v2-0003-Add-cache-for-recomputeNamespacePath.patch

From 54b79e769c2e593b4d0deab1a19f5891af002d8f Mon Sep 17 00:00:00 2001
From: Jeff Davis <jeff@j-davis.com>
Date: Wed, 26 Jul 2023 12:55:09 -0700
Subject: [PATCH v2 3/3] Add cache for recomputeNamespacePath().

When search_path is changed to something that was previously set, and
no invalidation happened in between, use the cached list of namespace
OIDs rather than recomputing them. This avoids syscache lookups and
ACL checks.

Important when the search_path changes frequently, such as when set in
proconfig.
---
 src/backend/catalog/namespace.c | 282 +++++++++++++++++++++++++-------
 1 file changed, 219 insertions(+), 63 deletions(-)

diff --git a/src/backend/catalog/namespace.c b/src/backend/catalog/namespace.c
index f457f778cb..3be2317d70 100644
--- a/src/backend/catalog/namespace.c
+++ b/src/backend/catalog/namespace.c
@@ -41,6 +41,7 @@
 #include "catalog/pg_ts_template.h"
 #include "catalog/pg_type.h"
 #include "commands/dbcommands.h"
+#include "common/hashfn.h"
 #include "funcapi.h"
 #include "mb/pg_wchar.h"
 #include "miscadmin.h"
@@ -109,11 +110,12 @@
  * activeSearchPath is always the actually active path; it points to
  * to baseSearchPath which is the list derived from namespace_search_path.
  *
- * If baseSearchPathValid is false, then baseSearchPath (and other
- * derived variables) need to be recomputed from namespace_search_path.
- * We mark it invalid upon an assignment to namespace_search_path or receipt
- * of a syscache invalidation event for pg_namespace.  The recomputation
- * is done during the next lookup attempt.
+ * If baseSearchPathValid is false, then baseSearchPath (and other derived
+ * variables) need to be recomputed from namespace_search_path, or retrieved
+ * from the search path cache if there haven't been any syscache
+ * invalidations.  We mark it invalid upon an assignment to
+ * namespace_search_path or receipt of a syscache invalidation event for
+ * pg_namespace.  The recomputation is done during the next lookup attempt.
  *
  * Any namespaces mentioned in namespace_search_path that are not readable
  * by the current user ID are simply left out of baseSearchPath; so
@@ -153,6 +155,22 @@ static Oid	namespaceUser = InvalidOid;
 
 /* The above four values are valid only if baseSearchPathValid */
 static bool baseSearchPathValid = true;
+static bool searchPathCacheValid = false;
+static MemoryContext SearchPathCacheContext = NULL;
+
+typedef struct SearchPathCacheKey
+{
+	const char	*searchPath;
+	Oid			 roleid;
+} SearchPathCacheKey;
+
+typedef struct SearchPathCacheEntry
+{
+	SearchPathCacheKey	*key;
+	List				*oidlist;
+	bool				 temp_missing;
+	char				 status;
+} SearchPathCacheEntry;
 
 /*
  * myTempNamespace is InvalidOid until and unless a TEMP namespace is set up
@@ -193,6 +211,44 @@ static bool MatchNamedCall(HeapTuple proctup, int nargs, List *argnames,
 						   bool include_out_arguments, int pronargs,
 						   int **argnumbers);
 
+/*
+ * Recomputing the namespace path can be costly when done frequently, such as
+ * when a function has search_path set in proconfig. Add a cache that can be
+ * used when the search_path changes to a value that was previously set, and
+ * no invalidation intervened.
+ *
+ * The cache is a simple hash table in its own memory context, with a key of
+ * search_path and role ID.
+ */
+
+static inline uint32
+nspcachekey_hash(SearchPathCacheKey *key)
+{
+	const unsigned char	*bytes = (const unsigned char *)key->searchPath;
+	int					 blen  = strlen(key->searchPath);
+
+	return hash_bytes(bytes, blen) ^ hash_uint32(key->roleid);
+}
+
+static inline bool
+nspcachekey_equal(SearchPathCacheKey *a, SearchPathCacheKey *b)
+{
+	return a->roleid == b->roleid &&
+		strcmp(a->searchPath, b->searchPath) == 0;
+}
+
+#define SH_PREFIX		nsphash
+#define SH_ELEMENT_TYPE	SearchPathCacheEntry
+#define SH_KEY_TYPE		SearchPathCacheKey *
+#define	SH_KEY			key
+#define SH_HASH_KEY(tb, key)   	nspcachekey_hash(key)
+#define SH_EQUAL(tb, a, b)		nspcachekey_equal(a, b)
+#define	SH_SCOPE		static inline
+#define SH_DECLARE
+#define SH_DEFINE
+#include "lib/simplehash.h"
+
+static nsphash_hash *SearchPathCache = NULL;
 
 /*
  * RangeVarGetRelidExtended
@@ -3370,6 +3426,7 @@ SetTempNamespaceState(Oid tempNamespaceId, Oid tempToastNamespaceId)
 	 */
 
 	baseSearchPathValid = false;	/* may need to rebuild list */
+	searchPathCacheValid = false;
 }
 
 
@@ -3633,28 +3690,19 @@ FindDefaultConversionProc(int32 for_encoding, int32 to_encoding)
 }
 
 /*
- * recomputeNamespacePath - recompute path derived variables if needed.
+ * Look up namespace IDs and perform ACL checks. Return newly-allocated list.
  */
-static void
-recomputeNamespacePath(void)
+static List *
+preprocessNamespacePath(const char *searchPath, Oid roleid,
+						bool *temp_missing)
 {
-	Oid			roleid = GetUserId();
 	char	   *rawname;
 	List	   *namelist;
 	List	   *oidlist;
-	List	   *newpath;
 	ListCell   *l;
-	bool		temp_missing;
-	Oid			firstNS;
-	bool		pathChanged;
-	MemoryContext oldcxt;
 
-	/* Do nothing if path is already valid. */
-	if (baseSearchPathValid && namespaceUser == roleid)
-		return;
-
-	/* Need a modifiable copy of namespace_search_path string */
-	rawname = pstrdup(namespace_search_path);
+	/* Need a modifiable copy */
+	rawname = pstrdup(searchPath);
 
 	/* Parse string into list of identifiers */
 	if (!SplitIdentifierString(rawname, ',', &namelist))
@@ -3671,7 +3719,7 @@ recomputeNamespacePath(void)
 	 * already been accepted.)	Don't make duplicate entries, either.
 	 */
 	oidlist = NIL;
-	temp_missing = false;
+	*temp_missing = false;
 	foreach(l, namelist)
 	{
 		char	   *curname = (char *) lfirst(l);
@@ -3691,10 +3739,8 @@ recomputeNamespacePath(void)
 				namespaceId = get_namespace_oid(rname, true);
 				ReleaseSysCache(tuple);
 				if (OidIsValid(namespaceId) &&
-					!list_member_oid(oidlist, namespaceId) &&
 					object_aclcheck(NamespaceRelationId, namespaceId, roleid,
-									ACL_USAGE) == ACLCHECK_OK &&
-					InvokeNamespaceSearchHook(namespaceId, false))
+									ACL_USAGE) == ACLCHECK_OK)
 					oidlist = lappend_oid(oidlist, namespaceId);
 			}
 		}
@@ -3702,16 +3748,12 @@ recomputeNamespacePath(void)
 		{
 			/* pg_temp --- substitute temp namespace, if any */
 			if (OidIsValid(myTempNamespace))
-			{
-				if (!list_member_oid(oidlist, myTempNamespace) &&
-					InvokeNamespaceSearchHook(myTempNamespace, false))
-					oidlist = lappend_oid(oidlist, myTempNamespace);
-			}
+				oidlist = lappend_oid(oidlist, myTempNamespace);
 			else
 			{
 				/* If it ought to be the creation namespace, set flag */
 				if (oidlist == NIL)
-					temp_missing = true;
+					*temp_missing = true;
 			}
 		}
 		else
@@ -3719,63 +3761,166 @@ recomputeNamespacePath(void)
 			/* normal namespace reference */
 			namespaceId = get_namespace_oid(curname, true);
 			if (OidIsValid(namespaceId) &&
-				!list_member_oid(oidlist, namespaceId) &&
 				object_aclcheck(NamespaceRelationId, namespaceId, roleid,
-								ACL_USAGE) == ACLCHECK_OK &&
-				InvokeNamespaceSearchHook(namespaceId, false))
+								ACL_USAGE) == ACLCHECK_OK)
 				oidlist = lappend_oid(oidlist, namespaceId);
 		}
 	}
 
+	pfree(rawname);
+	list_free(namelist);
+
+	return oidlist;
+}
+
+/*
+ * Remove duplicates, run namespace search hooks, and prepend
+ * implicitly-searched namespaces. Return newly-allocated list.
+ *
+ * All of this must happen after retrieving from cache, because we don't know
+ * what might invalidate the result from the last time the hook was
+ * invoked. It may seem that duplicate elimination is not dependent on the
+ * result of the hook, but if a hook returns different results on different
+ * calls for the same namespace ID, then it could affect the order in which
+ * that namespace appears in the final list.
+ */
+static List *
+finalNamespacePath(List *oidlist, Oid *firstNS)
+{
+	List		*finalPath = NIL;
+	ListCell	*lc;
+
+	foreach(lc, oidlist)
+	{
+		Oid namespaceId = lfirst_oid(lc);
+		if (!list_member_oid(finalPath, namespaceId) &&
+			InvokeNamespaceSearchHook(namespaceId, false))
+			finalPath = lappend_oid(finalPath, namespaceId);
+	}
+
 	/*
 	 * Remember the first member of the explicit list.  (Note: this is
 	 * nominally wrong if temp_missing, but we need it anyway to distinguish
 	 * explicit from implicit mention of pg_catalog.)
 	 */
-	if (oidlist == NIL)
-		firstNS = InvalidOid;
+	if (finalPath == NIL)
+		*firstNS = InvalidOid;
 	else
-		firstNS = linitial_oid(oidlist);
+		*firstNS = linitial_oid(finalPath);
 
 	/*
 	 * Add any implicitly-searched namespaces to the list.  Note these go on
 	 * the front, not the back; also notice that we do not check USAGE
 	 * permissions for these.
 	 */
-	if (!list_member_oid(oidlist, PG_CATALOG_NAMESPACE))
-		oidlist = lcons_oid(PG_CATALOG_NAMESPACE, oidlist);
+	if (!list_member_oid(finalPath, PG_CATALOG_NAMESPACE))
+		finalPath = lcons_oid(PG_CATALOG_NAMESPACE, finalPath);
 
 	if (OidIsValid(myTempNamespace) &&
-		!list_member_oid(oidlist, myTempNamespace))
-		oidlist = lcons_oid(myTempNamespace, oidlist);
+		!list_member_oid(finalPath, myTempNamespace))
+		finalPath = lcons_oid(myTempNamespace, finalPath);
+
+	return finalPath;
+}
+
+/*
+ * Retrieve search path information from the cache; or if not present,
+ * preprocess it and add it to the cache. Return OID list allocated in
+ * SearchPathCacheContext; caller should not free.
+ */
+static List *
+cachedNamespacePath(const char *searchPath, Oid roleid,
+					bool *temp_missing)
+{
+	SearchPathCacheKey		 cachekey;
+	SearchPathCacheEntry	*entry;
+	bool					 found;
+
+	/* clear the cache and start over when invalidated */
+	if (!searchPathCacheValid)
+	{
+		MemoryContextReset(SearchPathCacheContext);
+		/* arbitrary initial starting size of 16 elements */
+		SearchPathCache = nsphash_create(SearchPathCacheContext, 16, NULL);
+		searchPathCacheValid = true;
+	}
+
+	cachekey.searchPath = searchPath;
+	cachekey.roleid = roleid;
+
+	entry = nsphash_insert(SearchPathCache, &cachekey, &found);
+
+	if (!found)
+	{
+		MemoryContext		 oldcxt;
+		SearchPathCacheKey	*newkey;
+		List				*oidlist;
+
+		oidlist = preprocessNamespacePath(searchPath, roleid,
+										  &entry->temp_missing);
+
+		oldcxt = MemoryContextSwitchTo(SearchPathCacheContext);
+
+		newkey = palloc(sizeof(SearchPathCacheKey));
+		newkey->searchPath = pstrdup(searchPath);
+		newkey->roleid = roleid;
+
+		entry->key = newkey;
+		entry->oidlist = list_copy(oidlist);
+
+		MemoryContextSwitchTo(oldcxt);
+
+		/* oidlist already copied into cache entry */
+		list_free(oidlist);
+	}
+
+	*temp_missing = entry->temp_missing;
+	return entry->oidlist;
+}
+
+/*
+ * recomputeNamespacePath - recompute path derived variables if needed.
+ */
+static void
+recomputeNamespacePath(void)
+{
+	Oid			roleid = GetUserId();
+	List	   *oidlist;
+	Oid			firstNS;
+	bool		temp_missing;
+	bool		pathChanged;
+	List	   *finalPath;
+	MemoryContext	oldcxt;
+
+	/* Do nothing if path is already valid. */
+	if (baseSearchPathValid && namespaceUser == roleid)
+		return;
+
+	oidlist = cachedNamespacePath(namespace_search_path, roleid,
+								  &temp_missing);
+
+	/* Must save OID list in permanent storage. */
+	oldcxt = MemoryContextSwitchTo(TopMemoryContext);
+	finalPath = finalNamespacePath(oidlist, &firstNS);
+	MemoryContextSwitchTo(oldcxt);
 
-	/*
-	 * We want to detect the case where the effective value of the base search
-	 * path variables didn't change.  As long as we're doing so, we can avoid
-	 * copying the OID list unnecessarily.
-	 */
 	if (baseCreationNamespace == firstNS &&
 		baseTempCreationPending == temp_missing &&
-		equal(oidlist, baseSearchPath))
+		equal(finalPath, baseSearchPath))
 	{
 		pathChanged = false;
 	}
 	else
 	{
 		pathChanged = true;
-
-		/* Must save OID list in permanent storage. */
-		oldcxt = MemoryContextSwitchTo(TopMemoryContext);
-		newpath = list_copy(oidlist);
-		MemoryContextSwitchTo(oldcxt);
-
-		/* Now safe to assign to state variables. */
-		list_free(baseSearchPath);
-		baseSearchPath = newpath;
-		baseCreationNamespace = firstNS;
-		baseTempCreationPending = temp_missing;
 	}
 
+	/* Now safe to assign to state variables. */
+	list_free(baseSearchPath);
+	baseSearchPath = finalPath;
+	baseCreationNamespace = firstNS;
+	baseTempCreationPending = temp_missing;
+
 	/* Mark the path valid. */
 	baseSearchPathValid = true;
 	namespaceUser = roleid;
@@ -3791,11 +3936,6 @@ recomputeNamespacePath(void)
 	 */
 	if (pathChanged)
 		activePathGeneration++;
-
-	/* Clean up. */
-	pfree(rawname);
-	list_free(namelist);
-	list_free(oidlist);
 }
 
 /*
@@ -3950,6 +4090,7 @@ InitTempTableNamespace(void)
 	myTempNamespaceSubID = GetCurrentSubTransactionId();
 
 	baseSearchPathValid = false;	/* need to rebuild list */
+	searchPathCacheValid = false;
 }
 
 /*
@@ -3975,6 +4116,7 @@ AtEOXact_Namespace(bool isCommit, bool parallel)
 			myTempNamespace = InvalidOid;
 			myTempToastNamespace = InvalidOid;
 			baseSearchPathValid = false;	/* need to rebuild list */
+			searchPathCacheValid = false;
 
 			/*
 			 * Reset the temporary namespace flag in MyProc.  We assume that
@@ -4016,6 +4158,7 @@ AtEOSubXact_Namespace(bool isCommit, SubTransactionId mySubid,
 			myTempNamespace = InvalidOid;
 			myTempToastNamespace = InvalidOid;
 			baseSearchPathValid = false;	/* need to rebuild list */
+			searchPathCacheValid = false;
 
 			/*
 			 * Reset the temporary namespace flag in MyProc.  We assume that
@@ -4130,6 +4273,10 @@ assign_search_path(const char *newval, void *extra)
 	 * We mark the path as needing recomputation, but don't do anything until
 	 * it's needed.  This avoids trying to do database access during GUC
 	 * initialization, or outside a transaction.
+	 *
+	 * This does not invalidate the search path cache, so if this value had
+	 * been previously set and no syscache invalidations happened,
+	 * recomputation may not be necessary.
 	 */
 	baseSearchPathValid = false;
 }
@@ -4164,6 +4311,9 @@ InitializeSearchPath(void)
 	}
 	else
 	{
+		SearchPathCacheContext = AllocSetContextCreate(TopMemoryContext,
+													   "search_path processing cache",
+													   ALLOCSET_DEFAULT_SIZES);
 		/*
 		 * In normal mode, arrange for a callback on any syscache invalidation
 		 * of pg_namespace rows.
@@ -4173,6 +4323,7 @@ InitializeSearchPath(void)
 									  (Datum) 0);
 		/* Force search path to be recomputed on next use */
 		baseSearchPathValid = false;
+		searchPathCacheValid = false;
 	}
 }
 
@@ -4183,8 +4334,13 @@ InitializeSearchPath(void)
 static void
 NamespaceCallback(Datum arg, int cacheid, uint32 hashvalue)
 {
-	/* Force search path to be recomputed on next use */
+	/*
+	 * Force search path to be recomputed on next use, also invalidating the
+	 * search path cache (because namespace names, ACLs, or role names may
+	 * have changed).
+	 */
 	baseSearchPathValid = false;
+	searchPathCacheValid = false;
 }
 
 /*
-- 
2.34.1