By: EduardoS (no.delete@this.spam.com), April 27, 2012 1:27 pm
Room: Moderated Discussions
David Kanter (dkanter@realworldtech.com) on 4/27/12 wrote:
---------------------------
>Absolutely. In fact, it's more useful in the context of the K8 than Sandy Bridge.
>The K8 cannot re-order so you REALLY want to calculate every address as soon as possible.
It's not just to calculate address as soon as possible, it also help in packing x86 instructions inside the chip and reduce the size of the forwarding network, if there was two decoupled AGUs there would be 5 sources from the forwarding network per ALU, coupled there are only four*.
* Does anyone knows if being a power of 2 helps?
---------------------------
>Absolutely. In fact, it's more useful in the context of the K8 than Sandy Bridge.
>The K8 cannot re-order so you REALLY want to calculate every address as soon as possible.
It's not just to calculate address as soon as possible, it also help in packing x86 instructions inside the chip and reduce the size of the forwarding network, if there was two decoupled AGUs there would be 5 sources from the forwarding network per ALU, coupled there are only four*.
* Does anyone knows if being a power of 2 helps?



