PostgreSQL 14在MVCC上面有所增強，以保證在大批量客戶端連線場景下效能不會出現顯著下降（詳細可參見參考資料中的連結）。本節介紹了提交記錄“Don't compute global horizons while building snapshots.”中heapam.c中的修改以及涉及的相關資料結構。

### 一、資料結構
**GlobalVisState**
```C
/*
 * State for the GlobalVisTest* family of functions. Those functions can
 * e.g. be used to decide if a deleted row can be removed without violating
 * MVCC semantics: If the deleted row's xmax is not considered to be running
 * by anyone, the row can be removed.
 * 用於配合GlobalVisTest*函式族使用的資料結構。
 * 這些函式用於確定刪除的行在不違反MVCC語義的情況下是否可以被清除：
 * 如果被刪除的行的xmax與所有執行的事務無關，那麼可以清除。
 * 
 * To avoid slowing down GetSnapshotData(), we don't calculate a precise
 * cutoff XID while building a snapshot (looking at the frequently changing
 * xmins scales badly). Instead we compute two boundaries while building the
 * snapshot:
 *
 * 1) definitely_needed, indicating that rows deleted by XIDs >=
 *    definitely_needed are definitely still visible.
 *
 * 2) maybe_needed, indicating that rows deleted by XIDs < maybe_needed can
 *    definitely be removed
 *  
 *  為了避免影響GetSnapshotData()的效能，在構建快照時我們不會計算精確的cutoff XID，
 *  相反，我們會在構建快照時計算兩個邊界，分別是：
 *  1）definitely_needed，表示如果刪除事務的XIDs >= definitely_needed，則可見
 *  2）maybe_needed，表示被XIDs < maybe_needed 事務刪除的，可以確定被清除
 *
 * When testing an XID that falls in between the two (i.e. XID >= maybe_needed
 * && XID < definitely_needed), the boundaries can be recomputed (using
 * ComputeXidHorizons()) to get a more accurate answer. This is cheaper than
 * maintaining an accurate value all the time.
 * 對於XID >= maybe_needed && XID < definitely_needed的XID，
 * 邊界需要使用函式ComputeXidHorizons重新計算來獲取更精確的結果，這比在所有情況下都維護一個精確值成本要低
 *
 * As it is not cheap to compute accurate boundaries, we limit the number of
 * times that happens in short succession. See GlobalVisTestShouldUpdate().
 * 計算精確的邊界成本高昂，因此需要限制短時間內出現的次數，詳細參見GlobalVisTestShouldUpdate函式
 *
 * There are three backend lifetime instances of this struct, optimized for
 * different types of relations. As e.g. a normal user defined table in one
 * database is inaccessible to backends connected to another database, a test
 * specific to a relation can be more aggressive than a test for a shared
 * relation.  Currently we track four different states:
 * 對於不同的關係，出於優化的目的，該結構有三種不同的backend生命週期。
 * 比如某個資料庫中使用者定義的普通表對於連線到另外一個資料庫的backends是不可見的，
 * 對某個特定關係的驗證比起共享關係的驗證要更為積極。我們跟蹤四種不同的狀態：
 *
 * 1) GlobalVisSharedRels, which only considers an XID's
 *    effects visible-to-everyone if neither snapshots in any database, nor a
 *    replication slot's xmin, nor a replication slot's catalog_xmin might
 *    still consider XID as running.
 * 1）GlobalVisSharedRels，事務對該Relation的操作會影響所有資料庫中的快照、複製槽的xmin和複製槽的catalog_xmin。
 *
 * 2) GlobalVisCatalogRels, which only considers an XID's
 *    effects visible-to-everyone if neither snapshots in the current
 *    database, nor a replication slot's xmin, nor a replication slot's
 *    catalog_xmin might still consider XID as running.
 *
 *    I.e. the difference to GlobalVisSharedRels is that
 *    snapshot in other databases are ignored.
 * 2）GlobalVisCatalogRels，事務對該Relation的操作會影響當前資料庫中的快照、複製槽的xmin和複製槽的catalog_xmin
 *    與GlobalVisSharedRels不同之處是其他資料庫的快照會被忽略
 *
 * 3) GlobalVisDataRels, which only considers an XID's
 *    effects visible-to-everyone if neither snapshots in the current
 *    database, nor a replication slot's xmin consider XID as running.
 *
 *    I.e. the difference to GlobalVisCatalogRels is that
 *    replication slot's catalog_xmin is not taken into account.
 *
 * 3）GlobalVisDataRels，事務對該Relation的操作會影響當前資料庫中的快照、複製槽的xmin
 *    與GlobalVisCatalogRels不同的是catalog_xmin不作考慮
 *
 * 4) GlobalVisTempRels, which only considers the current session, as temp
 *    tables are not visible to other sessions.
 *
 * 4）GlobalVisTempRels，只考慮當前會話，臨時表對其他session不可見
 *
 * GlobalVisTestFor(relation) returns the appropriate state
 * for the relation.
 * GlobalVisTestFor(relation)返回關係正確的狀態
 *
 * The boundaries are FullTransactionIds instead of TransactionIds to avoid
 * wraparound dangers. There e.g. would otherwise exist no procarray state to
 * prevent maybe_needed to become old enough after the GetSnapshotData()
 * call.
 * 邊界使用FullTransactionIds而不是TransactionIds以避免因為回捲而導致的錯誤。
 * 例如，在GetSnapshotData()呼叫後，將不存在procarray狀態來防止maybe_needed變得32bit XID中存在的過老現象。
 *
 * The typedef is in the header.
 */
struct GlobalVisState
{
    /* XIDs >= are considered running by some backend */
    //大於等於該值的XIDs：其他活動backend可見
    FullTransactionId definitely_needed;

/* XIDs < are not considered to be running by any backend */
    //小於該值的XIDs：活動backend不可見
    FullTransactionId maybe_needed;
};

/*
 * A 64 bit value that contains an epoch and a TransactionId.  This is
 * wrapped in a struct to prevent implicit conversion to/from TransactionId.
 * Not all values represent valid normal XIDs.
 * 號外：PG 14已支援64 bit XID，令人厭煩的XID回捲問題不復存在
 */
typedef struct FullTransactionId
{
    uint64      value;
} FullTransactionId;

```

### 二、原始碼解讀

**diff heapam.c**
修改heapam.c，目的是去掉RecentGlobalXmin的訪問，改為通過GlobalVisXXX配合GlobalVisState Struct實現類似的功能
```Diff
--- a/src/backend/access/heap/heapam.c
+++ b/src/backend/access/heap/heapam.c
@@ -1517,6 +1517,7 @@ heap_hot_search_buffer(ItemPointer tid, Relation relation, Buffer buffer,
    bool        at_chain_start;
    bool        valid;
    bool        skip;
+   GlobalVisState *vistest = NULL;//新引入的資料結構，全域性訪問狀態，詳見GlobalVisState Struct
 
    /* If this is not the first call, previous call returned a (live!) tuple */
    if (all_dead)
@@ -1527,7 +1528,8 @@ heap_hot_search_buffer(ItemPointer tid, Relation relation, Buffer buffer,
    at_chain_start = first_call;
    skip = !first_call;
 
-   Assert(TransactionIdIsValid(RecentGlobalXmin));
+   /* XXX: we should assert that a snapshot is pushed or registered */
+   Assert(TransactionIdIsValid(RecentXmin));//去掉全域性變數RecentGlobalXmin的訪問，轉而確認快照入棧和註冊
    Assert(BufferGetBlockNumber(buffer) == blkno);
 
    /* Scan through possible multiple members of HOT-chain */
@@ -1616,9 +1618,14 @@ heap_hot_search_buffer(ItemPointer tid, Relation relation, Buffer buffer,
         * Note: if you change the criterion here for what is "dead", fix the
         * planner's get_actual_variable_range() function to match.
         */
-       if (all_dead && *all_dead &&
-           !HeapTupleIsSurelyDead(heapTuple, RecentGlobalXmin))
-           *all_dead = false;
+       if (all_dead && *all_dead)//去掉與RecentGlobalXmin相關的程式碼，改為從全域性訪問狀態結構體中獲取資訊
+       {
+           if (!vistest)
+               vistest = GlobalVisTestFor(relation);//在後續介紹
+
+           if (!HeapTupleIsSurelyDead(heapTuple, vistest))
+               *all_dead = false;
+       }
 
        /*
         * Check to see if HOT chain continues past this tuple; if so fetch
```

**GlobalVisTestFor**
```C
/*
 * If rel != NULL, return test state appropriate for relation, otherwise
 * return state usable for all relations.  The latter may consider XIDs as
 * not-yet-visible-to-everyone that a state for a specific relation would
 * already consider visible-to-everyone.
 * 如果rel有效（不為NULL），返回對應該關係的驗證狀態，否則返回對所有關係均可用的狀態。
 * 後續可能會認為XIDs為對所有仍未可見的狀態，該狀態對於某個指定的關係可能已經認為對所有事務可見。
 *
 * This needs to be called while a snapshot is active or registered, otherwise
 * there are wraparound and other dangers.
 * 該例程在快照活動或已註冊時呼叫，否則存在回捲或其他風險。
 *
 * See comment for GlobalVisState for details.
 * 詳細可參見GlobalVisState
 */
GlobalVisState *
GlobalVisTestFor(Relation rel)
{
    bool        need_shared;//共享？
    bool        need_catalog;//資料字典？
    GlobalVisState *state;//訪問狀態

/* XXX: we should assert that a snapshot is pushed or registered */
    //快照已入棧或註冊
    Assert(RecentXmin);

if (!rel)
        need_shared = need_catalog = true;//rel為NULL，則均值為true
    else
    {
        /*
         * Other kinds currently don't contain xids, nor always the necessary
         * logical decoding markers.
         * 還沒有包含xids的其他型別，或者通常是邏輯解碼標記器
         */
        Assert(rel->rd_rel->relkind == RELKIND_RELATION ||
               rel->rd_rel->relkind == RELKIND_MATVIEW ||
               rel->rd_rel->relkind == RELKIND_TOASTVALUE);

need_shared = rel->rd_rel->relisshared || RecoveryInProgress();//共享的關係（跨資料庫）或者正處於恢復過程
        need_catalog = IsCatalogRelation(rel) || RelationIsAccessibleInLogicalDecoding(rel);//資料字典表或者邏輯解碼可訪問的關係
    }

if (need_shared)
        state = &GlobalVisSharedRels;//全域性共享
    else if (need_catalog)
        state = &GlobalVisCatalogRels;//當前資料庫中的Catalog
    else if (RELATION_IS_LOCAL(rel))
        state = &GlobalVisTempRels;//臨時表
    else
        state = &GlobalVisDataRels;//常規表

Assert(FullTransactionIdIsValid(state->definitely_needed) &&
           FullTransactionIdIsValid(state->maybe_needed));

return state;
}

```

****
```

```

### 三、跟蹤分析
N/A

### 四、參考資料
1.[Improving connection scalability: GetSnapshotData()](https://www.postgresql.org/message-id/20200301083601.ews6hz5dduc3w2se%40alap3.anarazel.de)
2.[snapshot scalability: Don't compute global horizons while building snapshots.](https://git.postgresql.org/gitweb/?p=postgresql.git;a=commitdiff;h=dc7420c2c9274a283779ec19718d2d16323640c0)

PostgreSQL 原始碼解讀（249）- PG 14（Improving connection scalability）#1.md

相關文章