linux-kernel - Re: [PATCH] xen: Introduce 'xen_nopv' to disable PV extensions for HVM guests.

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <201407112309.s6BN9MFb016118@aserz7022.oracle.com>
Date:	Fri, 11 Jul 2014 19:09:02 -0400
From:	Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>
To:	"H. Peter Anvin" <hpa@...or.com>
Cc:	xen-devel@...ts.xenproject.org, konrad@...nel.org,
	boris.ostrovsky@...cle.com, linux-kernel@...r.kernel.org,
	david.vrabel@...rix.com
Subject: Re: [PATCH] xen: Introduce 'xen_nopv' to disable PV extensions for
 HVM guests.


On Jul 11, 2014 6:52 PM, "H. Peter Anvin" <hpa@...or.com> wrote:
>
> On 07/11/2014 12:53 PM, konrad@...nel.org wrote: 
> > 
> > diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt 
> > index c1b9aa8..5dcfa6e 100644 
> > --- a/Documentation/kernel-parameters.txt 
> > +++ b/Documentation/kernel-parameters.txt 
> > @@ -3691,6 +3691,10 @@ bytes respectively. Such letter suffixes can also be entirely omitted. 
> >  Disables the ticketlock slowpath using Xen PV 
> >  optimizations. 
> >  
> > + xen_nopv [X86,XEN] 
> > + Disables the PV optimizations forcing the guest to run 
> > + as generic HVM guest with no PV drivers. 
> > + 
> >  xirc2ps_cs= [NET,PCMCIA] 
> >  Format: 
> >  <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]] 
> > diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c 
> > index ffb101e..eb82239 100644 
> > --- a/arch/x86/xen/enlighten.c 
> > +++ b/arch/x86/xen/enlighten.c 
> > @@ -1826,8 +1826,19 @@ static void __init xen_hvm_guest_init(void) 
> >  xen_hvm_init_mmu_ops(); 
> >  } 
> >  
> > +static bool xen_nopv = false; 
> > +static __init int xen_parse_nopv(char *arg) 
> > +{ 
> > +       xen_nopv = true; 
> > +       return 0; 
> > +} 
> > +early_param("xen_nopv", xen_parse_nopv); 
> > + 
> >  static uint32_t __init xen_hvm_platform(void) 
> >  { 
> > + if (xen_nopv) 
> > + return 0; 
> > + 
> >  if (xen_pv_domain()) 
> >  return 0; 
> >  
> > @@ -1836,6 +1847,8 @@ static uint32_t __init xen_hvm_platform(void) 
> >  
> >  bool xen_hvm_need_lapic(void) 
> >  { 
> > + if (xen_nopv) 
> > + return false; 
> >  if (xen_pv_domain()) 
> >  return false; 
> >  if (!xen_hvm_domain()) 
> > 
>
> Any reason to not simply let xen_pv_domain() return false here?

It does return false already.

Did you mean to collapse them together?
>
> Also, the latter seems a bit odd? 
>
> -hpa 
>