fix dtk26.04 4nodes core dump.

Signed-off-by: root <root@host-10-212-17-3.cluster.local>

csrc/config.hpp

@@ -47,7 +47,7 @@ struct Config {
         EP_HOST_ASSERT(num_ranks <= NUM_MAX_NVL_PEERS or num_sms % (2 * NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL) == 0);
         const auto num_rdma_ranks = std::max(num_ranks / NUM_MAX_NVL_PEERS, 1);
         const auto num_nvl_ranks  = std::min(num_ranks, NUM_MAX_NVL_PEERS);
-        const int  num_channels   = num_ranks <=8 ? num_sms / 2 : num_sms / NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL;
+        const int  num_channels   = num_sms / 2;
 
         // 计算每个nvl通信数据包的数据量
         size_t num_single_nvl_bag_bytes =
@@ -83,7 +83,7 @@ struct Config {
         EP_HOST_ASSERT(num_ranks % NUM_MAX_NVL_PEERS == 0);
         EP_HOST_ASSERT(num_sms % NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL == 0);
         const int num_rdma_ranks = num_ranks / NUM_MAX_NVL_PEERS;
-        const int num_channels   = num_sms / NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL;
+        const int num_channels   = num_sms / 2;
 
         // 计算每个rdma通信数据包的数据量
         size_t num_single_rdma_bag_bytes = 

csrc/deep_ep.cu

@@ -809,8 +809,8 @@ Buffer::internode_dispatch(const torch::Tensor &x, const std::optional<torch::Te
     // here.
     pybind11::gil_scoped_release release;
 
-    const int num_channels = config.num_sms / NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL;
-    EP_HOST_ASSERT(config.num_sms % NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL == 0);
+    const int num_channels = config.num_sms / 2;
+    // EP_HOST_ASSERT(config.num_sms % NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL == 0);
     EP_HOST_ASSERT(0 < get_num_rdma_ranks() and get_num_rdma_ranks() <= NUM_MAX_RDMA_PEERS);
 
     bool cached_mode = cached_rdma_channel_prefix_matrix.has_value();
@@ -1130,8 +1130,8 @@ Buffer::internode_combine(
     const torch::Tensor &combined_nvl_head, const Config &config,
     std::optional<EventHandle> &previous_event, bool async, bool allocate_on_comm_stream) {
 #ifndef DISABLE_ROCSHMEM
-    const int num_channels = config.num_sms / NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL;
-    EP_HOST_ASSERT(config.num_sms % NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL == 0);
+    const int num_channels = config.num_sms / 2;
+    // EP_HOST_ASSERT(config.num_sms % NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL == 0);
 
     // Shape and contiguous checks
     EP_HOST_ASSERT(x.dim() == 2 and x.is_contiguous());

csrc/kernels/internode.cu

@@ -7,17 +7,6 @@
 
 #ifndef DISABLE_ROCSHMEM
 
-// 安全检查：确保宏已定义
-#ifndef HIP_VERSION_PATCH
-    #error "HIP_VERSION_PATCH not defined! Check your HIP installation."
-#endif
-// TODO: fix unroll warnings
-// #ifdef __clang__
-// #pragma clang diagnostic push
-// #pragma clang diagnostic ignored "-Wpass-failed"
-// #pragma clang diagnostic ignored "-Wdeprecated-volatile"
-// #endif // __clang__
-
 namespace deep_ep {
 
 namespace internode {
@@ -25,7 +14,7 @@ namespace internode {
 extern shmem_team_t cpu_rdma_team;
 
 struct SourceMeta {
-    int src_rdma_rank, is_token_in_nvl_rank_bits;   // sizeof(SourceMeta) = 8
+    int src_rdma_rank, is_token_in_nvl_rank_bits;
 
     EP_STATIC_ASSERT(NUM_MAX_NVL_PEERS == 8, "Invalid number of maximum NVL peers");
 
@@ -60,18 +49,16 @@ __host__ __device__ __forceinline__ int get_num_bytes_per_rdma_token(int hidden_
 
 __host__ __device__ __forceinline__ std::pair<int, int>
 get_rdma_clean_meta(int hidden_int4, int num_scales, int num_topk_idx, int num_topk_weights,
-                    int num_rdma_ranks, int num_rdma_recv_buffer_tokens, int num_channels) {
-
+                    int num_rdma_ranks, int num_rdma_recv_buffer_tokens, int num_sms) {
     // Return `int32_t` offset and count to clean
     return {(get_num_bytes_per_rdma_token(hidden_int4, num_scales, num_topk_idx, num_topk_weights) *
-           num_rdma_recv_buffer_tokens * num_rdma_ranks * 2 * num_channels) / sizeof(int),
-           (NUM_MAX_NVL_PEERS * 2 + 4) * num_rdma_ranks * 2 * num_channels};
+           num_rdma_recv_buffer_tokens * num_rdma_ranks * 2 * num_sms) / sizeof(int),
+           (NUM_MAX_NVL_PEERS * 2 + 4) * num_rdma_ranks * 2 * num_sms};
 }
-
 __host__ __device__ __forceinline__ std::pair<int, int>
 get_nvl_clean_meta(int hidden_int4, int num_scales, int num_topk_idx, int num_topk_weights,
                    int num_rdma_ranks, int num_nvl_ranks, int num_nvl_recv_buffer_tokens,
-                   int num_channels) {
+                   int num_sms) {
     // Return `int32_t` offset and to clean
     EP_STATIC_ASSERT(sizeof(SourceMeta) % sizeof(int) == 0,
                               "Invalid size of `SourceMeta`");
@@ -79,8 +66,8 @@ get_nvl_clean_meta(int hidden_int4, int num_scales, int num_topk_idx, int num_to
         (num_nvl_recv_buffer_tokens *
         (hidden_int4 * sizeof(int4) + num_scales * sizeof(float) + num_topk_idx * sizeof(int) +
         num_topk_weights * sizeof(float) + sizeof(SourceMeta)) *
-        num_nvl_ranks * num_channels) / sizeof(int),
-        num_nvl_ranks * (2 * num_rdma_ranks + 2) * num_channels,
+        num_nvl_ranks * num_sms) / sizeof(int),
+        num_nvl_ranks * (2 * num_rdma_ranks + 2) * num_sms,
     };
 }
 
@@ -92,9 +79,10 @@ __forceinline__ __device__ int translate_dst_rdma_rank(const int dst_rdma_rank,
 
 template <bool kLowLatencyMode>
 __forceinline__ __device__ void
-dushmem_barrier_with_same_gpu_idx(const shmem_team_t &rdma_team) {
+shmem_sync_with_same_gpu_idx(const shmem_team_t &rdma_team) {
     // NOTE: shmem_device_barrier_all() might be an issue as
     // it doesn't follow OpenSHMEM specification on ROCm
+    // kLowLatencyMode ? shmem_device_sync(rdma_team) : shmem_device_sync_all();
     kLowLatencyMode ? shmem_barrier(rdma_team) : shmem_device_barrier_all();
 }
 
@@ -123,7 +111,7 @@ notify_dispatch(const int *num_tokens_per_rank, int *moe_recv_counter_mapped, in
         // Communication with others
         // Global barrier: the first warp do intra-node sync, the second warp do internode sync
         if (thread_id == kWarpSize)
-            dushmem_barrier_with_same_gpu_idx<kLowLatencyMode>(rdma_team);
+            shmem_sync_with_same_gpu_idx<kLowLatencyMode>(rdma_team);
 
         barrier_block<NUM_MAX_NVL_PEERS>(barrier_signal_ptrs, nvl_rank);
 
@@ -175,7 +163,7 @@ notify_dispatch(const int *num_tokens_per_rank, int *moe_recv_counter_mapped, in
         __syncthreads();
 
         if (thread_id == 0)
-            dushmem_barrier_with_same_gpu_idx<kLowLatencyMode>(rdma_team);
+            shmem_sync_with_same_gpu_idx<kLowLatencyMode>(rdma_team);
 
         __syncthreads();
 
@@ -266,7 +254,7 @@ notify_dispatch(const int *num_tokens_per_rank, int *moe_recv_counter_mapped, in
 
         // Finally barrier
         if (thread_id == kWarpSize)
-            dushmem_barrier_with_same_gpu_idx<kLowLatencyMode>(rdma_team);
+            shmem_sync_with_same_gpu_idx<kLowLatencyMode>(rdma_team);
 
         barrier_block<NUM_MAX_NVL_PEERS>(barrier_signal_ptrs, nvl_rank);
     } else {
@@ -383,14 +371,12 @@ constexpr int get_num_topk_rdma_ranks(int num_rdma_ranks) {
     return num_rdma_ranks < 8 ? num_rdma_ranks : 8;
 }
 
-
-template <bool kLowLatencyMode,
-          int kNumRDMARanks,
-          bool kCachedMode,
+template <bool kLowLatencyMode, int kNumRDMARanks, bool kCachedMode,
           int kNumDispatchRDMASenderWarps,
           int kNumTopkRDMARanks = get_num_topk_rdma_ranks(kNumRDMARanks)>
-__global__ void __launch_bounds__(((1 + NUM_MAX_NVL_PEERS) * kWarpSize), 1)
-dispatch(int4 *recv_x, float *recv_x_scales, int64_t *recv_topk_idx, float *recv_topk_weights,
+__global__ void
+__launch_bounds__(((kNumDispatchRDMASenderWarps + 1 + NUM_MAX_NVL_PEERS) * kWarpSize), 1)
+    dispatch(int4 *recv_x, float *recv_x_scales, int64_t *recv_topk_idx, float *recv_topk_weights,
              SourceMeta *recv_src_meta, const int4 *x, const float *x_scales,
              const int64_t *topk_idx, const float *topk_weights, int *send_rdma_head,
              int *send_nvl_head, int *recv_rdma_channel_prefix_matrix,
@@ -403,739 +389,736 @@ dispatch(int4 *recv_x, float *recv_x_scales, int64_t *recv_topk_idx, float *recv
              int num_max_nvl_chunked_send_tokens, int num_max_nvl_chunked_recv_tokens, int rank,
              int num_ranks) {
     enum class WarpRole {
-        kRDMASender,            // 从x写入到RDMA发送缓存
-        kRDMASenderCoordinator, // 从RDMA发送缓存写入到远端rdma_rank接收缓存
-        kRDMAAndNVLForwarder,   // 从RDMA接收缓存转写到ipc nvl缓存
-        kForwarderCoordinator,  // 向远端RDMA确认接收
-        kNVLReceivers           // 从nvl缓存写入到recv_x
+        kRDMASender,
+        kRDMASenderCoordinator,
+        kRDMAAndNVLForwarder,
+        kForwarderCoordinator,
+        kNVLReceivers
     };
-#if !defined(FORCE_DUSHMEM_API) && !defined(ROCM_DISABLE_CTX)
-    __shared__ shmem_ctx_t ctx;
-    shmem_wg_ctx_create(&ctx);
-#endif
+
+    __shared__ rocshmem::rocshmem_ctx_t ctx;
+    rocshmem::rocshmem_wg_ctx_create(0, &ctx);
 
     const auto sm_id       = static_cast<int>(blockIdx.x);
     const auto num_threads = static_cast<int>(blockDim.x), num_warps = num_threads / kWarpSize;
-    const auto thread_id   = static_cast<int>(threadIdx.x), warp_id = thread_id / kWarpSize, lane_id = get_lane_id();
-    const auto num_channels = static_cast<int>(gridDim.x) / NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL,
-               channel_id   = sm_id / NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL;
+    const auto thread_id = static_cast<int>(threadIdx.x), warp_id = thread_id / kWarpSize,
+               lane_id      = get_lane_id();
+    const auto num_channels = static_cast<int>(gridDim.x) / 2, channel_id = sm_id / 2;
+    const bool is_forwarder = sm_id % 2 == 0;
     const auto rdma_rank = rank / NUM_MAX_NVL_PEERS, nvl_rank = rank % NUM_MAX_NVL_PEERS;
 
-    EP_STATIC_ASSERT(NUM_MAX_NVL_PEERS * sizeof(bool) == sizeof(uint64_t), "Invalid number of NVL peers");
-    EP_DEVICE_ASSERT(num_warps == 1 + NUM_MAX_NVL_PEERS);
-
     const auto role_meta = [=]() -> std::pair<WarpRole, int> {
-        if (sm_id % NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL == 0) {
-            if(warp_id < kNumDispatchRDMASenderWarps) {
-                return {WarpRole::kRDMASender, -1};
-            } else if(warp_id == kNumDispatchRDMASenderWarps) {
-                return {WarpRole::kRDMASenderCoordinator, -1};
-            }
-        } else if (sm_id % NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL == 1) {
-            if(warp_id < NUM_MAX_NVL_PEERS) {
+        if (is_forwarder) {
+            if (warp_id < NUM_MAX_NVL_PEERS) {
                 return {WarpRole::kRDMAAndNVLForwarder, (warp_id + channel_id) % NUM_MAX_NVL_PEERS};
             } else {
                 return {WarpRole::kForwarderCoordinator, warp_id - NUM_MAX_NVL_PEERS};
             }
+        } else if (warp_id < kNumDispatchRDMASenderWarps) {
+            return {WarpRole::kRDMASender, -1};
+        } else if (warp_id == kNumDispatchRDMASenderWarps) {
+            return {WarpRole::kRDMASenderCoordinator, -1};
         } else {
-            return {WarpRole::kNVLReceivers, (warp_id + channel_id + 1) % NUM_MAX_NVL_PEERS};
+            return {WarpRole::kNVLReceivers,
+                    (warp_id + channel_id - kNumDispatchRDMASenderWarps) % NUM_MAX_NVL_PEERS};
         }
     }();
 
     auto warp_role   = role_meta.first;
     auto target_rank = role_meta.second; // Not applicable for RDMA senders
+    EP_DEVICE_ASSERT(num_warps == kNumDispatchRDMASenderWarps + 1 + NUM_MAX_NVL_PEERS);
+
+    // Data checks
+    EP_DEVICE_ASSERT(num_topk <= kWarpSize);
 
     // RDMA symmetric layout
+    EP_STATIC_ASSERT(NUM_MAX_NVL_PEERS * sizeof(bool) == sizeof(uint64_t),
+                              "Invalid number of NVL peers");
     auto hidden_bytes = hidden_int4 * sizeof(int4);
-    auto num_bytes_per_rdma_token = get_num_bytes_per_rdma_token(hidden_int4, num_scales, num_topk, num_topk);
-    auto rdma_channel_data = SymBuffer<int8_t>(rdma_buffer_ptr, num_max_rdma_chunked_recv_tokens * num_bytes_per_rdma_token, kNumRDMARanks, channel_id, num_channels);
-    auto rdma_channel_meta = SymBuffer<int>(rdma_buffer_ptr, NUM_MAX_NVL_PEERS * 2 + 2, kNumRDMARanks, channel_id, num_channels);
-    auto rdma_channel_head = SymBuffer<uint64_t, false>(rdma_buffer_ptr, 1, kNumRDMARanks, channel_id, num_channels);
-    auto rdma_channel_tail = SymBuffer<uint64_t, false>(rdma_buffer_ptr, 1, kNumRDMARanks, channel_id, num_channels);
+    auto num_bytes_per_rdma_token =
+        get_num_bytes_per_rdma_token(hidden_int4, num_scales, num_topk, num_topk);
+    auto rdma_channel_data = SymBuffer<int8_t>(
+        rdma_buffer_ptr, num_max_rdma_chunked_recv_tokens * num_bytes_per_rdma_token, kNumRDMARanks,
+        channel_id, num_channels);
+    auto rdma_channel_meta = SymBuffer<int>(rdma_buffer_ptr, NUM_MAX_NVL_PEERS * 2 + 2,
+                                            kNumRDMARanks, channel_id, num_channels);
+    auto rdma_channel_head =
+        SymBuffer<uint64_t, false>(rdma_buffer_ptr, 1, kNumRDMARanks, channel_id, num_channels);
+    auto rdma_channel_tail =
+        SymBuffer<uint64_t, false>(rdma_buffer_ptr, 1, kNumRDMARanks, channel_id, num_channels);
 
     // NVL buffer layouts
-    // NOTES: `rs_wr_buffer_ptr` means "Read for Senders, Write for Receivers", `ws_rr_buffer_ptr` means "Write for Senders, Read for Receivers"
+    // NOTES: `rs_wr_buffer_ptr` means "Read for Senders, Write for Receivers", `ws_rr_buffer_ptr`
+    // means "Write for Senders, Read for Receivers"
     void *rs_wr_buffer_ptr = nullptr, *ws_rr_buffer_ptr = nullptr;
     int   rs_wr_rank = 0, ws_rr_rank = 0;
     if (warp_role == WarpRole::kRDMAAndNVLForwarder)
-        rs_wr_buffer_ptr = buffer_ptrs[nvl_rank], ws_rr_buffer_ptr = buffer_ptrs[target_rank], rs_wr_rank = nvl_rank, ws_rr_rank = target_rank;
+        rs_wr_buffer_ptr = buffer_ptrs[nvl_rank], ws_rr_buffer_ptr = buffer_ptrs[target_rank],
+        rs_wr_rank = nvl_rank, ws_rr_rank = target_rank;
     if (warp_role == WarpRole::kNVLReceivers)
-        rs_wr_buffer_ptr = buffer_ptrs[target_rank], ws_rr_buffer_ptr = buffer_ptrs[nvl_rank], rs_wr_rank = target_rank, ws_rr_rank = nvl_rank;
+        rs_wr_buffer_ptr = buffer_ptrs[target_rank], ws_rr_buffer_ptr = buffer_ptrs[nvl_rank],
+        rs_wr_rank = target_rank, ws_rr_rank = nvl_rank;
 
     // Allocate buffers
-    auto nvl_channel_x = AsymBuffer<int4>(ws_rr_buffer_ptr, num_max_nvl_chunked_recv_tokens * hidden_int4, NUM_MAX_NVL_PEERS, channel_id, num_channels, rs_wr_rank).advance_also(rs_wr_buffer_ptr);
-    auto nvl_channel_src_meta = AsymBuffer<SourceMeta>(ws_rr_buffer_ptr, num_max_nvl_chunked_recv_tokens, NUM_MAX_NVL_PEERS, channel_id, num_channels, rs_wr_rank).advance_also(rs_wr_buffer_ptr);
-    auto nvl_channel_x_scales = AsymBuffer<float>(ws_rr_buffer_ptr, num_max_nvl_chunked_recv_tokens * num_scales, NUM_MAX_NVL_PEERS, channel_id, num_channels, rs_wr_rank).advance_also(rs_wr_buffer_ptr);
-    auto nvl_channel_topk_idx = AsymBuffer<int>(ws_rr_buffer_ptr, num_max_nvl_chunked_recv_tokens * num_topk, NUM_MAX_NVL_PEERS, channel_id, num_channels, rs_wr_rank).advance_also(rs_wr_buffer_ptr);
-    auto nvl_channel_topk_weights = AsymBuffer<float>(ws_rr_buffer_ptr, num_max_nvl_chunked_recv_tokens * num_topk, NUM_MAX_NVL_PEERS, channel_id, num_channels, rs_wr_rank).advance_also(rs_wr_buffer_ptr);
-    auto nvl_channel_prefix_start = AsymBuffer<int>(ws_rr_buffer_ptr, kNumRDMARanks, NUM_MAX_NVL_PEERS, channel_id, num_channels, rs_wr_rank).advance_also(rs_wr_buffer_ptr);
-    auto nvl_channel_prefix_end = AsymBuffer<int>(ws_rr_buffer_ptr, kNumRDMARanks, NUM_MAX_NVL_PEERS, channel_id, num_channels, rs_wr_rank).advance_also(rs_wr_buffer_ptr);
-    auto nvl_channel_head = AsymBuffer<int>(rs_wr_buffer_ptr, 1, NUM_MAX_NVL_PEERS, channel_id, num_channels, ws_rr_rank).advance_also(ws_rr_buffer_ptr);
-    auto nvl_channel_tail = AsymBuffer<int>(ws_rr_buffer_ptr, 1, NUM_MAX_NVL_PEERS, channel_id, num_channels, rs_wr_rank).advance_also(rs_wr_buffer_ptr);
+    auto nvl_channel_x =
+        AsymBuffer<int4>(ws_rr_buffer_ptr, num_max_nvl_chunked_recv_tokens * hidden_int4,
+                         NUM_MAX_NVL_PEERS, channel_id, num_channels, rs_wr_rank)
+            .advance_also(rs_wr_buffer_ptr);
+    auto nvl_channel_src_meta =
+        AsymBuffer<SourceMeta>(ws_rr_buffer_ptr, num_max_nvl_chunked_recv_tokens, NUM_MAX_NVL_PEERS,
+                               channel_id, num_channels, rs_wr_rank)
+            .advance_also(rs_wr_buffer_ptr);
+    auto nvl_channel_x_scales =
+        AsymBuffer<float>(ws_rr_buffer_ptr, num_max_nvl_chunked_recv_tokens * num_scales,
+                          NUM_MAX_NVL_PEERS, channel_id, num_channels, rs_wr_rank)
+            .advance_also(rs_wr_buffer_ptr);
+    auto nvl_channel_topk_idx =
+        AsymBuffer<int>(ws_rr_buffer_ptr, num_max_nvl_chunked_recv_tokens * num_topk,
+                        NUM_MAX_NVL_PEERS, channel_id, num_channels, rs_wr_rank)
+            .advance_also(rs_wr_buffer_ptr);
+    auto nvl_channel_topk_weights =
+        AsymBuffer<float>(ws_rr_buffer_ptr, num_max_nvl_chunked_recv_tokens * num_topk,
+                          NUM_MAX_NVL_PEERS, channel_id, num_channels, rs_wr_rank)
+            .advance_also(rs_wr_buffer_ptr);
+    auto nvl_channel_prefix_start =
+        AsymBuffer<int>(ws_rr_buffer_ptr, kNumRDMARanks, NUM_MAX_NVL_PEERS, channel_id,
+                        num_channels, rs_wr_rank)
+            .advance_also(rs_wr_buffer_ptr);
+    auto nvl_channel_prefix_end =
+        AsymBuffer<int>(ws_rr_buffer_ptr, kNumRDMARanks, NUM_MAX_NVL_PEERS, channel_id,
+                        num_channels, rs_wr_rank)
+            .advance_also(rs_wr_buffer_ptr);
+    auto nvl_channel_head = AsymBuffer<int>(rs_wr_buffer_ptr, 1, NUM_MAX_NVL_PEERS, channel_id,
+                                            num_channels, ws_rr_rank)
+                                .advance_also(ws_rr_buffer_ptr);
+    auto nvl_channel_tail = AsymBuffer<int>(ws_rr_buffer_ptr, 1, NUM_MAX_NVL_PEERS, channel_id,
+                                            num_channels, rs_wr_rank)
+                                .advance_also(rs_wr_buffer_ptr);
 
     // RDMA sender warp synchronization
     __shared__ volatile int rdma_send_next_token_idx;
     __shared__ volatile int rdma_send_channel_tail[kNumRDMARanks];
     __shared__ volatile int rdma_send_channel_next_tail[kNumRDMARanks];
+    __shared__ volatile int rdma_sender_counter[1];
+    __shared__ volatile int rdma_forwarder_counter[1];
+    if (threadIdx.x == 0) {
+        rdma_sender_counter[0]    = 0;
+        rdma_forwarder_counter[0] = 0;
+    }
+    __syncthreads();
 
-    // NVL and RDMA coordinate Forward warp synchronization
+    auto sync_rdma_sender_smem = [&]() {
+        if (lane_id == 0) {
+            volatile int ret = __hip_atomic_fetch_add(&rdma_sender_counter[0], 1, __ATOMIC_RELAXED,
+                                                     __HIP_MEMORY_SCOPE_WORKGROUP);
+	    // volatile int ret = atomicAdd((int*)&rdma_sender_counter[0], 1);
+        }
+        syncwarp();
+        while (rdma_sender_counter[0] < (kNumDispatchRDMASenderWarps + 1)) {
+        }
+    };
+
+    // Forward warp synchronization
     __shared__ volatile int  forward_channel_head[NUM_MAX_NVL_PEERS][kNumRDMARanks];
     __shared__ volatile bool forward_channel_retired[NUM_MAX_NVL_PEERS];
+    // NOTE: Not sure that __syncthreads() is a suitable replacement
+    auto sync_forwarder_smem = [&]() {
+        if (lane_id == 0) {
+            volatile int ret = __hip_atomic_fetch_add(
+                 &rdma_forwarder_counter[0], 1, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_WORKGROUP);
+	    // volatile int ret = atomicAdd((int*)&rdma_forwarder_counter[0], 1);
+        }
+        syncwarp();
+        while (rdma_forwarder_counter[0] < (NUM_MAX_NVL_PEERS + 1)) {
+        }
+    };
 
-    // Place the main logic of your kernel here, using the parameters above.
-    if(warp_role == WarpRole::kRDMASender) {
-        /*
-        这段代码的主要功能是在一个CUDA内核中协调多个线程之间的RDMA发送操作。
-        它首先获取当前通道的任务范围，然后清理共享内存，接着计算并发送本通道中的令牌数量。
-        然后，它遍历所有的令牌，读取每个令牌的RDMA秩的存在性，获取顺序锁，计算下一个尾部位置，存储RDMA头部，更新最后一个令牌尾部，释放顺序锁，并广播尾部位置。
-        最后，它复制相关的数据到对称发送缓冲区。
-
-        kRDMASender主要目的是将发送信息x， x_scale，source_meta, topk_idx, topk_weight等信息填充进入rdma发送缓存，
-        期间要同步warp直接对token的依序操作，以及和kForwarderCoordinator, kRDMASenderCoordinator内存同步。
-        同时在复制操作时， 使用ld.global.nc.L1::no_allocate.L2::256B， st.global.L1::no_allocate减少L1/L2缓存使用。
-        */
-        // 获取任务范围
+    if (warp_role == WarpRole::kRDMASender) {
+        // Get tasks
         int token_start_idx, token_end_idx;
-        get_channel_task_range(num_tokens, num_channels, channel_id, token_start_idx, token_end_idx);
+        get_channel_task_range(num_tokens, num_channels, channel_id, token_start_idx,
+                               token_end_idx);
 
-        // 清理共享内存
-        EP_STATIC_ASSERT(kNumRDMARanks <= kWarpSize, "无效的RDMA秩数量");
-        if(warp_id == 0 && lane_id == 0) {
-            rdma_send_next_token_idx = token_start_idx;
-        }
-        if(warp_id == 0 && lane_id < kNumRDMARanks) {
-            rdma_send_channel_tail[lane_id]      = 0;
-            rdma_send_channel_next_tail[lane_id] = 0;
-        }
+        // Clean shared memory
+        EP_STATIC_ASSERT(kNumRDMARanks <= kWarpSize, "Invalid number of RDMA ranks");
+        (warp_id == 0 and lane_id == 0) ? (rdma_send_next_token_idx = token_start_idx) : 0;
+        (warp_id == 0 and lane_id < kNumRDMARanks) ? (rdma_send_channel_tail[lane_id] = 0) : 0;
+        (warp_id == 0 and lane_id < kNumRDMARanks) ? (rdma_send_channel_next_tail[lane_id] = 0) : 0;
 
-        // 发送本通道中的令牌数量，通过 `-value - 1` 表示
-        EP_STATIC_ASSERT(NUM_MAX_NVL_PEERS * 2 + 2 <= kWarpSize, "无效的NVL对等体数量");
-        // 对于每个目标RDMA秩，以warp为单位进行迭代。计算发送缓冲区的值，并存储在rdma_channel_meta.send_buffer中
-        // 用于填充rdma_channel_meta.send_buffer本节点发送到远端rank, rdma_rank的起始index和结束index
-        for(int dst_rdma_rank = warp_id; dst_rdma_rank < kNumRDMARanks; dst_rdma_rank += kNumDispatchRDMASenderWarps) {
-            auto dst_ptr = dst_rdma_rank == rdma_rank ? rdma_channel_meta.recv_buffer(dst_rdma_rank) : rdma_channel_meta.send_buffer(dst_rdma_rank);
+        // Send number of tokens in this channel by `-value - 1`
+        EP_STATIC_ASSERT(NUM_MAX_NVL_PEERS * 2 + 2 <= kWarpSize,
+                                  "Invalid number of NVL peers");
+        for (int dst_rdma_rank = warp_id; dst_rdma_rank < kNumRDMARanks;
+             dst_rdma_rank += kNumDispatchRDMASenderWarps) {
             if (lane_id < NUM_MAX_NVL_PEERS) {
-                dst_ptr[lane_id] = -(channel_id == 0 ? 0 : gbl_channel_prefix_matrix[(dst_rdma_rank * NUM_MAX_NVL_PEERS + lane_id) * num_channels + channel_id - 1]) - 1;
+                rdma_channel_meta.send_buffer(dst_rdma_rank)[lane_id] =
+                    -(channel_id == 0
+                          ? 0
+                          : gbl_channel_prefix_matrix
+                                [(dst_rdma_rank * NUM_MAX_NVL_PEERS + lane_id) * num_channels +
+                                 channel_id - 1]) -
+                    1;
             } else if (lane_id < NUM_MAX_NVL_PEERS * 2) {
-                dst_ptr[lane_id] = -gbl_channel_prefix_matrix[(dst_rdma_rank * NUM_MAX_NVL_PEERS + lane_id - NUM_MAX_NVL_PEERS) * num_channels + channel_id] - 1;
+                rdma_channel_meta.send_buffer(dst_rdma_rank)[lane_id] =
+                    -gbl_channel_prefix_matrix[(dst_rdma_rank * NUM_MAX_NVL_PEERS + lane_id -
+                                                NUM_MAX_NVL_PEERS) *
+                                                   num_channels +
+                                               channel_id] -
+                    1;
             } else if (lane_id == NUM_MAX_NVL_PEERS * 2) {
-                dst_ptr[lane_id] = -(channel_id == 0 ? 0 : rdma_channel_prefix_matrix[dst_rdma_rank * num_channels + channel_id - 1]) - 1;
+                rdma_channel_meta.send_buffer(dst_rdma_rank)[lane_id] =
+                    -(channel_id == 0 ? 0
+                                      : rdma_channel_prefix_matrix[dst_rdma_rank * num_channels +
+                                                                   channel_id - 1]) -
+                    1;
             } else if (lane_id == NUM_MAX_NVL_PEERS * 2 + 1) {
-                dst_ptr[lane_id] = -rdma_channel_prefix_matrix[dst_rdma_rank * num_channels + channel_id] - 1;
+                rdma_channel_meta.send_buffer(dst_rdma_rank)[lane_id] =
+                    -rdma_channel_prefix_matrix[dst_rdma_rank * num_channels + channel_id] - 1;
             }
 
-            syncwarp();
-            if (dst_rdma_rank != rdma_rank) {
-#if !defined(FORCE_DUSHMEM_API) && !defined(ROCM_DISABLE_CTX)
-                shmem_ctx_int_put_nbi_warp(ctx, 
-#else
-                shmemx_int_put_nbi_warp(
-#endif
-                rdma_channel_meta.recv_buffer(rdma_rank),
+            rocshmem::rocshmem_ctx_int_put_nbi_wave(
+                ctx, rdma_channel_meta.recv_buffer(rdma_rank),
                 rdma_channel_meta.send_buffer(dst_rdma_rank), NUM_MAX_NVL_PEERS * 2 + 2,
                 translate_dst_rdma_rank<kLowLatencyMode>(dst_rdma_rank, nvl_rank));
         }
-        }
 
-#if !defined(FORCE_DUSHMEM_API) && !defined(ROCM_DISABLE_CTX)
-        shmem_ctx_quiet(ctx);                
-#else
-        shmem_fence();
-#endif
+        rocshmem::rocshmem_ctx_quiet(ctx);
+        sync_rdma_sender_smem();
 
-        // sync_rdma_sender_smem();
-        __syncthreads();
-
-        // 遍历令牌并复制到缓冲区
+        // Iterate over tokens and copy into buffer
         int64_t token_idx;
         int     cached_rdma_channel_head = 0, last_rdma_tail_idx = -1;
-        auto send_buffer = lane_id == rdma_rank ? rdma_channel_data.recv_buffer(lane_id) : rdma_channel_data.send_buffer(lane_id);
-        for(token_idx = token_start_idx + warp_id; token_idx < token_end_idx; token_idx += kNumDispatchRDMASenderWarps) {
-            // 读取RDMA秩的存在性
+        auto    send_buffer = lane_id == rdma_rank ? rdma_channel_data.recv_buffer(lane_id)
+                                                   : rdma_channel_data.send_buffer(lane_id);
+        for (token_idx = token_start_idx + warp_id; token_idx < token_end_idx;
+             token_idx += kNumDispatchRDMASenderWarps) {
+            // Read RDMA rank existence
             uint64_t is_token_in_rank_uint64 = 0;
-            if(lane_id < kNumRDMARanks) {
-                is_token_in_rank_uint64 = *reinterpret_cast<const uint64_t*>(is_token_in_rank + token_idx * num_ranks + lane_id * NUM_MAX_NVL_PEERS);
-            }
+            if (lane_id < kNumRDMARanks)
+                is_token_in_rank_uint64 = *reinterpret_cast<const uint64_t *>(
+                    is_token_in_rank + token_idx * num_ranks + lane_id * NUM_MAX_NVL_PEERS);
 
-            // 获得处理数据的自旋锁，获得锁后才会处理一些数据信息
-            while(lane_id == 0 && rdma_send_next_token_idx != token_idx) {
-                // 等待
-            }
+            // Acquire sequential lock
+            while (lane_id == 0 and rdma_send_next_token_idx != token_idx)
+                ;
             syncwarp();
 
-            // 获取下一个尾部位置
+            // Acquire next tail
             int rdma_tail_idx = -1;
-            if(is_token_in_rank_uint64 != 0) {
+            if (is_token_in_rank_uint64 != 0) {
                 rdma_tail_idx = rdma_send_channel_next_tail[lane_id]++;
-
-                // 与kForwarderCoordinator相互配合，调节发送数据的频率
-                while(rdma_tail_idx - cached_rdma_channel_head >= num_max_rdma_chunked_recv_tokens) {
-                    cached_rdma_channel_head = static_cast<int>(ld_volatile_global(rdma_channel_head.buffer(lane_id)));
-                }
+                while (rdma_tail_idx - cached_rdma_channel_head >= num_max_rdma_chunked_recv_tokens)
+                    cached_rdma_channel_head =
+                        static_cast<int>(ld_volatile_global(rdma_channel_head.buffer(lane_id)));
             }
             syncwarp();
 
-            // 存储RDMA头部以供合并
-            if(lane_id < kNumRDMARanks && !kCachedMode) {
+            // Store RDMA head for combine
+            if (lane_id < kNumRDMARanks and not kCachedMode)
                 send_rdma_head[token_idx * kNumRDMARanks + lane_id] = rdma_tail_idx;
-            }
 
-            // 更新最后一个令牌尾部
-            if(last_rdma_tail_idx >= 0) {
-                st_release_cta(const_cast<int*>(rdma_send_channel_tail + lane_id), last_rdma_tail_idx + 1);
-            }
+            // Update last token tail
+            if (last_rdma_tail_idx >= 0)
+                st_release_cta(const_cast<const int *>(rdma_send_channel_tail + lane_id),
+                               last_rdma_tail_idx + 1);
             last_rdma_tail_idx = rdma_tail_idx;
 
-            // 释放顺序锁
-            if(lane_id == 0) {
-                rdma_send_next_token_idx += 1;
-            }
+            // Release sequential lock
+            lane_id == 0 ? (rdma_send_next_token_idx += 1) : 0;
 
-            // 广播尾部位置
+            // Broadcast tails
             SourceMeta src_meta;
             int        num_topk_ranks = 0, topk_ranks[kNumTopkRDMARanks];
-            void* dst_send_buffers[kNumTopkRDMARanks];
-            /*
-            该for循环主要功能是在一个CUDA内核中协调多个线程之间的RDMA发送操作
-            */
-            #pragma unroll
-            for(int i = 0, slot_idx; i < kNumRDMARanks; ++i) {
-                // 使用__shfl_sync函数在warp内同步并广播rdma_tail_idx的值
-                if((slot_idx = shfl_sync(rdma_tail_idx, i)) >= 0) {
-                    // warp 所有线程参与，rdma_tail_idx默认为-1， 只有对应rdma rank需要发送时， rdma_tail_idx才会>=0
-                    //  计算slot_idx在接收缓冲区中的位置
+            void      *dst_send_buffers[kNumTopkRDMARanks];
+#pragma unroll
+            for (int i = 0, slot_idx; i < kNumRDMARanks; ++i)
+                if ((slot_idx = shfl_sync(rdma_tail_idx, i)) >= 0) {
                     slot_idx                          = slot_idx % num_max_rdma_chunked_recv_tokens;
-
-                    // 存储当前RDMA秩到topk_ranks数组中
                     topk_ranks[num_topk_ranks]        = i;
-
-                    // 广播is_token_in_rank_uint64的值到所有线程，并解释为布尔数组
                     auto recv_is_token_in_rank_uint64 = broadcast(is_token_in_rank_uint64, i);
-                    auto recv_is_token_in_rank_values = reinterpret_cast<const bool*>(&recv_is_token_in_rank_uint64);
-
-                    // 如果当前lane_id等于num_topk_ranks，则更新src_meta
-                    if(lane_id == num_topk_ranks) {
+                    auto recv_is_token_in_rank_values =
+                        reinterpret_cast<const bool *>(&recv_is_token_in_rank_uint64);
+                    if (lane_id == num_topk_ranks)
                         src_meta = SourceMeta(rdma_rank, recv_is_token_in_rank_values);
-                    }
-
-                    // 计算目标发送缓冲区的地址，并存储在dst_send_buffers数组中
-                    // 获取到发送地址， num_topk_ranks-1 是需要发送的ranks数
-                    dst_send_buffers[num_topk_ranks++] = reinterpret_cast<uint8_t*>(broadcast(send_buffer, i)) + slot_idx * num_bytes_per_rdma_token;
-                }
+                    dst_send_buffers[num_topk_ranks++] =
+                        reinterpret_cast<uint8_t *>(broadcast(send_buffer, i)) +
+                        slot_idx * num_bytes_per_rdma_token;
                 }
             EP_DEVICE_ASSERT(num_topk_ranks <= kNumTopkRDMARanks);
 
-            //////////////// 复制数据到发送缓冲区 ////////////////
-            // 复制源元数据到对称发送缓冲区
-            if(lane_id < num_topk_ranks) {
-                st_na_global(reinterpret_cast<SourceMeta*>(dst_send_buffers[lane_id]), src_meta);
-            }
+            // Copy `x` into symmetric send buffer
+            auto st_broadcast = [=](const int key, const int4 &value) {
+                for (int j = 0; j < num_topk_ranks; ++j)
+                    st_na_global(reinterpret_cast<int4 *>(dst_send_buffers[j]) + key, value);
+            };
+            UNROLLED_WARP_COPY(5, lane_id, hidden_int4, 0, x + token_idx * hidden_int4,
+                               ld_nc_global, st_broadcast);
+            for (int i = 0; i < num_topk_ranks; ++i)
+                dst_send_buffers[i] = reinterpret_cast<int4 *>(dst_send_buffers[i]) + hidden_int4;
 
-            for(int i = 0; i < num_topk_ranks; ++i) {
-                dst_send_buffers[i] = reinterpret_cast<SourceMeta*>(dst_send_buffers[i]) + 1;
-            }
+            // Copy source metadata into symmetric send buffer
+            if (lane_id < num_topk_ranks)
+                st_na_global(reinterpret_cast<SourceMeta *>(dst_send_buffers[lane_id]), src_meta);
 
-            // 复制 `x` 到对称发送缓冲区
-            auto st_broadcast = [=](const int key, const int4& value) {
-                for(int j = 0; j < num_topk_ranks; ++j) {
-                    st_na_global(reinterpret_cast<int4*>(dst_send_buffers[j]) + key, value);
-                }
-            };
-            UNROLLED_WARP_COPY(5, lane_id, hidden_int4, 0, x + token_idx * hidden_int4, ld_nc_global, st_broadcast);
-            for(int i = 0; i < num_topk_ranks; ++i) {
-                dst_send_buffers[i] = reinterpret_cast<int4*>(dst_send_buffers[i]) + hidden_int4;
-            }
+            for (int i = 0; i < num_topk_ranks; ++i)
+                dst_send_buffers[i] = reinterpret_cast<SourceMeta *>(dst_send_buffers[i]) + 1;
 
-            // 复制 `x_scales` 到对称发送缓冲区
-            for(int i = lane_id; i < num_scales; i += kWarpSize) {
+            // Copy `x_scales` into symmetric send buffer
+            for (int i = lane_id; i < num_scales; i += kWarpSize) {
                 auto value = ld_nc_global(x_scales + token_idx * num_scales + i);
-                for(int j = 0; j < num_topk_ranks; ++j) {
-                    st_na_global(reinterpret_cast<float*>(dst_send_buffers[j]) + i, value);
-                }
-            }
-
-            for(int i = 0; i < num_topk_ranks; ++i) {
-                dst_send_buffers[i] = reinterpret_cast<float*>(dst_send_buffers[i]) + num_scales;
+                for (int j = 0; j < num_topk_ranks; ++j)
+                    st_na_global(reinterpret_cast<float *>(dst_send_buffers[j]) + i, value);
             }
+            for (int i = 0; i < num_topk_ranks; ++i)
+                dst_send_buffers[i] = reinterpret_cast<float *>(dst_send_buffers[i]) + num_scales;
 
-            // 复制 `topk_idx` 和 `topk_weights` 到对称发送缓冲区
-            for(int i = lane_id; i < num_topk * num_topk_ranks; i += kWarpSize) {
+            // Copy `topk_idx` and `topk_weights` into symmetric send buffer
+            for (int i = lane_id; i < num_topk * num_topk_ranks; i += kWarpSize) {
                 auto rank_idx = i / num_topk, copy_idx = i % num_topk;
-                auto idx_value = static_cast<int>(ld_nc_global(topk_idx + token_idx * num_topk + copy_idx));
+                auto idx_value =
+                    static_cast<int>(ld_nc_global(topk_idx + token_idx * num_topk + copy_idx));
                 auto weight_value = ld_nc_global(topk_weights + token_idx * num_topk + copy_idx);
-                st_na_global(reinterpret_cast<int*>(dst_send_buffers[rank_idx]) + copy_idx, idx_value);
-                st_na_global(reinterpret_cast<float*>(dst_send_buffers[rank_idx]) + num_topk + copy_idx, weight_value);
+                st_na_global(reinterpret_cast<int *>(dst_send_buffers[rank_idx]) + copy_idx,
+                             idx_value);
+                st_na_global(reinterpret_cast<float *>(dst_send_buffers[rank_idx]) + num_topk +
+                                 copy_idx,
+                             weight_value);
             }
         }
 
-        // 结尾部分
-        // 获取顺序锁
-        while(lane_id == 0 && rdma_send_next_token_idx != token_idx) {
-            // 等待
-        }
-
+        // Epilogue
+        // Acquire sequential lock
+        while (lane_id == 0 and rdma_send_next_token_idx != token_idx)
+            ;
         syncwarp();
 
-        // 更新最后一个令牌尾部
-        if(last_rdma_tail_idx >= 0) {
-            st_release_cta(const_cast<int*>(rdma_send_channel_tail + lane_id), last_rdma_tail_idx + 1);
-        }
+        // Update last token tail
+        if (last_rdma_tail_idx >= 0)
+            st_release_cta(const_cast<const int *>(rdma_send_channel_tail + lane_id),
+                           last_rdma_tail_idx + 1);
 
-        // 释放顺序锁
-        if(lane_id == 0) {
-            rdma_send_next_token_idx += 1;
-        }
-    } else if(warp_role == WarpRole::kRDMASenderCoordinator) {
-        /*
-        这段代码的主要功能是在一个CUDA内核中协调多个线程之间的RDMA发送操作。
-        它首先计算每个RDMA秩需要发送的令牌数，然后在所有RDMA秩之间循环，检查是否有令牌需要发送。
-        如果有，它将计算本次需要发出的令牌数，并发出相应的RDMA发送请求。
-        最后，它更新相关的尾部位置，以便下次循环时可以正确地计算需要发送的令牌数。
+        // Release sequential lock
+        lane_id == 0 ? (rdma_send_next_token_idx += 1) : 0;
+    } else if (warp_role == WarpRole::kRDMASenderCoordinator) {
+        // NOTES: in case of splitting the issued put at the end of the buffer
+        EP_DEVICE_ASSERT(
+            num_max_rdma_chunked_recv_tokens % num_max_rdma_chunked_send_tokens == 0);
 
-        kRDMASenderCoordinator使用了同sm内存一致性（ld.acquire.cta.s32），
-        dushmem内存一致性（dushmem_fence）和原子操作（dushmemx_signal_op），减少硬同步，提升整体效率。
-        */
-        if(warp_id > kNumDispatchRDMASenderWarps) {
-            return;
-        }
-        // 确保最大接收令牌数可以被最大发送令牌数整除，以避免缓冲区分割问题
-        EP_DEVICE_ASSERT(num_max_rdma_chunked_recv_tokens % num_max_rdma_chunked_send_tokens == 0);
+        // Synchronize shared memory
+        sync_rdma_sender_smem();
 
-        // 同步共享内存，确保所有线程在继续之前都达到了这一点
-        // sync_rdma_sender_smem();
-        __syncthreads();
-
-        // 计算当前通道需要发送的令牌数
+        // Get number of tokens to send for each RDMA rank
         int num_tokens_to_send = 0;
-        if(lane_id < kNumRDMARanks) {
+        if (lane_id < kNumRDMARanks) {
             num_tokens_to_send = rdma_channel_prefix_matrix[lane_id * num_channels + channel_id];
-            if(channel_id > 0)
-                num_tokens_to_send -= rdma_channel_prefix_matrix[lane_id * num_channels + channel_id - 1];
+            if (channel_id > 0)
+                num_tokens_to_send -=
+                    rdma_channel_prefix_matrix[lane_id * num_channels + channel_id - 1];
         }
 
-        // 记录上次发出的尾部位置
+        // Iterate all RDMA ranks
         int last_issued_tail = 0;
-        // 当有任何RDMA秩需要发送令牌时，继续循环
-        while(__any_sync(kFullWarpMask, num_tokens_to_send > 0)) {
-            for(int i = 0, synced_num_tokens_to_send; i < kNumRDMARanks; ++i) {
-                // 计算目标RDMA秩
+        while (__any_sync(kFullWarpMask, num_tokens_to_send > 0)) {
+            for (int i = 0, synced_num_tokens_to_send; i < kNumRDMARanks; ++i) {
                 int dst_rdma_rank         = (i + channel_id) % kNumRDMARanks;
-
-                // 获取同步后的需要发送的令牌数
                 synced_num_tokens_to_send = shfl_sync(num_tokens_to_send, dst_rdma_rank);
+                if (synced_num_tokens_to_send == 0)
+                    continue;
 
-                if(synced_num_tokens_to_send == 0)
-                    continue; // 如果没有令牌需要发送，则跳过
-
-                // 读取进度
+                // Read progress
                 auto synced_last_issued_tail = shfl_sync(last_issued_tail, dst_rdma_rank);
-                auto processed_tail          = ld_acquire_cta(const_cast<const int*>(rdma_send_channel_tail + dst_rdma_rank));
+                auto processed_tail =
+                    ld_acquire_cta(const_cast<const int *>(rdma_send_channel_tail + dst_rdma_rank));
                 auto num_tokens_processed = processed_tail - synced_last_issued_tail;
-
-                // 如果处理的令牌数不等于需要发送的令牌数，并且处理的令牌数小于最大发送令牌数，则跳过
-                if(num_tokens_processed != synced_num_tokens_to_send && num_tokens_processed < num_max_rdma_chunked_send_tokens)
+                if (num_tokens_processed != synced_num_tokens_to_send and
+                    num_tokens_processed < num_max_rdma_chunked_send_tokens)
                     continue;
 
-                // 计算本次需要发出的令牌数
-                auto num_tokens_to_issue = min(num_tokens_processed, num_max_rdma_chunked_send_tokens);
-                EP_DEVICE_ASSERT(num_tokens_to_issue >= 0 && num_tokens_to_issue <= synced_num_tokens_to_send);
-
-                // 发出RDMA发送请求
-                if(dst_rdma_rank != rdma_rank) {
+                // Issue RDMA send
+                auto num_tokens_to_issue =
+                    min(num_tokens_processed, num_max_rdma_chunked_send_tokens);
+                EP_DEVICE_ASSERT(num_tokens_to_issue >= 0 and
+                                          num_tokens_to_issue <= synced_num_tokens_to_send);
+                if (dst_rdma_rank != rdma_rank) {
                     auto dst_slot_idx = synced_last_issued_tail % num_max_rdma_chunked_recv_tokens;
-                    EP_DEVICE_ASSERT(dst_slot_idx + num_tokens_to_issue <= num_max_rdma_chunked_recv_tokens);
-#if !defined(FORCE_DUSHMEM_API) && !defined(ROCM_DISABLE_CTX)
-                    shmem_ctx_schar_put_nbi_warp(ctx,
-#else
-                    shmemx_int8_put_nbi_warp(
-#endif
+                    EP_DEVICE_ASSERT(dst_slot_idx + num_tokens_to_issue <=
+                                              num_max_rdma_chunked_recv_tokens);
+                    rocshmem::rocshmem_ctx_schar_put_nbi_wave(
+                        ctx,
                         rdma_channel_data.recv_buffer(rdma_rank) +
                             dst_slot_idx * num_bytes_per_rdma_token,
                         rdma_channel_data.send_buffer(dst_rdma_rank) +
                             dst_slot_idx * num_bytes_per_rdma_token,
                         num_bytes_per_rdma_token * num_tokens_to_issue,
                         translate_dst_rdma_rank<kLowLatencyMode>(dst_rdma_rank, nvl_rank));
-#if !defined(FORCE_DUSHMEM_API) && !defined(ROCM_DISABLE_CTX)
-                    shmem_ctx_quiet(ctx);                
-#else
-                    shmem_fence();
-#endif
+                    rocshmem::rocshmem_ctx_quiet(ctx);
                 } else {
-                    // 对于本地RDMA秩，使用较轻的内存屏障
+                    // Lighter fence for local RDMA rank
                     memory_fence();
                 }
 
-                // 更新尾部位置
+                // Update tails
                 syncwarp();
-                if(lane_id == dst_rdma_rank) {
+                if (lane_id == dst_rdma_rank) {
                     last_issued_tail += num_tokens_to_issue;
                     num_tokens_to_send -= num_tokens_to_issue;
-                    // 更新远端rdma 己方已发送的token数，用于做发送信息同步。用于与kRDMAAndNVLForwarder互相通信
-#if !defined(FORCE_DUSHMEM_API) && !defined(ROCM_DISABLE_CTX)
-                    shmem_ctx_ulong_atomic_add(ctx,
-#else
-                    shmem_signal_op_add(
-#endif
-                        rdma_channel_tail.buffer(rdma_rank), num_tokens_to_issue,
+
+                    rocshmem::rocshmem_ctx_ulong_atomic_add(
+                        ctx, rdma_channel_tail.buffer(rdma_rank), num_tokens_to_issue,
                         translate_dst_rdma_rank<kLowLatencyMode>(dst_rdma_rank, nvl_rank));
                 }
             }
-        } // while(__any(num_tokens_to_send > 0))
-    } else if(warp_role == WarpRole::kRDMAAndNVLForwarder) {
-        /*
-        这段代码的主要功能是在一个CUDA内核中协调从RDMA消费者到NVL生产者的转发操作。
-        它首先计算目标NVL秩和目标秩，然后等待相关的计数器到达。
-        接着，它检查目标队列是否为空，或者等待一个缓冲区被释放。
-        然后，它找到下一个源RDMA秩，并遍历RDMA缓冲区中的每一个令牌，复制相关的数据到NVL缓冲区。
-        最后，它同步头部和尾部索引，并标记通道为退役状态。
-        */
-        // RDMA消费者和NVL生产者
-        const auto dst_nvl_rank          = target_rank;                                       // 目标NVL秩
-        const auto dst_rank              = rdma_rank * NUM_MAX_NVL_PEERS + dst_nvl_rank;      // 目标秩
-        const auto dst_rank_expert_begin = dst_rank * (num_experts / num_ranks);              // 目标秩专家开始
-        const auto dst_rank_expert_end   = dst_rank_expert_begin + (num_experts / num_ranks); // 目标秩专家结束
-
-        // 等待计数器到达
+        }
+    } else if (warp_role == WarpRole::kRDMAAndNVLForwarder) {
+        // RDMA consumers and NVL producers
+        const auto dst_nvl_rank          = target_rank;
+        const auto dst_rank              = rdma_rank * NUM_MAX_NVL_PEERS + dst_nvl_rank;
+        const auto dst_rank_expert_begin = dst_rank * (num_experts / num_ranks);
+        const auto dst_rank_expert_end   = dst_rank_expert_begin + (num_experts / num_ranks);
+
+        // Wait counters to arrive
         int num_tokens_to_recv_from_rdma = 0, src_rdma_channel_prefix = 0;
         EP_DEVICE_ASSERT(kNumRDMARanks <= kWarpSize);
         auto start_time = wall_clock64();
-        if(lane_id < kNumRDMARanks) {
-            while(true) {
-                // 对应于kRDMASender中的数据写入
-                auto meta_0 = ld_volatile_global(rdma_channel_meta.recv_buffer(lane_id) + dst_nvl_rank);                     // 是nvl节点的起始地址
-                auto meta_1 = ld_volatile_global(rdma_channel_meta.recv_buffer(lane_id) + NUM_MAX_NVL_PEERS + dst_nvl_rank); // nvl节点的结束地址
-                auto meta_2 = ld_volatile_global(rdma_channel_meta.recv_buffer(lane_id) + NUM_MAX_NVL_PEERS * 2);            // 本rdma节点的起始地址
-                auto meta_3 = ld_volatile_global(rdma_channel_meta.recv_buffer(lane_id) + NUM_MAX_NVL_PEERS * 2 + 1);        // 本节点的结束地址
-                if(meta_0 < 0 && meta_1 < 0 && meta_2 < 0 && meta_3 < 0) {
-                    // 通知NVL秩
+        if (lane_id < kNumRDMARanks) {
+            while (true) {
+                auto meta_0 =
+                    ld_volatile_global(rdma_channel_meta.recv_buffer(lane_id) + dst_nvl_rank);
+                auto meta_1 = ld_volatile_global(rdma_channel_meta.recv_buffer(lane_id) +
+                                                 NUM_MAX_NVL_PEERS + dst_nvl_rank);
+                auto meta_2 = ld_volatile_global(rdma_channel_meta.recv_buffer(lane_id) +
+                                                 NUM_MAX_NVL_PEERS * 2);
+                auto meta_3 = ld_volatile_global(rdma_channel_meta.recv_buffer(lane_id) +
+                                                 NUM_MAX_NVL_PEERS * 2 + 1);
+                if (meta_0 < 0 and meta_1 < 0 and meta_2 < 0 and meta_3 < 0) {
+                    // Notify NVL ranks
                     int start_sum = -meta_0 - 1, end_sum = -meta_1 - 1;
-                    EP_DEVICE_ASSERT(start_sum >= 0 && end_sum >= 0 && end_sum >= start_sum);
-
-                    st_relaxed_sys_global(nvl_channel_prefix_start.buffer() + lane_id, -start_sum - 1);
+                    EP_DEVICE_ASSERT(start_sum >= 0 and end_sum >= 0 and
+                                              end_sum >= start_sum);
+                    st_relaxed_sys_global(nvl_channel_prefix_start.buffer() + lane_id,
+                                          -start_sum - 1);
                     st_relaxed_sys_global(nvl_channel_prefix_end.buffer() + lane_id, -end_sum - 1);
 
-                    // 保存从RDMA通道接收的令牌计数
+                    // Save RDMA channel received token count
                     src_rdma_channel_prefix        = -meta_2 - 1;
                     auto src_rdma_channel_prefix_1 = -meta_3 - 1;
-                    num_tokens_to_recv_from_rdma = src_rdma_channel_prefix_1 - src_rdma_channel_prefix; // 是远端 rdma_rank 会发送给当前节点的token数量
-                    if(!kCachedMode)
-                        recv_rdma_channel_prefix_matrix[lane_id * num_channels + channel_id] = src_rdma_channel_prefix_1;
-
-                    src_rdma_channel_prefix += lane_id == 0 ? 0 : recv_rdma_rank_prefix_sum[lane_id - 1]; // 对应的远端 rdma_rank 的起始index, 存在线程0之中
+                    num_tokens_to_recv_from_rdma =
+                        src_rdma_channel_prefix_1 - src_rdma_channel_prefix;
+                    if (not kCachedMode)
+                        recv_rdma_channel_prefix_matrix[lane_id * num_channels + channel_id] =
+                            src_rdma_channel_prefix_1;
+                    src_rdma_channel_prefix +=
+                        lane_id == 0 ? 0 : recv_rdma_rank_prefix_sum[lane_id - 1];
                     EP_DEVICE_ASSERT(num_tokens_to_recv_from_rdma >= 0);
                     break;
                 }
 
-                // 超时检查
-                if (wall_clock64() - start_time > NUM_TIMEOUT_CYCLES) {
-                    printf("DeepEP dispatch forwarder timeout (RDMA meta), channel: %d, RDMA: %d, nvl: %d, src RDMA lane: %d, dst NVL: %d, meta: %d, %d, %d, %d\n",
-                           channel_id, rdma_rank, nvl_rank, lane_id, dst_nvl_rank, meta_0, meta_1, meta_2, meta_3);
+                // Timeout check
+                long long int elapsed_time =
+                    wall_clock64() > start_time ? wall_clock64() - start_time : 0;
+                if (elapsed_time > NUM_TIMEOUT_CYCLES) {
+                    printf("DeepEP dispatch forwarder timeout (RDMA meta), channel: %d, RDMA: %d, "
+                           "nvl: %d, src RDMA lane: %d, dst NVL: %d, meta: %d, %d, %d, %d\n",
+                           channel_id, rdma_rank, nvl_rank, lane_id, dst_nvl_rank, meta_0, meta_1,
+                           meta_2, meta_3);
                     trap();
                 }
             }
         }
         syncwarp();
-
-        // 移动缓存的头部
+        // Shift cached head
         send_nvl_head += src_rdma_channel_prefix * NUM_MAX_NVL_PEERS + dst_nvl_rank;
 
-        // 等待共享内存被清理
-        // sync_forwarder_smem();
-        __syncthreads();
+        // Wait shared memory to be cleaned
+        sync_forwarder_smem();
 
-        // 开始准备处理接受数据，直到所有的数据接受完成。
-        // 转发从RDMA缓冲区的令牌
-        // 注意：总是从本地秩开始
+        // Forward tokens from RDMA buffer
+        // NOTES: always start from the local rank
         int src_rdma_rank            = sm_id % kNumRDMARanks;
         int cached_rdma_channel_head = 0, cached_rdma_channel_tail = 0;
         int cached_nvl_channel_head = 0, cached_nvl_channel_tail = 0, rdma_nvl_token_idx = 0;
-        while(__any_sync(kFullWarpMask, num_tokens_to_recv_from_rdma > 0)) {
-            // 检查nvl目标队列是否为空，或者等待一个缓冲区被释放
+        while (__any_sync(kFullWarpMask, num_tokens_to_recv_from_rdma > 0)) {
+            // Check destination queue emptiness, or wait a buffer to be released
             start_time = wall_clock64();
-
-            // 用于给kNVLReceivers进行互动，控制数据的传输速度
-            while(lane_id == 0) {
+            while (lane_id == 0) {
                 int num_used_slots = cached_nvl_channel_tail - cached_nvl_channel_head;
-                if(num_max_nvl_chunked_recv_tokens - num_used_slots >= num_max_nvl_chunked_send_tokens)
+                if (num_max_nvl_chunked_recv_tokens - num_used_slots >=
+                    num_max_nvl_chunked_send_tokens)
                     break;
                 cached_nvl_channel_head = ld_volatile_global(nvl_channel_head.buffer());
 
-                // 超时检查
-                if (wall_clock64() - start_time > NUM_TIMEOUT_CYCLES) {
-                    printf("DeepEP dispatch forwarder timeout (NVL check), channel: %d, RDMA: %d, nvl: %d, dst NVL: %d, head: %d, tail: %d\n",
-                           channel_id, rdma_rank, nvl_rank, dst_nvl_rank, ld_volatile_global(nvl_channel_head.buffer()), cached_nvl_channel_tail);
+                // Timeout check
+                long long int elapsed_time =
+                    wall_clock64() > start_time ? wall_clock64() - start_time : 0;
+                if (elapsed_time > NUM_TIMEOUT_CYCLES) {
+                    printf("DeepEP dispatch forwarder timeout (NVL check), channel: %d, RDMA: %d, "
+                           "nvl: %d, dst NVL: %d, head: %d, tail: %d\n",
+                           channel_id, rdma_rank, nvl_rank, dst_nvl_rank,
+                           ld_volatile_global(nvl_channel_head.buffer()), cached_nvl_channel_tail);
                     trap();
                 }
             }
             syncwarp();
 
-            // 找到下一个源RDMA秩（轮询）
+            // Find next source RDMA rank (round-robin)
             start_time = wall_clock64();
-            while(true) {
+            while (true) {
                 src_rdma_rank = (src_rdma_rank + 1) % kNumRDMARanks;
-                if(shfl_sync(num_tokens_to_recv_from_rdma, src_rdma_rank) > 0) {
-                    if(lane_id == src_rdma_rank && cached_rdma_channel_head == cached_rdma_channel_tail)
-                        cached_rdma_channel_tail = static_cast<int>(ld_acquire_sys_global(rdma_channel_tail.buffer(src_rdma_rank)));
-
-                    if(shfl_sync(cached_rdma_channel_tail > cached_rdma_channel_head, src_rdma_rank)) {
+                if (shfl_sync(num_tokens_to_recv_from_rdma, src_rdma_rank) > 0) {
+                    if (lane_id == src_rdma_rank and
+                        cached_rdma_channel_head == cached_rdma_channel_tail)
+                        cached_rdma_channel_tail = static_cast<int>(
+                            ld_relaxed_sys_global(rdma_channel_tail.buffer(src_rdma_rank)));
+                    if (shfl_sync(cached_rdma_channel_tail > cached_rdma_channel_head,
+                                  src_rdma_rank))
                         break;
                 }
-                }
 
-                // 超时检查
-                if (wall_clock64() - start_time > NUM_TIMEOUT_CYCLES and lane_id < kNumRDMARanks) {
-                    printf("DeepEP dispatch forwarder timeout (RDMA check), channel: %d, RDMA: %d, nvl: %d, dst NVL: %d, src RDMA lane: %d, head: %d, tail: %d, expected: %d\n",
-                           channel_id, rdma_rank, nvl_rank, dst_nvl_rank, lane_id, cached_rdma_channel_head, cached_rdma_channel_tail, num_tokens_to_recv_from_rdma);
+                // Timeout check
+                long long int elapsed_time =
+                    wall_clock64() > start_time ? wall_clock64() - start_time : 0;
+                if (elapsed_time > NUM_TIMEOUT_CYCLES and lane_id < kNumRDMARanks) {
+                    printf("DeepEP dispatch forwarder timeout (RDMA check), channel: %d, RDMA: %d, "
+                           "nvl: %d, dst NVL: %d, src RDMA lane: %d, head: %d, tail: %d, expected: "
+                           "%d\n",
+                           channel_id, rdma_rank, nvl_rank, dst_nvl_rank, lane_id,
+                           cached_rdma_channel_head, cached_rdma_channel_tail,
+                           num_tokens_to_recv_from_rdma);
                     trap();
                 }
             }
-
             auto src_rdma_head = shfl_sync(cached_rdma_channel_head, src_rdma_rank);
             auto src_rdma_tail = shfl_sync(cached_rdma_channel_tail, src_rdma_rank);
 
-            // 遍历RDMA缓冲区中的每一个令牌
-            for(int i = src_rdma_head, num_tokens_sent = 0; i < src_rdma_tail; ++i) {
+            // Iterate over every token from the RDMA buffer
+            for (int i = src_rdma_head, num_tokens_sent = 0; i < src_rdma_tail; ++i) {
                 auto  rdma_slot_idx = i % num_max_rdma_chunked_recv_tokens;
-                // 首先读取SourceMeta，对应到kRDMASenderCoordinator中 kRDMASender 的数据远程写入
-                void* shifted           = rdma_channel_data.recv_buffer(src_rdma_rank) + rdma_slot_idx * num_bytes_per_rdma_token;
-                auto src_meta           = ld_nc_global(reinterpret_cast<SourceMeta*>(reinterpret_cast<int8_t*>(shifted)));
-                if(lane_id == src_rdma_rank) {
-                    num_tokens_to_recv_from_rdma -= 1;
-                }
-
+                void *shifted       = rdma_channel_data.recv_buffer(src_rdma_rank) +
+                                rdma_slot_idx * num_bytes_per_rdma_token;
+                auto src_meta = ld_nc_global(reinterpret_cast<SourceMeta *>(
+                    reinterpret_cast<int8_t *>(shifted) + hidden_bytes));
+                lane_id == src_rdma_rank ? (num_tokens_to_recv_from_rdma -= 1) : 0;
                 bool is_in_dst_nvl_rank = src_meta.is_token_in_nvl_rank(dst_nvl_rank);
-                if(lane_id == src_rdma_rank) {
+                if (lane_id == src_rdma_rank) {
                     auto cached_head = is_in_dst_nvl_rank ? rdma_nvl_token_idx : -1;
                     rdma_nvl_token_idx += is_in_dst_nvl_rank;
-                    if(!kCachedMode)
+                    if (not kCachedMode)
                         send_nvl_head[i * NUM_MAX_NVL_PEERS] = cached_head;
                 }
-
-                if(!is_in_dst_nvl_rank)
+                if (not is_in_dst_nvl_rank)
                     continue;
 
-                // 获取一个空闲槽位
+                // Get an empty slot
                 int dst_slot_idx = (cached_nvl_channel_tail++) % num_max_nvl_chunked_recv_tokens;
 
-                // 设置 src和dst 位置
-                auto src_gpu_buffer_x = reinterpret_cast<int4*>(reinterpret_cast<int8_t*>(shifted) + sizeof(SourceMeta));
-                auto src_gpu_buffer_scales = reinterpret_cast<float*>(reinterpret_cast<int8_t*>(src_gpu_buffer_x) + hidden_bytes);
-                auto src_gpu_buffer_topk_idx = reinterpret_cast<int*>(reinterpret_cast<int8_t*>(src_gpu_buffer_scales) + num_scales * sizeof(float));
-                auto src_gpu_buffer_topk_weights = reinterpret_cast<float*>(reinterpret_cast<int8_t*>(src_gpu_buffer_topk_idx) + num_topk * sizeof(int));
+                // Copy data
+                UNROLLED_WARP_COPY(5, lane_id, hidden_int4,
+                                   nvl_channel_x.buffer() + dst_slot_idx * hidden_int4,
+                                   reinterpret_cast<int4 *>(shifted), ld_nc_global, st_na_global);
+                shifted = reinterpret_cast<int4 *>(shifted) + hidden_int4;
 
-                auto dst_gpu_buffer_x = nvl_channel_x.buffer() + dst_slot_idx * hidden_int4;
-                auto dst_gpu_buffer_scales = nvl_channel_x_scales.buffer() + dst_slot_idx * num_scales;
-                auto dst_gpu_buffer_topk_idx = nvl_channel_topk_idx.buffer() + dst_slot_idx * num_topk;
-                auto dst_gpu_buffer_topk_weights = nvl_channel_topk_weights.buffer() + dst_slot_idx * num_topk;
+                // Copy source meta
+                if (lane_id == 0)
+                    st_na_global(nvl_channel_src_meta.buffer() + dst_slot_idx, src_meta);
+                shifted = reinterpret_cast<SourceMeta *>(shifted) + 1;
 
-                if(lane_id == 0) {
-                    st_na_global(reinterpret_cast<int64_t*>(nvl_channel_src_meta.buffer() + dst_slot_idx), 
-                                 *reinterpret_cast<int64_t*>(&src_meta));
-                }
+                // Copy `x_scales`
+                UNROLLED_WARP_COPY(1, lane_id, num_scales,
+                                   nvl_channel_x_scales.buffer() + dst_slot_idx * num_scales,
+                                   reinterpret_cast<float *>(shifted), ld_nc_global, st_na_global);
+                shifted = reinterpret_cast<float *>(shifted) + num_scales;
 
-                UNROLLED_WARP_COPY(5, lane_id, hidden_int4,
-                    dst_gpu_buffer_x,
-                    src_gpu_buffer_x,
-                    ld_direct_global, st_na_global);
+                // Copy `topk_idx` and `topk_weights`
+                // NOTES: do not use `shifted` after this `if`, because only several lanes are
+                // shifted
+                if (lane_id < num_topk) {
+                    // Read
+                    auto idx_value    = ld_nc_global(reinterpret_cast<int *>(shifted) + lane_id);
+                    shifted           = reinterpret_cast<int *>(shifted) + num_topk;
+                    auto weight_value = ld_nc_global(reinterpret_cast<float *>(shifted) + lane_id);
+
+                    // Transform and write
+                    idx_value =
+                        (idx_value >= dst_rank_expert_begin and idx_value < dst_rank_expert_end)
+                            ? idx_value - dst_rank_expert_begin
+                            : -1;
+                    st_na_global(nvl_channel_topk_idx.buffer() + dst_slot_idx * num_topk + lane_id,
+                                 idx_value);
+                    weight_value = idx_value >= 0 ? weight_value : 0.0f;
+                    st_na_global(nvl_channel_topk_weights.buffer() + dst_slot_idx * num_topk +
+                                     lane_id,
+                                 weight_value);
+                }
 
-                UNROLLED_WARP_COPY(1, lane_id, num_scales,
-                    dst_gpu_buffer_scales,
-                    src_gpu_buffer_scales,
-                    ld_direct_global, st_na_global);
-
-                for(int t = lane_id; t < num_topk; t += kWarpSize) {
-                    int idx_val = ld_direct_global(reinterpret_cast<int*>(src_gpu_buffer_topk_idx) + t);
-                    float w_val = ld_direct_global(reinterpret_cast<float*>(src_gpu_buffer_topk_weights) + t);
-                    int new_idx = (idx_val >= dst_rank_expert_begin && idx_val < dst_rank_expert_end) 
-                                ? (idx_val - dst_rank_expert_begin) : -1;
-                    float new_w = (new_idx != -1) ? w_val : 0.0f;
-                    dst_gpu_buffer_topk_idx[t] = new_idx;
-                    dst_gpu_buffer_topk_weights[t] = new_w;
-                }
-
-                // 在NVL缓冲区不足的情况下，提前停止
-                if((++num_tokens_sent) == num_max_nvl_chunked_send_tokens)
+                // In case of insufficient NVL buffers, early stopping
+                if ((++num_tokens_sent) == num_max_nvl_chunked_send_tokens)
                     src_rdma_tail = i + 1;
             }
 
-            // 同步头部索引
-            if(lane_id == src_rdma_rank)
-                forward_channel_head[dst_nvl_rank][src_rdma_rank] = (cached_rdma_channel_head = src_rdma_tail);
+            // Sync head index
+            if (lane_id == src_rdma_rank)
+                forward_channel_head[dst_nvl_rank][src_rdma_rank] =
+                    (cached_rdma_channel_head = src_rdma_tail);
 
-            // 移动尾部索引，与kNVLReceivers互相通信使用
+            // Move tail index
             syncwarp();
-            if(lane_id == 0) {
-                st_release_sys_global(nvl_channel_tail.buffer(), cached_nvl_channel_tail);
-            }
+            if (lane_id == 0)
+                st_relaxed_sys_global(nvl_channel_tail.buffer(), cached_nvl_channel_tail);
         }
 
         // Retired
         syncwarp();
-        if(lane_id == 0) {
+        if (lane_id == 0)
             forward_channel_retired[dst_nvl_rank] = true;
-        }
-    } else if(warp_role == WarpRole::kForwarderCoordinator) {
-        /*
-        这段代码的主要功能是在一个CUDA内核中协调转发器的逻辑。
-        它首先检查当前warp是否是额外的转发器协调warp，如果是，则直接退出。
-        然后，它清理共享内存，并初始化转发通道的头部和退役状态。
-        接着，它进入一个无限循环，在循环中，它找到最小的头部，如果所有的通道都已退役，则退出循环。
-        否则，它更新远程头部，并进行纳秒级睡眠，以让其他warp工作。
-        */
+    } else if (warp_role == WarpRole::kForwarderCoordinator) {
         // Extra warps for forwarder coordinator should exit directly
-        if (warp_id > NUM_MAX_NVL_PEERS)
+        if (target_rank > 0)
             return;
 
-        // 转发warp协调器
-        EP_STATIC_ASSERT(kNumRDMARanks <= kWarpSize, "无效的RDMA对等体数量");
-        // 清理共享内存
-        EP_STATIC_ASSERT(NUM_MAX_NVL_PEERS <= kWarpSize, "无效的NVL对等体数量");
-#pragma unroll
-        for(int i = lane_id; i < kNumRDMARanks * NUM_MAX_NVL_PEERS; i += kWarpSize)
+        // Forward warp coordinator
+        EP_STATIC_ASSERT(kNumRDMARanks <= kWarpSize, "Invalid number of RDMA peers");
+
+        // Clean shared memory
+        EP_STATIC_ASSERT(NUM_MAX_NVL_PEERS <= kWarpSize, "Invalid number of NVL peers");
+        for (int i = lane_id; i < kNumRDMARanks * NUM_MAX_NVL_PEERS; i += kWarpSize)
             forward_channel_head[i % NUM_MAX_NVL_PEERS][i / NUM_MAX_NVL_PEERS] = 0;
-        if(lane_id < NUM_MAX_NVL_PEERS)
+        if (lane_id < NUM_MAX_NVL_PEERS)
             forward_channel_retired[lane_id] = false;
-        // sync_forwarder_smem();
-        __syncthreads();
+        sync_forwarder_smem();
 
         int last_head = 0, target_rdma = lane_id < kNumRDMARanks ? lane_id : 0;
-
-        while(true) {
-            // 找到最小的头部
+        while (true) {
+            // Find minimum head
             int min_head = std::numeric_limits<int>::max();
 #pragma unroll
-            for(int i = 0; i < NUM_MAX_NVL_PEERS; ++i)
-                if(!forward_channel_retired[i])
+            for (int i = 0; i < NUM_MAX_NVL_PEERS; ++i)
+                if (not forward_channel_retired[i])
                     min_head = min(min_head, forward_channel_head[i][target_rdma]);
-
-            if(__all_sync(kFullWarpMask, min_head == std::numeric_limits<int>::max())) {
+            if (__all_sync(kFullWarpMask, min_head == std::numeric_limits<int>::max()))
                 break;
-            }
 
-            // 更新远程头部
-            if(min_head != std::numeric_limits<int>::max() && min_head >= last_head + num_max_rdma_chunked_send_tokens && lane_id < kNumRDMARanks){
-#if !defined(FORCE_DUSHMEM_API) && !defined(ROCM_DISABLE_CTX)
-                shmem_ctx_ulong_atomic_add(ctx,
-#else
-                shmem_signal_op_add(
-#endif
-                    rdma_channel_head.buffer(rdma_rank), min_head - last_head,
+            // Update remote head
+            if (min_head != std::numeric_limits<int>::max() and
+                min_head >= last_head + num_max_rdma_chunked_send_tokens and
+                lane_id < kNumRDMARanks) {
+                rocshmem::rocshmem_ctx_ulong_atomic_add(
+                    ctx, rdma_channel_head.buffer(rdma_rank), min_head - last_head,
                     translate_dst_rdma_rank<kLowLatencyMode>(lane_id, nvl_rank));
                 last_head = min_head;
             }
 
-            // 纳秒级睡眠并让其他warp工作 // Nanosleep and let other warps work
+            // Nanosleep and let other warps work
             __builtin_amdgcn_s_sleep(NUM_WAIT_CYCLES_TIMES_64);
         }
-    } else if(warp_role == WarpRole::kNVLReceivers) {
-        if(warp_id >= NUM_MAX_NVL_PEERS) {
-            return;
-        }
-
-        // Place the main logic of your kernel here, using the parameters above.
-        // NVL消费者
-        // 从屏障结果中检索秩偏移（每个通道的寄存器存储一个RDMA秩）
+    } else {
+        // NVL consumers
+        // Retrieve rank offset from barrier results (each lane's register stores an RDMA rank)
         int src_nvl_rank = target_rank, total_offset = 0;
-        EP_STATIC_ASSERT(kNumRDMARanks <= kWarpSize, "无效的RDMA对等体数量");
-        if(lane_id < kNumRDMARanks && lane_id * NUM_MAX_NVL_PEERS + src_nvl_rank > 0)
+        EP_STATIC_ASSERT(kNumRDMARanks <= kWarpSize, "Invalid number of RDMA peers");
+        if (lane_id < kNumRDMARanks and lane_id * NUM_MAX_NVL_PEERS + src_nvl_rank > 0)
             total_offset = recv_gbl_rank_prefix_sum[lane_id * NUM_MAX_NVL_PEERS + src_nvl_rank - 1];
 
-        // 接收通道偏移
+        // Receive channel offsets
         int  start_offset = 0, end_offset = 0, num_tokens_to_recv;
         auto start_time = wall_clock64();
-
-        while(lane_id < kNumRDMARanks) {
+        while (lane_id < kNumRDMARanks) {
             start_offset = ld_volatile_global(nvl_channel_prefix_start.buffer() + lane_id);
             end_offset   = ld_volatile_global(nvl_channel_prefix_end.buffer() + lane_id);
-            if(start_offset < 0 && end_offset < 0) {
+            if (start_offset < 0 and end_offset < 0) {
                 start_offset = -start_offset - 1, end_offset = -end_offset - 1;
                 total_offset += start_offset;
                 break;
             }
-            // 超时检查
-            if (wall_clock64() - start_time > NUM_TIMEOUT_CYCLES) {
-                printf("DeepEP dispatch NVL receiver timeout, channel: %d, RDMA: %d, nvl: %d, src RDMA: %d, src nvl: %d, start: %d, end: %d\n",
-                        channel_id, rdma_rank, nvl_rank, lane_id, src_nvl_rank, start_offset, end_offset);
+
+            // Timeout check
+            long long int elapsed_time =
+                wall_clock64() > start_time ? wall_clock64() - start_time : 0;
+            if (elapsed_time > NUM_TIMEOUT_CYCLES) {
+                printf("DeepEP dispatch NVL receiver timeout, channel: %d, RDMA: %d, nvl: %d, src "
+                       "RDMA: %d, src nvl: %d, start: %d, end: %d\n",
+                       channel_id, rdma_rank, nvl_rank, lane_id, src_nvl_rank, start_offset,
+                       end_offset);
                 trap();
             }
         }
-
         num_tokens_to_recv = warp_reduce_sum(end_offset - start_offset);
 
-        // 保存以供合并使用
-        if(lane_id < kNumRDMARanks && !kCachedMode)
-            recv_gbl_channel_prefix_matrix[(lane_id * NUM_MAX_NVL_PEERS + src_nvl_rank) * num_channels + channel_id] = total_offset;
+        // Save for combine usage
+        if (lane_id < kNumRDMARanks and not kCachedMode)
+            recv_gbl_channel_prefix_matrix[(lane_id * NUM_MAX_NVL_PEERS + src_nvl_rank) *
+                                               num_channels +
+                                           channel_id] = total_offset;
         syncwarp();
 
         int cached_channel_head_idx = 0, cached_channel_tail_idx = 0;
-        while(num_tokens_to_recv > 0) {
-            // 通过通道0检查通道状态
+        while (num_tokens_to_recv > 0) {
+            // Check channel status by lane 0
             start_time = wall_clock64();
-            while(lane_id == 0) {
-                // 准备复制
-                if(cached_channel_head_idx != cached_channel_tail_idx)
+            while (lane_id == 0) {
+                // Ready to copy
+                if (cached_channel_head_idx != cached_channel_tail_idx)
                     break;
-                cached_channel_tail_idx = ld_acquire_sys_global(nvl_channel_tail.buffer());
-                // 超时检查
-                if (wall_clock64() - start_time > NUM_TIMEOUT_CYCLES) {
-                    printf("DeepEP dispatch NVL receiver timeout, channel: %d, RDMA: %d, nvl: %d, src NVL: %d, head: %d, tail: %d\n",
-                            channel_id, rdma_rank, nvl_rank, src_nvl_rank, cached_channel_head_idx, cached_channel_tail_idx);
+                cached_channel_tail_idx = ld_relaxed_sys_global(nvl_channel_tail.buffer());
+
+                // Timeout check
+                long long int elapsed_time =
+                    wall_clock64() > start_time ? wall_clock64() - start_time : 0;
+                if (elapsed_time > NUM_TIMEOUT_CYCLES) {
+                    printf("DeepEP dispatch NVL receiver timeout, channel: %d, RDMA: %d, nvl: %d, "
+                           "src NVL: %d, head: %d, tail: %d\n",
+                           channel_id, rdma_rank, nvl_rank, src_nvl_rank, cached_channel_head_idx,
+                           cached_channel_tail_idx);
                     trap();
                 }
             }
 
-            // 同步队列尾部
+            // Sync queue tail
             cached_channel_tail_idx = shfl_sync(cached_channel_tail_idx, 0);
 
-            // 复制数据
+            // Copy data
             int num_recv_tokens = cached_channel_tail_idx - cached_channel_head_idx;
-            for(int chunk_idx = 0; chunk_idx < num_recv_tokens; ++chunk_idx, --num_tokens_to_recv) {
-                int token_idx_in_buffer = (cached_channel_head_idx++) % num_max_nvl_chunked_recv_tokens;
+            for (int chunk_idx = 0; chunk_idx < num_recv_tokens;
+                 ++chunk_idx, --num_tokens_to_recv) {
+                int token_idx_in_buffer =
+                    (cached_channel_head_idx++) % num_max_nvl_chunked_recv_tokens;
                 auto    meta = ld_nc_global(nvl_channel_src_meta.buffer() + token_idx_in_buffer);
                 int64_t recv_token_idx = shfl_sync(total_offset, meta.src_rdma_rank);
                 (lane_id == meta.src_rdma_rank) ? (total_offset += 1) : 0;
 
-                // 复制数据
-                UNROLLED_WARP_COPY(5,
-                                lane_id,
-                                hidden_int4,
-                                recv_x + recv_token_idx * hidden_int4,
+                // Copy data
+                UNROLLED_WARP_COPY(5, lane_id, hidden_int4, recv_x + recv_token_idx * hidden_int4,
                                    nvl_channel_x.buffer() + token_idx_in_buffer * hidden_int4,
-                                ld_nc_global,
-                                st_na_global);
+                                   ld_nc_global, st_na_global);
 
-                // 复制源元数据
-                if(lane_id == 0 && !kCachedMode)
+                // Copy source meta
+                if (lane_id == 0 and not kCachedMode)
                     st_na_global(recv_src_meta + recv_token_idx, meta);
 
-                // 复制比例
-                UNROLLED_WARP_COPY(1,
-                                lane_id,
-                                num_scales,
+                // Copy scales
+                UNROLLED_WARP_COPY(1, lane_id, num_scales,
                                    recv_x_scales + recv_token_idx * num_scales,
                                    nvl_channel_x_scales.buffer() + token_idx_in_buffer * num_scales,
-                                ld_nc_global,
-                                st_na_global);
+                                   ld_nc_global, st_na_global);
 
-                // 复制 `topk_idx` 和 `topk_weights`
-                if(lane_id < num_topk) {
+                // Copy `topk_idx` and `topk_weights`
+                if (lane_id < num_topk) {
                     auto recv_idx   = recv_token_idx * num_topk + lane_id;
                     auto buffer_idx = token_idx_in_buffer * num_topk + lane_id;
-                    st_na_global(recv_topk_idx + recv_idx, static_cast<int64_t>(ld_nc_global(nvl_channel_topk_idx.buffer() + buffer_idx)));
-                    st_na_global(recv_topk_weights + recv_idx, ld_nc_global(nvl_channel_topk_weights.buffer() + buffer_idx));
+                    st_na_global(recv_topk_idx + recv_idx,
+                                 static_cast<int64_t>(
+                                     ld_nc_global(nvl_channel_topk_idx.buffer() + buffer_idx)));
+                    st_na_global(recv_topk_weights + recv_idx,
+                                 ld_nc_global(nvl_channel_topk_weights.buffer() + buffer_idx));
                 }
             }
 
-            // 移动队列
+            // Move queue
             syncwarp();
-            if(lane_id == 0) {
+            if (lane_id == 0)
                 st_relaxed_sys_global(nvl_channel_head.buffer(), cached_channel_head_idx);
         }
-        } // while(num_tokens_to_recv > 0)
     }
-#if !defined(FORCE_DUSHMEM_API) && !defined(ROCM_DISABLE_CTX)
-    shmem_wg_ctx_destroy(&ctx);
-#endif
+    rocshmem::rocshmem_wg_ctx_destroy(&ctx);
 }
 
 void dispatch(void *recv_x, float *recv_x_scales, int64_t *recv_topk_idx, float *recv_topk_weights,
@@ -1152,8 +1135,6 @@ void dispatch(void *recv_x, float *recv_x_scales, int64_t *recv_topk_idx, float
               int num_ranks, bool is_cached_dispatch, hipStream_t stream, int num_channels,
               bool low_latency_mode) {
     constexpr int kNumDispatchRDMASenderWarps = 7;
-    // Make sure never OOB
-    EP_HOST_ASSERT(static_cast<int64_t>(num_scales) * scale_hidden_stride < std::numeric_limits<int>::max());
 
 #define DISPATCH_LAUNCH_CASE(num_rdma_ranks)                                                       \
     {                                                                                              \
@@ -1181,8 +1162,8 @@ void dispatch(void *recv_x, float *recv_x_scales, int64_t *recv_topk_idx, float
     EP_HOST_ASSERT((topk_idx == nullptr) == (topk_weights == nullptr));
     EP_HOST_ASSERT((recv_topk_idx == nullptr) == (recv_topk_weights == nullptr));
 
-    SETUP_LAUNCH_CONFIG(num_channels * NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL,
-                        (1 + NUM_MAX_NVL_PEERS) * kWarpSize, stream);
+    SETUP_LAUNCH_CONFIG(num_channels * 2,
+                        (kNumDispatchRDMASenderWarps + 1 + NUM_MAX_NVL_PEERS) * kWarpSize, stream);
     SWITCH_RDMA_RANKS(DISPATCH_LAUNCH_CASE);
 #undef DISPATCH_LAUNCH_CASE
 }
@@ -1209,7 +1190,7 @@ cached_notify(const int rdma_clean_offset, const int rdma_num_int_clean, const i
     if (sm_id == 0) {
         // Barrier for RDMA
         if (thread_id == kWarpSize)
-            dushmem_barrier_with_same_gpu_idx<kLowLatencyMode>(rdma_team);
+            shmem_sync_with_same_gpu_idx<kLowLatencyMode>(rdma_team);
 
         // Barrier for NVL
         barrier_block<NUM_MAX_NVL_PEERS>(barrier_signal_ptrs, nvl_rank);
@@ -1228,7 +1209,7 @@ cached_notify(const int rdma_clean_offset, const int rdma_num_int_clean, const i
 
         // Barrier again
         if (thread_id == kWarpSize)
-            dushmem_barrier_with_same_gpu_idx<kLowLatencyMode>(rdma_team);
+            shmem_sync_with_same_gpu_idx<kLowLatencyMode>(rdma_team);
 
         // Barrier again
         barrier_block<NUM_MAX_NVL_PEERS>(barrier_signal_ptrs, nvl_rank);
@@ -1236,13 +1217,13 @@ cached_notify(const int rdma_clean_offset, const int rdma_num_int_clean, const i
         if (is_cached_dispatch)
             return;
 
+        EP_DEVICE_ASSERT(num_warps >= num_channels);
         EP_DEVICE_ASSERT(num_rdma_ranks <= kWarpSize);
 
         // Iterate in reverse order
-        for (int channel_id = warp_id; channel_id < num_channels; channel_id += num_warps) {
-            if (lane_id < num_rdma_ranks) {
+        if (lane_id < num_rdma_ranks and warp_id < num_channels) {
             int token_start_idx, token_end_idx;
-                get_channel_task_range(num_combined_tokens, num_channels, channel_id, token_start_idx,
+            get_channel_task_range(num_combined_tokens, num_channels, warp_id, token_start_idx,
                                    token_end_idx);
 
             // NOTES: `1 << 25` is a heuristic large number
@@ -1257,26 +1238,26 @@ cached_notify(const int rdma_clean_offset, const int rdma_num_int_clean, const i
                 }
             }
         }
-        }
     } else {
         if (is_cached_dispatch)
             return;
 
-        EP_DEVICE_ASSERT(rdma_channel_prefix_matrix != nullptr and rdma_rank_prefix_sum != nullptr);
+        EP_DEVICE_ASSERT(num_warps >= num_channels);
+        EP_DEVICE_ASSERT(rdma_channel_prefix_matrix != nullptr and
+                                  rdma_rank_prefix_sum != nullptr);
         EP_STATIC_ASSERT(NUM_MAX_NVL_PEERS <= kWarpSize, "Too many NVL peers");
         constexpr int num_clean_sms = 2;
 
-        for (int channel_id = warp_id; channel_id < num_channels; channel_id += num_warps) {
-            if (lane_id < NUM_MAX_NVL_PEERS ) {
+        if (lane_id < NUM_MAX_NVL_PEERS and warp_id < num_channels) {
             for (int dst_rdma_rank = sm_id - num_clean_sms; dst_rdma_rank < num_rdma_ranks;
-                    dst_rdma_rank += num_channels * NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL - num_clean_sms) {
+                 dst_rdma_rank += num_channels * 2 - num_clean_sms) {
                 // Iterate in reverse order
                 int token_start_idx =
-                        channel_id == 0
+                    warp_id == 0
                         ? 0
-                            : rdma_channel_prefix_matrix[dst_rdma_rank * num_channels + channel_id - 1];
+                        : rdma_channel_prefix_matrix[dst_rdma_rank * num_channels + warp_id - 1];
                 int token_end_idx =
-                        rdma_channel_prefix_matrix[dst_rdma_rank * num_channels + channel_id];
+                    rdma_channel_prefix_matrix[dst_rdma_rank * num_channels + warp_id];
                 int shift = dst_rdma_rank == 0 ? 0 : rdma_rank_prefix_sum[dst_rdma_rank - 1];
                 token_start_idx += shift, token_end_idx += shift;
 
@@ -1294,7 +1275,6 @@ cached_notify(const int rdma_clean_offset, const int rdma_num_int_clean, const i
             }
         }
     }
-    }
 }
 
 void cached_notify(int hidden_int4, int num_scales, int num_topk_idx, int num_topk_weights,
@@ -1305,7 +1285,7 @@ void cached_notify(int hidden_int4, int num_scales, int num_topk_idx, int num_to
                    int num_max_nvl_chunked_recv_tokens, int **barrier_signal_ptrs, int rank,
                    hipStream_t stream, int64_t num_rdma_bytes, int64_t num_nvl_bytes,
                    bool is_cached_dispatch, bool low_latency_mode) {
-    const int  num_threads    = ::min(1024, ::max(128, kWarpSize * num_channels));
+    const int  num_threads    = ::max(128, kWarpSize * num_channels);
     const auto num_rdma_ranks = num_ranks / NUM_MAX_NVL_PEERS;
 
     // Get clean meta
@@ -1321,11 +1301,11 @@ void cached_notify(int hidden_int4, int num_scales, int num_topk_idx, int num_to
                        num_nvl_bytes);
     EP_HOST_ASSERT(num_rdma_bytes < std::numeric_limits<int>::max());
     EP_HOST_ASSERT(num_nvl_bytes < std::numeric_limits<int>::max());
-    EP_HOST_ASSERT(num_channels * NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL > 2);
+    EP_HOST_ASSERT(num_channels * 2 > 2);
 
     // Launch kernel
     auto cached_notify_func = low_latency_mode ? cached_notify<true> : cached_notify<false>;
-    SETUP_LAUNCH_CONFIG(num_channels * NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL, num_threads, stream);
+    SETUP_LAUNCH_CONFIG(num_channels * 2, num_threads, stream);
     LAUNCH_KERNEL_NON_COOPERATIVE(
         &cfg, cached_notify_func, rdma_clean_meta.first, rdma_clean_meta.second,
         nvl_clean_meta.first, nvl_clean_meta.second, combined_rdma_head, num_combined_tokens,
@@ -1334,45 +1314,49 @@ void cached_notify(int hidden_int4, int num_scales, int num_topk_idx, int num_to
         cpu_rdma_team);
 }
 
-template <int kNumRanks, typename dtype_t, int kMaxNumRanks, bool kUseMLS, typename GetAddrFn, typename ReceiveTWFn>
+template <int kNumRanks, typename dtype_t, int kMaxNumRanks, int kWidth, typename ReceiveFn, typename ReceiveTWFn>
 __device__ int combine_token(bool is_token_in_rank, int head_idx,
                              int lane_id, int hidden_int4, int num_topk,
                              int4* combined_row, float* combined_topk_weights,
-                             int num_max_recv_tokens,
-                             const GetAddrFn& get_addr_fn, const ReceiveTWFn& recv_tw_fn) {
+                             int num_max_recv_tokens, const ReceiveFn& recv_fn, const ReceiveTWFn& recv_tw_fn) {
     constexpr auto kDtypePerInt4 = sizeof(int4) / sizeof(dtype_t);
 
     // Broadcast current heads
     // Lane `i` holds the head of rank `i` and `is_token_in_rank`
-    EP_STATIC_ASSERT(kMaxNumRanks <= kWarpSize, "Too many ranks");
+    EP_STATIC_ASSERT(kMaxNumRanks <= kWidth, "Too many ranks");
     int num_topk_ranks = 0, topk_ranks[kMaxNumRanks], slot_indices[kMaxNumRanks];
     #pragma unroll
-    for (int i = 0; i < kNumRanks; ++ i) if (shfl_sync(is_token_in_rank, i)) {
-        slot_indices[num_topk_ranks] = shfl_sync(head_idx, i) % num_max_recv_tokens;
+    for (int i = 0; i < kNumRanks; ++ i) if (shfl_sync(is_token_in_rank, i, kWidth)) {
+        slot_indices[num_topk_ranks] = shfl_sync(head_idx, i, kWidth) % num_max_recv_tokens;
         topk_ranks[num_topk_ranks ++] = i;
     }
     EP_DEVICE_ASSERT(num_topk_ranks <= kMaxNumRanks);
 
     // Reduce data    
     #pragma unroll
-    for (int i = lane_id; i < hidden_int4; i += kWarpSize) {
-        // Read buffers
-        float values[kDtypePerInt4] = {0};  // 8 × 4B = 32B
+    for (int i = lane_id; i < hidden_int4; i += kWidth) {
+        float values[kDtypePerInt4] = {0};
+
+        // Temporary buffer
+        int4 temp;
+
         #pragma unroll
         for (int j = 0; j < num_topk_ranks; ++j) {
-            int4 recv_value = ld_nc_global(get_addr_fn(topk_ranks[j], slot_indices[j], i));
-            auto recv_dtypes = reinterpret_cast<const dtype_t*>(&recv_value);
+            temp = recv_fn(topk_ranks[j], slot_indices[j], i);
+            const dtype_t* d = reinterpret_cast<const dtype_t*>(&temp);
+
             #pragma unroll
             for (int k = 0; k < kDtypePerInt4; ++k)
-                values[k] += static_cast<float>(recv_dtypes[k]);
+                values[k] += static_cast<float>(d[k]);
         }
 
-        // Cast back to `dtype_t` and write
         int4 out_int4;
-        auto out_dtypes = reinterpret_cast<dtype_t*>(&out_int4);
+        dtype_t* out_dtypes = reinterpret_cast<dtype_t*>(&out_int4);
+
         #pragma unroll
-        for (int j = 0; j < kDtypePerInt4; ++ j)
+        for (int j = 0; j < kDtypePerInt4; ++j)
             out_dtypes[j] = static_cast<dtype_t>(values[j]);
+
         st_na_global(combined_row + i, out_int4);
     }
 
@@ -1389,87 +1373,98 @@ __device__ int combine_token(bool is_token_in_rank, int head_idx,
     return topk_ranks[0];
 }
 
-template <bool kLowLatencyMode,
-          int kNumRDMARanks,
-          typename dtype_t,
+template<bool kLowLatencyMode,
+         int kNumRDMARanks, typename dtype_t,
          int kNumCombineForwarderWarps,
          int kNumTopkRDMARanks = get_num_topk_rdma_ranks(kNumRDMARanks),
          int kNumWarpsPerForwarder = (kNumCombineForwarderWarps / kNumRDMARanks > 0) ? kNumCombineForwarderWarps / kNumRDMARanks : 1,
          int kNumForwarders = kNumRDMARanks * kNumWarpsPerForwarder,
-          int kNumRDMAReceivers     = kNumForwarders>
-__global__ void __launch_bounds__((1 + NUM_MAX_NVL_PEERS) * kWarpSize, 1) 
-combine(int4 *combined_x, float *combined_topk_weights, const bool *is_combined_token_in_rank,
-            const int4 *x, const float *topk_weights, const int4 *bias_0, const int4 *bias_1,
-            const int *combined_rdma_head, const int *combined_nvl_head, const SourceMeta *src_meta,
-            const int *rdma_channel_prefix_matrix, const int *rdma_rank_prefix_sum,
-            const int *gbl_channel_prefix_matrix, int num_tokens, int num_combined_tokens,
-            int hidden, int num_topk, void *rdma_buffer_ptr, int num_max_rdma_chunked_send_tokens,
-            int num_max_rdma_chunked_recv_tokens, void **buffer_ptrs,
-            int num_max_nvl_chunked_send_tokens, int num_max_nvl_chunked_recv_tokens, int rank,
-            int num_ranks) {
+         int kNumRDMAReceivers = kNumRDMARanks <=8 ? kNumForwarders + NUM_MAX_NVL_PEERS / 2: kNumForwarders + NUM_MAX_NVL_PEERS,
+         int kBlockThreads = (kNumRDMARanks > 8) ? ((NUM_MAX_NVL_PEERS + kNumForwarders) * kEmulatedWarpSize + kWarpSize) : ((NUM_MAX_NVL_PEERS/2 + 1 + kNumForwarders) * kWarpSize) >
+__global__ void __launch_bounds__(kBlockThreads, 1)
+combine(int4* combined_x, float* combined_topk_weights,
+        const bool* is_combined_token_in_rank,
+        const int4* x, const float* topk_weights, const int4 *bias_0, const int4 *bias_1,
+        const int* combined_rdma_head, const int* combined_nvl_head,
+        const SourceMeta* src_meta, const int* rdma_channel_prefix_matrix, const int* rdma_rank_prefix_sum, const int* gbl_channel_prefix_matrix,
+        int num_tokens, int num_combined_tokens, int hidden, int num_topk,
+        void* rdma_buffer_ptr, int num_max_rdma_chunked_send_tokens, int num_max_rdma_chunked_recv_tokens,
+        void** buffer_ptrs, int num_max_nvl_chunked_send_tokens, int num_max_nvl_chunked_recv_tokens,
+        int rank, int num_ranks) {
     enum class WarpRole {
         kNVLSender,
         kNVLAndRDMAForwarder,
         kRDMAReceiver,
-        kRDMACoordinator,
-        kNVLCoordinator
+        kCoordinator
     };
+
+    constexpr auto kNVLPeersHyb = (kNumRDMARanks > 8) ? NUM_MAX_NVL_PEERS : NUM_MAX_NVL_PEERS / 2;
+    constexpr auto  kWarpHyb = kNumRDMARanks > 8 ? kEmulatedWarpSize : kWarpSize;
+    const auto sm_id = static_cast<int>(blockIdx.x);
+    const auto num_threads = static_cast<int>(blockDim.x);
+    const int num_warps = kNumRDMARanks > 8 ? (num_threads / kEmulatedWarpSize - 1) : (num_threads / kWarpSize);
+    auto thread_id = static_cast<int>(threadIdx.x);
+    const auto num_channels = static_cast<int>(gridDim.x) / 2, channel_id = sm_id / 2;
+    const bool is_rdma_receiver_sm = sm_id % 2 == 1;
+
 #if !defined(FORCE_DUSHMEM_API) && !defined(ROCM_DISABLE_CTX)
     __shared__ shmem_ctx_t ctx;
     shmem_wg_ctx_create(&ctx);
-#endif
-    EP_STATIC_ASSERT(kNumCombineForwarderWarps <= kWarpSize, "Invalid number of forwarder warps");
 
-    const auto sm_id       = static_cast<int>(blockIdx.x);
-    const auto num_threads = static_cast<int>(blockDim.x), num_warps = num_threads / kWarpSize;
-    const auto thread_id   = static_cast<int>(threadIdx.x), warp_id = thread_id / kWarpSize, lane_id = get_lane_id();
-    const auto num_channels = static_cast<int>(gridDim.x) / NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL,
-               channel_id   = sm_id / NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL;
+#endif
 
+    EP_DEVICE_ASSERT(num_topk <= kEmulatedWarpSize);
+    EP_DEVICE_ASSERT(hidden % (sizeof(int4) / sizeof(dtype_t)) == 0);
     const auto hidden_int4 = hidden / (sizeof(int4) / sizeof(dtype_t));
 
     // NOTES: we decouple a channel into 2 SMs
     const auto rdma_rank = rank / NUM_MAX_NVL_PEERS, nvl_rank = rank % NUM_MAX_NVL_PEERS;
-
-    const auto role_meta = [=]() -> std::pair<WarpRole, int> {
-        if (sm_id % NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL == 1) {
-            return {WarpRole::kNVLSender, (warp_id + channel_id) % NUM_MAX_NVL_PEERS};
-        } else if (sm_id % NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL == 0) {
-            if(warp_id < kNumForwarders) {
-                return {WarpRole::kNVLAndRDMAForwarder, (warp_id + channel_id) % kNumForwarders};
+    auto role_meta = [=]() -> std::pair<WarpRole, int> {
+        auto warp_id = thread_id / kWarpHyb;
+        if (not is_rdma_receiver_sm) {
+            if (warp_id < kNVLPeersHyb) {
+                auto shuffled_warp_id = warp_id;
+                shuffled_warp_id = (shuffled_warp_id + channel_id) % kNVLPeersHyb;
+                return {WarpRole::kNVLSender, shuffled_warp_id};
+            } else if (warp_id < kNVLPeersHyb + kNumForwarders) {
+                auto shuffled_warp_id = warp_id - kNVLPeersHyb;
+                shuffled_warp_id = (shuffled_warp_id + channel_id) % kNumForwarders;
+                return {WarpRole::kNVLAndRDMAForwarder, shuffled_warp_id};
             } else {
-                return {WarpRole::kRDMACoordinator, 0};
+                return {WarpRole::kCoordinator, 0};
             }
         } else {
-            if(warp_id < kNumForwarders) {
+            if (warp_id < kNVLPeersHyb + kNumForwarders) {
                 return {WarpRole::kRDMAReceiver, warp_id};
             } else {
-                return {WarpRole::kNVLCoordinator, 0};
+                return {WarpRole::kCoordinator, 0};
             }
         }
     }();
-
     auto warp_role = role_meta.first;
-    auto target_rank = role_meta.second; // Not applicable for RDMA senders
+    auto warp_id = role_meta.second;
 
-    EP_DEVICE_ASSERT(num_warps == NUM_MAX_NVL_PEERS + 1);
     auto num_max_nvl_chunked_recv_tokens_per_rdma = num_max_nvl_chunked_recv_tokens / kNumRDMARanks;
 
     // This approach is designed to sync multiple warps in a loop
     constexpr int num_sync_large_iteration = 64;
-    constexpr int rdma_warp_counters = kNumRDMARanks * num_sync_large_iteration;
-    __shared__ volatile int sync_large_warp_counters[2 * rdma_warp_counters];   
-    for (int i = thread_id; i < 2 * rdma_warp_counters; i += num_threads) {
+    __shared__ volatile int rdma_receiver_counter[1];
+    __shared__ volatile int rdma_forwarder_counter[1];
+    __shared__ volatile uint8_t sync_large_warp_counters[2 * kNumRDMARanks * num_sync_large_iteration ];
+    if (threadIdx.x==0){
+        rdma_receiver_counter[0] = 0;
+        rdma_forwarder_counter[0] = 0;
+    }
+    
+    for (int i = thread_id; i < 2 * kNumRDMARanks * num_sync_large_iteration; i += num_threads) {
         sync_large_warp_counters[i] = 0;
     }
     __syncthreads();
-
     if (warp_role == WarpRole::kNVLSender) {
-        if(warp_id >= NUM_MAX_NVL_PEERS) {
-            return;
-        }
+        // NVL producers
+        const int dst_nvl_rank = kNumRDMARanks <= 8 ? (warp_id * 2 + (thread_id % kWarpSize) / kEmulatedWarpSize) : warp_id;  
+        auto lane_id = get_lane_id() % kEmulatedWarpSize;
 
-        const auto dst_nvl_rank = target_rank;
         // NVL layouts
         // NOTES: to avoid deadlocks, we use separate NVL buffers for different RDMA sources
         auto dst_buffer_ptr = buffer_ptrs[dst_nvl_rank], local_buffer_ptr = buffer_ptrs[nvl_rank];
@@ -1481,7 +1476,7 @@ combine(int4 *combined_x, float *combined_topk_weights, const bool *is_combined_
 
         // Get tasks for each RDMA lane
         int token_start_idx = 0, token_end_idx = 0;
-        if(lane_id < kNumRDMARanks) {
+        if (lane_id < kNumRDMARanks) {
             int prefix_idx = (lane_id * NUM_MAX_NVL_PEERS + dst_nvl_rank) * num_channels + channel_id;
             token_start_idx = gbl_channel_prefix_matrix[prefix_idx];
             token_end_idx = (prefix_idx == num_channels * num_ranks - 1) ? num_tokens : gbl_channel_prefix_matrix[prefix_idx + 1];
@@ -1490,94 +1485,81 @@ combine(int4 *combined_x, float *combined_topk_weights, const bool *is_combined_
 
         // NOTES: here the cached value of each lane is only responsible for a single RDMA buffer
         int cached_channel_head_idx = 0, cached_channel_tail_idx = 0;
-        EP_STATIC_ASSERT(kNumRDMARanks <= kWarpSize, "Invalid number of RDMA peers");
+        EP_STATIC_ASSERT(kNumRDMARanks <= kEmulatedWarpSize, "Invalid number of RDMA peers");
 
         // Iterate over all tokens and send by chunks
-        while(true) {
+        while (true) {
             // Exit if possible
-            if(__all_sync(kFullWarpMask, token_start_idx >= token_end_idx))
+            if (__all_sync(kFullWarpMask, token_start_idx >= token_end_idx))
                 break;
 
             // Decide next RDMA buffer to send
             bool is_lane_ready = false;
             auto start_time = wall_clock64();
-
-            while(true) {
+            while (true) {
                 int num_used_slots = cached_channel_tail_idx - cached_channel_head_idx;
-                is_lane_ready      = lane_id < kNumRDMARanks and token_start_idx < token_end_idx and
-                                num_max_nvl_chunked_recv_tokens_per_rdma - num_used_slots >= num_max_nvl_chunked_send_tokens;
-
-                if(__any_sync(kFullWarpMask, is_lane_ready))
+                is_lane_ready = lane_id < kNumRDMARanks and token_start_idx < token_end_idx and num_max_nvl_chunked_recv_tokens_per_rdma - num_used_slots >= num_max_nvl_chunked_send_tokens;
+                if(__any_sync(kFirstHalfMask, is_lane_ready))
+                    break;
+                if(__any_sync(kSecondHalfMask, is_lane_ready))
                     break;
-
                 // Retry
-                if(lane_id < kNumRDMARanks and token_start_idx < token_end_idx)
+                if (lane_id < kNumRDMARanks and token_start_idx < token_end_idx)
                     cached_channel_head_idx = ld_volatile_global(nvl_channel_head.buffer() + lane_id);
 
                 // Timeout check
-                if(wall_clock64() - start_time > NUM_TIMEOUT_CYCLES and lane_id < kNumRDMARanks) {
-                    printf("DeepEP combine NVL sender timeout, channel: %d, RDMA: %d, nvl: %d, dst NVL: %d, "
-                        "RDMA lane: %d, head: %d, tail: %d, start: %d, end: %d\n",
-                        channel_id,
-                        rdma_rank,
-                        nvl_rank,
-                        dst_nvl_rank,
-                        lane_id,
-                        ld_volatile_global(nvl_channel_head.buffer() + lane_id),
-                        cached_channel_tail_idx,
-                        token_start_idx,
-                        token_end_idx);
+                long long int elapsed_time = wall_clock64() > start_time ? wall_clock64() - start_time : 0;
+                if (elapsed_time > NUM_TIMEOUT_CYCLES and lane_id < kNumRDMARanks) {
+                    printf("DeepEP combine NVL sender timeout, channel: %d, RDMA: %d, nvl: %d, dst NVL: %d, RDMA lane: %d, head: %d, tail: %d, start: %d, end: %d\n",
+                           channel_id, rdma_rank, nvl_rank, dst_nvl_rank, lane_id, ld_volatile_global(nvl_channel_head.buffer() + lane_id), cached_channel_tail_idx,
+                           token_start_idx, token_end_idx);
                     trap();
                 }
+                __builtin_amdgcn_s_sleep(1);
             }
 
             // Sync token start index and count
-            for(int current_rdma_idx = 0; current_rdma_idx < kNumRDMARanks; ++current_rdma_idx) {
-                if(shfl_sync((token_start_idx >= token_end_idx) or (not is_lane_ready), current_rdma_idx))
+            for (int current_rdma_idx = 0; current_rdma_idx < kNumRDMARanks; ++ current_rdma_idx) {
+                if (shfl_sync((token_start_idx >= token_end_idx) or (not is_lane_ready), current_rdma_idx, kEmulatedWarpSize))
                     continue;
 
                 // Sync token start index
-                auto token_idx          = static_cast<int64_t>(shfl_sync(token_start_idx, current_rdma_idx));
-                int num_tokens_in_chunk = shfl_sync(min(num_max_nvl_chunked_send_tokens, token_end_idx - token_start_idx), current_rdma_idx);
+                auto token_idx = static_cast<int64_t>(shfl_sync(token_start_idx, current_rdma_idx, kEmulatedWarpSize));
+                int num_tokens_in_chunk = shfl_sync(min(num_max_nvl_chunked_send_tokens, token_end_idx - token_start_idx), current_rdma_idx, kEmulatedWarpSize);
 
                 // Send by chunk
-                for(int chunk_idx = 0; chunk_idx < num_tokens_in_chunk; ++chunk_idx, ++token_idx) {
+                for (int chunk_idx = 0; chunk_idx < num_tokens_in_chunk; ++ chunk_idx, ++ token_idx) {
                     // Get an empty slot
                     int dst_slot_idx = 0;
-                    if(lane_id == current_rdma_idx) {
-                        dst_slot_idx = (cached_channel_tail_idx++) % num_max_nvl_chunked_recv_tokens_per_rdma;
+                    if (lane_id == current_rdma_idx) {
+                        dst_slot_idx = (cached_channel_tail_idx ++) % num_max_nvl_chunked_recv_tokens_per_rdma;
                         dst_slot_idx = current_rdma_idx * num_max_nvl_chunked_recv_tokens_per_rdma + dst_slot_idx;
                     }
-                    dst_slot_idx = shfl_sync(dst_slot_idx, current_rdma_idx);
+                    dst_slot_idx = shfl_sync(dst_slot_idx, current_rdma_idx, kEmulatedWarpSize);
 
                     // Copy data
                     auto shifted_x_buffers = nvl_channel_x.buffer() + dst_slot_idx * hidden_int4;
                     auto shifted_x = x + token_idx * hidden_int4;
-                    UNROLLED_WARP_COPY(5, lane_id, hidden_int4, shifted_x_buffers, shifted_x, ld_nc_global, st_na_global);
+                    UNROLLED_WARP_COPY_EMULATED(5, lane_id, hidden_int4, shifted_x_buffers, shifted_x, ld_nc_global, st_na_global);
 
                     // Copy source meta
-                    if(lane_id == 0)
+                    if (lane_id == num_topk)
                         st_na_global(nvl_channel_src_meta.buffer() + dst_slot_idx, ld_nc_global(src_meta + token_idx));
 
                     // Copy `topk_weights`
-                    if(lane_id < num_topk)
-                        st_na_global(nvl_channel_topk_weights.buffer() + dst_slot_idx * num_topk + lane_id,
-                                    ld_nc_global(topk_weights + token_idx * num_topk + lane_id));
+                    if (lane_id < num_topk)
+                        st_na_global(nvl_channel_topk_weights.buffer() + dst_slot_idx * num_topk + lane_id, ld_nc_global(topk_weights + token_idx * num_topk + lane_id));
                 }
                 lane_id == current_rdma_idx ? (token_start_idx = static_cast<int>(token_idx)) : 0;
             }
 
             // Move queue tail
             syncwarp();
-            if(lane_id < kNumRDMARanks and is_lane_ready) {
-                st_release_sys_global(nvl_channel_tail.buffer() + lane_id, cached_channel_tail_idx);
-            }
+            if (lane_id < kNumRDMARanks and is_lane_ready)
+                st_relaxed_sys_global(nvl_channel_tail.buffer() + lane_id, cached_channel_tail_idx);
         }
     } else {
-        if(warp_id > kNumForwarders) {
-            return;
-        }
-
+        auto lane_id = get_lane_id() % kWarpHyb;
         // Combiners and coordinators
         // RDMA symmetric layout
         auto hidden_bytes = hidden_int4 * sizeof(int4);
@@ -1604,38 +1586,48 @@ combine(int4 *combined_x, float *combined_topk_weights, const bool *is_combined_
         __shared__ volatile int rdma_receiver_rdma_head[kNumRDMAReceivers][kNumRDMARanks];
         __shared__ volatile bool rdma_receiver_retired[kNumRDMAReceivers];
 
+        auto sync_forwarder_smem = [&]() {
+            if (lane_id==0) {
+                volatile int ret = __hip_atomic_fetch_add(&rdma_forwarder_counter[0], 1, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_WORKGROUP);
+            }
+            syncwarp();
+            while(rdma_forwarder_counter[0]<(kNumForwarders + 1)){}
+        };
+        auto sync_rdma_receiver_smem = [&]() {
+            if (lane_id==0) {
+                volatile int ret = __hip_atomic_fetch_add(&rdma_receiver_counter[0], 1, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_WORKGROUP);
+            }
+            syncwarp();
+            while(rdma_receiver_counter[0]<(kNumRDMAReceivers+1)){} 
+        };
+
         if (warp_role == WarpRole::kNVLAndRDMAForwarder) {
             // Receive from NVL ranks and forward to RDMA ranks
             // NOTES: this part is using "large warps" for each RDMA ranks
-            const auto dst_rdma_rank = target_rank / kNumWarpsPerForwarder;
-            const auto sub_warp_id   = target_rank % kNumWarpsPerForwarder;
+            const auto dst_rdma_rank = warp_id / kNumWarpsPerForwarder;
+            const auto sub_warp_id = warp_id % kNumWarpsPerForwarder;
             auto send_buffer = dst_rdma_rank == rdma_rank ? rdma_channel_data.recv_buffer(dst_rdma_rank) : rdma_channel_data.send_buffer(dst_rdma_rank);
-            // auto sync_large_warp     = [=]() {
-            //     if(kNumWarpsPerForwarder == 1) {
-            //         syncwarp();
-            //     } else {
-            //         // asm volatile("bar.sync %0, %1;" ::"r"(dst_rdma_rank + 2), "r"(kNumWarpsPerForwarder * kWarpSize));
-            //         // __syncthreads();
-            //         syncwarp();
-            //     }
-            // };
             auto sync_large_warp = [=](const int iter, const int mode) {
                 if (kNumWarpsPerForwarder == 1) {
                     syncwarp();
                 } else {
                     // LDS index to store for sync
-                        int lds_dst_rdma_rank = dst_rdma_rank + (iter % num_sync_large_iteration) * kNumRDMARanks + mode * rdma_warp_counters;
+                    int lds_dst_rdma_rank = dst_rdma_rank + (iter % num_sync_large_iteration) * kNumRDMARanks + mode * kNumRDMARanks * num_sync_large_iteration;
                     //reset index in the LDS to avoid race condition due to warp scheduling
-                        int reset_idx =         dst_rdma_rank + ((iter + num_sync_large_iteration/2) % num_sync_large_iteration) * kNumRDMARanks + mode * rdma_warp_counters;
-                        auto start_time = wall_clock64();
+                    int reset_idx =         dst_rdma_rank + ((iter + num_sync_large_iteration/2) % num_sync_large_iteration) * kNumRDMARanks + mode * kNumRDMARanks * num_sync_large_iteration;
+                    // if (lane_id==0)
+                    //     printf("rank %d dst_rdma_rank %d iter %d  warp_id %d  val %d\n", rank, dst_rdma_rank, iter, warp_id, sync_large_warp_counters[lds_dst_rdma_rank]);
+                    auto start_time = clock64();
                     if (lane_id == 0){
-                            volatile int ret = atomicAdd((int*)&sync_large_warp_counters[lds_dst_rdma_rank], 1);
+                        volatile int ret = __hip_atomic_fetch_add(
+                            &sync_large_warp_counters[lds_dst_rdma_rank], 1,
+                            __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_WORKGROUP);
                     }
                     syncwarp();
                     //The while(...) loop polls the counter until all warps have arrived
                     if (lane_id == 0){
                         while (sync_large_warp_counters[lds_dst_rdma_rank] < (kNumWarpsPerForwarder)){
-                                if (wall_clock64() - start_time > NUM_TIMEOUT_CYCLES) {
+                            if (clock64() - start_time > NUM_TIMEOUT_CYCLES) {
                                 printf("DeepEP combine sync timeout. current num_sync_large_iteration %d. double it.\n", num_sync_large_iteration );
                                 trap();
                             }
@@ -1648,9 +1640,11 @@ combine(int4 *combined_x, float *combined_topk_weights, const bool *is_combined_
                     syncwarp();
                 }
             };
-            EP_STATIC_ASSERT(kNumWarpsPerForwarder == 1 or kNumRDMARanks + 2 <= kNumCombineForwarderWarps, "Barriers are not enough");
+            EP_STATIC_ASSERT(kNumWarpsPerForwarder == 1 or kNumRDMARanks + 2 <= 16, "Barriers are not enough");
 
-            // Advance to the corresponding NVL buffer， 基于原本指针进行的地址偏移
+            // In case of running less than 8 nodes
+            constexpr bool kUseWave = (kNumRDMARanks <= 8);
+            // Advance to the corresponding NVL buffer
             nvl_channel_x.advance(dst_rdma_rank * num_max_nvl_chunked_recv_tokens_per_rdma * hidden_int4);
             nvl_channel_src_meta.advance(dst_rdma_rank * num_max_nvl_chunked_recv_tokens_per_rdma);
             nvl_channel_topk_weights.advance(dst_rdma_rank * num_max_nvl_chunked_recv_tokens_per_rdma * num_topk);
@@ -1659,10 +1653,9 @@ combine(int4 *combined_x, float *combined_topk_weights, const bool *is_combined_
 
             // Clean shared memory and sync
             EP_STATIC_ASSERT(NUM_MAX_NVL_PEERS <= kWarpSize, "Invalid number of NVL peers");
-            lane_id < NUM_MAX_NVL_PEERS ? (forwarder_nvl_head[target_rank][lane_id] = 0) : 0;
-            lane_id == 0 ? (forwarder_retired[target_rank] = false) : false;
-            // sync_forwarder_smem();
-            __syncthreads();
+            lane_id < NUM_MAX_NVL_PEERS ? (forwarder_nvl_head[warp_id][lane_id] = 0) : 0;
+            lane_id == 0 ? (forwarder_retired[warp_id] = false) : false;
+            sync_forwarder_smem();
 
             // Get count and cached head
             int cached_nvl_channel_tail_idx = 0;
@@ -1673,87 +1666,104 @@ combine(int4 *combined_x, float *combined_topk_weights, const bool *is_combined_
             combined_nvl_head += num_tokens_prefix * NUM_MAX_NVL_PEERS;
 
             // Iterate over all tokens and combine by chunks
-            for(int token_start_idx = 0; token_start_idx < num_tokens_to_combine; token_start_idx += num_max_rdma_chunked_send_tokens) {
+            for (int token_start_idx = 0; token_start_idx < num_tokens_to_combine; token_start_idx += num_max_rdma_chunked_send_tokens) {
                 // Check destination queue emptiness, or wait a buffer to be released
                 auto token_end_idx = min(token_start_idx + num_max_rdma_chunked_send_tokens, num_tokens_to_combine);
                 auto num_chunked_tokens = token_end_idx - token_start_idx;
                 auto start_time = wall_clock64();
-                while(sub_warp_id == 0 and lane_id == 0) {
+                while (sub_warp_id == 0 and lane_id == 0) {
                     // Inequality: `num_max_rdma_chunked_recv_tokens - (tail - head) >= num_chunked_tokens`
                     // Here, `token_start_idx` is the actual tail
                     int num_used_slots = token_start_idx - ld_volatile_global(rdma_channel_head.buffer(dst_rdma_rank));
-
-                    if(num_max_rdma_chunked_recv_tokens - num_used_slots >= num_chunked_tokens)
+                    if (num_max_rdma_chunked_recv_tokens - num_used_slots >= num_chunked_tokens)
                         break;
 
                     // Timeout check
-                    if (wall_clock64() - start_time > NUM_TIMEOUT_CYCLES) {
+                    long long int elapsed_time = wall_clock64() > start_time ? wall_clock64() - start_time : 0;
+                    if (elapsed_time > NUM_TIMEOUT_CYCLES) {
                         printf("DeepEP combine forwarder (RDMA check) timeout, channel: %d, RDMA: %d, nvl: %d, dst RDMA: %d, head: %ld, tail: %d, chunked: %d\n",
                                channel_id, rdma_rank, nvl_rank, dst_rdma_rank, ld_volatile_global(rdma_channel_head.buffer(dst_rdma_rank)), token_start_idx, num_chunked_tokens);
                         trap();
                     }
                 }
-                // sync_large_warp();
                 sync_large_warp(token_start_idx, 0);
-
                 // Combine and write to the RDMA buffer
-                for(int token_idx = token_start_idx + sub_warp_id; token_idx < token_end_idx; token_idx += kNumWarpsPerForwarder) {
+                for (int token_idx = token_start_idx + sub_warp_id; token_idx < token_end_idx; token_idx += kNumWarpsPerForwarder) {
                     // Read expected head
                     EP_STATIC_ASSERT(kNumRDMARanks <= kWarpSize, "Invalid number of RDMA peers");
                     int expected_head = -1;
-                    if(lane_id < NUM_MAX_NVL_PEERS)
+                    if (lane_id < NUM_MAX_NVL_PEERS)
                         expected_head = ld_nc_global(combined_nvl_head + token_idx * NUM_MAX_NVL_PEERS + lane_id);
 
                     // Wait lanes to be ready
                     start_time = wall_clock64();
-                    while(cached_nvl_channel_tail_idx <= expected_head) {
-                        cached_nvl_channel_tail_idx = ld_acquire_sys_global(nvl_channel_tail.buffer(lane_id));
-
+                    while (cached_nvl_channel_tail_idx <= expected_head) {
+                        cached_nvl_channel_tail_idx = ld_relaxed_sys_global(nvl_channel_tail.buffer(lane_id));
                         // Timeout check
-                        if (wall_clock64() - start_time > NUM_TIMEOUT_CYCLES and lane_id < NUM_MAX_NVL_PEERS) {
+                        long long int elapsed_time = wall_clock64() > start_time ? wall_clock64() - start_time : 0;
+                        if (elapsed_time > NUM_TIMEOUT_CYCLES and lane_id < NUM_MAX_NVL_PEERS) {
                             printf("DeepEP combine forwarder (NVL check) timeout, channel: %d, RDMA: %d, nvl: %d, src NVL: %d, dst RDMA: %d, tail: %d, waiting: %d, total: %d, sub: %d, large: %d, expected: %d\n",
                                    channel_id, rdma_rank, nvl_rank, lane_id, dst_rdma_rank, cached_nvl_channel_tail_idx, token_idx, num_tokens_to_combine, sub_warp_id, kNumWarpsPerForwarder, expected_head);
                             trap();
                         }
+                        __builtin_amdgcn_s_sleep(1);
                     }
 
                     // Combine current token
                     auto rdma_slot_idx = token_idx % num_max_rdma_chunked_recv_tokens;
                     void* shifted = send_buffer + rdma_slot_idx * num_bytes_per_rdma_token;
-                    auto get_addr_fn = [&](int src_nvl_rank, int slot_idx, int hidden_int4_idx) -> int4* { return reinterpret_cast<int4*>(nvl_channel_x.buffer(src_nvl_rank) + slot_idx * hidden_int4) + hidden_int4_idx; };
+                    auto recv_fn = [&](int src_nvl_rank, int slot_idx, int hidden_int4_idx) -> int4 { return ld_nc_global(nvl_channel_x.buffer(src_nvl_rank) + slot_idx * hidden_int4 + hidden_int4_idx); };
                     auto recv_tw_fn = [&](int src_nvl_rank, int slot_idx, int topk_idx) -> float { return ld_nc_global(nvl_channel_topk_weights.buffer(src_nvl_rank) + slot_idx * num_topk + topk_idx); };
-                    combine_token<NUM_MAX_NVL_PEERS, dtype_t, NUM_MAX_NVL_PEERS, true>(expected_head >= 0,
+                    combine_token<NUM_MAX_NVL_PEERS, dtype_t, NUM_MAX_NVL_PEERS, kWarpHyb>(expected_head >= 0,
                                                                                      expected_head, lane_id,
                                                                                      hidden_int4, num_topk,
                                                                                      reinterpret_cast<int4*>(shifted),
                                                                                      reinterpret_cast<float*>(reinterpret_cast<int8_t*>(shifted) + hidden_bytes + sizeof(SourceMeta)),
-                                                                                    num_max_nvl_chunked_recv_tokens_per_rdma, 
-                                                                                    get_addr_fn, recv_tw_fn);
+                                                                                     num_max_nvl_chunked_recv_tokens_per_rdma, recv_fn, recv_tw_fn);
 
                     // Update head
-                    if(lane_id < NUM_MAX_NVL_PEERS) {
-                        expected_head < 0 ? (forwarder_nvl_head[target_rank][lane_id] = -expected_head - 1)
-                                        : (forwarder_nvl_head[target_rank][lane_id] = expected_head + 1);
+                    if (lane_id < NUM_MAX_NVL_PEERS)
+                        expected_head < 0 ? (forwarder_nvl_head[warp_id][lane_id] = -expected_head - 1) : (forwarder_nvl_head[warp_id][lane_id] = expected_head + 1);
                 }
-                }
-                // sync_large_warp();
                 sync_large_warp(token_start_idx, 1);
-
                 // Issue RDMA send
-                if(sub_warp_id == kNumWarpsPerForwarder - 1) {
-                    if(dst_rdma_rank != rdma_rank) {
+                // TODO: Switch back to put_nbi_wave function
+              
+                if (sub_warp_id == kNumWarpsPerForwarder - 1 ) {
+                    if (dst_rdma_rank != rdma_rank) {
                         auto rdma_slot_idx = token_start_idx % num_max_rdma_chunked_recv_tokens;
-#if !defined(FORCE_DUSHMEM_API) && !defined(ROCM_DISABLE_CTX)
-                        shmem_ctx_schar_put_nbi_warp(ctx,
-#else
+                    #ifdef FORCE_DUSHMEM_API
                         shmemx_int8_put_nbi_warp(
-#endif
                             rdma_channel_data.recv_buffer(rdma_rank) +
                                     rdma_slot_idx * num_bytes_per_rdma_token,
                                 rdma_channel_data.send_buffer(dst_rdma_rank) +
                                     rdma_slot_idx * num_bytes_per_rdma_token,
                                 num_chunked_tokens * num_bytes_per_rdma_token,
                                 translate_dst_rdma_rank<kLowLatencyMode>(dst_rdma_rank, nvl_rank));
+                    #else
+                        #if !defined(ROCM_DISABLE_CTX)
+                        if constexpr (kUseWave){
+                            shmem_ctx_schar_put_nbi_warp(ctx,
+                                                 rdma_channel_data.recv_buffer(rdma_rank) + rdma_slot_idx * num_bytes_per_rdma_token,
+                                                 rdma_channel_data.send_buffer(dst_rdma_rank) + rdma_slot_idx * num_bytes_per_rdma_token,
+                                                 num_chunked_tokens * num_bytes_per_rdma_token,
+                                                 translate_dst_rdma_rank<kLowLatencyMode>(dst_rdma_rank, nvl_rank));
+                        }
+                        #else
+                        if constexpr (kUseWave){
+                            shmemx_int8_put_nbi_warp(rdma_channel_data.recv_buffer(rdma_rank) + rdma_slot_idx * num_bytes_per_rdma_token,
+                                                 rdma_channel_data.send_buffer(dst_rdma_rank) + rdma_slot_idx * num_bytes_per_rdma_token,
+                                                 num_chunked_tokens * num_bytes_per_rdma_token,
+                                                 translate_dst_rdma_rank<kLowLatencyMode>(dst_rdma_rank, nvl_rank));
+                        } else {
+                            if (lane_id == 0)
+                                shmemx_int8_put_nbi(rdma_channel_data.recv_buffer(rdma_rank) + rdma_slot_idx * num_bytes_per_rdma_token,
+                                                 rdma_channel_data.send_buffer(dst_rdma_rank) + rdma_slot_idx * num_bytes_per_rdma_token,
+                                                 num_chunked_tokens * num_bytes_per_rdma_token,
+                                                 translate_dst_rdma_rank<kLowLatencyMode>(dst_rdma_rank, nvl_rank));
+                        }
+                        #endif
+                    #endif
 #if !defined(FORCE_DUSHMEM_API) && !defined(ROCM_DISABLE_CTX)
                         shmem_ctx_quiet(ctx);
 #else
@@ -1765,7 +1775,7 @@ combine(int4 *combined_x, float *combined_topk_weights, const bool *is_combined_
 
                     // Write new RDMA tail
                     syncwarp();
-                    if(lane_id == 0) {
+                    if (lane_id == 0)
 #if !defined(FORCE_DUSHMEM_API) && !defined(ROCM_DISABLE_CTX)
                         shmem_ctx_ulong_atomic_add(ctx,
 #else
@@ -1775,157 +1785,90 @@ combine(int4 *combined_x, float *combined_topk_weights, const bool *is_combined_
                                            translate_dst_rdma_rank<kLowLatencyMode>(dst_rdma_rank, nvl_rank));
                 }
             }
-            }
 
             // Retired
             syncwarp();
-            if(lane_id == 0) {
-                forwarder_retired[target_rank] = true;
-            }
-        } else if (warp_role == WarpRole::kRDMACoordinator) {
-            // Coordinator
-            // Sync shared memory status
-            // sync_forwarder_smem();
-            __syncthreads();
-            constexpr int num_warps_per_rdma_rank = kNumForwarders / kNumRDMARanks;
-
-            int last_nvl_head[kNumRDMARanks] = {0};
-            int dst_nvl_rank                 = lane_id < NUM_MAX_NVL_PEERS ? lane_id : 0;
-
-            while(true) {
-                // Retired
-                if(__all_sync(kFullWarpMask, lane_id >= kNumForwarders or forwarder_retired[lane_id]))
-                    break;
-
-                {
-                    // Find minimum head for NVL ranks
-                    #pragma unroll
-                    for(int i = 0; i < kNumRDMARanks; ++i) {
-                        int min_head = std::numeric_limits<int>::max();
-                        #pragma unroll
-                        for(int j = 0; j < num_warps_per_rdma_rank; ++j)
-                            if(not forwarder_retired[i * num_warps_per_rdma_rank + j])
-                                min_head = min(min_head, forwarder_nvl_head[i * num_warps_per_rdma_rank + j][dst_nvl_rank]);
-
-                        if(min_head != std::numeric_limits<int>::max() and min_head > last_nvl_head[i] and lane_id < NUM_MAX_NVL_PEERS) {
-                            st_relaxed_sys_global(nvl_channel_head.buffer_by(dst_nvl_rank) + i, last_nvl_head[i] = min_head);
-                        }
-                    }
-                }
-
-                // Nanosleep and let other warps work
-                __builtin_amdgcn_s_sleep(NUM_WAIT_CYCLES_TIMES_64);
-            }
-        } else if(warp_role == WarpRole::kRDMAReceiver) {
+            if (lane_id == 0)
+                forwarder_retired[warp_id] = true;
+        } else if (warp_role == WarpRole::kRDMAReceiver) {
             // Receive from RDMA ranks and write to the output tensor
             // Clean shared memory and sync
             EP_DEVICE_ASSERT(kNumRDMARanks <= kWarpSize);
-            lane_id < kNumRDMARanks ? (rdma_receiver_rdma_head[target_rank][lane_id] = 0) : 0;
-            lane_id == 0 ? (rdma_receiver_retired[target_rank] = false) : 0;
-            // sync_rdma_receiver_smem();
-            __syncthreads();
+            lane_id < kNumRDMARanks ? (rdma_receiver_rdma_head[warp_id][lane_id] = 0) : 0;
+            lane_id == 0 ? (rdma_receiver_retired[warp_id] = false) : 0;
+            sync_rdma_receiver_smem();
 
             // The same tokens as the dispatch process
             int token_start_idx, token_end_idx;
             get_channel_task_range(num_combined_tokens, num_channels, channel_id, token_start_idx, token_end_idx);
 
-            // ==================== Token 级展开 x4 ====================
-            constexpr int kTokenUnroll = 4;
+            // Iterate over all tokens and combine
             int cached_channel_tail_idx = 0;
-
-            for (int64_t base = token_start_idx + target_rank;
-                base < token_end_idx;
-                base += (int64_t)kNumRDMAReceivers * kTokenUnroll) {
-
-                // ---- Phase 1: 批量预取所有 token 的 expected_head ----
-                int cached_expected_head[kTokenUnroll];
-                int max_expected_head = -1;
-
-                #pragma unroll
-                for (int u = 0; u < kTokenUnroll; ++u) {
-                    int64_t tidx = base + (int64_t)u * kNumRDMAReceivers;
-                    cached_expected_head[u] = -1;
-                    if (tidx < token_end_idx && lane_id < kNumRDMARanks) {
-                        int expected_head = ld_nc_global(combined_rdma_head + tidx * kNumRDMARanks + lane_id);
-                        cached_expected_head[u] = expected_head;
-                        if (expected_head > max_expected_head) max_expected_head = expected_head;
-                    }
+            for (int64_t token_idx = token_start_idx + warp_id; token_idx < token_end_idx; token_idx += kNumRDMAReceivers) {
+                // Read expected head
+                EP_STATIC_ASSERT(kNumRDMARanks <= kWarpSize, "Invalid number of RDMA peers");
+                int expected_head = -1;
+                if (lane_id < kNumRDMARanks) {
+                    expected_head = ld_nc_global(combined_rdma_head + token_idx * kNumRDMARanks + lane_id);
+                    (expected_head < 0) ? (rdma_receiver_rdma_head[warp_id][lane_id] = -expected_head - 1) : (rdma_receiver_rdma_head[warp_id][lane_id] = expected_head);
                 }
 
-                // ---- Phase 2: 一次等待，覆盖所有 token ----
-                if (max_expected_head >= 0) {
+                // Wait lanes to be ready
                 auto start_time = wall_clock64();
-                    while (cached_channel_tail_idx <= max_expected_head) {
-                        cached_channel_tail_idx = static_cast<int>(
-                            ld_acquire_sys_global(rdma_channel_tail.buffer(lane_id)));
-                        if (wall_clock64() - start_time > NUM_TIMEOUT_CYCLES) {
-                            printf("DeepEP combine RDMA receiver timeout (unroll x%d), "
-                                "ch: %d, rdma: %d, nvl: %d, lane: %d, "
-                                "tail: %d, wait: %d\n",
-                                kTokenUnroll, channel_id, rdma_rank, nvl_rank,
-                                lane_id, cached_channel_tail_idx, max_expected_head);
+                while (cached_channel_tail_idx <= expected_head) {
+                    cached_channel_tail_idx = static_cast<int>(ld_acquire_sys_global(rdma_channel_tail.buffer(lane_id)));
+
+                    // Timeout check
+                    long long int elapsed_time = wall_clock64() > start_time ? wall_clock64() - start_time : 0;
+                    if (elapsed_time > NUM_TIMEOUT_CYCLES) {
+                        printf("DeepEP combine RDMA receiver timeout, channel: %d, RDMA: %d, nvl: %d, src RDMA: %d, tail: %d, waiting: %ld, expect: %d\n",
+                               channel_id, rdma_rank, nvl_rank, lane_id, cached_channel_tail_idx, token_idx, expected_head);
                         trap();
                     }
-                    }
+                    __builtin_amdgcn_s_sleep(1);
                 }
                 syncwarp();
 
-                // ---- Phase 3: 批量处理所有就绪 token ----
-                #pragma unroll
-                for (int u = 0; u < kTokenUnroll; ++u) {
-                    int64_t tidx = base + (int64_t)u * kNumRDMAReceivers;
-                    if (tidx < token_end_idx) {
-                        int expected_head = cached_expected_head[u];
                 // Combine current token
-                        auto get_addr_fn = [&](int src_rdma_rank, int slot_idx, int hidden_int4_idx) -> int4* { return reinterpret_cast<int4*>(rdma_channel_data.recv_buffer(src_rdma_rank) + slot_idx * num_bytes_per_rdma_token) + hidden_int4_idx; };
+                auto recv_fn = [&](int src_rdma_rank, int slot_idx, int hidden_int4_idx) -> int4 { return ld_nc_global(reinterpret_cast<const int4*>(rdma_channel_data.recv_buffer(src_rdma_rank) + slot_idx * num_bytes_per_rdma_token) + hidden_int4_idx);};
                 auto recv_tw_fn = [&](int src_rdma_rank, int slot_idx, int topk_idx) -> float { return ld_nc_global(reinterpret_cast<const float*>(rdma_channel_data.recv_buffer(src_rdma_rank) + slot_idx * num_bytes_per_rdma_token + hidden_bytes + sizeof(SourceMeta)) + topk_idx);};
-
-                        combine_token<kNumRDMARanks, dtype_t, kNumTopkRDMARanks, false>(
-                            expected_head >= 0, expected_head, lane_id,
+                combine_token<kNumRDMARanks, dtype_t, kNumTopkRDMARanks, kWarpHyb>(expected_head >= 0,
+                                                                                            expected_head, lane_id,
                                                                                             hidden_int4, num_topk,
-                            combined_x + tidx * hidden_int4,
-                            combined_topk_weights + tidx * num_topk,
-                            num_max_rdma_chunked_recv_tokens,
-                            get_addr_fn, recv_tw_fn);
-
-                        if (lane_id < kNumRDMARanks) {
-                            rdma_receiver_rdma_head[target_rank][lane_id] =
-                                expected_head < 0 ? -expected_head - 1 : expected_head;
-                        }
-                    }
-                }
+                                                                                            combined_x + token_idx * hidden_int4,
+                                                                                            combined_topk_weights + token_idx * num_topk,
+                                                                                            num_max_rdma_chunked_recv_tokens, recv_fn, recv_tw_fn);
             }
 
             // Retired
             syncwarp();
-            if (lane_id == 0) {
-                rdma_receiver_retired[target_rank] = true;
-            }
-        } else if(warp_role == WarpRole::kNVLCoordinator) {
+            if (lane_id == 0)
+                rdma_receiver_retired[warp_id] = true;
+        } else {
+            auto lane_id = get_lane_id();
             // Coordinator
             // Sync shared memory status
-            // sync_rdma_receiver_smem();
-            __syncthreads();
+            is_rdma_receiver_sm ? sync_rdma_receiver_smem() : sync_forwarder_smem();
             const auto num_warps_per_rdma_rank = kNumForwarders / kNumRDMARanks;
 
             int last_rdma_head = 0;
             int last_nvl_head[kNumRDMARanks] = {0};
             int dst_rdma_rank = lane_id < kNumRDMARanks ? lane_id : 0;
             int dst_nvl_rank = lane_id < NUM_MAX_NVL_PEERS ? lane_id : 0;
-
-            while(true) {
+            EP_STATIC_ASSERT(kNumCombineForwarderWarps <= kWarpSize, "Invalid number of forwarder warps");
+            while (true) {
                 // Retired
-                if(__all_sync(kFullWarpMask, lane_id >= kNumRDMAReceivers or rdma_receiver_retired[lane_id]))
+                if (is_rdma_receiver_sm and __all_sync(kFullWarpMask, lane_id >= kNumRDMAReceivers or rdma_receiver_retired[lane_id]))
+                    break;
+                if (not is_rdma_receiver_sm and __all_sync(kFullWarpMask, lane_id >= kNumForwarders or forwarder_retired[lane_id]))
                     break;
 
                 // Find minimum head for RDMA ranks
-                {
+                if (is_rdma_receiver_sm) {
                     int min_head = std::numeric_limits<int>::max();
                     #pragma unroll
-                    for(int i = 0; i < kNumRDMAReceivers; ++i)
-                        if(not rdma_receiver_retired[i])
+                    for (int i = 0; i < kNumRDMAReceivers; ++ i) if (not rdma_receiver_retired[i])
                         min_head = min(min_head, rdma_receiver_rdma_head[i][dst_rdma_rank]);
-
                     if (min_head != std::numeric_limits<int>::max() and min_head >= last_rdma_head + num_max_rdma_chunked_send_tokens and lane_id < kNumRDMARanks) {
 #if !defined(FORCE_DUSHMEM_API) && !defined(ROCM_DISABLE_CTX)
                         shmem_ctx_ulong_atomic_add(ctx,
@@ -1937,6 +1880,17 @@ combine(int4 *combined_x, float *combined_topk_weights, const bool *is_combined_
 
                         last_rdma_head = min_head;
                     }
+                } else {
+                    // Find minimum head for NVL ranks
+                    #pragma unroll
+                    for (int i = 0; i < kNumRDMARanks; ++ i) {
+                        int min_head = std::numeric_limits<int>::max();
+                        #pragma unroll
+                        for (int j = 0; j < num_warps_per_rdma_rank; ++ j) if (not forwarder_retired[i * num_warps_per_rdma_rank + j])
+                            min_head = min(min_head, forwarder_nvl_head[i * num_warps_per_rdma_rank + j][dst_nvl_rank]);
+                        if (min_head != std::numeric_limits<int>::max() and min_head > last_nvl_head[i] and lane_id < NUM_MAX_NVL_PEERS)
+                            st_relaxed_sys_global(nvl_channel_head.buffer_by(dst_nvl_rank) + i, last_nvl_head[i] = min_head);
+                    }
                 }
 
                 // Nanosleep and let other warps work
@@ -1949,46 +1903,80 @@ combine(int4 *combined_x, float *combined_topk_weights, const bool *is_combined_
 #endif
 }
 
-void combine(hipDataType type, void *combined_x, float *combined_topk_weights,
-             const bool *is_combined_token_in_rank, const void *x, const float *topk_weights,
-             const void *bias_0, const void *bias_1, const int *combined_rdma_head,
-             const int *combined_nvl_head, const void *src_meta,
-             const int *rdma_channel_prefix_matrix, const int *rdma_rank_prefix_sum,
-             const int *gbl_channel_prefix_matrix, int num_tokens, int num_combined_tokens,
-             int hidden, int num_topk, void *rdma_buffer_ptr, int num_max_rdma_chunked_send_tokens,
-             int num_max_rdma_chunked_recv_tokens, void **buffer_ptrs,
-             int num_max_nvl_chunked_send_tokens, int num_max_nvl_chunked_recv_tokens, int rank,
-             int num_ranks, hipStream_t stream, int num_channels, bool low_latency_mode) {
-    constexpr int kNumCombineForwarderWarps = 8;
 
-#define COMBINE_LAUNCH_CASE(num_rdma_ranks)                                                        \
-    {                                                                                              \
-        auto combine_func =                                                                        \
-            low_latency_mode                                                                       \
-                ? combine<true, num_rdma_ranks, hip_bfloat16, kNumCombineForwarderWarps>           \
-                : combine<false, num_rdma_ranks, hip_bfloat16, kNumCombineForwarderWarps>;         \
-        LAUNCH_KERNEL_NON_COOPERATIVE(                                                             \
-            &cfg, combine_func, reinterpret_cast<int4 *>(combined_x), combined_topk_weights,       \
-            is_combined_token_in_rank, reinterpret_cast<const int4 *>(x), topk_weights,            \
-            reinterpret_cast<const int4 *>(bias_0), reinterpret_cast<const int4 *>(bias_1),        \
-            combined_rdma_head, combined_nvl_head, reinterpret_cast<const SourceMeta *>(src_meta), \
-            rdma_channel_prefix_matrix, rdma_rank_prefix_sum, gbl_channel_prefix_matrix,           \
-            num_tokens, num_combined_tokens, hidden, num_topk, rdma_buffer_ptr,                    \
-            num_max_rdma_chunked_send_tokens, num_max_rdma_chunked_recv_tokens, buffer_ptrs,       \
-            num_max_nvl_chunked_send_tokens, num_max_nvl_chunked_recv_tokens, rank, num_ranks);    \
-    }                                                                                              \
-    break
-
-    int num_rdma_ranks           = num_ranks / NUM_MAX_NVL_PEERS;
-    auto num_warps_per_forwarder = std::max(kNumCombineForwarderWarps / num_rdma_ranks, 1);
-    int num_forwarder_warps      = num_rdma_ranks * num_warps_per_forwarder;
-    EP_HOST_ASSERT(num_forwarder_warps >= NUM_MAX_NVL_PEERS);
-    EP_HOST_ASSERT(num_forwarder_warps > 0 and num_forwarder_warps % num_rdma_ranks == 0);
+void combine(hipDataType type,
+             void* combined_x,
+             float* combined_topk_weights,
+             const bool* is_combined_token_in_rank,
+             const void* x,
+             const float* topk_weights,
+             const void* bias_0,
+             const void* bias_1,
+             const int* combined_rdma_head,
+             const int* combined_nvl_head,
+             const void* src_meta,
+             const int* rdma_channel_prefix_matrix,
+             const int* rdma_rank_prefix_sum,
+             const int* gbl_channel_prefix_matrix,
+             int num_tokens,
+             int num_combined_tokens,
+             int hidden,
+             int num_topk,
+             void* rdma_buffer_ptr,
+             int num_max_rdma_chunked_send_tokens,
+             int num_max_rdma_chunked_recv_tokens,
+             void** buffer_ptrs,
+             int num_max_nvl_chunked_send_tokens,
+             int num_max_nvl_chunked_recv_tokens,
+             int rank,
+             int num_ranks,
+             hipStream_t stream,
+             int num_channels,
+             bool low_latency_mode)
+{
+    const int num_rdma_ranks = num_ranks / NUM_MAX_NVL_PEERS;
+
+    EP_HOST_ASSERT(num_rdma_ranks > 0);
     EP_HOST_ASSERT(num_max_nvl_chunked_recv_tokens % num_rdma_ranks == 0);
     EP_HOST_ASSERT(num_max_nvl_chunked_recv_tokens / num_rdma_ranks > std::max(num_max_rdma_chunked_send_tokens, num_max_nvl_chunked_send_tokens));
     EP_HOST_ASSERT(type == HIP_R_16BF);
 
-    SETUP_LAUNCH_CONFIG(num_channels * NUM_INTERNODE_DISPATCH_BLOCKS_PER_CHANNEL, (NUM_MAX_NVL_PEERS + 1) * kWarpSize, stream);
+    // One case per compile-time NR specialization.
+#define COMBINE_LAUNCH_CASE(NR) {                                                        \
+    /* Per-case compile-time constants */                                                \
+    constexpr int kNumCombineForwarderWarps = (NR < 9) ? 10 : 16;                        \
+    constexpr int kWarpsPerForwarder = (kNumCombineForwarderWarps/NR) > 0                \
+                                         ? (kNumCombineForwarderWarps/NR) : 1;           \
+    constexpr int kNumForwarders = NR * kWarpsPerForwarder;                              \
+    constexpr int kBlockThreads = (NR > 8)                                               \
+        ? ((NUM_MAX_NVL_PEERS + kNumForwarders) * kEmulatedWarpSize + kWarpSize)         \
+        : ((NUM_MAX_NVL_PEERS/2 + 1 + kNumForwarders) * kWarpSize);                      \
+                                                                                         \
+    SETUP_LAUNCH_CONFIG(num_channels * 2, kBlockThreads, stream);                        \
+                                                                                         \
+    using scalar_t =  hip_bfloat16;                                                      \
+    auto fn = low_latency_mode                                                           \
+        ? combine<true,  NR, scalar_t, kNumCombineForwarderWarps>                        \
+        : combine<false, NR, scalar_t, kNumCombineForwarderWarps>;                       \
+                                                                                         \
+    /* Launch (backend-specific) */                                                      \
+                                                                             \
+    LAUNCH_KERNEL_NON_COOPERATIVE(&cfg, fn,                                                            \
+            reinterpret_cast<int4*>(combined_x), combined_topk_weights, is_combined_token_in_rank, \
+            reinterpret_cast<const int4*>(x), topk_weights,  \
+            reinterpret_cast<const int4 *>(bias_0), reinterpret_cast<const int4 *>(bias_1), \
+            combined_rdma_head, combined_nvl_head,                                       \
+            reinterpret_cast<const SourceMeta*>(src_meta), rdma_channel_prefix_matrix,   \
+            rdma_rank_prefix_sum, gbl_channel_prefix_matrix,                             \
+            num_tokens, num_combined_tokens, hidden, num_topk,                           \
+            rdma_buffer_ptr, num_max_rdma_chunked_send_tokens,                           \
+            num_max_rdma_chunked_recv_tokens,                                            \
+            buffer_ptrs, num_max_nvl_chunked_send_tokens,                                \
+            num_max_nvl_chunked_recv_tokens,                                             \
+            rank, num_ranks);                                                            \
+} break
+
+    // Dispatch on the runtime num_rdma_ranks, but each case is compile-time specialized.
     SWITCH_RDMA_RANKS(COMBINE_LAUNCH_CASE);
 #undef COMBINE_LAUNCH_CASE
 }
@@ -1997,8 +1985,4 @@ void combine(hipDataType type, void *combined_x, float *combined_topk_weights,
 
 } // namespace deep_ep
 
-// #ifdef __clang__
-// #pragma clang diagnostic pop
-// #endif // __clang__
-
 #endif