Skip to content

add fastmath flag #2732

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
wants to merge 1 commit into
base: master
Choose a base branch
from
Draft

add fastmath flag #2732

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 3 additions & 2 deletions src/CUDAKernels.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,9 +16,10 @@ export CUDABackend
struct CUDABackend <: KA.GPU
prefer_blocks::Bool
always_inline::Bool
fastmath::Bool
end

CUDABackend(; prefer_blocks=false, always_inline=false) = CUDABackend(prefer_blocks, always_inline)
CUDABackend(; prefer_blocks=false, always_inline=false, fastmath=false) = CUDABackend(prefer_blocks, always_inline, fastmath)

KA.allocate(::CUDABackend, ::Type{T}, dims::Tuple) where T = CuArray{T}(undef, dims)
KA.zeros(::CUDABackend, ::Type{T}, dims::Tuple) where T = CUDA.zeros(T, dims)
Expand Down Expand Up @@ -100,7 +101,7 @@ function (obj::KA.Kernel{CUDABackend})(args...; ndrange=nothing, workgroupsize=n
maxthreads = nothing
end

kernel = @cuda launch=false always_inline=backend.always_inline maxthreads=maxthreads obj.f(ctx, args...)
kernel = @cuda launch=false always_inline=backend.always_inline fastmath=backend.fastmath maxthreads=maxthreads obj.f(ctx, args...)

# figure out the optimal workgroupsize automatically
if KA.workgroupsize(obj) <: KA.DynamicSize && workgroupsize === nothing
Expand Down