.net StopWatch 的计时不一致
inconsistent timing from .net StopWatch
我有一个使用 GPU (NVIDA GTX980) 进行图像处理的 C# 和 .NET 应用程序。有 4 个阶段,我将 CPU 同步到 GPU(时间上没有重叠)以进行计时。但这些数字并没有加起来。
Launch() will do a async launch of the GPU kernel) but synchronize()
will wait till it is done.
- 总计: t阈值:4.2827ms
- t直方图:3.7714ms
- tHistogramSum: 0.1065ms
- tIQR:3.8603 毫秒
- tThresholdOnly:0.4126ms
这是怎么回事?
public static void threshold()
{
Stopwatch watch = new Stopwatch();
watch.Start();
gpu.Lock();
dim3 block = new dim3(tileWidthBig, tileHeightBig);
dim3 grid = new dim3(Frame.width / tileWidthBig, Frame.height / tileHeightBig);
gpu.Launch(grid, block).gHistogram(gForeground, gPercentile, gInfo);
gpu.Synchronize();
tHistogram = watch.Elapsed.TotalMilliseconds;
block = new dim3(1024);
grid = new dim3(1);
gpu.Launch(grid, block).gSumHistogram(gPercentile);
gpu.Synchronize();
tHistogramSum = watch.Elapsed.TotalMilliseconds - tHistogram;
gpu.Launch(grid, block).gIQR(gPercentile, gInfo);
gpu.Synchronize();
tIQR = watch.Elapsed.TotalMilliseconds - tHistogramSum;
block = new dim3(256, 4);
grid = new dim3(Frame.width / 256, Frame.height / 4);
gpu.Launch(grid, block).gThreshold(gForeground, gMask, gInfo);
gpu.Synchronize();
tThresholdOnly = watch.Elapsed.TotalMilliseconds - tIQR;
gpu.Unlock();
watch.Stop();
tThreshold = watch.Elapsed.TotalMilliseconds;
}
由于 TotalMilliseconds 不断递增并且您试图找出时间点之间的差异,因此您需要在第二个差异之后减去前面差异的总和,因此:
tIQR = watch.Elapsed.TotalMillisconds - (tHistogram + tHistogramSum);
&
tThresholdOnly= watch.Elapsed.TotalMillisconds - (tHistogram + tHistogramSum + tIQR);
我有一个使用 GPU (NVIDA GTX980) 进行图像处理的 C# 和 .NET 应用程序。有 4 个阶段,我将 CPU 同步到 GPU(时间上没有重叠)以进行计时。但这些数字并没有加起来。
Launch() will do a async launch of the GPU kernel) but synchronize() will wait till it is done.
- 总计: t阈值:4.2827ms
- t直方图:3.7714ms
- tHistogramSum: 0.1065ms
- tIQR:3.8603 毫秒
- tThresholdOnly:0.4126ms
这是怎么回事?
public static void threshold()
{
Stopwatch watch = new Stopwatch();
watch.Start();
gpu.Lock();
dim3 block = new dim3(tileWidthBig, tileHeightBig);
dim3 grid = new dim3(Frame.width / tileWidthBig, Frame.height / tileHeightBig);
gpu.Launch(grid, block).gHistogram(gForeground, gPercentile, gInfo);
gpu.Synchronize();
tHistogram = watch.Elapsed.TotalMilliseconds;
block = new dim3(1024);
grid = new dim3(1);
gpu.Launch(grid, block).gSumHistogram(gPercentile);
gpu.Synchronize();
tHistogramSum = watch.Elapsed.TotalMilliseconds - tHistogram;
gpu.Launch(grid, block).gIQR(gPercentile, gInfo);
gpu.Synchronize();
tIQR = watch.Elapsed.TotalMilliseconds - tHistogramSum;
block = new dim3(256, 4);
grid = new dim3(Frame.width / 256, Frame.height / 4);
gpu.Launch(grid, block).gThreshold(gForeground, gMask, gInfo);
gpu.Synchronize();
tThresholdOnly = watch.Elapsed.TotalMilliseconds - tIQR;
gpu.Unlock();
watch.Stop();
tThreshold = watch.Elapsed.TotalMilliseconds;
}
由于 TotalMilliseconds 不断递增并且您试图找出时间点之间的差异,因此您需要在第二个差异之后减去前面差异的总和,因此:
tIQR = watch.Elapsed.TotalMillisconds - (tHistogram + tHistogramSum);
&
tThresholdOnly= watch.Elapsed.TotalMillisconds - (tHistogram + tHistogramSum + tIQR);