[PATCH] svm: amend c/s 24990:322300fd2ebd (fake BU_CFG MSR on AMD revF)

[PATCH] svm: amend c/s 24990:322300fd2ebd (fake BU_CFG MSR on AMD revF)

[Jan Beulich]

[-- Attachment #1: Type: text/plain, Size: 580 bytes --]

Let's restrict such a hack to the known affected family.

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

--- a/xen/arch/x86/hvm/svm/svm.c
+++ b/xen/arch/x86/hvm/svm/svm.c
@@ -1505,7 +1505,7 @@ static int svm_msr_read_intercept(unsign
         if ( rdmsr_safe(msr, *msr_content) == 0 )
             break;
 
-        if ( msr == MSR_F10_BU_CFG )
+        if ( boot_cpu_data.x86 == 0xf && msr == MSR_F10_BU_CFG )
         {
             /* Win2k8 x64 reads this MSR on revF chips, where it
              * wasn't publically available; it uses a magic constant




[-- Attachment #2: amend-24990.patch --]
[-- Type: text/plain, Size: 641 bytes --]

svm: amend c/s 24990:322300fd2ebd (fake BU_CFG MSR on AMD revF)

Let's restrict such a hack to the known affected family.

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

--- a/xen/arch/x86/hvm/svm/svm.c
+++ b/xen/arch/x86/hvm/svm/svm.c
@@ -1505,7 +1505,7 @@ static int svm_msr_read_intercept(unsign
         if ( rdmsr_safe(msr, *msr_content) == 0 )
             break;
 
-        if ( msr == MSR_F10_BU_CFG )
+        if ( boot_cpu_data.x86 == 0xf && msr == MSR_F10_BU_CFG )
         {
             /* Win2k8 x64 reads this MSR on revF chips, where it
              * wasn't publically available; it uses a magic constant

[-- 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] svm: amend c/s 24990:322300fd2ebd (fake BU_CFG MSR on AMD revF)

[Keir Fraser]

On 15/03/2012 12:34, "Jan Beulich" <JBeulich@suse.com> wrote:

> Let's restrict such a hack to the known affected family.
> 
> Signed-off-by: Jan Beulich <jbeulich@suse.com>

Acked-by: Keir Fraser <keir@xen.org>

> --- a/xen/arch/x86/hvm/svm/svm.c
> +++ b/xen/arch/x86/hvm/svm/svm.c
> @@ -1505,7 +1505,7 @@ static int svm_msr_read_intercept(unsign
>          if ( rdmsr_safe(msr, *msr_content) == 0 )
>              break;
>  
> -        if ( msr == MSR_F10_BU_CFG )
> +        if ( boot_cpu_data.x86 == 0xf && msr == MSR_F10_BU_CFG )
>          {
>              /* Win2k8 x64 reads this MSR on revF chips, where it
>               * wasn't publically available; it uses a magic constant
> 
> 
> 
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xen.org
> http://lists.xen.org/xen-devel

Re: [PATCH] svm: amend c/s 24990:322300fd2ebd (fake BU_CFG MSR on AMD revF)

[George Dunlap]

On Thu, 2012-03-15 at 12:34 +0000, Jan Beulich wrote:
> Let's restrict such a hack to the known affected family.
> 
> Signed-off-by: Jan Beulich <jbeulich@suse.com>

I haven't tested it, but in principle:
Acked-by: George Dunlap <george.dunlap@eu.citrix.com>

Thanks Jan.

I'll see if I can get this back-ported to the XS patchqueue, so we can
test to make sure it works on all the hardware we have available.

 -George

> 
> --- a/xen/arch/x86/hvm/svm/svm.c
> +++ b/xen/arch/x86/hvm/svm/svm.c
> @@ -1505,7 +1505,7 @@ static int svm_msr_read_intercept(unsign
>          if ( rdmsr_safe(msr, *msr_content) == 0 )
>              break;
>  
> -        if ( msr == MSR_F10_BU_CFG )
> +        if ( boot_cpu_data.x86 == 0xf && msr == MSR_F10_BU_CFG )
>          {
>              /* Win2k8 x64 reads this MSR on revF chips, where it
>               * wasn't publically available; it uses a magic constant
> 
> 
>