linux-btrfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jeff Mahoney <jeffm@suse.com>
To: Linux Btrfs List <linux-btrfs@vger.kernel.org>
Cc: David Sterba <dsterba@suse.com>, David Sterba <dsterba@suse.cz>
Subject: [patch 33/35] btrfs: handle errors when excluding super extents
Date: Wed, 21 Mar 2012 21:11:37 -0400	[thread overview]
Message-ID: <20120322011137.642386355@suse.com> (raw)
In-Reply-To: 20120322011104.212214639@suse.com

Some images produced by btrfs-image cause a mount crash in rmap block
when excluding superblock extents. Handle the errors gracefully.

Signed-off-by: David Sterba <dsterba@suse.cz>
---
 fs/btrfs/extent-tree.c |   19 ++++++++++++++++---
 fs/btrfs/volumes.c     |   20 +++++++++++++++++---
 2 files changed, 33 insertions(+), 6 deletions(-)

diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c
index 0bdb9fa..8055d58 100644
--- a/fs/btrfs/extent-tree.c
+++ b/fs/btrfs/extent-tree.c
@@ -250,10 +250,17 @@ static int exclude_super_stripes(struct btrfs_root *root,
 
 	for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) {
 		bytenr = btrfs_sb_offset(i);
+		if (bytenr <= cache->key.objectid + cache->key.offset)
+			break;
+		logical = NULL;
 		ret = btrfs_rmap_block(&root->fs_info->mapping_tree,
 				       cache->key.objectid, bytenr,
 				       0, &logical, &nr, &stripe_len);
-		BUG_ON(ret); /* -ENOMEM */
+		if (ret < 0) {
+			kfree(logical);
+			free_excluded_extents(root, cache);
+			goto out;
+		}
 
 		while (nr--) {
 			cache->bytes_super += stripe_len;
@@ -264,7 +271,9 @@ static int exclude_super_stripes(struct btrfs_root *root,
 
 		kfree(logical);
 	}
-	return 0;
+
+out:
+	return ret;
 }
 
 static struct btrfs_caching_control *
@@ -7665,7 +7674,11 @@ int btrfs_read_block_groups(struct btrfs_root *root)
 		 * info has super bytes accounted for, otherwise we'll think
 		 * we have more space than we actually do.
 		 */
-		exclude_super_stripes(root, cache);
+		ret = exclude_super_stripes(root, cache);
+		if (ret < 0) {
+			kfree(cache);
+			goto error;
+		}
 
 		/*
 		 * check for two cases, either we are full, and therefore
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index f818f48..3400fbd 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -3911,12 +3911,22 @@ int btrfs_rmap_block(struct btrfs_mapping_tree *map_tree,
 	u64 length;
 	u64 stripe_nr;
 	int i, j, nr = 0;
+	int ret = 0;
 
 	read_lock(&em_tree->lock);
 	em = lookup_extent_mapping(em_tree, chunk_start, 1);
 	read_unlock(&em_tree->lock);
 
-	BUG_ON(!em || em->start != chunk_start);
+	if (!em)
+		return -ENOENT;
+
+	if (em->start != chunk_start) {
+		printk(KERN_DEBUG "btrfs: rmap em->start %llu != %llu\n",
+				em->start, chunk_start);
+		ret = -EINVAL;
+		goto out_error;
+	}
+
 	map = (struct map_lookup *)em->bdev;
 
 	length = em->len;
@@ -3926,7 +3936,10 @@ int btrfs_rmap_block(struct btrfs_mapping_tree *map_tree,
 		do_div(length, map->num_stripes);
 
 	buf = kzalloc(sizeof(u64) * map->num_stripes, GFP_NOFS);
-	BUG_ON(!buf); /* -ENOMEM */
+	if (!buf) {
+		ret = -ENOMEM;
+		goto out_error;
+	}
 
 	for (i = 0; i < map->num_stripes; i++) {
 		if (devid && map->stripes[i].dev->devid != devid)
@@ -3960,8 +3973,9 @@ int btrfs_rmap_block(struct btrfs_mapping_tree *map_tree,
 	*naddrs = nr;
 	*stripe_len = map->stripe_len;
 
+out_error:
 	free_extent_map(em);
-	return 0;
+	return ret;
 }
 
 static void btrfs_end_bio(struct bio *bio, int err)
-- 
1.7.9




  parent reply	other threads:[~2012-03-22  1:11 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-03-22  1:11 [patch 00/35] btrfs: Error Handling Patchset Jeff Mahoney
2012-03-22  1:11 ` [patch 01/35] btrfs: Add btrfs_panic() Jeff Mahoney
2012-03-22  1:11 ` [patch 02/35] btrfs: Catch locking failures in {set,clear,convert}_extent_bit Jeff Mahoney
2012-03-22  1:11 ` [patch 03/35] btrfs: Panic on bad rbtree operations Jeff Mahoney
2012-03-22  1:11 ` [patch 04/35] btrfs: Fix kfree of member instead of structure Jeff Mahoney
2012-03-22  1:11 ` [patch 05/35] btrfs: Simplify btrfs_insert_root Jeff Mahoney
2012-03-22  1:11 ` [patch 06/35] btrfs: clean_tree_block should panic on observed memory corruption and return void Jeff Mahoney
2012-03-22  1:11 ` [patch 07/35] btrfs: avoid NULL deref in btrfs_reserve_extent with DEBUG_ENOSPC Jeff Mahoney
2012-03-22  1:11 ` [patch 08/35] btrfs: Remove set bits return from clear_extent_bit Jeff Mahoney
2012-03-22  2:03   ` Liu Bo
2012-03-22  2:34     ` Jeff Mahoney
2012-03-22  1:11 ` [patch 09/35] btrfs: find_and_setup_root error push-up Jeff Mahoney
2012-03-22  1:11 ` [patch 10/35] btrfs: btrfs_update_root " Jeff Mahoney
2012-03-22  1:11 ` [patch 11/35] btrfs: Simplify btrfs_submit_bio_hook Jeff Mahoney
2012-03-22  1:11 ` [patch 12/35] btrfs: Factor out tree->ops->merge_bio_hook call Jeff Mahoney
2012-03-22  1:11 ` [patch 13/35] btrfs: ->submit_bio_hook error push-up Jeff Mahoney
2012-03-22  1:11 ` [patch 14/35] btrfs: __add_reloc_root " Jeff Mahoney
2012-03-22  1:11 ` [patch 15/35] btrfs: return void in functions without error conditions Jeff Mahoney
2012-03-22  1:11 ` [patch 16/35] btrfs: drop gfp_t from lock_extent Jeff Mahoney
2012-03-22  1:11 ` [patch 17/35] btrfs: split extent_state ops Jeff Mahoney
2012-03-22  1:56   ` Liu Bo
2012-03-22  2:35     ` Jeff Mahoney
2012-03-22  1:11 ` [patch 18/35] btrfs: btrfs_drop_snapshot should return int Jeff Mahoney
2012-03-22  1:11 ` [patch 19/35] btrfs: Dont BUG_ON errors from btrfs_create_subvol_root() Jeff Mahoney
2012-03-22  1:11 ` [patch 20/35] btrfs: Dont BUG_ON() errors in update_ref_for_cow() Jeff Mahoney
2012-03-22  1:11 ` [patch 21/35] btrfs: Dont BUG_ON kzalloc error in btrfs_lookup_csums_range() Jeff Mahoney
2012-03-22  1:11 ` [patch 22/35] btrfs: Dont BUG_ON errors in __finish_chunk_alloc() Jeff Mahoney
2012-03-22  1:11 ` [patch 23/35] btrfs: Go readonly on bad extent refs in update_ref_for_cow() Jeff Mahoney
2012-03-22  1:11 ` [patch 24/35] btrfs: Dont BUG_ON errors from update_ref_for_cow() Jeff Mahoney
2012-03-22  1:11 ` [patch 25/35] btrfs: Go readonly on tree errors in balance_level Jeff Mahoney
2012-03-22  1:11 ` [patch 26/35] btrfs: Dont BUG_ON insert errors in btrfs_alloc_dev_extent() Jeff Mahoney
2012-03-22  1:11 ` [patch 27/35] btrfs: Remove BUG_ON from __btrfs_alloc_chunk() Jeff Mahoney
2012-03-22  1:11 ` [patch 28/35] btrfs: Remove BUG_ON from __finish_chunk_alloc() Jeff Mahoney
2012-03-22  1:11 ` [patch 29/35] btrfs: add varargs to btrfs_error Jeff Mahoney
2012-03-22  1:11 ` [patch 30/35] btrfs: enhance transaction abort infrastructure Jeff Mahoney
2012-03-22  1:11 ` [patch 32/35] btrfs: Fix busyloop in transaction_kthread() Jeff Mahoney
2012-03-22  1:11 ` Jeff Mahoney [this message]
2012-03-23 17:21   ` [patch 33/35] btrfs: handle errors when excluding super extents David Sterba
2012-03-22  1:11 ` [patch 34/35] btrfs: enhance superblock sanity checks Jeff Mahoney
2012-03-22  1:11 ` [patch 35/35] btrfs: disallow unequal data/metadata blocksize for mixed block groups Jeff Mahoney
2012-03-22 16:02   ` Stefan Behrens
2012-03-23 17:13     ` David Sterba

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20120322011137.642386355@suse.com \
    --to=jeffm@suse.com \
    --cc=dsterba@suse.com \
    --cc=dsterba@suse.cz \
    --cc=linux-btrfs@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).