radeonsi: cosmetic changes to the clear/copy_buffer shader, make 12B clears faster, rewrite the microbenchmark
This is preparation before adding support for unaligned clears and copies.
Due to an influx of spam, we have had to impose restrictions on new accounts. Please see this wiki page for instructions on how to get full permissions. Sorry for the inconvenience.
This is preparation before adding support for unaligned clears and copies.