CUDA GPU memtest

4 Reviews

Downloads: 80 This Week

Last Update: 2015-09-04

Download

Get an email when there's a new version of CUDA GPU memtest

Linux

A GPU memory test utility for NVIDIA and AMD GPUs using well established patterns from memtest86/memtest86+ as well as additional stress tests. The tests are designed to find hardware and soft errors. The code is written in CUDA and OpenCL.

Features

bug fixes
opencl memory tests available for NVIDIA and AMD GPUs

Project Activity

See All Activity >

License

Other License

Follow CUDA GPU memtest

CUDA GPU memtest Web Site

Other Useful Business Software

Managed Cloud Hosting Platform | Nexcess

For growing digital businesses and engineering teams that need reliable, fully managed cloud infrastructure to run high-performance applications.

The managed cloud solution engineered for simplicity, with built-in governance and risk-mitigation, plus a bill you can actually forecast.

Learn More

Rate This Project

User Ratings

4.0 out of 5 stars

★★★★★

★★★★

★★★

★★

★

ease 1 of 5 2 of 5 3 of 5 4 of 5 5 of 5 0 / 5

features 1 of 5 2 of 5 3 of 5 4 of 5 5 of 5 0 / 5

design 1 of 5 2 of 5 3 of 5 4 of 5 5 of 5 0 / 5

support 1 of 5 2 of 5 3 of 5 4 of 5 5 of 5 0 / 5

User Reviews

Filter Reviews:

All

nsimakov Posted 2013-04-06

Just tested my new GPU with cudagpumemtest. Very useful tool. It allows to exclude one crucial component out of equation when you are trying to find out why your software is not working. It is a good addition to memtest86/memtest86+ test. I have tested all my CUDA-enabled GPUs with cudagpumemtest as I tested my RAM with memtest86/memtest86+. Very good practice. I am very glad that such tools exists
fabtar Posted 2012-09-29

Compile needs few steps on Opensuse, adding the scorot repository for cudatools and changing the path in order to include all nvcc executables. cuda memtest is very simple to use and is the only tools around which looks good, so, it rules. Little tip: You should document the expected output in case of error, what am I going to see when an error occours?
sepben Posted 2012-09-05

cannot compile GPU Memtest with Dev-CPP :( Compiler is saying: Compiler: Default compiler Building Makefile: "C:\Dev-Cpp\Makefile.win" Führt make... aus make.exe -f "C:\Dev-Cpp\Makefile.win" all g++.exe -c ../Temp/cuda_memtest-1.2.3/misc.cpp -o ../Temp/cuda_memtest-1.2.3/misc.o -I"C:/Dev-Cpp/lib/gcc/mingw32/3.4.2/include" -I"C:/Dev-Cpp/include/c++/3.4.2/backward" -I"C:/Dev-Cpp/include/c++/3.4.2/mingw32" -I"C:/Dev-Cpp/include/c++/3.4.2" -I"C:/Dev-Cpp/include" In file included from ../Temp/cuda_memtest-1.2.3/pthread.h:4, from ../Temp/cuda_memtest-1.2.3/cuda_memtest.h:45, from ../Temp/cuda_memtest-1.2.3/misc.cpp:4: ../Temp/cuda_memtest-1.2.3/sched.h:25: error: expected constructor, destructor, or type conversion before "int" ../Temp/cuda_memtest-1.2.3/sched.h:25: error: expected `,' or `;' before "int" In file included from ../Temp/cuda_memtest-1.2.3/pthread.h:5, from ../Temp/cuda_memtest-1.2.3/cuda_memtest.h:45, from ../Temp/cuda_memtest-1.2.3/misc.cpp:4: C:/Dev-Cpp/include/signal.h:46: error: expected constructor, destructor, or type conversion before "typedef" C:/Dev-Cpp/include/signal.h:46: error: expected `,' or `;' before "typedef" In file included from ../Temp/cuda_memtest-1.2.3/cuda_memtest.h:45, from ../Temp/cuda_memtest-1.2.3/misc.cpp:4: ../Temp/cuda_memtest-1.2.3/pthread.h:30: error: expected constructor, destructor, or type conversion before "typedef" ../Temp/cuda_memtest-1.2.3/pthread.h:30: error: expected `,' or `;' before "typedef" ../Temp/cuda_memtest-1.2.3/pthread.h:49: error: `_pthread_descr' does not name a type ../Temp/cuda_memtest-1.2.3/pthread.h:101: error: `_pthread_descr' does not name a type In file included from ../Temp/cuda_memtest-1.2.3/misc.cpp:4: ../Temp/cuda_memtest-1.2.3/cuda_memtest.h:52: error: expected constructor, destructor, or type conversion before "extern" ../Temp/cuda_memtest-1.2.3/cuda_memtest.h:52: error: expected `,' or `;' before "extern" ../Temp/cuda_memtest-1.2.3/cuda_memtest.h:53: error: declaration of `_iob' as array of references ../Temp/cuda_memtest-1.2.3/cuda_memtest.h:54: error: declaration of `_iob' as array of references ../Temp/cuda_memtest-1.2.3/cuda_memtest.h:58: error: thread-local storage not supported for this target make.exe: *** [../Temp/cuda_memtest-1.2.3/misc.o] Error 1 Ausführung beendet
leewa2000 Posted 2011-07-08

I'm glad that I found this utility. Nvidia doesn't really have many diagnostic tools for their GPUs, especially for the T10 based GPUs. We were able to confirm GPU memory errors with this utility. Many thanks to the developers.

Additional Project Details

Operating Systems

Linux

Intended Audience

Developers, System Administrators

Programming Language

C, C++

Related Categories

C++ Hardware Drivers, C Hardware Drivers

Registered

2009-06-01

Similar Business Software

Pogo Linux

Our best-selling Intel® Modular Servers and AMD Epyc™ Servers deliver next-generation 3rd Gen Intel® Xeon® Scalable, Intel Core i9®, and AMD Epyc™ processor innovations for cloud computing, virtualization, big data, highly-transactional workloads, and more. High-density space and...

See Software
RUGGEDCOM Edge Routers

RUGGEDCOM’s family of industrial Edge routers include the RUGGEDCOM RX1400 and the RUGGEDCOM RM1224, space-saving, small form-factor rugged devices that offer reliable, high-bandwidth WLAN or 4G LTE connectivity over short and long distances for remote networks in harsh environments. Designed...

See Software
Lenovo ThinkSystem Rack Servers

From growing small businesses to enterprise workloads, Lenovo rack servers offer unmatched value, flexibility, and industry-leading efficiency to meet mission-critical demands with legendary quality and reliability. Storage flexibility support for M.2, U.2 NVMe, 4x simple-swap/hot-swap 3.5-inch...

See Software
RB2011iLS-IN

Desktop metal case, 5xEthernet, 5xGigabit Ethernet, SFP cage, PoE out on port 10, 600MHz CPU, 64MB RAM, RouterOS L4. RB2011iLS is our first product with SFP cage for gigabit SFP transceivers and an ideal solution for FTTH (Fiber To The Home) implementations. The RB2011iLS has 11 ports, Gigabit...

See Software
Supermicro CloudDC

All-in-one rackmount platform for cloud data centers. Compact 2U system that supports up to two double-width GPUs in a 25.5" chassis. 4-12 SATA/SAS drive bays with optional full NVMe support in select SKUs. 2 or 4 PCI-E x16 slots and dual AIOM (OCP 3.0 superset) slots for maximum data...

See Software
hAP ax³

hAP ax³ is our most powerful AX device with the best wireless network coverage so far. It features a modern quad-core ARM CPU running at 1.8 GHz and enough memory (1GB RAM + 128 MB NAND) for most tasks. Complex firewall rules, IPsec hardware encryption, Wireguard, BGP, advanced routing, or...

See Software

Report inappropriate content

Managed Cloud Hosting Platform | Nexcess

For growing digital businesses and engineering teams that need reliable, fully managed cloud infrastructure to run high-performance applications.

The managed cloud solution engineered for simplicity, with built-in governance and risk-mitigation, plus a bill you can actually forecast.

Learn More

Recommended Projects

NVIDIA GPU Operator
NVIDIA GPU Operator creates/configures/manages GPUs atop Kubernetes
clinfo
clinfo - openCL INFOrmation Simple linux script that provides information regarding cpu-gpu components by querying devices and comparing them against known data. CPUs : Intel(>Netburst), GPUs : Amd (>3xxx), Nvidia(>8)
Ricks-Lab GPU Utilities
A set of utilities for monitoring and customizing GPU performance
NVIDIA Container Toolkit
Build and run Docker containers leveraging NVIDIA GPUs
Triton
Development repository for the Triton language and compiler