A method of video processing is described. The method includes determining chroma weights used for determining a chroma prediction block of a chroma block of a current block of a video by blending predictions of the chroma block according to a rule, and performing a conversion between the current block and a coded representation of the video according to the determining, wherein the rule specifies that the chroma weights are determined from luma weights of a collocated luma block of the current block; wherein the current block is coded with a geometric partitioning mode.