linux-kernel - [PATCH 2/3] KVM: x86: Use explicit case-statements for MCx banks in {g,s}et_msr

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <20220512222716.4112548-3-seanjc@google.com>
Date:   Thu, 12 May 2022 22:27:15 +0000
From:   Sean Christopherson <seanjc@...gle.com>
To:     Paolo Bonzini <pbonzini@...hat.com>
Cc:     Sean Christopherson <seanjc@...gle.com>,
        Vitaly Kuznetsov <vkuznets@...hat.com>,
        Wanpeng Li <wanpengli@...cent.com>,
        Jim Mattson <jmattson@...gle.com>,
        Joerg Roedel <joro@...tes.org>, kvm@...r.kernel.org,
        linux-kernel@...r.kernel.org, Jue Wang <juew@...gle.com>
Subject: [PATCH 2/3] KVM: x86: Use explicit case-statements for MCx banks in {g,s}et_msr_mce()

Use an explicit case statement to grab the full range of MCx bank MSRs
in {g,s}et_msr_mce(), and manually check only the "end" (the number of
banks configured by userspace may be less than the max).  The "default"
trick works, but is a bit odd now, and will be quite odd if/when support
for accessing MCx_CTL2 MSRs is added, which has near identical logic.

Hoist "offset" to function scope so as to avoid curly braces for the case
statement, and because MCx_CTL2 support will need the same variables.

Opportunstically clean up the comment about allowing bit 10 to be cleared
from bank 4.

No functional change intended.

Cc: Jue Wang <juew@...gle.com>
Signed-off-by: Sean Christopherson <seanjc@...gle.com>
---
 arch/x86/kvm/x86.c | 71 ++++++++++++++++++++++++----------------------
 1 file changed, 37 insertions(+), 34 deletions(-)

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index bc6db58975dc..7e685ea9882b 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -3206,6 +3206,7 @@ static int set_msr_mce(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 	unsigned bank_num = mcg_cap & 0xff;
 	u32 msr = msr_info->index;
 	u64 data = msr_info->data;
+	u32 offset, last_msr;
 
 	switch (msr) {
 	case MSR_IA32_MCG_STATUS:
@@ -3219,32 +3220,33 @@ static int set_msr_mce(struct kvm_vcpu *vcpu, struct msr_data *msr_info)
 			return 1;
 		vcpu->arch.mcg_ctl = data;
 		break;
+	case MSR_IA32_MC0_CTL ... MSR_IA32_MCx_CTL(KVM_MAX_MCE_BANKS) - 1:
+		last_msr = MSR_IA32_MCx_CTL(bank_num) - 1;
+		if (msr > last_msr)
+			return 1;
+
+		offset = array_index_nospec(msr - MSR_IA32_MC0_CTL,
+					    last_msr + 1 - MSR_IA32_MC0_CTL);
+
+		/*
+		 * Only 0 or all 1s can be written to IA32_MCi_CTL, all other
+		 * values are architecturally undefined.  But, some Linux
+		 * kernels clear bit 10 in bank 4 to workaround a BIOS/GART TLB
+		 * issue on AMD K8s, allow bit 10 to be clear when setting all
+		 * other bits in order to avoid an uncaught #GP in the guest.
+		 */
+		if ((offset & 0x3) == 0 &&
+		    data != 0 && (data | (1 << 10)) != ~(u64)0)
+			return 1;
+
+		/* MCi_STATUS */
+		if (!msr_info->host_initiated && (offset & 0x3) == 1 &&
+		    data != 0 && !can_set_mci_status(vcpu))
+			return 1;
+
+		vcpu->arch.mce_banks[offset] = data;
+		break;
 	default:
-		if (msr >= MSR_IA32_MC0_CTL &&
-		    msr < MSR_IA32_MCx_CTL(bank_num)) {
-			u32 offset = array_index_nospec(
-				msr - MSR_IA32_MC0_CTL,
-				MSR_IA32_MCx_CTL(bank_num) - MSR_IA32_MC0_CTL);
-
-			/* only 0 or all 1s can be written to IA32_MCi_CTL
-			 * some Linux kernels though clear bit 10 in bank 4 to
-			 * workaround a BIOS/GART TBL issue on AMD K8s, ignore
-			 * this to avoid an uncatched #GP in the guest
-			 */
-			if ((offset & 0x3) == 0 &&
-			    data != 0 && (data | (1 << 10)) != ~(u64)0)
-				return 1;
-
-			/* MCi_STATUS */
-			if (!msr_info->host_initiated &&
-			    (offset & 0x3) == 1 && data != 0) {
-				if (!can_set_mci_status(vcpu))
-					return 1;
-			}
-
-			vcpu->arch.mce_banks[offset] = data;
-			break;
-		}
 		return 1;
 	}
 	return 0;
@@ -3789,6 +3791,7 @@ static int get_msr_mce(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata, bool host)
 	u64 data;
 	u64 mcg_cap = vcpu->arch.mcg_cap;
 	unsigned bank_num = mcg_cap & 0xff;
+	u32 offset, last_msr;
 
 	switch (msr) {
 	case MSR_IA32_P5_MC_ADDR:
@@ -3806,16 +3809,16 @@ static int get_msr_mce(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata, bool host)
 	case MSR_IA32_MCG_STATUS:
 		data = vcpu->arch.mcg_status;
 		break;
-	default:
-		if (msr >= MSR_IA32_MC0_CTL &&
-		    msr < MSR_IA32_MCx_CTL(bank_num)) {
-			u32 offset = array_index_nospec(
-				msr - MSR_IA32_MC0_CTL,
-				MSR_IA32_MCx_CTL(bank_num) - MSR_IA32_MC0_CTL);
+	case MSR_IA32_MC0_CTL ... MSR_IA32_MCx_CTL(KVM_MAX_MCE_BANKS) - 1:
+		last_msr = MSR_IA32_MCx_CTL(bank_num) - 1;
+		if (msr > last_msr)
+			return 1;
 
-			data = vcpu->arch.mce_banks[offset];
-			break;
-		}
+		offset = array_index_nospec(msr - MSR_IA32_MC0_CTL,
+					    last_msr + 1 - MSR_IA32_MC0_CTL);
+		data = vcpu->arch.mce_banks[offset];
+		break;
+	default:
 		return 1;
 	}
 	*pdata = data;
-- 
2.36.0.550.gb090851708-goog