| Name |
Phase |
Title |
| Synthesis |
Avoid assigning a signal multiple times in the same sequential path. 避免同一时序路径下对一个信号进行多次赋值 |
|
| Chip |
Do not overload a wire with too many drivers. 禁止对同一信号线进行多源驱动 |
|
| Constraints general |
Do not have simultaneously active drivers on one signal. 同一个信号禁止同时有两个及以上有效驱动器 |
|
| Chip |
Use tri-states to control bidirectional bus mode. 使用三态控制双向总线 |
|
| Chip |
Only tri-state elements should be connected to the bidirectional buses. 只有三态逻辑可连接到双向总线上 |
|
| Constraints clocks |
Do not use gated clocks (FPGA). 不适用门控时钟（FPGA） |
|
| Constraints clocks |
Isolate gated clocks to a separate clock generator instance. 门控时钟应独立设计为一个时钟生成模块 |
|
| Constraints clocks |
Gated clocks can be used only at top level (ASIC). 只能在顶层使用门控时钟（ASIC） |
|
| Constraints clocks |
Do not use flip-flop output as a clock. 不将触发器的输出作为时钟使用 |
|
| Constraints clocks |
Do not connect clocks to anything other than flip-flop clock pins. 仅将时钟信号连接到触发器的时钟引脚 |
|
| Constraints resets |
Do not connect resets to anything other than flip-flop reset pins. 仅将复位信号连接至触发器的复位引脚 |
|
| Chip |
Do not use combinatorial logic in reset lines. 复位信号链路中不使用组合逻辑 |
|
| Constraints resets |
Avoid internally generated resets. 避免使用内部产生的复位信号 |
|
| Constraints resets |
Place the reset generator instance at the top-level of design hierarchy. 在顶层例化复位信号产生模块 |
|
| Constraints resets |
A flip-flop inference should have one asynchronous control. 触发器应具有异步控制端口 |
|
| Constraints resets |
Do not use the same signal as clock and reset. 不将一个信号同时用做时钟信号和复位信号 |
|
| Synthesis |
Do not use resets with mixed polarity. 复位信号使用单一极性 |
|
| Synthesis |
All registers should have a reset control. 所有寄存器都应具有复位控制端口 |
|
| Parse |
Use the 'std_logic_1164' standard package whenever it is possible. 尽量使用‘std_logic_1164’库 |
|
| Parse |
Use the 'numeric_std' package instead of 'std_logic_arith' and 'std_logic_unsigned' packages. 使用'numeric_std' 尽量不使用'std_logic_arith' 和 'std_logic_unsigned' |
|
| Elaboration |
Use 'std_logic' only for I/O ports. 仅对I/O端口使用 'std_logic' |
|
| Parse |
Do not use unsafe data types inside architecture. 在结构体不使用不安全的数据类型 |
|
| Parse |
Specify range for objects of 'integer' data type. 'integer'数据类型对象须声明范围 |
|
| Parse |
Do not use the 'bit' data type. 禁止使用 'bit'数据类型 |
|
| Parse |
Do not use the 'enum_encoding' attribute for enumerations. 枚举对象禁止使用'enum_encoding'属性 |
|
| Parse |
Specify range for interface objects of 'std_logic_vector' data type. ' std_logic_vector '数据类型对象须声明位宽 |
|
| Parse |
Use descending range for one-dimensional objects. 使用降序声明一维数据对象的范围 |
|
| Elaboration |
Specify '0' as least significant bit for one-dimensional objects. 使用‘0’表示一维数据对象的最低位 |
|
| Parse |
Use simple signals only for non-constant indexing. 数据的非常量索引应使用简单的信号表示（索引不能使用表达式） |
|
| Parse |
All declared objects should be used in the description. 声明的所有对象都应被使用 |
|
| Elaboration |
All declared signals should be used in the description. 声明的所有信号都应被使用 |
|
| Parse |
Use constants or generics instead of hard-coded numbers. 使用常量或者参量代替数字 |
|
| Parse |
Avoid Hard-Coded vector reset Assignments. 避免在复位中直接使用数字矢量（将复位值定义为常量或者参量） |
|
| Parse |
Do not use ordered port connections. 端口映射禁止使用顺序映射 |
|
| Parse |
Match port and connected signal names. 匹配信号名称及所连接的端口名称 |
|
| Parse |
Avoid unconnected input ports in component instance. 例化时，避免模块输入端口悬空 |
|
| Parse |
Avoid unconnected output ports in component instance. 例化时，避免模块输出端口悬空 |
|
| Constraints clocks |
Do not use a clock signal as a data. 禁止将时钟信号用为数据 |
|
| Elaboration |
Avoid mismatching in assignment ranges. 避免赋值两端范围不匹配 |
|
| Elaboration |
Match bit widths of relational operator arguments. 条件运算符两边的数据位宽应匹配 |
|
| Elaboration |
Bit widths of component port and bind signal should match. 例化时，信号及其连接的端口数据位宽应匹配 |
|
| Synthesis |
Define all the necessary signals in the sensitivity list. 敏感列表包含所有必需的信号 |
|
| Synthesis |
Do not define unnecessary objects in the sensitivity list. 禁止敏感列表中包含非必需的信号 |
|
| Synthesis |
Do not use a signal within the same process statement it was assigned. 不在信号赋值的过程中使用该信号 |
|
| Parse |
Do not read global signals in a function body. 函数中不适用全局信号 |
|
| Synthesis |
Function return value should be defined in all the possible cases. 函数的返回值在任何条件下都应是确定的 |
|
| Parse |
A function should return at the end of its body. 函数应在其函数体的结尾处返回 |
|
| Parse |
Function should not return in conditional branches. 禁止函数在条件语句分支中返回 |
|
| Synthesis |
Use proper encoding type for FSM states. 有限状态机使用合适的编码 |
|
| Synthesis |
Use constants to define FSM state vectors. 使用常量定义有限状态机的向量 |
|
| Synthesis |
Each FSM should have reset control. 每一个有限状态机都应具有复位控制 |
|
| Synthesis |
All illegal or undefined states should transition to a defined state in a case of state corruption. 有限状态机应具备从非法或未定义状态中自动跳转到定义的状态中 |
|
| Synthesis |
Avoid unreachable states in FSM descriptions. 避免有限状态机中包含不可达状态 |
|
| Synthesis |
Avoid deadlock states in FSM descriptions. 避免有限状态机中死锁状态 |
|
| Chip |
All FSM inputs should be synchronous to the state machine clock. 所有有限状态机的输入都应利用其主时钟进行同步后使用 |
|
| Chip |
Port description order should follow a pattern. 端口定义顺序应服从某一模式 |
|
| Parse |
Do not describe multiple statements in one line. 禁止一行中包含多条语句 |
|
| Parse |
Do not describe multiple association elements in one line. 禁止一行中包含多个元素 |
|
| Parse |
Do not describe multiple declarations per line. 禁止一行中包含多个声明 |
|
| Parse |
Label statements whenever it is possible. 尽量对语句进行标识 |
|
| Parse |
Always use ending label or constructs name at the end of the statements and declarations. 在语句和声明的结尾处使用标识 |
|
| Parse |
Do not use the same names for different objects. 禁止对不同对象使用同一名称 |
|
| Parse |
Indent each level of the source code. 代码逐层缩进 |
|
| Parse |
Use spaces instead of tabs. 使用空格，而不使用占位符 |
|
| Parse |
Avoid large design files. 避免单一设计过大 |
|
| Parse |
Describe one design unit per file. 每一个设计文件仅包含一个设计模块 |
|
| Parse |
Ensure consistent file header. 确保一致的文件头注释 |
|
| Parse |
Ensure sufficient comment density. 确保充足的注释 |
|
| Parse |
Ensure commenting of design constructs. 确保设计结构具有注释 |
|
| Parse |
Do not describe same objects using both upper and lower letter case. 禁止同一对象一处使用大写，在另一处使用小写（同一对象名称应大小写保持不变） |
|
| Elaboration |
Do not use multiple waveform and optional delay expression in assignments. 禁止在赋值中使用多个波形和可选的延迟表达式 |
|
| Parse |
Avoid using records in RTL description. 禁止在寄存器传输级设计中使用records |
|
| Synthesis |
Do not use register initialization assignments. 禁止对寄存器定义时赋初值 |
|
| Synthesis |
Avoid Undriven Signals. 禁止未经驱动的信号 |
|
| Constraints general |
Avoid long combinational paths. 禁止长组合逻辑路径 |
|
| Parse |
Limit the use of nested 'if' and 'case' statements. 限制if和case语句的嵌套级数 |
|
| Parse |
Case statement should always have 'others' choice. Case语句必须包含others分支 |
|
| Synthesis |
Do not describe unreachable conditions. 禁止包含不可达条件分支 |
|
| Parse |
Do not use 'with select' statement. 禁止使用’with select’语句 |
|
| Chip |
Do not connect inout port directly to input/output ports. 禁止直接将inout端口连接至input或者output端口 |
|
| Constraints general |
Avoid delay chains in digital logic. 禁止在数字逻辑中包含延时 |
|
| Constraints general |
Associate each signal with only one non-tri-state driver. 每个信号仅使用一个非三态源进行驱动 |
|
| Constraints general |
Avoid feed-throughs in digital logic. 避免在数字逻辑中使用输入输出直连 |
|
| Constraints general |
Do not use combinatorial feedbacks. 禁止使用组合逻辑反馈 |
|
| Elaboration |
Do not describe multiple independent conditions in the process. 禁止在process中使用多个独立的条件 |
|
| Constraints general |
Avoid latches as much as possible. 尽量避免使用锁存器 |
|
| Chip |
Each register should be controllable from its inputs. 应能从外部输入引脚控制每个寄存器 |
|
| Synthesis |
Avoid using variables to infer sequential logic. 避免时序逻辑中使用变量 |
相关文章
- 10-09iOS基础 - Objective-C编码规范
- 10-09Python的编码规范（PEP 8 & Google Python guide）
- 10-09前端开发安全编码规范
- 10-09Python的编码规范
- 10-09Python编码规范（养成好的编码习惯很重要）
- 10-09戏说前端之CSS编码规范
- 10-09七、CSS简介及编码规范
- 10-09CSS编码规范
- 10-09基于神经网络视频编码的通用测试规范
- 10-09Python 编码规范 PEP8