CUDA报错: Cannot create Cublas handle. Cublas won't be available. 以及:Check failed: status == CUBLAS_STATUS_SUCCESS (1 vs. 0) CUBLAS_STATUS_NOT_INITIALIZED
Error描述:
aita@aita-Alienware-Area-51-R5:~/AITA2/daisida/ssd-github/caffe$ make runtest -j8
.build_release/tools/caffe
caffe: command line brew
usage: caffe <command> <args> commands:
train train or finetune a model
test score a model
device_query show GPU diagnostic information
time benchmark model execution time Flags from tools/caffe.cpp:
-gpu (Optional; run in GPU mode on given device IDs separated by ','.Use
'-gpu all' to run on all available GPUs. The effective training batch
size is multiplied by the number of devices.) type: string default: ""
-iterations (The number of iterations to run.) type: int32 default: 50
-level (Optional; network level.) type: int32 default: 0
-model (The model definition protocol buffer text file.) type: string
default: ""
-phase (Optional; network phase (TRAIN or TEST). Only used for 'time'.)
type: string default: ""
-sighup_effect (Optional; action to take when a SIGHUP signal is received:
snapshot, stop or none.) type: string default: "snapshot"
-sigint_effect (Optional; action to take when a SIGINT signal is received:
snapshot, stop or none.) type: string default: "stop"
-snapshot (Optional; the snapshot solver state to resume training.)
type: string default: ""
-solver (The solver definition protocol buffer text file.) type: string
default: ""
-stage (Optional; network stages (not to be confused with phase), separated
by ','.) type: string default: ""
-weights (Optional; the pretrained weights to initialize finetuning,
separated by ','. Cannot be set simultaneously with snapshot.)
type: string default: ""
.build_release/test/test_all.testbin 0 --gtest_shuffle
Cuda number of devices: 3
Setting to use device 0
Current device id: 0
Current device name: GeForce GTX 1080 Ti
Note: Randomizing tests' orders with a seed of 48866 .
[==========] Running 2361 tests from 309 test cases.
[----------] Global test environment set-up.
[----------] 7 tests from DetectionOutputLayerTest/2, where TypeParam = caffe::GPUDevice<float>
[ RUN ] DetectionOutputLayerTest/2.TestForwardShareLocationTopK
E0103 00:37:53.042623 19470 common.cpp:113] Cannot create Cublas handle. Cublas won't be available.
[ OK ] DetectionOutputLayerTest/2.TestForwardShareLocationTopK (219 ms)
[ RUN ] DetectionOutputLayerTest/2.TestForwardNoShareLocationNeg0TopK
[ OK ] DetectionOutputLayerTest/2.TestForwardNoShareLocationNeg0TopK (2 ms)
[ RUN ] DetectionOutputLayerTest/2.TestSetup
[ OK ] DetectionOutputLayerTest/2.TestSetup (1 ms)
[ RUN ] DetectionOutputLayerTest/2.TestForwardNoShareLocationNeg0
[ OK ] DetectionOutputLayerTest/2.TestForwardNoShareLocationNeg0 (2 ms)
[ RUN ] DetectionOutputLayerTest/2.TestForwardNoShareLocation
[ OK ] DetectionOutputLayerTest/2.TestForwardNoShareLocation (2 ms)
[ RUN ] DetectionOutputLayerTest/2.TestForwardShareLocation
[ OK ] DetectionOutputLayerTest/2.TestForwardShareLocation (1 ms)
[ RUN ] DetectionOutputLayerTest/2.TestForwardNoShareLocationTopK
[ OK ] DetectionOutputLayerTest/2.TestForwardNoShareLocationTopK (2 ms)
[----------] 7 tests from DetectionOutputLayerTest/2 (229 ms total) [----------] 2 tests from EuclideanLossLayerTest/2, where TypeParam = caffe::GPUDevice<float>
[ RUN ] EuclideanLossLayerTest/2.TestGradient
F0103 00:37:53.068140 19470 math_functions.cu:110] Check failed: status == CUBLAS_STATUS_SUCCESS (1 vs. 0) CUBLAS_STATUS_NOT_INITIALIZED
*** Check failure stack trace: ***
@ 0x7f9210daf5cd google::LogMessage::Fail()
@ 0x7f9210db1433 google::LogMessage::SendToLog()
@ 0x7f9210daf15b google::LogMessage::Flush()
@ 0x7f9210db1e1e google::LogMessageFatal::~LogMessageFatal()
@ 0x7f920c7ad43a caffe::caffe_gpu_dot<>()
@ 0x7f920c7ec7c3 caffe::EuclideanLossLayer<>::Forward_gpu()
@ 0x48ae96 caffe::Layer<>::Forward()
@ 0x48d445 caffe::GradientChecker<>::CheckGradientSingle()
@ 0x4aea53 caffe::GradientChecker<>::CheckGradientExhaustive()
@ 0x848f0c caffe::EuclideanLossLayerTest_TestGradient_Test<>::TestBody()
@ 0xa17c23 testing::internal::HandleExceptionsInMethodIfSupported<>()
@ 0xa1123a testing::Test::Run()
@ 0xa11388 testing::TestInfo::Run()
@ 0xa11465 testing::TestCase::Run()
@ 0xa1273f testing::internal::UnitTestImpl::RunAllTests()
@ 0xa12a63 testing::UnitTest::Run()
@ 0x47a98d main
@ 0x7f920ba46830 __libc_start_main
@ 0x483b49 _start
@ (nil) (unknown)
Makefile:526: recipe for target 'runtest' failed
make: *** [runtest] Aborted (core dumped)
解决方案1:
sudo rm -rf .nv/
解决方案2:
I realized that there was an error with my CUDA installation, specifically with the cuBLAS library. You can check if yours has the same problem by running the sample program simpleCUBLAS:
- cd /usr/local/cuda/samples/7_CUDALibraries/simpleCUBLAS # check if your samples are in the same directory
- make
- ./simpleCUBLAS
I was getting an error when I tried to run it, so I reinstalled CUDA 8.0 and it solved the issue.
此前尝试过:
CUDA_VISIBLE_DEVICES=2
原因是这个Demo默认使用所有探测到的CUDA 设备,而实验室的CUDA设备还有很多人在用,这会造成问题(可能是冲突或者资源不够,或者不被允许个人使用这么多个?)
所以在运行的命令行前面要加上一定的限制:
CUDA_VISIBLE_DEVICES=2 ./build/examples/openpose/openpose.bin --net_resolution "160x80" --video examples/media/video.avi
使得被探测到的设备数量只有两个。
然后发现还是运行不了,为什么呢?
因为OS X上的SHELL并不具备显示远程窗口的功能,
所以使用MobaXterm(在WIN10下),然后就成功了,因为这个软件直接内置了X server的接口
CUDA报错: Cannot create Cublas handle. Cublas won't be available. 以及:Check failed: status == CUBLAS_STATUS_SUCCESS (1 vs. 0) CUBLAS_STATUS_NOT_INITIALIZED的更多相关文章
- 【err】开启Persistence-M模式-Check failed: err == CUBLAS_STATUS_SUCCESS (1 vs. 0) : Create cublas handle failed
前言 安装好CUDA.CUDNN.NVIDIA driver之后,使用mxnet框架的时候出现该错误,本文记录该问题的解决方法. 环境 ubuntu 16.04 MxNet Cuda9.0 Nvidi ...
- caffe报错:cudnn.hpp:86] Check failed: status == CUDNN_STATUS_SUCCESS (3 vs. 0) CUDNN_STATUS_BAD_PARAM 原因
在实际项目中出现的该问题,起初以为是cudnn版本的问题,后来才定位到在网络进行reshape操作的时候 input_layer->Reshape({(), input_layer->sh ...
- Could not create cudnn handle: CUDNN_STATUS_INTERNAL_ERROR tensorflow-1.13.1和1.14windows版本目前不支持CUDA10.0
报错出现 Could not create cudnn handle: CUDNN_STATUS_INTERNAL_ERROR tensorflow-1.13.1和1.14windows版本目前不支持 ...
- Eclipse配置tomcat8.5.7报错:The Apache Tomcat installation at this directory is version 8.5.27. A Tomcat 8.0 installation is...
Eclipse配置tomcat8.5.7报错:The Apache Tomcat installation at this directory is version 8.5.27. A Tomcat ...
- 记CM+kerberos环境停电后无法启动报错An error: (java.security.PrivilegedActionException: javax.security.sasl.SaslException: GSS initiate failed [Caused by GSSException: No valid credentials provided (Mechanism leve
公司突然停电,然后cm环境无法重启,报错 An error: (java.security.PrivilegedActionException: javax.security.sasl.SaslExc ...
- YII报错笔记:<pre>PHP Notice 'yii\base\ErrorException' with message 'Uninitialized string offset: 0' in /my/test/project/iot/vendor/yiisoft/yii2/base/Model.php:778
YII常见报错笔记 报错返回的代码如下: <pre>PHP Notice 'yii\base\ErrorException' with message 'Uninitialized str ...
- 【CUDA开发】 Check failed: error == cudaSuccess (8 vs. 0) invalid device function
最近在复现R-CNN一系列的实验时,配置代码环境真是花费了不少时间.由于对MATLAB不熟悉,实验采用的都是github上rbg大神的Python版本.在配置Faster R-CNN时,编译没有问题, ...
- Spring-test使用JUnit时,测试类autowired报错,create bean error
Spring-test使用JUnit时,测试类里面使用autowired会报错, 报create bean error...... 但是controller里面@autowired可以正常运行的. 在 ...
- nfs客户端报错解决Stale file handle
NFS故障: 场景:客户端挂载是好的.服务端磁盘满了,重新给挂了一快.客户端df -h nfs挂载消失. 客户端报错:Stale file handle 现象如下: [root@test63-spri ...
随机推荐
- 【LOJ】#2494. 「AHOI / HNOI2018」寻宝游戏
题面 题解 第\(i\)个数之前的符号是或那么记为0,是与就记为1,得到一个二进数x 然后按位分开考虑,如果这一行是1那么记为1,如果这一位数位0记为0,得到一个二进制数\(b_i\) 第\(N\)行 ...
- H5中标签Canvas实现图像动画
一:主题部分 1.介绍 canvas可以实现画图功能,所以只要通过js的控制,就可以实现简单的动画效果. 这里主要是两个程序,一个小球来回上下弹跳,一个是吹气球. 2.弹跳程序 <!DOCTYP ...
- 002.NTP服务端搭建
一 安装及准备 1.1 安装NTP [root@server ~]# yum -y install ntp #也可下载之后rpm安装,或者源码安装 1.2 NTP服务地址 http://www.ntp ...
- POJ1743 Musical Theme [后缀数组+分组/并查集]
Musical Theme Time Limit: 1000MS Memory Limit: 30000K Total Submissions: 27539 Accepted: 9290 De ...
- [BZOJ2124]等差子序列/[CF452F]Permutation
[BZOJ2124]等差子序列/[CF452F]Permutation 题目大意: 一个\(1\sim n\)的排列\(A_{1\sim n}\),询问是否存在\(i,j(i<j)\),使得\( ...
- div中嵌套img元素,4px空白
写布局的过程中遇到一个问题,在div中嵌套了img元素,没有指定div的高度,嵌套之后div高度始终比img高4个像素,没有设置内外边距,找不到原因. 除非强制div的高度为img的高度,才能使div ...
- Western Subregional of NEERC, Minsk, Wednesday, November 4, 2015 Problem A. A + B
Problem A. A + B 题目连接: http://opentrains.snarknews.info/~ejudge/team.cgi?SID=c75360ed7f2c7022&al ...
- STM32F4 Timer Internal Trigger Connection
The Timers can be cascaded to make more complex timing relationships, or longer periods. Internally ...
- ARM LDR/STR, LDM/STM 指令
这里比较下容易混淆的四条指令,已经在这4条指令的混淆上花费了很多精力,现在做个小结,LDR,STR,LDM,STM这四条指令, 关于LDM和STM的说明,见另外一个说明文件,说明了这两个文件用于栈操作 ...
- ISO 7816-4: GET RESPONSE and ENVELOPE command
http://www.cardwerk.com/smartcards/smartcard_standard_ISO7816-4_7_transmission_interindustry_command ...