It seems that both mx0 & mx1.topicbox.com are down this morning so the lists will be a tad quiet!

So in lieu of that, if anyone has some time I'd appreciate a review on illumos 18082 (fenix).

BUG 18082: bhyve virtio-console: defer PORT_NAME until guest signals PORT_READY (In Progress)

↳ illumos.org/issues/18082 | code.illumos.org/c/illumos-gate/+/4721

Eeesh, I'm looking in socktpi.c and sockstr.c.... is it just me or should we be cv_wait_sig()-ing here a lot more than we do today?

Have $CUSTOMER with locked processes in sotpi_connect() where it's stuck on cv_wait() because the peer got killed before it could return a T_*_ACK primitive.

you're description is making me twitch :)

just mentioning those files is making me twitch

I'm pretty damned sure this all needs to be more tolerant of signals ONLY for AF_UNIX-y things.

I'm also seeing uses of cv_wait_sig_swap() that suggest... well... just using cv_wait_sig() may be hazardous in some cases.

Speaking of reviews then: code.illumos.org/c/illumos-gate/+/4723

→ CODE REVIEW 4723: 18084 Update to tzdata 2026b (NEW) | illumos.org/issues/18084

For any cases where someone wants to RTI something, please just drop a direct e-mail to me or one of the other core team members and we'll try to take care of it. We've reached out to topicbox, but are waiting to hear back.

danmcd: _swap() means that the thread can get swapped out while it's asleep

we only _swap_ threads under extremes of pressure that are rarely seen in practice, and that at least a couple people (maybe it was you jclulow?) have suspected don't work

danmcd: but that's the same code path that is removing processes wholesale in the (pre-virtual memory) sense of "swapping"

Thank you for that reality check.

seagl.org/news/2026/04/24/CfP-Open in case anyone wants to present at a free conference in Seattle in November.

code.illumos.org/c/illumos-gate/+/4725 is also a thing folks may be interested in that was something I had in the background for ixi.

→ CODE REVIEW 4725: 17868 ps could know about microstates (NEW) | illumos.org/issues/17868