Input Methods for Linux (Desktop and Mobile)

Size: px

Start display at page:

Download "Input Methods for Linux (Desktop and Mobile)"

穷圜尚
5 years ago
Views:

1 STM32 簡介 Jim Huang ( 黃敬群 ) <jserv@0xlab.org> Developer, 0xlab July 10, 2012

2 Rights to Copy Copyright xlab Latest update: July 5, 2012 Attribution ShareAlike 3.0 You are free to copy, distribute, display, and perform the work to make derivative works to make commercial use of the work Under the following conditions Attribution. You must give the original author credit. Share Alike. If you alter, transform, or build upon this work, you may distribute the resulting work only under a license identical to this one. For any reuse or distribution, you must make clear to others the license terms of this work. Any of these conditions can be waived if you get permission from the copyright holder. Your fair use and other rights are in no way affected by the above. License text:

3 術語 ART Accelerator : ST s adaptive real-time memory accelerator CMSIS: Cortex microcontroller software interface standard MCU: microcontroller unit DSC: digital signal controller DSP: digital signal processor FPU: floating point unit RTC: real-time clock MPU: memory protection unit FSMC: flexible static memory controller

4 Agenda (1) STM32 平台資訊 (2) ARM Cortex-M 概述 (3) STM32 延伸

5 STM32 平台資訊

6 STM32F4 系列高效數位訊號控制 Cortex-M4 FPU Single precision Ease of use Better code efficiency Faster time to market Eliminate scaling and saturation Easier support for meta-language tools MCU Ease of use of C programming Interrupt handling Ultra-low power DSP Cortex-M4 Harvard architecture Single-cycle MAC Barrel shifter

7 低功耗設計 230 μa/mhz, 38.6 ma at 168 MHz executing Coremark benchmark from Flash memory (with peripherals off), made possible with: ST s 90 nm process allowing the CPU core to run at only 1.2 V Typical values in VBAT mode ART Accelerator reducing the number of accesses to Flash Voltage scaling to optimize performance/power consumption VDD min down to 1.7 V Low-power modes with backup SRAM and RTC support

8 STM32F 系列產品線 le b i t pa m co n i -p ers o t in umb p 0 25 part n r e Ov

9 4 product series STM32F 系列產品線

10 STM32F4 block diagram Feature highlight 168 MHz Cortex-M4 CPU Floating point unit (FPU) ART Accelerator TM Multi-level AHB bus matrix 1-Mbyte Flash, 192-Kbyte SRAM 1.7 to 3.6 V supply RTC: <1 µa typ, sub second accuracy 2x full duplex I²S 3x 12-bit ADC 0.41 µs/2.4 MSPS 168 MHz timers

11 STM32F4 應用領域 Points of sale/inventory management Industrial automation and solar panels Transportation Medical Building Security/fire/HVAC Test and measurement Consumer Communication

12 ARM Cortex-M 概述

13 ARM Cortex-M3 哈佛架構具有分支預測功能的三級 pipeline 指令集 Thumb-2 具備硬體除法與單週期乘法的 ALU Cortex-M3 NVIC MPU (optional) ETM (optional)

14 ARM Cortex 系列 2004 年推出 ARMv7 架构应用在 CortexTM 处理器家族 Cortex 处理器分为三种 A 系列适用于复杂操作系统和应用程序 R 系列适用于嵌入式实时系统 M 系列适用于对成本敏感的深嵌入式领域

15 ARM Cortex 系列 ARMv7-A ( 应用程序 ) 内存控制器 (MMU) 支持虚拟存储器满足多任务操作系统的需要 ARMv7-R ( 实时系统 ) 内存保护单元 (MPU) 低延迟可预测性实时需要支持原有嵌入式系统的升级 ARMv7-M ( 微控制器 ) 最少的门数目深嵌入式 (Deeply embedded) 可与 ARMv7-R 紧密协作

16 ARM Cortex-M3 哈佛结构指令总线和数据总线分离允许并行地取指和数据存储 1.25DMIPS/MHz 0.19mW/MHz Thumb-2 指令集拥有 32 位的性能和 16 位的代码密度单周期乘法和硬件除法 Cortex-M3 核内部集成了嵌入式高速中断控制器低中断响应时间最低可达 6 个 CPU 周期内部中断从低功耗模式被唤醒也只需 6 个 CPU 周期相比于 ARM7TDMI Cortex-M3 内核要快 35% 且减少了 45% 的代码

18 ARM Cortex-M3 三级流水线取指解码和执行单信号周期乘法 Source Destination Cycles 16b x 16b 32b 1 32b x 16b 32b 1 32b x 32b 32b 1 32b x 32b 64b 3-7* 硬件除法 UDIV & SDIV 指令执行需要 2 到 12 周期取决与被除数与除数两者越相近指令完成越快指令可以被中断丢弃 / 重启

19 ARM Cortex-M3 vs. ARM7 ARM7TDMI-S 体系结构 v4t Cortex-M3 v7m ARM (32-bit) & Thumb (16-bit) Thumb-2 (Merged 32/16-bit) DMIPS/MHz 0.74 Thumb / 0.93 ARM 1.25 Thumb-2 流水线 3-Stage 3-Stage + Branch Speculation 中断 FIQ / IRQ 指令集 NMI, SysTick and up to 240 interrupts. Integrated NVIC Interrupt Controller up to Priorities 中断响应时间 Cycles (Depending on LSM) 12 Cycles (6 when Tail Chaining) 存储器印射 Undefined Architecture Defined 系统状态 PSR. 6 modes. 20 Banked regs xpsr. 2 modes. Stacked regs (1 bank) 睡眠模式 No Three Cortex-M3 附加的特性简化了跟踪调试接口的管脚从 9 脚减少到 2 或 3 脚硬件中断处理不需要汇编代码综合的原子位操作改进了数据存储扩展的数据观测点 & Flash 保护技术嵌入式的睡眠控制和掉电模式可选择的小型的存储器保护单元 (MPU) 和嵌入式跟踪宏单元 (ETM)

20 Register R0 R8 R1 R9 R2 R10 R3 R11 R4 R12 R5 R13 (SP) R6 R14 (LR) R7 R15 (PC) PSR PRIMASK R0 R12 General Purpose R13 Stack Pointer R14 Link Register R15 Program Counter FAULTMASK CONTROL BASEPRI 20

21 Memory Map CM3 has a fixed memory map Easy to port software 4GB Address Space STM32 Peripherals SRAM Flash 21

22 Reset SP from address 0x0 PC from address 0x4 Address is mapped to Flash 0x0004 0x0000 Reset Vec. Initial SP

23 STM32 延伸

24 STM32 延伸 ART 6502 sumulator Quake

ART ART_work 是 ART 板在 PC 上的配套软件修改至 Arduino IDE ART 是使用 STM32F407VG 芯片兼容 Arduino 接口核心小板 ART_work 携带 ARM embedded 版本的 GNU GCC 并采用 STM32 DFU 作为下载方式从而能够使用一条 USB 线把程序下载和 RT-Thread 里的 shell 操作都搞定 ART

25 ART ART_work 是 ART 板在 PC 上的配套软件修改至 Arduino IDE ART 是使用 STM32F407VG 芯片兼容 Arduino 接口核心小板 ART_work 携带 ARM embedded 版本的 GNU GCC 并采用 STM32 DFU 作为下载方式从而能够使用一条 USB 线把程序下载和 RT-Thread 里的 shell 操作都搞定 ART 的创新有几点兼容 Arduino 不仅是 Arduino 硬件接口也兼容 Arduino 软件语法源文件 ART_work 的 examples 都未经修改直接引用 Arduino 的例程使用实时多任务的 RT-Thread 操作系统也相应的获得其中的功能特性虚拟文件系统默认加载只读文件系统 FAT 文件系统 pthreads 库等 Arduino 程序编译形成单独可执行的程序这个是在 ARM Cortex-M3/M4 上未使用 MMU 而实现的一种伪进程技术依赖于这种方式把多个 Arduino 程序进行并行化形成多任务目的

26 6502 simulator! posts/ez6agpxhvug After a few optimizations, and actually compiling with -O3 instead of -Os, the simulator is effectively doing a 17.8 MHz more benchmarking in the meantime-- with the more intensive instructions, the sim slows down to about 3MHz. Boo. On the other hand, EhBasic has been added to the mix, and a bug with the BMI instruction was fixed.

27 Quake posts/ez6agpxhvug /albums/ /

28 參考資料 STM32 F4 series, High-performance Cortex -M4 MCU, STMicroelectronics Embedded Programming with the GNU Toolchain, Vijay Kumar B.

ARM Cortex-M3 (STM32F) STMicroelectronics ( ST) STM32F103 Core: ARM 32-bit Cortex -M3 CPU 72 MHz, 90 DMIPS with 1.25 DMIPS/MHz Single-cycle multiplica

CP Chip Power ARM Cortex-M3 (STM32F) ARM Cortex-M3 (STM32F) STMicroelectronics ( ST) STM32F103 Core: ARM 32-bit Cortex -M3 CPU 72 MHz, 90 DMIPS with 1.25 DMIPS/MHz Single-cycle multiplication and hardware