bcachefs: bch2_write_op_error() now prints info about data update

A user has been seeing the "error verifying existing checksum while
rewriting existing data (memory corruption?)" error.

This generally indicates a hardware issue (and that may be the case
here), but it might also indicate a bug, in which case we need more
information to look for patterns.

Reported-by: Roland Vet <vet.roland@protonmail.com>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>

5 files changed, 80 insertions, 35 deletions

diff --git a/fs/bcachefs/compress.c b/fs/bcachefs/compress.c
index 114bf2f3879f..31467f77930f 100644
--- a/fs/bcachefs/compress.c
+++ b/fs/bcachefs/compress.c
@@ -271,8 +271,8 @@ int bch2_bio_uncompress_inplace(struct bch_write_op *op,
 	if (crc->uncompressed_size << 9	> c->opts.encoded_extent_max ||
 	    crc->compressed_size << 9	> c->opts.encoded_extent_max) {
 		struct printbuf buf = PRINTBUF;
-		bch2_write_op_error(&buf, op);
-		prt_printf(&buf, "error rewriting existing data: extent too big");
+		bch2_write_op_error(&buf, op, op->pos.offset,
+				    "extent too big to decompress");
 		bch_err_ratelimited(c, "%s", buf.buf);
 		printbuf_exit(&buf);
 		return -EIO;
@@ -283,8 +283,8 @@ int bch2_bio_uncompress_inplace(struct bch_write_op *op,
 	if (__bio_uncompress(c, bio, data.b, *crc)) {
 		if (!c->opts.no_data_io) {
 			struct printbuf buf = PRINTBUF;
-			bch2_write_op_error(&buf, op);
-			prt_printf(&buf, "error rewriting existing data: decompression error");
+			bch2_write_op_error(&buf, op, op->pos.offset,
+					    "decompression error");
 			bch_err_ratelimited(c, "%s", buf.buf);
 			printbuf_exit(&buf);
 		}
diff --git a/fs/bcachefs/error.c b/fs/bcachefs/error.c
index c8fc58fab958..3f93a5a6bbfa 100644
--- a/fs/bcachefs/error.c
+++ b/fs/bcachefs/error.c
@@ -580,3 +580,9 @@ int bch2_inum_snap_offset_err_msg_trans(struct btree_trans *trans, struct printb
 	prt_printf(out, " offset %llu: ", pos.offset << 8);
 	return 0;
 }
+
+void bch2_inum_snap_offset_err_msg(struct bch_fs *c, struct printbuf *out,
+				  struct bpos pos)
+{
+	bch2_trans_do(c, bch2_inum_snap_offset_err_msg_trans(trans, out, pos));
+}
diff --git a/fs/bcachefs/error.h b/fs/bcachefs/error.h
index 76da0e88cee8..b3cc69f29fd9 100644
--- a/fs/bcachefs/error.h
+++ b/fs/bcachefs/error.h
@@ -243,5 +243,6 @@ int bch2_inum_offset_err_msg_trans(struct btree_trans *, struct printbuf *, subv
 void bch2_inum_offset_err_msg(struct bch_fs *, struct printbuf *, subvol_inum, u64);
 
 int bch2_inum_snap_offset_err_msg_trans(struct btree_trans *, struct printbuf *, struct bpos);
+void bch2_inum_snap_offset_err_msg(struct bch_fs *, struct printbuf *, struct bpos);
 
 #endif /* _BCACHEFS_ERROR_H */
diff --git a/fs/bcachefs/io_write.c b/fs/bcachefs/io_write.c
index 076e39474610..738bdbfbdb14 100644
--- a/fs/bcachefs/io_write.c
+++ b/fs/bcachefs/io_write.c
@@ -396,29 +396,61 @@ static int bch2_write_index_default(struct bch_write_op *op)
 
 /* Writes */
 
-static void __bch2_write_op_error(struct printbuf *out, struct bch_write_op *op,
-				  u64 offset)
+void bch2_write_op_error_trans(struct btree_trans *trans, struct printbuf *out,
+			       struct bch_write_op *op, u64 offset, const char *fmt, ...)
 {
-	bch2_inum_offset_err_msg(op->c, out,
-				 (subvol_inum) { op->subvol, op->pos.inode, },
-				 offset << 9);
-	prt_printf(out, "write error%s: ",
-		   op->flags & BCH_WRITE_move ? "(internal move)" : "");
-}
+	if (op->subvol)
+		lockrestart_do(trans,
+			bch2_inum_offset_err_msg_trans(trans, out,
+						       (subvol_inum) { op->subvol, op->pos.inode, },
+						       offset << 9));
+	else {
+		struct bpos pos = op->pos;
+		pos.offset = offset;
+		lockrestart_do(trans, bch2_inum_snap_offset_err_msg_trans(trans, out, pos));
+	}
 
-void bch2_write_op_error(struct printbuf *out, struct bch_write_op *op)
-{
-	__bch2_write_op_error(out, op, op->pos.offset);
+	prt_str(out, "write error: ");
+
+	va_list args;
+	va_start(args, fmt);
+	prt_vprintf(out, fmt, args);
+	va_end(args);
+
+	if (op->flags & BCH_WRITE_move) {
+		struct data_update *u = container_of(op, struct data_update, op);
+
+		prt_printf(out, "\n  from internal move ");
+		bch2_bkey_val_to_text(out, op->c, bkey_i_to_s_c(u->k.k));
+	}
 }
 
-static void bch2_write_op_error_trans(struct btree_trans *trans, struct printbuf *out,
-				      struct bch_write_op *op, u64 offset)
+void bch2_write_op_error(struct printbuf *out, struct bch_write_op *op, u64 offset,
+			 const char *fmt, ...)
 {
-	bch2_inum_offset_err_msg_trans(trans, out,
-				       (subvol_inum) { op->subvol, op->pos.inode, },
-				       offset << 9);
-	prt_printf(out, "write error%s: ",
-		   op->flags & BCH_WRITE_move ? "(internal move)" : "");
+	if (op->subvol)
+		bch2_inum_offset_err_msg(op->c, out,
+					 (subvol_inum) { op->subvol, op->pos.inode, },
+					 offset << 9);
+	else {
+		struct bpos pos = op->pos;
+		pos.offset = offset;
+		bch2_inum_snap_offset_err_msg(op->c, out, pos);
+	}
+
+	prt_str(out, "write error: ");
+
+	va_list args;
+	va_start(args, fmt);
+	prt_vprintf(out, fmt, args);
+	va_end(args);
+
+	if (op->flags & BCH_WRITE_move) {
+		struct data_update *u = container_of(op, struct data_update, op);
+
+		prt_printf(out, "\n  from internal move ");
+		bch2_bkey_val_to_text(out, op->c, bkey_i_to_s_c(u->k.k));
+	}
 }
 
 void bch2_submit_wbio_replicas(struct bch_write_bio *wbio, struct bch_fs *c,
@@ -561,8 +593,8 @@ static void __bch2_write_index(struct bch_write_op *op)
 			struct bkey_i *insert = bch2_keylist_front(&op->insert_keys);
 
 			struct printbuf buf = PRINTBUF;
-			__bch2_write_op_error(&buf, op, bkey_start_offset(&insert->k));
-			prt_printf(&buf, "btree update error: %s", bch2_err_str(ret));
+			bch2_write_op_error(&buf, op, bkey_start_offset(&insert->k),
+					    "btree update error: %s", bch2_err_str(ret));
 			bch_err_ratelimited(c, "%s", buf.buf);
 			printbuf_exit(&buf);
 		}
@@ -1114,8 +1146,8 @@ do_write:
 csum_err:
 	{
 		struct printbuf buf = PRINTBUF;
-		bch2_write_op_error(&buf, op);
-		prt_printf(&buf, "error verifying existing checksum while rewriting existing data (memory corruption?)");
+		bch2_write_op_error(&buf, op, op->pos.offset,
+				    "error verifying existing checksum while rewriting existing data (memory corruption?)");
 		bch_err_ratelimited(c, "%s", buf.buf);
 		printbuf_exit(&buf);
 	}
@@ -1211,8 +1243,8 @@ static void bch2_nocow_write_convert_unwritten(struct bch_write_op *op)
 			struct bkey_i *insert = bch2_keylist_front(&op->insert_keys);
 
 			struct printbuf buf = PRINTBUF;
-			bch2_write_op_error_trans(trans, &buf, op, bkey_start_offset(&insert->k));
-			prt_printf(&buf, "btree update error: %s", bch2_err_str(ret));
+			bch2_write_op_error_trans(trans, &buf, op, bkey_start_offset(&insert->k),
+						  "btree update error: %s", bch2_err_str(ret));
 			bch_err_ratelimited(c, "%s", buf.buf);
 			printbuf_exit(&buf);
 		}
@@ -1379,8 +1411,8 @@ err:
 
 	if (ret) {
 		struct printbuf buf = PRINTBUF;
-		bch2_write_op_error(&buf, op);
-		prt_printf(&buf, "%s(): btree lookup error: %s", __func__, bch2_err_str(ret));
+		bch2_write_op_error(&buf, op, op->pos.offset,
+				    "%s(): btree lookup error: %s", __func__, bch2_err_str(ret));
 		bch_err_ratelimited(c, "%s", buf.buf);
 		printbuf_exit(&buf);
 		op->error = ret;
@@ -1502,8 +1534,8 @@ err:
 			if (unlikely(ret < 0)) {
 				if (!(op->flags & BCH_WRITE_alloc_nowait)) {
 					struct printbuf buf = PRINTBUF;
-					bch2_write_op_error(&buf, op);
-					prt_printf(&buf, "%s(): %s", __func__, bch2_err_str(ret));
+					bch2_write_op_error(&buf, op, op->pos.offset,
+							    "%s(): %s", __func__, bch2_err_str(ret));
 					bch_err_ratelimited(c, "%s", buf.buf);
 					printbuf_exit(&buf);
 				}
@@ -1634,8 +1666,8 @@ CLOSURE_CALLBACK(bch2_write)
 
 	if (unlikely(bio->bi_iter.bi_size & (c->opts.block_size - 1))) {
 		struct printbuf buf = PRINTBUF;
-		bch2_write_op_error(&buf, op);
-		prt_printf(&buf, "misaligned write");
+		bch2_write_op_error(&buf, op, op->pos.offset,
+				    "misaligned write");
 		printbuf_exit(&buf);
 		op->error = -EIO;
 		goto err;
diff --git a/fs/bcachefs/io_write.h b/fs/bcachefs/io_write.h
index 02cca52be0bd..bf942566a8eb 100644
--- a/fs/bcachefs/io_write.h
+++ b/fs/bcachefs/io_write.h
@@ -20,7 +20,13 @@ static inline void bch2_latency_acct(struct bch_dev *ca, u64 submit_time, int rw
 void bch2_submit_wbio_replicas(struct bch_write_bio *, struct bch_fs *,
 			       enum bch_data_type, const struct bkey_i *, bool);
 
-void bch2_write_op_error(struct printbuf *out, struct bch_write_op *op);
+__printf(5, 6)
+void bch2_write_op_error_trans(struct btree_trans *trans, struct printbuf *out,
+			       struct bch_write_op *op, u64, const char *, ...);
+
+__printf(4, 5)
+void bch2_write_op_error(struct printbuf *out, struct bch_write_op *op, u64,
+			 const char *, ...);
 
 #define BCH_WRITE_FLAGS()		\
 	x(alloc_nowait)			\