[patch] KVM: fix hypercall cache/decache

[patch] KVM: fix hypercall cache/decache

[Ingo Molnar]

Subject: [patch] KVM: fix hypercall cache/decache
From: Ingo Molnar <mingo-X9Un+BFzKDI@public.gmane.org>

a recent change to kvm_hypercall() flipped around the order of 
->cache_regs() and ->decache_regs(), but that is bogus and causes 
ESP/RIP corruption: ->cache_regs() will 'cache' the CPU-side registers 
into the vcpu structure, and ->decache_regs() will sync it back to the 
CPU.

fix this by using the correct order again.

Signed-off-by: Ingo Molnar <mingo-X9Un+BFzKDI@public.gmane.org>
---
 drivers/kvm/kvm_main.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

Index: linux/drivers/kvm/kvm_main.c
===================================================================
--- linux.orig/drivers/kvm/kvm_main.c
+++ linux/drivers/kvm/kvm_main.c
@@ -1177,7 +1177,7 @@ int kvm_hypercall(struct kvm_vcpu *vcpu,
 {
 	unsigned long nr, a0, a1, a2, a3, a4, a5, ret;
 
-	kvm_arch_ops->decache_regs(vcpu);
+	kvm_arch_ops->cache_regs(vcpu);
 	ret = -KVM_EINVAL;
 #ifdef CONFIG_X86_64
 	if (is_long_mode(vcpu)) {
@@ -1204,7 +1204,8 @@ int kvm_hypercall(struct kvm_vcpu *vcpu,
 		;
 	}
 	vcpu->regs[VCPU_REGS_RAX] = ret;
-	kvm_arch_ops->cache_regs(vcpu);
+	kvm_arch_ops->decache_regs(vcpu);
+
 	return 1;
 }
 EXPORT_SYMBOL_GPL(kvm_hypercall);

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: [patch] KVM: fix hypercall cache/decache

[Avi Kivity]

Ingo Molnar wrote:
> Subject: [patch] KVM: fix hypercall cache/decache
> From: Ingo Molnar <mingo-X9Un+BFzKDI@public.gmane.org>
>
> a recent change to kvm_hypercall() flipped around the order of 
> ->cache_regs() and ->decache_regs(), but that is bogus and causes 
> ESP/RIP corruption: ->cache_regs() will 'cache' the CPU-side registers 
> into the vcpu structure, and ->decache_regs() will sync it back to the 
> CPU.
>
> fix this by using the correct order again.
>
>   

That's 57d78025d84fb607aa335d015a79b257517aa209 in my tree...


-- 
error compiling committee.c: too many arguments to function


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV

Re: [patch] KVM: fix hypercall cache/decache

[Ingo Molnar]

* Avi Kivity <avi-atKUWr5tajBWk0Htik3J/w@public.gmane.org> wrote:

> Ingo Molnar wrote:
> >Subject: [patch] KVM: fix hypercall cache/decache
> >From: Ingo Molnar <mingo-X9Un+BFzKDI@public.gmane.org>
> >
> >a recent change to kvm_hypercall() flipped around the order of 
> >->cache_regs() and ->decache_regs(), but that is bogus and causes 
> >ESP/RIP corruption: ->cache_regs() will 'cache' the CPU-side registers 
> >into the vcpu structure, and ->decache_regs() will sync it back to the 
> >CPU.
> >
> >fix this by using the correct order again.
> 
> That's 57d78025d84fb607aa335d015a79b257517aa209 in my tree...

ok :)

	Ingo

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys-and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV