All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Roedel, Joerg" <Joerg.Roedel@amd.com>
To: Avi Kivity <avi@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>,
	"kvm@vger.kernel.org" <kvm@vger.kernel.org>
Subject: Re: [PATCH 08/13] KVM: SVM: Add intercept checks for SVM instructions
Date: Mon, 28 Mar 2011 14:18:43 +0200	[thread overview]
Message-ID: <20110328121843.GC2085@amd.com> (raw)
In-Reply-To: <4D907A35.9090008@redhat.com>

On Mon, Mar 28, 2011 at 08:08:21AM -0400, Avi Kivity wrote:
> On 03/28/2011 12:46 PM, Joerg Roedel wrote:
> > This patch adds the necessary code changes in the
> > instruction emulator and the extensions to svm.c to
> > implement intercept checks for the svm instructions.
> >
> > diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
> > index 485a09f..9b22f5f 100644
> > --- a/arch/x86/kvm/svm.c
> > +++ b/arch/x86/kvm/svm.c
> > @@ -3896,6 +3896,14 @@ static struct __x86_intercept {
> >   	[x86_intercept_sidt]		= POST_MEM(SVM_EXIT_IDTR_READ),
> >   	[x86_intercept_lgdt]		= POST_MEM(SVM_EXIT_GDTR_WRITE),
> >   	[x86_intercept_lidt]		= POST_MEM(SVM_EXIT_IDTR_WRITE),
> > +	[x86_intercept_vmrun]		= POST_EX(SVM_EXIT_VMRUN),
> > +	[x86_intercept_vmmcall]		= POST_EX(SVM_EXIT_VMMCALL),
> 
> VMMCALL is not privileged, is it?

Right, thanks. Here is the updated patch.


>From aeaa38bf81c9386aeff4368c2b23ec95c0d08e4c Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Fri, 11 Mar 2011 11:53:39 +0100
Subject: [PATCH 08/13] KVM: SVM: Add intercept checks for SVM instructions

This patch adds the necessary code changes in the
instruction emulator and the extensions to svm.c to
implement intercept checks for the svm instructions.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
---
 arch/x86/kvm/emulate.c |   23 ++++++++++++++++++++++-
 arch/x86/kvm/svm.c     |    8 ++++++++
 2 files changed, 30 insertions(+), 1 deletions(-)

diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index ec2e9ad..8b8f63d 100644
--- a/arch/x86/kvm/emulate.c
+++ b/arch/x86/kvm/emulate.c
@@ -75,6 +75,8 @@
 #define Stack       (1<<13)     /* Stack instruction (push/pop) */
 #define Group       (1<<14)     /* Bits 3:5 of modrm byte extend opcode */
 #define GroupDual   (1<<15)     /* Alternate decoding of mod == 3 */
+#define RMExt       (1<<16)     /* Opcode extension in ModRM r/m if mod == 3 */
+
 /* Misc flags */
 #define VendorSpecific (1<<22) /* Vendor specific instruction */
 #define NoAccess    (1<<23) /* Don't access memory (lea/invlpg/verr etc) */
@@ -2349,6 +2351,7 @@ static int em_mov(struct x86_emulate_ctxt *ctxt)
 #define D(_y) { .flags = (_y) }
 #define DI(_y, _i) { .flags = (_y), .intercept = x86_intercept_##_i }
 #define N    D(0)
+#define EXT(_f, _e) { .flags = ((_f) | RMExt), .u.group = (_e) }
 #define G(_f, _g) { .flags = ((_f) | Group), .u.group = (_g) }
 #define GD(_f, _g) { .flags = ((_f) | Group | GroupDual), .u.gdual = (_g) }
 #define I(_f, _e) { .flags = (_f), .u.execute = (_e) }
@@ -2363,6 +2366,17 @@ static int em_mov(struct x86_emulate_ctxt *ctxt)
 		D2bv(((_f) & ~Lock) | DstAcc | SrcImm)
 
 
+static struct opcode group7_rm3[] = {
+	DI(SrcNone | ModRM | Priv, vmrun),
+	DI(SrcNone | ModRM       , vmmcall),
+	DI(SrcNone | ModRM | Priv, vmload),
+	DI(SrcNone | ModRM | Priv, vmsave),
+	DI(SrcNone | ModRM | Priv, stgi),
+	DI(SrcNone | ModRM | Priv, clgi),
+	DI(SrcNone | ModRM | Priv, skinit),
+	DI(SrcNone | ModRM | Priv, invlpga),
+};
+
 static struct opcode group1[] = {
 	X7(D(Lock)), N
 };
@@ -2406,7 +2420,7 @@ static struct group_dual group7 = { {
 	DI(SrcMem | ModRM | ByteOp | Priv | NoAccess, invlpg),
 }, {
 	D(SrcNone | ModRM | Priv | VendorSpecific), N,
-	N, D(SrcNone | ModRM | Priv | VendorSpecific),
+	N, EXT(0, group7_rm3),
 	DI(SrcNone | ModRM | DstMem | Mov, smsw), N,
 	DI(SrcMem16 | ModRM | Mov | Priv, lmsw), N,
 } };
@@ -2601,6 +2615,7 @@ static struct opcode twobyte_table[256] = {
 #undef G
 #undef GD
 #undef I
+#undef EXT
 
 #undef D2bv
 #undef I2bv
@@ -2778,6 +2793,12 @@ done_prefixes:
 			opcode = g_mod3[goffset];
 		else
 			opcode = g_mod012[goffset];
+
+		if (opcode.flags & RMExt) {
+			goffset = c->modrm & 7;
+			opcode = opcode.u.group[goffset];
+		}
+
 		c->d |= opcode.flags;
 	}
 
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 485a09f..9b22f5f 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -3896,6 +3896,14 @@ static struct __x86_intercept {
 	[x86_intercept_sidt]		= POST_MEM(SVM_EXIT_IDTR_READ),
 	[x86_intercept_lgdt]		= POST_MEM(SVM_EXIT_GDTR_WRITE),
 	[x86_intercept_lidt]		= POST_MEM(SVM_EXIT_IDTR_WRITE),
+	[x86_intercept_vmrun]		= POST_EX(SVM_EXIT_VMRUN),
+	[x86_intercept_vmmcall]		= POST_EX(SVM_EXIT_VMMCALL),
+	[x86_intercept_vmload]		= POST_EX(SVM_EXIT_VMLOAD),
+	[x86_intercept_vmsave]		= POST_EX(SVM_EXIT_VMSAVE),
+	[x86_intercept_stgi]		= POST_EX(SVM_EXIT_STGI),
+	[x86_intercept_clgi]		= POST_EX(SVM_EXIT_CLGI),
+	[x86_intercept_skinit]		= POST_EX(SVM_EXIT_SKINIT),
+	[x86_intercept_invlpga]		= POST_EX(SVM_EXIT_INVLPGA),
 };
 
 #undef POST_EX
-- 
1.7.1

-- 
AMD Operating System Research Center

Advanced Micro Devices GmbH Einsteinring 24 85609 Dornach
General Managers: Alberto Bozzo, Andrew Bowd
Registration: Dornach, Landkr. Muenchen; Registerger. Muenchen, HRB Nr. 43632


  reply	other threads:[~2011-03-28 12:18 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-03-28 10:46 [PATCH 0/13] KVM: Make the instruction emulator aware of Nested Virtualization v3 Joerg Roedel
2011-03-28 10:46 ` [PATCH 01/13] KVM: x86 emulator: add framework for instruction intercepts Joerg Roedel
2011-03-28 10:46 ` [PATCH 02/13] KVM: x86 emulator: add SVM intercepts Joerg Roedel
2011-03-28 10:46 ` [PATCH 03/13] KVM: X86: Don't write-back cpu-state on X86EMUL_INTERCEPTED Joerg Roedel
2011-03-28 10:46 ` [PATCH 04/13] KVM: X86: Add x86 callback for intercept check Joerg Roedel
2011-03-28 10:46 ` [PATCH 05/13] KVM: SVM: Add intercept check for emulated cr accesses Joerg Roedel
2011-03-28 10:46 ` [PATCH 06/13] KVM: SVM: Add intercept check for accessing dr registers Joerg Roedel
2011-03-28 10:46 ` [PATCH 07/13] KVM: SVM: Add intercept checks for descriptor table accesses Joerg Roedel
2011-03-28 12:35   ` Avi Kivity
2011-03-28 13:56     ` Roedel, Joerg
2011-03-28 14:34       ` Avi Kivity
2011-03-28 10:46 ` [PATCH 08/13] KVM: SVM: Add intercept checks for SVM instructions Joerg Roedel
2011-03-28 12:08   ` Avi Kivity
2011-03-28 12:18     ` Roedel, Joerg [this message]
2011-03-28 10:46 ` [PATCH 09/13] KVM: SVM: Add intercept checks for remaining group7 instructions Joerg Roedel
2011-03-28 12:15   ` Avi Kivity
2011-03-28 10:46 ` [PATCH 10/13] KVM: SVM: Add intercept checks for remaining twobyte instructions Joerg Roedel
2011-03-28 12:29   ` Avi Kivity
2011-03-28 10:46 ` [PATCH 11/13] KVM: SVM: Add intercept checks for one-byte instructions Joerg Roedel
2011-03-28 10:46 ` [PATCH 12/13] KVM: SVM: Add checks for IO instructions Joerg Roedel
2011-03-28 12:28   ` Avi Kivity
2011-03-31  7:14     ` Roedel, Joerg
2011-03-31  9:18       ` Avi Kivity
2011-03-31  9:42         ` Roedel, Joerg
2011-03-31 10:03           ` Avi Kivity
2011-03-31 10:28             ` Roedel, Joerg
2011-03-28 10:46 ` [PATCH 13/13] KVM: SVM: Remove nested sel_cr0_write handling code Joerg Roedel
  -- strict thread matches above, loose matches on Subject: below --
2011-03-25  9:29 [PATCH 0/13] KVM: Make the instruction emulator aware of Nested Virtualization v2 Joerg Roedel
2011-03-25  9:29 ` [PATCH 08/13] KVM: SVM: Add intercept checks for SVM instructions Joerg Roedel

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20110328121843.GC2085@amd.com \
    --to=joerg.roedel@amd.com \
    --cc=avi@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=mtosatti@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.