world leader in high performance signal processing
Trace: » nbench
Table of Contents

nbench

nbench is a Linux/Unix port of release 2 of BYTE Magazine's BYTEmark benchmark program (previously known as BYTE's Native Mode Benchmarks). These are Native Mode (a.k.a. Algorithm Level) tests; benchmarks designed to expose the capabilities of a system's CPU, FPU, and memory system. Read all about it at BYTE's benchmark page.

The benchmark program takes less than 10 minutes to run (on most machines) and compares the system it is run on to two benchmark systems (a Dell Pentium 90 with 256 KB cache running MSDOS and an AMD K6/233 with 512 KB cache running Linux).

root:/> nbench

BYTEmark* Native Mode Benchmark ver. 2 (10/95)
Index-split by Andrew D. Balsa (11/97)
Linux/Unix* port by Uwe F. Mayer (12/96,11/97)

TEST                : Iterations/sec.  : Old Index   : New Index
                    :                  : Pentium 90* : AMD K6/233*
--------------------:------------------:-------------:------------
NUMERIC SORT        :           86.84  :       2.23  :       0.73
STRING SORT         :           10.02  :       4.48  :       0.69
BITFIELD            :      2.1645e+07  :       3.71  :       0.78
FP EMULATION        :          13.095  :       6.28  :       1.45
FOURIER             :          27.918  :       0.03  :       0.02
ASSIGNMENT          :          1.6257  :       6.19  :       1.60
IDEA                :          300.49  :       4.60  :       1.36
HUFFMAN             :          112.99  :       3.13  :       1.00
NEURAL NET          :        0.033139  :       0.05  :       0.02
LU DECOMPOSITION    :          1.1078  :       0.06  :       0.04
==========================ORIGINAL BYTEMARK RESULTS==========================
INTEGER INDEX       : 4.137
FLOATING-POINT INDEX: 0.046
Baseline (MSDOS*)   : Pentium* 90, 256 KB L2-cache, Watcom* compiler 10.0
==============================LINUX DATA BELOW===============================
CPU                 : Analog Devices ADSP-BF537 600(MHz CCLK) 120(MHz SCLK) 600MHz
L2 Cache            : 16 KB(L1 icache) 32 KB(L1 dcache-wb) 0 KB(L2 cache)
OS                  : Linux 2.6.22.19-ADI-2008R1.5-svn5350
C compiler          : bfin-uclinux-gcc
libc                : static
MEMORY INDEX        : 0.952
INTEGER INDEX       : 1.097
FLOATING-POINT INDEX: 0.025
Baseline (LINUX)    : AMD K6/233*, 512 KB L2-cache, gcc 2.7.2.3, libc-5.4.38
* Trademarks are property of their respective holder.
root:/>

If this application/benchmark is profiled, the top 13 functions account for 90% of the CPU Utilization.

Total CPU % Application Function
14.23% nbench ___muldf3
13.62% nbench ___adddf3
8.77% nbench _DoAssignIteration
7.79% nbench _cipher_idea
7.72% nbench _DoBitfieldIteration
7.62% nbench _DoNumSortIteration
7.62% libuClibc-0.9.29.so ___GI_memmove
6.96% nbench _DoHuffIteration
3.77% nbench _DivideInternalFPF
3.61% nbench _DoEmFloatIteration
3.30% nbench ___divdf3
2.69% libm-0.9.29.so ___udivsi3
2.37% nbench _DoNNetIteration

Results

Flags size Baseline (MSDOS*)
Pentium* 90\\, 256 KB L2-cache, Watcom* compiler 10.0
Baseline (LINUX)\AMD K6/233*\\512 KB L2-cache, gcc 2.7.2.3, libc-5.4.38
-Os 80352 INTEGER INDEX : 3.653 FLOATING-POINT INDEX: 0.046 MEMORY INDEX : 0.880 INTEGER INDEX : 0.936 FLOATING-POINT INDEX: 0.026
-Os -mfast-fp 77672 INTEGER INDEX : 3.826 FLOATING-POINT INDEX: 0.186 MEMORY INDEX : 0.880 INTEGER INDEX : 1.015 FLOATING-POINT INDEX: 0.103