Messages in this thread | | | Date | Thu, 18 Feb 2010 09:26:07 -0800 | From | "H. Peter Anvin" <> | Subject | Re: [PATCH 34/35] x86: use num_processors for possible cpus |
| |
On 02/17/2010 06:38 PM, Yinghai Lu wrote: >> >> I'm confused by this one. This would seem to mean that unless you're >> specifying possible_cpus= then you are not treating anything as >> hotpluggable. >> >> This is clearly wrong, and it would appear to go the wrong direction in >> terms of what is safe. >> >> What am I missing here? > > per_cpu setup will allocate some mem for every cpu according to possible cpus. >
Yes, and I have repeatedly requested that we allocate the memory on the first up of a disabled CPU rather than eagerly, but in *most* configurations the amount is relatively small.
> no spec says that we should disable cpus entries in MADT as hotplug cpus.
Reality seems to, though. Consider the bug report that led to patch 681ee44d40d7c93b42118320e4620d07d8704fd6 for example.
> also how many system that support hotplug cpus there?
In terms of physical machines, not all that many; however, it is getting common to use CPU hotplug for virtual environments so they can be dynamically scaled.
-hpa
| |