public inbox for kvm@vger.kernel.org
 help / color / mirror / Atom feed
From: Patrick Bellasi <derkling@google.com>
To: Borislav Petkov <bp@alien8.de>
Cc: Sean Christopherson <seanjc@google.com>,
	Paolo Bonzini <pbonzini@redhat.com>,
	 Josh Poimboeuf <jpoimboe@redhat.com>,
	Pawan Gupta <pawan.kumar.gupta@linux.intel.com>,
	x86@kernel.org,  kvm@vger.kernel.org,
	linux-kernel@vger.kernel.org,
	 Patrick Bellasi <derkling@matbug.net>
Subject: Re: Re: Re: [PATCH] x86/bugs: KVM: Add support for SRSO_MSR_FIX
Date: Thu, 13 Feb 2025 13:44:08 +0000	[thread overview]
Message-ID: <20250213134408.2931040-1-derkling@google.com> (raw)
In-Reply-To: <20250213105304.1888660-1-derkling@google.com>

And, of course, this bit:

> diff --git a/tools/arch/x86/include/asm/msr-index.h b/tools/arch/x86/include/asm/msr-index.h
> index 3ae84c3b8e6db..1372a569fb585 100644
> --- a/tools/arch/x86/include/asm/msr-index.h
> +++ b/tools/arch/x86/include/asm/msr-index.h
> @@ -717,6 +717,7 @@
> >
>  /* Zen4 */
>  #define MSR_ZEN4_BP_CFG                 0xc001102e
> +#define MSR_ZEN4_BP_CFG_BP_SPEC_REDUCE_BIT 4

has to be added to arch/x86/include/asm/msr-index.h as well.

Following (yet another) updated versions accounting for this.

Best,
Patrick

---

From: Borislav Petkov <bp@alien8.de>

Add support for

  CPUID Fn8000_0021_EAX[31] (SRSO_MSR_FIX). If this bit is 1, it
  indicates that software may use MSR BP_CFG[BpSpecReduce] to mitigate
  SRSO.

enable this BpSpecReduce bit to mitigate SRSO across guest/host
boundaries.

Co-developed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Signed-off-by: Patrick Bellasi <derkling@google.com>
---
 Documentation/admin-guide/hw-vuln/srso.rst | 20 ++++++++++++++++++++
 arch/x86/include/asm/cpufeatures.h         |  1 +
 arch/x86/include/asm/msr-index.h           |  1 +
 arch/x86/kernel/cpu/bugs.c                 | 21 +++++++++++++++++----
 arch/x86/kvm/svm/svm.c                     | 14 ++++++++++++++
 tools/arch/x86/include/asm/msr-index.h     |  1 +
 6 files changed, 54 insertions(+), 4 deletions(-)

diff --git a/Documentation/admin-guide/hw-vuln/srso.rst b/Documentation/admin-guide/hw-vuln/srso.rst
index 2ad1c05b8c883..49680ab99c393 100644
--- a/Documentation/admin-guide/hw-vuln/srso.rst
+++ b/Documentation/admin-guide/hw-vuln/srso.rst
@@ -104,6 +104,26 @@ The possible values in this file are:
 
    (spec_rstack_overflow=ibpb-vmexit)
 
+ * 'Mitigation: Reduced Speculation':
+
+   This mitigation gets automatically enabled when the above one "IBPB on
+   VMEXIT" has been selected and the CPU supports the BpSpecReduce bit.
+
+   It gets automatically enabled on machines which have the
+   SRSO_USER_KERNEL_NO=1 CPUID bit. In that case, the code logic is to switch
+   to the above =ibpb-vmexit mitigation because the user/kernel boundary is
+   not affected anymore and thus "safe RET" is not needed.
+
+   After enabling the IBPB on VMEXIT mitigation option, the BpSpecReduce bit
+   is detected (functionality present on all such machines) and that
+   practically overrides IBPB on VMEXIT as it has a lot less performance
+   impact and takes care of the guest->host attack vector too.
+
+   Currently, the mitigation uses KVM's user_return approach
+   (kvm_set_user_return_msr()) to set the BpSpecReduce bit when a vCPU runs
+   a guest and reset it upon return to host userspace or when the KVM module
+   is unloaded. The intent being, the small perf impact of BpSpecReduce should
+   be incurred only when really necessary.
 
 
 In order to exploit vulnerability, an attacker needs to:
diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h
index 508c0dad116bc..c46754298507b 100644
--- a/arch/x86/include/asm/cpufeatures.h
+++ b/arch/x86/include/asm/cpufeatures.h
@@ -468,6 +468,7 @@
 #define X86_FEATURE_IBPB_BRTYPE		(20*32+28) /* MSR_PRED_CMD[IBPB] flushes all branch type predictions */
 #define X86_FEATURE_SRSO_NO		(20*32+29) /* CPU is not affected by SRSO */
 #define X86_FEATURE_SRSO_USER_KERNEL_NO	(20*32+30) /* CPU is not affected by SRSO across user/kernel boundaries */
+#define X86_FEATURE_SRSO_BP_SPEC_REDUCE	(20*32+31) /* BP_CFG[BpSpecReduce] can be used to mitigate SRSO for VMs (SRSO_MSR_FIX in AMD docs). */
 
 /*
  * Extended auxiliary flags: Linux defined - for features scattered in various
diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h
index 9a71880eec070..6bbc8836d6766 100644
--- a/arch/x86/include/asm/msr-index.h
+++ b/arch/x86/include/asm/msr-index.h
@@ -720,6 +720,7 @@
 
 /* Zen4 */
 #define MSR_ZEN4_BP_CFG                 0xc001102e
+#define MSR_ZEN4_BP_CFG_BP_SPEC_REDUCE_BIT 4
 #define MSR_ZEN4_BP_CFG_SHARED_BTB_FIX_BIT 5
 
 /* Fam 19h MSRs */
diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c
index a5d0998d76049..d2007dbfcc1cc 100644
--- a/arch/x86/kernel/cpu/bugs.c
+++ b/arch/x86/kernel/cpu/bugs.c
@@ -2522,6 +2522,7 @@ enum srso_mitigation {
 	SRSO_MITIGATION_SAFE_RET,
 	SRSO_MITIGATION_IBPB,
 	SRSO_MITIGATION_IBPB_ON_VMEXIT,
+	SRSO_MITIGATION_BP_SPEC_REDUCE,
 };
 
 enum srso_mitigation_cmd {
@@ -2539,7 +2540,8 @@ static const char * const srso_strings[] = {
 	[SRSO_MITIGATION_MICROCODE]		= "Vulnerable: Microcode, no safe RET",
 	[SRSO_MITIGATION_SAFE_RET]		= "Mitigation: Safe RET",
 	[SRSO_MITIGATION_IBPB]			= "Mitigation: IBPB",
-	[SRSO_MITIGATION_IBPB_ON_VMEXIT]	= "Mitigation: IBPB on VMEXIT only"
+	[SRSO_MITIGATION_IBPB_ON_VMEXIT]	= "Mitigation: IBPB on VMEXIT only",
+	[SRSO_MITIGATION_BP_SPEC_REDUCE]	= "Mitigation: Reduced Speculation"
 };
 
 static enum srso_mitigation srso_mitigation __ro_after_init = SRSO_MITIGATION_NONE;
@@ -2578,7 +2580,7 @@ static void __init srso_select_mitigation(void)
 	    srso_cmd == SRSO_CMD_OFF) {
 		if (boot_cpu_has(X86_FEATURE_SBPB))
 			x86_pred_cmd = PRED_CMD_SBPB;
-		return;
+		goto out;
 	}
 
 	if (has_microcode) {
@@ -2590,7 +2592,7 @@ static void __init srso_select_mitigation(void)
 		 */
 		if (boot_cpu_data.x86 < 0x19 && !cpu_smt_possible()) {
 			setup_force_cpu_cap(X86_FEATURE_SRSO_NO);
-			return;
+			goto out;
 		}
 
 		if (retbleed_mitigation == RETBLEED_MITIGATION_IBPB) {
@@ -2670,6 +2672,12 @@ static void __init srso_select_mitigation(void)
 
 ibpb_on_vmexit:
 	case SRSO_CMD_IBPB_ON_VMEXIT:
+		if (boot_cpu_has(X86_FEATURE_SRSO_BP_SPEC_REDUCE)) {
+			pr_notice("Reducing speculation to address VM/HV SRSO attack vector.\n");
+			srso_mitigation = SRSO_MITIGATION_BP_SPEC_REDUCE;
+			break;
+		}
+
 		if (IS_ENABLED(CONFIG_MITIGATION_IBPB_ENTRY)) {
 			if (has_microcode) {
 				setup_force_cpu_cap(X86_FEATURE_IBPB_ON_VMEXIT);
@@ -2691,7 +2699,12 @@ static void __init srso_select_mitigation(void)
 	}
 
 out:
-	pr_info("%s\n", srso_strings[srso_mitigation]);
+
+	if (srso_mitigation != SRSO_MITIGATION_BP_SPEC_REDUCE)
+		setup_clear_cpu_cap(X86_FEATURE_SRSO_BP_SPEC_REDUCE);
+
+	if (srso_mitigation != SRSO_MITIGATION_NONE)
+		pr_info("%s\n", srso_strings[srso_mitigation]);
 }
 
 #undef pr_fmt
diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
index 7640a84e554a6..6ea3632af5807 100644
--- a/arch/x86/kvm/svm/svm.c
+++ b/arch/x86/kvm/svm/svm.c
@@ -257,6 +257,7 @@ DEFINE_PER_CPU(struct svm_cpu_data, svm_data);
  * defer the restoration of TSC_AUX until the CPU returns to userspace.
  */
 static int tsc_aux_uret_slot __read_mostly = -1;
+static int zen4_bp_cfg_uret_slot __ro_after_init = -1;
 
 static const u32 msrpm_ranges[] = {0, 0xc0000000, 0xc0010000};
 
@@ -1540,6 +1541,11 @@ static void svm_prepare_switch_to_guest(struct kvm_vcpu *vcpu)
 	    (!boot_cpu_has(X86_FEATURE_V_TSC_AUX) || !sev_es_guest(vcpu->kvm)))
 		kvm_set_user_return_msr(tsc_aux_uret_slot, svm->tsc_aux, -1ull);
 
+	if (cpu_feature_enabled(X86_FEATURE_SRSO_BP_SPEC_REDUCE))
+		kvm_set_user_return_msr(zen4_bp_cfg_uret_slot,
+					BIT_ULL(MSR_ZEN4_BP_CFG_BP_SPEC_REDUCE_BIT),
+					BIT_ULL(MSR_ZEN4_BP_CFG_BP_SPEC_REDUCE_BIT));
+
 	svm->guest_state_loaded = true;
 }
 
@@ -5306,6 +5312,14 @@ static __init int svm_hardware_setup(void)
 
 	tsc_aux_uret_slot = kvm_add_user_return_msr(MSR_TSC_AUX);
 
+	if (cpu_feature_enabled(X86_FEATURE_SRSO_BP_SPEC_REDUCE)) {
+		zen4_bp_cfg_uret_slot = kvm_add_user_return_msr(MSR_ZEN4_BP_CFG);
+		if (WARN_ON_ONCE(zen4_bp_cfg_uret_slot < 0)) {
+			r = -EIO;
+			goto err;
+		}
+	}
+
 	if (boot_cpu_has(X86_FEATURE_AUTOIBRS))
 		kvm_enable_efer_bits(EFER_AUTOIBRS);
 
diff --git a/tools/arch/x86/include/asm/msr-index.h b/tools/arch/x86/include/asm/msr-index.h
index 3ae84c3b8e6db..1372a569fb585 100644
--- a/tools/arch/x86/include/asm/msr-index.h
+++ b/tools/arch/x86/include/asm/msr-index.h
@@ -717,6 +717,7 @@
 
 /* Zen4 */
 #define MSR_ZEN4_BP_CFG                 0xc001102e
+#define MSR_ZEN4_BP_CFG_BP_SPEC_REDUCE_BIT 4
 #define MSR_ZEN4_BP_CFG_SHARED_BTB_FIX_BIT 5
 
 /* Fam 19h MSRs */
-- 
2.48.1.601.g30ceb7b040-goog


  reply	other threads:[~2025-02-13 13:44 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-12-02 12:04 [PATCH v2 0/4] x86/bugs: Adjust SRSO mitigation to new features Borislav Petkov
2024-12-02 12:04 ` [PATCH v2 1/4] x86/bugs: Add SRSO_USER_KERNEL_NO support Borislav Petkov
2024-12-10  6:53   ` Josh Poimboeuf
2024-12-10 15:37     ` Borislav Petkov
2024-12-11  7:53       ` Josh Poimboeuf
2024-12-11 20:38         ` Borislav Petkov
2024-12-11 22:35           ` Sean Christopherson
2024-12-16 17:21             ` Borislav Petkov
2024-12-02 12:04 ` [PATCH v2 2/4] KVM: x86: Advertise SRSO_USER_KERNEL_NO to userspace Borislav Petkov
2024-12-02 12:04 ` [PATCH v2 3/4] x86/bugs: KVM: Add support for SRSO_MSR_FIX Borislav Petkov
2024-12-11 22:27   ` Sean Christopherson
2024-12-16 17:31     ` Borislav Petkov
2024-12-16 18:51       ` Sean Christopherson
2024-12-17  9:34         ` Borislav Petkov
2024-12-30 11:14         ` Borislav Petkov
2025-01-08 13:38           ` Sean Christopherson
2025-01-08 15:49             ` Borislav Petkov
2025-01-08 17:18               ` Sean Christopherson
2025-01-08 18:14                 ` Borislav Petkov
2025-01-08 18:37                   ` Jim Mattson
2025-01-08 19:14                     ` Borislav Petkov
2025-01-08 19:43                       ` Jim Mattson
2025-01-08 19:45                         ` Borislav Petkov
2025-01-11 12:52                   ` [PATCH] " Borislav Petkov
2025-01-17 18:56                     ` Sean Christopherson
2025-01-18 15:26                       ` Borislav Petkov
2025-01-23 16:25                         ` Sean Christopherson
2025-01-23 17:01                           ` Borislav Petkov
2025-01-23 18:04                             ` Sean Christopherson
2025-01-24 12:58                               ` Borislav Petkov
2025-02-11 19:19                                 ` Jim Mattson
2025-02-11 20:51                                   ` Borislav Petkov
2025-02-13 10:53                             ` Patrick Bellasi
2025-02-13 13:44                               ` Patrick Bellasi [this message]
2025-02-13 14:28                                 ` Borislav Petkov
2025-02-13 17:50                                   ` Patrick Bellasi
2025-02-14 20:10                                     ` Borislav Petkov
2025-02-15  0:57                                       ` Yosry Ahmed
2025-02-15  9:15                                         ` Borislav Petkov
2025-02-17  5:47                                           ` Yosry Ahmed
2025-02-17 15:26                                             ` Borislav Petkov
2025-02-15 12:53                                       ` Borislav Petkov
2025-02-17  5:59                                         ` Yosry Ahmed
2025-02-17 16:07                                           ` Borislav Petkov
2025-02-17 19:56                                             ` Yosry Ahmed
2025-02-17 20:20                                               ` Borislav Petkov
2025-02-17 20:32                                                 ` Yosry Ahmed
2025-02-18 11:13                                                   ` [PATCH final?] " Borislav Petkov
2025-02-18 14:42                                                     ` Patrick Bellasi
2025-02-18 15:34                                                       ` Borislav Petkov
2025-04-29 13:25                                                     ` x86/bugs: KVM: Add support for SRSO_MSR_FIX, back for moar Borislav Petkov
2025-04-30 23:33                                                       ` Sean Christopherson
2025-05-01  0:42                                                         ` Michael Larabel
2025-05-01  8:19                                                         ` Borislav Petkov
2025-05-01 16:56                                                           ` Sean Christopherson
2025-05-05 15:25                                                             ` Borislav Petkov
2025-05-05 15:40                                                               ` Kaplan, David
2025-05-05 15:47                                                                 ` Borislav Petkov
2025-05-05 16:30                                                                 ` Sean Christopherson
2025-05-05 16:42                                                                   ` Kaplan, David
2025-05-05 18:03                                                                     ` Sean Christopherson
2025-05-05 18:25                                                                       ` Kaplan, David
2024-12-02 12:04 ` [PATCH v2 4/4] Documentation/kernel-parameters: Fix a typo in kvm.enable_virt_at_load text Borislav Petkov
2024-12-03 14:30 ` [PATCH v2 0/4] x86/bugs: Adjust SRSO mitigation to new features Nikolay Borisov

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=20250213134408.2931040-1-derkling@google.com \
    --to=derkling@google.com \
    --cc=bp@alien8.de \
    --cc=derkling@matbug.net \
    --cc=jpoimboe@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pawan.kumar.gupta@linux.intel.com \
    --cc=pbonzini@redhat.com \
    --cc=seanjc@google.com \
    --cc=x86@kernel.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox