> The checkpoint dumped is 'cpt.003825856942'. Then I config the phase period of > Sampler to be [3825856942, 1e15] That period is relative. The first number is the number of cycles that you want to warm up. Sounds like you only want to warm up for one cycle.
> and change spec-surge-server.rcS back to: the rcS file is only run the first time. If you're loading from a checkpoint, the rcS file is ignored because you're past the point where it was loaded. > After these configuration, I run './m5.opt run.py -d output/output138 > -ESYSTEM=Sampler -EMEMORY=HTX -ETEST=SURGE_SPECWEB -EDUMPFILE=138'. The way this is specified, I don't see how you're running from a checkpoint. You probably need to do something like: -ECKPT_FILE=path/to/cpt.0123456789 > I start the simulation a day before, but now the sever still stay at the point > 'starting httpd...' and will not move on. I think the problem is that you haven't started from a checkpoint, and the way you've specified your sampler periods means that you're actually still stuck loading the httpd. This is a slow process normally. It is made orders of magnitude slower by the fact that you're executing that part in the most detailed mode now. Hope this helps. Nathan Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ m5sim-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/m5sim-users
