[mpiwg-abi] asynchronous voting to determine what requires more discussion (ie a meeting)

[Christoph Niethammer]

Hello Jeff,

Here my answers maybe repeating some of my comments from the last telco:
1. MPI Status: The current proposal looks good to me

2. Buffer addresses - I do not have any objections

3. Bsend overhead - I don't feel competent enough for a found comment here

4. Looking at the string constants I am a bit concerned about the limit of
MPI_MAX_INFO_VAL choice of only 1024.
We have, e.g., the Info keys "io_node_list" or "filename" where I could imagine that we could run into problems.
The "io_node_list" key could be a problem on lager systems that come with more than 100 cabinets - assuming for hostnames typically around 9 characters/15 byte string IP + 1 comma for separation.
The "filename" key could be a problem if we interpret the description in the standard - "This hint specifies the file name used when the file was opened." - to include the path.
This is unlimited in length for nearly all common filesystems (thinking ext4, lustre, ...). However, Linux will limit the combined filenames/path length to 4096 characters for us.

Best
Christoph


----- Original Message -----
From: "Jeff Hammond" <jeff.science at gmail.com>
To: "mpiwg-abi" <mpiwg-abi at lists.mpi-forum.org>
Sent: Monday, 24 April, 2023 09:10:33
Subject: [mpiwg-abi] asynchronous voting to determine what requires more discussion (ie a meeting)

I do not think we need a meeting unless things are complicated or
controversial.

I posted a Google Sheet link on Slack so we can vote on a bunch of things
asynchronously.  If we don't have a consensus, I'll schedule a meeting.

1. MPI_Status object definition.
2. Buffer address constants.
3. BSEND overhead
4. String length constants

We discussed string constants a bunch already but many people didn't
comment, so it is not clear what the overall group attitude is.

I can share the Google Sheet link if necessary, but I won't post it on a
public mailing list.

Thanks,

Jeff

-- 
Jeff Hammond
jeff.science at gmail.com
http://jeffhammond.github.io/

-- 
mpiwg-abi mailing list
mpiwg-abi at lists.mpi-forum.org
https://lists.mpi-forum.org/mailman/listinfo/mpiwg-abi