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. An earlier version of this patch was reviewd by Nathan Bossart. This version simplifies a few things and is safer in case of OOM. Discussion: https://www.postgresql.org/message-id/abf4ce8804e0e05dff8c1725ae6a8ed28b7d66e0.camel%40j-davis.com Reviewed-by: Nathan Bossart
2023-11-14 17:51:03 -08:00 · 2023-11-14 17:51:03 -08:00 · f26c2368dc
parent 7f6bc3b356
commit f26c2368dc
1 changed files with 341 additions and 63 deletions
--- 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 +110,13 @@
 * 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 or pg_authid.  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 +156,27 @@ 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;		/* namespace OIDs that pass ACL checks */
+	List	   *finalPath;		/* cached final computed search path */
+	Oid			firstNS;		/* first explicitly-listed namespace */
+	bool		temp_missing;
+	bool		forceRecompute; /* force recompute of finalPath */
+
+	/* needed for simplehash */
+	char		status;
+}			SearchPathCacheEntry;

 /*
 * myTempNamespace is InvalidOid until and unless a TEMP namespace is set up
@ -206,6 +230,133 @@ 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 search path cache
+ * that can be used by recomputeNamespacePath().
+ *
+ * The search path cache is based on a wrapper around a simplehash hash table
+ * (nsphash, defined below). The spcache wrapper deals with OOM while trying
+ * to initialize a key, and also offers a more convenient API.
+ */
+
+static inline uint32
+spcachekey_hash(SearchPathCacheKey key)
+{
+	const unsigned char *bytes = (const unsigned char *) key.searchPath;
+	int			blen = strlen(key.searchPath);
+
+	return hash_combine(hash_bytes(bytes, blen),
+						hash_uint32(key.roleid));
+}
+
+static inline bool
+spcachekey_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)   	spcachekey_hash(key)
+#define SH_EQUAL(tb, a, b)		spcachekey_equal(a, b)
+#define SH_SCOPE		static inline
+#define SH_DECLARE
+#define SH_DEFINE
+#include "lib/simplehash.h"
+
+/*
+ * We only expect a small number of unique search_path strings to be used. If
+ * this cache grows to an unreasonable size, reset it to avoid steady-state
+ * memory growth. Most likely, only a few of those entries will benefit from
+ * the cache, and the cache will be quickly repopulated with such entries.
+ */
+#define SPCACHE_RESET_THRESHOLD		256
+
+static nsphash_hash * SearchPathCache = NULL;
+
+/*
+ * Create search path cache.
+ */
+static void
+spcache_init(void)
+{
+	Assert(SearchPathCacheContext);
+
+	if (SearchPathCache)
+		return;
+
+	/* arbitrary initial starting size of 16 elements */
+	SearchPathCache = nsphash_create(SearchPathCacheContext, 16, NULL);
+	searchPathCacheValid = true;
+}
+
+/*
+ * Reset and reinitialize search path cache.
+ */
+static void
+spcache_reset(void)
+{
+	Assert(SearchPathCacheContext);
+	Assert(SearchPathCache);
+
+	MemoryContextReset(SearchPathCacheContext);
+	SearchPathCache = NULL;
+
+	spcache_init();
+}
+
+static uint32
+spcache_members(void)
+{
+	return SearchPathCache->members;
+}
+
+/*
+ * Look up or insert entry in search path cache.
+ *
+ * Initialize key safely, so that OOM does not leave an entry without a valid
+ * key. Caller must ensure that non-key contents are properly initialized.
+ */
+static SearchPathCacheEntry *
+spcache_insert(const char *searchPath, Oid roleid)
+{
+	SearchPathCacheEntry *entry;
+	bool		found;
+	SearchPathCacheKey cachekey = {
+		.searchPath = searchPath,
+		.roleid = roleid
+	};
+
+	/*
+	 * If a new entry is created, we must ensure that it's properly
+	 * initialized. Set the cache invalid temporarily, so that if the
+	 * MemoryContextStrdup() below raises an OOM, the cache will be reset on
+	 * the next use, clearing the uninitialized entry.
+	 */
+	searchPathCacheValid = false;
+
+	entry = nsphash_insert(SearchPathCache, cachekey, &found);
+
+	/* ensure that key is initialized and the rest is zeroed */
+	if (!found)
+	{
+		entry->key.searchPath = MemoryContextStrdup(SearchPathCacheContext, searchPath);
+		entry->key.roleid = roleid;
+		entry->oidlist = NIL;
+		entry->finalPath = NIL;
+		entry->firstNS = InvalidOid;
+		entry->temp_missing = false;
+		entry->forceRecompute = false;
+		/* do not touch entry->status, used by simplehash */
+	}
+
+	searchPathCacheValid = true;
+	return entry;
+}

 /*
 * RangeVarGetRelidExtended
@ -3630,6 +3781,7 @@ SetTempNamespaceState(Oid tempNamespaceId, Oid tempToastNamespaceId)
 	 */

 	baseSearchPathValid = false;	/* may need to rebuild list */
+	searchPathCacheValid = false;
 }


@ -3893,28 +4045,19 @@ 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))
@ -3931,7 +4074,7 @@ 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);
@ -3951,10 +4094,8 @@ 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);
 			}
 		}
@ -3962,16 +4103,12 @@ 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
@ -3979,63 +4116,192 @@ 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.
+ *
+ * If an object_access_hook is present, this must always be recalculated. 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))
+		{
+			if (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 there, fill
+ * it. The returned entry is valid only until the next call to this function.
+ *
+ * We also determine if the newly-computed finalPath is the same as the
+ * prevPath passed by the caller (i.e. a no-op or a real change?). It's more
+ * efficient to check for a change in this function than the caller, because
+ * we can avoid unnecessary temporary copies of the previous path.
+ */
+static const SearchPathCacheEntry *
+cachedNamespacePath(const char *searchPath, Oid roleid, List *prevPath,
+					bool *same)
+{
+	MemoryContext oldcxt;
+	SearchPathCacheEntry *entry;
+	List	   *prevPathCopy = NIL;
+
+	spcache_init();
+
+	/* invalidate cache if necessary */
+	if (!searchPathCacheValid || spcache_members() >= SPCACHE_RESET_THRESHOLD)
+	{
+		/* prevPath will be destroyed; make temp copy for later comparison */
+		prevPathCopy = list_copy(prevPath);
+		prevPath = prevPathCopy;
+		spcache_reset();
+	}
+
+	entry = spcache_insert(searchPath, roleid);

 	/*
-	 * 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.
+	 * An OOM may have resulted in a cache entry with mising 'oidlist' or
+	 * 'finalPath', so just compute whatever is missing.
 	 */
-	if (baseCreationNamespace == firstNS &&
-		baseTempCreationPending == temp_missing &&
-		equal(oidlist, baseSearchPath))
+
+	if (entry->oidlist == NIL)
+	{
+		oldcxt = MemoryContextSwitchTo(SearchPathCacheContext);
+		entry->oidlist = preprocessNamespacePath(searchPath, roleid,
+												 &entry->temp_missing);
+		MemoryContextSwitchTo(oldcxt);
+	}
+
+	/*
+	 * If a hook is set, we must recompute finalPath from the oidlist each
+	 * time, because the hook may affect the result. This is still much faster
+	 * than recomputing from the string (and doing catalog lookups and ACL
+	 * checks).
+	 */
+	if (entry->finalPath == NIL || object_access_hook ||
+		entry->forceRecompute)
+	{
+		/*
+		 * Do not free the stale value of entry->finalPath until we've
+		 * performed the comparison, in case it's aliased by prevPath (which
+		 * can only happen when recomputing due to an object_access_hook).
+		 */
+		List	   *finalPath;
+
+		oldcxt = MemoryContextSwitchTo(SearchPathCacheContext);
+		finalPath = finalNamespacePath(entry->oidlist,
+									   &entry->firstNS);
+		MemoryContextSwitchTo(oldcxt);
+
+		*same = equal(prevPath, finalPath);
+
+		list_free(entry->finalPath);
+		entry->finalPath = finalPath;
+
+		/*
+		 * If an object_access_hook set when finalPath is calculated, the
+		 * result may be affected by the hook. Force recomputation of
+		 * finalPath the next time this cache entry is used, even if the
+		 * object_access_hook is not set at that time.
+		 */
+		entry->forceRecompute = object_access_hook ? true : false;
+	}
+	else
+	{
+		/* use cached version of finalPath */
+		*same = equal(prevPath, entry->finalPath);
+	}
+
+	list_free(prevPathCopy);
+
+	return entry;
+}
+
+/*
+ * recomputeNamespacePath - recompute path derived variables if needed.
+ */
+static void
+recomputeNamespacePath(void)
+{
+	Oid			roleid = GetUserId();
+	bool		newPathEqual;
+	bool		pathChanged;
+	const		SearchPathCacheEntry *entry;
+
+	/* Do nothing if path is already valid. */
+	if (baseSearchPathValid && namespaceUser == roleid)
+		return;
+
+	entry = cachedNamespacePath(namespace_search_path, roleid, baseSearchPath,
+								&newPathEqual);
+
+	if (baseCreationNamespace == entry->firstNS &&
+		baseTempCreationPending == entry->temp_missing &&
+		newPathEqual)
 	{
 		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. */
+	baseSearchPath = entry->finalPath;
+	baseCreationNamespace = entry->firstNS;
+	baseTempCreationPending = entry->temp_missing;
+
 	/* Mark the path valid. */
 	baseSearchPathValid = true;
 	namespaceUser = roleid;
@ -4051,11 +4317,6 @@ recomputeNamespacePath(void)
 	 */
 	if (pathChanged)
 		activePathGeneration++;
-
-	/* Clean up. */
-	pfree(rawname);
-	list_free(namelist);
-	list_free(oidlist);
 }

 /*
@ -4210,6 +4471,7 @@ InitTempTableNamespace(void)
 	myTempNamespaceSubID = GetCurrentSubTransactionId();

 	baseSearchPathValid = false;	/* need to rebuild list */
+	searchPathCacheValid = false;
 }

 /*
@ -4235,6 +4497,7 @@ 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
@ -4276,6 +4539,7 @@ 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
@ -4399,6 +4663,10 @@ 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;
 }
@ -4433,6 +4701,10 @@ 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 or pg_authid rows. (Changing a role name may affect
@ -4446,6 +4718,7 @@ InitializeSearchPath(void)
 									  (Datum) 0);
 		/* Force search path to be recomputed on next use */
 		baseSearchPathValid = false;
+		searchPathCacheValid = false;
 	}
 }

@ -4456,8 +4729,13 @@ 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;
 }

 /*