Simple CbYCrY conversion to Grayscale Image(only Y) takes 8 MIPS. Help needed how to optimize using cache I am attaching sample code for your reference
pucInTemp - Input Buffer
pucOutTemp - Output Buffer
-----------------------------------------------------------------------
pucInTemp = pInputBuffer+1;
for(uiCount = 0; uiCount <720*480; uiCount++)
{
*pucOutTemp = *pucInTemp;
pucOutTemp++;
pucInTemp = pucInTemp+2;
}