Valentin Churavy
Valentin Churavy
> Preference-loading depends on the path you launch Julia from. See below. It depends on your project/load path, not the path you launch Julia from. > For non-devved packages, if...
I don't think it's in a publication, but around 10mins in @tgymnich has some in his talk at EnzymeCon https://youtu.be/nPN_Z5j6JDM?feature=shared
@tgymnich do you remember what machine you used for these measurements?
8/16 should be a safe bet.
Could you try with Julia 1.10?
@afilogo thank you very much for your contribution. I am going to bring this PR to the finish line in #2476
cc: @Zentrik
On LLVM 16 ```julia using CUDA, LLVM.Interop function kernel_aligned(X, Y) assume(UInt(pointer(X)) % (2*sizeof(eltype(X))) == 0) @inbounds begin I = threadIdx().x v1 = CUDA.unsafe_cached_load(pointer(Y), 2*I-1,Val(8)) v2 = CUDA.unsafe_cached_load(pointer(Y), 2*I, Val(8)) v1...
So the backends that I am worried about is Metal and to a lesser extend Intel.
* **#562** 👈 (View in Graphite) * **#556** : 1 other dependent PR ([#558](https://github.com/JuliaGPU/KernelAbstractions.jl/pull/558) ) * `main` This stack of pull requests is managed by Graphite. Learn more about stacking.