[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: xen_acpi_processor driver is bound to dom0 vcpu count



On Fri, May 23, 2025 at 12:22:44AM +0100, me@xxxxxxxxx wrote:
> I think I’ve just uncovered a rather nasty bug in the xen_acpi_processor 
> driver in dom0. If the vcpu count in dom0 is set to anything other than the 
> exact number of physical cores, the xen_acpi_processor kernel driver will 
> fail to upload the C-state information for those cores to Xen, resulting in 
> Xen never knowing about the C-states, which significantly impacts battery 
> life on mobile platforms.
> 

While I dislike this bug, I don't think it qualifies as "nasty" since it
doesn't severely break anything.  Your system ends up using rather more
power than it should, but everything else continues to works as expected.

The simplest workaround is to not use dom0_max_vcpus, and instead do
`xl vcpu-set 0 #` to offline vCPUs after boot.  This leaves domain 0
allocating per-processor memory for unused processors, but does reduce
domain 0's vCPU use.

I believe this is partially bugs with Xen's ACPI support, but also an
issue with the Linux kernel module.  I've observed patches going by which
appeared aimed towards this problem, but for now this remains.


-- 
(\___(\___(\______          --=> 8-) EHM <=--          ______/)___/)___/)
 \BS (    |         ehem+sigmsg@xxxxxxx  PGP 87145445         |    )   /
  \_CS\   |  _____  -O #include <stddisclaimer.h> O-   _____  |   /  _/
8A19\___\_|_/58D2 7E3D DDF4 7BA6 <-PGP-> 41D1 B375 37D0 8714\_|_/___/5445





 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.