After re-reading Ben's replies and following the link to the discussion
in the Squeak VM list I think I've finally understood this correctly.
I've updated the text based on Ben's input to:
--
This VM uses a separate heartbeat thread to update its internal clock
and handle events.�� For best operation, this thread should run at a
higher priority, however the VM was unable to change the priority.�� The
effect is that heavily loaded systems may experience some latency
issues.�� If this occurs, please create the configuration file
in /etc/security/limits.d/ as shown below:
cat <<END | sudo tee /etc/security/limits.d/VMNAME.conf
*�� �� �� ��hard�� �� rtprio�� 2
*�� �� �� ��soft�� �� rtprio�� 2
END
and report to the appropriate mailing list for your smalltalk dialect
whether this improves behaviour.
I'm seeing the message printed out three times, which is a little
strange, but haven't looked in to it.
There are two issues in fogbugz:
For the VM: https://pharo.fogbugz.com/f/cases/19890/Linux-Heartbeat- VM-text-and-instructions
For the build scripts: https://pharo.fogbugz.com/f/cases/19891/Suggested- improvement-for-linux-CONFIG- file
Given how close the 6.0 release date is and even though there is a
bug in the instructions for enabling elevated thread privilege, I've
marked this for Pharo 7.0.�� Of course, if Eliot or someone else decides
to put it in 6.0, that's fine by me.�� As before, if there aren't any
conflicting comments I'll submit the pull requests in a couple of days.
Thanks again for all your assistance,
Alistair