Re: [HACKERS] Do we need the gcc feature "__builtin_expect" topromote the branches prediction?
От | Julien Rouhaud |
---|---|
Тема | Re: [HACKERS] Do we need the gcc feature "__builtin_expect" topromote the branches prediction? |
Дата | |
Msg-id | d696d555-e8ff-3fbb-7998-1596b188d198@dalibo.com обсуждение исходный текст |
Ответ на | Re: [HACKERS] Do we need the gcc feature "__builtin_expect" topromote the branches prediction? (Craig Ringer <craig.ringer@2ndquadrant.com>) |
Ответы |
Re: [HACKERS] Do we need the gcc feature "__builtin_expect" topromote the branches prediction?
|
Список | pgsql-hackers |
On 02/06/2017 12:50, Craig Ringer wrote: > > > On 2 Jun. 2017 16:42, "Hao Lee" <mixtrue@gmail.com > <mailto:mixtrue@gmail.com>> wrote: > > Hi all, > There is a lot of "if statement" in system, and GCC provides > a feature,"__builtin_expect", which let compilers know which branch > is mostly run. > > > Compilers and CPUs are really good at guessing this. > > Humans are wrong about it more than we'd like too. +1 > > It's surprisingly rarely s good idea to use branch prediction hints. > > See the vsrious Linux kernel discussions about this. > > If you find concrete sites of frequent or costly branch mis-prediction > please point them out, with benchmarks. And also see this thread: https://www.postgresql.org/message-id/CAKJS1f8yqRW3qx2CO9r4bqqvA2Vx68=3awbh8CJWTP9zXeoHMw@mail.gmail.com BTW Andres added support for likely/unlikely in aa3ca5e3dd6, but its usage is still really limited. -- Julien Rouhaud http://dalibo.com - http://dalibo.org
В списке pgsql-hackers по дате отправления: