firtst release
Revision | 09bf8f7b62fa4918de109988ed85139dcb9cfbf6 (tree) |
---|---|
Zeit | 2014-12-22 11:07:09 |
Autor | Kyotaro Horiguchi <horiguchi.kyotaro@lab....> |
Commiter | Kyotaro Horiguchi |
Catch up the changes of pg_stat_statements.c
The change has nothing to do with pg_hint_plan but it's necessary to
catch it up in order to keep the difference minimal for the ease of
maintenance.
@@ -965,10 +965,13 @@ pgss_ProcessUtility(Node *parsetree, const char *queryString, | ||
965 | 965 | * calculated from the query tree) would be used to accumulate costs of |
966 | 966 | * ensuing EXECUTEs. This would be confusing, and inconsistent with other |
967 | 967 | * cases where planning time is not included at all. |
968 | + * | |
969 | + * Likewise, we don't track execution of DEALLOCATE. | |
968 | 970 | */ |
969 | 971 | if (pgss_track_utility && pgss_enabled() && |
970 | 972 | !IsA(parsetree, ExecuteStmt) && |
971 | - !IsA(parsetree, PrepareStmt)) | |
973 | + !IsA(parsetree, PrepareStmt) && | |
974 | + !IsA(parsetree, DeallocateStmt)) | |
972 | 975 | { |
973 | 976 | instr_time start; |
974 | 977 | instr_time duration; |