Nvidia nvml nvmlUnit_t query

Question

Finding that code examples for the nvml API for nvidia cards is just really sparse.

Before any nvml calls could be conducted CMAKE required:

target_link_libraries(04_nvml_testing "/usr/lib/x86_64-linux-gnu/libnvidia-ml.so")

Code snippet:

    nvmlReturn_t result;
    unsigned int temp;
    nvmlDevice_t device;

    result = nvmlInit();    

    nvmlUnit_t unit;
    unsigned int myint;
    result = nvmlUnitGetHandleByIndex(0, &unit);

I can read the GPU temperature fine, but getting the nvmlUnit_t value of the card is required before a lot of API calls can be made.

This code block inside Clion is kicking: NVML_ERROR_INVALID_ARGUMENT

Also there is references to 'available to s-Series devices' whatever that is..

Answer 1

This is not a full answer but a the API reference for this. ''' nvmlReturn_t nvmlUnitGetHandleByIndex (unsigned int index, nvmlUnit_t *unit)

The index of the target unit, >= 0 and < unitCount unit Reference in which to return the unit handle

Returns ‣ NVML_SUCCESS if unit has been set

‣ NVML_ERROR_UNINITIALIZED if the library has not been successfully initialized

‣ NVML_ERROR_INVALID_ARGUMENT if index is invalid or unit is NULL

‣ NVML_ERROR_UNKNOWN on any unexpected error

Description

Acquire the handle for a particular unit, based on its index.

For S-class products. << - what is a S-class product.

Valid indices are derived from the unitCount returned by

nvmlUnitGetCount(). For example, if unitCount is 2 the valid indices

are 0 and 1, corresponding to UNIT 0 and UNIT 1.