Write out each pass, including the final answer, in a parallel reduction algorithms described in Figure 5.13 and Figure 5.15 of the GPU textbook using to find the sum of the following vector: [0 3 4 6 1 2 2 8 3 3 2 1 4 1 5 6] Analyze the thread divergence for both the algorithms.

1. _shared_float partialSum[SIZE); partialSum(threadIdx.x] = X[blockIdx.x*blockDim.x+threadIdx.x]; 2. unsigned int t = threadIdx.x; 3. for (unsigned int stride = 1; stride FIGURE 5.13 A simple sum reduction Kernel. 8 5. 1. - shared_float partialSum[SIZE]; partialSum(threadIdx.x) = X[blockIdx.x*blockDim.x+threadIdx.x]; 2. unsigned int t = threadIdx.x; 3. for (unsigned int stride - blockDim.x/2; stride >= 1; stride = stride>>1) _syncthreads(); if (t FIGURE 5.15 A kernel with fewer thread divergence. 6