Re: CVS commit: src/sys/arch/xen/xen

[Manuel Bouyer <bouyer%antioche.eu.org@localhost>]

On Fri, Oct 17, 2008 at 11:06:55AM +0900, YAMAMOTO Takashi wrote:
> > On Fri, Oct 10, 2008 at 03:25:56PM +0900, YAMAMOTO Takashi wrote:
> > > hi,
> > > 
> > > > Module Name:    src
> > > > Committed By:   bouyer
> > > > Date:           Sat Sep 20 20:36:09 UTC 2008
> > > > 
> > > > Modified Files:
> > > >         src/sys/arch/xen/xen: privcmd.c
> > > > 
> > > > Log Message:
> > > > Fix IOCTL_PRIVCMD_MMAPBATCH: don't blindly map requested pages 
> > > > read/write
> > > > but use the map's protection bits, as the hypervisor may refuse 
> > > > read/write
> > > > mappings for some entries. Now suspend/resume of domUs should work
> > > > from a NetBSD dom0, provided that the domU's kernel supports it.
> > > > 
> > > > From Jean-Yves Migeon.
> > > > 
> > > > 
> > > > To generate a diff of this commit:
> > > > cvs rdiff -r1.29 -r1.30 src/sys/arch/xen/xen/privcmd.c
> > > > 
> > > > Please note that diffs are not public domain; they are subject to the
> > > > copyright notices on the relevant files.
> > > 
> > > what prevents the 'entry' pointer from getting invalid
> > > while you release the vm_map lock?
> > 
> > Ha yes, entry is a pointer and not a vm_map_entry. I guess the attached 
> > patch
> > would fix it ?
> > 
> > -- 
> > Manuel Bouyer <bouyer%antioche.eu.org@localhost>
> >      NetBSD: 26 ans d'experience feront toujours la difference
> > --
> 
> it should workaround the problem, yes.
> (i'm not sure if it's correct semantics for this ioctl.)

What do you think the semantics should be ?
AFAIK linux uses the permissions of the old pages for the new one.
We used to always map it read/write but Xen refuses it in some cases.

-- 
Manuel Bouyer, LIP6, Universite Paris VI.           
Manuel.Bouyer%lip6.fr@localhost
     NetBSD: 26 ans d'experience feront toujours la difference
--