Add an additional usage note about using __predict_*() only in hot spots,
as suggested by Chris Demetriou.
This commit is contained in:
parent
7ae9f5b5e9
commit
9e79857b5c
|
@ -1,4 +1,4 @@
|
||||||
/* $NetBSD: cdefs.h,v 1.33 2000/05/08 18:36:00 thorpej Exp $ */
|
/* $NetBSD: cdefs.h,v 1.34 2000/05/08 22:41:38 thorpej Exp $ */
|
||||||
|
|
||||||
/*
|
/*
|
||||||
* Copyright (c) 1991, 1993
|
* Copyright (c) 1991, 1993
|
||||||
|
@ -215,6 +215,12 @@
|
||||||
* * Other than that, if you don't know the liklyhood of a test
|
* * Other than that, if you don't know the liklyhood of a test
|
||||||
* succeeding from empirical or other `hard' evidence, don't
|
* succeeding from empirical or other `hard' evidence, don't
|
||||||
* make predictions.
|
* make predictions.
|
||||||
|
*
|
||||||
|
* * These are meant to be used in places that are run `a lot'.
|
||||||
|
* It is wasteful to make predictions in code that is run
|
||||||
|
* seldomly (e.g. at subsystem initialization time) as the
|
||||||
|
* basic block reordering that this affects can often generate
|
||||||
|
* larger code.
|
||||||
*/
|
*/
|
||||||
#if __GNUC_PREREQ__(2, 96)
|
#if __GNUC_PREREQ__(2, 96)
|
||||||
#define __predict_true(exp) __builtin_expect(((exp) != 0), 1)
|
#define __predict_true(exp) __builtin_expect(((exp) != 0), 1)
|
||||||
|
|
Loading…
Reference in New Issue