kernel oops due to unaligned access with lswi

[Paul Mackerras]

David Edelsohn writes:

> 	The information that you received about lwsi are overly
> simplistic.  The instructions are neither overly good nor overly bad --
> they should not be used for everything, but neither should they be avoided
> at all cost.  They are particularly good for producing compact code and
> preserving the instruction cache.  Remember, programming, including
> assembly language programming, is an art.

In the experiments that I did comparing different memcpy loops, I
didn't find any combination of alignment, size of copy and processor
implementation where using string loads/stores was as fast as than
ordinary loads and stores.  So I am inclined to think that the only
advantage of the string load/store instructions is in saving a little
bit of icache.

Paul.

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/