Re: powerpc & unaligned block moves with fp registers

[degger at fhm dot edu]

On 10 Nov, mike stump wrote:

>> Pessimistic or not, we should try to avoid misaligned accesses on any
>> cost since they are in a lucky case at least magnitude slower.
 
> I reject the notion that we should at any cost.  What if it costs are
> 10% on the SPEC2000 performance benchmark?  What if we are using -Os,
> and it makes code 6% larger?  A good change, will be backed up by size
> and time numbers from a variety of benchmarks.

Ok, I take my "on any cost" back. -Os is of course a different picture,
when optimising for size alignment doesn't play such a big role. Anyway
the mentioned cases are pretty rare (at least I haven't seen much
wrongly aligned structures here in the past) and thusly I also think
that aligning them properly would make such a big different in code
size.

--
Servus,
       Daniel