On 03/31, Rajesh Paulraj wrote: > *This looks highly suspect to me. Why would the destination stride > bedifferent depending on the subpel offsets? You're using MAX_CU_SIZE > here**but > dstStride is used everywhere below.* > > luma and chroma are using the same C function filterPixelToShort_C. The > dstStride is MAX_CU_SIZE for luma and MAX_CU_SIZE/2 for chroma i420,i422, > MAX_CU_SIZE for i444. For earlier C function we passed > dstStride(MAX_CU_SIZE) directly to C function.
this was wrong before then, it shouldn't continue this way. if you are going to pass a destination stride, pass the correct one. And the test harness needs to test likely strides for each block size -- Steve Borho _______________________________________________ x265-devel mailing list [email protected] https://mailman.videolan.org/listinfo/x265-devel
