Subject: Re: Possible bug in arm32 strongarm optimisations.
To: Thor Lancelot Simon <tls@rek.tjls.com>
From: Ignatios Souvatzis <ignatios@theory.cs.uni-bonn.de>
List: port-arm32
Date: 10/25/2000 21:32:31
On Wed, Oct 25, 2000 at 12:44:14PM -0400, Thor Lancelot Simon wrote:
> And, sometimes, it can even be a win.  If you have an algorithm that wants
> to do rotates, and you can structure it so that it does them by way of
> unaligned loads/stores -- they're "free", they take no extra cycles, since
> you had to get the data in/out anyway...

Hm, only there isn't a way to get them in or out of that memory quickly, in
an unrotated way.

OTOH, every register-toregister ALU operation of the ARMs comes with a free
shift or rotate of one of the operands.

Regards,
	-is