As suggested by bcr@ and others shorten examples output:

I've tried opposite approach of adding LOWPRIO flag instead and using it only for background operations in few places, and marking BIOs without it high-priority in ATA/SCSI. But while testing it I've noticed that disk random IOPS drop to almost non-NCQ level on a mix of different priorities. And I am measuring the same on both WD and HGST. I don't understand what is going on there, may be I am missing something, but that is unacceptable trade-off to me. I've uploaded my present patch in case somebody wish to play, but probably won't commit it in this state.

Rebase and fix libreoffice

I only know that I know nothing...

Good to go.

In D26975#602036, @cyprien_cypou.net wrote:

I think from GICv2 it would be assumed that the OS could write to the SGI bits. Software cannot know all existing implementations.

Looks good to me. Thanks!

I agree. Maybe you can shorten the examples to a few lines and indicate the rest with ... without loosing too much context?

Andrew, I'm sorry but I don't quite understand why we crossing.
I will try to repeat my objections in a more compact/consistent form.
I have 3 objections:

If doing such reorg, it might make sense to de-commission syscall_thread_enter/exit, renaming underscored functions back to syscall_thread_.... Then you would fetch SY_THR_STATIC once and cache it in local var.

Set domain.end address (max guest addr + 1) to the VM_MAXUSER_ADDR

Add missing chunk.

Fix thinko.

Regen.

Regen.

Hi, any thoughts on this?

Approved with gusto :)

Approved :)

If booting from EFI, updating gptzfsboot won't do anything. We likely need additional documentation on actually updating the contents of the ESP as well.

Excellent, consistency is king. ;-)

Workable version.

Ha, my calendar says the same. What a coincidence. Let's fix it.

Upon further inspection, I really like luaunit -- it can do TAP output, so it would be almost no work to hook up to the existing infrastructure. The default output format is actually really nice for diagnosing problems (see: https://luaunit.readthedocs.io/en/latest/#output-formats), which makes it more enticing.

In D26998#602308, @debdrup wrote:

This is an exceptionally good idea that I should have thought of, since I've been following the calendar discussion on the mailing lists.
However, the dates aren't quite right - the "new" official deadlines are listed in ./doc/en_US.ISO8859-1/htdocs/news/status/README and basically amount to being the last day of a given quarter (barring the exception for late reports and portmgr@).

Update dates