Memory-mapped I/O And port I/O
2012-06-28 10:21
417 查看
1.1 Memory-mapped I/O
Ø I/O设备的寄存器和内存被映射到CPU的内存地址空间Ø CPU访问I/O设备的方式和操作普通内存的方式是一样的。这样简化了I/O操作。
Ø I/O设备和普通内存共享相同的总线信号(地址、数据、控制)
Ø 因为I/O操作要比内存操作慢，所以这种方式，会降低总线的速率，造成内存操作的性能下降
Ø 由与此类型的CPU架构采用了单独的地址空间，所以CPU的逻辑、设计都变的简单了，CPU的运行速度变快了，RISC计算机就是采取这种设计方式。
1.2 port I/O
Ø 有单独的指令来进行I/O访问，在intel架构中，通过in/out指令完成Ø 有单独的I/O地址空间
Ø 有单独的I/O总线
内容来自用户分享和网络整理，不保证内容的准确性，如有侵权内容，可联系管理员处理 
相关文章推荐
- Memory-mapped I/O and port-mapped I/O
- Access Physical Memory, Port and PCI Configuration Space
- 到底什么是I/O Port，什么是Memory-mapped I/O
- 整理一下 PCI的Memory Mapped IO vs Port IO
- IO的端口映射和内存映射 （Port mapped I/O 和 Memory mapped I/O说明）
- 整理一下 PCI的Memory Mapped IO vs Port IO
- Memory-mapped I/O vs port-mapped I/O
- Memory-mapped I/O vs port-mapped I/O
- tomcat调优方案Maximum number of threads (200) created for connector with address null and port 8091
- DEBUGGING AND OPTIMIZING MEMORY
- Could not find action or result There is no Action mapped for namespace / and action name . - [unkno
- Dynamic Memory and Smart Poniters(I)
- Dual-Port Block RAM with Two Write Ports and Byte-wide Write Enable in Read-First Mode
- How to install Windbg and get your first memory dump
- Avoid memory copying between user space and kernel space
- Method and apparatus for training a memory signal via an error signal of a memory
- There is no Action mapped for namespace / and action name UserAction
- Memory-mapped I/O
- CPU-Cache-and-Memory-Ordering