[Feature][Kernel]FusedMoE LoRA (#21229)

Signed-off-by: wuchen <[email protected]>
Signed-off-by: banjuede <[email protected]>
Signed-off-by: Chen Wu <[email protected]>
Signed-off-by: Danielle Robinson <[email protected]>
Signed-off-by: Jee Jee Li <[email protected]>
Signed-off-by: bk-201 <[email protected]>
Co-authored-by: wuchen <[email protected]>
Co-authored-by: Nathan Van Gheem <[email protected]>
Co-authored-by: banjuede <[email protected]>
Co-authored-by: Danielle Robinson <[email protected]>
Co-authored-by: Jee Jee Li <[email protected]>
Co-authored-by: bk-201 <[email protected]>

Author: 7 people

.buildkite/test-pipeline.yaml

@@ -384,7 +384,12 @@ steps:
       --num-shards=$$BUILDKITE_PARALLEL_JOB_COUNT \
       --ignore=lora/test_chatglm3_tp.py \
       --ignore=lora/test_llama_tp.py \
-      --ignore=lora/test_llm_with_multi_loras.py
+      --ignore=lora/test_llm_with_multi_loras.py \
+      --ignore=lora/test_olmoe_tp.py \
+      --ignore=lora/test_deepseekv2_tp.py \
+      --ignore=lora/test_gptoss.py \
+      --ignore=lora/test_qwen3moe_tp.py
+      
   parallelism: 4
 
 - label: PyTorch Compilation Unit Tests # 15min
@@ -1065,6 +1070,7 @@ steps:
     - pytest -v -s -x lora/test_chatglm3_tp.py
     - pytest -v -s -x lora/test_llama_tp.py
     - pytest -v -s -x lora/test_llm_with_multi_loras.py
+    - pytest -v -s -x lora/test_olmoe_tp.py
 
 
 - label: Weight Loading Multiple GPU Test  # 33min

CMakeLists.txt

@@ -883,6 +883,7 @@ target_compile_definitions(_C PRIVATE CUTLASS_ENABLE_DIRECT_CUDA_DRIVER_CALL=1)
 set(VLLM_MOE_EXT_SRC
   "csrc/moe/torch_bindings.cpp"
   "csrc/moe/moe_align_sum_kernels.cu"
+  "csrc/moe/moe_lora_align_sum_kernels.cu"
   "csrc/moe/topk_softmax_kernels.cu")
 
 if(VLLM_GPU_LANG STREQUAL "CUDA")

csrc/moe/moe_lora_align_sum_kernels.cu

@@ -0,0 +1,173 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <time.h>
+#include <torch/all.h>
+#include <ATen/cuda/CUDAContext.h>
+#include <c10/cuda/CUDAGuard.h>
+
+#include <ATen/ATen.h>
+#include <ATen/cuda/Atomic.cuh>
+
+#include "../cuda_compat.h"
+#include "../dispatch_utils.h"
+#include "core/math.hpp"
+
+namespace {
+
+__device__ __forceinline__ int32_t index(int32_t total_col, int32_t row,
+                                         int32_t col) {
+  return row * total_col + col;
+}
+
+}  // namespace
+
+// TODO: Refactor common parts with moe_align_sum_kernels
+template <typename scalar_t, typename token_cnts_t>
+__global__ void moe_lora_align_sum_kernel(
+    scalar_t* __restrict__ topk_ids, int32_t* token_lora_mapping,
+    int64_t block_size, int num_experts, int max_loras, size_t numel,
+    int max_num_tokens_padded, int max_num_m_blocks,
+    int32_t* __restrict__ sorted_token_ids, int32_t* __restrict__ expert_ids,
+    int topk_num, int32_t* total_tokens_post_pad) {
+  const size_t tokens_per_thread = div_ceil(numel, blockDim.x);
+  const size_t start_idx = threadIdx.x * tokens_per_thread;
+
+  int lora_id = blockIdx.x;
+  extern __shared__ int32_t shared_mem[];
+  int32_t* cumsum = shared_mem;
+  token_cnts_t* tokens_cnts = (token_cnts_t*)(shared_mem + num_experts + 1);
+
+  // Initialize sorted_token_ids with numel
+  for (size_t it = threadIdx.x; it < max_num_tokens_padded; it += blockDim.x) {
+    sorted_token_ids[lora_id * max_num_tokens_padded + it] = numel;
+  }
+
+  // Initialize expert_ids with -1
+  for (size_t it = threadIdx.x; it < max_num_m_blocks; it += blockDim.x) {
+    expert_ids[lora_id * max_num_m_blocks + it] = -1;
+  }
+
+  // Initialize total_tokens_post_pad with 0
+  if (threadIdx.x == 0) {
+    total_tokens_post_pad[lora_id] = 0;
+  }
+
+  for (int i = 0; i < num_experts; ++i) {
+    tokens_cnts[index(num_experts, threadIdx.x + 1, i)] = 0;
+  }
+
+  for (int i = start_idx; i < numel && i < start_idx + tokens_per_thread; ++i) {
+    int mask = token_lora_mapping[i / topk_num] == lora_id;
+    int idx = index(num_experts, threadIdx.x + 1, topk_ids[i]);
+    tokens_cnts[idx] += mask;
+  }
+
+  __syncthreads();
+
+  // For each expert we accumulate the token counts from the different threads.
+  if (threadIdx.x < num_experts) {
+    tokens_cnts[index(num_experts, 0, threadIdx.x)] = 0;
+    for (int i = 1; i <= blockDim.x; ++i) {
+      tokens_cnts[index(num_experts, i, threadIdx.x)] +=
+          tokens_cnts[index(num_experts, i - 1, threadIdx.x)];
+    }
+  }
+
+  __syncthreads();
+
+  // We accumulate the token counts of all experts in thread 0.
+  if (threadIdx.x == 0) {
+    cumsum[0] = 0;
+    for (int i = 1; i <= num_experts; ++i) {
+      cumsum[i] = cumsum[i - 1] +
+                  div_ceil(tokens_cnts[index(num_experts, blockDim.x, i - 1)],
+                           block_size) *
+                      block_size;
+    }
+    total_tokens_post_pad[lora_id] = static_cast<int32_t>(cumsum[num_experts]);
+  }
+
+  __syncthreads();
+
+  /**
+   * For each expert, each thread processes the tokens of the corresponding
+   * blocks and stores the corresponding expert_id for each block.
+   */
+  if (threadIdx.x < num_experts) {
+    for (int i = cumsum[threadIdx.x]; i < cumsum[threadIdx.x + 1];
+         i += block_size) {
+      expert_ids[index(max_num_m_blocks, lora_id, i / block_size)] =
+          threadIdx.x;
+    }
+  }
+
+  for (int i = start_idx; i < numel && i < start_idx + tokens_per_thread; ++i) {
+    int32_t expert_id = topk_ids[i];
+    /** The cumsum[expert_id] stores the starting index of the tokens that the
+     * expert with expert_id needs to process, and
+     * tokens_cnts[threadIdx.x][expert_id] stores the indices of the tokens
+     * processed by the expert with expert_id within the current thread's token
+     * shard.
+     */
+    int32_t rank_post_pad =
+        tokens_cnts[index(num_experts, threadIdx.x, expert_id)] +
+        cumsum[expert_id];
+
+    int mask = (int)token_lora_mapping[i / topk_num] == lora_id;
+    atomicAdd(
+        &sorted_token_ids[index(max_num_tokens_padded, lora_id, rank_post_pad)],
+        (i - numel) * mask);
+    tokens_cnts[index(num_experts, threadIdx.x, expert_id)] += mask;
+  }
+}
+
+void moe_lora_align_block_size(torch::Tensor topk_ids,
+                               torch::Tensor token_lora_mapping,
+                               int64_t num_experts, int64_t block_size,
+                               int64_t max_loras,
+                               torch::Tensor sorted_token_ids,
+                               torch::Tensor expert_ids,
+                               torch::Tensor num_tokens_post_pad) {
+  const int topk_num = topk_ids.size(1);
+
+  int max_num_tokens_padded = topk_ids.numel() + num_experts * (block_size - 1);
+
+  TORCH_CHECK(block_size > 0, "block_size should be greater than 0. ");
+  max_num_tokens_padded = round_to_next_multiple_of(
+      max_num_tokens_padded, static_cast<int>(block_size));
+  int max_num_m_blocks = div_ceil(max_num_tokens_padded, block_size);
+
+  int device_max_shared_mem;
+  auto dev = topk_ids.get_device();
+  cudaDeviceGetAttribute(&device_max_shared_mem,
+                         cudaDevAttrMaxSharedMemoryPerBlockOptin, dev);
+  const cudaStream_t stream = at::cuda::getCurrentCUDAStream();
+
+  const int32_t num_thread = max((int32_t)num_experts, 128);  // WARP_SIZE,
+  TORCH_CHECK(num_thread <= 1024,
+              "num_thread must be less than 1024, "
+              "and fallback is not implemented yet.");
+  const int32_t shared_mem = (num_thread + 1) * num_experts * sizeof(int32_t) +
+                             (num_experts + 1) * sizeof(int32_t);
+
+  if (shared_mem > device_max_shared_mem) {
+    TORCH_CHECK(false,
+                "Shared memory usage exceeds device limit, and global memory "
+                "fallback is not implemented yet.");
+  }
+
+  VLLM_DISPATCH_INTEGRAL_TYPES(
+      topk_ids.scalar_type(), "moe_lora_align_sum_kernel", [&] {
+        dim3 blockDim(num_thread);
+        auto kernel = moe_lora_align_sum_kernel<scalar_t, int32_t>;
+        AT_CUDA_CHECK(VLLM_DevFuncAttribute_SET_MaxDynamicSharedMemorySize(
+            (void*)kernel, shared_mem));
+        kernel<<<max_loras, blockDim, shared_mem, stream>>>(
+            topk_ids.data_ptr<scalar_t>(),
+            token_lora_mapping.data_ptr<int32_t>(), block_size, num_experts,
+            max_loras, topk_ids.numel(), max_num_tokens_padded,
+            max_num_m_blocks, sorted_token_ids.data_ptr<int32_t>(),
+            expert_ids.data_ptr<int32_t>(), topk_num,
+            num_tokens_post_pad.data_ptr<int32_t>());
+      });
+}

csrc/moe/moe_ops.h

@@ -20,6 +20,13 @@ void batched_moe_align_block_size(int64_t max_tokens_per_batch,
                                   torch::Tensor expert_ids,
                                   torch::Tensor num_tokens_post_pad);
 
+void moe_lora_align_block_size(torch::Tensor topk_ids,
+                               torch::Tensor token_lora_mapping,
+                               int64_t num_experts, int64_t block_size,
+                               int64_t max_loras,
+                               torch::Tensor sorted_token_ids,
+                               torch::Tensor expert_ids,
+                               torch::Tensor num_tokens_post_pad);
 #ifndef USE_ROCM
 torch::Tensor moe_wna16_gemm(torch::Tensor input, torch::Tensor output,
                              torch::Tensor b_qweight, torch::Tensor b_scales,

csrc/moe/torch_bindings.cpp

@@ -33,6 +33,18 @@ TORCH_LIBRARY_EXPAND(TORCH_EXTENSION_NAME, m) {
   m.impl("batched_moe_align_block_size", torch::kCUDA,
          &batched_moe_align_block_size);
 
+  // Aligning the number of tokens to be processed by each expert such
+  // that it is divisible by the block size.
+  m.def(
+      "moe_lora_align_block_size(Tensor topk_ids,"
+      "                     Tensor token_lora_mapping,"
+      "                     int num_experts,"
+      "                     int block_size, int max_loras, "
+      "                     Tensor !sorted_token_ids,"
+      "                     Tensor !experts_ids,"
+      "                     Tensor !num_tokens_post_pad) -> () ");
+  m.impl("moe_lora_align_block_size", torch::kCUDA, &moe_lora_align_block_size);
+
 #ifndef USE_ROCM
   m.def(
       "moe_wna16_gemm(Tensor input, Tensor! output, Tensor b_qweight, "

tests/lora/conftest.py

@@ -230,6 +230,26 @@ def tinyllama_lora_files():
     return snapshot_download(repo_id="jashing/tinyllama-colorist-lora")
 
 
+@pytest.fixture(scope="session")
+def deepseekv2_lora_files():
+    return snapshot_download(repo_id="wuchen01/DeepSeek-V2-Lite-Chat-All-LoRA")
+
+
+@pytest.fixture(scope="session")
+def gptoss20b_lora_files():
+    return snapshot_download(repo_id="LevinZheng/gpt-oss-20b-lora-adapter")
+
+
+@pytest.fixture(scope="session")
+def qwen3moe_lora_files():
+    return snapshot_download(repo_id="jeeejeee/qwen3-moe-text2sql-spider")
+
+
+@pytest.fixture(scope="session")
+def olmoe_lora_files():
+    return snapshot_download(repo_id="jeeejeee/olmoe-instruct-text2sql-spider")
+
+
 @pytest.fixture
 def reset_default_device():
     """

tests/lora/test_deepseekv2_tp.py

@@ -0,0 +1,97 @@
+# SPDX-License-Identifier: Apache-2.0
+# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
+
+import vllm
+from vllm.lora.request import LoRARequest
+
+from ..utils import multi_gpu_test
+
+MODEL_PATH = "deepseek-ai/DeepSeek-V2-Lite-Chat"
+
+PROMPT_TEMPLATE = "<｜begin▁of▁sentence｜>You are a helpful assistant.\n\nUser: {context}\n\nAssistant:"  # noqa: E501
+
+
+def generate_and_test(llm: vllm.LLM, lora_path: str, lora_id: int):
+    prompts = [
+        PROMPT_TEMPLATE.format(context="Who are you?"),
+    ]
+    sampling_params = vllm.SamplingParams(temperature=0, max_tokens=64)
+    outputs = llm.generate(
+        prompts,
+        sampling_params,
+        lora_request=LoRARequest(str(lora_id), lora_id, lora_path) if lora_id else None,
+    )
+    # Print the outputs.
+    generated_texts: list[str] = []
+    for output in outputs:
+        prompt = output.prompt
+        generated_text = output.outputs[0].text.strip()
+        generated_texts.append(generated_text)
+        print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")
+    # return generated_texts
+    expected_lora_output = [
+        "I am \u5f20\u5b50\u8c6a, an AI assistant developed by \u9648\u58eb\u680b.",  # noqa: E501
+    ]
+    for i in range(len(expected_lora_output)):
+        assert generated_texts[i].startswith(expected_lora_output[i])
+
+
+def test_deepseekv2_lora(deepseekv2_lora_files):
+    # We enable enforce_eager=True here to reduce VRAM usage for lora-test CI,
+    # Otherwise, the lora-test will fail due to CUDA OOM.
+    llm = vllm.LLM(
+        MODEL_PATH,
+        max_model_len=1024,
+        enable_lora=True,
+        max_loras=4,
+        enforce_eager=True,
+        trust_remote_code=True,
+        enable_chunked_prefill=True,
+    )
+    generate_and_test(llm, deepseekv2_lora_files, 1)
+
+
+def test_deepseekv2(deepseekv2_lora_files):
+    # We enable enforce_eager=True here to reduce VRAM usage for lora-test CI,
+    # Otherwise, the lora-test will fail due to CUDA OOM.
+    llm = vllm.LLM(
+        MODEL_PATH,
+        max_model_len=1024,
+        enable_lora=True,
+        max_loras=4,
+        enforce_eager=True,
+        trust_remote_code=True,
+    )
+    generate_and_test(llm, deepseekv2_lora_files, 1)
+
+
+@multi_gpu_test(num_gpus=2)
+def test_deepseekv2_tp2(deepseekv2_lora_files):
+    # We enable enforce_eager=True here to reduce VRAM usage for lora-test CI,
+    # Otherwise, the lora-test will fail due to CUDA OOM.
+    llm = vllm.LLM(
+        MODEL_PATH,
+        max_model_len=1024,
+        enable_lora=True,
+        max_loras=4,
+        enforce_eager=True,
+        trust_remote_code=True,
+        tensor_parallel_size=2,
+    )
+    generate_and_test(llm, deepseekv2_lora_files, 2)
+
+
+@multi_gpu_test(num_gpus=4)
+def test_deepseekv2_tp4(deepseekv2_lora_files):
+    # We enable enforce_eager=True here to reduce VRAM usage for lora-test CI,
+    # Otherwise, the lora-test will fail due to CUDA OOM.
+    llm = vllm.LLM(
+        MODEL_PATH,
+        max_model_len=1024,
+        enable_lora=True,
+        max_loras=4,
+        enforce_eager=True,
+        trust_remote_code=True,
+        tensor_parallel_size=4,
+    )
+    generate_and_test(llm, deepseekv2_lora_files, 2)