Nvidia Memory Error

Just a guess but the opengl driver is probably trying to allocate more ram than you have available not gpu memory.

Nvidia memory error. Professional nvidia gpus the tesla and quadro products are equipped with error correcting code ecc memory which allows the system to detect when memory errors occur. The 970 has two different segments of memory it can access as nvidia recently explained to pc perspective. Reports runtime executions errors such as device or user stack overflows and illegal instructions. Reports detailed information about global memory access errors such as out of bounds accesses and misaligned memory accesses including instruction offet in cuda function kernel name or source file and line number.

If the cuda debugger detects an mmu fault when running a kernel it will not be able to specify the exact location of the fault. Smaller single bit errors are transparently corrected. This feature is used to prevent known degraded memory locations from being used. In this case enable the cuda memory checker and restart debugging and the cuda memory checker will pinpoint the exact statements that are triggering the fault.

Row remapping is a hardware mechanism to improve the reliability of frame buffer memory on nvidia a100. The cuda memory checker detects problems in global and shared memory. Nvidia a100 gpu memory error management da 09826 001 vr450 6 chapter 6.