By: dess (dess.delete@this.nospam.com), May 22, 2007 4:09 pm
Room: Moderated Discussions
EduardoS (no@spam.com) on 5/22/07 wrote:
---------------------------
>An AMD's macro-op can have up to, and in this order:
>An address calculation, a load, an operation and a store
"Can have"? I don't get it. I would think an address calculation bolongs to the _execution_ of a macro-op.
>But AMD only call uop the address calculation and the store.
Hmmm, AMD docs say:
"A macro-op is a fixed-length instruction that:
•Expresses, at most, one integer or floating-point operation and one load and/or store operation.
•Is the primary unit of work managed (that is, dispatched and retired) by the processor.
A micro-op is a fixed-length instruction that:
•Expresses one and only one of the primitive operations that the processor can perform (for example, a load).
•Is executed by the processor’s execution units."
>Intel fused-uop can be a load+exec or store adress+store data, Intel call a load
>uop the entire process of calculating and address and loading the data.
And what do you think how an AMD processor is executing f.ex. a "load" micro-op?
ps. why do you say "Stop comparing apples to oranges"? Where I handled the two firm's given terms like there were the same, or so?
---------------------------
>An AMD's macro-op can have up to, and in this order:
>An address calculation, a load, an operation and a store
"Can have"? I don't get it. I would think an address calculation bolongs to the _execution_ of a macro-op.
>But AMD only call uop the address calculation and the store.
Hmmm, AMD docs say:
"A macro-op is a fixed-length instruction that:
•Expresses, at most, one integer or floating-point operation and one load and/or store operation.
•Is the primary unit of work managed (that is, dispatched and retired) by the processor.
A micro-op is a fixed-length instruction that:
•Expresses one and only one of the primitive operations that the processor can perform (for example, a load).
•Is executed by the processor’s execution units."
>Intel fused-uop can be a load+exec or store adress+store data, Intel call a load
>uop the entire process of calculating and address and loading the data.
And what do you think how an AMD processor is executing f.ex. a "load" micro-op?
ps. why do you say "Stop comparing apples to oranges"? Where I handled the two firm's given terms like there were the same, or so?
| Topic | Posted By | Date |
|---|---|---|
| Barcelona Article Online | David Kanter | 05/16/07 03:20 AM |
| Barcelona Article Online | PiedPiper | 05/16/07 05:12 AM |
| Yes, I left out a sentence there. Fixed (NT) | David Kanter | 05/16/07 12:07 PM |
| Barcelona Article Online | anonymous | 05/16/07 06:01 AM |
| Barcelona Article Online | Anonymous | 05/16/07 06:28 PM |
| Barcelona Article Online | anonymous | 05/16/07 07:52 PM |
| Barcelona Article Online | Anonymous1 | 05/16/07 07:08 AM |
| Barcelona Article Online | Dean M | 05/16/07 11:09 AM |
| Barcelona Article Online | David Kanter | 05/16/07 12:38 PM |
| Barcelona Article Online | Dean M | 05/16/07 02:10 PM |
| Barcelona Article Online | IntelUser2000 | 05/16/07 02:59 PM |
| Barcelona Article Online | Linus Torvalds | 05/16/07 03:24 PM |
| Barcelona Article Online | David Kanter | 05/16/07 04:57 PM |
| Barcelona Article Online | Michael S | 05/17/07 05:07 AM |
| Barcelona Article Online | IntelUser2000 | 05/18/07 08:58 PM |
| 8 socket servers | Doug Siebert | 05/16/07 04:58 PM |
| 8 socket servers | Michael S | 05/17/07 05:20 AM |
| 8 socket servers | Joe Chang | 05/17/07 07:38 AM |
| 8 socket servers | Alex Jones | 05/17/07 09:35 AM |
| 8 socket servers | Jose | 05/23/07 08:23 AM |
| 8 socket servers | Michael S | 05/23/07 11:37 AM |
| 8 socket servers | anonymous | 05/26/07 03:49 PM |
| 8 socket servers | Joe Chang | 05/27/07 01:46 PM |
| 8 socket servers | Doug Siebert | 05/23/07 09:56 PM |
| 8 socket servers | Joe Chang | 05/24/07 04:33 AM |
| 8 socket servers | Anonymous | 05/24/07 11:18 AM |
| 8 socket servers | Doug Siebert | 05/24/07 10:47 PM |
| 8 socket servers | Linus Torvalds | 05/25/07 10:35 AM |
| 8 socket servers | Nick | 05/25/07 02:29 AM |
| Performance estimation seems odd | Hotar | 05/17/07 01:54 AM |
| Performance estimation seems odd | David Kanter | 05/17/07 08:38 AM |
| microops vs macroops on page 4 | Peter Lund | 05/17/07 12:04 PM |
| microops vs macroops on page 4 | David Kanter | 05/21/07 04:51 PM |
| microops vs macroops on page 4 | EduardoS | 05/21/07 05:42 PM |
| microops vs macroops on page 4 | dess | 05/21/07 07:00 PM |
| Barcelona Article Online | Peter Lund | 05/17/07 12:25 PM |
| macro-op vs. micro-op | dess | 05/21/07 07:24 AM |
| macro-op vs. micro-op | David Kanter | 05/21/07 04:38 PM |
| macro-op vs. micro-op | dess | 05/21/07 06:15 PM |
| macro-op vs. micro-op | David Kanter | 05/22/07 12:11 AM |
| macro-op vs. micro-op | dess | 05/22/07 03:56 AM |
| macro-op vs. micro-op | Gipsel | 05/22/07 05:05 AM |
| macro-op vs. micro-op | dess | 05/22/07 05:52 AM |
| macro-op vs. micro-op | anonymous | 05/22/07 06:14 AM |
| macro-op vs. micro-op | dess | 05/22/07 06:44 AM |
| macro-op vs. micro-op | EduardoS | 05/22/07 02:19 PM |
| macro-op vs. micro-op | dess | 05/24/07 08:52 AM |
| Stop comparing apples to oranges | EduardoS | 05/22/07 02:30 PM |
| Stop comparing apples to oranges | dess | 05/22/07 04:09 PM |
| Stop comparing apples to oranges | dess | 05/22/07 04:30 PM |
| Stop comparing apples to oranges | EduardoS | 05/22/07 04:31 PM |
| Stop comparing... apples to oranges? | dess | 05/24/07 09:30 AM |
| Stop comparing apples to oranges | anonymous | 05/22/07 08:12 PM |
| Stop comparing apples to oranges | EduardoS | 05/23/07 02:50 PM |
| macro-op vs. micro-op | anonymous | 05/22/07 06:08 AM |
| macro-op vs. micro-op | dess | 05/22/07 06:40 AM |
| macro-op vs. micro-op | anonymous | 05/22/07 06:48 AM |
| macro-op vs. micro-op | dess | 05/21/07 08:30 PM |
| macro-op vs. micro-op | anonymous | 05/22/07 06:44 AM |
| macro-op vs. micro-op | dess | 05/24/07 09:38 AM |
| macro-op vs. micro-op | Michael S | 05/22/07 05:26 AM |



