Hi Manuel,
The bug has been fixed and a patch is awaiting approval for release. The bug occurs only when a COM group involves atom 1.
Update.11 will fix this. Jason can you provide a timeframe for it’s release?
In the meantime, if it is possible to not use atom 1 in any of your groups in your system, you shouldn’t see the error.
Thanks,
Perri
On Feb 25, 2015, at 8:29 PM, Ross Walker <ross.rosswalker.co.uk> wrote:
> Hi Manuel,
>
> Could you please send me you input files: mdin, prmtop and inpcrd file directly to my email (instead of the list) and I will try to reproduce this.
>
> All the best
> Ross
>
>> On Feb 25, 2015, at 8:19 PM, mmaestre <mmaestre.gate.sinica.edu.tw> wrote:
>>
>> Dear Amber users/developers
>>
>> A while ago I posted the message I am reproducing below. I haven't received any comments on my problem, and I still am not able to solve it by myself. I've noticed, though, that the message does not appear as a new thread, but as a reply to an unrelated thread. This is probably my fault, as I probably started it by replying to the unrelated thread and editing the message. Thus, I am hoping that that's the reason for the absence of replies, and that someone might give me a hand ;)
>> I am now re-posting the message, using a completely new thread.
>>
>> Thanks a lot, and sorry if this was unnecessary, but I'd be really interested in getting this to run!
>>
>> Manuel
>>
>> Here is my original message:
>>
>>
>> Der Amber users,
>>
>> I've been dealing lately with a protein-dna complex, in which I want to compare the dna behavior in the presence and absence of an experimentally determined (via FRET) distance restraint.
>> Since, since last month, COM restraints are available in pmemd.cuda, I decided to give them a try.
>> My results are not really encouraging...
>> When using the following control file:
>>
>> &cntrl
>> imin=0,irest=0,ntx=7,ioutfm=1,nmropt=1
>> ntb=2,pres0=1.0,ntp=1,barostat=2
>> taup=2.0,
>> ntr=1,ntc=2,ntf=2,
>> tempi=300.0,temp0=300.0,ntt=3,ig=-1,gamma_ln=5
>> nstlim=1000000,dt=0.002,
>> ntpr=10000,ntwx=10000,ntwr=100000
>> cut=8
>> /
>> &wt type='DUMPFREQ', istep1=2 /
>> &wt type='REST', istep1=0,istep2=1000000,value1=0.1,
>> value2=1, /
>> &wt type='END' /
>> DISANG=res.RST
>> DUMPAVE=bent_res_immo_min5.rst
>> LISTIN=POUT
>> LISTOUT=POUT
>> END
>>
>> With the following restraint file:
>> &rst iat= -1,-1, r1= 49.8, r2= 50.8, r3=55.76, r4=56.76, rk2=1, rk3=0.1, ir6=1, ialtd=0, igr1=1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63, igr2=579,580,581,582,583,584,585,586,587,588,589,590,591,592,593,594,595,596,597,598,599,600,601,602,603,604,605,606,607,608,609,610,611,612,613,614,615,616,617,618,619,620,621,622,623,624,625,626,627,628,629,630,631,632,633,634,635,636,637 /
>>
>> I consistently get ~20 setps, after which the following error appears:
>> Error: an illegal memory access was encountered launching kernel kClearForces
>> cudaFree GpuBuffer::Deallocate failed an illegal memory access was encountered
>>
>> I've tried it on GTX780 and on K20 cards, with the same results.
>>
>> If I run the same simulation on CPUs, or if I modify the restraint file to present an atom based restraint, instead of a center of mass based one (i.e. i delete igr1 and igr2, and substitute iat= -1,-1, for iat= 1,579) and run it on either of the GPUs, the simulation can be run successfully.
>>
>> I am not sure if my input files are bad, or if there might be any bugs on the COM implementation, but just thought it might be useful to post it here, and get some comments.
>>
>> Thank you very much,
>> Manuel
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Feb 27 2015 - 12:00:02 PST
