Date: | Wed, 6 Jul 2011 10:53:47 +0800 (CST) |
---|---|
From: | argon_zh <argon_zh@xxxxxxx> |
Subject: | [Gems-users] warm application checkpoint |
Hello, I have read the GEMS isca tutorial and some former links on making warmup application checkpoint (https://lists.cs.wisc.edu/archive/gems-users/2006-March/msg00143.shtml, https://lists.cs.wisc.edu/archive/gems-users/2006-October/msg00099.shtml). However, I still have some questions on it. I attacked the procedure from https://lists.cs.wisc.edu/archive/gems-users/2006-March/msg00143.shtml. I post my questions on it. 1) first run (Only need Simics to simulation) ---------------------------------------------------------------------------------------- * The Console command: ./BARNES < input-warm * write-configuration CHKPT_DIR/barnes-cold-16p.check (generate cold checkpoint) ----------------------------------------------------------------------------------------- Q1. Do I need to enable magic break at this step? Q2. How to decide the size of input-warm? 2) second run (Need Simics+Ruby) ------------------------------------------------------------------------------------------- load "cold" checkpoint generated in the first run, set up ruby *The Console command: ./BARNES < input-warm * ruby0.save-caches barnes.caches.gz (generate warmup cache file) write-configuration CHKPT_DIR/barnes-warm-16p.check (generate warm checkpoint)
Q3. Do I need to enable magic break at this step?
|
[← Prev in Thread] | Current Thread | [Next in Thread→] |
---|---|---|
|
Previous by Date: | Re: [Gems-users] compile SPLASH2, Hamid Reza Khaleghzadeh |
---|---|
Next by Date: | Re: [Gems-users] How to collect multicore coherence traffic, jianhual |
Previous by Thread: | Re: [Gems-users] Some Problems about the MOESI_directory cache coherence protocol, sheng ma |
Next by Thread: | [Gems-users] write-allocate or not???, Hamid Reza Khaleghzadeh |
Indexes: | [Date] [Thread] |