Re: [AMBER] sebomd and qmmm2 test failing in serial/parallel for Amber14

From: Gerald Monard <Gerald.Monard.univ-lorraine.fr>
Date: Thu, 19 Jun 2014 12:09:55 +0200

Dear Qingfen,

As a workaround, if you want really to keep gcc-4.8.1, you should
disable all strong optimizations (see gcc-4.8.2 changelogs, gcc-4.8.1
has problems with code optimization).

What you can do is the following: after the './configure gnu' command,
manually inspect the 'config.h' file in $AMBERHOME and replace all '-O3
-mtune=native' and '-O3' commands by '-O'. I tested it with gcc-4.8.1
and it works.

Gerald.

P.S.: However, I still suggest upgrading your compiler and/or your OS.

On 06/18/2014 09:44 PM, Gerald Monard wrote:
> Hi,
>
> I have compiled gcc-4.8.1 and gcc-4.8.2 on a Scientific Linux 5.7
> (equivalent to CentOS 5.7). With gcc-4.8.1, the sebomd test fails (I
> checked only sebomd tests). With gcc-4.8.2, sebomd tests don't fail.
> I would then suggest upgrading your gcc version...
>
> Gerald.
>
> On 06/18/2014 12:03 PM, 余青芬 wrote:
>> Dear Gerald,
>>
>> I try compiling with the default gcc-4.1.2 on CentOS 5.1, and the sebomd test errors in serial and parallel (2, 4 and 8 processes) persist. It is appreciated if you could send me some putative fixes for the sebomd part.
>>
>> We currently cannot update the CentOS, and I didn't compile with Intel or PGI.
>>
>> Furthermore, there are more test errors using gcc-4.1.2 than using gcc-4.8.1. It seems that Amber14 does not compile properly with the default gcc-4.1.2.
>>
>> Qingfen
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> --
>> Qingfen Yu
>> Center for High Performance Computing
>> Shenzhen Institutes of Advanced Technology
>> Chinese Academy of Sciences
>> 1068 Xueyuan Avenue, Shenzhen University Town
>> Shenzhen, P.R. China, 518055
>>
>>
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>>
>

-- 
____________________________________________________________________________
  Prof. Gerald MONARD
  SRSMC, Université de Lorraine, CNRS
  Boulevard des Aiguillettes B.P. 70239
  F-54506 Vandoeuvre-les-Nancy, FRANCE
  e-mail : Gerald.Monard.univ-lorraine.fr
  tel.   : +33 (0)383.684.381
  fax    : +33 (0)383.684.371
  web    : http://www.monard.info
____________________________________________________________________________
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Jun 19 2014 - 03:30:02 PDT
Custom Search