Messages in this thread | | | Date | Wed, 05 Mar 2008 09:28:51 -0800 | From | "H. Peter Anvin" <> | Subject | Re: [RFC] use realmode code to reserve end-of-conventional-memory to 1MB |
| |
Eduardo Habkost wrote: >>> >>> I am aware that it would take more work to tell all kernel code that it >>> shouldn't look for BIOS data on this region when running as a domU guest, >>> but it seems that it would be a better solution. >>> >> No, the right thing is for Xen to not try to map RAM in this area. > > Why? If the Xen host is telling us there is valid RAM in this area, > why can't we use it? > > Existing Xen hosts use those physical address ranges for RAM. We can't > fix this on the e820 map on the guest side without making otherwise-valid > RAM unused. >
x86 has a number of historical assumptions, both in kernel and in userspace (consider dmidecode!) Trying to go against them is a losing proposition, *AND* a headache for the maintainers, which will have to consider "oh yes, and Xen does this dumb thing which goes against what all the hardware does."
-hpa
| |