mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Sean Christopherson <seanjc@google.com>
To: Sean Christopherson <seanjc@google.com>,
	Paolo Bonzini <pbonzini@redhat.com>,
	 Thomas Gleixner <tglx@linutronix.de>,
	Borislav Petkov <bp@alien8.de>,
	Peter Zijlstra <peterz@infradead.org>,
	 Josh Poimboeuf <jpoimboe@kernel.org>
Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
	 Pawan Gupta <pawan.kumar.gupta@linux.intel.com>,
	Brendan Jackman <jackmanb@google.com>
Subject: [PATCH v4 6/8] KVM: VMX: Bundle all L1 data cache flush mitigation code together
Date: Thu, 30 Oct 2025 17:30:38 -0700	[thread overview]
Message-ID: <20251031003040.3491385-7-seanjc@google.com> (raw)
In-Reply-To: <20251031003040.3491385-1-seanjc@google.com>

Move vmx_l1d_flush(), vmx_cleanup_l1d_flush(), and the vmentry_l1d_flush
param code up in vmx.c so that all of the L1 data cache flushing code is
bundled together.  This will allow conditioning the mitigation code on
CONFIG_CPU_MITIGATIONS=y with minimal #ifdefs.

No functional change intended.

Reviewed-by: Brendan Jackman <jackmanb@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 arch/x86/kvm/vmx/vmx.c | 174 ++++++++++++++++++++---------------------
 1 file changed, 87 insertions(+), 87 deletions(-)

diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c
index 5af2338c7cb8..55962146fc34 100644
--- a/arch/x86/kvm/vmx/vmx.c
+++ b/arch/x86/kvm/vmx/vmx.c
@@ -302,6 +302,16 @@ static int vmx_setup_l1d_flush(enum vmx_l1d_flush_state l1tf)
 	return 0;
 }
 
+static void vmx_cleanup_l1d_flush(void)
+{
+	if (vmx_l1d_flush_pages) {
+		free_pages((unsigned long)vmx_l1d_flush_pages, L1D_CACHE_ORDER);
+		vmx_l1d_flush_pages = NULL;
+	}
+	/* Restore state so sysfs ignores VMX */
+	l1tf_vmx_mitigation = VMENTER_L1D_FLUSH_AUTO;
+}
+
 static int vmentry_l1d_flush_parse(const char *s)
 {
 	unsigned int i;
@@ -352,6 +362,83 @@ static int vmentry_l1d_flush_get(char *s, const struct kernel_param *kp)
 	return sysfs_emit(s, "%s\n", vmentry_l1d_param[l1tf_vmx_mitigation].option);
 }
 
+/*
+ * Software based L1D cache flush which is used when microcode providing
+ * the cache control MSR is not loaded.
+ *
+ * The L1D cache is 32 KiB on Nehalem and later microarchitectures, but to
+ * flush it is required to read in 64 KiB because the replacement algorithm
+ * is not exactly LRU. This could be sized at runtime via topology
+ * information but as all relevant affected CPUs have 32KiB L1D cache size
+ * there is no point in doing so.
+ */
+static noinstr void vmx_l1d_flush(struct kvm_vcpu *vcpu)
+{
+	int size = PAGE_SIZE << L1D_CACHE_ORDER;
+
+	/*
+	 * This code is only executed when the flush mode is 'cond' or
+	 * 'always'
+	 */
+	if (static_branch_likely(&vmx_l1d_flush_cond)) {
+		bool flush_l1d;
+
+		/*
+		 * Clear the per-vcpu flush bit, it gets set again if the vCPU
+		 * is reloaded, i.e. if the vCPU is scheduled out or if KVM
+		 * exits to userspace, or if KVM reaches one of the unsafe
+		 * VMEXIT handlers, e.g. if KVM calls into the emulator.
+		 */
+		flush_l1d = vcpu->arch.l1tf_flush_l1d;
+		vcpu->arch.l1tf_flush_l1d = false;
+
+		/*
+		 * Clear the per-cpu flush bit, it gets set again from
+		 * the interrupt handlers.
+		 */
+		flush_l1d |= kvm_get_cpu_l1tf_flush_l1d();
+		kvm_clear_cpu_l1tf_flush_l1d();
+
+		if (!flush_l1d)
+			return;
+	}
+
+	vcpu->stat.l1d_flush++;
+
+	if (static_cpu_has(X86_FEATURE_FLUSH_L1D)) {
+		native_wrmsrq(MSR_IA32_FLUSH_CMD, L1D_FLUSH);
+		return;
+	}
+
+	asm volatile(
+		/* First ensure the pages are in the TLB */
+		"xorl	%%eax, %%eax\n"
+		".Lpopulate_tlb:\n\t"
+		"movzbl	(%[flush_pages], %%" _ASM_AX "), %%ecx\n\t"
+		"addl	$4096, %%eax\n\t"
+		"cmpl	%%eax, %[size]\n\t"
+		"jne	.Lpopulate_tlb\n\t"
+		"xorl	%%eax, %%eax\n\t"
+		"cpuid\n\t"
+		/* Now fill the cache */
+		"xorl	%%eax, %%eax\n"
+		".Lfill_cache:\n"
+		"movzbl	(%[flush_pages], %%" _ASM_AX "), %%ecx\n\t"
+		"addl	$64, %%eax\n\t"
+		"cmpl	%%eax, %[size]\n\t"
+		"jne	.Lfill_cache\n\t"
+		"lfence\n"
+		:: [flush_pages] "r" (vmx_l1d_flush_pages),
+		    [size] "r" (size)
+		: "eax", "ebx", "ecx", "edx");
+}
+
+static const struct kernel_param_ops vmentry_l1d_flush_ops = {
+	.set = vmentry_l1d_flush_set,
+	.get = vmentry_l1d_flush_get,
+};
+module_param_cb(vmentry_l1d_flush, &vmentry_l1d_flush_ops, NULL, 0644);
+
 static __always_inline void vmx_disable_fb_clear(struct vcpu_vmx *vmx)
 {
 	u64 msr;
@@ -404,12 +491,6 @@ static void vmx_update_fb_clear_dis(struct kvm_vcpu *vcpu, struct vcpu_vmx *vmx)
 		vmx->disable_fb_clear = false;
 }
 
-static const struct kernel_param_ops vmentry_l1d_flush_ops = {
-	.set = vmentry_l1d_flush_set,
-	.get = vmentry_l1d_flush_get,
-};
-module_param_cb(vmentry_l1d_flush, &vmentry_l1d_flush_ops, NULL, 0644);
-
 static u32 vmx_segment_access_rights(struct kvm_segment *var);
 
 void vmx_vmexit(void);
@@ -6672,77 +6753,6 @@ int vmx_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath)
 	return ret;
 }
 
-/*
- * Software based L1D cache flush which is used when microcode providing
- * the cache control MSR is not loaded.
- *
- * The L1D cache is 32 KiB on Nehalem and later microarchitectures, but to
- * flush it is required to read in 64 KiB because the replacement algorithm
- * is not exactly LRU. This could be sized at runtime via topology
- * information but as all relevant affected CPUs have 32KiB L1D cache size
- * there is no point in doing so.
- */
-static noinstr void vmx_l1d_flush(struct kvm_vcpu *vcpu)
-{
-	int size = PAGE_SIZE << L1D_CACHE_ORDER;
-
-	/*
-	 * This code is only executed when the flush mode is 'cond' or
-	 * 'always'
-	 */
-	if (static_branch_likely(&vmx_l1d_flush_cond)) {
-		bool flush_l1d;
-
-		/*
-		 * Clear the per-vcpu flush bit, it gets set again if the vCPU
-		 * is reloaded, i.e. if the vCPU is scheduled out or if KVM
-		 * exits to userspace, or if KVM reaches one of the unsafe
-		 * VMEXIT handlers, e.g. if KVM calls into the emulator.
-		 */
-		flush_l1d = vcpu->arch.l1tf_flush_l1d;
-		vcpu->arch.l1tf_flush_l1d = false;
-
-		/*
-		 * Clear the per-cpu flush bit, it gets set again from
-		 * the interrupt handlers.
-		 */
-		flush_l1d |= kvm_get_cpu_l1tf_flush_l1d();
-		kvm_clear_cpu_l1tf_flush_l1d();
-
-		if (!flush_l1d)
-			return;
-	}
-
-	vcpu->stat.l1d_flush++;
-
-	if (static_cpu_has(X86_FEATURE_FLUSH_L1D)) {
-		native_wrmsrq(MSR_IA32_FLUSH_CMD, L1D_FLUSH);
-		return;
-	}
-
-	asm volatile(
-		/* First ensure the pages are in the TLB */
-		"xorl	%%eax, %%eax\n"
-		".Lpopulate_tlb:\n\t"
-		"movzbl	(%[flush_pages], %%" _ASM_AX "), %%ecx\n\t"
-		"addl	$4096, %%eax\n\t"
-		"cmpl	%%eax, %[size]\n\t"
-		"jne	.Lpopulate_tlb\n\t"
-		"xorl	%%eax, %%eax\n\t"
-		"cpuid\n\t"
-		/* Now fill the cache */
-		"xorl	%%eax, %%eax\n"
-		".Lfill_cache:\n"
-		"movzbl	(%[flush_pages], %%" _ASM_AX "), %%ecx\n\t"
-		"addl	$64, %%eax\n\t"
-		"cmpl	%%eax, %[size]\n\t"
-		"jne	.Lfill_cache\n\t"
-		"lfence\n"
-		:: [flush_pages] "r" (vmx_l1d_flush_pages),
-		    [size] "r" (size)
-		: "eax", "ebx", "ecx", "edx");
-}
-
 void vmx_update_cr8_intercept(struct kvm_vcpu *vcpu, int tpr, int irr)
 {
 	struct vmcs12 *vmcs12 = get_vmcs12(vcpu);
@@ -8677,16 +8687,6 @@ __init int vmx_hardware_setup(void)
 	return r;
 }
 
-static void vmx_cleanup_l1d_flush(void)
-{
-	if (vmx_l1d_flush_pages) {
-		free_pages((unsigned long)vmx_l1d_flush_pages, L1D_CACHE_ORDER);
-		vmx_l1d_flush_pages = NULL;
-	}
-	/* Restore state so sysfs ignores VMX */
-	l1tf_vmx_mitigation = VMENTER_L1D_FLUSH_AUTO;
-}
-
 void vmx_exit(void)
 {
 	allow_smaller_maxphyaddr = false;
-- 
2.51.1.930.gacf6e81ea2-goog


  parent reply	other threads:[~2025-10-31  0:30 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-10-31  0:30 [PATCH v4 0/8] x86/bugs: KVM: L1TF and MMIO Stale Data cleanups Sean Christopherson
2025-10-31  0:30 ` [PATCH v4 1/8] x86/bugs: Use VM_CLEAR_CPU_BUFFERS in VMX as well Sean Christopherson
2025-10-31 11:30   ` Brendan Jackman
2025-11-01  1:46     ` Pawan Gupta
2025-11-03 18:18   ` Pawan Gupta
2025-11-07 19:05     ` Borislav Petkov
2025-11-11 22:03       ` Sean Christopherson
2025-11-12 10:23         ` Borislav Petkov
2025-11-12 18:19           ` Pawan Gupta
2025-11-12 18:17       ` Pawan Gupta
2025-11-07 18:59   ` Borislav Petkov
2025-11-12 18:02     ` Pawan Gupta
2025-10-31  0:30 ` [PATCH v4 2/8] x86/bugs: Decouple ALTERNATIVE usage from VERW macro definition Sean Christopherson
2025-10-31 11:37   ` Brendan Jackman
2025-10-31 17:43     ` Sean Christopherson
2025-11-01  4:13   ` Pawan Gupta
2025-11-03 17:00     ` Sean Christopherson
2025-11-03 17:40       ` Pawan Gupta
2025-11-12 12:15       ` Borislav Petkov
2025-10-31  0:30 ` [PATCH v4 3/8] x86/bugs: Use an X86_FEATURE_xxx flag for the MMIO Stale Data mitigation Sean Christopherson
2025-10-31 11:44   ` Brendan Jackman
2025-10-31 21:47     ` Sean Christopherson
2025-11-03 10:49       ` Brendan Jackman
2025-10-31 22:28   ` Pawan Gupta
2025-10-31 22:37     ` Sean Christopherson
2025-10-31 22:50       ` Pawan Gupta
2025-11-12 14:46   ` Borislav Petkov
2025-11-12 18:24     ` Pawan Gupta
2025-10-31  0:30 ` [PATCH v4 4/8] KVM: VMX: Handle MMIO Stale Data in VM-Enter assembly via ALTERNATIVES_2 Sean Christopherson
2025-10-31 12:32   ` Brendan Jackman
2025-10-31 21:44     ` Sean Christopherson
2025-11-03 10:51       ` Brendan Jackman
2025-10-31 23:55   ` Pawan Gupta
2025-11-01  3:41     ` Pawan Gupta
2025-11-03  9:17     ` Peter Zijlstra
2025-11-03 17:37       ` Pawan Gupta
2025-11-03 17:46   ` Pawan Gupta
2025-11-12 16:41   ` Borislav Petkov
2025-11-12 17:15     ` Sean Christopherson
2025-11-12 18:38       ` Borislav Petkov
2025-11-12 20:30         ` Sean Christopherson
2025-11-12 23:01           ` Pawan Gupta
2025-11-13 14:20           ` Borislav Petkov
2025-11-13 22:01             ` Sean Christopherson
2025-10-31  0:30 ` [PATCH v4 5/8] x86/bugs: KVM: Move VM_CLEAR_CPU_BUFFERS into SVM as SVM_CLEAR_CPU_BUFFERS Sean Christopherson
2025-10-31 12:34   ` Brendan Jackman
2025-11-13 15:03   ` Borislav Petkov
2025-11-13 15:37     ` Sean Christopherson
2025-11-13 16:19       ` Borislav Petkov
2025-10-31  0:30 ` Sean Christopherson [this message]
2025-11-03 18:26   ` [PATCH v4 6/8] KVM: VMX: Bundle all L1 data cache flush mitigation code together Pawan Gupta
2025-10-31  0:30 ` [PATCH v4 7/8] KVM: VMX: Disable L1TF L1 data cache flush if CONFIG_CPU_MITIGATIONS=n Sean Christopherson
2025-10-31 12:37   ` Brendan Jackman
2025-10-31  0:30 ` [PATCH v4 8/8] KVM: x86: Unify L1TF flushing under per-CPU variable Sean Christopherson
2025-10-31 11:22 ` [PATCH v4 0/8] x86/bugs: KVM: L1TF and MMIO Stale Data cleanups Brendan Jackman
2025-10-31 17:36   ` Sean Christopherson
2025-11-04 10:58     ` Brendan Jackman

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=20251031003040.3491385-7-seanjc@google.com \
    --to=seanjc@google.com \
    --cc=bp@alien8.de \
    --cc=jackmanb@google.com \
    --cc=jpoimboe@kernel.org \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pawan.kumar.gupta@linux.intel.com \
    --cc=pbonzini@redhat.com \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    /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

all inboxes | Powered by JetHome®