Hi Dan,
You are right, sorry. What I do is I have a separate tree i use for
development, so whenever I run the tests I'm using the
sander/sander.MPI from $AMBERHOME/bin. When switching between
development/production versions, I just change AMBERHOME.
Sorry for the confusion.
Gustavo Seabra
Professor Adjunto
Departamento de Química Fundamental
Universidade Federal de Pernambuco
Fone: +55-81-2126-7417
On Sun, Mar 7, 2010 at 11:45 AM, Daniel Roe <daniel.r.roe.gmail.com> wrote:
> Hi Gustavo,
>
> Are you sure you aren't just running the sander.MPI in your $AMBERHOME/bin
> directory? If I do something like
>
> export TESTsander=ThisWontWork
>
> and run:
>
> make test
>
> I indeed get something like:
>
> cd cytosine && ./Run.cytosine
> ThisWontWork: Command not found.
> ./Run.cytosine: Program error
> make: *** [test.sander.BASIC] Error 1
>
> However, if I do the same thing and run:
>
> make test.parallel
>
> (after setting DO_PARALLEL and setting up MPI etc) I get:
>
> cd cytosine && ./Run.cytosine
> diffing cytosine.out.save with cytosine.out
> PASSED
>
> the reason being that in the makefile the parallel tests are set up in this
> way:
>
> test.sander.BASIC.MPI:
> export TESTsander=$(AMBERHOME)/exe/sander.MPI; $(MAKE) -k
> test.sander.BASIC
> export TESTsander=$(AMBERHOME)/exe/sander.MPI; cd 4096wat &&
> ./Run.column_fft
> ...
>
> So whatever you set as TESTsander gets overwritten to be what is in
> $AMBERHOME/exe. I have done a cvs update on the makefile and its current -
> is it set up differently in your makefile? Or am I missing your point?
>
> -Dan
>
> On Sat, Mar 6, 2010 at 10:58 AM, Gustavo Seabra <gustavo.seabra.gmail.com>wrote:
>
>> On Sat, Mar 6, 2010 at 1:36 AM, Daniel Roe <daniel.r.roe.gmail.com> wrote:
>> > I will take responsibility for this. I was testing the code for Carlos,
>> but
>> > I forgot that the test Makefile overwrites the TESTsander variable for
>> the
>> > parallel tests. So when I ran the parallel tests I wasn't testing what I
>> set
>> > TESTsander to, I was testing the default ($(AMBERHOME)/exe/sander.MPI).
>> As a
>> > bonus the error doesn't show up when the fdist subroutine isnt called
>> (i.e
>> > REM runs etc with 1 thread per group still passed the parallel tests).
>> Sorry
>> > everyone.
>>
>> That's weird... Something else may be broken. For me, it has always
>> worked to set TESTsander to point to any executable I wanted, (serial
>> or parallel, sander or pmemd) and the tests used the correct
>> executable.
>>
>> > Can anyone suggest an easy way to modify the parallel tests so that you
>> can
>> > set a variable to point to the binary you want to test (like TESTsander
>> in
>> > the serial tests). I suppose you could have a SANDERMPI variable in the
>> > makefile or something that TESTsander gets set to, like
>>
>> If just using:
>>
>> $ export TESTsander=/path/to/the/parallel/executable/I/want/to/test
>>
>> isn't working, something else must be wrong.
>>
>> Gustavo Seabra
>> Professor Adjunto
>> Departamento de Química Fundamental
>> Universidade Federal de Pernambuco
>> Fone: +55-81-2126-7450 Ext. 5021
>>
>> _______________________________________________
>> AMBER-Developers mailing list
>> AMBER-Developers.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber-developers
>>
> _______________________________________________
> AMBER-Developers mailing list
> AMBER-Developers.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber-developers
>
_______________________________________________
AMBER-Developers mailing list
AMBER-Developers.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber-developers
Received on Mon Mar 08 2010 - 04:00:04 PST
