Gabriel Baraldi
Gabriel Baraldi
@nanosoldier `runbenchmarks(!"scalar", vs=":master")`
@nanosoldier `runbenchmarks(!"scalar", vs=":master")`
I took a big look at it. There's still a couple regressions, but it seems to be a pretty clear overall win. If anyone wants to take a further look...
Do we want to run a pkgeval? Im slightly worried about the fact that I had to modify passes.
@nanosoldier `runbenchmarks(ALL, vs=":master")`
`["union", "array", ("map", "*", "Float64", "(false, true)")]` seems to be quite a large regression in my mac 2x
@nanosoldier `runbenchmarks(!"scalar", vs=":master")`
Using ```julia function fib(n::Int) n @benchmark fib(20) BenchmarkTools.Trial: 1571 samples with 1 evaluation. Range (min … max): 2.895 ms … 7.492 ms ┊ GC (min … max): 0.00% … 0.00%...
I wonder if we should rely on hipconfig more for this. Because it would be nice to not rely on the ROCM_PATH, also the device libs can be put into...
https://github.com/JuliaPackaging/Yggdrasil/blob/master/0_RootFS/gcc_common.jl and https://github.com/JuliaPackaging/Yggdrasil/blob/master/0_RootFS/gcc_sources.jl are the important files. @iains, sorry for the ping, but you are probably the most knowledgeable person on this.