fix a bug caused by hipcc lambda value capture (#55612)

lishicheng1996 · web-flow · commit 8db3ff1f8daa · 2023-07-25T16:15:06.000+08:00
diff --git a/paddle/phi/kernels/funcs/blas/blas_impl.hip.h b/paddle/phi/kernels/funcs/blas/blas_impl.hip.h
@@ -1173,6 +1173,56 @@ void Blas<phi::GPUContext>::BatchedGEMM(CBLAS_TRANSPOSE transA,
   });
 }
 
+template <>
+template <>
+inline void Blas<phi::GPUContext>::BatchedGEMM(CBLAS_TRANSPOSE transA,
+                                               CBLAS_TRANSPOSE transB,
+                                               int M,
+                                               int N,
+                                               int K,
+                                               float16 alpha,
+                                               const float16 *A,
+                                               const float16 *B,
+                                               float16 beta,
+                                               float16 *C,
+                                               int batchCount,
+                                               int64_t strideA,
+                                               int64_t strideB) const {
+  // Note that cublas follows fortran order, so the order is different from
+  // the cblas convention.
+  int lda = (transA == CblasNoTrans) ? K : M;
+  int ldb = (transB == CblasNoTrans) ? N : K;
+  int ldc = N;
+  rocblas_operation cuTransA = (transA == CblasNoTrans)
+                                   ? rocblas_operation_none
+                                   : rocblas_operation_transpose;
+  rocblas_operation cuTransB = (transB == CblasNoTrans)
+                                   ? rocblas_operation_none
+                                   : rocblas_operation_transpose;
+  const int64_t strideC = M * N;
+  context_.CublasCall([&](rocblas_handle handle) {
+    PADDLE_ENFORCE_GPU_SUCCESS(phi::dynload::rocblas_hgemm_strided_batched(
+        handle,
+        cuTransB,
+        cuTransA,
+        N,
+        M,
+        K,
+        reinterpret_cast<const rocblas_half *>(&alpha),
+        reinterpret_cast<const rocblas_half *>(B),
+        ldb,
+        strideB,
+        reinterpret_cast<const rocblas_half *>(A),
+        lda,
+        strideA,
+        reinterpret_cast<const rocblas_half *>(&beta),
+        reinterpret_cast<rocblas_half *>(C),
+        ldc,
+        strideC,
+        batchCount));
+  });
+}
+
 // note(wangran16): unknown bug. parameters dislocation when calling
 // GEMM_STRIDED_BATCH<float> and GEMM_STRIDED_BATCH<double>
 template <>
