[HACKERS] Re: is possible cache tupledesc templates in execution plan?significant performance issue, maybe bug?

Поиск
Список
Период
Сортировка
От Pavel Stehule
Тема [HACKERS] Re: is possible cache tupledesc templates in execution plan?significant performance issue, maybe bug?
Дата
Msg-id CAFj8pRAUEKdfn-_gLz0ng0P+htd7sJ-m9As3PSfd5mQmbrogww@mail.gmail.com
обсуждение исходный текст
Ответ на [HACKERS] Re: is possible cache tupledesc templates in execution plan?significant performance issue, maybe bug?  (Pavel Stehule <pavel.stehule@gmail.com>)
Список pgsql-hackers




The following workaround is working

create view xxxx as  select CISLOEXEKUCE, MT.ID_NAJDATSPLT
      from najzalobpr MT, najvzallok A1,
                    NAJZALOBST A2, NAJZALOBCE A3 where
                    MT.ID_NAJVZALLOK= A1.ID_NAJVZALLOK AND
                    A1.ID_NAJZALOBST=A2.ID_NAJZALOBST AND
                    A2.ID_NAJZALOBCE= A3.ID_NAJZALOBCE;


and function should be changed to

BEGIN
   BEGIN
    select CISLOEXEKUCE INTO STRICT mADRA
      from xxxx
      WHERE id_najdatsplt = mID_najdatsplt LIMIT 1;
    EXCEPTION
      WHEN OTHERS THEN
        mADRA := NULL;
    END;



  Result:=mADRA;
  return(Result);
end;

So this issue is really related to tupleDesc management

I found a bug in this workaround. It doesn't work



 
Pavel




В списке pgsql-hackers по дате отправления:

Предыдущее
От: Pavel Stehule
Дата:
Сообщение: Re: [HACKERS] is possible cache tupledesc templates in executionplan? significant performance issue, maybe bug?
Следующее
От: Peter Geoghegan
Дата:
Сообщение: Re: [HACKERS] [COMMITTERS] pgsql: Fix freezing of a dead HOT-updated tuple