<div dir="ltr">Hi All,<br><div><br></div><div>Wanted to share an update on this reading. I added an update to the proposal yesterday (before cutoff) that introduces an explicit "default" memory kind. My hope is that this will address several concerns that were raised at the virtual meeting. Namely,</div><div><br></div><div>1. Use cases where users want to add a kind in addition to the default were challenging/confusing and are now easy ("default;additional_kinds")</div><div>2. The default kind must include "system" and "mpi". After re-reading Section 9.2, I believe this is not introducing a new constraint.</div><div>3. MPI may not ignore a request to support the default kind. This relieves users from checking for basic support they have relied on for many years.</div><div><br></div><div>If you've already read the proposal in preparation for the meeting, you can find these changes in commit 80a67b3.</div><div><br></div><div>Cheers,</div><div> ~Jim.</div><div><br></div></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Fri, Feb 17, 2023 at 5:38 PM Jim Dinan <<a href="mailto:james.dinan@gmail.com">james.dinan@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr">Hi All,<br><div><br></div><div>I would like to read the Memory Allocation Kind Info proposal at the upcoming Forum meeting:</div><div><br></div><div><a href="https://github.com/mpi-forum/mpi-issues/issues/580" target="_blank">https://github.com/mpi-forum/mpi-issues/issues/580</a></div><div><br></div><div> ~Jim.</div></div>
</blockquote></div>