RTL Compiler之synthesis steps
1 synthesis steps
1) Search Paths
rc:/> set_attribute lib_search_path path /
rc:/> set_attribute script_search_path path /
rc:/> set_attribute hdl_search_path path /
2) Target Technology Library
rc:/> set_attribute library lib_name.lib
- Target Library: Design Compiler uses the target library to build a circuit. During mapping, Design Compiler selects functionally correct gates from the target library. It also calculates the timing of the circuit, using the vendor-supplied timing data for these gates.
3) Loading the HDL Files
- sequentially: read_hdl file1.v read_hdl file2.v read_hdl file3.v
- simultaneously: read_hdl { file1.v file2.v file3.v }
read_hdl [-v1995 | -v2001 | -sv
| -vhdl [-library library_name ]
| -netlist]
[-define macro=value ] ... file_list
4) Performing Elaboration
This is only required for top-level design. The elaborate command automatically elaborates the top-level design and all of its references.
5) Applying Constraints
- Operating conditions; clock waveforms; I/O timing
6) Applying Optimization Constraints
7) Performing Synthesis
rc:\> synthesize -to_mapped
8) Analyzing the Synthesis Results
- generate a detailed area report, use report area
- generate a detailed gate selection and area report, use report gates
- generate a detailed timing report, including the worst critical path of the current design, use report timing
9) Exporting the Design
- gate-level netlist rc:/> write_hdl > design.v
- design constraints rc:/> write_script > constraints.g
- constraints in SDC format rc:/> write_sdc > constraints.sdc
10) Exiting RTL Compiler
quit or exit
2 recommended flow
#general setup
#--------------
set_attributer lib_search_path ...
set_attribute hdl_search_path ..
#load the library
#------------------------------
set_attribute library library_name
#load and elaborate the design
#------------------------------
read_hdl design.v
elaborate
#specify timing and design constraints
#--------------------------------------
read_sdc sdc_file
# add optimization constraints
#----------------------------
.....
#synthesize the design
#---------------------
synthesize -to_mapped
#analyze design
------------------
report area
report timing
report gates
#export design
#-------------
write_hdl > dessign.vm
write_sdc > constraints.sdc
write_script > constraints.g
# export design for EDI
#-----------------------
write_design [-basename string ] [-gzip_files] [-tcf]
[-encounter] [-hierarchical] [ design ]
3 working in the top-level(root) directory
Root is a special object that contains all other objects represented as a ‘tree’ underneath it.
The root object is always present in RTL Compiler and is represented by a “/”, as shown in above Figure.
Root attributes contain information about all loaded designs.
RTL Compiler之synthesis steps的更多相关文章
- RTL Compiler之synthesis flow
1 generic RTL Compiler work flow 2 invoking RTL compiler RTL Compiler is invoked from the operating ...
- RTL Compiler之Example
Synthesis = Translation + Logic Optimization + Mapping Step 1 Source files 1) make directory mkdir ...
- RTL Compiler之Technology Library
1 Target Library Design Compiler uses the target library to build a circuit. During mapping, Design ...
- 行为级和RTL级的区别(转)
转自:http://hi.baidu.com/renmeman/item/5bd83496e3fc816bf14215db RTL级,registertransferlevel,指的是用寄存器这一级别 ...
- Verilog的数据流、行为、结构化与RTL级描述
Verilog语言可以有多种方式来描述硬件,同时,使用这些描述方式,又可以在多个抽象层次上设计硬件,这是Verilog语言的重要特征. 在Verilog语言中,有以下3种最基本的描述方式: 数据流描述 ...
- RTL 与 technology schematic的区别,包含概念与实例
2013-06-25 16:40:45 下面是xilinx官网上的问答贴: http://china.xilinx.com/support/answers/41500.htm#solution The ...
- 1.2 the structure of a compiler
Compiler 1.2 the structure of a compiler Compiler : analysis and synthesis syntactically 语法上的 sema ...
- Tcl之Read files for synthesis
The following file is to read all design files into syntehsis tool automatically, like Cadence RTL C ...
- backend flow
在PD之后,netlist中会多出很多DCAP元件(去耦电容,减少IR-Drop)或者filter cell(保证芯片均匀度要求) 还有一些antenna cell也就是一些diode用来泻流,防止天 ...
随机推荐
- 21、Java并发性和多线程-Java中的锁
以下内容转自http://ifeve.com/locks/: 锁像synchronized同步块一样,是一种线程同步机制,但比Java中的synchronized同步块更复杂.因为锁(以及其它更高级的 ...
- NYOJ 16 矩形嵌套 (DAG上的DP)
矩形嵌套 时间限制:3000 ms | 内存限制:65535 KB 难度:4 描写叙述 有n个矩形,每个矩形能够用a,b来描写叙述.表示长和宽.矩形X(a,b)能够嵌套在矩形Y(c,d)中当且仅当 ...
- [Vue @Component] Pass Props to Vue Functional Templates
Functional templates allow you to create components consisting of only the template tag and exposing ...
- 从打击App刷榜看苹果的底线
这两天苹果打击App刷榜者的消息刷屏了,从腾讯科技.appying多个媒体渠道看到,<安居客>.<友秘>.<微在>.<秦时明月2>.<悟空与貂蝉& ...
- Linux Framebuffer驱动剖析之中的一个—软件需求
嵌入式企鹅圈将以本文作为2015年的终结篇,以回应第一篇<Linux字符设备驱动剖析>.嵌入式企鹅圈一直专注于嵌入式Linux和物联网IOT双方面的原创技术分享,稍后会公布嵌入式企鹅圈的2 ...
- 【Swift】学习笔记(八)——类和结构体
1.类和结构体的定义 <pre name="code" class="html">struct Resolution { var width = 0 ...
- 快速编译system.img和boot.img的方法【转】
本文转载自:http://www.cnblogs.com/wanqieddy/archive/2012/10/22/2734024.html 快速编译system.img,可以使用这个命令: #mak ...
- WPF,Silverlight与XAML读书笔记(3) - 标记扩展
hystar的.Net世界 博客园 首页 新闻 新随笔 联系 管理 订阅 随笔- 103 文章- 0 评论- 107 WPF,Silverlight与XAML读书笔记(3) - 标记扩展 说 ...
- P2597 [ZJOI2012]灾难 拓扑排序
这个题有点意思,正常写法肯定会T,然后需要优化.先用拓扑排序重构一遍树,然后进行一个非常神奇的操作:把每个点放在他的食物的lca上,然后计算的时候直接dfs全加上就行了.为什么呢,因为假如你的食物的l ...
- B1877 [SDOI2009]晨跑 费用流
其实之前写过一个板子,但是一点印象都没有,所以今天重写了一下,顺便把这个题当成板子就行了. 其实费用流就是把bfs换成spfa,但是中间有一个原则,就是费用优先,在费用(就是c)上跑spfa,顺便求出 ...