SPDK Jenkins CI system will be shut down
Scheduled planned electrical and network works on campus.
Shutdown is planned for July 24th 2:00 PM GMT to July 27th 8:00AM GMT.
How does that affect us?
CI will be unable to pick up changes and perform tests during this time.
is NVMe hotplug functionality as implemented limited to PCIe transport or
does it also work for other transports? If it's currently PCIe only, are
there any plans to extend the support to RDMA/TCP?
Just a quick reminder about the importance of doing code reviews. Code reviews are the single best way to keep our code quality high *before* it lands in the SPDK repo. Although it is indeed the case that it’s the maintainer’s jobs to approve patches before they are merged, doing code reviews to keep software quality high is everyone’s job ☺
I understand you are the author of this, I didn't find any
documentation on that in the spdk docs. Could you prepare a quick memo
and post it here and hopefully later have something more official?
For never-sleeping framework like spdk/dpdk there's always the kinda
critical question of the break-down to idle (we didn't get anything
new to process) vs busy (we have some work to do) %% (e.g portions of
a second when sampled repeatly) -- when I looked on the idle/busy
columns the numbers were in us but it seems they didn't sum to 1M. So
what is expected there?