[ILUVATAR_GPU] Add logic to apply patches to python files in install script && Fix the segment fault that occurred after linking with the NCCL library. (#1762)

Author: tianyuzhou668

backends/iluvatar_gpu/CMakeLists.txt

@@ -15,7 +15,6 @@ cmake_minimum_required(VERSION 3.10)
 set(PROJ_NAME "paddle-iluvatar-gpu")
 project(${PROJ_NAME} CXX C CUDA)
 
-set(PLUGIN_VERSION "0.0.1")
 set(TARGET_NAME ${PROJ_NAME})
 
 set(CMAKE_MODULE_PATH "${CMAKE_SOURCE_DIR}/cmake")
@@ -35,6 +34,7 @@ include(external/xxhash)
 include(external/zlib)
 include(external/protobuf)
 
+set(PLUGIN_VERSION ${PADDLE_VERSION})
 set(PROTO_FILE "${PADDLE_SOURCE_DIR}/paddle/phi/core/external_error.proto")
 get_filename_component(PROTO_WE "${PROTO_FILE}" NAME_WE)
 
@@ -253,7 +253,6 @@ target_link_libraries(
           protobuf
           external_error_proto
           cuinfer
-          # May cause a segment fault when the program exits
           nccl)
 
 include_directories(BEFORE ${PADDLE_SOURCE_DIR})

backends/iluvatar_gpu/build_paddle.sh

@@ -24,7 +24,6 @@ fi
 BUILD_TEST=${BUILD_TEST:-1}
 COREX_ARCH=${COREX_ARCH:-ivcore11}
 export CMAKE_CUDA_ARCHITECTURES=${COREX_ARCH}
-export PADDLE_VERSION=${PADDLE_VERSION:-3.0.0}
 
 CURRENT_DIR=$(pwd)
 PADDLE_SOURCE_DIR="${CURRENT_DIR}/../../Paddle"

backends/iluvatar_gpu/install_paddle.sh

@@ -20,6 +20,7 @@ PYTHON_PATH=$(which python3)
 PYTHON_DIST_PATH=${TARGET_DIR}/lib/python3/dist-packages
 
 PKG_DIR="build_pip"
+PKGCPU_NAME="paddlepaddle"
 PKG_NAME="paddle_iluvatar_gpu"
 
 if [[ ! -d ${PKG_DIR} ]]; then
@@ -43,6 +44,8 @@ if [[ "${TARGET_DIR}" != ""  ]]; then
   rm -rf ./tmp
   echo "Paddle installed in ${PYTHON_DIST_PATH}; please add it to your PYTHONPATH."
 else
+  ${PYTHON_PATH} -m pip uninstall ${PKGCPU_NAME} -y
+  ${PYTHON_PATH} -m pip install  --pre paddlepaddle -i https://www.paddlepaddle.org.cn/packages/nightly/cpu/
   ${PYTHON_PATH} -m pip uninstall ${PKG_NAME} -y
   ${PYTHON_PATH} -m pip install ${PKG_DIR}/${latest_pkg} || exit
 fi

backends/iluvatar_gpu/kernels/cuda_kernels/c_embedding_grad_kernel.cu

@@ -0,0 +1,148 @@
+// Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+#include "glog/logging.h"
+#include "paddle/common/flags.h"
+#include "paddle/phi/backends/gpu/gpu_context.h"
+#include "paddle/phi/backends/gpu/gpu_primitives.h"
+#include "paddle/phi/core/kernel_registry.h"
+#include "paddle/phi/kernels/c_embedding_kernel.h"
+#include "paddle/phi/kernels/funcs/eigen/common.h"
+#include "paddle/phi/kernels/funcs/embedding_grad.h"
+
+COMMON_DECLARE_int64(embedding_deterministic);
+
+namespace phi {
+
+static constexpr int kNumCUDAThreads = 512;
+static constexpr int kNumMaximumNumBlocks = 4096;
+
+static inline int NumBlocks(const int N) {
+  return std::min((N + kNumCUDAThreads - 1) / kNumCUDAThreads,
+                  kNumMaximumNumBlocks);
+}
+
+template <typename T, typename IndexT>
+__global__ void CEmbeddingGrad(T* table,
+                               const T* output,
+                               const IndexT* ids,
+                               const int rows,
+                               const int columns,
+                               const int64_t N,
+                               const int64_t start_idx,
+                               const int64_t end_idx,
+                               const int64_t limit) {
+  CUDA_KERNEL_LOOP(i, limit) {
+    size_t row = i / columns;
+    size_t col = i % columns;
+    auto id = ids[row];
+    if (id >= start_idx && id < end_idx) {
+      auto real_idx = id - start_idx;
+      phi::CudaAtomicAdd(&table[real_idx * columns + col], output[i]);
+    }
+  }
+}
+
+template <typename T, typename Context>
+void CEmbeddingGradKernel(const Context& dev_ctx,
+                          const DenseTensor& w,
+                          const DenseTensor& ids,
+                          const DenseTensor& out_grad,
+                          int64_t start_index,
+                          DenseTensor* w_grad) {
+  int N = w_grad->dims()[0];
+  int D = w_grad->dims()[1];
+  int K = ids.numel();
+
+  auto limit = K * D;
+  int blocks = NumBlocks(limit);
+  int threads = kNumCUDAThreads;
+
+  const T* d_output = out_grad.data<T>();
+  T* d_table = dev_ctx.template Alloc<T>(w_grad);
+
+  auto t = EigenVector<T>::Flatten(*w_grad);
+  t.device(*dev_ctx.eigen_device()) = t.constant(static_cast<T>(0));
+
+  const auto& index_type = ids.dtype();
+  if (FLAGS_embedding_deterministic == 1) {
+    if (index_type == phi::DataType::INT32) {
+      phi::funcs::LaunchEmbeddingGradDeterministicKernel<T, int32_t>(
+          dev_ctx,
+          ids.data<int32_t>(),
+          d_output,
+          d_table,
+          N,
+          D,
+          K,
+          start_index);
+      return;
+    } else if (index_type == phi::DataType::INT64) {
+      phi::funcs::LaunchEmbeddingGradDeterministicKernel<T, int64_t>(
+          dev_ctx,
+          ids.data<int64_t>(),
+          d_output,
+          d_table,
+          N,
+          D,
+          K,
+          start_index);
+      return;
+    }
+  } else {
+    if (FLAGS_embedding_deterministic > 1) {
+      VLOG(2) << "Run grad kernel of embedding with single thread.";
+      blocks = 1;
+    }
+    const int64_t end_idx = start_index + N;
+    if (index_type == phi::DataType::INT32) {
+      CEmbeddingGrad<T, int32_t>
+          <<<blocks, threads, 0, dev_ctx.stream()>>>(d_table,
+                                                     d_output,
+                                                     ids.data<int32_t>(),
+                                                     K,
+                                                     D,
+                                                     N,
+                                                     start_index,
+                                                     end_idx,
+                                                     limit);
+      return;
+    } else if (index_type == phi::DataType::INT64) {
+      CEmbeddingGrad<T, int64_t>
+          <<<blocks, threads, 0, dev_ctx.stream()>>>(d_table,
+                                                     d_output,
+                                                     ids.data<int64_t>(),
+                                                     K,
+                                                     D,
+                                                     N,
+                                                     start_index,
+                                                     end_idx,
+                                                     limit);
+      return;
+    }
+  }
+  PADDLE_THROW(common::errors::InvalidArgument(
+      "The data type of Input(Ids) must be int32 or int64."));
+}
+
+}  // namespace phi
+
+PD_REGISTER_PLUGIN_KERNEL(c_embedding_grad,
+                          iluvatar_gpu,
+                          ALL_LAYOUT,
+                          phi::CEmbeddingGradKernel,
+                          float,
+                          phi::dtype::bfloat16,
+                          phi::dtype::float16,
+                          phi::dtype::complex<float>) {}

backends/iluvatar_gpu/kernels/cuda_kernels/c_embedding_grad_kernel_register.cc

@@ -0,0 +1,122 @@
+// Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved.
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+#include "paddle/phi/backends/gpu/gpu_context.h"
+#include "paddle/phi/core/kernel_registry.h"
+#include "paddle/phi/kernels/c_embedding_kernel.h"
+
+namespace phi {
+
+static constexpr int kNumCUDAThreads = 512;
+static constexpr int kNumMaximumNumBlocks = 4096;
+
+static inline int NumBlocks(const int N) {
+  return std::min((N + kNumCUDAThreads - 1) / kNumCUDAThreads,
+                  kNumMaximumNumBlocks);
+}
+
+template <typename T, typename IndexT>
+__global__ void CEmbedding(T* out,
+                           const T* table,
+                           const IndexT* ids,
+                           const int rows,
+                           const int columns,
+                           const int64_t N,
+                           const int64_t start_idx,
+                           const int64_t end_idx,
+                           const int64_t limit,
+                           const int64_t vocab_size) {
+  CUDA_KERNEL_LOOP(i, limit) {
+    size_t row = i / columns;
+    size_t col = i % columns;
+    auto id = ids[row];
+
+    PADDLE_ENFORCE(
+        id >= 0 && (vocab_size < 0 || id < vocab_size),
+        "The index is out of bounds, "
+        "please check whether the dimensions of index and "
+        "input meet the requirements. It should "
+        "be less than [%d] and greater than or equal to 0, but received [%d]",
+        vocab_size,
+        id);
+    if (id >= start_idx && id < end_idx) {
+      auto real_idx = id - start_idx;
+      out[i] = table[real_idx * columns + col];
+    } else {
+      out[i] = static_cast<T>(0);
+    }
+  }
+}
+
+template <typename T, typename Context>
+void CEmbeddingKernel(const Context& dev_ctx,
+                      const DenseTensor& w,
+                      const DenseTensor& ids,
+                      int64_t start_index,
+                      int64_t vocab_size,
+                      DenseTensor* out) {
+  size_t N = w.dims()[0];
+  size_t D = w.dims()[1];
+  size_t K = ids.numel();
+
+  const int64_t end_idx = start_index + N;
+
+  auto* table = w.data<T>();
+  auto* output = dev_ctx.template Alloc<T>(out);
+
+  auto limit = K * D;
+  int blocks = NumBlocks(limit);
+  int threads = kNumCUDAThreads;
+
+  const auto& index_type = ids.dtype();
+  if (index_type == phi::DataType::INT32) {
+    CEmbedding<T, int32_t>
+        <<<blocks, threads, 0, dev_ctx.stream()>>>(output,
+                                                   table,
+                                                   ids.data<int32_t>(),
+                                                   K,
+                                                   D,
+                                                   N,
+                                                   start_index,
+                                                   end_idx,
+                                                   limit,
+                                                   vocab_size);
+
+  } else if (index_type == phi::DataType::INT64) {
+    CEmbedding<T, int64_t>
+        <<<blocks, threads, 0, dev_ctx.stream()>>>(output,
+                                                   table,
+                                                   ids.data<int64_t>(),
+                                                   K,
+                                                   D,
+                                                   N,
+                                                   start_index,
+                                                   end_idx,
+                                                   limit,
+                                                   vocab_size);
+  } else {
+    PADDLE_THROW(common::errors::Unavailable(
+        "GPU c_embedding ids only support int32 or int64."));
+  }
+}
+}  // namespace phi
+
+PD_REGISTER_PLUGIN_KERNEL(c_embedding,
+                          iluvatar_gpu,
+                          ALL_LAYOUT,
+                          phi::CEmbeddingKernel,
+                          float,
+                          phi::dtype::bfloat16,
+                          phi::dtype::float16,
+                          phi::dtype::complex<float>) {}

backends/iluvatar_gpu/kernels/cuda_kernels/c_embedding_kernel.cu

@@ -152,6 +152,7 @@ void FlashAttnUnpaddedBaseKernel(
   flashAttnInfo.softmax_scale = std::sqrt(1.f / head_size);
   flashAttnInfo.dropout_prob = is_test ? 0.0f : dropout;
   flashAttnInfo.is_causal = causal;
+  flashAttnInfo.causal_mode = 1;
   // flashAttnInfo.is_alibi              = use_alibi;
   // flashAttnInfo.alibi_mode            = alibi_mode;
   flashAttnInfo.return_softmax_lse = true;