-
Notifications
You must be signed in to change notification settings - Fork 205
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
LLVM not found #681
Comments
That's strange. Could you post the full output of |
|
Well, that's weird. Can you try the following in a clean session: using Libdl
Libdl.dlopen("C:\Users\Emirhan Kurtulu�\AppData\Local\Programs\Julia 1.5.3\bin\LLVM.dll") |
� character is "ş" which is non-ascii, is there any chance that this is the problem? |
Maybe; try: @show Base.libllvm_path()
@show String(Base.libllvm_path())
using Libdl
Libdl.dlopen(String(Base.libllvm_path())) |
|
Oh right, you're using Julia 1.5 which doesn't have |
It seems to be finding the LLVM.dll but does not work during precompilation. I really could not back trace where the error occurs. |
And you can |
It seems I cannot, but Julia IO is capable of finding the file:
As far as I reviewed the Libdl library it is just an export of the inner C library so it is incapable of dealing with unicode characters. I do not believe that such an extension of Julia functionality is required; however, a warning (or directly an error since this will already result in unexpected behavior) must be given by CUDA library. If you concur, I can prepare a pull request on this. |
FWIW, this isn't a CUDA.jl issue, or CUDA related. It's an issue of Julia or LLVM.jl, but I can't transfer the issue outside of this organization. Can you try Julia 1.6 to see if the new mechanism there fixes this issue? |
It seems to work with 1.6. I also changed the directory names so that such a problem will never exist. Thank you. |
Describe the bug
I am using Julia 1.5.3. When I add Cuda package via Pkg.add("CUDA") and use it I am getting the following errors:
I tried to pin down the problem but could not get any leads. Pytorch and Tensorflow successfully uses my GPU without a problem. I tried using Knet.jl and Flux.jl to see if the problem is on the library but when I saw that I get the same error on both of them, I decided to submit the issue here. I checked "Julia 1.5.3\bin\LLVM.dll" manually to ascertain that the file exists.
Version info
Details on Julia:
julia> versioninfo()
Julia Version 1.5.3
Commit 788b2c77c1 (2020-11-09 13:37 UTC)
Platform Info:
OS: Windows (x86_64-w64-mingw32)
CPU: Intel(R) Core(TM) i7-10875H CPU @ 2.30GHz
WORD_SIZE: 64
LIBM: libopenlibm
LLVM: libLLVM-9.0.1 (ORCJIT, skylake)
The text was updated successfully, but these errors were encountered: