UVM TLM（Transaction Level Modeling）1.0事务级建模介绍

大家好，欢迎来到IT知识分享网。

目录

1. 基本概念

1.1 TLM定义

1.2 put & get & transport 操作

1.3 PORT & EXPORT & IMP

1.4 ANALYSIS_PORT(EXPORT) & ANALYSIS_IMP

1.5 阻塞赋值与非阻塞赋值

2. UVM通信方式

2.1 一对一（PORT–EXPORT–IMP）单向端口通信

2.2 一对多（ANALYSIS_PORT–ANALYSIS_IMP）单向端口通信

2.3 FIFO双向通信

1. 基本概念

1.1 TLM定义

TLM（Transaction Level Modeling）是源于SystemC的一种通信标准，这种通信方式每次传输一个transaction，一个transaction是指将一组信息（如地址，数据等等）封装成一个包发送出去，参数如下代码。

class simple_trans extends uvm_sequence_item; rand data_t data; rand addr_t addr; rand enum {WRITE,READ} kind; constraint c1 { addr < 16’h2000; } ... endclass

1.2 put & get & transport 操作

put操作：通信动作发起者A将一个transaction发送给通信动作接收者B，A的端口称为PORT，B的端口称为EXPORT，数据流从A流向B；

get操作：通信动作发起者A向B索取一个transaction，数据流从B流向A；

transport操作：通信动作发起者A向B提交了通信请求，B返还给A一个transaction，数据从A流到B，再从B流向A，相当于一次put加get操作；

1.3 PORT & EXPORT & IMP

PORT端口：通信动作发起者的端口，有阻塞赋值和非阻塞赋值的区分，优先级高于EXPORT；常用的UVM PORT类型如下：

#参数T代表要传输的类型 uvm_blocking_put_port#(T); uvm_nonblocking_put_port#(T); uvm_put_port#(T); uvm_blocking_get_port#(T); uvm_nonblocking_get_port#(T); uvm_get_port#(T); uvm_blocking_peek_port#(T); uvm_nonblocking_peek_port#(T); uvm_peek_port#(T); uvm_blocking_get_peek_port#(T); uvm_nonblocking_get_peek_port#(T); uvm_get_peek_port#(T); uvm_blocking_transport_port#(REQ, RSP); uvm_nonblocking_transport_port#(REQ, RSP); uvm_transport_port#(REQ, RSP);

EXPORT端口： 通信动作中转方，有阻塞赋值和非阻塞赋值的区分，优先级高于IMP，常用的UVM EXPORT有：

#参数T代表要传输的类型 uvm_blocking_put_export#(T); uvm_nonblocking_put_export#(T); uvm_put_export#(T); uvm_blocking_get_export#(T); uvm_nonblocking_get_export#(T); uvm_get_export#(T); uvm_blocking_peek_export#(T); uvm_nonblocking_peek_export#(T); uvm_peek_export#(T); uvm_blocking_get_peek_export#(T); uvm_nonblocking_get_peek_export#(T); uvm_get_peek_export#(T); uvm_blocking_transport_export#(REQ, RSP); uvm_nonblocking_transport_export#(REQ, RSP); uvm_transport_export#(REQ, RSP);

IMP端口：通信动作接收方，有阻塞赋值和非阻塞赋值的区分，优先级最低，常用的UVM IMP有：

#参数T代表要传输的类型 uvm_blocking_put_imp#(T, IMP); uvm_nonblocking_put_imp#(T, IMP); uvm_put_imp#(T, IMP); uvm_blocking_get_imp#(T, IMP); uvm_nonblocking_get_imp#(T, IMP); uvm_get_imp#(T, IMP); uvm_blocking_peek_imp#(T, IMP); uvm_nonblocking_peek_imp#(T, IMP); uvm_peek_imp#(T, IMP); uvm_blocking_get_peek_imp#(T, IMP); uvm_nonblocking_get_peek_imp#(T, IMP); uvm_get_peek_imp#(T, IMP); uvm_blocking_transport_imp#(REQ, RSP, IMP); uvm_nonblocking_transport_imp#(REQ, RSP, IMP); uvm_transport_imp#(REQ, RSP, IMP);

只有高优先级的端口才可向低优先级端口发起上述三种操作，一个PORT可以连接到一个IMP，并发起三种操作，反之不行；15种PORT，EXPORT，IMP一一对应。

1.4 ANALYSIS_PORT(EXPORT) & ANALYSIS_IMP

analysis_port(export)和port端口类似，区别如下：

1）一个analysis_port可以连接多个analysis_imp；

2）没有阻塞赋值和非阻塞赋值的区分；

3）只有write一种通信操作；

1.5 阻塞赋值与非阻塞赋值

阻塞赋值（blocking）：通信动作发起方发起的request未被响应，会一直等待接收方的响应，直至request被响应才能发起下一个request，阻塞时会消耗时间；

非阻塞赋值（nonblocking）：通信动作发起方不关心第一个request是否被响应，会继续发下一个request；

2. UVM通信方式

2.1 一对一（PORT–EXPORT–IMP）单向端口通信

通过connect函数建立起port-export-imp的关系，从而进行一对一端口通信；

只有通信动作发起方才可以调用connect函数；

imp端口必须作为通信的最后节点；

1）port-export-imp端口相连，在B的代码里必须定义put函数，否则会出错，代码如下：

//A.sv class A extends uvm_component; `uvm_component_utils(A) uvm_blocking_put_port#(my_transaction) A_port; ... endclass task A::main_phase(uvm_phase phase); my_transaction tr; repeat(10) begin #10; tr = new("tr"); assert(tr.randomize()); A_port.put(tr); end endtask //B.sv class B extends uvm_component; `uvm_component_utils(B) uvm_blocking_put_export#(my_transaction) B_export; uvm_blocking_put_imp#(my_transaction, B) B_imp; ... endclass function void B::connect_phase(uvm_phase phase); super.connect_phase(phase); B_export.connect(B_imp); endfunction function void B::put(my_transaction tr); `uvm_info("B", "receive a transaction", UVM_LOW) tr.print(); endfunction //my_env.sv class my_env extends uvm_env; A A_inst; B_inst; ... endclass virtual function void build_phase(uvm_phase phase); A_inst = A::type_id::create("A_inst", this); B_inst = B::type_id::create("B_inst", this); endfunction ... function void my_env::connect_phase(uvm_phase phase); super.connect_phase(phase); A_inst.A_port.connect(B_inst.B_export); endfunction

2）port-imp端口相连，根据A.sv种PORT端口的类型决定在B.sv种定义哪种put/get函数，详见下表

参考代码如下（uvm_blocking_put类型）：

//A.sv和my_env.sv的代码保持不变 //B.sv class B extends uvm_component; `uvm_component_utils(B) uvm_blocking_put_imp#(my_transaction, B) B_imp; ... endclass function void B::put(my_transaction tr); `uvm_info("B", "receive a transaction", UVM_LOW) tr.print(); endfunction

2.2 一对多（ANALYSIS_PORT–ANALYSIS_IMP）单向端口通信

1）analysis_port支持一对一，一对多通信，需要在B.sv中定义一个write函数，示例代码如下：

//A.sv class A extends uvm_component; `uvm_component_utils(A) uvm_analysis_port#(my_transaction) A_ap; ... endclass task A::main_phase(uvm_phase phase); my_transaction tr; repeat(10) begin #10; tr = new("tr"); assert(tr.randomize()); A_ap.write(tr); end endtask //B.sv class B extends uvm_component; `uvm_component_utils(B) uvm_analysis_imp#(my_transaction, B) B_imp; ... endclass function void B::write(my_transaction tr); `uvm_info("B", "receive a transaction", UVM_LOW) tr.print(); endfunction //my_env.sv class my_env extends uvm_env; A A_inst; B_inst; ... endclass virtual function void build_phase(uvm_phase phase); A_inst = A::type_id::create("A_inst", this); B_inst = B::type_id::create("B_inst", this); endfunction ... function void my_env::connect_phase(uvm_phase phase); super.connect_phase(phase); A_inst.A_ap.connect(B_inst.B_imp); A_inst.A_ap.connect(C_inst.C_imp);//一个port连接多个imp，C的定义和B类似 endfunction 

2） 一个component有多个IMP，使用imp_decl，示例如下：

//my_scoreboard.sv `uvm_analysis_imp_decl(_monitor) `uvm_analysis_imp_decl(_model) class my_scoreboard extends uvm_scoreboard; my_transaction expect_queue[$]; uvm_analysis_imp_monitor#(my_transaction, my_scoreboard) monitor_imp; uvm_analysis_imp_model#(my_transaction, my_scoreboard) model_imp; ... extern function void write_monitor(my_transaction tr); extern function void write_model(my_transaction tr); extern virtual task main_phase(uvm_phase phase); endclass function void my_scoreboard::write_model(my_transaction tr); expect_queue.push_back(tr); endfunction function void my_scoreboard::write_monitor(my_transaction tr); my_transaction tmp_tran; bit result; if(expect_queue.size() > 0) begin ... end endfunction

2.3 FIFO双向通信

待更新