From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Wed May 07 2008 - 13:23:36 CDT
Hi Lei,
this warning should not have any effect on namd, which does not
generally use the thread migration features of charm. Are you sure your
FEPs in both cases are sufficiently sampled? I've never seen any
aberrant results coming from the build on ranger. Are you using Jim's
build, or did you compile it yourself?
Best,
Peter
Lei Shi wrote:
> Dear namdlist,
>
> The problem I have is different from the previous thread about running
> NAMD with ranger_at_tacc...
>
> I have been following the instruction from
> http://www.ks.uiuc.edu/Research/namd/wiki/index.cgi?NamdAtTexas
>
> to run NAMD at Ranger(TACC).
>
> I noticed the following error
>
> "Charm warning> Randomization of stack pointer is turned on in Kernel,
> run 'echo 0 > /proc/sys/kernel/randomize_va_space' as root to disable
> it. Thread migration may not work!"
>
>
> I cannot tell the impact of this on the regular MD, because the system
> seems maintaining sanity. But for FEP/MD, the free energy number I got
> doubles what I got locally, which I assumes is what is supposed to be.
>
> It is unlikely I can login as a root, so is it something I need to
> talk to the staff there, or it's a compilation thing?
>
>
> Thanks.
> Lei
>
This archive was generated by hypermail 2.1.6 : Sun Jul 06 2008 - 19:18:32 CDT