Open MPI logo

Open MPI Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Development mailing list

Subject: Re: [OMPI devel] XML request
From: Ralph Castain (rhc_at_[hidden])
Date: 2009-09-01 22:14:26


Ah, if only it were that simple!

Unfortunately, we need the xml stuff totally separated from anything
else that might come out via stdout or stderr. That's the problem Greg
and I are trying to solve.

The -xml-file option appears to be the only solution at the moment.

Thanks for the thought, though...

On Sep 1, 2009, at 11:43 AM, Neil Ludban wrote:

>> Date: Mon, 31 Aug 2009 20:26:20 -0500
>> From: Greg Watson <g.watson_at_[hidden]>
>> Subject: Re: [OMPI devel] XML request
>> To: Open MPI Developers <devel_at_[hidden]>
>> Message-ID: <7C3BD80B-2E0D-4188-BA0A-9B4963AE2964_at_[hidden]>
>> Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes
>>
>> Hey Ralph,
>>
>> Unfortunately I don't think this is going to work for us. Most of the
>> time we're starting the mpirun command using the ssh exec or shell
>> service, neither of which provide any mechanism for reading from file
>> descriptors other than 1 or 2. The only alternatives I see are:
>>
>> 1. Provide a separate command that starts mpirun at the end of a pipe
>> that is connected to the fd passed using the -xml-fd argument. This
>> command would need to be part of the OMPI distribution, because the
>> whole purpose of the XML was to provide an out-of-the-box experience
>> when using PTP with OMPI.
>>
>> 2. Implement an -xml-file option, but I could write the code for you.
>>
>> 3. Go back to limiting XML output to the map only.
>>
>> None of these are particularly ideal. If you can think of anything
>> else, let me know.
>
> I don't think there's a nicer way to do this:
>
> if (xml_output) {
> FILE *fxml = fopen("/dev/stdout", "w");
> freopen("/dev/stderr", "w", stdout);
> }
>
> You could also replace file descriptors 1 and 2 with pipes and
> read data back in, convert to XML, and send it on to the dup'd
> fd 1. But this would cause problems if there were any debug
> printfs that were timing sensitive.
> _______________________________________________
> devel mailing list
> devel_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/devel