Optimize moe_align_block_size for deepseek_v3 (#12850)

Signed-off-by: mgoin <mgoin64@gmail.com>

csrc/moe/moe_align_sum_kernels.cu

@@ -198,26 +198,27 @@ __global__ void moe_align_block_size_global_mem_kernel(
 }
 
 // taken from
-// https://github.com/sgl-project/sglang/commit/ded9fcd09a43d5e7d5bb31a2bc3e9fc21bf65d2a
+// https://github.com/sgl-project/sglang/commit/cdae77b03dfc6fec3863630550b45bbfc789f957
 template <typename scalar_t>
 __global__ void sgl_moe_align_block_size_kernel(
     scalar_t* __restrict__ topk_ids, int32_t* sorted_token_ids,
     int32_t* expert_ids, int32_t* total_tokens_post_pad, int32_t num_experts,
     int32_t block_size, size_t numel, int32_t* cumsum) {
   __shared__ int32_t shared_counts[32][8];
-  __shared__ int32_t local_offsets[256];
 
   const int warp_id = threadIdx.x / 32;
-  const int lane_id = threadIdx.x % 32;
   const int experts_per_warp = 8;
   const int my_expert_start = warp_id * experts_per_warp;
 
+  // Initialize shared_counts for this warp's experts
   for (int i = 0; i < experts_per_warp; ++i) {
     if (my_expert_start + i < num_experts) {
       shared_counts[warp_id][i] = 0;
     }
   }
 
+  __syncthreads();
+
   const size_t tokens_per_thread = CEILDIV(numel, blockDim.x);
   const size_t start_idx = threadIdx.x * tokens_per_thread;
 
@@ -230,6 +231,7 @@ __global__ void sgl_moe_align_block_size_kernel(
 
   __syncthreads();
 
+  // Single thread computes cumulative sum and total tokens
   if (threadIdx.x == 0) {
     cumsum[0] = 0;
     for (int i = 1; i <= num_experts; ++i) {
@@ -246,19 +248,28 @@ __global__ void sgl_moe_align_block_size_kernel(
 
   __syncthreads();
 
+  // Assign expert IDs to blocks
   if (threadIdx.x < num_experts) {
     for (int i = cumsum[threadIdx.x]; i < cumsum[threadIdx.x + 1];
          i += block_size) {
       expert_ids[i / block_size] = threadIdx.x;
     }
-    local_offsets[threadIdx.x] = cumsum[threadIdx.x];
   }
+}
 
-  __syncthreads();
-
-  for (int i = start_idx; i < numel && i < start_idx + tokens_per_thread; ++i) {
+// taken from
+// https://github.com/sgl-project/sglang/commit/cdae77b03dfc6fec3863630550b45bbfc789f957
+template <typename scalar_t>
+__global__ void sgl_moe_token_sort_kernel(scalar_t* __restrict__ topk_ids,
+                                          int32_t* sorted_token_ids,
+                                          int32_t* cumsum_buffer,
+                                          size_t numel) {
+  const size_t tid = blockIdx.x * blockDim.x + threadIdx.x;
+  const size_t stride = blockDim.x * gridDim.x;
+
+  for (size_t i = tid; i < numel; i += stride) {
     int32_t expert_id = topk_ids[i];
-    int32_t rank_post_pad = atomicAdd(&local_offsets[expert_id], 1);
+    int32_t rank_post_pad = atomicAdd(&cumsum_buffer[expert_id], 1);
     sorted_token_ids[rank_post_pad] = i;
   }
 }
@@ -377,23 +388,34 @@ void sgl_moe_align_block_size(torch::Tensor topk_ids, int64_t num_experts,
                               torch::Tensor experts_ids,
                               torch::Tensor num_tokens_post_pad) {
   const cudaStream_t stream = at::cuda::getCurrentCUDAStream();
+  TORCH_CHECK(num_experts == 256,
+              "sgl_moe_align_block_size kernel only supports deepseek v3.");
+
   VLLM_DISPATCH_INTEGRAL_TYPES(
       topk_ids.scalar_type(), "sgl_moe_align_block_size_kernel", [&] {
-        // calc needed amount of shared mem for `tokens_cnts` and `cumsum`
-        // tensors
+        // calc needed amount of shared mem for `cumsum` tensors
         auto options_int =
             torch::TensorOptions().dtype(torch::kInt).device(topk_ids.device());
-        // torch::Tensor token_cnts_buffer =
-        //     torch::empty({(num_experts + 1) * num_experts}, options_int);
         torch::Tensor cumsum_buffer =
-            torch::empty({num_experts + 1}, options_int);
+            torch::zeros({num_experts + 1}, options_int);
 
-        auto kernel = vllm::moe::sgl_moe_align_block_size_kernel<scalar_t>;
-        kernel<<<1, 1024, 0, stream>>>(
+        auto align_kernel =
+            vllm::moe::sgl_moe_align_block_size_kernel<scalar_t>;
+        align_kernel<<<1, 1024, 0, stream>>>(
             topk_ids.data_ptr<scalar_t>(), sorted_token_ids.data_ptr<int32_t>(),
             experts_ids.data_ptr<int32_t>(),
             num_tokens_post_pad.data_ptr<int32_t>(), num_experts, block_size,
             topk_ids.numel(), cumsum_buffer.data_ptr<int32_t>());
+
+        const int block_threads = 256;
+        const int num_blocks =
+            (topk_ids.numel() + block_threads - 1) / block_threads;
+        const int max_blocks = 65535;
+        const int actual_blocks = std::min(num_blocks, max_blocks);
+        auto sort_kernel = vllm::moe::sgl_moe_token_sort_kernel<scalar_t>;
+        sort_kernel<<<actual_blocks, block_threads, 0, stream>>>(
+            topk_ids.data_ptr<scalar_t>(), sorted_token_ids.data_ptr<int32_t>(),
+            cumsum_buffer.data_ptr<int32_t>(), topk_ids.numel());
       });
 }
 

vllm/model_executor/layers/fused_moe/fused_moe.py

@@ -596,7 +596,7 @@ def moe_align_block_size(
                                       dtype=torch.int32,
                                       device=topk_ids.device)
     if num_experts >= 224:
-        if envs.VLLM_ENABLE_MOE_ALIGN_BLOCK_SIZE_TRITON:
+        if envs.VLLM_ENABLE_MOE_ALIGN_BLOCK_SIZE_TRITON or num_experts != 256:
             moe_align_block_size_triton(
                 topk_ids,
                 num_experts,
@@ -606,6 +606,7 @@ def moe_align_block_size(
                 num_tokens_post_pad,
             )
         else:
+            # Currently requires num_experts=256
             ops.sgl_moe_align_block_size(
                 topk_ids,
                 num_experts,