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, 174 insertions, 0 deletions

diff --git a/drivers/md/bcache/io.c b/drivers/md/bcache/io.c
new file mode 100644
index 000000000..2386d08bf
--- /dev/null
+++ b/drivers/md/bcache/io.c
@@ -0,0 +1,174 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Some low level IO code, and hacks for various block layer limitations
+ *
+ * Copyright 2010, 2011 Kent Overstreet <kent.overstreet@gmail.com>
+ * Copyright 2012 Google, Inc.
+ */
+
+#include "bcache.h"
+#include "bset.h"
+#include "debug.h"
+
+#include <linux/blkdev.h>
+
+/* Bios with headers */
+
+void bch_bbio_free(struct bio *bio, struct cache_set *c)
+{
+	struct bbio *b = container_of(bio, struct bbio, bio);
+
+	mempool_free(b, &c->bio_meta);
+}
+
+struct bio *bch_bbio_alloc(struct cache_set *c)
+{
+	struct bbio *b = mempool_alloc(&c->bio_meta, GFP_NOIO);
+	struct bio *bio = &b->bio;
+
+	bio_init_inline(bio, NULL, meta_bucket_pages(&c->cache->sb), 0);
+
+	return bio;
+}
+
+void __bch_submit_bbio(struct bio *bio, struct cache_set *c)
+{
+	struct bbio *b = container_of(bio, struct bbio, bio);
+
+	bio->bi_iter.bi_sector	= PTR_OFFSET(&b->key, 0);
+	bio_set_dev(bio, c->cache->bdev);
+
+	b->submit_time_us = local_clock_us();
+	closure_bio_submit(c, bio, bio->bi_private);
+}
+
+void bch_submit_bbio(struct bio *bio, struct cache_set *c,
+		     struct bkey *k, unsigned int ptr)
+{
+	struct bbio *b = container_of(bio, struct bbio, bio);
+
+	bch_bkey_copy_single_ptr(&b->key, k, ptr);
+	__bch_submit_bbio(bio, c);
+}
+
+/* IO errors */
+void bch_count_backing_io_errors(struct cached_dev *dc, struct bio *bio)
+{
+	unsigned int errors;
+
+	WARN_ONCE(!dc, "NULL pointer of struct cached_dev");
+
+	/*
+	 * Read-ahead requests on a degrading and recovering md raid
+	 * (e.g. raid6) device might be failured immediately by md
+	 * raid code, which is not a real hardware media failure. So
+	 * we shouldn't count failed REQ_RAHEAD bio to dc->io_errors.
+	 */
+	if (bio->bi_opf & REQ_RAHEAD) {
+		pr_warn_ratelimited("%pg: Read-ahead I/O failed on backing device, ignore\n",
+				    dc->bdev);
+		return;
+	}
+
+	errors = atomic_add_return(1, &dc->io_errors);
+	if (errors < dc->error_limit)
+		pr_err("%pg: IO error on backing device, unrecoverable\n",
+			dc->bdev);
+	else
+		bch_cached_dev_error(dc);
+}
+
+void bch_count_io_errors(struct cache *ca,
+			 blk_status_t error,
+			 int is_read,
+			 const char *m)
+{
+	/*
+	 * The halflife of an error is:
+	 * log2(1/2)/log2(127/128) * refresh ~= 88 * refresh
+	 */
+
+	if (ca->set->error_decay) {
+		unsigned int count = atomic_inc_return(&ca->io_count);
+
+		while (count > ca->set->error_decay) {
+			unsigned int errors;
+			unsigned int old = count;
+			unsigned int new = count - ca->set->error_decay;
+
+			/*
+			 * First we subtract refresh from count; each time we
+			 * successfully do so, we rescale the errors once:
+			 */
+
+			count = atomic_cmpxchg(&ca->io_count, old, new);
+
+			if (count == old) {
+				count = new;
+
+				errors = atomic_read(&ca->io_errors);
+				do {
+					old = errors;
+					new = ((uint64_t) errors * 127) / 128;
+					errors = atomic_cmpxchg(&ca->io_errors,
+								old, new);
+				} while (old != errors);
+			}
+		}
+	}
+
+	if (error) {
+		unsigned int errors = atomic_add_return(1 << IO_ERROR_SHIFT,
+						    &ca->io_errors);
+		errors >>= IO_ERROR_SHIFT;
+
+		if (errors < ca->set->error_limit)
+			pr_err("%pg: IO error on %s%s\n",
+			       ca->bdev, m,
+			       is_read ? ", recovering." : ".");
+		else
+			bch_cache_set_error(ca->set,
+					    "%pg: too many IO errors %s\n",
+					    ca->bdev, m);
+	}
+}
+
+void bch_bbio_count_io_errors(struct cache_set *c, struct bio *bio,
+			      blk_status_t error, const char *m)
+{
+	struct bbio *b = container_of(bio, struct bbio, bio);
+	struct cache *ca = c->cache;
+	int is_read = (bio_data_dir(bio) == READ ? 1 : 0);
+
+	unsigned int threshold = op_is_write(bio_op(bio))
+		? c->congested_write_threshold_us
+		: c->congested_read_threshold_us;
+
+	if (threshold) {
+		unsigned int t = local_clock_us();
+		int us = t - b->submit_time_us;
+		int congested = atomic_read(&c->congested);
+
+		if (us > (int) threshold) {
+			int ms = us / 1024;
+
+			c->congested_last_us = t;
+
+			ms = min(ms, CONGESTED_MAX + congested);
+			atomic_sub(ms, &c->congested);
+		} else if (congested < 0)
+			atomic_inc(&c->congested);
+	}
+
+	bch_count_io_errors(ca, error, is_read, m);
+}
+
+void bch_bbio_endio(struct cache_set *c, struct bio *bio,
+		    blk_status_t error, const char *m)
+{
+	struct closure *cl = bio->bi_private;
+
+	bch_bbio_count_io_errors(c, bio, error, m);
+	bio_put(bio);
+	closure_put(cl);
+}