Daniel/master-thesis
1
0
Fork 0
Code Issues 3 Pull Requests Actions Packages Projects 1 Releases Wiki Activity

benchmarking: redid inbounds tests because first test run did not use inbounds everywhere
Some checks failed
CI / Julia ${{ matrix.version }} - ${{ matrix.os }} - ${{ matrix.arch }} - ${{ github.event_name }} (x64, ubuntu-latest, 1.10) (push) Has been cancelled
Details
CI / Julia ${{ matrix.version }} - ${{ matrix.os }} - ${{ matrix.arch }} - ${{ github.event_name }} (x64, ubuntu-latest, 1.6) (push) Has been cancelled
Details
CI / Julia ${{ matrix.version }} - ${{ matrix.os }} - ${{ matrix.arch }} - ${{ github.event_name }} (x64, ubuntu-latest, pre) (push) Has been cancelled
Details

Browse Source
This commit is contained in:
Daniel
2025-04-06 14:29:07 +02:00
parent 20fcbab4ca
commit 258d33c338
3 changed files with 5 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
package/src/Transpiler.jl
View File

@ -71,12 +71,12 @@ function evaluate(expressions::Vector{Expr}, variables::Matrix{Float32}, paramet
cudaResults = CuArray{Float32}(undef, variableCols, length(expressions))
# execute each kernel (also try doing this with Threads.@threads. Since we can have multiple grids, this might improve performance)
@inbounds for i in eachindex(kernels)
for kernel in kernels
# config = launch_configuration(kernels[i])
threads = min(variableCols, 256)
blocks = cld(variableCols, threads)
cudacall(kernels[i], (CuPtr{Float32},CuPtr{Float32},CuPtr{Float32}), cudaVars, cudaParams, cudaResults; threads=threads, blocks=blocks)
cudacall(kernel, (CuPtr{Float32},CuPtr{Float32},CuPtr{Float32}), cudaVars, cudaParams, cudaResults; threads=threads, blocks=blocks)
end
return cudaResults