pisses me off to no end

[Jed Brown]

Barry Smith wrote:
> 
> On Jun 10, 2009, at 5:32 PM, Jed Brown wrote:
> 
>> Matthew Knepley wrote:
>>
>>> I am not opposed to this, but it ONLY works for mpich. Is there any
>>> generality for mpicc?
>>
>> FWIW, the -show option also works for Open MPI, HP MPI, and Quadrics MPI
>> (i.e. all that I've had recent contact with).  The standard doesn't
>> specify anything (not even the presence of wrapper compilers).
>                                     Which are the worst possible
> solution perpetuated on the community by MPI! Why doesn't have package
> in the world provide compiler wrappers, I'm sure that would work great :-(

Heh, then we end up with wrappers like h5pcc because -lhdf5_hl -lhdf5 is
too hard.  MPI is a bit special (not many "libraries" requires a special
launcher program), but although I use MPI wrappers (it's a pain not to),
they are a shitty solution.

Jed

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 260 bytes
Desc: OpenPGP digital signature
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20090611/1997a7af/attachment.sig>