If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
|
Thread Tools | Display Modes |
#11
|
|||
|
|||
Can someone explain step by step how one avoid many conditionalin forth as described in Moore Fourth essay?
On Jan 14, 4:23*am, Terje Mathisen "terje.mathisen at tmsw.no"
wrote: Joe keane wrote: In , Andy (Super) *wrote: Problem is that eager threads seldom last long, so it is hard for eager threads to overcome the cost of migrating to another core. Essentially creating a new thread, for every -branch-, falls into the 'that don't sound right'. I've thought about this in the context "execute alternating instructions from both sides of the branch, adding implicit predicates to them", so that when the branch retires the predicates become known and half the instructions are cancelled. This reminds me of a (not so useful) idea I posted here 21 Oct 2010, "Move on 'commit' predication?" Message-ID: http://groups.google.com/group/comp....b5e80e5fd36288 It was basically just a (wrong-headed?) microarchitecture for supporting dynamic predication (of hammock branches) where one path was less likely but uncertain enough to justify predication. (One of the issues of eager execution is recognizing joining.) |
#12
|
|||
|
|||
Can someone explain step by step how one avoid many conditionalin forth as described in Moore Fourth essay?
On 1/13/2012 8:16 PM, Joe keane wrote:
In , Andy (Super) wrote: Problem is that eager threads seldom last long, so it is hard for eager threads to overcome the cost of migrating to another core. Essentially creating a new thread, for every -branch-, falls into the 'that don't sound right'. Nobody would seriously propose a new thread for every branch. You would only consider eager threading for flakey branches, where you don't know which way they are going to go. And won't know for a while. Myself, I think non-threading approaches to eager execution are better. E.g. the discussion of short forward branches - emit colde that computes both sides of such a hammock, with operations like Phi nodes to bring back rto a single dataflow. |
|
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Upgrade Report [Step-By-Step: Tune Up Your Windows XP PC - 05/25/2004] | Ablang | General | 0 | June 30th 04 03:08 AM |
Anyone give me a step by step to setting up raid 0? | Mike Barnard | Gigabyte Motherboards | 6 | April 12th 04 10:06 AM |
Newbie question - online step by step??? | Moldy | Overclocking AMD Processors | 0 | December 8th 03 11:56 AM |
INCREDIBLE ! I can't tune the FSB step by step on an A7N8X-X !!! ASUS a liar ? | GreatShot | Asus Motherboards | 3 | August 2nd 03 07:25 PM |