diff mbox series

[Resend,RFC,V4,13/13] x86/HV: Not set memory decrypted/encrypted during kexec alloc/free page in IVM

Message ID 20210707154629.3977369-14-ltykernel@gmail.com (mailing list archive)
State RFC
Headers show
Series x86/Hyper-V: Add Hyper-V Isolation VM support | expand

Checks

Context Check Description
netdev/tree_selection success Guessing tree name failed - patch did not apply

Commit Message

Tianyu Lan July 7, 2021, 3:46 p.m. UTC 
From: Tianyu Lan <Tianyu.Lan@microsoft.com>

Hyper-V Isolation VM reuses set_memory_decrypted/encrypted function
and not needs to decrypted/encrypted in arch_kexec_post_alloc(pre_free)
_pages just likes AMD SEV VM. So skip them.

Signed-off-by: Tianyu Lan <Tianyu.Lan@microsoft.com>
---
 arch/x86/kernel/machine_kexec_64.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

Comments

Dave Hansen July 7, 2021, 4:14 p.m. UTC | #1 
On 7/7/21 8:46 AM, Tianyu Lan wrote:
> @@ -598,7 +599,7 @@ void arch_kexec_unprotect_crashkres(void)
>   */
>  int arch_kexec_post_alloc_pages(void *vaddr, unsigned int pages, gfp_t gfp)
>  {
> -	if (sev_active())
> +	if (sev_active() || hv_is_isolation_supported())
>  		return 0;
>  
>  	/*
> @@ -611,7 +612,7 @@ int arch_kexec_post_alloc_pages(void *vaddr, unsigned int pages, gfp_t gfp)
>  
>  void arch_kexec_pre_free_pages(void *vaddr, unsigned int pages)
>  {
> -	if (sev_active())
> +	if (sev_active() || hv_is_isolation_supported())
>  		return;

You might want to take a look through the "protected guest" patches.  I
think this series is touching a few of the same locations that TDX and
recent SEV work touch.

https://lore.kernel.org/lkml/20210618225755.662725-5-sathyanarayanan.kuppuswamy@linux.intel.com/
Tianyu Lan July 8, 2021, 1:54 p.m. UTC | #2 
Hi Dave:
      Thanks for your review.

On 7/8/2021 12:14 AM, Dave Hansen wrote:
> On 7/7/21 8:46 AM, Tianyu Lan wrote:
>> @@ -598,7 +599,7 @@ void arch_kexec_unprotect_crashkres(void)
>>    */
>>   int arch_kexec_post_alloc_pages(void *vaddr, unsigned int pages, gfp_t gfp)
>>   {
>> -	if (sev_active())
>> +	if (sev_active() || hv_is_isolation_supported())
>>   		return 0;
>>   
>>   	/*
>> @@ -611,7 +612,7 @@ int arch_kexec_post_alloc_pages(void *vaddr, unsigned int pages, gfp_t gfp)
>>   
>>   void arch_kexec_pre_free_pages(void *vaddr, unsigned int pages)
>>   {
>> -	if (sev_active())
>> +	if (sev_active() || hv_is_isolation_supported())
>>   		return;
> 
> You might want to take a look through the "protected guest" patches.  I
> think this series is touching a few of the same locations that TDX and
> recent SEV work touch.
> 
> https://lore.kernel.org/lkml/20210618225755.662725-5-sathyanarayanan.kuppuswamy@linux.intel.com/

Thanks for reminder. You are right. There will be a generic API to check 
"proteced guest" type.
diff mbox series

Patch

diff --git a/arch/x86/kernel/machine_kexec_64.c b/arch/x86/kernel/machine_kexec_64.c
index c078b0d3ab0e..0cadc64b6873 100644
--- a/arch/x86/kernel/machine_kexec_64.c
+++ b/arch/x86/kernel/machine_kexec_64.c
@@ -26,6 +26,7 @@ 
 #include <asm/kexec-bzimage64.h>
 #include <asm/setup.h>
 #include <asm/set_memory.h>
+#include <asm/mshyperv.h>
 
 #ifdef CONFIG_ACPI
 /*
@@ -598,7 +599,7 @@  void arch_kexec_unprotect_crashkres(void)
  */
 int arch_kexec_post_alloc_pages(void *vaddr, unsigned int pages, gfp_t gfp)
 {
-	if (sev_active())
+	if (sev_active() || hv_is_isolation_supported())
 		return 0;
 
 	/*
@@ -611,7 +612,7 @@  int arch_kexec_post_alloc_pages(void *vaddr, unsigned int pages, gfp_t gfp)
 
 void arch_kexec_pre_free_pages(void *vaddr, unsigned int pages)
 {
-	if (sev_active())
+	if (sev_active() || hv_is_isolation_supported())
 		return;
 
 	/*