be something like this.   Both the demos are configured for 9600baud to communicate with PC. And the IRDA communication is at 2400baud.      The figure in page 13 of this document is referred while development. http://www.vishay.com/docs/82513/physical.pdf                                                                            Transmission - 1. The bit width is calculated depending on the required IRDA baud and PIT1 modulo is adjusted accordingly. There are 3 MOD values initialized one after the other to generate required IR frame. 2. The data to be transmitted is sent over the HyperTerminal @ 9600baud. This data is accordingly transmitted on PTB1.      The green waveform in the below oscilloscope snap is the IR frame from the above picture. The data being sent is 'e', the hex value of which is 0x65 which can be made out from the waveform. 0x65 is 0xa6 in reverse since LSB is sent out first in the IRDA protocol[The first logic high is the start bit. In IRDA transmission, logic low is data '1' and logic high is data '0'].                                                                                      Reception - 1. ACMP0 is configured to generate interrupts on either the rising edge or the falling edge. 2. The falling edge of the start bit triggers ACMP. 3. The logic received on PTA1 is decoded to arrive at the received data. 4. The falling edge on ACMP channel decides logic 0 and the PIT0 interrupt decides the logic 1.      The below waveform is the one probed at PTA1.                                                                     Thanks, Pramod TM

To do: In this exercise, you realize a windscreen wiper with the Tower. The LEDs simulate the moving wiper by running from left to right and back again. The user interface is simple: With pushbutton1, the time delay interval is increased in steps of 1 second starting from 5 seconds delay down to 1 second and further to continuous wiping. With pushbutton2 the delay is decreased in the same way. If the delay is decreased, the wiper should start immediately, because right now there is a need for wiping. This is not true for increasing the delay. In this case, wait till the current delay time is over, set the new delay. In technical terms: Use asynchronous interval reset for decrease and synchronous reset for increase. Use active wait loops instead of the timer in this Kinetis exercise. Hint: Have you already switched a case? Result: TWR_K60_wiper_poll.zip

Project Summary MonkeyListen uses the FRDM-K20D50 board (which has a Cortex M4 core with DSP instructions) with the FRDM-OLED shield so you can  make your very own spectrum analyzer display  The end result will be a functional DSP system that will analyze incoming audio content via an electret microphone on FRDM-OLED board and display the spectral content.   The example code will also show you how to plot time domain data (a simple audio scope!),  Frequency domain data (via an FFT) and a time-frequency plot (spectrogram).  Extra I/O are provided to hack the code and create your own DMM or oscilloscope. The FRDM-OLED shield also has an optional RS-485 interface for doing cool things like driving a DMX lighting system! Skills Developed: Spectral Analysis via FFT OLED display interfacing Electret microphone Interfacing Soldering SOIC8 and 1206 surface mount devices Cortex CMSIS DSP Library Materials: FRDM-K20D50 FRDM-OLED Development Tools Install Codewarrior 10.5 for Microcontrollers (Eclipse) Special Edition to your  machine Example Code Get the latest copy from Github Prerequisite Videos: All of the videos are organized on a Youtube playlist: H.I.T. #2: MonkeyListen - YouTube MonkeyListen WatchMe1st MonkeyListen Demo: Time Domain + FFT Mode MonkeyListen Demo: Spectrogram Mode Spectral Analysis for Embedded Systems Part 1 & Part 2 Getting Started with the ARM CMSIS DSP FFT library Introduction to Fixed Point Math for Embedded Systems Part 1, 2 and 3 The q31_t (Q.31) number format for the CMSIS DSP libraries FRDM-OLED Overview EEVblog #611 - Electret Microphone Design Loading and Configuring the MonkeyListen Example Software Step 1:  Get a FRDM-OLED      You can order raw PCBs yourself from OSHPark or your favorite PCB vendor.      The bill of materials for the FRDM-OLED is included with the build package on the FRDM-OLED page.     Please let us know if you are interested in a pre-assembled version.   If there is enough demand we will get some built via a Kickstarter campaign   Don't be afraid to build it yourself,  Soldering is fun!  There is plenty of good stuff on the web on how to do SMT soldering.   All of the parts on the board are fairly simply once you get the hang of it and everything can be hand soldered  The key is having some decent tools. Step 2: Put it Together Attach  The FRDM-OLED to the FRDM-K20D50.   The FRDM-K20D50 comes with female headers that you can solder on so the boards can be easily separated. Step 3: Download Download the example software.   The video "Loading and Configuring the MonkeyListen Example Software" will step you though downloading the program and doing some basic configuration. Step 4: Say Something Speak, Sing and Yell.... Step 5: Hack! Do something else cool.....  Make a DMM,  or an o-scope....

近期有客户提出需求，要求通过外部Flash编程工具烧写Flash Program Flash IFR区域。 目前P&E Cyclone MAX和Segger J-Flash均无法实现对IFR区域编程。 客户可以使用软件的方式来编程IFR提供的单次烧写区域，存储客户产品信息，例如MAC地址等。 Program Flash单次烧写区域提供了64个字节，只允许烧写一次，通过Program Once和Read Once命令来读写这个区域。 下图为单次烧写区域在Prgoram Flash IFR的具体位置， IFR独立于FTFL Flash空间，可以理解成另外一个Flash模块。 Program Once和Read Once命令每次调用可以读取Program Flash单次烧写区域的4个字节，通过命令参数的数据索引号可以通过多次操作遍历整个64个字节。 附件中的例程使用Program Once命令编写MAC地址到单次烧写区域，之后通过Read Once命令读取MAC地址信息。 例程环境： IAR Workbench + TWR-K60D100M

Hi,All Our team have developed the K60's peripheral derives lib which is open source and open source firmware library. the The open source lib have these feature just as follow: 1\The setup code based on the CMSIS Standard; 2\The lib include most of peripherals of K60 such as ADC,DAC,DMA,CAN,FTM,LPTMR, ENET,FLASH,FLexBUS,GPIO,IIC,MCG,PDB,SPI,USB,TSI,UART,WDOG,SDHC,PIT,etc. 3\All of the peripheral initiate function are based on the a structure variable Struct format. 4\Add parts of Freescale USB Stack into the lib such as USB CDC and USB HID mouse. 5\We also develop relative example projects to demonstrate how to use the library for the peripheral. All of the project are created by the IAR for ARM Ver.6.4. The attach is the code. We divide the code into two parts. One part is driver lib, and another part is project. PS 1:All of the code comments are Chinese, please forgive. PS 2:All of drivers is on the \lib\LPLD\HW.      The K60 project need Unzip into the \porject\. PS 3:The code has many place that need update and improve. if you have any doubt and opinions,please contact us.support[AT]lpld.cn Best Regards Wang

Hello, The below attachment is the code for UART with the input data as a character..But i need the input data to be a STRING...Can anyone please post the code for UART with input as a string using MK22D5.

There with phase shifting when using two different FTM modules to output PWM signals. Although the two FTM modules using the same clock source (bus clock), there still exists the phase shifting status. Please check attached video about phase shifting. FTM Global Time Base(GTB) introduction The global time base (GTB) is a FTM function that allows the synchronization of multiple FTM modules on a chip. The following figure shows an example of the GTB feature used to synchronize two FTM modules. In this case, the FTM A and B channels can behave as if just one FTM module was used, that is, a global time base. K65’s FTM0 provides the only source for the FTM global time base. The other FTM modules can share the time base as shown in the following figure: The code description:    // Configure ftm params with frequency 2MHz for CLK        ftm_pwm_param_t ftmParamCLK = {             .mode                   = kFtmEdgeAlignedPWM,   //PWM mode             .edgeMode               = kFtmLowTrue,           //PWM Low-true pulses (clear Output on match-on)             .uFrequencyHZ           = 200000u,         //2MHz clock frequency             .uDutyCyclePercent      = 50,                //Duty cycle 50%             .uFirstEdgeDelayPercent = 0,                        };    //using FTM0 & FTM3 GTB feature    FTM_HAL_SetClockSource (FTM0, kClock_source_FTM_None);   //disable FTM0 clock source    FTM_HAL_SetClockSource (FTM3, kClock_source_FTM_None);   //disable FTM3 clock source      FTM_HAL_SetGlobalTimeBaseCmd(FTM0, true);   //enable FTM0 GTBEEN    FTM_HAL_SetBdmMode(FTM0, kFtmBdmMode_11);   //enable FTM0 BDMMODE    FTM_HAL_SetGlobalTimeBaseCmd(FTM3, true);   //enable FTM3 GTBEEN    FTM_HAL_SetBdmMode(FTM3, kFtmBdmMode_11);   //enable FTM3 BDMMODE        FTM_HAL_SetClockSource (FTM0, kClock_source_FTM_SystemClk);   //disable FTM0 clock source    FTM_HAL_SetClockSource (FTM3, kClock_source_FTM_SystemClk);   //disable FTM3 clock source      FTM_HAL_SetCounter(FTM0, 0U);        //clear TFM0 counter value to 0    FTM_HAL_SetCounter(FTM3, 0U);        //clear FTM3 counter value to 0      FTM_HAL_SetGlobalTimeBaseOutputCmd(FTM0, true);      //enale FTM0 GTBEOUT Please check attached video about after using GTB feature, the FTM0_CH4 and FTM3_CH1 PWM output signals. How to output two PWM output signals at one FTM module with KSDK? The two PWM output signal will provides the same clock frequency with different duty cycle. The two PWM output signal need use the same PWM mode. Please check below code to enable K65’s FTM0 module output two PWM signals. // Configure ftm params with frequency 2MHz for CLK        ftm_pwm_param_t ftmParamCLK = {             .mode                   = kFtmEdgeAlignedPWM,   //PWM mode             .edgeMode               = kFtmLowTrue,           //PWM Low-true pulses (clear Output on match-on)             .uFrequencyHZ           = 200000u,         //2MHz clock frequency             .uDutyCyclePercent      = 50,                //Duty cycle 50%             .uFirstEdgeDelayPercent = 0,                        }; // Configure ftm params with frequency 2MHz for CLK                ftm_pwm_param_t ftmParamSH =         {              .mode                   = kFtmEdgeAlignedPWM,   //PWM mode              .edgeMode               = kFtmLowTrue,   //PWM Low-true pulses (clear Output on match-on)              .uFrequencyHZ           = 200000u,        //2MHz clock frequency              .uDutyCyclePercent      = 75,     //Duty cycle 75%              .uFirstEdgeDelayPercent = 0,         };     // Initialize FTM module,     // configure for software trigger.     memset(&ftmInfo, 0, sizeof(ftmInfo));     ftmInfo.syncMethod = kFtmUseSoftwareTrig;  //Using software trigger PWM synchronization     FTM_DRV_Init(BOARD_FTM_INSTANCE, &ftmInfo);  //FTM0 initialization     FTM_DRV_SetClock(BOARD_FTM_INSTANCE, kClock_source_FTM_SystemClk, kFtmDividedBy1); //Enable FTM0 counter clock     FTM_DRV_PwmStart(BOARD_FTM_INSTANCE, &ftmParamCLK, BOARD_FTM_CHANNEL); //Enable PWM output at FTM0_CH4     FTM_HAL_SetClockSource(FTM0, kClock_source_FTM_None); //Disable FTM0 counter clock     FTM_DRV_PwmStart(BOARD_FTM_INSTANCE, &ftmParamSH, BOARD_FTM_CHANNEL5);   //Enable PWM output at FTM0_CH5 The tested code also be attached, please using it with KSDK V1.2 software. Wish it helps.

       RSA is a major cryptosystem in the public key cryptography. It is wildly used nowadays.  But why are we interested in elliptic    curve cryptography?   1. The smaller parameters can be used in elliptic curve cryptography (ECC) than with  RSA systems at a given security level.       2. In particular, private-key operations (such as signature generation and decryption) for ECC are many times more efficient        than RSA private-key operations.     3. Public-key operations (such as signature verification and encryption) for ECC are more efficient than RSA if a  bigger       encryption exponent e is selected for RSA.           The advantages offered by ECC can be important in environments where processing     power, storage, bandwidth, or power consumption is constrained.              KL26Z MCU is 48 MHz ARM Cortex-M0+ core, there are only 128K ROM and 16K RAM on chip. When you want to use the    public key cryptography, the elliptic curve cryptography may be a good choice for this low cost MCU.        This example implements a simple ECC certification from PC to USB.  After certification,     USB can get encrypted data from PC and decrypt these data to plain text.   KL26-Freedom : 1.   Open the porject file at folder kecc/build/cw/KECC/kl26_ecc/test   2.  Build the image and flush it with OpenSDA.   3. Open the Tera Term and start running image on KL26Z     PC running environment:     1. Install  these packets on PC:         python-2.7.3.msi, pycrypto-2.6.win32-py2.7.exe, pyserial-2.7.win32.exe and pywin32-219.win32-py2.7         2. Connect USB cable to KL26Z Freedom and install the CDC driver with       Freescale_CDC_Driver_Kinetis.inf driver     3. Open the application folder and run the usbkey.py.                 4. If USB admits  PC certification message, then PC can put encrypted data to USB.               This  demo only tested on WIN7 64bit. IF you can't open the CDC port, you need reset the KL26Z freedom board and restart the usbkey.py.     This ECC demo uses a library including some security toolkits for ECC certification.     1. Signature and verify : ECDSA     2. Key exchanging         :ECCRYPT    3.  Digest                         : SHA1    4. Symmetric   cipher     : AES and base64      All tools and codes are in following KECC.zip. Original Attachment has been moved to: kecc.zip

介绍通过使用Kinetis KL系列以及K系列通过不同方式驱动液晶。

The documentation is only for eFlexPWM module of KV58, it describes the feature of nano-edge PWM, the mechanism of nano-edge PWM, and give the waveform which can describe the feature of nano-edge PWM. The attachment includes the brief introduction of nano edge PWM, the waveform of nano edge PWM, and the code which runs on TWR-KV58 and KDS3.0. Original Attachment has been moved to: eFlexPWMNanoEdgeKV58_2.rar

UART Presentation (universal asynchronous receiver/transmitter) by Ali Piña, Freescale TIC. Module Explanation Connection Diagram Hands-On. Polling mode. Interrupt Mode. Presentación de UART  (universal asynchronous receiver/transmitter) by Ali Piña, Freescale TIC. a.       Explicación del modulo. b.      Diagrama de conexión. c.       Hands-On. Modo de Poleo Modo de Interrupción

Many customers reported that their ADC function works on FRDM-KL27Z board but meet issue on their own board. We need to pay attention to the difference between the ADC reference voltages of different packages (on board MKL27Z64VLH4 is 64LQFP package). This tip introduce the ADC Reference Options on KL17/KL27 32/36pin package Part number involved: 32-pins 36-pins MKL17Z32VFM4 MKL17Z32VDA4 MKL17Z64VFM4 MKL17Z64VDA4 MKL27Z32VFM4 MKL27Z32VDA4 MKL27Z64VFM4 MKL27Z64VDA4 PTE30/VREF_OUT- connected as the primary reference option on 36-pin and below packages VDDA/VSSA - connected as the VALT reference option   ADCx_SC2[REFSEL] selects the voltage reference source used for conversions.   About the primary reference option: When on-chip 1.2V VREF is enabled, PTE30 pin must be used as VREF_OUT and has to be configured as an analog input, such as ADC0_SE23 (PORTE_PCR30[MUX] = 000). Notice: this pin needs to connect a capacitor to ground.   PTE30 can also be used as an external reference voltage input as long as PTE30 is configured as analog input and VREF module is disabled. It means you can connect external reference voltage to PTE30 pin and use it as ADC reference voltage. (For example 3.3V) KL17P64M48SF2RM     Kinetis KL17: 48MHz Cortex-M0+ 32-64KB Flash (32-64pin) (REV 4.1) KL27P64M48SF2RM     Kinetis KL27: 48MHz Cortex-M0+ 32-64KB Flash (32-64pin) (REV 4.1)

Kinetis M0+ FAQ文档的版本更新历史： Version 1: Basic structure; Version 2: Add 2 items Top highlights, 7 items Common and 7 items Tools; Version 3: add "Cyclone Max 使用步骤及注意事项"; Version 4,5,6: Adjust the FAQ structure; Version 7: Add "Kinetis L 、E、V、M系列选型指南"; Version 8: Add "开源工具"; Version 9: Add "Kinetis SDK"; Version 10,11,12: Add "智能插座"; Version 13: Add “FAQ使用规范” and "Kinetis Bootloader", and Readjust the "参考设计与方案"; Version 14, 15: Add the new motor control solution link into the "参考设计与方案"; Version 16: Add Blogs to "常用网站资料"; Version 17: Push the Kinetis M0+ FAQ Version log histors' information into subpage of "Kinetis M0+ FAQ 版本更新历史"; Version 18: Add "飞思卡尔MAPS开发板资料" into "常用网站资料"; Version 19: Replace "iBeacon" with "BLE"; Version 20, 21: Remove "WIFI" solution, Add "SPI接口读写SD卡"; Version 22: Rename "软件和例程" with "软件和文档", Add "常用的应用笔记" into "软件和文档" list; Click here to return to Kinetis M0+ FAQ main page...​

Hi everybody, The original document was moved to KSDK subspace. You can find it in the next link: https://community.freescale.com/docs/DOC-102612 Regards, Carlos

There is a popular WIFI platform called “GoKit” in China. This testing kit can be use to do some customized application. Not only WIFI communication, kit also support other functions. You can find interfaces listed as below. GoKit Interfaces: I try to use FRDM-KL02 to communicate with this kit to do a WIFI communication application. Board connection as below. This platform has two running mode. One is AirLink mode, and another is normal running mode. AirLink mode is used to WIFI communication or pair. Go to AirLink mode steps: Power on FRDM-KL02 Long press key1 to reset WIFI module. Wait until RED led on. Short press Key2 to go into configuration mode, wait until RED led flash on WIFI module. Open demo APP, select “adding device”, input SSID password. Waiting for configuration finish. Command Format HOF: 2bytes, value 0xFFFF Length: 2bytes Cmd:1byte SN:1byte Flags:2bytes DATA: Xbytes Checksum:1byte WIFI acquire device information MCU inform WIFI into configure mode MCU reset WIFI WIFI inform MCU status WIFI ask for reset Illegal command For detail code, I will post another thread for your reference.

Freescale Semiconductor is to demonstrate its Kinetis L series microcontrollers (MCUs) built on the ARM Cortex-M0+ processor at DESIGN West in San Jose, California, with alpha sampling due to start in the second quarter of 2012. Freescale  says the ability to demonstrate these devices is possible due to its  close partnership between ARM during the Cortex-M0+ core development  process and as a lead partner provided  input that helped ARM define and  develop the processor. The devices are slated for applications  such as domestic appliances, portable medical systems, smart meters,  lighting, power and motor control systems. "Our close partnership  with ARM throughout the design and development of their new core has  positioned us as the first MCU supplier to produce and demonstrate an MCU based on the Cortex-M0+ and continues our strategy of driving to  market new products based on the ARM architecture," said Reza  Kazerounian, senior vice president and general manager of Freescale’s  Automotive, Industrial and Multi-Market Solutions Group. Mike  Inglis, executive vice president and general manager of ARM’s Processor  Division, added "With the addition of the L series to their Kinetis  line, Freescale is creating one of the industry’s broadest, most  scalable ARM Cortex-M MCU portfolios, ranging from very low-cost,  entry-level products based on the ARM Cortex-M0+ processor, up to 4 MB,  200 MHz devices based on the Cortex-M4 processor." Manufactured  using Freescale’s low-leakage, 90 nm thin film storage (TFS) process  technology, the Kinetis L series will have a selection of on-chip flash  memory densities and analog, connectivity and HMI peripheral options. Upward  migration through the Kinetis portfolio is available via compatible Kinetis K series devices (built on the ARM Cortex-M4 processor) that  provide access to DSP performance and advanced feature integration. The  ARM Cortex-M0+ processor includes a reduced two-stage pipeline,  allowing faster branch instruction execution, single-cycle access to I/O  and critical peripherals, optimized access to program memory, linear 4  GB address space that removes the need for paging, reducing software  complexity and ensuring a more 8-bit-like user experience and a micro  trace buffer, providing a low-cost trace solution that allows faster bug  identification and correction without the need for additional I/O  resources. Freescale will demonstrate the ARM Cortex-M0+ core at its exhibition booth #1604 at DESIGN West , March 26-29 at the San Jose McEnery Convention Center.

The Technical University in Brasov (Romania) has set up a Medical Teaching lab featuring the Freescale Tower Kit K53 with Oximeter MED-SPO2 and 2 electrode system EKG MED-EKG. Find below the material associated to this lab led by Prof. Sorin-Aurel Moraru from the Faculty of Engineering. He can be contacted at smoraru@unitbv.ro The course is in Romanian and include an overview of the hardware and exercises

Using a MK70FN1M0VMJ15 processor with MII and ADC2_SE16 enabled. Also using Codewarrior 10.6 and MQX4.022. Anytime I initialize the ADC2_SE16 the webserver stops working. I have delayed the initialization of the ADC2_SE16 and found that webserver works fine till initialization. I have also noticed that when the ADC2_SE16 is initialized first the webserver never gets link active. My question is that even though these are different ports and no pin is shared could the possibly be affecting each other? Hi Lee: I have been looking for information related to this problem but at this moment I have not find anything, In this case I would like to know if you are using a development board from Freescale or if you are using a custom board? Have a nice day, Perla Moncada ----------------------------------------------------------------------------------------------------------------------- Note: If this post answers your question, please click the Correct Answer button. Thank you! ----------------------------------------------------------------------------------------------------------------------- It is a custom board, with MII setup instead of RMII. I am concerned that it may be a chip issue, and have no real way to test that possibility with the Development board. This document was generated from the following discussion: 

Here you will find both the code and project files for the USB Mouse project. In this project the USB module is configured as a device, the X and Y coordinates to move the cursor are obtained from the accelerometer measurements. Once the code is loaded it is necessary to disconnect the USB cable from the J26 USB connector and plug it to the K64 USB connector. Once the device enumerates you can use it as an air mouse. The left and right click buttons have not been enabled. To compile the project you must import the following libraries: USBMouse.h FXOS8700Q.h Code: #include "mbed.h" #include "USBMouse.h" #include "FXOS8700Q.h" //I2C lines for FXOS8700Q accelerometer/magnetometer FXOS8700Q_acc acc( PTE25, PTE24, FXOS8700CQ_SLAVE_ADDR1); USBMouse mouse; int main() {     acc.enable();     float faX, faY, faZ;     int16_t x = 0;     int16_t y = 0;       while (1)     {         //acc.getAxis(acc_data);         acc.getX(&faX);         acc.getY(&faY);         x = 10*faX;         y = 10*faY;               mouse.move(x, y);         wait(0.001);     } }

在电机控制，Audio等很多应用中，我们经常会用到一些常见的正余弦，矩阵变换，FFT等一些DSP函数，提到DSP库，通常会想到使用ARM 公司提供的 CMSIS 库。CMSIS 库是ARM和一些半导体厂家针对Cortex-M系列制定的一套接口标准，包括针对内核操作的CMSIS-CORE API，针对DSP应用的CMSIS-DSP Library，针对RTOS的CMSIS-RTOS API，与外设接口的CMSIS-SVD以及提供Debug访问接口的CMSIS-DAP。 其中，又以DSP应用的CMSIS-DSP 库的应用最为广泛。针对Cortex-M4中的DSP功能，CMSIS-DSP部分提供了超过60多种功能的DSP算法库，尤其是随着Cortex-M4中集成了FPU硬件单元，CMSIS-DSP 库的应用也越来越广泛。 在KEIL 和IAR中都集成了对CMSIS的支持，然而在CodeWarrior中没有直接支持CMSIS，需要用户移植到自己的CW工程中，所以就需要使用者了解CMSIS的结构，手动添加库文件和头文件，并完成一些重要的编译参数配置。特别是有些芯片支持FPU浮点运算单元，有些不支持，在配置选项上差别很大。在飞思卡尔Kinetis系列芯片中，FPU浮点运算单元也是一个可选的部件，只有在名称中带有FN和FX的芯片才支持FPU硬件浮点功能，如MK60FN1M0, MK60FX512。本文档分别介绍在使用和不使用FPU的情况下如何一步步移植CMSIS的DSP库到自己的CodeWarrior工程中。 需要注意的是 FPU 单元是指的芯片上的一个独立于 CPU 处理的浮点运算单元，整个单元在大多数厂家的芯片中都是可以被使能和关闭的。相对于芯片，编译器也设置了相应的 FPU 功能开启/关闭的选项，在编译时需要告诉编译器是否开启 FPU 功能。编译器一旦开启 FPU 功能，在处理单精度浮点运算的语句时就会用带 V-开头的汇编指令进行编译。如果编译器使能了 FPU 功能，而芯片未开启 FPU 单元，程序运行到浮点语句时就会出现异常。相反，如果编译器未使能 FPU 功能，芯片即使开启了 FPU单元，程序还是会按照未使能 FPU 的代码进行处理。在本例程中，为对比分析是否采用FPU的编译指令差别以及在板的执行效率，选用Kinetis K70FN1M为实验对象。 硬件平台：TWR-K70F120M核心板      软件环境：CodeWarrior v10.5        CMSIS版本 ：V3.2 一. 准备工作： 下载CMSIS的库，当前最新的版本为V3.2，解压后名称为CMSIS-SP-00300-r3p2-00rel1，其目录结构如下图。分别包含CMSIS-DSP, CMSIS-RTOS和CMSIS-SVD的库文件。在本Cortex M4的CMSIS-DSP的应用中，真正用到的文件包括CMSIS\Include中CM4相关的头文件，CMSIS\Lib\GCC文件夹中的库文件libarm_cortexM4l_math.a(软浮点)和libarm_cortexM4lf_math.a(硬浮点)，以及Device\ARM\ARMCM4\include中外设访问相关的两个头文件。 二. 不使用K70的FPU浮点运算单元，移植CMSIS的DSP库到CW的步骤；      Step 1:    新建一个Baremental的工程，选择器件器件MK70FN1M0（支持硬件FPU）；     Step 2:    选择Floating Point浮点运算实现的类型，即指定编译器将C代码编译成汇编代码时使用的规则； 在Floating Point四个选项含义如下： Software选项：表示不使用FPU硬件，而是使用GCC的整数算术运算来模拟浮点运算； Handware(-mfloat –abi=hard) 选项：表示使用FPU硬件来进行浮点运算，函数的参数直接传递到FPU的寄存器(s0-d0)中； Handware(-mfloat –abi=softfp)选项：表示使用FPU硬件来进行浮点运算，但是函数的参数传递到整数寄存器(r0-r3）中，然后再传递到FPU中； Handware(-mfloat –abi=softfp –fshort -double)选项：其配置项同上，只不过使能了fshort  double功能，并且此处的double数据的宽度等同于float； 有兴趣研究各个选项意义的可以参考CW for MCU技术文档的第3章，在本例程中使用的是软浮点，所以选择Software项。需要注意的是：此配置选项仅出现支持FPU硬件单元的芯片工程中，如MK60FN1M0, MK60FX512等，否则默认没有此选项，默认为软件浮点。 Step 3:    点击“Next”进入下图，选择使用Processor Expert，点击“Finish“完成工程的建立； Step 4:    进入当前工程的文件夹，新建文件夹CMSIS，从之前在准备步骤解压的CMSIS文件包\...\CMSIS-SP-00300-r3p2-00rel1\CMSIS中拷贝Include和Lib文件夹到当前工程新建的CMSIS文件夹。另外，拷贝\...\CMSIS-SP-00300-r3p2-00rel1\Device\ARM\ARMCM4\Include中的ARMCM4.h和system_ARMCM4.h到当前工程新建的CMSIS文件夹中； Step 5:    回到CodeWarrior主界面选择新建的工程文件，F5刷新可以看到CMSIS出现在工程文件中。其中Include是CMSIS库的一些头文件，包括M0+/M3/M4的一些头文件；在Lib文件中是已经编译好的库文件，ARM文件夹是使用在KEIL IDE中的库文件，G++文件夹是使用在IAR中的库文件，而由于当前CW工程使用的GCC的编译器，所以GCC文件夹才是CW需要的，因此，为缩减工程大小可以删除ARM和G++文件夹；     Step 6:    打开工程属性框，选择Target Processor的Float ABI为No FPU； Step 7:    在GCC Complier的Defined symbols中添加编译的宏定义ARM_MATH_CM4； Step 8:    在GCC Complier的Include paths中添加CMSIS库的头文件，路径为：工程目录\CMSIS\Include； Step 9:    在GCC C Linker的Miscellaneous项的Other objects中指定使用的库文件(位于CMSIS\Lib\GCC文件夹中)。因为本例程中不使用FPU，所以选择libarm_cortexM4l_math.a，此处需要特别注意，否则编译会报错； Step 10: 在ProcessorExpert.c文件中添加代码； #include <math.h> #include "arm_math.h" #define DELTA    (0.000001f) const float32_t testRefOutput_f32 = 1.000000000; const float32_t radians=1.047197533333333; float32_t  cosOutput, sinOutput, diff; float32_t cosSquareOutput,sinSquareOutput,testOutput; int main(void) {   int m,n;   PE_low_level_init();   cosOutput = arm_cos_f32(radians); /*求正余弦*/                   sinOutput = arm_sin_f32(radians);                                 arm_mult_f32(&cosOutput, &cosOutput, &cosSquareOutput, 1); /*求积运算*/   arm_mult_f32(&sinOutput, &sinOutput, &sinSquareOutput, 1);               arm_add_f32(&cosSquareOutput, &sinSquareOutput, &testOutput, 1); /*求和运算*/   diff = fabsf(testRefOutput_f32 - testOutput);  /* 求绝对值 */                 if(diff > DELTA)   {                         while(1)                                 {                                   for(m=0;m<2000;m++)                                                 for(n=0;n<200;n++){};                                                 D7_NegVal();                                 }     }                 } Step 11:    编译并下载Debug，在  sinOutput = arm_sin_f32(radians);处设置断点，可以看到CMSIS-DSP库中的正余弦浮点数运算函数运算正常，其反汇编的得到为普通的ARM指令(FPU 单元汇编指令通常在普通指令前加字母V，仅在 FPU 功能被使能时使用)，完成一个正弦计算； 至此，完成了不使用K70的FPU浮点运算单元，移植CMSIS的DSP库到CW的步骤。     三.     使用K70的FPU浮点运算单元，移植CMSIS的DSP库到Codewarrior的步骤 使用K70的FPU硬件浮点运算单元，移植CMSIS的DSP库到Codewarrior的方法有两种：一种是按照上面软浮点的方式Step By Step的建立工程，步骤和上面二的步骤基本一致，主要的两个区别在于：(1). Step 2要选择Handware(-mfloat –abi=hard) 选项，(2). Step 9 改为libarm_cortexM4lf_math.a。另外一种就是在上面工程的基础上修改配置选项。鉴于方便，本教程采用第二种方案，完成在Codewarrior中调用CMSIS的DSP库，通过K70的FPU浮点运算单元执行浮点运算，即硬浮点。 Step 1:    打开工程属性对话框，选择Target Processor的Float ABI为FPU with hard vfp passing(-mfloat –abi=hard)； Step 2:    在GCC Complier的Defined symbols中添加编译的宏定义：_VFPV4； Step 3: 在GCC C Linker的Libraries项的library search path中指定链接的规则，把上面工程中默认的"${MCUToolsBaseDir}/ARM_GCC_Support/ewl/lib/armv7e-m"修改为 "${MCUToolsBaseDir}/ARM_GCC_Support/ewl/lib/armv7e-m/fpu"； Step 4:    在GCC C Linker的Miscellaneous项的Other objects中指定使用的库文件(位于CMSIS\Lib\GCC文件夹中)。因为本例程中使用FPU，所以选择libarm_cortexM4lf_math.a，此处需要特别注意，否则编译会报错； Step 5:    完成以上配置后，编译工程并下载调试（此处建议编译之前先Clean一下整个工程），同样，在  sinOutput = arm_sin_f32(radians);处设置断点，可以看到CMSIS-DSP库中的正余弦浮点数运算函数运算正常，其反汇编的得到为FPU指令(FPU指令通常是指在普通指令前加字母V，仅在 FPU功能被使能时使用)，并且在Register观察窗口中也多了个FPU寄存器列表，感兴趣的读者可以对比一下和前面实验汇编出代码的差异，此处不再赘述； 至此，分别完成了使用和不适用K70的FPU浮点运算单元情况下， CMSIS的DSP库到Codewarrior的移植。在实际应用中需要用到更多的DSP函数，在项目中直接调用即可。下一步， Just Enjoy The Convenience of  The CMSIS! 有一点需要说明的是，前文中讲到使用FPU单元需要两步设置：(1). 在编译器中开启相应的 FPU 功能选项；(2). 开启芯片FPU 单元。似乎我们前面的设置是完成了第一个步骤，然而第二个步骤呢？仔细查看_arm_atart.c文件，可以发现代码_fp_init()正是完成了开启芯片FPU单元的过程，如下图，是一个条件编译函数，这也就解释了为什么在上面Step 2中定义了_VFPV4，其本质也就是使能芯片的FPU单元，其具体实现可以查看ARM手册的第74页的描述。 #ifdef   __VFPV4__        //Step 2中的宏定义                 __fp_init();      // 开启芯片的FPU单元 #endif