-
Cheng Chen authored
With shift, convolve copy no longer needs 32-bit multiplication of two 8-bit numbers. Thus we can implement it with sse2 instead of sse4. Change-Id: I63e8ba414383a24f820bad4a6c607f222ec40ec2
3f2b57d8
With shift, convolve copy no longer needs 32-bit multiplication of two 8-bit numbers. Thus we can implement it with sse2 instead of sse4. Change-Id: I63e8ba414383a24f820bad4a6c607f222ec40ec2