qbit/go
1
0
Fork 0
mirror of https://github.com/golang/go synced 2024-11-17 23:34:52 -07:00
Code Releases Activity

math: speed up Log on amd64

Browse Source 
After https://golang.org/cl/31490 we break false
output dependency for CVTS.. in compiler generated code.
I've looked through asm code, which uses CVTS..
and added XOR to the only case where it affected performance.

Log-6                  21.6ns ± 0%  19.9ns ± 0%  -7.87%  (p=0.000 n=10+10)

Change-Id: I25d9b405e3041a3839b40f9f9a52e708034bb347
Reviewed-on: https://go-review.googlesource.com/38771
Run-TryBot: Ilya Tocar <ilya.tocar@intel.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
This commit is contained in:
Ilya Tocar 2017-03-28 14:09:28 -05:00
parent 41fd8d6401
commit 4f579cc65b
1 changed files with 1 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
src/math/log_amd64.s
View File

@ -41,6 +41,7 @@ TEXT ·Log(SB),NOSPLIT,$0
SHRQ $52, BX SHRQ $52, BX
ANDL $0x7FF, BX ANDL $0x7FF, BX
SUBL $0x3FE, BX SUBL $0x3FE, BX
XORPS X1, X1 // break dependency for CVTSL2SD
CVTSL2SD BX, X1 // x1= k, x2= f1 CVTSL2SD BX, X1 // x1= k, x2= f1
// if f1 < math.Sqrt2/2 { k -= 1; f1 *= 2 } // if f1 < math.Sqrt2/2 { k -= 1; f1 *= 2 }
MOVSD $HSqrt2, X0 // x0= 0.7071, x1= k, x2= f1 MOVSD $HSqrt2, X0 // x0= 0.7071, x1= k, x2= f1