Use STRICT_ASSIGN() for exp2f() and exp2() instead of a volatile
variable hack for exp2f() only. The volatile variable had a surprisingly large cost for exp2f() -- 19 cycles or 15% on i386 in the worst case observed. This is only partly explained by there being several references to the variable, only one of which benefited from it being volatile. Arches that have working assignment are likely to benefit even more from not having any volatile variable. exp2() now has a chance of working with extra precision on i386. exp2() has even more references to the variable, so it would have been pessimized more by simply declaring the variable as volatile. Even the temporary volatile variable for STRICT_ASSIGN costs 5-10% on i386, (A64) so I will change STRICT_ASSIGN() to do an ordinary assignment until i386 defaults to extra precision.
This commit is contained in:
parent
fa7fdac725
commit
684217d889
@ -364,7 +364,7 @@ exp2(double x)
|
||||
}
|
||||
|
||||
/* Reduce x, computing z, i0, and k. */
|
||||
t = x + redux;
|
||||
STRICT_ASSIGN(double, t, x + redux);
|
||||
GET_LOW_WORD(i0, t);
|
||||
i0 += TBLSIZE / 2;
|
||||
k = (i0 >> TBLBITS) << 20;
|
||||
|
@ -27,6 +27,8 @@
|
||||
#include <sys/cdefs.h>
|
||||
__FBSDID("$FreeBSD$");
|
||||
|
||||
#include <float.h>
|
||||
|
||||
#include "math.h"
|
||||
#include "math_private.h"
|
||||
|
||||
@ -91,8 +93,7 @@ float
|
||||
exp2f(float x)
|
||||
{
|
||||
double tv;
|
||||
float r, z;
|
||||
volatile float t; /* prevent gcc from using too much precision */
|
||||
float r, t, z;
|
||||
uint32_t hx, hr, ix, i0;
|
||||
int32_t k;
|
||||
|
||||
@ -115,7 +116,7 @@ exp2f(float x)
|
||||
}
|
||||
|
||||
/* Reduce x, computing z, i0, and k. */
|
||||
t = x + redux;
|
||||
STRICT_ASSIGN(float, t, x + redux);
|
||||
GET_FLOAT_WORD(i0, t);
|
||||
i0 += TBLSIZE / 2;
|
||||
k = (i0 >> TBLBITS) << 23;
|
||||
|
Loading…
x
Reference in New Issue
Block a user