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

Replace cub::Min with cuda::minimum #4948

Merged
merged 6 commits into from
Mar 6, 2025
+18 −4
Merged

Replace cub::Min with cuda::minimum #4948

Changes from 3 commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
5ebec03
Replace `cub::Min` with `cuda::minimum`
miscco Feb 25, 2025
e021874
Avoid availability issue
miscco Feb 26, 2025
5ed169d
Merge branch 'branch-25.04' into drop_cub_min
miscco Feb 26, 2025
d18312a
Add FIXME comment
miscco Feb 27, 2025
2ca80b3
Fix formatting
miscco Feb 28, 2025
721e35c
Merge branch 'branch-25.04' into drop_cub_min
miscco Mar 5, 2025
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
20 changes: 16 additions & 4 deletions cpp/src/prims/detail/per_v_transform_reduce_e.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -589,6 +589,12 @@ __global__ static void per_v_transform_reduce_e_mid_degree(
}
}

#if CCCL_MAJOR_VERSION >= 3
using cuda::minimum;
#else
using minimum = cub::Min;
#endif

if (edge_partition_e_mask) {
if constexpr (update_major && std::is_same_v<ReduceOp, reduce_op::any<T>>) {
auto rounded_up_local_degree =
Expand All @@ -601,7 +607,7 @@ __global__ static void per_v_transform_reduce_e_mid_degree(
e_op_result = call_e_op(i);
}
first_valid_lane_id = WarpReduce(temp_storage[threadIdx.x / raft::warp_size()])
.Reduce(e_op_result ? lane_id : raft::warp_size(), cub::Min());
.Reduce(e_op_result ? lane_id : raft::warp_size(), minimum{});
first_valid_lane_id = __shfl_sync(raft::warp_full_mask(), first_valid_lane_id, int{0});
if (lane_id == first_valid_lane_id) { reduced_e_op_result = *e_op_result; }
if (first_valid_lane_id != raft::warp_size()) { break; }
Expand Down Expand Up @@ -634,7 +640,7 @@ __global__ static void per_v_transform_reduce_e_mid_degree(
e_op_result = call_e_op(i);
}
first_valid_lane_id = WarpReduce(temp_storage[threadIdx.x / raft::warp_size()])
.Reduce(e_op_result ? lane_id : raft::warp_size(), cub::Min());
.Reduce(e_op_result ? lane_id : raft::warp_size(), minimum{});
first_valid_lane_id = __shfl_sync(raft::warp_full_mask(), first_valid_lane_id, int{0});
if (lane_id == first_valid_lane_id) { reduced_e_op_result = *e_op_result; }
if (first_valid_lane_id != raft::warp_size()) { break; }
Expand Down Expand Up @@ -781,6 +787,12 @@ __global__ static void per_v_transform_reduce_e_high_degree(
}
}

#if CCCL_MAJOR_VERSION >= 3
using cuda::minimum;
#else
using minimum = cub::Min;
#endif

if (edge_partition_e_mask) {
if constexpr (update_major && std::is_same_v<ReduceOp, reduce_op::any<T>>) {
auto rounded_up_local_degree =
Expand All @@ -799,7 +811,7 @@ __global__ static void per_v_transform_reduce_e_high_degree(
.Reduce(e_op_result
? threadIdx.x
: per_v_transform_reduce_e_kernel_high_degree_reduce_any_block_size,
cub::Min());
minimum{});
if (threadIdx.x == 0) { output_thread_id = first_valid_thread_id; }
__syncthreads();
first_valid_thread_id = output_thread_id;
Expand Down Expand Up @@ -843,7 +855,7 @@ __global__ static void per_v_transform_reduce_e_high_degree(
.Reduce(e_op_result
? threadIdx.x
: per_v_transform_reduce_e_kernel_high_degree_reduce_any_block_size,
cub::Min());
minimum{});
if (threadIdx.x == 0) { output_thread_id = first_valid_thread_id; }
__syncthreads();
first_valid_thread_id = output_thread_id;
Expand Down