I asked about increasing it and was told that it can be done by editing
a value in one of the .h files or something, but the reason it wasn't
higher was that it didn't scale nicely, and they wanted to reword how it
I would definitely be interested in this, as we run video transcoding
(uses a LOT of cpu) on 24 or 32 core machines, and we'd like to pass
more of the cores into the VM.
You can bump it by modifying:
sys/amd64/include/vmm.h:#define VM_MAXCPU 16 /* maximum
virtual cpus */
... and rebuilding world+kernel.
Let us know how performance goes with larger values.
It will probably be a while before this value can be dynamic, but
that's the goal.
firstname.lastname@example.org mailing list
To unsubscribe, send any mail to