This is cuda 4.2 and cuda 5.0
i have real data (4 byte), image with 1280 columns, 720 rows, 10 frames.
cudaExtent extent = make_cudaExtent (xRes * sizeof(float), yRes, gNumFramesInKernel); err = cudaMalloc3D( &PPtr, extent);
extent is properly built.
PPtr returns pitch and xsize as identical numbers [that is, both are in unbits oif bytes]; BUT documentation says that xsize and ysize of cudaPitchedPtr are logical units [pixels in my case] while pitch is the stride – that is number of bytes from the start of one row to the next row.
shouldnt the returned cudaPitchedPtr conform to the documentation? what happens if i modify by:
PPtr.xsize = PPtr.xsize / 4;