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

Re: [Xen-devel] [PATCH] Make "xm mem-set" be lower bound on domX-min-mem



Bugger. :-)

But this is good to know! Obviously it stresses the need to keep track of the initial memory allocation, to set limits on subsequent shrinking. Fortunately we already do this in the (new) Xen CIM provider model by maintaining both the so-called 'recorded' memory allocation size/settings (aka initial size) as well as the 'current' memory allocation size/setting (eg after ballooning). So at least we have the data needed to enforce a lower limit.

thnx!

- Gareth

Dr. Gareth S. Bestor
IBM Linux Technology Center
M/S DES2-01
15300 SW Koll Parkway, Beaverton, OR 97006
503-578-3186, T/L 775-3186, Fax 503-578-3186

Inactive hide details for Keir Fraser <Keir.Fraser@xxxxxxxxxxxx>Keir Fraser <Keir.Fraser@xxxxxxxxxxxx>


          Keir Fraser <Keir.Fraser@xxxxxxxxxxxx>

          05/21/06 01:45 AM


To

Gareth S Bestor/Beaverton/IBM@IBMUS

cc

xen-devel@xxxxxxxxxxxxxxxxxxx

Subject

Re: [Xen-devel] [PATCH] Make "xm mem-set" be lower bound on domX-min-mem


On 20 May 2006, at 22:27, Gareth S Bestor wrote:

>  An absolute limit might be easier to handle - and expose to users -
> than a relative one, especially up-front in CIM where it exposes
> min/max limits on resource allocations. Or is it really <2% of
> whatever the original memory allocation is when things go to
> kabluwey... ?

Unfortunately so: some critical kernel data structures are sized
proportional to the maximum amount of memory that may be allocated to
the guest. These data structures do not shrink as you remove memory
from the guest.

 -- Keir


GIF image

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel

 


Rackspace

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