# Profiling Julia CUDA code missing 'CUDA HW'
**URL:** https://discourse.julialang.org/t/profiling-julia-cuda-code-missing-cuda-hw/76061
**Category:** GPU
**Created:** [February 9, 2022, 7:18am UTC](https://discourse.julialang.org/t/profiling-julia-cuda-code-missing-cuda-hw/76061 "2022-02-09T07:18:46Z")
**Posts on this page:** 8
**Page:** 1
### Author:  [@torrance](https://discourse.julialang.org/u/torrance)
#### Post date: [February 9, 2022, 7:18am UTC](https://discourse.julialang.org/t/profiling-julia-cuda-code-missing-cuda-hw/76061/1 "2022-02-09T07:18:46Z")
I’m attempting to profile a custom CUDA kernel using NVIDIA’s nsight-systems (ie. nsys), as recommended by the CUDA.jl project.
nsys correctly gives me profiling results for nvcc compiled code, however when I run it against Julia, I only see CPU thread entries. It is missing the _CUDA HW_ entry in the timeline where I am expecting to see information about the kernel execution, memory copies, etc.
Has anyone else experienced this, and is there a fix?
---
### Author:  [@maleadt](https://discourse.julialang.org/u/maleadt)
#### Post date: [February 9, 2022, 8:30am UTC](https://discourse.julialang.org/t/profiling-julia-cuda-code-missing-cuda-hw/76061/2 "2022-02-09T08:30:10Z")
Are you actually profiling GPU code? Can you try with a simple `CUDA.@profile CuArray([1]) .+ 1` or so?
---
### Author:  [@torrance](https://discourse.julialang.org/u/torrance)
#### Post date: [February 9, 2022, 8:47am UTC](https://discourse.julialang.org/t/profiling-julia-cuda-code-missing-cuda-hw/76061/3 "2022-02-09T08:47:10Z")
Yes I believe I am. I’ve reduced the minimum code to this:
```
nsys profile julia -e 'using CUDA; arr = rand(100); arrd = CuArray(arr); map(arrd) do x; sin(x); end; println(Array(arrd))'
```
Actually, if I force `--cuda-memory-usage=true`, I do in fact get a “CUDA HW” header, but it does not include any information about the CUDA threads. See attached screenshot for an example.
---
### Author:  [@maleadt](https://discourse.julialang.org/u/maleadt)
#### Post date: [February 9, 2022, 8:48am UTC](https://discourse.julialang.org/t/profiling-julia-cuda-code-missing-cuda-hw/76061/4 "2022-02-09T08:48:33Z")
Strange. Are there actual CUDA API calls in there? Can you share the .nsys-rep file?
---
### Author:  [@torrance](https://discourse.julialang.org/u/torrance)
#### Post date: [February 9, 2022, 8:53am UTC](https://discourse.julialang.org/t/profiling-julia-cuda-code-missing-cuda-hw/76061/5 "2022-02-09T08:53:24Z")
Sure please see the .nsys-rep file here: [Sign in to your account](https://curtin-my.sharepoint.com/:u:/g/personal/277966k_curtin_edu_au/ETHAh1saleNCt3ONCA_6gRABj1aLQCR9cWGqBO8hEml0iA?e=ykjfbZ)
---
### Author:  [@maleadt](https://discourse.julialang.org/u/maleadt)
#### Post date: [February 9, 2022, 9:01am UTC](https://discourse.julialang.org/t/profiling-julia-cuda-code-missing-cuda-hw/76061/6 "2022-02-09T09:01:07Z")
Kernel launches are definitely there, but the `CUDA profiling might have not been started correctly.` is suspicious. Maybe try upgrading to the latest NSight.
---
### Author:  [@maleadt](https://discourse.julialang.org/u/maleadt)
#### Post date: [February 9, 2022, 9:17am UTC](https://discourse.julialang.org/t/profiling-julia-cuda-code-missing-cuda-hw/76061/7 "2022-02-09T09:17:30Z")
Also, which version of CUDA.jl? Could you post the output of running under JULIA\_DEBUG=CUDA?
---
### Author:  [@torrance](https://discourse.julialang.org/u/torrance)
#### Post date: [February 9, 2022, 9:46am UTC](https://discourse.julialang.org/t/profiling-julia-cuda-code-missing-cuda-hw/76061/8 "2022-02-09T09:46:17Z")
I can confirm its happening with other non-Julia code. So this would appear to be a system issue, rather than Julia specific. I’ll try to get one of the HPC staff onto it.
Thanks for your help @maleadt Tim!