Skip to content
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

Fix AMDGPU test #27

Merged
merged 1 commit into from
Feb 15, 2024
+4 −4
Merged

Fix AMDGPU test #27

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
8 changes: 4 additions & 4 deletions ext/JACCAMDGPU/JACCAMDGPU.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,8 +26,8 @@ function JACC.parallel_reduce(N::I, f::F, x...) where {I<:Integer,F<:Function}
blocks = ceil(Int, N / threads)
ret = AMDGPU.zeros(Float64, blocks)
rret = AMDGPU.zeros(Float64, 1)
@roc groupsize = threads gridsize = threads * blocks localmem = 512 * sizeof(Float64) _parallel_reduce_amdgpu(N, ret, f, x...)
@roc groupsize = threads gridsize = threads localmem = 512 * sizeof(Float64) reduce_kernel_amdgpu(blocks, ret, rret)
@roc groupsize = threads gridsize = threads * blocks _parallel_reduce_amdgpu(N, ret, f, x...)
@roc groupsize = threads gridsize = threads reduce_kernel_amdgpu(blocks, ret, rret)
return rret

end
Expand All @@ -40,8 +40,8 @@ function JACC.parallel_reduce((M, N), f::F, x...) where {F<:Function}
Nblocks = ceil(Int, N / Nthreads)
ret = AMDGPU.zeros(Float64, (Mblocks, Nblocks))
rret = AMDGPU.zeros(Float64, 1)
@roc groupsize = (Mthreads, Nthreads) gridsize = (Mblocks * Mthreads, Nblocks * Nthreads) localmem = 16 * 16 * sizeof(Float64) _parallel_reduce_amdgpu_MN((M, N), ret, f, x...)
@roc groupsize = (Mblocks, Nblocks) gridsize = (Mblocks, Nblocks) localmem = 16 * 16 * sizeof(Float64) reduce_kernel_amdgpu_MN((Mblocks, Nblocks), ret, rret)
@roc groupsize = (Mthreads, Nthreads) gridsize = (Mblocks * Mthreads, Nblocks * Nthreads) _parallel_reduce_amdgpu_MN((M, N), ret, f, x...)
@roc groupsize = (Mblocks, Nblocks) gridsize = (Mblocks, Nblocks) reduce_kernel_amdgpu_MN((Mblocks, Nblocks), ret, rret)
return rret
end

Expand Down
Loading