snapshot scalability: Move PGXACT->xmin back to PGPROC.

Now that xmin isn't needed for GetSnapshotData() anymore, it leads to
unnecessary cacheline ping-pong to have it in PGXACT, as it is updated
considerably more frequently than the other PGXACT members.

After the changes in dc7420c2c92, this is a very straight-forward change.

For highly concurrent, snapshot acquisition heavy, workloads this change alone
can significantly increase scalability. E.g. plain pgbench on a smaller 2
socket machine gains 1.07x for read-only pgbench, 1.22x for read-only pgbench
when submitting queries in batches of 100, and 2.85x for batches of 100
'SELECT';.  The latter numbers are obviously not to be expected in the
real-world, but micro-benchmark the snapshot computation
scalability (previously spending ~80% of the time in GetSnapshotData()).

Author: Andres Freund <[email protected]>
Reviewed-By: Robert Haas <[email protected]>
Reviewed-By: Thomas Munro <[email protected]>
Reviewed-By: David Rowley <[email protected]>
Discussion: https://postgr.es/m/[email protected]

2 files changed, 8 insertions, 8 deletions

diff --git a/src/backend/replication/logical/snapbuild.c b/src/backend/replication/logical/snapbuild.c
index 3089f0d5ddc..e9701ea7221 100644
--- a/src/backend/replication/logical/snapbuild.c
+++ b/src/backend/replication/logical/snapbuild.c
@@ -553,8 +553,8 @@ SnapBuildInitialSnapshot(SnapBuild *builder)
 		elog(ERROR, "cannot build an initial slot snapshot, not all transactions are monitored anymore");
 
 	/* so we don't overwrite the existing value */
-	if (TransactionIdIsValid(MyPgXact->xmin))
-		elog(ERROR, "cannot build an initial slot snapshot when MyPgXact->xmin already is valid");
+	if (TransactionIdIsValid(MyProc->xmin))
+		elog(ERROR, "cannot build an initial slot snapshot when MyProc->xmin already is valid");
 
 	snap = SnapBuildBuildSnapshot(builder);
 
@@ -575,7 +575,7 @@ SnapBuildInitialSnapshot(SnapBuild *builder)
 	}
 #endif
 
-	MyPgXact->xmin = snap->xmin;
+	MyProc->xmin = snap->xmin;
 
 	/* allocate in transaction context */
 	newxip = (TransactionId *)
diff --git a/src/backend/replication/walsender.c b/src/backend/replication/walsender.c
index 460ca3f947f..3f756b470af 100644
--- a/src/backend/replication/walsender.c
+++ b/src/backend/replication/walsender.c
@@ -1964,7 +1964,7 @@ PhysicalReplicationSlotNewXmin(TransactionId feedbackXmin, TransactionId feedbac
 	ReplicationSlot *slot = MyReplicationSlot;
 
 	SpinLockAcquire(&slot->mutex);
-	MyPgXact->xmin = InvalidTransactionId;
+	MyProc->xmin = InvalidTransactionId;
 
 	/*
 	 * For physical replication we don't need the interlock provided by xmin
@@ -2093,7 +2093,7 @@ ProcessStandbyHSFeedbackMessage(void)
 	if (!TransactionIdIsNormal(feedbackXmin)
 		&& !TransactionIdIsNormal(feedbackCatalogXmin))
 	{
-		MyPgXact->xmin = InvalidTransactionId;
+		MyProc->xmin = InvalidTransactionId;
 		if (MyReplicationSlot != NULL)
 			PhysicalReplicationSlotNewXmin(feedbackXmin, feedbackCatalogXmin);
 		return;
@@ -2135,7 +2135,7 @@ ProcessStandbyHSFeedbackMessage(void)
 	 * risk already since a VACUUM could already have determined the horizon.)
 	 *
 	 * If we're using a replication slot we reserve the xmin via that,
-	 * otherwise via the walsender's PGXACT entry. We can only track the
+	 * otherwise via the walsender's PGPROC entry. We can only track the
 	 * catalog xmin separately when using a slot, so we store the least of the
 	 * two provided when not using a slot.
 	 *
@@ -2148,9 +2148,9 @@ ProcessStandbyHSFeedbackMessage(void)
 	{
 		if (TransactionIdIsNormal(feedbackCatalogXmin)
 			&& TransactionIdPrecedes(feedbackCatalogXmin, feedbackXmin))
-			MyPgXact->xmin = feedbackCatalogXmin;
+			MyProc->xmin = feedbackCatalogXmin;
 		else
-			MyPgXact->xmin = feedbackXmin;
+			MyProc->xmin = feedbackXmin;
 	}
 }