lkml.org 
[lkml]   [2012]   [Oct]   [24]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
SubjectRe: [PATCH 05/10] xen/pvh: Extend vcpu_guest_context, p2m, event, and XenBus.
From
Date
On Tue, 2012-10-23 at 19:12 +0100, Konrad Rzeszutek Wilk wrote:
[...]
> + /* GUEST_GDTR_BASE and */
> + ctxt->u.pvh.gdtaddr = (unsigned long)gdt;
> + /* GUEST_GDTR_LIMIT in the VMCS. */
> + ctxt->u.pvh.gdtsz = (unsigned long)(GDT_SIZE - 1);
> +
> [...]
> - ctxt->gdt_frames[0] = gdt_mfn;
> - ctxt->gdt_ents = GDT_ENTRIES;
> + ctxt->u.pv.gdt_frames[0] = gdt_mfn;
> + ctxt->u.pv.gdt_ents = GDT_ENTRIES;

I guess I've been told and forgotten but why does this need to differ
between PV and PVH? Can the hypervisor not take the gfn in gdt_frames[0]
= gdt_mfn and figure out the gdtaddr from it? Is this because n the PV
case the H/V loads the GDT with the address of its own mapping of the
gdt_frames but in the PVH case we have no such mapping because the
pagetables don't have a Xen region in them?

It's worthy of a comment in any case.

[...]
> + /* PVH TBD/FIXME: future work */
[...]
> + /* PVH: TBD/FIXME: debug and fix eio map to work with pvh */

Do we really need these TODOs inline in the code? Especially in generic
code.

Ian.



\
 
 \ /
  Last update: 2012-10-26 09:01    [W:0.087 / U:1.400 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site