On Mon, 12 Dec 2005, Jeff Squyres wrote:
In short -- we ended up agreeing with you. :-)
Whew, I'm surprised given the enthusiasm which you showed when seeing Paul's code ! Really, I thought that you will choose Paul's code with only the conditions changed as expressed in my last e-mail... as to give satisfaction to both of us :-)
But I would like to stress the fact that my conclusions came out only from reading the kernel code; I only did some tests on the kernels that I have running at the moment (RHEL3 and RHEL4 on i386). So some real-world testing is still needed, especially on the kernels that were "different".
As a personal sidenote -- it sucks that we all had to do this much research to figure this out.
Well, I consider this to be a result of applying commercial interest to Linux. The distributions should not have backported those changes until a final, stable API was established, but when you want to be the first to claim "I support this and that"...
Thanks for your diligence in pestering us about this! :-)
Eh, don't mention it! I want Open MPI to work :-) -- Bogdan Costescu IWR - Interdisziplinaeres Zentrum fuer Wissenschaftliches Rechnen Universitaet Heidelberg, INF 368, D-69120 Heidelberg, GERMANY Telephone: +49 6221 54 8869, Telefax: +49 6221 54 8868 E-mail: bogdan.coste...@iwr.uni-heidelberg.de
