Skip to contentSkip to conversation

x86/svm: work around Virtual VMLOAD/VMSAVE bug on Naples and Rome

PATCH Applied Igor Mammedov 2 participants 2 messages May 19 – Sep 25 KVM

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 x86/svm: work around Virtual VMLOAD/VMSAVE bug on Naples and Rome

AMD Family 17h models 0x00-0x0f (Naples/Zen+) and 0x30-0x3f (Rome/Zen2) have a hardware bug where Virtual VMLOAD/VMSAVE causes spurious VMEXITs on VMLOAD/VMSAVE even with intercepts disabled, when the VMCB physical address as seen by L1 falls in [0x78000000, 0x80000000) range. Reserve this range on affected CPUs so the page allocator would never allocate the VMCB there.

Given that it's relatively old CPUs + nested env + only performance impact, it's not worth fixing on KVM side (which could involve messing with allocator or reallocating VMCB, until it's not in affected range). Hence a quirk here, to prevent tests failures where we can't do anything about them.

---
there is not official EOL dates on both, but using a newer generation(s) release dates it appears that both are effectively discontinued for ~3-5 years

v2:
* Sean Christopherson <[email protected]>
- s/has_vls_bug/has_vls_spurious_vmexit_bug/

---
 x86/svm.c | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)
x86/svm.c +26 -0
index a85da905..3f2c7390 100644
@@ -334,6 +334,25 @@ static void setup_npt(void)
 	__setup_mmu_range(pml4e, 0, size, X86_MMU_MAP_USER);
 }
 
+#define VLS_SPURIOUS_VMEXIT_START	0x78000000ULL
+#define VLS_SPURIOUS_VMEXIT_END	0x80000000ULL
+
+/*
+ * Naples (Fam17h/00-0f) and Rome (Fam17h/30-3f) cause spurious VMEXITs
+ * on VMLOAD/VMSAVE when the VMCB address falls in [0x78000000, 0x80000000).
+ */
+static bool has_vls_spurious_vmexit_bug(void)
+{
+	u32 sig = cpuid(1).a;
+	u32 fam = x86_family(sig);
+	u32 model = x86_model(sig);
+
+	if (fam != 0x17)
+		return false;
+
+	return model <= 0x0f || (model >= 0x30 && model <= 0x3f);
+}
+
 static void setup_svm(void)
 {
 	void *hsave = alloc_page();
@@ -413,6 +432,13 @@ int run_svm_tests(int ac, char **av, struct svm_test *svm_tests)
 		return report_summary();
 	}
 
+	if (has_vls_spurious_vmexit_bug()) {
+		phys_addr_t addr;
+
+		for (addr = VLS_SPURIOUS_VMEXIT_START; addr < VLS_SPURIOUS_VMEXIT_END; addr += PAGE_SIZE)
+			reserve_pages(addr, 1);
+	}
+
 	setup_svm();
 
 	vmcb = alloc_page();
Signature · 1 line
2.43.0
reviewer Applied
#1 lore raw
Quoted 8 lines · Igor Mammedov

AMD Family 17h models 0x00-0x0f (Naples/Zen+) and 0x30-0x3f (Rome/Zen2) have a hardware bug where Virtual VMLOAD/VMSAVE causes spurious VMEXITs on VMLOAD/VMSAVE even with intercepts disabled, when the VMCB physical address as seen by L1 falls in [0x78000000, 0x80000000) range. Reserve this range on affected CPUs so the page allocator would never allocate the VMCB there.

[...]

Applied to kvm-x86 next, thanks!

[1/1] x86/svm: work around Virtual VMLOAD/VMSAVE bug on Naples and Rome

      https://github.com/kvm-x86/kvm-unit-tests/commit/1cb376031ed1

--
https://github.com/kvm-x86/kvm-unit-tests/tree/next