未验证 提交 6b6f7a21 编写于 作者: M ming1753 提交者: GitHub

Fc fp16 (#44558)

* (modified) fc support fp16

* __CUDA_ARCH__ version

* delete half

* delete half
上级 19d9c736
......@@ -36,14 +36,6 @@ struct FcTypeTraits<double> {
typedef double4 Type;
};
#if defined(PADDLE_WITH_CUDA)
#include <cuda_fp16.h>
template <>
struct FcTypeTraits<float16> {
typedef half2 Type;
};
#else
struct float16_4 {
float16 x, y, z, w;
};
......@@ -52,7 +44,6 @@ template <>
struct FcTypeTraits<float16> {
typedef float16_4 Type;
};
#endif
template <typename T, bool DoRelu>
__global__ void bias_relu_v4(const int num, const T* bias, T* data, int K) {
......@@ -126,95 +117,12 @@ void AddReluKernel(
}
}
#if defined(PADDLE_WITH_CUDA)
template <bool DoRelu>
__global__ void bias_relu_v2(const int num,
const half2* bias,
half2* data,
int K) {
int tid = blockIdx.x * blockDim.x + threadIdx.x;
if (tid < num) {
int bias_idx = tid % K;
const half2 bias_ptr = bias[bias_idx];
const half2 in_ptr = data[tid];
half2 packed_val = __hadd2(bias_ptr, in_ptr);
if (DoRelu) {
#if __CUDA_ARCH__ >= 800
packed_val = __hmax2(__half2(0, 0), packed_val);
#else
packed_val = __hmul2(__hgt2(__half2(0, 0), packed_val), packed_val);
#endif
}
data[tid] = packed_val;
}
}
template <bool DoRelu, int BlockDim>
__global__ void InplaceAddReluKernel(const int N,
const half* bias,
half* data) {
int offset = blockIdx.x * N;
for (int i = threadIdx.x; i < N; i += BlockDim) {
half temp;
#if defined(__HIPCC__) || __CUDA_ARCH__ >= 350
temp = __ldg(data + offset + i) + __ldg(bias + i);
#else
temp = data[offset + i] + bias[i];
#endif
if (DoRelu) {
#if __CUDA_ARCH__ >= 800
data[offset + i] = __hmax(0, temp);
#else
data[offset + i] = __hmul(__hgt(temp, 0), temp);
#endif
} else {
data[offset + i] = temp;
}
}
}
template <>
void AddReluKernel(cudaStream_t stream,
const int M,
const int N,
float16* Y,
const float16* B,
bool relu) {
if (N % 2 == 0) {
const int threads = 256;
const int num = M * N / 2;
const int blocks = (num + threads - 1) / threads;
typedef typename FcTypeTraits<float16>::Type trans_type;
auto* bias_ptr_v2 = reinterpret_cast<const trans_type*>(B);
auto* data_ptr_v2 = reinterpret_cast<trans_type*>(Y);
if (relu) {
bias_relu_v2<true><<<blocks, threads, 0, stream>>>(
num, bias_ptr_v2, data_ptr_v2, N / 2);
} else {
bias_relu_v2<false><<<blocks, threads, 0, stream>>>(
num, bias_ptr_v2, data_ptr_v2, N / 2);
}
} else {
const int threads = 256;
const int blocks = M;
auto* halfB = reinterpret_cast<const half*>(B);
auto* halfY = reinterpret_cast<half*>(Y);
if (relu) {
InplaceAddReluKernel<true, threads>
<<<blocks, threads, 0, stream>>>(N, halfB, halfY);
} else {
InplaceAddReluKernel<false, threads>
<<<blocks, threads, 0, stream>>>(N, halfB, halfY);
}
}
}
#else
template <bool DoRelu, int BlockDim>
__global__ void InplaceAddReluKernel(const int N,
const float16* bias,
float16* data) {
int offset = blockIdx.x * N;
for (int i = threadIdx.x; i < N; i += BlockDim) {
float16 temp;
temp = data[offset + i] + bias[i];
......@@ -260,7 +168,6 @@ void AddReluKernel(gpuStream_t stream,
}
}
}
#endif
template <typename DeviceContext, typename T>
void FCFunctor<DeviceContext, T>::operator()(const DeviceContext& context,
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册