# AMDGPU: LD.LLD Problem

**URL:** <https://discourse.julialang.org/t/amdgpu-ld-lld-problem/96878>\
**Category:** New to Julia\
**Tags:** amdgpu\
**Created:** [March 31, 2023, 5:54am UTC](https://discourse.julialang.org/t/amdgpu-ld-lld-problem/96878 "2023-03-31T05:54:28Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![0xFF](https://avatars.discourse-cdn.com/v4/letter/0/5fc32e/32.png) [@0xFF](https://discourse.julialang.org/u/0xFF)\
**Post date:** [March 31, 2023, 5:54am UTC](https://discourse.julialang.org/t/amdgpu-ld-lld-problem/96878/1 "2023-03-31T05:54:28Z")

</div>

Archlinux user, trying to install Julia for some primitive GPU calculations. ROCm is installed, Clang/LLVM too.

Running Julia as `JULIA_AMDGPU_CORE_MUST_LOAD=1 julia`, then `pkg> add AMDGPU`, then `test AMDGPU`.

Test fails with the following message:

```plaintext
Warning: Agent creation failed.
The GPU node has an unrecognized id.

┌ Warning: LLD is unavailable, compilation functionality will be disabled.
│ Reason: LLD_jll is not available on this platform
└ @ AMDGPU ~/.julia/packages/AMDGPU/bzHD4/src/AMDGPU.jl:280

...

ERROR: LoadError: InitError: Failed to find ld.lld, but ld.lld must exist, bailing out
Stacktrace:
 [1] __init__ ()
   @ AMDGPU ~/.julia/packages/AMDGPU/bzHD4/src/AMDGPU.jl:287

```

I’m not sure what’s my problem. I tried to play with LD path, but no luck so far. Any ideas?

---

<div class="post-metadata">

**Author:** ![giordano](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/giordano/32/2166_2.png) [@giordano](https://discourse.julialang.org/u/giordano)\
**Post date:** [April 1, 2023, 1:48am UTC](https://discourse.julialang.org/t/amdgpu-ld-lld-problem/96878/2 "2023-04-01T01:48:30Z")

</div>

How did you install Julia on archlinux? The package in the archlinux repository is [known to be broken](https://wiki.archlinux.org/title/Julia), if you used that please direct your complaints to the maintainers of that package and then use instead the [`julia-bin`](https://aur.archlinux.org/packages/julia-bin/) or [`juliaup`](https://aur.archlinux.org/packages/juliaup) packages in AUR. Note that juliaup allows you to maintain multiple versions of Julia at the same time.