tags 242366 pending thanks This was fixed in 1.2.5.3-2, which was uploaded on December 22 and is waiting in the NEW queue. In the meantime, you can get it at http://lyre.mit.edu/~powell/mpich/
On Thu, 2004-08-05 at 15:04 -0400, Adam C Powell IV wrote: > Hi Eray, it's good to hear from you! > > On Fri, 2004-07-30 at 14:57, Eray Ozkural (exa) wrote: > > Package: mpich > > Version: 1.2.5.2-1 > > Severity: normal > > > > > > I think it should be sufficient that the client code has > > > > #include <mpi.h> > > > > Without having to specify the directory under which mpi.h resides. > > I don't think this is standard practice. For example, PETSc includes > the entire path to mpi.h in a -I compile flag, I think this is the way > it's supposed to be done. If you have a counterexample from the mpich > suite, let me know and I'll try to change things so it works. > > > Also seems to be missing is the mpi++.h file. hello++.cc (in examples/pi) > > would not compile in this situation! > > D'oh! I'll try to fix that. Thanks. > > I'm a bit over my head now, the bugs in mpich are piling up and I'm > trying to fix them all. Should be done by sometime next week. > > Thanks for the report, > > -Adam P. -Adam -- GPG fingerprint: D54D 1AEE B11C CE9B A02B C5DD 526F 01E8 564E E4B6 Welcome to the best software in the world today cafe! http://www.take6.com/albums/greatesthits.html -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
