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, 108 insertions, 0 deletions

diff --git a/io_uring/advise.c b/io_uring/advise.c
new file mode 100644
index 000000000..0073f74e3
--- /dev/null
+++ b/io_uring/advise.c
@@ -0,0 +1,108 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/fs.h>
+#include <linux/file.h>
+#include <linux/mm.h>
+#include <linux/slab.h>
+#include <linux/namei.h>
+#include <linux/io_uring.h>
+
+#include <uapi/linux/fadvise.h>
+#include <uapi/linux/io_uring.h>
+
+#include "io_uring.h"
+#include "advise.h"
+
+struct io_fadvise {
+	struct file			*file;
+	u64				offset;
+	u64				len;
+	u32				advice;
+};
+
+struct io_madvise {
+	struct file			*file;
+	u64				addr;
+	u64				len;
+	u32				advice;
+};
+
+int io_madvise_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+#if defined(CONFIG_ADVISE_SYSCALLS) && defined(CONFIG_MMU)
+	struct io_madvise *ma = io_kiocb_to_cmd(req, struct io_madvise);
+
+	if (sqe->buf_index || sqe->splice_fd_in)
+		return -EINVAL;
+
+	ma->addr = READ_ONCE(sqe->addr);
+	ma->len = READ_ONCE(sqe->off);
+	if (!ma->len)
+		ma->len = READ_ONCE(sqe->len);
+	ma->advice = READ_ONCE(sqe->fadvise_advice);
+	req->flags |= REQ_F_FORCE_ASYNC;
+	return 0;
+#else
+	return -EOPNOTSUPP;
+#endif
+}
+
+int io_madvise(struct io_kiocb *req, unsigned int issue_flags)
+{
+#if defined(CONFIG_ADVISE_SYSCALLS) && defined(CONFIG_MMU)
+	struct io_madvise *ma = io_kiocb_to_cmd(req, struct io_madvise);
+	int ret;
+
+	WARN_ON_ONCE(issue_flags & IO_URING_F_NONBLOCK);
+
+	ret = do_madvise(current->mm, ma->addr, ma->len, ma->advice);
+	io_req_set_res(req, ret, 0);
+	return IOU_COMPLETE;
+#else
+	return -EOPNOTSUPP;
+#endif
+}
+
+static bool io_fadvise_force_async(struct io_fadvise *fa)
+{
+	switch (fa->advice) {
+	case POSIX_FADV_NORMAL:
+	case POSIX_FADV_RANDOM:
+	case POSIX_FADV_SEQUENTIAL:
+		return false;
+	default:
+		return true;
+	}
+}
+
+int io_fadvise_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
+{
+	struct io_fadvise *fa = io_kiocb_to_cmd(req, struct io_fadvise);
+
+	if (sqe->buf_index || sqe->splice_fd_in)
+		return -EINVAL;
+
+	fa->offset = READ_ONCE(sqe->off);
+	fa->len = READ_ONCE(sqe->addr);
+	if (!fa->len)
+		fa->len = READ_ONCE(sqe->len);
+	fa->advice = READ_ONCE(sqe->fadvise_advice);
+	if (io_fadvise_force_async(fa))
+		req->flags |= REQ_F_FORCE_ASYNC;
+	return 0;
+}
+
+int io_fadvise(struct io_kiocb *req, unsigned int issue_flags)
+{
+	struct io_fadvise *fa = io_kiocb_to_cmd(req, struct io_fadvise);
+	int ret;
+
+	WARN_ON_ONCE(issue_flags & IO_URING_F_NONBLOCK && io_fadvise_force_async(fa));
+
+	ret = vfs_fadvise(req->file, fa->offset, fa->len, fa->advice);
+	if (ret < 0)
+		req_set_fail(req);
+	io_req_set_res(req, ret, 0);
+	return IOU_COMPLETE;
+}