sortperm! fails when output is UInt vector

[chrstphrbrns]

julia> CUDA.versioninfo()
CUDA runtime 12.2, artifact installation
CUDA driver 12.0
NVIDIA driver 525.125.6

CUDA libraries: 
- CUBLAS: 12.2.4
- CURAND: 10.3.3
- CUFFT: 11.0.8
- CUSOLVER: 11.5.1
- CUSPARSE: 12.1.2
- CUPTI: 20.0.0
- NVML: 12.0.0+525.125.6

Julia packages: 
- CUDA: 4.4.0
- CUDA_Driver_jll: 0.6.0+2
- CUDA_Runtime_jll: 0.8.0+0

Toolchain:
- Julia: 1.11.0-DEV.304
- LLVM: 15.0.7
- PTX ISA support: 3.2, 4.0, 4.1, 4.2, 4.3, 5.0, 6.0, 6.1, 6.3, 6.4, 6.5, 7.0, 7.1, 7.2, 7.3, 7.4, 7.5
- Device capability support: sm_37, sm_50, sm_52, sm_53, sm_60, sm_61, sm_62, sm_70, sm_72, sm_75, sm_80, sm_86

2 devices:
  0: NVIDIA GeForce RTX 2080 Ti (sm_75, 10.013 GiB / 11.000 GiB available)
  1: NVIDIA GeForce RTX 2080 Ti (sm_75, 10.745 GiB / 11.000 GiB available)

julia> c=cu(collect(1:1000000));

julia> out=CuVector{UInt}(undef,length(c));

julia> sortperm!(out,c);
# ERROR: InvalidIRError: compiling MethodInstance for (::GPUArrays.var"#broadcast_kernel#26")(::CUDA.CuKernelContext, ::CuDeviceVector{UInt64, 1}, ::Base.Broadcast.Broadcasted{CUDA.CuArrayStyle{…}, Tuple{…}, typeof(identity), Tuple{…}}, ::Int64) resulted in invalid LLVM IR
# Reason: unsupported call to an unknown function (call to jl_f_tuple)
# ...

julia> out=CuVector{Int}(undef,length(c));

julia> sortperm!(out,c);
# no error

julia> out=CuVector{UInt32}(undef,length(c));

julia> sortperm!(out,c);
# no error

[maleadt]

The problem is that you're swapping out and c; the indices has to be the first argument.
It shouldn't error like that though, so let me push a fix.

[chrstphrbrns]

out is just an uninitialized output vector. c is the data I'm sorting (pointlessly, in this MWE)