KVM: SEV: Do cache maintenance on the source VM during intra-host migrationgrafted

Manually perform cache maintenance on the source VM during intra-host
migration to ensure no stale data is left in CPU caches after the VM is
destroyed.  Because the source VM is "converted" to a non-SEV VM, KVM's
memory reclaim flows won't trigger cache maintenance, e.g. when all guest
memory is reclaimed in response to detaching from the mmu_notifier.

Note, relying on the destination VM to do cache maintenance isn't an option
as KVM doesn't require identical guest memory configurations, i.e. the
source VM may have access to memory that the destination VM does not.
Enforcing equivalent memory configurations is infeasible, as it would
require a *deep* comparison of memslots, e.g. to verify that not only are
the memslot identical, but what the memslots point at is also identical.

Fixes: b56639318bb2 ("KVM: SEV: Add support for SEV intra host migration")
Cc: stable@vger.kernel.org
Reported-by: Stefan Teodorescu <fane@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-ID: <20260923163721.1584779-3-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

1 files changed, 45 insertions, 0 deletions

diff --git a/include/uapi/asm-generic/sembuf.h b/include/uapi/asm-generic/sembuf.h
new file mode 100644
index 000000000..f54e48fc9
--- /dev/null
+++ b/include/uapi/asm-generic/sembuf.h
@@ -0,0 +1,45 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+#ifndef __ASM_GENERIC_SEMBUF_H
+#define __ASM_GENERIC_SEMBUF_H
+
+#include <asm/bitsperlong.h>
+#include <asm/ipcbuf.h>
+
+/*
+ * The semid64_ds structure for most architectures (though it came from x86_32
+ * originally). Note extra padding because this structure is passed back and
+ * forth between kernel and user space.
+ *
+ * semid64_ds was originally meant to be architecture specific, but
+ * everyone just ended up making identical copies without specific
+ * optimizations, so we may just as well all use the same one.
+ *
+ * 64 bit architectures use a 64-bit long time field here, while
+ * 32 bit architectures have a pair of unsigned long values.
+ *
+ * On big-endian systems, the padding is in the wrong place for
+ * historic reasons, so user space has to reconstruct a time_t
+ * value using
+ *
+ * user_semid_ds.sem_otime = kernel_semid64_ds.sem_otime +
+ *		((long long)kernel_semid64_ds.sem_otime_high << 32)
+ *
+ * Pad space is left for 2 miscellaneous 32-bit values
+ */
+struct semid64_ds {
+	struct ipc64_perm sem_perm;	/* permissions .. see ipc.h */
+#if __BITS_PER_LONG == 64
+	long		sem_otime;	/* last semop time */
+	long		sem_ctime;	/* last change time */
+#else
+	unsigned long	sem_otime;	/* last semop time */
+	unsigned long	sem_otime_high;
+	unsigned long	sem_ctime;	/* last change time */
+	unsigned long	sem_ctime_high;
+#endif
+	unsigned long	sem_nsems;	/* no. of semaphores in array */
+	unsigned long	__unused3;
+	unsigned long	__unused4;
+};
+
+#endif /* __ASM_GENERIC_SEMBUF_H */