Update bcachefs sources to cbb2e45634dd bcachefs: fix simulateously upgrading & downgrading

Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>

9 files changed, 166 insertions, 90 deletions

diff --git a/libbcachefs/bset.c b/libbcachefs/bset.c
index bb73ba90..74bf8eb9 100644
--- a/libbcachefs/bset.c
+++ b/libbcachefs/bset.c
@@ -68,6 +68,12 @@ void bch2_dump_bset(struct bch_fs *c, struct btree *b,
 	     _k = _n) {
 		_n = bkey_p_next(_k);
 
+		if (!_k->u64s) {
+			printk(KERN_ERR "block %u key %5zu - u64s 0? aieee!\n", set,
+			       _k->_data - i->_data);
+			break;
+		}
+
 		k = bkey_disassemble(b, _k, &uk);
 
 		printbuf_reset(&buf);
diff --git a/libbcachefs/btree_io.c b/libbcachefs/btree_io.c
index 378579bb..33db48e2 100644
--- a/libbcachefs/btree_io.c
+++ b/libbcachefs/btree_io.c
@@ -524,7 +524,8 @@ static void btree_err_msg(struct printbuf *out, struct bch_fs *c,
 	prt_printf(out, "at btree ");
 	bch2_btree_pos_to_text(out, c, b);
 
-	prt_printf(out, "\n  node offset %u", b->written);
+	prt_printf(out, "\n  node offset %u/%u",
+		   b->written, btree_ptr_sectors_written(&b->key));
 	if (i)
 		prt_printf(out, " bset u64s %u", le16_to_cpu(i->u64s));
 	prt_str(out, ": ");
@@ -830,6 +831,23 @@ static int bset_key_invalid(struct bch_fs *c, struct btree *b,
 		(rw == WRITE ? bch2_bkey_val_invalid(c, k, READ, err) : 0);
 }
 
+static bool __bkey_valid(struct bch_fs *c, struct btree *b,
+			 struct bset *i, struct bkey_packed *k)
+{
+	if (bkey_p_next(k) > vstruct_last(i))
+		return false;
+
+	if (k->format > KEY_FORMAT_CURRENT)
+		return false;
+
+	struct printbuf buf = PRINTBUF;
+	struct bkey tmp;
+	struct bkey_s u = __bkey_disassemble(b, k, &tmp);
+	bool ret = __bch2_bkey_invalid(c, u.s_c, btree_node_type(b), READ, &buf);
+	printbuf_exit(&buf);
+	return ret;
+}
+
 static int validate_bset_keys(struct bch_fs *c, struct btree *b,
 			 struct bset *i, int write,
 			 bool have_retry, bool *saw_error)
@@ -845,6 +863,7 @@ static int validate_bset_keys(struct bch_fs *c, struct btree *b,
 	     k != vstruct_last(i);) {
 		struct bkey_s u;
 		struct bkey tmp;
+		unsigned next_good_key;
 
 		if (btree_err_on(bkey_p_next(k) > vstruct_last(i),
 				 -BCH_ERR_btree_node_read_err_fixable,
@@ -859,12 +878,8 @@ static int validate_bset_keys(struct bch_fs *c, struct btree *b,
 				 -BCH_ERR_btree_node_read_err_fixable,
 				 c, NULL, b, i,
 				 btree_node_bkey_bad_format,
-				 "invalid bkey format %u", k->format)) {
-			i->u64s = cpu_to_le16(le16_to_cpu(i->u64s) - k->u64s);
-			memmove_u64s_down(k, bkey_p_next(k),
-					  (u64 *) vstruct_end(i) - (u64 *) k);
-			continue;
-		}
+				 "invalid bkey format %u", k->format))
+			goto drop_this_key;
 
 		/* XXX: validate k->u64s */
 		if (!write)
@@ -885,11 +900,7 @@ static int validate_bset_keys(struct bch_fs *c, struct btree *b,
 				  c, NULL, b, i,
 				  btree_node_bad_bkey,
 				  "invalid bkey: %s", buf.buf);
-
-			i->u64s = cpu_to_le16(le16_to_cpu(i->u64s) - k->u64s);
-			memmove_u64s_down(k, bkey_p_next(k),
-					  (u64 *) vstruct_end(i) - (u64 *) k);
-			continue;
+			goto drop_this_key;
 		}
 
 		if (write)
@@ -906,21 +917,45 @@ static int validate_bset_keys(struct bch_fs *c, struct btree *b,
 			prt_printf(&buf, " > ");
 			bch2_bkey_to_text(&buf, u.k);
 
-			bch2_dump_bset(c, b, i, 0);
-
 			if (btree_err(-BCH_ERR_btree_node_read_err_fixable,
 				      c, NULL, b, i,
 				      btree_node_bkey_out_of_order,
-				      "%s", buf.buf)) {
-				i->u64s = cpu_to_le16(le16_to_cpu(i->u64s) - k->u64s);
-				memmove_u64s_down(k, bkey_p_next(k),
-						  (u64 *) vstruct_end(i) - (u64 *) k);
-				continue;
-			}
+				      "%s", buf.buf))
+				goto drop_this_key;
 		}
 
 		prev = k;
 		k = bkey_p_next(k);
+		continue;
+drop_this_key:
+		next_good_key = k->u64s;
+
+		if (!next_good_key ||
+		    (BSET_BIG_ENDIAN(i) == CPU_BIG_ENDIAN &&
+		     version >= bcachefs_metadata_version_snapshot)) {
+			/*
+			 * only do scanning if bch2_bkey_compat() has nothing to
+			 * do
+			 */
+
+			if (!__bkey_valid(c, b, i, (void *) ((u64 *) k + next_good_key))) {
+				for (next_good_key = 1;
+				     next_good_key < (u64 *) vstruct_last(i) - (u64 *) k;
+				     next_good_key++)
+					if (__bkey_valid(c, b, i, (void *) ((u64 *) k + next_good_key)))
+						goto got_good_key;
+
+			}
+
+			/*
+			 * didn't find a good key, have to truncate the rest of
+			 * the bset
+			 */
+			next_good_key = (u64 *) vstruct_last(i) - (u64 *) k;
+		}
+got_good_key:
+		le16_add_cpu(&i->u64s, -next_good_key);
+		memmove_u64s_down(k, bkey_p_next(k), (u64 *) vstruct_end(i) - (u64 *) k);
 	}
 fsck_err:
 	printbuf_exit(&buf);
@@ -1007,8 +1042,8 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca,
 
 			nonce = btree_nonce(i, b->written << 9);
 
-			csum_bad = bch2_crc_cmp(b->data->csum,
-				csum_vstruct(c, BSET_CSUM_TYPE(i), nonce, b->data));
+			struct bch_csum csum = csum_vstruct(c, BSET_CSUM_TYPE(i), nonce, b->data);
+			csum_bad = bch2_crc_cmp(b->data->csum, csum);
 			if (csum_bad)
 				bch2_io_error(ca, BCH_MEMBER_ERROR_checksum);
 
@@ -1016,7 +1051,10 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca,
 				     -BCH_ERR_btree_node_read_err_want_retry,
 				     c, ca, b, i,
 				     bset_bad_csum,
-				     "invalid checksum");
+				     "%s",
+				     (printbuf_reset(&buf),
+				      bch2_csum_err_msg(&buf, BSET_CSUM_TYPE(i), b->data->csum, csum),
+				      buf.buf));
 
 			ret = bset_encrypt(c, i, b->written << 9);
 			if (bch2_fs_fatal_err_on(ret, c,
@@ -1045,8 +1083,8 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca,
 				     "unknown checksum type %llu", BSET_CSUM_TYPE(i));
 
 			nonce = btree_nonce(i, b->written << 9);
-			csum_bad = bch2_crc_cmp(bne->csum,
-				csum_vstruct(c, BSET_CSUM_TYPE(i), nonce, bne));
+			struct bch_csum csum = csum_vstruct(c, BSET_CSUM_TYPE(i), nonce, bne);
+			csum_bad = bch2_crc_cmp(bne->csum, csum);
 			if (csum_bad)
 				bch2_io_error(ca, BCH_MEMBER_ERROR_checksum);
 
@@ -1054,7 +1092,10 @@ int bch2_btree_node_read_done(struct bch_fs *c, struct bch_dev *ca,
 				     -BCH_ERR_btree_node_read_err_want_retry,
 				     c, ca, b, i,
 				     bset_bad_csum,
-				     "invalid checksum");
+				     "%s",
+				     (printbuf_reset(&buf),
+				      bch2_csum_err_msg(&buf, BSET_CSUM_TYPE(i), bne->csum, csum),
+				      buf.buf));
 
 			ret = bset_encrypt(c, i, b->written << 9);
 			if (bch2_fs_fatal_err_on(ret, c,
diff --git a/libbcachefs/btree_iter.c b/libbcachefs/btree_iter.c
index 6e8e9ba5..fa298289 100644
--- a/libbcachefs/btree_iter.c
+++ b/libbcachefs/btree_iter.c
@@ -2818,34 +2818,11 @@ void *__bch2_trans_kmalloc(struct btree_trans *trans, size_t size)
 	return p;
 }
 
-#include "sb-members.h"
-
 static inline void check_srcu_held_too_long(struct btree_trans *trans)
 {
-	if (trans->srcu_held && time_after(jiffies, trans->srcu_lock_time + HZ * 10)) {
-		struct printbuf buf = PRINTBUF;
-
-		prt_str(&buf, "btree node read time:\n");
-		bch2_time_stats_to_text(&buf, &trans->c->times[BCH_TIME_btree_node_read]);
-
-		prt_str(&buf, "btree node read_done time:\n");
-		bch2_time_stats_to_text(&buf, &trans->c->times[BCH_TIME_btree_node_read_done]);
-
-		for_each_member_device(trans->c, ca) {
-			prt_printf(&buf, "device %u read time:\n", ca->dev_idx);
-			bch2_time_stats_to_text(&buf, &ca->io_latency[READ]);
-		}
-
-		struct btree_transaction_stats *s = btree_trans_stats(trans);
-		prt_str(&buf, "transaction duration:\n");
-		bch2_time_stats_to_text(&buf, &s->duration);
-
-		WARN(trans->srcu_held && time_after(jiffies, trans->srcu_lock_time + HZ * 10),
-		     "btree trans held srcu lock (delaying memory reclaim) for %lu seconds",
-		     (jiffies - trans->srcu_lock_time) / HZ);
-		bch2_print_string_as_lines(KERN_ERR, buf.buf);
-		printbuf_exit(&buf);
-	}
+	WARN(trans->srcu_held && time_after(jiffies, trans->srcu_lock_time + HZ * 10),
+	     "btree trans held srcu lock (delaying memory reclaim) for %lu seconds",
+	     (jiffies - trans->srcu_lock_time) / HZ);
 }
 
 void bch2_trans_srcu_unlock(struct btree_trans *trans)
diff --git a/libbcachefs/checksum.h b/libbcachefs/checksum.h
index 13998388..1b8c2c10 100644
--- a/libbcachefs/checksum.h
+++ b/libbcachefs/checksum.h
@@ -45,6 +45,29 @@ struct bch_csum bch2_checksum(struct bch_fs *, unsigned, struct nonce,
 	bch2_checksum(_c, _type, _nonce, _start, vstruct_end(_i) - _start);\
 })
 
+static inline void bch2_csum_to_text(struct printbuf *out,
+				     enum bch_csum_type type,
+				     struct bch_csum csum)
+{
+	const u8 *p = (u8 *) &csum;
+	unsigned bytes = type < BCH_CSUM_NR ? bch_crc_bytes[type] : 16;
+
+	for (unsigned i = 0; i < bytes; i++)
+		prt_hex_byte(out, p[i]);
+}
+
+static inline void bch2_csum_err_msg(struct printbuf *out,
+				     enum bch_csum_type type,
+				     struct bch_csum expected,
+				     struct bch_csum got)
+{
+	prt_printf(out, "checksum error: got ");
+	bch2_csum_to_text(out, type, got);
+	prt_str(out, " should be ");
+	bch2_csum_to_text(out, type, expected);
+	prt_printf(out, " type %s", bch2_csum_types[type]);
+}
+
 int bch2_chacha_encrypt_key(struct bch_key *, struct nonce, void *, size_t);
 int bch2_request_key(struct bch_sb *, struct bch_key *);
 #ifndef __KERNEL__
diff --git a/libbcachefs/fs-io.c b/libbcachefs/fs-io.c
index 9fea8976..98bd5bab 100644
--- a/libbcachefs/fs-io.c
+++ b/libbcachefs/fs-io.c
@@ -194,16 +194,6 @@ int bch2_fsync(struct file *file, loff_t start, loff_t end, int datasync)
 	struct bch_fs *c = inode->v.i_sb->s_fs_info;
 	int ret;
 
-	/*
-	 * check if unlinked, disable/defer until relink
-	 */
-
-	/*
-	 * also: add a mode where a file is a tmpfile until fully,
-	 * asynchronously written
-	 */
-
-
 	ret = file_write_and_wait_range(file, start, end);
 	if (ret)
 		goto out;
diff --git a/libbcachefs/io_read.c b/libbcachefs/io_read.c
index 5c2d118e..3c574d88 100644
--- a/libbcachefs/io_read.c
+++ b/libbcachefs/io_read.c
@@ -642,12 +642,17 @@ csum_err:
 		goto out;
 	}
 
+	struct printbuf buf = PRINTBUF;
+	buf.atomic++;
+	prt_str(&buf, "data ");
+	bch2_csum_err_msg(&buf, crc.csum_type, rbio->pick.crc.csum, csum);
+
 	bch_err_inum_offset_ratelimited(ca,
 		rbio->read_pos.inode,
 		rbio->read_pos.offset << 9,
-		"data checksum error: expected %0llx:%0llx got %0llx:%0llx (type %s)",
-		rbio->pick.crc.csum.hi, rbio->pick.crc.csum.lo,
-		csum.hi, csum.lo, bch2_csum_types[crc.csum_type]);
+		"data %s", buf.buf);
+	printbuf_exit(&buf);
+
 	bch2_io_error(ca, BCH_MEMBER_ERROR_checksum);
 	bch2_rbio_error(rbio, READ_RETRY_AVOID, BLK_STS_IOERR);
 	goto out;
diff --git a/libbcachefs/journal_io.c b/libbcachefs/journal_io.c
index c5bc5824..b0f4dd49 100644
--- a/libbcachefs/journal_io.c
+++ b/libbcachefs/journal_io.c
@@ -27,11 +27,15 @@ static struct nonce journal_nonce(const struct jset *jset)
 	}};
 }
 
-static bool jset_csum_good(struct bch_fs *c, struct jset *j)
+static bool jset_csum_good(struct bch_fs *c, struct jset *j, struct bch_csum *csum)
 {
-	return bch2_checksum_type_valid(c, JSET_CSUM_TYPE(j)) &&
-		!bch2_crc_cmp(j->csum,
-			      csum_vstruct(c, JSET_CSUM_TYPE(j), journal_nonce(j), j));
+	if (!bch2_checksum_type_valid(c, JSET_CSUM_TYPE(j))) {
+		*csum = (struct bch_csum) {};
+		return false;
+	}
+
+	*csum = csum_vstruct(c, JSET_CSUM_TYPE(j), journal_nonce(j), j);
+	return !bch2_crc_cmp(j->csum, *csum);
 }
 
 static inline u32 journal_entry_radix_idx(struct bch_fs *c, u64 seq)
@@ -934,6 +938,7 @@ static int journal_read_bucket(struct bch_dev *ca,
 	u64 offset = bucket_to_sector(ca, ja->buckets[bucket]),
 	    end = offset + ca->mi.bucket_size;
 	bool saw_bad = false, csum_good;
+	struct printbuf err = PRINTBUF;
 	int ret = 0;
 
 	pr_debug("reading %u", bucket);
@@ -966,7 +971,7 @@ reread:
 				 * found on a different device, and missing or
 				 * no journal entries will be handled later
 				 */
-				return 0;
+				goto out;
 			}
 
 			j = buf->data;
@@ -983,12 +988,12 @@ reread:
 				ret = journal_read_buf_realloc(buf,
 							vstruct_bytes(j));
 				if (ret)
-					return ret;
+					goto err;
 			}
 			goto reread;
 		case JOURNAL_ENTRY_NONE:
 			if (!saw_bad)
-				return 0;
+				goto out;
 			/*
 			 * On checksum error we don't really trust the size
 			 * field of the journal entry we read, so try reading
@@ -997,7 +1002,7 @@ reread:
 			sectors = block_sectors(c);
 			goto next_block;
 		default:
-			return ret;
+			goto err;
 		}
 
 		/*
@@ -1007,20 +1012,28 @@ reread:
 		 * bucket:
 		 */
 		if (le64_to_cpu(j->seq) < ja->bucket_seq[bucket])
-			return 0;
+			goto out;
 
 		ja->bucket_seq[bucket] = le64_to_cpu(j->seq);
 
-		csum_good = jset_csum_good(c, j);
+		enum bch_csum_type csum_type = JSET_CSUM_TYPE(j);
+		struct bch_csum csum;
+		csum_good = jset_csum_good(c, j, &csum);
+
 		if (bch2_dev_io_err_on(!csum_good, ca, BCH_MEMBER_ERROR_checksum,
-				       "journal checksum error"))
+				       "%s",
+				       (printbuf_reset(&err),
+					prt_str(&err, "journal "),
+					bch2_csum_err_msg(&err, csum_type, j->csum, csum),
+					err.buf)))
 			saw_bad = true;
 
 		ret = bch2_encrypt(c, JSET_CSUM_TYPE(j), journal_nonce(j),
 			     j->encrypted_start,
 			     vstruct_end(j) - (void *) j->encrypted_start);
 		bch2_fs_fatal_err_on(ret, c,
-				"error decrypting journal entry: %i", ret);
+				"error decrypting journal entry: %s",
+				bch2_err_str(ret));
 
 		mutex_lock(&jlist->lock);
 		ret = journal_entry_add(c, ca, (struct journal_ptr) {
@@ -1039,7 +1052,7 @@ reread:
 		case JOURNAL_ENTRY_ADD_OUT_OF_RANGE:
 			break;
 		default:
-			return ret;
+			goto err;
 		}
 next_block:
 		pr_debug("next");
@@ -1048,7 +1061,11 @@ next_block:
 		j = ((void *) j) + (sectors << 9);
 	}
 
-	return 0;
+out:
+	ret = 0;
+err:
+	printbuf_exit(&err);
+	return ret;
 }
 
 static CLOSURE_CALLBACK(bch2_journal_read_device)
diff --git a/libbcachefs/recovery.c b/libbcachefs/recovery.c
index e1f0da6a..72521460 100644
--- a/libbcachefs/recovery.c
+++ b/libbcachefs/recovery.c
@@ -696,8 +696,11 @@ static int bch2_run_recovery_passes(struct bch_fs *c)
 
 	while (c->curr_recovery_pass < ARRAY_SIZE(recovery_pass_fns)) {
 		if (should_run_recovery_pass(c, c->curr_recovery_pass)) {
+			unsigned pass = c->curr_recovery_pass;
+
 			ret = bch2_run_recovery_pass(c, c->curr_recovery_pass);
-			if (bch2_err_matches(ret, BCH_ERR_restart_recovery))
+			if (bch2_err_matches(ret, BCH_ERR_restart_recovery) ||
+			    (ret && c->curr_recovery_pass < pass))
 				continue;
 			if (ret)
 				break;
diff --git a/libbcachefs/super-io.c b/libbcachefs/super-io.c
index ea869217..55926b81 100644
--- a/libbcachefs/super-io.c
+++ b/libbcachefs/super-io.c
@@ -612,7 +612,6 @@ int bch2_sb_from_fs(struct bch_fs *c, struct bch_dev *ca)
 
 static int read_one_super(struct bch_sb_handle *sb, u64 offset, struct printbuf *err)
 {
-	struct bch_csum csum;
 	size_t bytes;
 	int ret;
 reread:
@@ -628,7 +627,9 @@ reread:
 
 	if (!uuid_equal(&sb->sb->magic, &BCACHE_MAGIC) &&
 	    !uuid_equal(&sb->sb->magic, &BCHFS_MAGIC)) {
-		prt_printf(err, "Not a bcachefs superblock");
+		prt_str(err, "Not a bcachefs superblock (got magic ");
+		pr_uuid(err, sb->sb->magic.b);
+		prt_str(err, ")");
 		return -BCH_ERR_invalid_sb_magic;
 	}
 
@@ -651,17 +652,16 @@ reread:
 		goto reread;
 	}
 
-	if (BCH_SB_CSUM_TYPE(sb->sb) >= BCH_CSUM_NR) {
+	enum bch_csum_type csum_type = BCH_SB_CSUM_TYPE(sb->sb);
+	if (csum_type >= BCH_CSUM_NR) {
 		prt_printf(err, "unknown checksum type %llu", BCH_SB_CSUM_TYPE(sb->sb));
 		return -BCH_ERR_invalid_sb_csum_type;
 	}
 
 	/* XXX: verify MACs */
-	csum = csum_vstruct(NULL, BCH_SB_CSUM_TYPE(sb->sb),
-			    null_nonce(), sb->sb);
-
+	struct bch_csum csum = csum_vstruct(NULL, csum_type, null_nonce(), sb->sb);
 	if (bch2_crc_cmp(csum, sb->sb->csum)) {
-		prt_printf(err, "bad checksum");
+		bch2_csum_err_msg(err, csum_type, sb->sb->csum, csum);
 		return -BCH_ERR_invalid_sb_csum;
 	}
 
@@ -1088,13 +1088,22 @@ bool bch2_check_version_downgrade(struct bch_fs *c)
 	/*
 	 * Downgrade, if superblock is at a higher version than currently
 	 * supported:
+	 *
+	 * c->sb will be checked before we write the superblock, so update it as
+	 * well:
 	 */
-	if (BCH_SB_VERSION_UPGRADE_COMPLETE(c->disk_sb.sb) > bcachefs_metadata_version_current)
+	if (BCH_SB_VERSION_UPGRADE_COMPLETE(c->disk_sb.sb) > bcachefs_metadata_version_current) {
 		SET_BCH_SB_VERSION_UPGRADE_COMPLETE(c->disk_sb.sb, bcachefs_metadata_version_current);
-	if (c->sb.version > bcachefs_metadata_version_current)
+		c->sb.version_upgrade_complete = bcachefs_metadata_version_current;
+	}
+	if (c->sb.version > bcachefs_metadata_version_current) {
 		c->disk_sb.sb->version = cpu_to_le16(bcachefs_metadata_version_current);
-	if (c->sb.version_min > bcachefs_metadata_version_current)
+		c->sb.version = bcachefs_metadata_version_current;
+	}
+	if (c->sb.version_min > bcachefs_metadata_version_current) {
 		c->disk_sb.sb->version_min = cpu_to_le16(bcachefs_metadata_version_current);
+		c->sb.version_min = bcachefs_metadata_version_current;
+	}
 	c->disk_sb.sb->compat[0] &= cpu_to_le64((1ULL << BCH_COMPAT_NR) - 1);
 	return ret;
 }
@@ -1261,6 +1270,11 @@ void bch2_sb_to_text(struct printbuf *out, struct bch_sb *sb,
 	pr_uuid(out, sb->uuid.b);
 	prt_newline(out);
 
+	prt_printf(out, "Magic number:");
+	prt_tab(out);
+	pr_uuid(out, sb->magic.b);
+	prt_newline(out);
+
 	prt_str(out, "Device index:");
 	prt_tab(out);
 	prt_printf(out, "%u", sb->dev_idx);