对仿真glbl.v文件的理解
Simulation, UniSim, SimPrim - How do I use the "glbl.v" module in a Verilog simulation?
Description
How do I use the "glbl.v" module in a Verilog simulation?
Solution
The "glbl.v" module connects the Global Set/Reset and Global Tristate signals to the design. In order to properly reset the design in a Verilog simulation, the "glbl.v" module must be compiled and loaded along with the design. The "glbl.v" module is located at "$XILINX/verilog/src/glbl.v".
Using 6.1i design tools and later
In the 6.1i design tools, the "glbl.v" module was modified to automatically pulse GSR (FPGA Global Set/Reset) and PRLD (CPLD Global Set/Reset) for the first 100 ns of simulation. Code was also added to automatically pulse Global Tristate (GTS), but the default pulse is 0 ns.
For exact commands on how to compile and load the "glbl.v" in ModelSim, see the following solutions:
(Xilinx Answer 1078) - Behavioral Simulation
(Xilinx Answer 10177) - Post-PAR Timing Simulation
For additional information, reference the Synthesis and Simulation Design Guide:
http://toolbox.xilinx.com/docsan/xilinx6/books/docs/sim/sim.pdf
In Chapter 6, Verifying Your Design, there is a section on "Understanding the Global Reset and Tristate for Simulation."
Using 5.1i/5.2i design tools and earlier versions
Prior to the 6.1i release, the "glbl.v" module did not automatically pulse the GSR or PRLD signal. It is therefore necessary to drive GSR or PRLD and/or GTS from the testbench. This is the code that needs to be added to the testbench:
reg GSR;
assign glbl.GSR = GSR;
reg GTS;
assign glbl.GTS = GTS;
initial begin
GSR = 1;
#100 GSR = 0;
end
NOTE 1: For CPLD designs, replace GSR with PRLD in above the code.
NOTE 2: GTS can also be driven, but it is generally not necessary unless you are doing a board-level simulation.
For exact commands on how to compile and load the "glbl.v" in ModelSim, see the following solutions:
(Xilinx Answer 1078) - Behavioral Simulation
(Xilinx Answer 10177) - Post-PAR Timing Simulation
For additional information, reference the Synthesis and Simulation Design Guide:
http://toolbox.xilinx.com/docsan/xilinx6/books/docs/sim/sim.pdf
In Chapter 6, Verifying Your Design, there is a section on "Understanding the Global Reset and Tristate for Simulation."
Xilinx FPGAs have register (flip-flops and latches) set/reset circuitry that pulses at the end of the configuration mode. This pulse is automatic and does not need to be programmed. All the flip-flops and latches receive this pulse through a dedicated global GSR (Global Set-Reset) net. The registers either set or reset, depending on how the registers are defined.
For some device families, it is important to address the built-in reset circuitry behavior in your designs starting with the first simulation to ensure that the simulations agree at the three primary points.
For the Virtex and Spartan-II device families, Xilinx recommends using the manual reset instead of the dedicated GSR circuitry. This is because the implementation tools use the high-speed backbone routing for Reset signals, thus making them faster than the dedicated global routing which transports the GSR signal. However, for the XC4000 and Spartan device families, GSR is the better method of propagating the global reset signal.
For the XC4000 and Spartan device families, if you do not simulate GSR behavior prior to synthesis and place and route, your RTL and possibly post-synthesis simulations might not initialize to the same state as your post-route timing simulation. As a result, the various design descriptions will not be functionally equivalent and your simulation results will not match. Some synthesis tools can identify, from the behavioral description, the GSR net, and will place the STARTUP module on the net to direct the implementation tools to use the global network. However, other synthesis tools interpret behavioral descriptions literally, and will introduce additional logic into your design to implement a function. Without specific instructions to use device global networks, the Xilinx implementation tools will use general purpose logic and interconnect resources to redundantly build functions already provided by the silicon.
If GSR behavior is not described, the chip will initialize during configuration, and the post-route netlist will include this net that must be driven during simulation. This section includes the methodology to describe this behavior, as well as the GTS behavior for output buffers.
In addition to the set/reset pulse, all output buffers are set to a high impedance state during configuration mode with the dedicated global output tristate enable (GTS) net.
The GSR net requires special handling during synthesis, simulation, and implementation to prevent them from being assigned to normally routed nets, which uses valuable routing resources and degrades design performance. The GSR net receives a reset-on-configuration pulse from the initialization controller, as shown in the following figure.
Figure 6-2 Built-in FPGA Initialization Circuitry
This pulse occurs during the configuration mode of the FPGA. However, for ease of simulation, it is usually inserted at time zero of the test bench, before logical simulation is initiated. The pulse width is device-dependent and can vary widely, depending on process voltage and temperature changes. The pulse is guaranteed to be long enough to overcome all net delays on the reset special-purpose net. The parameter for the pulse width is TPOR, as described in The Programmable Logic Data Book.
The tristate-on-configuration circuit shown in the "Built-in FPGA Initialization Circuitry" also occurs during the configuration mode of the FPGA. Just as for the reset-on-configuration simulation, it is usually inserted at time zero of the test bench before logical simulation is initiated. The pulse drives all outputs to the tristate condition they are in during the configuration of the FPGA. All general-purpose outputs are affected whether they are regular, tristate, or bi-directional outputs during normal operation. This ensures that the outputs do not erroneously drive other devices as the FPGA is being configured. The pulse width is device-dependent and can vary widely with process and temperature changes. The pulse is guaranteed to be long enough to overcome all net delays on the GTS net. The generating circuitry is separate from the reset-on-configuration circuit. The pulse width parameter is TPOR, as described in The Programmable Logic Data Book. Simulation models use this pulse width parameter for determining HDL simulation for global reset and tristate circuitry.
If a global set/reset is desired for behavioral simulation, it must be included in the behavioral code. Any described register in the code must have a common signal that will asynchronously set or reset the register depending on the desired result. Similarly, if a global tristate-state is desired for simulation, it should be described in the code as well.
参考
https://wenku.baidu.com/view/fb602c53ad02de80d4d8402b.html
对仿真glbl.v文件的理解的更多相关文章
- QuestaSim自动化仿真之do文件
一.编写基本的do文件 下面按照实际仿真的步骤来说明do文件中需要用到的各个tcl命令. 1.quit -sim ---- 退出原来的仿真工程: 2.cd ---- 设置工作目录的路径,就是仿真工程路 ...
- modelsim仿真中 do文件的写法技巧
网上的关于DO文件的编写好像资料不多,比较杂,所以本人总结一下常用的简单语法,方便大家查看.其实本人也刚接触DO文件没多久,有纰漏很正常,欢迎指正批评,互相学习.PS:写得有点乱 还有一个值得注意 ...
- VCS仿真生成fsdb文件(Verilog)
VCS仿真生成fsdb文件(Verilog) 一.环境 Linux 平台 csh环境 VCS 64bit Verdi3 二.开始仿真 1. 联合仿真环境配置 a.在testbench中加入如下语句: ...
- VCS仿真生成vpd文件(verilog)
VCS仿真生成vpd文件(verilog) 一.环境与文件 Linux平台 csh环境 VCS 64bit 代码文件请参考<一个简单的Verilog计数器模型> 二.开始仿真 1.com ...
- 在头文件中声明class 类 与 include类所在的头文件区别---理解
在头文件中声明class 类 与 include类所在的头文件的理解: 在头文件中,声明类 它告诉编译器:存在这样的类.而实际的类则可以位于同一个编译单元中,也可以放在其他编译单元中.没有这个类原型, ...
- 如何对多个文件进行MODELSIM仿真? (由于是一个很大的项目,不可能把所有MODULE都放在一个文件里。 如何在ModelSim中对多个.V文件进行仿真?)
可以将所有要编译的所有文件的名字做一个list.新建一个文本文档,重命名为vflist vflist内容例子如下(src为文件夹):src/base_addr_chk.vsrc/config_mux. ...
- Modelsim自动化仿真之do文件书写
创建本地库 vlib ./work You must use vlib rather than operating system commands to creat a library directo ...
- C 头文件阅读理解
__BEGIN_DECLS ..... ..... __END_DECLS 很多时候,为了使 C 代码和 C++ 代码保持互相兼容的过程调用接口,需要在 C++ 代码里加上 extern " ...
- stm32f10x.h文件分析理解
今天再看过半年前自己写的这篇发现自己当时理解有误,stm32f10x.h与库开发并未存在太大关系,只是一个最为重要的寄存器地址到寄存器结构体变量的映射. stm32f10x.h 这个头文件是STM32 ...
随机推荐
- java8新特性: lambda表达式:直接获得某个list/array/对象里面的字段集合
java8新特性: lambda表达式:直接获得某个list/array/对象里面的字段集合 比如,我有一张表: entity Category.java service CategoryServic ...
- spring cloud: Hystrix(八):turbine集群监控(dashboard)
turbine是聚合服务器发送事件流数据的一个工具,hystrix的监控中,只能监控单个节点,实际生产中都为集群, 因此可以通过turbine来监控集群下hystrix的metrics情况,通过eur ...
- week02 课堂作业
测试一:(点此看原题目) 运行结果: 测试二:(点此看原题目) 运行结果: 测试三:(点此看原题目) 运行结果:
- every day a practice —— morning(4)
If there’s one thing New Yorkers love more than discovering a new secret remedy, it’s telling other ...
- 安卓中使用OkHttp发送数据请求的两种方式(同、异步的GET、POST) 示例-- Android基础
1.首先看一下最终效果的截图,看看是不是你想要的,这个年代大家都很忙,开门见山很重要! 简要说下,点击不同按钮可以实现通过不同的方式发送OkHttp请求,并返回数据,这里请求的是网页,所以返回的都是些 ...
- canvas绘图在手机上边缘粗糙
辛辛苦苦用canvas绘图,做好动画后,想看看效果,结果在手机上一打开,效果是有了,但是边缘特别粗糙,这怎么可以呢,有一种说法是你用canvas绘图时候在手机retain屏幕上,他把一个像素分为两个像 ...
- adobe
使用adobe acrobat pro dc可以处理pdf,自动识别,编辑pdf,将pdf导出为word.(收费可破解)
- python3 设置滚动条
#!python3#coding=utf-8from selenium import webdriverfrom selenium.webdriver.common.by import Byimpor ...
- 『TensotFlow』RNN中文文本_上
中文文字预处理流程 文本处理 读取+去除特殊符号 按照字段长度排序 辅助数据结构生成 生成 {字符:出现次数} 字典 生成按出现次数排序好的字符list 生成 {字符:序号} 字典 生成序号list ...
- github上fork了别人的项目后,再同步更新别人的提交(转)
原文地址:github上fork了别人的项目后,再同步更新别人的提交 我从github网站和用git命令两种方式说一下. github网站上操作 打开自己的仓库,进入code下面. 点击new pul ...