Performance Test of the nioii Processor

In this paper, the performance of the niosii processor's economical type niosii/E and fast type niosii/F are tested in different optimization methods.Code.
1. printf // print the file

2. usleep (1000) // sleep time

3. iowr_altera_avalon_pio_data // port read/write

The test code is as follows:

# Include <stdio. h>
# Include "system. H"
# Include "unistd. H"
# Include "altera_avalon_performance_counter.h"
# Include "altera_avalon_pio_regs.h"

Int main ()
{
Perf_reset (performance_counter_base );
Perf_start_measuring (performance_counter_base );
Perf_begin (performance_counter_base, 1 );
Printf ("hello from NiO II! \ N "); // test printf
Perf_end (performance_counter_base, 1 );
Perf_begin (performance_counter_base, 2 );
Usleep (1000); // test usleep
Perf_end (performance_counter_base, 2 );
Perf_begin (performance_counter_base, 3 );
// Iowr_altera_avalon_pio_data (pio_base, 0x00); // test IOWR
IOWR (pio_base, 0, 0x00 );
Perf_end (performance_counter_base, 3 );
Perf_stop_measuring (performance_counter_base );
Perf_print_formatted_report (performance_counter_base, alt_get_cpu_freq (),
3, "printf", "usleep", "IOWR ");
Return 0;
}

The test results are as follows:

Table 1: niosii/e, not optimized
Section	%	Time (SEC)	Time (clocks)	Occurrences
Printf	20.9	0.00113	56317	1
Usleep	79	0.00427	213400	1
IOWR	0.0459	0.00000	124	1

Table 2: NiO II/e, optimization: optimize-03

Section	%	Time (SEC)	Time (clocks)	Occurrences
Printf	13.9	0.00065	32648	1
Usleep	86	0.00404	201865	1
IOWR	0.0383	0.00000	90	1

Table 3: NiO II/e, optimization: optimize-0 s

Section	%	Time (SEC)	Time (clocks)	Occurrences
Printf	13.4	0.00063	31439	1
Usleep	86.5	0.00405	202681	1
IOWR	0.0358	0.00000	84	1

Table 4: niosii/F, not optimized

Section	%	Time (SEC)	Time (clocks)	Occurrences
Printf	18.7	0.00023	11387	1
Usleep	81.1	0.00099	49290	1
IOWR	0.0428	0.00000	26	1

Table 5: NiO II/F, optimization: optimize-03

Section	%	Time (SEC)	Time (clocks)	Occurrences
Printf	11	0.00012	5969	1
Usleep	88.9	0.00097	48281	1
IOWR	0.0147	0.00000	8	1

Table 6: NiO II/F, optimization: optimize-0 s

Section	%	Time (SEC)	Time (clocks)	Occurrences
Printf	12.1	0.00013	6653	1
Usleep	87.8	0.00097	48315	1
IOWR	0.0473	0.00000	26	1

By comparing the data in the above six tables, the performance of nioⅱ/E and nioⅱ/F is still quite different. From the data in the table, the performance difference is more than 4 times, after the same processor optimization, the performance is improved by about 25% compared with before optimization. For-03 optimization and-0 s optimization, the performance is basically the same. In addition, usleep (1000), that is, the latency of 1 ms is set only when it is run in nioⅱ/F. The actual latency in nioⅱ/E isProgramAbout 4 times of the specified time (1 ms), that is, 4 ms.

according to the information provided by Altera:
In niosii performance benchmarks (alteras document) are this dmips ratio: niosii/F-1.105, niosii/S-0.518,
If the 50 MHz clock is used, the dmips of the cup is
niosii/F: 1.105*50 = 55.25 dmips
niosii/S: 0.518*50 = 25.9 dmips
niosii/E: 0.107*50 = 5.35 dmips
the condition for this performance is to run in onchip-MEM, and optimize is-03. Cycloneii device. The program in this article runs in external SDRAM