cuda_memtest | Fork of CUDA GPU memtest eyeglasses

cuda_memtest is a C++ library typically used in Testing applications. cuda_memtest has no bugs, it has no vulnerabilities and it has low support. You can download it from GitHub.

Test 0 [Walking 1 bit] This test changes one bit a time in memory address to see it goes to a different memory location. It is designed to test the address wires. Test 1 [Own address test] Each Memory location is filled with its own address. The next kernel checks if the value in each memory location still agrees with the address. Test 2 [Moving inversions, ones&zeros] This test uses the moving inversions algorithm with patterns of all ones and zeros. Test 3 [Moving inversions, 8 bit pat] This is the same as test 1 but uses a 8 bit wide pattern of "walking" ones and zeros. This test will better detect subtle errors in "wide" memory chips. Test 4 [Moving inversions, random pattern] Test 4 uses the same algorithm as test 1 but the data pattern is a random number and it's complement. This test is particularly effective in finding difficult to detect data sensitive errors. The random number sequence is different with each pass so multiple passes increase effectiveness. Test 5 [Block move, 64 moves] This test stresses memory by moving block memories. Memory is initialized with shifting patterns that are inverted every 8 bytes. Then blocks of memory are moved around. After the moves are completed the data patterns are checked. Because the data is checked only after the memory moves are completed it is not possible to know where the error occurred. The addresses reported are only for where the bad pattern was found. Test 6 [Moving inversions, 32 bit pat] This is a variation of the moving inversions algorithm that shifts the data pattern left one bit for each successive address. The starting bit position is shifted left for each pass. To use all possible data patterns 32 passes are required. This test is quite effective at detecting data sensitive errors but the execution time is long. Test 7 [Random number sequence] This test writes a series of random numbers into memory. A block (1 MB) of memory is initialized with random patterns. These patterns and their complements are used in moving inversions test with rest of memory. Test 8 [Modulo 20, random pattern] A random pattern is generated. This pattern is used to set every 20th memory location in memory. The rest of the memory location is set to the complimemnt of the pattern. Repeat this for 20 times and each time the memory location to set the pattern is shifted right. Test 9 [Bit fade test, 90 min, 2 patterns] The bit fade test initializes all of memory with a pattern and then sleeps for 90 minutes. Then memory is examined to see if any memory bits have changed. All ones and all zero patterns are used. This test takes 3 hours to complete. The Bit Fade test is disabled by default. Test 10 [memory stress test] Stress memory as much as we can. A random pattern is generated and a kernel of large grid size and block size is launched to set all memory to the pattern. A new read and write kernel is launched immediately after the previous write kernel to check if there is any errors in memory and set the memory to the compliment. This process is repeated for 1000 times for one pattern. The kernel is written as to achieve the maximum bandwidth between the global memory and GPU. This will increase the chance of catching software error. In practice, we found this test quite useful to flush hardware errors as well.