Paddle Debug版本编译过程中，报错HTTP page not retrieved

[wangguan1995]

bug描述 Describe the Bug

commit id 0a85d41775961fc85b8d3aa8a843194d80dc9f00
gcc (GCC) 12.1.0
g++ (GCC) 12.1.0
GNU ld (GNU Binutils) 2.33.1

export https_proxy=
export http_proxy=
cmake .. \
    -GNinja \
    -DCMAKE_BUILD_TYPE=Debug \
    -DPY_VERSION=3.10 \
    -DWITH_GPU=ON \
    -DWITH_AVX=OFF \
    -DWITH_DISTRIBUTE=ON \
    -DCMAKE_C_COMPILER=/usr/local/gcc-12.1/bin/gcc \
    -DCMAKE_CXX_COMPILER=/usr/local/gcc-12.1/bin/g++ \
 2>&1 | tee cmake_0421.log

其他补充信息 Additional Supplementary Information

No response

[GuoxiaWang]

这个没找到正常，没找到的话，就会重新编译。

[wangguan1995]

具体编译时报错：

<pre style="white-space: pre-wrap;">
FAILED: paddle/phi/CMakeFiles/phi_core.dir/kernels/fusion/gpu/fused_dropout_add_grad_kernel.cu.o 
/usr/local/bin/ccache /usr/local/cuda/bin/nvcc -forward-unknown-to-host-compiler -DBRPC_WITH_GLOG -DCUDA_TOOLKIT_ROOT_DIR=\"/usr/local/cuda\" -DCUDA_VERSION_MAJOR=\"12\" -DCUDA_VERSION_MINOR=\"3\" -DCUDNN_MAJOR_VERSION=\"9\" -DEIGEN_USE_GPU -DGLOG_NO_ABBREVIATED_SEVERITIES -DLAPACK_FOUND -DPADDLE_DISABLE_PROFILER -DPADDLE_DLL_EXPORT -DPADDLE_ON_INFERENCE -DPADDLE_USE_PTHREAD_BARRIER -DPADDLE_USE_PTHREAD_SPINLOCK -DPADDLE_VERSION=0.0.0 -DPADDLE_VERSION_INTEGER=0 -DPADDLE_WITH_AVX512F -DPADDLE_WITH_CCCL -DPADDLE_WITH_CRYPTO -DPADDLE_WITH_CUDA -DPADDLE_WITH_CUPTI -DPADDLE_WITH_CUTLASS -DPADDLE_WITH_DGC -DPADDLE_WITH_DISTRIBUTE -DPADDLE_WITH_DNNL -DPADDLE_WITH_FLASHATTN -DPADDLE_WITH_GLOO -DPADDLE_WITH_MEMORY_EFFICIENT_ATTENTION -DPADDLE_WITH_MKLML -DPADDLE_WITH_NCCL -DPADDLE_WITH_POCKETFFT -DPADDLE_WITH_PSCORE -DPADDLE_WITH_RPC -DPADDLE_WITH_SSE3 -DPADDLE_WITH_XBYAK -DPHI_INNER -DPHI_SHARED -DSPCONV_WITH_CUTLASS=0 -DSTATIC_IR -DTRT_PLUGIN_FP16_AVAILABLE -DXBYAK64 -DXBYAK_NO_OP_NAMES -DYAML_CPP_STATIC_DEFINE -Dphi_core_EXPORTS -I../third_party/cccl/thrust -I../third_party/cccl/libcudacxx/include -I../third_party/cccl/cub -I. -I../paddle/fluid/framework/io -Ithird_party/install/zlib/include -Ithird_party/install -Ithird_party/install/gflags/include -Ithird_party/install/glog/include -I../third_party/cutlass -I../third_party/cutlass/include -I../third_party/cutlass/tools/util/include -I../third_party/eigen3 -I../third_party/threadpool -I../third_party/dlpack/include -Ithird_party/install/xxhash/include -Ithird_party/install/warpctc/include -Ithird_party/install/warprnnt/include -Ithird_party/install/utf8proc/include -Ithird_party/install/mklml/include -Ithird_party/install/onednn/include -Ithird_party/install/protobuf/include -I../third_party/nlohmann_json/include -Ithird_party/install/yaml-cpp/include -I/usr/include/python3.10 -I/usr/local/lib/python3.10/dist-packages/numpy/core/include -Ithird_party/pybind/src/extern_pybind/include -Ithird_party/install/gtest/include -I../third_party/cccl -Ithird_party/install/gloo/include -Ithird_party/install/snappy/include -Ithird_party/install/leveldb/include -Ithird_party/install/brpc/include -Ithird_party/install/libmct/include -Ithird_party/install/rocksdb/include -Ithird_party/install/jemalloc/include -Ithird_party/install/afs_api/include -Ithird_party/install/xbyak/include -Ithird_party/install/xbyak/include/xbyak -Ithird_party/install/dgc/include -Ithird_party/install/cryptopp/include -Ithird_party/pocketfft/src -Ithird_party/install/flashattn/include -I/usr/local/cuda/targets/x86_64-linux/include -I/usr/local/cuda/include -I../ -I../paddle/fluid/framework/io --cudart shared -D_MWAITXINTRIN_H_INCLUDED -D__STRICT_ANSI__ -Wno-deprecated-gpu-targets  -gencode arch=compute_80,code=sm_80 -w --expt-relaxed-constexpr --expt-extended-lambda  -Xcompiler="-Wall" -Xcompiler="-Wextra" -Xcompiler="-Werror" -Xcompiler="-fPIC" -Xcompiler="-fno-omit-frame-pointer" -Xcompiler="-Wno-unused-parameter" -Xcompiler="-Wno-unused-function" -Xcompiler="-Wno-error=literal-suffix" -Xcompiler="-Wno-error=unused-local-typedefs" -Xcompiler="-Wno-error=unused-function" -Xcompiler="-Wno-error=array-bounds" -Xcompiler="-msse3" -g -Xcompiler=-fPIC -std=c++17 -MD -MT paddle/phi/CMakeFiles/phi_core.dir/kernels/fusion/gpu/fused_dropout_add_kernel.cu.o -MF paddle/phi/CMakeFiles/phi_core.dir/kernels/fusion/gpu/fused_dropout_add_kernel.cu.o.d -x cu -c ../paddle/phi/kernels/fusion/gpu/fused_dropout_add_kernel.cu -o paddle/phi/CMakeFiles/phi_core.dir/kernels/fusion/gpu/fused_dropout_add_kernel.cu.o
../paddle/phi/kernels/funcs/dropout_impl.cu.h(363): error: identifier "seed_tensor_ptr" is undefined
               seed_tensor_ptr == nullptr
               ^

1 error detected in the compilation of "../paddle/phi/kernels/fusion/gpu/fused_dropout_add_kernel.cu".

错误原因分析
未定义标识符：
seed_tensor_ptr 在代码中被使用，但编译器找不到其定义。可能原因包括：
拼写错误：变量名拼写错误（如 seed_tensor_ptr vs. seed_tensor_ptr_）。
作用域问题：变量未在当前作用域声明，或声明在条件编译块中未启用。
头文件缺失：定义 seed_tensor_ptr 的头文件未包含。
代码上下文：
错误出现在 dropout_impl.cu.h 的第 363 行，检查该行附近的代码逻辑：

[wangguan1995]

错误原因分析
未定义的引用：
cudaStreamEndCapture@@libcudart.so.12 是 CUDA 库中的函数，链接器（ld）找不到其定义。
这通常是因为链接命令中未包含 CUDA 运行时库（-lcudart）。
DSO missing from command line：
动态共享对象（DSO）缺失，表明链接器在命令中未找到所需的库文件。

FAILED: paddle/phi/tools/print_phi_kernels 
: && /usr/local/gcc-12.1/bin/g++ -Wno-error=uninitialized -Wno-error=deprecated-declarations -Wno-deprecated-declarations -Wno-error=maybe-uninitialized -std=c++17 -m64 -fPIC -fno-omit-frame-pointer -pipe -ffunction-sections -fdata-sections -Werror -Wall -Wextra -Wno-unused-parameter -Wno-unused-function -Wno-error=array-bounds -Wno-error=ignored-attributes -Wno-error=int-in-bool-context -Wimplicit-fallthrough=0 -Wno-error=deprecated-copy -Wno-ignored-qualifiers -Wno-ignored-attributes -Wno-parentheses -fopenmp -msse3 -g  paddle/phi/tools/CMakeFiles/print_phi_kernels.dir/print_phi_kernels.cc.o -o paddle/phi/tools/print_phi_kernels -L/workspace/Paddle/Paddle/build/third_party/install/onednn/lib   -L/workspace/Paddle/Paddle/build/third_party/install/mklml/lib -Wl,-rpath,/workspace/Paddle/Paddle/build/paddle/phi:/workspace/Paddle/Paddle/build/paddle/common:/workspace/Paddle/Paddle/build/third_party/install/onednn/lib:/workspace/Paddle/Paddle/build/third_party/install/mklml/lib  -lpthread  third_party/install/glog/lib/libglog.a  third_party/install/yaml-cpp/lib/libyaml-cpp.a  paddle/phi/libphi.so  paddle/phi/libphi_gpu.so  paddle/phi/libphi_core.so  paddle/common/libcommon.so  -lrt  paddle/phi/api/profiler/libphi_profiler_proto.a  paddle/phi/core/distributed/auto_parallel/libauto_parallel_proto.a  third_party/install/xxhash/lib/libxxhash.a  libcblas.a  third_party/install/utf8proc/lib/libutf8proc.a  paddle/phi/core/framework/libpass_desc_proto.a  paddle/phi/core/framework/libtrainer_desc_proto.a  paddle/phi/core/framework/libframework_proto.a  paddle/phi/core/framework/libdata_feed_proto.a  paddle/phi/core/framework/libheter_service_proto.a  paddle/phi/core/libexternal_error_proto.a  third_party/install/protobuf/lib/libprotobuf.a  libonednn.a  -ldnnl  -liomp5  third_party/install/gloo/lib/libgloo.a  paddle/phi/backends/dynload/libdynload_mklml.a  -lpthread  third_party/install/glog/lib/libglog.a  third_party/install/yaml-cpp/lib/libyaml-cpp.a  -L/workspace/Paddle/Paddle/build/third_party/install/mklml/lib -liomp5 -Wl,--as-needed  third_party/install/dgc/lib/libdgc.a -pthread -ldl -lrt -lz -lssl -lcrypto && :
/usr/local/bin/ld: paddle/phi/tools/CMakeFiles/print_phi_kernels.dir/print_phi_kernels.cc.o: undefined reference to symbol 'cudaStreamEndCapture@@libcudart.so.12'
/usr/local/bin/ld: /usr/local/cuda-12.3/targets/x86_64-linux/lib/libcudart.so.12: error adding symbols: DSO missing from command line
collect2: error: ld returned 1 exit status

[0x3878f]

错误原因分析 未定义的引用： cudaStreamEndCapture@@libcudart.so.12 是 CUDA 库中的函数，链接器（ld）找不到其定义。 这通常是因为链接命令中未包含 CUDA 运行时库（-lcudart）。 DSO missing from command line： 动态共享对象（DSO）缺失，表明链接器在命令中未找到所需的库文件。

问题解决了吗？-DCMAKE_BUILD_TYPE=Release 可以编译过吗？

[wangguan1995]

现在需要编译Debug版本