| 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
| ||
|
Message-Id: <20220720011350.1024134-22-sashal@kernel.org>
Date: Tue, 19 Jul 2022 21:13:30 -0400
From: Sasha Levin <sashal@...nel.org>
To: linux-kernel@...r.kernel.org, stable@...r.kernel.org
Cc: Josh Poimboeuf <jpoimboe@...nel.org>,
Peter Zijlstra <peterz@...radead.org>,
Borislav Petkov <bp@...e.de>, Sasha Levin <sashal@...nel.org>,
tglx@...utronix.de, mingo@...hat.com, bp@...en8.de,
dave.hansen@...ux.intel.com, x86@...nel.org, seanjc@...gle.com,
pbonzini@...hat.com, pawan.kumar.gupta@...ux.intel.com,
sblbir@...zon.com, kim.phillips@....com,
alexandre.chartre@...cle.com, kvm@...r.kernel.org
Subject: [PATCH AUTOSEL 5.15 22/42] KVM: VMX: Prevent guest RSB poisoning attacks with eIBRS
From: Josh Poimboeuf <jpoimboe@...nel.org>
[ Upstream commit fc02735b14fff8c6678b521d324ade27b1a3d4cf ]
On eIBRS systems, the returns in the vmexit return path from
__vmx_vcpu_run() to vmx_vcpu_run() are exposed to RSB poisoning attacks.
Fix that by moving the post-vmexit spec_ctrl handling to immediately
after the vmexit.
Signed-off-by: Josh Poimboeuf <jpoimboe@...nel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
Signed-off-by: Borislav Petkov <bp@...e.de>
Signed-off-by: Sasha Levin <sashal@...nel.org>
---
arch/x86/include/asm/nospec-branch.h | 1 +
arch/x86/kernel/cpu/bugs.c | 4 +++
arch/x86/kvm/vmx/run_flags.h | 1 +
arch/x86/kvm/vmx/vmenter.S | 49 +++++++++++++++++++++-------
arch/x86/kvm/vmx/vmx.c | 48 +++++++++++++++------------
arch/x86/kvm/vmx/vmx.h | 1 +
6 files changed, 73 insertions(+), 31 deletions(-)
diff --git a/arch/x86/include/asm/nospec-branch.h b/arch/x86/include/asm/nospec-branch.h
index 6189609bede3..4ef9cc93d584 100644
--- a/arch/x86/include/asm/nospec-branch.h
+++ b/arch/x86/include/asm/nospec-branch.h
@@ -220,6 +220,7 @@ static inline void indirect_branch_prediction_barrier(void)
/* The Intel SPEC CTRL MSR base value cache */
extern u64 x86_spec_ctrl_base;
+extern u64 x86_spec_ctrl_current;
extern void write_spec_ctrl_current(u64 val, bool force);
/*
diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c
index c59db48472dc..13a2ee4737f6 100644
--- a/arch/x86/kernel/cpu/bugs.c
+++ b/arch/x86/kernel/cpu/bugs.c
@@ -199,6 +199,10 @@ void __init check_bugs(void)
#endif
}
+/*
+ * NOTE: For VMX, this function is not called in the vmexit path.
+ * It uses vmx_spec_ctrl_restore_host() instead.
+ */
void
x86_virt_spec_ctrl(u64 guest_spec_ctrl, u64 guest_virt_spec_ctrl, bool setguest)
{
diff --git a/arch/x86/kvm/vmx/run_flags.h b/arch/x86/kvm/vmx/run_flags.h
index 57f4c664ea9c..edc3f16cc189 100644
--- a/arch/x86/kvm/vmx/run_flags.h
+++ b/arch/x86/kvm/vmx/run_flags.h
@@ -3,5 +3,6 @@
#define __KVM_X86_VMX_RUN_FLAGS_H
#define VMX_RUN_VMRESUME (1 << 0)
+#define VMX_RUN_SAVE_SPEC_CTRL (1 << 1)
#endif /* __KVM_X86_VMX_RUN_FLAGS_H */
diff --git a/arch/x86/kvm/vmx/vmenter.S b/arch/x86/kvm/vmx/vmenter.S
index ddc3bf85db33..8641ea74a307 100644
--- a/arch/x86/kvm/vmx/vmenter.S
+++ b/arch/x86/kvm/vmx/vmenter.S
@@ -33,9 +33,10 @@
/**
* __vmx_vcpu_run - Run a vCPU via a transition to VMX guest mode
- * @vmx: struct vcpu_vmx * (forwarded to vmx_update_host_rsp)
+ * @vmx: struct vcpu_vmx *
* @regs: unsigned long * (to guest registers)
- * @flags: VMX_RUN_VMRESUME: use VMRESUME instead of VMLAUNCH
+ * @flags: VMX_RUN_VMRESUME: use VMRESUME instead of VMLAUNCH
+ * VMX_RUN_SAVE_SPEC_CTRL: save guest SPEC_CTRL into vmx->spec_ctrl
*
* Returns:
* 0 on VM-Exit, 1 on VM-Fail
@@ -54,6 +55,12 @@ SYM_FUNC_START(__vmx_vcpu_run)
#endif
push %_ASM_BX
+ /* Save @vmx for SPEC_CTRL handling */
+ push %_ASM_ARG1
+
+ /* Save @flags for SPEC_CTRL handling */
+ push %_ASM_ARG3
+
/*
* Save @regs, _ASM_ARG2 may be modified by vmx_update_host_rsp() and
* @regs is needed after VM-Exit to save the guest's register values.
@@ -149,25 +156,23 @@ SYM_INNER_LABEL(vmx_vmexit, SYM_L_GLOBAL)
mov %r15, VCPU_R15(%_ASM_AX)
#endif
- /* IMPORTANT: RSB must be stuffed before the first return. */
- FILL_RETURN_BUFFER %_ASM_BX, RSB_CLEAR_LOOPS, X86_FEATURE_RETPOLINE
-
- /* Clear RAX to indicate VM-Exit (as opposed to VM-Fail). */
- xor %eax, %eax
+ /* Clear return value to indicate VM-Exit (as opposed to VM-Fail). */
+ xor %ebx, %ebx
.Lclear_regs:
/*
- * Clear all general purpose registers except RSP and RAX to prevent
+ * Clear all general purpose registers except RSP and RBX to prevent
* speculative use of the guest's values, even those that are reloaded
* via the stack. In theory, an L1 cache miss when restoring registers
* could lead to speculative execution with the guest's values.
* Zeroing XORs are dirt cheap, i.e. the extra paranoia is essentially
* free. RSP and RAX are exempt as RSP is restored by hardware during
- * VM-Exit and RAX is explicitly loaded with 0 or 1 to return VM-Fail.
+ * VM-Exit and RBX is explicitly loaded with 0 or 1 to hold the return
+ * value.
*/
+ xor %eax, %eax
xor %ecx, %ecx
xor %edx, %edx
- xor %ebx, %ebx
xor %ebp, %ebp
xor %esi, %esi
xor %edi, %edi
@@ -185,6 +190,28 @@ SYM_INNER_LABEL(vmx_vmexit, SYM_L_GLOBAL)
/* "POP" @regs. */
add $WORD_SIZE, %_ASM_SP
+ /*
+ * IMPORTANT: RSB filling and SPEC_CTRL handling must be done before
+ * the first unbalanced RET after vmexit!
+ *
+ * For retpoline, RSB filling is needed to prevent poisoned RSB entries
+ * and (in some cases) RSB underflow.
+ *
+ * eIBRS has its own protection against poisoned RSB, so it doesn't
+ * need the RSB filling sequence. But it does need to be enabled
+ * before the first unbalanced RET.
+ */
+
+ FILL_RETURN_BUFFER %_ASM_CX, RSB_CLEAR_LOOPS, X86_FEATURE_RETPOLINE
+
+ pop %_ASM_ARG2 /* @flags */
+ pop %_ASM_ARG1 /* @vmx */
+
+ call vmx_spec_ctrl_restore_host
+
+ /* Put return value in AX */
+ mov %_ASM_BX, %_ASM_AX
+
pop %_ASM_BX
#ifdef CONFIG_X86_64
pop %r12
@@ -204,7 +231,7 @@ SYM_INNER_LABEL(vmx_vmexit, SYM_L_GLOBAL)
ud2
.Lvmfail:
/* VM-Fail: set return value to 1 */
- mov $1, %eax
+ mov $1, %_ASM_BX
jmp .Lclear_regs
SYM_FUNC_END(__vmx_vcpu_run)
diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index bac08c265ddf..f668df996668 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -842,6 +842,14 @@ unsigned int __vmx_vcpu_run_flags(struct vcpu_vmx *vmx)
if (vmx->loaded_vmcs->launched)
flags |= VMX_RUN_VMRESUME;
+ /*
+ * If writes to the SPEC_CTRL MSR aren't intercepted, the guest is free
+ * to change it directly without causing a vmexit. In that case read
+ * it after vmexit and store it in vmx->spec_ctrl.
+ */
+ if (unlikely(!msr_write_intercepted(vmx, MSR_IA32_SPEC_CTRL)))
+ flags |= VMX_RUN_SAVE_SPEC_CTRL;
+
return flags;
}
@@ -6664,6 +6672,26 @@ void noinstr vmx_update_host_rsp(struct vcpu_vmx *vmx, unsigned long host_rsp)
}
}
+void noinstr vmx_spec_ctrl_restore_host(struct vcpu_vmx *vmx,
+ unsigned int flags)
+{
+ u64 hostval = this_cpu_read(x86_spec_ctrl_current);
+
+ if (!cpu_feature_enabled(X86_FEATURE_MSR_SPEC_CTRL))
+ return;
+
+ if (flags & VMX_RUN_SAVE_SPEC_CTRL)
+ vmx->spec_ctrl = __rdmsr(MSR_IA32_SPEC_CTRL);
+
+ /*
+ * If the guest/host SPEC_CTRL values differ, restore the host value.
+ */
+ if (vmx->spec_ctrl != hostval)
+ native_wrmsrl(MSR_IA32_SPEC_CTRL, hostval);
+
+ barrier_nospec();
+}
+
static fastpath_t vmx_exit_handlers_fastpath(struct kvm_vcpu *vcpu)
{
switch (to_vmx(vcpu)->exit_reason.basic) {
@@ -6799,26 +6827,6 @@ static fastpath_t vmx_vcpu_run(struct kvm_vcpu *vcpu)
/* The actual VMENTER/EXIT is in the .noinstr.text section. */
vmx_vcpu_enter_exit(vcpu, vmx, __vmx_vcpu_run_flags(vmx));
- /*
- * We do not use IBRS in the kernel. If this vCPU has used the
- * SPEC_CTRL MSR it may have left it on; save the value and
- * turn it off. This is much more efficient than blindly adding
- * it to the atomic save/restore list. Especially as the former
- * (Saving guest MSRs on vmexit) doesn't even exist in KVM.
- *
- * For non-nested case:
- * If the L01 MSR bitmap does not intercept the MSR, then we need to
- * save it.
- *
- * For nested case:
- * If the L02 MSR bitmap does not intercept the MSR, then we need to
- * save it.
- */
- if (unlikely(!msr_write_intercepted(vmx, MSR_IA32_SPEC_CTRL)))
- vmx->spec_ctrl = native_read_msr(MSR_IA32_SPEC_CTRL);
-
- x86_spec_ctrl_restore_host(vmx->spec_ctrl, 0);
-
/* All fields are clean at this point */
if (static_branch_unlikely(&enable_evmcs)) {
current_evmcs->hv_clean_fields |=
diff --git a/arch/x86/kvm/vmx/vmx.h b/arch/x86/kvm/vmx/vmx.h
index 35476115cbd0..a8b8150252bb 100644
--- a/arch/x86/kvm/vmx/vmx.h
+++ b/arch/x86/kvm/vmx/vmx.h
@@ -383,6 +383,7 @@ void vmx_set_virtual_apic_mode(struct kvm_vcpu *vcpu);
struct vmx_uret_msr *vmx_find_uret_msr(struct vcpu_vmx *vmx, u32 msr);
void pt_update_intercept_for_msr(struct kvm_vcpu *vcpu);
void vmx_update_host_rsp(struct vcpu_vmx *vmx, unsigned long host_rsp);
+void vmx_spec_ctrl_restore_host(struct vcpu_vmx *vmx, unsigned int flags);
unsigned int __vmx_vcpu_run_flags(struct vcpu_vmx *vmx);
bool __vmx_vcpu_run(struct vcpu_vmx *vmx, unsigned long *regs,
unsigned int flags);
--
2.35.1
Powered by blists - more mailing lists