[PATCH] x86/nested HAP: don't BUG() on legitimate error

[PATCH] x86/nested HAP: don't BUG() on legitimate error

[Jan Beulich]

[-- Attachment #1: Type: text/plain, Size: 498 bytes --]

set_p2m_entry() can fail without there being a bug in the code - crash
the domain rather than the host in that case.

Signed-off-by: Jan Beulich <jbeulich@suse.com>

--- a/xen/arch/x86/mm/hap/nested_hap.c
+++ b/xen/arch/x86/mm/hap/nested_hap.c
@@ -133,7 +133,7 @@ nestedhap_fix_p2m(struct vcpu *v, struct
         gdprintk(XENLOG_ERR,
 		"failed to set entry for %#"PRIx64" -> %#"PRIx64"\n",
 		L2_gpa, L0_gpa);
-        BUG();
+        domain_crash(p2m->domain);
     }
 }
 




[-- Attachment #2: x86-nHAP-fix-p2m-failure.patch --]
[-- Type: text/plain, Size: 543 bytes --]

x86/nested HAP: don't BUG() on legitimate error

set_p2m_entry() can fail without there being a bug in the code - crash
the domain rather than the host in that case.

Signed-off-by: Jan Beulich <jbeulich@suse.com>

--- a/xen/arch/x86/mm/hap/nested_hap.c
+++ b/xen/arch/x86/mm/hap/nested_hap.c
@@ -133,7 +133,7 @@ nestedhap_fix_p2m(struct vcpu *v, struct
         gdprintk(XENLOG_ERR,
 		"failed to set entry for %#"PRIx64" -> %#"PRIx64"\n",
 		L2_gpa, L0_gpa);
-        BUG();
+        domain_crash(p2m->domain);
     }
 }
 

[-- Attachment #3: Type: text/plain, Size: 126 bytes --]

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

Re: [PATCH] x86/nested HAP: don't BUG() on legitimate error

[Andrew Cooper]

[-- Attachment #1.1: Type: text/plain, Size: 890 bytes --]

On 11/04/14 11:28, Jan Beulich wrote:
> set_p2m_entry() can fail without there being a bug in the code - crash
> the domain rather than the host in that case.

It would be useful to indicate which conditions.

>
> Signed-off-by: Jan Beulich <jbeulich@suse.com>

Irrespective of the conditions causing a failed p2m update, this is
certainly a good change

Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>

>
> --- a/xen/arch/x86/mm/hap/nested_hap.c
> +++ b/xen/arch/x86/mm/hap/nested_hap.c
> @@ -133,7 +133,7 @@ nestedhap_fix_p2m(struct vcpu *v, struct
>          gdprintk(XENLOG_ERR,
>  		"failed to set entry for %#"PRIx64" -> %#"PRIx64"\n",
>  		L2_gpa, L0_gpa);
> -        BUG();
> +        domain_crash(p2m->domain);
>      }
>  }
>  
>
>
>
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel


[-- Attachment #1.2: Type: text/html, Size: 1968 bytes --]

[-- Attachment #2: Type: text/plain, Size: 126 bytes --]

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

Re: [PATCH] x86/nested HAP: don't BUG() on legitimate error

[Jan Beulich]

>>> On 11.04.14 at 13:08, <andrew.cooper3@citrix.com> wrote:
> On 11/04/14 11:28, Jan Beulich wrote:
>> set_p2m_entry() can fail without there being a bug in the code - crash
>> the domain rather than the host in that case.
> 
> It would be useful to indicate which conditions.

Out of memory when splitting a huge page; I thought this was
pretty obvious, but maybe I've been working too much on all
that paging code lately...

Jan

Re: [PATCH] x86/nested HAP: don't BUG() on legitimate error

[Tim Deegan]

At 11:28 +0100 on 11 Apr (1397212094), Jan Beulich wrote:
> set_p2m_entry() can fail without there being a bug in the code - crash
> the domain rather than the host in that case.
> 
> Signed-off-by: Jan Beulich <jbeulich@suse.com>

Acked-by: Tim Deegan <tim@xen.org>