Valentin Churavy
Valentin Churavy
Yeah that's the scoped value performance issue that was introduced https://github.com/JuliaLang/julia/issues/53584#issuecomment-1979089217
You are on an Apple M3/M4? On my system there is a overhead, but it is much smaller. ``` [ Info: FD BenchmarkTools.Trial: 10000 samples with 996 evaluations. Range (min...
For the fourth derivative I start to see the overhead grow: ``` [ Info: Fourth derivative [ Info: FD BenchmarkTools.Trial: 10000 samples with 991 evaluations. Range (min … max): 28.631...
I run into this a while back, hooking into the perf mechanism mentioned earlier would be ideal since it would simplify the work needed for new languages. I wrote https://github.com/vchuravy/UProbes.jl...
Seen on PkgEval https://s3.amazonaws.com/julialang-reports/nanosoldier/pkgeval/by_hash/dabb822_vs_aacfcf0/MPIReco.primary.log
Generally we won't be able to read LLVM IR being produced by a newer version of LLVM. This is why I said earlier this summer that I believe for a...
> Will Julia 1.11 use LLVM 16? Maybe? It heavily depends on the bandwidth folks have (I currently can't work on it) and @gbaraldi is busy with a lot of...
Will run Oceananigans testuite over this alone later this today/this week.
@wsmoses some concerns by @jpsamaroo in https://github.com/JuliaGPU/GPUCompiler.jl/pull/359#issuecomment-1420848278
This is essentially the evolution I have in mind with #562 Aligning KA relatively closely to OpenCL/SPIRV semantics My milestones for KA 0.10 is essentially 1. CPU to POCL transition...