首页 > 评测 > ARM中的R,一款Cortex R内核开发板:三星ARTIK 055开发体验

ARM中的R,一款Cortex R内核开发板:三星ARTIK 055开发体验

  • 作者:zhanzr
  • 来源:21ic
  • [导读]
  • 三星公司的Exynos系列处理器针对IoT市场目前推出了两款: Exynos i S111, Exynos i T200. S111是针对NB-IoT市场的, T200即本文评测的这款针对WiFi市场. 就T200而言, 无论是开发工具, 还是配套SDK的完善度, 都是比较优秀的. 但是缺点是来的太迟了, T200这款芯片可以大约对标乐鑫的ESP32.

Build Time: 2018-07-17 16:02:30

System Time: 01 Jan 2010, 00:00:00 [s] UTC Hardware RTC Support

ARTIK SDK version: 1.8.0

TASH>>demo_main

Sample:

1: channel: 0, value: 17

2: channel: 1, value: 21

3: channel: 2, value: 14

4: channel: 3, value: 18

cJSON Version: 1.7.7

{

"name": "Jack (\"Bee\") Nimble",

"format": {

"type": "rect",

"width": 1920.00000000000000000,

"height": 1080.00000000000000000,

"interlace": false,

"frame rate": 24.00000000000000000

}

}

["Sunday", "Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday"]

[[0.00000000000000000, -1.00000000000000000, 0.00000000000000000], [1.00000000000000000, 0.00000000000000000, 0.00000000000000000], [0.00000000000000000, 0.00000000000000000, 1.00000000000000000]]

{

"Image": {

"Width": 800.00000000000000000,

"Height": 600.00000000000000000,

"Title": "View from 15th Floor",

"Thumbnail": {

"Url": "http:/*www.example.com/image/481989943",

"Height": 125.00000000000000000,

"Width": "100"

},

"IDs": [116.00000000000000000, 943.00000000000000000, 234.00000000000000000, 38793.00000000000000000]

}

}

[{

"precision": "zip",

"Latitude": 37.76680000000000348,

"Longitude": -122.39589999999999748,

"Address": "",

"City": "SAN FRANCISCO",

"State": "CA",

"Zip": "94107",

"Country": "US"

}, {

"precision": "zip",

"Latitude": 37.37199100000000129,

"Longitude": -122.02599999999999625,

"Address": "",

"City": "SUNNYVALE",

"State": "CA",

"Zip": "94085",

"Country": "US"

}]

{

"number": null

}

Demo2: Coremark测试

对于新的处理器进行一些benchmark大概是所有程序员的一种强迫症, 至少作者是这样的人. 但是对于这个评测需要注意:除非做大刀阔斧的hack, 否则用户程序只是TizenRT这个操作系统的一个task. 如果要做严肃而公平的评测, 应该使用baremetal环境进行评测. 为了简单起见, 这里不做大刀阔斧的hack. 但是要注意, 这个coremark得分不是T200的Cortex R4内核的最权威最精确的体现.

还有要留意的是将优化等级配到最高: -O3. 实验证明-O0与-O3的结果相差4.3倍.

image17.png

图 优化等级配到最高

具体的修改请参考后面的代码连接, 这里直接贴出结果:

portable_init Sep 27 2018 02:28:11 start

2K performance run parameters for coremark.

CoreMark Size : 666

Total ticks : 3430

Total time (secs): 34

Iterations/Sec : 729

Iterations : 25000

Compiler version : GCC4.9.3 20150529 (release) [ARM/embedded-4_9-branch revision 227977]

Compiler flags : -std=gnu11 -O3 -ffunction-sections -fdata-sections -fno-builtin -fno-strict-aliasing -fno-strength-reduce -fomit-frame-pointer -Wstrict-prototypes -Wshadow -Wno-implicit-function-declaration -Wno-unused-function -Wno-unused-but-set-variable -DAPP_STACKSIZE=18432 -DAPP_PRIORITY=100 -mcpu=cortex-r4 -mfpu=vfpv3 -g

Memory location : STACK

seedcrc : 0xe9f5

[0]crclist : 0xe714

[0]crcmatrix : 0x1fd7

[0]crcstate : 0x8e3a

[0]crcfinal : 0xcc42

Correct operation validated. See readme.txt for run and reporting rules.

CoreMark 1.0 : 729 / GCC4.9.3 20150529 (release) [ARM/embedded-4_9-branch revision 227977] -std=gnu11 -O3 -ffunction-sections -fdata-sections -fno-builtin -fno-strict-aliasing -fno-strength-reduce -fomit-frame-pointer -Wstrict-prototypes -Wshadow -Wno-implicit-function-declaration -Wno-unused-function -Wno-unused-but-set-variable -DAPP_STACKSIZE=18432 -DAPP_PRIORITY=100 -mcpu=cortex-r4 -mfpu=vfpv3 -g / STACK

portable_fini Sep 27 2018 02:28:11 end

再啰嗦一遍, 这个结果不能代表T200的Cortex R4内核的最精确的Coremark得分. 能代表的是用户的普通APP开发模式下该内核的性能表现. 即使这样对于320MHz的内核也算不错了. 400MHz的STM32H7的得分是2000左右, 但那是Baremetal的得分.

  • 本文系21ic原创,未经许可禁止转载!

网友评论