Exploiting Choice: Instruction Fetch and Issue on an Implementable Simultaneous Multithreading Proce
Posted 大大大大大圣归来
tags:
篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了Exploiting Choice: Instruction Fetch and Issue on an Implementable Simultaneous Multithreading Proce相关的知识,希望对你有一定的参考价值。
Time
2020.10.31
Summary
Research Objective
Problem Statement
Method(s)
The methodology in this paper closely follows the simulation and measurement methodology of[27]
Evaluation
Conclusion
Notes
RR.4.2 has four output buses, each two instructions wide.
RR.4.2有四个输出总线,每一个是两指令宽度。
The efficiency of the entire processor is affected by the quality of instructions fetched.
https://blog.csdn.net/shuiliusheng/article/details/84255514
Words
interleaving 交织
decoupled 解耦的
branch target buffer (BTB) 分支目标缓存???
pattern history table (PHT) 图形历史表???
implementation 实现
emulation 仿真
Multiflow 多流程
scheduling 调度
differentiate 鉴别
reap benefits 收获利益
appear to 似乎
culprit 罪魁祸首
simultaneous multi-threading(SMT) 同时多线程
定义:在超标量结构的基础上增加线程上下文控制,允许在同一个时钟周期内发射来自不同线程的多条指令,以充分利用功能部件的一种处理器结构。
exploit 利用
misalignment 错位
fragmentation 碎片
fetch 读取
fetch block fragmentation(our term for the various factors that prevent us from fetching the
maximum number of instructions)
term 术语
indistinguishable 难以区分
finely 精细地
multiplexer 多路选择器
multiplexed 多路复用
circuitry 电路
replicated 复制的
additions 附加物
MUX 复用器
bank conflict 存储体冲突
当被访问的存储体没有恢复时又出现对该存储体新访问的现象
negligible 微不足道
extensive 广泛的
rename register 换名寄存器
寄存器重命名,由于x86指令的限制,可以使用的寄存器就那么几个,但现在CPU的内部实际的寄存器数量是高于x86指令中的寄存器数量的,所以就要在执行指令的时候把X86寄存器重新映射到内部实际寄存器
,寄存器重命名可以提高指令执行的并行度。
dependency chains 依赖链
assume 假设
desirable 可取的
attack 破坏
clog 阻塞
outstanding 优秀的
heuristics 启发式
saturated 饱和的
Cache miss 高速缓存缺失
gain 实现
metrics 指标
Intelligent fetching 智能获取
weighted 加权的
blockked 被阻塞
lookups 查找
Sentence
fetch efficiency, by partitioning the fetch unit among threads
通过在线程之间划分获取单元来获取效率
fetch effectiveness, by improving the quality of the instructions fetched
通过提高所提取指令的质量来获取有效性
four each from two threads
两个线程各四个
reduced by a factor of 2 and 4
减少了2到4倍
splitting the fetch over multiple threads.
将获取指令分为多个线程
we stand to lose
我们承受的损失
we can fetch around cache misses:
我们可以获取缓存未命中
TimtLine
Using a combination of partitioning the fetch unit, intelligent
fetching, and early I cache tag Iookups
以上是关于Exploiting Choice: Instruction Fetch and Issue on an Implementable Simultaneous Multithreading Proce的主要内容,如果未能解决你的问题,请参考以下文章
论文总结Exploiting a Joint Embedding Space for Generalized Zero-Shot Semantic Segmentation
论文导读Autoregressive models are one example of modern ML techniques exploiting joint distribution fa
ParticleSfM:Exploiting Dense Point Trajectories for Localizing Moving Cameras in the Wild——论文笔记
文献内容提取和翻译记录:Exploiting synergies of global land cover products for carbon cycle modeling