Bugfix and improvements in multixact.c

Поиск
Список
Период
Сортировка
От Maxim Orlov
Тема Bugfix and improvements in multixact.c
Дата
Msg-id CACG=ezbLUG-OD1osAW3OchOMxZtdxHh2itYR9Zhh-a13wEBEQw@mail.gmail.com
обсуждение исходный текст
Ответы Re: Bugfix and improvements in multixact.c
Re: Bugfix and improvements in multixact.c
Список pgsql-hackers
Hi!

While working on a making multix xact offsets 64-bit [0] I've discovered a minor issue.  The 
thing is that type 'xid' is used in all macro, but it doesn't correct.  Appropriate MultiXactId or 
MultiXactOffset should be used, actually.

And the second thing, as Heikki Linnakangas points out, args naming is also misleading.

Since, these problems are not the point of thread [0], I decided to create this discussion.
And here is the patch set addressing mentioned issues (0001 and 0002).

Additionally, I made an optional patch 0003 to switch from macro to inline functions.  For 
me, personally, use of macro functions is justified if we are dealing with different argument 
types, to make polymorphic call.  Which is not the case here.  So, we can have more 
control over types and still generate the same code in terms of speed.  
See https://godbolt.org/z/KM8voadhs Starting from O1 function is inlined, thus no 
overhead is noticeable.  Anyway, it's up to the actual commiter to decide does it worth it
or not.  Again, this particular patch 0003 is completely optional.

As always, any opinions and reviews are very welcome!


--
Best regards,
Maxim Orlov.
Вложения

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Michail Nikolaev
Дата:
Сообщение: Re: Issues with ON CONFLICT UPDATE and REINDEX CONCURRENTLY
Следующее
От: "David E. Wheeler"
Дата:
Сообщение: Re: jsonpath: Missing Binary Execution Path?