Open MPI logo

Open MPI Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |   all Development mailing list

Subject: Re: [OMPI devel] Should visibility and memchecker abort configure?
From: Ralph Castain (rhc_at_[hidden])
Date: 2008-10-06 14:01:58


Not exactly - see below

On Oct 6, 2008, at 11:56 AM, Jeff Squyres wrote:

> I think the real issue here is that --enable-debug (or the presence
> of the .svn or .hg directories) *implies* several other options,
> such as --enable-visibility and --enable-memchecker.
>
> As I understand it: the user has *not* specifically asked for --
> enable-visibility, but is getting a failure if it can't be delivered
> because --enable-debug was specified. Is that right? If so, that's
> downright weird -- because I configure/compile with the PGI
> compilers with --enable-debug and simply get a build that does not
> include visibility (i.e., "ompi_info | grep visibil" results in
> "Symbol visibility support: no") -- the configure/build does not
> abort.

Not quite. In this case, we specifically did include --enable-
visibility so that it would be there for those compilers that support
it. We didn't realize that visibility was going to be included for a
debug build even if we -didn't- request it.

>
>
> Additionally, I agree that if the memchecker/valgrind component
> cannot deliver what it should, it should disable itself silently/
> without error *unless* the valgrind component was specifically
> requested (which, in this case, it sounds like it was not). So if
> we're not doing that, it's a bug.

Yes, in this case it is a bug. Shiqing is on holiday, but already
contacted me about it. We'll deal with it upon his return.

>
>
>
>
> On Oct 5, 2008, at 5:15 AM, Ralf Wildenhues wrote:
>
>> Hello,
>>
>> if you allow me my 2 cents:
>>
>> At configure time, it is possible to distinguish between several
>> different user inputs:
>> - the user typed --enable-foo,
>> - the user typed --disable-foo or --enable-foo=no,
>> - the user typed --enable-foo=ARG (ARG is available for further
>> inspection),
>> - the user used none of the above.
>>
>> IIUC you have already sorted out the visibility issue by using the
>> last,
>> and the memchecker issue is about having an incompatible version
>> installed. One typical semantics would be to not try to use the
>> feature
>> at all if --disable-foo was explicitly passed.
>>
>> Hope that helps.
>>
>> Cheers,
>> Ralf
>> _______________________________________________
>> devel mailing list
>> devel_at_[hidden]
>> http://www.open-mpi.org/mailman/listinfo.cgi/devel
>
>
> --
> Jeff Squyres
> Cisco Systems
>
> _______________________________________________
> devel mailing list
> devel_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/devel