Open MPI logo

Open MPI Development Mailing List Archives

  |   Home   |   Support   |   FAQ   |  

This web mail archive is frozen.

This page is part of a frozen web archive of this mailing list.

You can still navigate around this archive, but know that no new mails have been added to it since July of 2016.

Click here to be taken to the new web archives of this list; it includes all the mails that are in this frozen archive plus all new mails that have been sent to the list since it was migrated to the new archives.

Subject: Re: [OMPI devel] ibcm private header file
From: Jeff Squyres (jsquyres_at_[hidden])
Date: 2008-07-23 09:12:16


This has now been fixed as of https://svn.open-mpi.org/trac/ompi/changeset/18995
; MTT should [finally] run tonight without problems.

On Jul 22, 2008, at 12:38 PM, Jeff Squyres wrote:

> On Jul 22, 2008, at 12:32 PM, Roland Dreier wrote:
>
>>> So I think using byteorder.h is not a good idea (note the warning).
>>> How about just having two #defines, picking the Right one based on
>>> WORDS_BIGENDIAN?
>>
>> On Linux, the public <byteswap.h> header might have what you need.
>> For that matter <infiniband/arch.h> has htonll() defined.
>
> The problematic line is:
>
> #define IB_CM_ASSIGN_SERVICE_ID __cpu_to_be64(0x0200000000000000ULL)
>
> There's no reason this has to be byte-swapped at compile time,
> right? Using htonll or OMPI's hton64() should be sufficient, right?
>
> --
> Jeff Squyres
> Cisco Systems
>
> _______________________________________________
> devel mailing list
> devel_at_[hidden]
> http://www.open-mpi.org/mailman/listinfo.cgi/devel

-- 
Jeff Squyres
Cisco Systems