什麼是實現定義爲使用的OpenCL內核的OpenCL內核實現一個簡單的數學公式
誤差函數時要考慮的最佳實踐?
A,B和C是三維浮法陣列,\ delta是克羅內克三角洲。 (N,M)=(2,7)或(N,M)=(2,23)的典型值。
天真的實現(下面給出)比CPU版本慢幾個數量級。
感謝,
T.
__kernel void cl_bilinear_alg(
__global float * A,
__global float * B,
__global float * C,
__global const int M,
__global const int N,
__global float * R)
{
int index = get_global_id(0);
int N2 = N * N;
int mat_offset = index * N2 * M;
float s1, s2, err = 0.0f;
for (int i = 0; i < N; ++i)
{
for (int j = 0; j < N; ++j)
{
for (int k = 0; k < N; ++k)
{
for (int l = 0; l < N; ++l)
{
for (int m = 0; m < N; ++m)
{
for (int n = 0; n < N; ++n)
{
s1 = (n == i) * (j == k) * (l == m);
s2 = 0;
for (int r = 0; r < M; ++r)
{
s2 += A[mat_offset + r * N2 + i * N + j] *
B[mat_offset + r * N2 + k * N + l] *
C[mat_offset + r * N2 + m * N + n];
}
err += (s2 - s1) * (s2 - s1);
}
}
}
}
}
}
R[index] = err;
}
UPDATE
主要的目標是一塊GeForce GTX 570,雖然這可能在未來改變。
UPDATE2
明確地矢量化的代碼,移動位本地內存,展開一些循環並通過預先計算克羅內克產品的內核後,代碼如下:
__kernel void cl_bilinear_alg(__global const float * A,
__global const float * B,
__global const float * C,
__global const int N,
__global const int M,
__global const float * kron,
__global float * R)
{
__private int index = get_global_id(0);
__private int cM = ceil(M/4.0f);
__private int N2 = N*N;
__private int N4 = N2*N2;
__private int mat_offset = index * N2 * M;
__private float s1, s2, err = 0;
__private float4 vzero = (float4) (0.0f, 0.0f, 0.0f, 0.0f);
__local float4 va[54], vb[54], vc[54];
for (int ij = 0, k = 0; ij < N2; ++ij)
{
int r = 0;
for (; r < M/4; r += 4, ++k)
{
int idx0 = mat_offset + N2 * r + ij;
int idx1 = mat_offset + N2 * (r + 1) + ij;
int idx2 = mat_offset + N2 * (r + 2) + ij;
int idx3 = mat_offset + N2 * (r + 3) + ij;
va[k] = (float4) (A[idx0], A[idx1], A[idx2], A[idx3]);
vb[k] = (float4) (B[idx0], B[idx1], B[idx2], B[idx3]);
vc[k] = (float4) (C[idx0], C[idx1], C[idx2], C[idx3]);
}
if (M % 4)
{
float buffa[4] = {0}, buffb[4] = {0}, buffc[4] = {0};
for (; r < M; ++r)
{
int idx = mat_offset + N2 * r + ij;
buffa[r % 4] = A[idx];
buffb[r % 4] = B[idx];
buffc[r % 4] = C[idx];
}
va[k] = vload4(0, buffa);
vb[k] = vload4(0, buffb);
vc[k++] = vload4(0, buffc);
}
}
for (int ij = 0; ij < N2; ++ij)
{
for (int kl = 0; kl < N2; ++kl)
{
for (int mn = 0; mn < N2; ++mn)
{
s1 = kron[ij * N4 + kl * N2 + mn];
s2 = 0;
for (int r = 0; r < cM; ++r)
s2 += dot(va[cM * ij + r], mad(vb[cM * kl + r], vc[cM * mn + r], vzero));
//the most expensive line
err += (s2 - s1) * (s2 - s1);
}
}
}
R[index] = err;
}
通過將這些相比於天真的實施,觀察到4倍的速度增加。此外,顯示所有最昂貴的行是錯誤更新,即
err += (s2 - s1) * (s2 - s1);
有什麼建議嗎?
設備的架構是什麼?如果是帶矢量寄存器的intel,則可以使用SIMD寄存器獲得更高的速度,速度可提高5倍 – ardiyu07
「天真」版本作爲單個工作項目運行?如果是這樣,你正在浪費GPU的總運算峯值容量的99.8%,所以它應該不會太過出人意料,它的速度很慢。 – talonmies
@talonmies:不,每個工作項目都在工作一個單獨的問題。 – user92382