On Mon, May 17, 2021 at 8:16 AM Christoph Hellwig <hch@xxxxxx> wrote:
On Sat, May 15, 2021 at 12:18:01PM +0200, Arnd Bergmann wrote:
From: Arnd Bergmann <arnd@xxxxxxxx>
Most per-architecture versions of these functions are broken
in some form, and they are almost certainly slower than the
generic code as well.
This version is fairly slow because it always does byte accesses
even for aligned data, and its checks for user_addr_max() differ
from the generic code.
Remove the ones for arc and instead use the generic version.
Same comment as for hexaon before.
Changed now.
Arnd
