龙芯3A5000LL与i7-10700的UnixBench跑分对比

龙芯

架构:           loongarch64
字节序:         Little Endian
CPU:             4
在线 CPU 列表:  0-3
每个核的线程数: 1
每个座的核数:   4
座:             1
NUMA 节点:      1
CPU 系列:       Loongson-64bit
型号名称:       Loongson-3A5000LL
BogoMIPS:       4600.00
L1d 缓存:       64K
L1i 缓存:       64K
L2 缓存:        256K
L3 缓存:        16384K

英特尔

架构:        x86_64
处理器操作模式:      32-bit, 64-bit
字节序:          Little Endian
地址大小:       39 bits physical, 48 bits virtual
CPU:              16
在线 CPU 列表: 0-15
每个核的线程数:  2
每个座的核数:  8
座:           1
NUMA 节点:        1
Vendor ID:           GenuineIntel
型号:               165
型号名称:          Intel(R) Core(TM) i7-10700 CPU @ 2.90GHz
CPU MHz:             2900.000
CPU max MHz:         4800.0000
CPU min MHz:         800.0000
BogoMIPS:            5799.77
虚拟化:      VT-x
L1d 缓存:           32K
L1i 缓存:           32K
L2 缓存:            256K
L3 缓存:            16384K

龙芯跑分:

4 CPUs in system; running 1 parallel copy of testsDhrystone 2 using register variables       29466399.7 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3950.2 MWIPS (9.4 s, 7 samples)
Execl Throughput                               4801.2 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        574277.5 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          154530.5 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1661865.0 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1112434.3 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 288865.4 lps   (10.0 s, 7 samples)
Process Creation                              11881.8 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   8327.0 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   2704.2 lpm   (60.0 s, 2 samples)
System Call Overhead                        1217616.5 lps   (10.0 s, 7 samples)System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   29466399.7   2525.0
Double-Precision Whetstone                       55.0       3950.2    718.2
Execl Throughput                                 43.0       4801.2   1116.6
File Copy 1024 bufsize 2000 maxblocks          3960.0     574277.5   1450.2
File Copy 256 bufsize 500 maxblocks            1655.0     154530.5    933.7
File Copy 4096 bufsize 8000 maxblocks          5800.0    1661865.0   2865.3
Pipe Throughput                               12440.0    1112434.3    894.2
Pipe-based Context Switching                   4000.0     288865.4    722.2
Process Creation                                126.0      11881.8    943.0
Shell Scripts (1 concurrent)                     42.4       8327.0   1963.9
Shell Scripts (8 concurrent)                      6.0       2704.2   4507.0
System Call Overhead                          15000.0    1217616.5    811.7========
System Benchmarks Index Score                                        1342.8------------------------------------------------------------------------
4 CPUs in system; running 4 parallel copies of testsDhrystone 2 using register variables      117353855.4 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                    15802.3 MWIPS (9.4 s, 7 samples)
Execl Throughput                              17546.7 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        770178.5 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          200917.9 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       2579517.0 KBps  (30.0 s, 2 samples)
Pipe Throughput                             4430140.5 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                1119742.5 lps   (10.0 s, 7 samples)
Process Creation                              42837.8 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  20755.5 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   2977.2 lpm   (60.1 s, 2 samples)
System Call Overhead                        4115032.3 lps   (10.0 s, 7 samples)System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0  117353855.4  10056.0
Double-Precision Whetstone                       55.0      15802.3   2873.2
Execl Throughput                                 43.0      17546.7   4080.6
File Copy 1024 bufsize 2000 maxblocks          3960.0     770178.5   1944.9
File Copy 256 bufsize 500 maxblocks            1655.0     200917.9   1214.0
File Copy 4096 bufsize 8000 maxblocks          5800.0    2579517.0   4447.4
Pipe Throughput                               12440.0    4430140.5   3561.2
Pipe-based Context Switching                   4000.0    1119742.5   2799.4
Process Creation                                126.0      42837.8   3399.8
Shell Scripts (1 concurrent)                     42.4      20755.5   4895.2
Shell Scripts (8 concurrent)                      6.0       2977.2   4962.0
System Call Overhead                          15000.0    4115032.3   2743.4========
System Benchmarks Index Score                                        3445.2

i7-10700跑分:

16 CPUs in system; running 1 parallel copy of testsDhrystone 2 using register variables       59349705.1 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     9540.0 MWIPS (8.9 s, 7 samples)
Execl Throughput                               7200.7 lps   (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       1613123.6 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          425426.4 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       4720434.2 KBps  (30.0 s, 2 samples)
Pipe Throughput                             2990172.7 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 308606.1 lps   (10.0 s, 7 samples)
Process Creation                              15796.8 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  18735.9 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   8475.1 lpm   (60.0 s, 2 samples)
System Call Overhead                        3500159.7 lps   (10.0 s, 7 samples)System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   59349705.1   5085.7
Double-Precision Whetstone                       55.0       9540.0   1734.6
Execl Throughput                                 43.0       7200.7   1674.6
File Copy 1024 bufsize 2000 maxblocks          3960.0    1613123.6   4073.5
File Copy 256 bufsize 500 maxblocks            1655.0     425426.4   2570.6
File Copy 4096 bufsize 8000 maxblocks          5800.0    4720434.2   8138.7
Pipe Throughput                               12440.0    2990172.7   2403.7
Pipe-based Context Switching                   4000.0     308606.1    771.5
Process Creation                                126.0      15796.8   1253.7
Shell Scripts (1 concurrent)                     42.4      18735.9   4418.9
Shell Scripts (8 concurrent)                      6.0       8475.1  14125.2
System Call Overhead                          15000.0    3500159.7   2333.4========
System Benchmarks Index Score                                        2945.4------------------------------------------------------------------------
16 CPUs in system; running 16 parallel copies of testsDhrystone 2 using register variables      519769901.7 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                   111240.3 MWIPS (10.3 s, 7 samples)
Execl Throughput                              47864.5 lps   (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       1538356.3 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          397724.3 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       4889086.9 KBps  (30.0 s, 2 samples)
Pipe Throughput                            21435187.2 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                3008238.6 lps   (10.0 s, 7 samples)
Process Creation                             102306.7 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  99855.0 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                  12952.2 lpm   (60.0 s, 2 samples)
System Call Overhead                       10496632.4 lps   (10.0 s, 7 samples)System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0  519769901.7  44539.0
Double-Precision Whetstone                       55.0     111240.3  20225.5
Execl Throughput                                 43.0      47864.5  11131.3
File Copy 1024 bufsize 2000 maxblocks          3960.0    1538356.3   3884.7
File Copy 256 bufsize 500 maxblocks            1655.0     397724.3   2403.2
File Copy 4096 bufsize 8000 maxblocks          5800.0    4889086.9   8429.5
Pipe Throughput                               12440.0   21435187.2  17230.9
Pipe-based Context Switching                   4000.0    3008238.6   7520.6
Process Creation                                126.0     102306.7   8119.6
Shell Scripts (1 concurrent)                     42.4      99855.0  23550.7
Shell Scripts (8 concurrent)                      6.0      12952.2  21587.0
System Call Overhead                          15000.0   10496632.4   6997.8========
System Benchmarks Index Score                                       10944.6

结果:

从跑分结果来看,目前龙芯的单核性能还不到10代i7的一半,多核性能就更羸弱了,在使用过程中也有相当程度的卡顿,龙芯还需要加油啊。

跑分结果仅供参考,如果以后遇到其他版本的龙芯接着测试。

Published by

风君子

独自遨游何稽首 揭天掀地慰生平

发表回复

您的电子邮箱地址不会被公开。 必填项已用 * 标注