linux-btrfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Liu Bo <bo.li.liu@oracle.com>
To: Qu Wenruo <quwenruo@cn.fujitsu.com>
Cc: linux-btrfs@vger.kernel.org, vegard.nossum@oracle.com, sterba@suse.com
Subject: Re: [PATCH 2/2] Btrfs: add valid checks for chunk loading
Date: Tue, 3 May 2016 16:36:25 -0700	[thread overview]
Message-ID: <20160503233625.GG21008@localhost.localdomain> (raw)
In-Reply-To: <42231399-bac4-1c41-bb43-f2c7b4d1a0a0@cn.fujitsu.com>

On Tue, May 03, 2016 at 09:12:01AM +0800, Qu Wenruo wrote:
> 
> 
> Liu Bo wrote on 2016/05/02 11:15 -0700:
> >To prevent fuzz filesystem images from panic the whole system,
> >we need various validation checks to refuse to mount such an image
> >if btrfs finds any invalid value during loading chunks, including
> >both sys_array and regular chunks.
> >
> >Note that these checks may not be sufficient to cover all corner cases,
> >feel free to add more checks.
> 
> Looks good for me.
> 
> But would you mind to do extra check on some minor members like owner,
> io_align, io_width and sub_stripes?
> Since we have a dedicated function now, if not too hard, it's never a bad
> idea to check every member for best robust.

OK.

> 
> Especially sub_stripes, as it seems to be used by division in
> btrfs_rmap_block().

It makes sense to check sub_stripes, but io_align/width are not used at
this moment, I'd leave it for the future.

What do you think?

Thanks,

-liubo

> 
> Thanks,
> Qu
> >
> >Reported-by: Vegard Nossum <vegard.nossum@oracle.com>
> >Reported-by: Quentin Casasnovas <quentin.casasnovas@oracle.com>
> >Signed-off-by: Liu Bo <bo.li.liu@oracle.com>
> >---
> > fs/btrfs/volumes.c | 84 +++++++++++++++++++++++++++++++++++++++++++-----------
> > 1 file changed, 68 insertions(+), 16 deletions(-)
> >
> >diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
> >index bd0f45f..1075573 100644
> >--- a/fs/btrfs/volumes.c
> >+++ b/fs/btrfs/volumes.c
> >@@ -6206,27 +6206,23 @@ struct btrfs_device *btrfs_alloc_device(struct btrfs_fs_info *fs_info,
> > 	return dev;
> > }
> >
> >-static int read_one_chunk(struct btrfs_root *root, struct btrfs_key *key,
> >-			  struct extent_buffer *leaf,
> >-			  struct btrfs_chunk *chunk)
> >+/* Return -EIO if any error, otherwise return 0. */
> >+static int btrfs_check_chunk_valid(struct btrfs_root *root,
> >+				   struct extent_buffer *leaf,
> >+				   struct btrfs_chunk *chunk, u64 logical)
> > {
> >-	struct btrfs_mapping_tree *map_tree = &root->fs_info->mapping_tree;
> >-	struct map_lookup *map;
> >-	struct extent_map *em;
> >-	u64 logical;
> > 	u64 length;
> > 	u64 stripe_len;
> >-	u64 devid;
> >-	u8 uuid[BTRFS_UUID_SIZE];
> >-	int num_stripes;
> >-	int ret;
> >-	int i;
> >+	u16 num_stripes;
> >+	u16 sub_stripes;
> >+	u64 type;
> >
> >-	logical = key->offset;
> > 	length = btrfs_chunk_length(leaf, chunk);
> > 	stripe_len = btrfs_chunk_stripe_len(leaf, chunk);
> > 	num_stripes = btrfs_chunk_num_stripes(leaf, chunk);
> >-	/* Validation check */
> >+	sub_stripes = btrfs_chunk_sub_stripes(leaf, chunk);
> >+	type = btrfs_chunk_type(leaf, chunk);
> >+
> > 	if (!num_stripes) {
> > 		btrfs_err(root->fs_info, "invalid chunk num_stripes: %u",
> > 			  num_stripes);
> >@@ -6237,24 +6233,70 @@ static int read_one_chunk(struct btrfs_root *root, struct btrfs_key *key,
> > 			  "invalid chunk logical %llu", logical);
> > 		return -EIO;
> > 	}
> >+	if (btrfs_chunk_sector_size(leaf, chunk) != root->sectorsize) {
> >+		btrfs_err(root->fs_info, "invalid chunk sectorsize %llu",
> >+			  (unsigned long long)btrfs_chunk_sector_size(leaf,
> >+								      chunk));
> >+		return -EIO;
> >+	}
> > 	if (!length || !IS_ALIGNED(length, root->sectorsize)) {
> > 		btrfs_err(root->fs_info,
> > 			"invalid chunk length %llu", length);
> > 		return -EIO;
> > 	}
> >-	if (!is_power_of_2(stripe_len)) {
> >+	if (stripe_len != BTRFS_STRIPE_LEN) {
> > 		btrfs_err(root->fs_info, "invalid chunk stripe length: %llu",
> > 			  stripe_len);
> > 		return -EIO;
> > 	}
> > 	if (~(BTRFS_BLOCK_GROUP_TYPE_MASK | BTRFS_BLOCK_GROUP_PROFILE_MASK) &
> >-	    btrfs_chunk_type(leaf, chunk)) {
> >+	    type) {
> > 		btrfs_err(root->fs_info, "unrecognized chunk type: %llu",
> > 			  ~(BTRFS_BLOCK_GROUP_TYPE_MASK |
> > 			    BTRFS_BLOCK_GROUP_PROFILE_MASK) &
> > 			  btrfs_chunk_type(leaf, chunk));
> > 		return -EIO;
> > 	}
> >+	if ((type & BTRFS_BLOCK_GROUP_RAID10 && sub_stripes == 0) ||
> >+	    (type & BTRFS_BLOCK_GROUP_RAID1 && num_stripes < 1) ||
> >+	    (type & BTRFS_BLOCK_GROUP_RAID5 && num_stripes < 2) ||
> >+	    (type & BTRFS_BLOCK_GROUP_RAID5 && num_stripes < 3) ||
> >+	    (type & BTRFS_BLOCK_GROUP_DUP && num_stripes > 2) ||
> >+	    ((type & BTRFS_BLOCK_GROUP_PROFILE_MASK) == 0 &&
> >+	     num_stripes != 1)) {
> >+		btrfs_err(root->fs_info, "Invalid num_stripes:sub_stripes %u:%u for profile %llu",
> >+			  num_stripes, sub_stripes,
> >+			  type & BTRFS_BLOCK_GROUP_PROFILE_MASK);
> >+		return -EIO;
> >+	}
> >+
> >+	return 0;
> >+}
> >+
> >+static int read_one_chunk(struct btrfs_root *root, struct btrfs_key *key,
> >+			  struct extent_buffer *leaf,
> >+			  struct btrfs_chunk *chunk)
> >+{
> >+	struct btrfs_mapping_tree *map_tree = &root->fs_info->mapping_tree;
> >+	struct map_lookup *map;
> >+	struct extent_map *em;
> >+	u64 logical;
> >+	u64 length;
> >+	u64 stripe_len;
> >+	u64 devid;
> >+	u8 uuid[BTRFS_UUID_SIZE];
> >+	int num_stripes;
> >+	int ret;
> >+	int i;
> >+
> >+	logical = key->offset;
> >+	length = btrfs_chunk_length(leaf, chunk);
> >+	stripe_len = btrfs_chunk_stripe_len(leaf, chunk);
> >+	num_stripes = btrfs_chunk_num_stripes(leaf, chunk);
> >+	/* Validation check */
> >+	ret = btrfs_check_chunk_valid(root, leaf, chunk, logical);
> >+	if (ret)
> >+		return ret;
> >
> > 	read_lock(&map_tree->map_tree.lock);
> > 	em = lookup_extent_mapping(&map_tree->map_tree, logical, 1);
> >@@ -6502,6 +6544,7 @@ int btrfs_read_sys_array(struct btrfs_root *root)
> > 	u32 array_size;
> > 	u32 len = 0;
> > 	u32 cur_offset;
> >+	u64 type;
> > 	struct btrfs_key key;
> >
> > 	ASSERT(BTRFS_SUPER_INFO_SIZE <= root->nodesize);
> >@@ -6568,6 +6611,15 @@ int btrfs_read_sys_array(struct btrfs_root *root)
> > 				break;
> > 			}
> >
> >+			type = btrfs_chunk_type(sb, chunk);
> >+			if ((type & BTRFS_BLOCK_GROUP_SYSTEM) == 0) {
> >+				printk(KERN_ERR
> >+	    "BTRFS: invalid chunk type %llu in sys_array at offset %u\n",
> >+					type, cur_offset);
> >+				ret = -EIO;
> >+				break;
> >+			}
> >+
> > 			len = btrfs_chunk_item_size(num_stripes);
> > 			if (cur_offset + len > array_size)
> > 				goto out_short_read;
> >
> 
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

  reply	other threads:[~2016-05-03 23:36 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-05-02 18:15 [PATCH 1/2] Btrfs: add more valid checks for superblock Liu Bo
2016-05-02 18:15 ` [PATCH 2/2] Btrfs: add valid checks for chunk loading Liu Bo
2016-05-03  1:12   ` Qu Wenruo
2016-05-03 23:36     ` Liu Bo [this message]
2016-05-05  1:03       ` Qu Wenruo
2016-05-03  5:53   ` Anand Jain
2016-05-03 23:33     ` Liu Bo
2016-05-04 13:56   ` David Sterba
2016-05-13 23:57     ` Liu Bo
2016-05-17 13:37       ` David Sterba
2016-05-02 18:23 ` [PATCH 1/2] Btrfs: add more valid checks for superblock Liu Bo
2016-05-03  1:02 ` Qu Wenruo
2016-05-03 23:32   ` Liu Bo
2016-05-04 13:23   ` David Sterba
2016-05-04 17:44     ` Liu Bo
2016-05-05  1:08       ` Qu Wenruo
2016-05-06 14:35         ` David Sterba
2016-05-09  1:31           ` Qu Wenruo
2016-05-13 18:14             ` Liu Bo
2016-05-13 23:42               ` Qu Wenruo
2016-05-17 13:47                 ` David Sterba
2016-05-04 13:29 ` David Sterba
2016-05-04 17:40   ` Liu Bo
2016-05-06 14:39     ` 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=20160503233625.GG21008@localhost.localdomain \
    --to=bo.li.liu@oracle.com \
    --cc=linux-btrfs@vger.kernel.org \
    --cc=quwenruo@cn.fujitsu.com \
    --cc=sterba@suse.com \
    --cc=vegard.nossum@oracle.com \
    /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).