日志

uvm_sequence_base (2011-10-24 16:26:00)

已有 1790 次阅读| 2015-3-25 11:24 |个人分类:uvm

转

liu_uestc的博客

的一篇文章

先抄过来再看。

uvm_sequence_base

(2011-10-24 16:26:00)

转载▼

标签：

it

分类： uvm

preface

uvm_sequence_base是一个比较关键的类，该类实现了一个item的构成的流;同时该类本省也继承自item,这样该类也可以包含若干uvm_sequence_base构成的流。

一个sequence的执行大致分为几种方法，这些方法都直接或者间接地调用了<start_item>/<finish_item> 或者调用宏`uvm_do_* macros。下边把这几种方法简单介绍下：

1.Executing sequences via <start>:

   A sequence's <start> method has a ~parent_sequence~ argument that controls
   whether <pre_do>, <mid_do>, and <post_do> are called *in the parent*
   sequence. It also has a ~call_pre_post~ argument that controls whether its
    <pre_body> and <post_body> methods are called.
  The sequence execution flow looks like

   User code
   sub_seq.randomize(...); // optional
   sub_seq.start(seqr, parent_seq, priority, *call_pre_post*)

   The following methods are called, in order
   sub_seq.pre_body           (task)  if call_pre_post==1
     parent_seq.pre_do(0)     (task)  if parent_sequence!=null
     parent_seq.mid_do(this)  (func)  if parent_sequence!=null
   sub_seq.body               (task)  YOUR STIMULUS CODE
     parent_seq.post_do(this) (func)  if parent_sequence!=null
   sub_seq.post_body          (task)  if call_pre_post==1

2.Executing sub-sequences via <start_item>/<finish_item> or `uvm_do macros:

   A sequence can also be indirectly started as a child in the <body> of a
parent sequence. The child sequence's <start> method is called indirectly
via calls to its <start_item>/<finish_item> methods or by invoking
any of the `uvm_do macros. Child sequences can also be started by
the predefined sequences, <uvm_random_sequence> and
<uvm_exhaustive_sequence>. In all these cases, <start> is called with
~call_pre_post~ set to 0, preventing the started sequence's <pre_body> and
<post_body> methods from being called. During execution of the
child sequence, the parent's <pre_do>, <mid_do>, and <post_do> methods
are called

  User code

| parent_seq.start_item(sub_seq, priority);
| sub_seq.randomize(...);
| parent_seq.finish_item(sub_seq);
|
| or
|
| `uvm_do_with_prior(seq_seq, { constraints }, priority)
|

The following methods are called, in order

|
|   parent_seq.pre_do(0)        (task)
|   parent_req.mid_do(sub_seq)  (func)
|     sub_seq.body              (task)
|   parent_seq.post_do(sub_seq) (func)

Remember, it is the *parent* sequence's pre|mid|post_do that are called, not
the sequence being executed.
3. Executing sequence items via <start_item>/<finish_item> or `uvm_do macros:

  Items are started in the <body> of a parent sequence via calls to
<start_item>/<finish_item> or invocations of any of the `uvm_do
macros. The <pre_do>, <mid_do>, and <post_do> methods of the parent
sequence will be called as the item is executed.
User code

| parent_seq.start_item(item, priority);
| sub_seq.randomize(...) [with {constraints}];
| parent_seq.finish_item(item);
|
| or
|
| `uvm_do_with_prior(item, constraints, priority)
|

The following methods are called, in order

|
|   sequencer.wait_for_grant(prior) (task) \ start_item  \
|   parent_seq.pre_do(1)            (task) /              \
|                                                      `uvm_do* macros
|   parent_seq.mid_do(item)         (func) \              /
|   sequencer.send_request(item)    (func)  \finish_item /
|   sequencer.wait_for_item_done()  (task)  /
|   parent_seq.post_do(item)        (func) /
4.属性

  protected uvm_sequence_state m_sequence_state;//枚举，查看base globals
            int                m_next_transaction_id = 1;//下一个数据的id
  local     int                m_priority = -1;
            int                m_tr_handle;//在recorder中的句柄
            int                m_wait_for_grant_semaphore;//记录的是发起了多少个发送数据的请求

// Each sequencer will assign a sequence id. When a sequence is talking to multiple
// sequencers, each sequence_id is managed seperately

//一个sequence在sequencer中注册一次就会分配一个全局的sequence_id
protected int m_sqr_seq_ids[int]

  protected uvm_sequence_item response_queue[$];//存储来自driver的响应
  protected int               response_queue_depth = 8;
  protected bit               response_queue_error_report_disabled = 0;//比如driver返回的响应在队列中溢出了打印报告，这个就是来控制打印不打印的

  rand bit is_randomized = 0;
  constraint c_randomized {
    is_randomized == 1;
  }

protected process m_sequence_process;//squence的进程
local bit m_use_response_handler = 0;

static string type_name = "uvm_sequence_base";

  // bits to detect if is_relevant()/wait_for_relevant() are implemented
  local bit is_rel_default;
  local bit wait_rel_default;
5. function new (string name = "uvm_sequence");//设置sequence的状态为m_sequence_state = CREATED;发出的传输数据的请求为m_wait_for_grant_semaphore = 0

6.virtual function bit is_item();返回零，表示不是item，妈的，这就是通过属性实现多态性的余孽

7.function uvm_sequence_state_enum get_sequence_state();

8.task wait_for_sequence_state(uvm_sequence_state_enum state);等待m_sequence_state == state

  //--------------------------
  // Group: Sequence Execution
  //--------------------------

9.  virtual task start (uvm_sequencer_base sequencer,
                      uvm_sequence_base parent_sequence = null,
                      integer this_priority = 100,
                      bit call_pre_post = 1);

9.1 如果sequence还没执行完成，就报告出错

9.2 this_priority检查这个值设置是否违规

9.3 clear_response_queue();

9.4 把arguments中的值设置到属性

9.5 分结构化的sequence和单个sequence分别调用sequencer的记录器记录本实例

9.6 设置sequence_id

9.7 清空sequencer_id_sequence_id队列

9.8 在对应的sequencer中注册本sequence

9.9 开一个进程m_sequence_process = process::self();实现1中的调用过程

9.10 sequencer中的记录器结束记录

9.11 调用m_sequencer.m_sequence_exiting(this)执行

10. callbacks null

virtual task pre_body();

virtual task pre_do(bit is_item);

virtual function void mid_do(uvm_sequence_item this_item);

virtual task body();

virtual function void post_do(uvm_sequence_item this_item);

virtual task post_body();

11.uvm_phase starting_phase;在sequencer中的适当阶段本sequence中的start会被调用

  //------------------------
  // Group: Sequence Control
  //------------------------

12.function void set_priority (int value);

13.function int get_priority();

14.virtual function bit is_relevant();sequencer会对所有发出请求的sequence并且没有阻塞的sequence进行仲裁，仲裁前会检查is_relevent，如果这个返回0就会等待这个sequence的is_relevent返回1，在让这个sequence参加仲裁；

15. virtual task wait_for_relevant();参考14，被sequencer调用

16. task lock(uvm_sequencer_base sequencer = null);//让sequencer把本sequence锁住，就是暂时不要仲裁给本sequence

17.task grab(uvm_sequencer_base sequencer = null);//告诉sequencer本sequence应该放在总裁队列的grabers的末尾，相对优先的被仲裁

18.function void unlock(uvm_sequencer_base sequencer = null);

19.function void ungrab(uvm_sequencer_base sequencer = null);

20.function bit is_blocked();

21.function bit has_lock();

22. function void kill();对m_sequence_process的进程的操作

23.virtual function void do_kill(); null

24.function void m_kill();参考22

  //-------------------------------
  // Group: Sequence Item Execution
  //-------------------------------

25. protected function uvm_sequence_item create_item(uvm_object_wrapper type_var,
uvm_sequencer_base l_sequencer, string name);

//调用工厂创建一个uvm_sequence_item并对一些属性惊醒设置

26.   virtual task m_start_item_or_seq(uvm_sequencer_base sequencer,
                                   uvm_sequence_item parent_seq,
                                   int set_priority);//改写了item中的函数

27.  virtual task m_finish_item(uvm_sequencer_base sequencer,
                             uvm_sequence_item parent_seq,
                             int set_priority = -1);//改写item中的函数

28.virtual task wait_for_grant(int item_priority = -1, bit lock_request = 0);//调用sequencer中的本函数，该函数让sequence给sequencer发起一个发包请求，sequencer进行仲裁后返回，sequence得到返回后立即调用send_request。当然这些操作都是在sequencer中间接调用sequence中的东西来实现的

29. virtual function void send_request(uvm_sequence_item request, bit rerandomize = 0);//调用sequencer中的send_request实现，该函数中调用sequence中的函数，设置sequence产生的数据，如transaction_id然后传递给driver

30.virtual task wait_for_item_done(int transaction_id = -1);等待driver用完item给出反馈

// Group: Response API
//--------------------

31. function void use_response_handler(bit enable);

function bit get_use_response_handler();

32.virtual function void response_handler(uvm_sequence_item response);//null

33.function void set_response_queue_error_report_disabled(bit value);

function bit get_response_queue_error_report_disabled();

34.function void set_response_queue_depth(int value);

function int get_response_queue_depth();

35. virtual function void put_response (uvm_sequence_item response_item);

36. virtual task get_base_response(output uvm_sequence_item response, input int transaction_id = -1);//在response_queque中找到对应的东西

  //------------------------
  // Group- Sequence Library DEPRECATED
  //------------------------

37. rand int unsigned seq_kind;

38.function int get_seq_kind(string type_name);//调用sequencerm_sequencer.get_seq_kind(type_name)实现

39. function uvm_sequence_base get_sequence(int unsigned req_kind);//返回m_seq_type = m_sequencer.sequences[req_kind];

40. task do_sequence_kind(int unsigned req_kind);//该函数返回一个req_kind的sequence

41. function uvm_sequence_base get_sequence_by_name(string seq_name);//返回一个seq_name的sequence

42.task create_and_start_sequence_by_name(string seq_name);//create_and_start_sequence_by_name

43.function int m_get_sqr_sequence_id(int sequencer_id, bit update_sequence_id);

44.function void m_set_sqr_sequence_id(int sequencer_id, int sequence_id);

45.virtual function uvm_sequence_item create_request ();

46.virtual function uvm_sequence_item create_response ();

location:

C:\Documents and Settings\zhliu\Desktop\uvm-1.0p1\src\seq\uvm_sequence_base.svh

收藏分享邀请举报

全部作者的其他最新日志

发表评论评论 (1 个评论)

回复 luyuan_0922 2015-3-25 19:06: 排版还是不行，，算了，贴上原文链接： http://blog.sina.com.cn/s/blog_466496f30100ygis.html :lol :lol :lol

luyuan_0922

0
周排名
0
月排名
0
总排名
0
关注
4
粉丝
1
好友
1
获赞
11
评论
1002
访问数

luyuan_0922的个人空间 https://blog.eetop.cn/106870 [收藏] [复制] [分享] [RSS]

日志

uvm_sequence_base (2011-10-24 16:26:00)

liu_uestc的博客

uvm_sequence_base

uvm

sequence

sequence_base

it

全部作者的其他最新日志

发表评论评论 (1 个评论)

luyuan_0922

0

0

0

0

4

1

1

11

1002

站长推荐 /2

luyuan_0922的个人空间 https://blog.eetop.cn/106870 [收藏] [复制] [分享] [RSS]

日志

uvm_sequence_base (2011-10-24 16:26:00)

uvm_sequence_base

全部作者的其他最新日志

发表评论 评论 (1 个评论)

1

11

1002

站长推荐 /2

发表评论评论 (1 个评论)