Bug Report #2803

Trying to optimise call_user_func_array with a switch is pointless

Added by Mike Parkin about 4 years ago. Updated almost 4 years ago.

Status:ClosedStart date:04/16/2010
Priority:NormalDue date:
Assignee:John Heathco% Done:

0%

Category:Modules:ORM
Target version:v3.0.8
Resolution:fixed Points:

Description

The switch might be faster, but we are talking such tiny margins here i think it would be better to just have a clean codebase rather than hacks to boost micro performance. x% of 0 is still 0.

some tests:
http://files.droplr.com/files/785785/PmR2h.LittleSnapper.png

http://github.com/kohana/codebench/commit/d2677eeada4847f8854eaeb6fcf8c284763d423c

Associated revisions

History

#1 Updated by Woody Gilk almost 4 years ago

  • Status changed from New to Assigned
  • Target version set to v3.0.5

Agreed. We're talking about a 0.00001 second difference across 100,000 requests.

#2 Updated by Woody Gilk almost 4 years ago

  • Target version changed from v3.0.5 to v3.0.6

#3 Updated by Woody Gilk almost 4 years ago

  • Target version changed from v3.0.6 to v3.0.7

#4 Updated by Woody Gilk almost 4 years ago

  • Target version changed from v3.0.7 to v3.0.8

Also available in: Atom PDF