linux-kernel - [RFC PATCH v2 11/35] KVM: selftests: Skip guest memory checks for APIC MMIO mapping

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: <20250923050942.206116-12-Neeraj.Upadhyay@amd.com>
Date: Tue, 23 Sep 2025 10:39:18 +0530
From: Neeraj Upadhyay <Neeraj.Upadhyay@....com>
To: <kvm@...r.kernel.org>, <seanjc@...gle.com>, <pbonzini@...hat.com>
CC: <linux-kernel@...r.kernel.org>, <Thomas.Lendacky@....com>,
	<nikunj@....com>, <Santosh.Shukla@....com>, <Vasant.Hegde@....com>,
	<Suravee.Suthikulpanit@....com>, <bp@...en8.de>, <David.Kaplan@....com>,
	<huibo.wang@....com>, <naveen.rao@....com>, <pgonda@...gle.com>,
	<linux-kselftest@...r.kernel.org>, <shuah@...nel.org>, <tiala@...rosoft.com>
Subject: [RFC PATCH v2 11/35] KVM: selftests: Skip guest memory checks for APIC MMIO mapping

The __virt_pg_map() helper function, used to build guest page tables,
calls vm_is_gpa_protected() to determine the correct memory protection
attributes for a given Guest Physical Address (GPA). This check works
by looking up the GPA in the VM's defined physical memory regions.

This approach fails when mapping the local APIC's base address
(0xfee00000), as this is a Memory-Mapped I/O (MMIO) region, not part of
the guest's RAM. Consequently, the lookup fails, and virt_pg_map()
aborts with the error: "No vm physical memory at 0xfee00000".

Fix this by introducing a special case for the APIC base address. When
the physical address being mapped is APIC_DEFAULT_GPA, bypass the
vm_is_gpa_protected() call. Since the APIC MMIO region is always
treated as shared, explicitly set the shared bit in its Page Table Entry
(PTE) and return immediately.

This allows selftests to correctly map the APIC MMIO region, which is
necessary for tests involving protected guests.

Signed-off-by: Neeraj Upadhyay <Neeraj.Upadhyay@....com>
---
 tools/testing/selftests/kvm/lib/x86/processor.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/tools/testing/selftests/kvm/lib/x86/processor.c b/tools/testing/selftests/kvm/lib/x86/processor.c
index d72eb96efb7c..181f1943c4be 100644
--- a/tools/testing/selftests/kvm/lib/x86/processor.c
+++ b/tools/testing/selftests/kvm/lib/x86/processor.c
@@ -8,6 +8,7 @@
 #include "kvm_util.h"
 #include "processor.h"
 #include "sev.h"
+#include "apic.h"

 #ifndef NUM_INTERRUPTS
 #define NUM_INTERRUPTS 256
@@ -227,6 +228,11 @@ void __virt_pg_map(struct kvm_vm *vm, uint64_t vaddr, uint64_t paddr, int level)
 		    "PTE already present for 4k page at vaddr: 0x%lx", vaddr);
 	*pte = PTE_PRESENT_MASK | PTE_WRITABLE_MASK | (paddr & PHYSICAL_PAGE_MASK);

+	if (paddr == APIC_DEFAULT_GPA) {
+		*pte |= vm->arch.s_bit;
+		return;
+	}
+
 	/*
 	 * Neither SEV nor TDX supports shared page tables, so only the final
 	 * leaf PTE needs manually set the C/S-bit.
-- 
2.34.1