Re: [Xen-devel] [PATCH v2 17/19] x86/hvm: Avoid __hvm_copy() raising #PF behind the emulators back

[Paul Durrant <Paul.Durrant@xxxxxxxxxx>]

> -----Original Message-----
> From: Andrew Cooper
> Sent: 28 November 2016 12:58
> To: Paul Durrant <Paul.Durrant@xxxxxxxxxx>; Xen-devel <xen-
> devel@xxxxxxxxxxxxx>
> Cc: Jan Beulich <JBeulich@xxxxxxxx>; Tim (Xen.org) <tim@xxxxxxx>; Jun
> Nakajima <jun.nakajima@xxxxxxxxx>; Kevin Tian <kevin.tian@xxxxxxxxx>
> Subject: Re: [PATCH v2 17/19] x86/hvm: Avoid __hvm_copy() raising #PF
> behind the emulators back
> 
> On 28/11/16 11:56, Paul Durrant wrote:
> >> -----Original Message-----
> >> From: Andrew Cooper [mailto:andrew.cooper3@xxxxxxxxxx]
> >> Sent: 28 November 2016 11:14
> >> To: Xen-devel <xen-devel@xxxxxxxxxxxxx>
> >> Cc: Andrew Cooper <Andrew.Cooper3@xxxxxxxxxx>; Jan Beulich
> >> <JBeulich@xxxxxxxx>; Paul Durrant <Paul.Durrant@xxxxxxxxxx>; Tim
> >> (Xen.org) <tim@xxxxxxx>; Jun Nakajima <jun.nakajima@xxxxxxxxx>;
> Kevin
> >> Tian <kevin.tian@xxxxxxxxx>
> >> Subject: [PATCH v2 17/19] x86/hvm: Avoid __hvm_copy() raising #PF
> behind
> >> the emulators back
> >>
> >> Drop the call to hvm_inject_page_fault() in __hvm_copy(), and require
> >> callers
> >> to inject the pagefault themselves.
> >>
> >> No functional change.
> > That's not the way it looks on the face of it. You've indeed removed the 
> > call
> to hvm_inject_page_fault() but some of the callers now call
> x86_emul_pagefault(). I'd call that a functional change... clearly the change
> you intended, but still a functional change.
> 
> Hmm - I suppose I am confusing no functional change in the hypervisor
> with no functional change as observed by a guest.
> 

Yes, I was thinking from the PoV of someone looking at this patch years later 
and saying 'hang on a minute...'. Saying 'no guest-observable behavioural 
change' is much clearer I think.

  Paul

> ~Andrew

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
https://lists.xen.org/xen-devel