Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1732851 > unrolled thread

Re: [RFC Part1 PATCH v3 13/17] x86/io: Unroll string I/O when SEV is active

Started byBorislav Petkov <bp@suse.de>
First post2017-09-15 14:30 +0200
Last post2017-09-15 18:30 +0200
Articles 3 — 1 participant

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: [RFC Part1 PATCH v3 13/17] x86/io: Unroll string I/O when SEV is  active Borislav Petkov <bp@suse.de> - 2017-09-15 14:30 +0200
    Re: [RFC Part1 PATCH v3 13/17] x86/io: Unroll string I/O when SEV is  active Borislav Petkov <bp@suse.de> - 2017-09-15 16:50 +0200
      Re: [RFC Part1 PATCH v3 13/17] x86/io: Unroll string I/O when SEV is  active Borislav Petkov <bp@suse.de> - 2017-09-15 18:30 +0200

#1732851 — Re: [RFC Part1 PATCH v3 13/17] x86/io: Unroll string I/O when SEV is active

FromBorislav Petkov <bp@suse.de>
Date2017-09-15 14:30 +0200
SubjectRe: [RFC Part1 PATCH v3 13/17] x86/io: Unroll string I/O when SEV is active
Message-ID<upXUS-2J1-17@gated-at.bofh.it>
On Tue, Aug 22, 2017 at 06:52:48PM +0200, Borislav Petkov wrote:
> As always, the devil is in the detail.

Ok, actually we can make this much simpler by using a static key. A
conceptual patch below - I only need to fix that crazy include hell I'm
stepping into with this.

In any case, we were talking about having a static branch already so
this fits the whole strategy.

---
diff --git a/arch/x86/mm/mem_encrypt.c b/arch/x86/mm/mem_encrypt.c
index d174b1c4a99e..e45369158632 100644
--- a/arch/x86/mm/mem_encrypt.c
+++ b/arch/x86/mm/mem_encrypt.c
@@ -45,6 +45,8 @@ EXPORT_SYMBOL_GPL(sme_me_mask);
 unsigned int sev_enabled __section(.data) = 0;
 EXPORT_SYMBOL_GPL(sev_enabled);
 
+DEFINE_STATIC_KEY_FALSE(__sev);
+
 /* Buffer used for early in-place encryption by BSP, no locking needed */
 static char sme_early_buffer[PAGE_SIZE] __aligned(PAGE_SIZE);
 
@@ -790,6 +792,7 @@ void __init __nostackprotector sme_enable(struct boot_params *bp)
 		/* SEV state cannot be controlled by a command line option */
 		sme_me_mask = me_mask;
 		sev_enabled = 1;
+		static_branch_enable(&__sev);
 		return;
 	}
 
diff --git a/include/linux/mem_encrypt.h b/include/linux/mem_encrypt.h
index ea0831a8dbe2..f3ab965a3d6a 100644
--- a/include/linux/mem_encrypt.h
+++ b/include/linux/mem_encrypt.h
@@ -13,6 +13,8 @@
 #ifndef __MEM_ENCRYPT_H__
 #define __MEM_ENCRYPT_H__
 
+#include <linux/jump_label.h>
+
 #ifndef __ASSEMBLY__
 
 #ifdef CONFIG_ARCH_HAS_MEM_ENCRYPT
@@ -26,6 +28,8 @@
 
 #endif	/* CONFIG_ARCH_HAS_MEM_ENCRYPT */
 
+extern struct static_key_false __sev;
+
 static inline bool sme_active(void)
 {
 	return (sme_me_mask && !sev_enabled);
@@ -33,7 +37,7 @@ static inline bool sme_active(void)
 
 static inline bool sev_active(void)
 {
-	return (sme_me_mask && sev_enabled);
+	return static_branch_unlikely(&__sev);
 }
 
 static inline unsigned long sme_get_me_mask(void)

-- 
Regards/Gruss,
    Boris.

SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg)
-- 

[toc] | [next] | [standalone]


#1732917

FromBorislav Petkov <bp@suse.de>
Date2017-09-15 16:50 +0200
Message-ID<uq06m-4g8-23@gated-at.bofh.it>
In reply to#1732851
On Fri, Sep 15, 2017 at 09:13:00AM -0500, Brijesh Singh wrote:
> thanks for the suggestion Boris, it will make patch much simpler.
> I will try this out.

It won't build - this was supposed to show the general idea.

I need to figure out the include hell first.

-- 
Regards/Gruss,
    Boris.

SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg)
-- 

[toc] | [prev] | [next] | [standalone]


#1732945

FromBorislav Petkov <bp@suse.de>
Date2017-09-15 18:30 +0200
Message-ID<uq1F8-5pk-13@gated-at.bofh.it>
In reply to#1732917
On Fri, Sep 15, 2017 at 09:48:53AM -0500, Brijesh Singh wrote:
> I see the similar issue with non SEV guest with my simple patch below.
> Guest will reboot as soon as it tries to enable the key.

Can't do it there as the pagetable is not setup yet and you're probably
getting a #PF on any of the derefs down the static_key_enable() path.

I guess one possible place to enable the static key would be in
mem_encrypt_init() where everything should be set up already.

-- 
Regards/Gruss,
    Boris.

SUSE Linux GmbH, GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg)
-- 

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web