No x86 will not be getting a kernel cage, at least not yet. This
project is just integrating the ability to hot add cpus and memory to
a running system and thus no need for a kernel cage. There are future
phases in this project but the direction/implementation of those
phases has not been solidified yet.
--Mike
On Mar 8, 2010, at 12:54 PM, [email protected] wrote:
Does this mean x86 is getting a kernel cage?
-j
On Sun, Mar 07, 2010 at 08:22:59PM -0800, Michael Corcoran wrote:
Hi All,
It was suggested that I use this alias to try and hammer out a name
for
a new package I'm planning on delivering as part of the x86 cpu/
memory
hotplug project. I have a webrev of the packaging portions of things
available at:
http://cr.opensolaris.org/~mec/ips_local.2/
This new package will contain all of the bits needed to enable the
addition of cpus or memory to a running x86 system. This includes
the
acpihpd daemon, managed by the svc:/platform/i86pc/acpihpd:default
SMF
service, who's manifest is also included in the package, as well as
the
various dr based drivers which are needed to support this feature.
The current name I have for the package is:
system/dynamic-reconfiguration/i86pc
Other suggestions have been:
system/dynamic-reconfiguration/generic
system/dynamic-reconfiguration/apci-hotplug
system/dynamic-reconfiguration/x86
system/dynamic-reconfiguration
Some additional background which makes all of this a bit more
confusing :)
There already exist sparc packages for similar DR based features:
system/kernel/dynamic-reconfiguration/sun-fire-15000
system/kernel/dynamic-reconfiguration/sun-fire-880
system/kernel/dynamic-reconfiguration/ultra-enterprise-10000
With IPS, it seems like there should be no need for all of these
packages and a single system/kernel/dynamic-reconfiguration package
might make sense. I know IPS is smart enough to handle x86 and sparc
bits in the same package, but I don't know if it's smart enough to
distinguish between different sparc platforms. I would prefer not
to do
this consolidation myself, if it is a good idea, and instead file
an RFE
for it so that my project is not impacted by the extra testing this
would require in which case we'd go with the
system/dynamic-reconfiguration name.
So, please let me know your opinions on this.
Thanks,
Mike
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
_______________________________________________
pkg-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/pkg-discuss
