Newsgroups: comp.parallel.mpi
From: thakur@abacus.mcs.anl.gov (Rajeev Thakur)
Subject: Re: TYPE_LB
Organization: MCS, Argonne National Laboratory
Date: Fri, 9 Aug 1996 23:2:40 GMT
Message-ID: <83963176010956@abacus.mcs.anl.gov>

In article <83963004710914@abacus.mcs.anl.gov>,
Rajeev Thakur <thakur@abacus.mcs.anl.gov> wrote:
>My understanding of the Standard, pg. 71, lines 26-28, is that 
>the LB of the second type is 0. This is because eric_t is a derived
>datatype and is therefore not of type LB itself. So the condition 
>"such that type_j = LB" does not hold. 
>
>Rajeev

Well, that is not quite right. From a second reading, I think LB
is indeed 100 in this case.

Rajeev


>In article <4ufbgd$22j@murrow.corp.sgi.com>,
>Eric Salo <salo@mrjones.engr.sgi.com> wrote:
>>Let's say that I create a new MPI datatype, eric_t, and I explicitly
>>set the LB to 0 with a TYPE_LB. I then create a second datatype, which
>>has a single integer at displacement 0 and an eric_t at offset 100.
>>
>>Question: What is the LB of that second type?
>>
>>If I'm reading the standard correctly, it is 100, which seems really
>>silly...
>>
>>Eric Salo         Silicon Graphics Inc.             "Do you know what the
>>(415)933-2998     2011 N. Shoreline Blvd, 8U-802     last Xon said, just
>>salo@sgi.com      Mountain View, CA   94043-1389     before he died?"




