Stack Overflow
Stack Overflow

Questions tagged [nvml]

A C-based API for monitoring and managing various states of the NVIDIA GPU devices.

A C-based API for monitoring and managing various states of the NVIDIA GPU devices.

34 questions
1
vote
0 answers

How to reset nvidia gpu after error in golang through golang-nvml

is there any command to reset nvidia gpu after error happened in golang through golang-nvml? i only found how to get GPU infos API in golang-nvml…
LibAndLab
  • 41
  • 4
1
vote
0 answers

How to check bus utilization / bus load for GPU during ML inference?

I am running an ML inference for image recognition on the GPU using onnxruntime and I am seeing an upper limit for how much performance improvement batching of images is giving me - there is reduction in inference time upto around batch_size of 8,…
sn710
  • 581
  • 5
  • 20
1
vote
1 answer

Which function returns "nvmlDevice_t" type variable in cuda/nvml library?

I am working with gpu's and want to get the serial numbers of the gpu's.In NVIDIA Management Library there is a function that I can use. The function prototype is. nvmlReturn_t nvmlDeviceGetSerial ( nvmlDevice_t device, char* serial, unsigned int …
arm
  • 56
  • 1
  • 12
1
vote
1 answer

How do NVML and NVAPI compare?

I want to get some basic GPU data: name, RAM size, and do temperature monitoring. From NVIDIA docs, it's not clear which library to use. Is NVAPI a legacy API which should be avoided?
Andrey Moiseev
  • 3,914
  • 7
  • 48
  • 64
1
vote
1 answer

nvmlDeviceGetPowerManagementMode() always returning NVML_ERROR_INVALID_ARGUMENT?

I am writing a code to measure the power usage of an NVIDIA Tesla K20 GPU (Kepler architecture) periodically using the NVML API. Variables: nvmlReturn_t result; nvmlEnableState_t pmmode; nvmlDevice_t nvmlDeviceID; unsigned int powerInt; Basic…
Kajal
  • 581
  • 11
  • 24
1
vote
1 answer

NVML code doesn't compile

I am implementing an example program with nvml library as shown at https://devtalk.nvidia.com/default/topic/504951/how-to-call-nvml-apis-/ The program is as follows: #include #include const char *…
Pranjal
  • 659
  • 1
  • 7
  • 24
1
vote
1 answer

Nvidia-smi showing fan speed as not available

My machine has nvidia Tesla K20m gpu. I would like to know gpu utilzation, memory utilization, temperature and fan speed. So I have used nvidia-smi to know the details. Nvidia-smi log is as follows ==============NVSMI LOG============== Timestamp …
Alvin
  • 940
  • 2
  • 13
  • 27
0
votes
0 answers

How to check the gpu wakeups related to call DxgkNotifyProcessFreezeCallout?

I have a tool(written in C#) that uses few nvidia third party apis to collect some data. nvapi.dll and nvml.dll This tool is running continously and once every 15 mins and uses the delegates from these dlls to collect some particular data. I have…
Jasmeet
  • 1,315
  • 11
  • 23
0
votes
0 answers

Why below NVML code is not getting compiled even when include directory added for nvml.h?

I am trying to build a C-code for NVML for A5000 GPU. I got a code from internet which is as below. #include #include ///usr/include/hwloc/nvml.h int main() { nvmlReturn_t result; nvmlDevice_t device; nvmlMemory_t…
Aravind D. Chakravarti
  • 69
  • 1
  • 8
0
votes
1 answer

Passing uint array to Dll to be populated (Nvidia NVML library)

I'm attempting to pass a uint array into the NVML function nvmlDeviceGetAccountingPids(Doc here) from C#, here's a minimum working sample of what I have so far: { public const string NVMLDLL = "nvml.dll"; public static class…
codingNewb
  • 470
  • 1
  • 8
  • 23
0
votes
0 answers

NVML cannot load methods "NVMLError_FunctionNotFound"

I have a 1080Ti GPU with CUDA 10.2, NVIDIA driver 440.59 and pynvml version 11.4.1 running on Ubuntu 16.04. 1. I am trying to profile my PyTorch code using scalene. When I run my code as scalene main.py I get the following error: Error in program…
user529295
  • 189
  • 2
  • 11
0
votes
1 answer

Nvidia nvml nvmlUnit_t query

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: …
user794667
  • 53
  • 5
0
votes
0 answers

Error: bool undeclared (first use in the function). Have already included in the code

I'm trying to compile Perl bindings to Nvidia Management library (NVML). When I use the makefile, following errors appear: /usr/lib/powerpc64le-linux-gnu/perl/5.22/CORE/handy.h:117:34: error: ‘bool’ undeclared (first use in this function) #define…
user3509540
  • 41
  • 1
  • 7
0
votes
1 answer

Using nvidia-smi what is the best strategy to capture power

I am using Tesla K20c and measuring power with nvidia-smi as my application is run. My problem is power consumption does not reach a steady state but keeps rising. For example, if my application runs for 100 iterations, power reaches 106W(in 4…
Tania
  • 418
  • 3
  • 20
0
votes
2 answers

NVML supported on Jetson TK1?

I installed NVML on Jetson TK1 and compiled a cuda program. The compilation does not show any error but when running it shows the error /NVML-installed-path/usr/src/gdk/nvml/lib//libnvidia-ml.so: file not recognized: File format not…
Barath
  • 1
  • 3
1
2
3