y450 archlinux cuda6.5
January 28, 2018 4:11 PM
archlinux是最新更新版本，gcc版本到了7.几，太新了。
[qiangge@lqspc ~]$ gcc --version
gcc (GCC) 7.2.1 20180116
Copyright © 2017 Free Software Foundation, Inc.
本程序是自由软件；请参看源代码的版权声明。本软件没有任何担保；
包括没有适销性和某一专用目的下的适用性担保。
这系统对中文翻译的不太习惯哈。
总体步骤
- 确认安装的archlinux比较新，不想降级gcc等。
- 确认y450的笔记本显卡型号，g 110M。
- 确定可以安装的cuda版本。这个地方走过弯路，开始直接
pa cuda，结果就给我装了个9.1的版本。反复测试发现安装失败。经过查询显卡型号（上一步）支持的计算能力（compute capability?希望没拼错）只是支持1.2以下，后来安装完发现是1.1.而1.2以下的最多安装cuda-6.5以前的版本。 yaourt cuda找到相关版本安装（上一步），安装过程中遇到/tmp不够用，新建个目录挂载到/tmp，冲掉了内存挂载的/tmp，这样可以充分利用硬盘空间来操作。之所以不够用因为内存只有8G，这样默认/tmp就只有4G，废话了。- 安装完后测试
/opt/cuda/samples的devicequery例子，最好拷贝到自己的/home目录吧。 - 开始不能编译任何例子，有两个错误。主要参考cuda社区解决。
(1)Here is a patch to /usr/include/bits/floatn.h for avoiding __FLOAT128 only when compiling via NVCC
(2)Here is how to use other GCC compiing via NVCC
- 第一个错误是floatn.h错误。参考论坛解决，本质上是判断条件里面添加一个条件，就是
不编译cuda代码的意思。 - 第二个错误是默认的gcc版本太新了，cuda65不支持，那就采用5试试看（参考下一步方法），发现这只能编译devicequery。于是经过google，知道必须4.7左右。本机yaourt编译4.7失败，当然依然要/tmp，编译个编译器真的很容易失败，浪费了好几天的电费哈。上海电费蛮贵的，尤其是租房，呜呜。那么总有解决办法吧，参考资料在archlinux的yaourt源里面。作者提到了要动态库加上软连接，
sudo ln -s /usr/lib/libisl.so /usr/lib/libisl.so.10 && sudo ldconfig
不然会失败，当然作为折腾专家，我必须先不加看看效果，果然不行
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.4/cc1plus: error while loading shared libraries: libisl.so.10: cannot open shared object file: No such file or directory
make: *** [Makefile:196：bandwidthTest.o] 错误 1
加上还提示另外一个错误，这个是作者没考虑的吧，哈哈
/usr/lib/gcc/x86_64-unknown-linux-gnu/4.7.4/cc1plus: error while loading shared libraries: libmpfr.so.4: cannot open shared object file: No such file or directory
解决办法是相同的思路，相似的代码，读者自行思考哈。
9. 解决gcc问题的方法有两个，本质是一个事情，请看参考1和参考2。最后的效果
[qiangge@lqspc ~]$ ll /opt/cuda/
bin/ jre/ libnvvp/ samples/
doc/ lib/ NVIDIA_SLA_cuDNN_Support.txt share/
extras/ lib64/ nvvm/ src/
include/ libnsight/ open64/ tools/
[qiangge@lqspc ~]$ ll /opt/cuda/bin/gcc/
总用量 8.0K
drwxr-xr-x 2 root 4.0K 1月 28 22:52 .
lrwxrwxrwx 1 root 16 1月 28 22:52 gcc -> /usr/bin/gcc-4.7
lrwxrwxrwx 1 root 16 1月 28 22:52 cpp -> /usr/bin/cpp-4.7
lrwxrwxrwx 1 root 16 1月 28 22:52 g++ -> /usr/bin/g++-4.7
drwxr-xr-x 4 root 4.0K 1月 22 09:45 ..
[qiangge@lqspc ~]$
[qiangge@lqspc 1_Utilities]$ cd bandwidthTest/
[qiangge@lqspc bandwidthTest]$ nvidia-smi
Mon Jan 29 00:01:22 2018
+------------------------------------------------------+
| NVIDIA-SMI 340.106 Driver Version: 340.106 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 GeForce G 110M Off | 0000:01:00.0 N/A | N/A |
| N/A 52C P12 N/A / N/A | 50MiB / 255MiB | N/A Default |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Compute processes: GPU Memory |
| GPU PID Process name Usage |
|=============================================================================|
| 0 Not Supported |
+-----------------------------------------------------------------------------+
[qiangge@lqspc bandwidthTest]$
[qiangge@lqspc bandwidthTest]$ ./bandwidthTest
[CUDA Bandwidth Test] - Starting...
Running on...
Device 0: GeForce G 110M
Quick Mode
Host to Device Bandwidth, 1 Device(s)
PINNED Memory Transfers
Transfer Size (Bytes) Bandwidth(MB/s)
33554432 2551.5
Device to Host Bandwidth, 1 Device(s)
PINNED Memory Transfers
Transfer Size (Bytes) Bandwidth(MB/s)
33554432 1675.0
Device to Device Bandwidth, 1 Device(s)
PINNED Memory Transfers
Transfer Size (Bytes) Bandwidth(MB/s)
33554432 6319.8
Result = PASS
[qiangge@lqspc bandwidthTest]$
[qiangge@lqspc 1_Utilities]$ cd deviceQuery
[qiangge@lqspc deviceQuery]$ ls
deviceQuery deviceQuery.cpp deviceQuery.o Makefile NsightEclipse.xml readme.txt
[qiangge@lqspc deviceQuery]$ ./deviceQuery
./deviceQuery Starting...
CUDA Device Query (Runtime API) version (CUDART static linking)
Detected 1 CUDA Capable device(s)
Device 0: "GeForce G 110M"
CUDA Driver Version / Runtime Version 6.5 / 6.5
CUDA Capability Major/Minor version number: 1.1
Total amount of global memory: 256 MBytes (268107776 bytes)
( 2) Multiprocessors, ( 8) CUDA Cores/MP: 16 CUDA Cores
GPU Clock rate: 1000 MHz (1.00 GHz)
Memory Clock rate: 700 Mhz
Memory Bus Width: 64-bit
Maximum Texture Dimension Size (x,y,z) 1D=(8192), 2D=(65536, 32768), 3D=(2048, 2048, 2048)
Maximum Layered 1D Texture Size, (num) layers 1D=(8192), 512 layers
Maximum Layered 2D Texture Size, (num) layers 2D=(8192, 8192), 512 layers
Total amount of constant memory: 65536 bytes
Total amount of shared memory per block: 16384 bytes
Total number of registers available per block: 8192
Warp size: 32
Maximum number of threads per multiprocessor: 768
Maximum number of threads per block: 512
Max dimension size of a thread block (x,y,z): (512, 512, 64)
Max dimension size of a grid size (x,y,z): (65535, 65535, 1)
Maximum memory pitch: 2147483647 bytes
Texture alignment: 256 bytes
Concurrent copy and kernel execution: Yes with 1 copy engine(s)
Run time limit on kernels: Yes
Integrated GPU sharing Host Memory: No
Support host page-locked memory mapping: Yes
Alignment requirement for Surfaces: Yes
Device has ECC support: Disabled
Device supports Unified Addressing (UVA): No
Device PCI Bus ID / PCI location ID: 1 / 0
Compute Mode:
< Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >
deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 6.5, CUDA Runtime Version = 6.5, NumDevs = 1, Device0 = GeForce G 110M
Result = PASS
[qiangge@lqspc deviceQuery]$
配置虽然低，学习可能够用吧，不行就去买个新点的台式二手显卡？二手是不是抠门了呢？的确是，但是其实自己不用买，公司有1080TI显卡，可以加班学习用就行了。这里只是想自己安装一次，并且可以简单用来学习、练习和测试。同时帮朋友解决了y550上cuda65，那个显卡是g 240m的样子，最多也是1.2的计算能力。但是他用的Ubuntu。臃肿的Ubuntu还不是我的菜。之后又发现自己硬盘快满了，原来是需要pacman -Sc一下了。回头考虑配置一下自动清除不安装的包吧。