On 16.02.2012 15:33, David A Case wrote:
> On Fri, Jan 27, 2012, Jan-Philip Gehrcke wrote:
>>
>> We're using Amber 11 (including all bugfixes up to last week) and
>> observed a compiler-dependent error while parsing a DISANG file in
>> sander. For ICC 12.1, things work properly, while for GCC 4.6.2, sander
>> crashes. Details below.
>
> Just a short, not-so-helpful note: the test case runs fine for me with gcc
> 4.6.1 on Mac OSX, 10.6.8.
>
>>
>> Result on Amber 11 compiled with GCC 4.6.2 on CentOS 5.7
>> ========================================================
>> Sander crashes "immediately".. presumably while parsing the `rest` file.
>> min2.out:
>>
>>> Error: Invalid atom or grouping specified in restraint.
>>> restraint = "torsion [:70.C1, :70.O5, :70.C5, :70.C4]"
>
> A quibble: sander stops with an error message, which is somewhat different
> than a "crash".
Right you are :-)
> Have you tried removing the spaces after the commas above?
> Can you tell if all "natural language" restraints fail, or just certain ones?
I will work on these things.
> We'll continue to try (using virtual machines) to reproduce the error.
> If anyone has easy access to CentOS 5.7 (or other Linux with gcc 4.6.2) they
> might try it. The original test case is attached.
I should clarify and add that CentOS 5.7 originally comes with GCC 4.1.2
(which makes certain test cases fail and is not recommended for Amber,
right?). Therefore, I had built the GCC 4.6.2 (which I had used to build
Amber(tools)) from source using GCC 4.1.2.
Now, it would definitely be good to reproduce this problem on a system
that natively comes with GCC 4.6.2, e.g. Fedora 16. I'll try to do so.
>
> ...thx...dac
>
>
>
>
> _______________________________________________
> 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 Mon Feb 20 2012 - 11:30:02 PST
