|
|
|
|
|
|
|
|
|
|
xen-devel
[Xen-devel] Re: x86_64 SMP support (status update)
* Nakajima, Jun (jun.nakajima@xxxxxxxxx) wrote:
> Chris Wright wrote:
> > Here's an updated patch that fixes that faulting. It uses gdt per
> > page like i386 is doing, still crashing upon schedule. Why does each
> > gdt entry need it's own page?
>
> Each gdt entry? I think gdt should be per vcpu, but I don't understand
> "each gdt entry". Can you please point to the code you are talking
> about?
Yes, the table is in arch/xen/x86_64/kernel/head.S (cpu_gdt_table, which
has 16 8 byte entries per cpu). However rather than using that table directly,
we must copy the entries into a per cpu page. Xen side has assumptions
regarding gdt being page sized (8 bytes * 512 entries, e.g. in
xen/arch/x86/mm.c in alloc_segdesc_page). That help?
thanks,
-chris
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel
|
<Prev in Thread] |
Current Thread |
[Next in Thread>
|
- [Xen-devel] RE: x86_64 SMP support (status update), Nakajima, Jun
- RE: [Xen-devel] Re: x86_64 SMP support (status update), Nakajima, Jun
- RE: [Xen-devel] Re: x86_64 SMP support (status update), Nakajima, Jun
- RE: [Xen-devel] RE: x86_64 SMP support (status update), Li, Xin B
- RE: [Xen-devel] RE: x86_64 SMP support (status update), Li, Xin B
|
|
|
|
|