Hi Crystal,

I think there is probably a big endian issue with this approach. On a
big endian framebuffer without RI_BSWAP, the mixed pixel pairs would
be reversed.

Also, b and f are signed int values, so the cast can corrupt the
adjacent pixel. For example, white after byte swapping is 0xc7c7c700:

    (uint64_t)f            /* 0xffffffffc7c7c700 */
    (uint64_t)(uint32_t)f  /* 0x00000000c7c7c700 */

It needs the additional cast through uint32_t before combining the
two colours.

Ted