From: Qu Wenruo <quwenruo@cn.fujitsu.com>
To: <linux-btrfs@vger.kernel.org>
Cc: <lukas.lueg@gmail.com>
Subject: [PATCH 4/4] btrfs-progs: Add extra chunk item check to avoid btrfs-progs crash.
Date: Wed, 13 May 2015 17:15:36 +0800 [thread overview]
Message-ID: <1431508536-7275-5-git-send-email-quwenruo@cn.fujitsu.com> (raw)
In-Reply-To: <1431508536-7275-1-git-send-email-quwenruo@cn.fujitsu.com>
Adds extra check when reading a chunk item:
1) Check chunk type.
Don't allow any unsupported type/profile bit.
2) Check num_stripes
Any chunk item should contain at least one stripe.
For system chunk, the chunk item size(calculated by btrfs_stripe size *
(num_stripes - 1) + btrfs_chunk size) should not exceed
BTRFS_SYSTEM_CHUNK_SIZE(2048).
For normal chunk, the chunk item size(calculated) should match the chunk
item size.
3) Check num_stripes/sub_stripes against chunk profile.
Num_stripes/sub_stripes must meet its lower limit for its chunk profile.
Reported-by: Lukas Lueg <lukas.lueg@gmail.com>
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
---
volumes.c | 60 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++---
1 file changed, 57 insertions(+), 3 deletions(-)
diff --git a/volumes.c b/volumes.c
index 16dbf64..77cc305 100644
--- a/volumes.c
+++ b/volumes.c
@@ -1575,9 +1575,14 @@ static struct btrfs_device *fill_missing_device(u64 devid)
return device;
}
+/*
+ * Slot is used to verfy the chunk item is valid
+ *
+ * For sys chunk in superblock, pass -1 to indicate sys chunk.
+ */
static int read_one_chunk(struct btrfs_root *root, struct btrfs_key *key,
struct extent_buffer *leaf,
- struct btrfs_chunk *chunk)
+ struct btrfs_chunk *chunk, int slot)
{
struct btrfs_mapping_tree *map_tree = &root->fs_info->mapping_tree;
struct map_lookup *map;
@@ -1615,6 +1620,51 @@ static int read_one_chunk(struct btrfs_root *root, struct btrfs_key *key,
map->type = btrfs_chunk_type(leaf, chunk);
map->sub_stripes = btrfs_chunk_sub_stripes(leaf, chunk);
+ /* Check on chunk item type */
+ if (map->type & ~(BTRFS_BLOCK_GROUP_TYPE_MASK |
+ BTRFS_BLOCK_GROUP_PROFILE_MASK)) {
+ fprintf(stderr, "Unknown chunk type bits: %llu\n",
+ map->type & ~(BTRFS_BLOCK_GROUP_TYPE_MASK |
+ BTRFS_BLOCK_GROUP_PROFILE_MASK));
+ ret = -EIO;
+ goto out;
+ }
+
+ /*
+ * Btrfs_chunk contains at least one stripe, and for sys_chunk
+ * it can't exceed the system chunk array size
+ * For normal chunk, it should match its chunk item size.
+ */
+ if (num_stripes < 1 ||
+ (slot == -1 && sizeof(struct btrfs_stripe) * num_stripes >
+ BTRFS_SYSTEM_CHUNK_ARRAY_SIZE) ||
+ (slot >= 0 && sizeof(struct btrfs_stripe) * (num_stripes - 1) >
+ btrfs_item_size_nr(leaf, slot))) {
+ fprintf(stderr, "Invalid num_stripes: %u\n",
+ num_stripes);
+ ret = -EIO;
+ goto out;
+ }
+
+ /*
+ * Device number check against profile
+ */
+ if ((map->type & BTRFS_BLOCK_GROUP_RAID10 && num_stripes < 4 &&
+ map->sub_stripes < 2) ||
+ (map->type & BTRFS_BLOCK_GROUP_RAID1 && num_stripes < 2) ||
+ (map->type & BTRFS_BLOCK_GROUP_RAID5 && num_stripes < 3) ||
+ (map->type & BTRFS_BLOCK_GROUP_RAID6 && num_stripes < 4) ||
+ (map->type & BTRFS_BLOCK_GROUP_DUP && num_stripes != 1) ||
+ ((map->type & BTRFS_BLOCK_GROUP_PROFILE_MASK) == 0 &&
+ num_stripes != 1)) {
+ fprintf(stderr,
+ "Invalid num_stripes:sub_stripes %u:%u for profile %llu\n",
+ num_stripes, map->sub_stripes,
+ map->type & BTRFS_BLOCK_GROUP_PROFILE_MASK);
+ ret = -EIO;
+ goto out;
+ }
+
for (i = 0; i < num_stripes; i++) {
map->stripes[i].physical =
btrfs_stripe_offset_nr(leaf, chunk, i);
@@ -1635,6 +1685,9 @@ static int read_one_chunk(struct btrfs_root *root, struct btrfs_key *key,
BUG_ON(ret);
return 0;
+out:
+ free(map);
+ return ret;
}
static int fill_device_from_item(struct extent_buffer *leaf,
@@ -1773,7 +1826,7 @@ int btrfs_read_sys_array(struct btrfs_root *root)
if (key.type == BTRFS_CHUNK_ITEM_KEY) {
chunk = (struct btrfs_chunk *)(ptr - (u8 *)super_copy);
- ret = read_one_chunk(root, &key, sb, chunk);
+ ret = read_one_chunk(root, &key, sb, chunk, -1);
if (ret)
break;
num_stripes = btrfs_chunk_num_stripes(sb, chunk);
@@ -1835,7 +1888,8 @@ int btrfs_read_chunk_tree(struct btrfs_root *root)
} else if (found_key.type == BTRFS_CHUNK_ITEM_KEY) {
struct btrfs_chunk *chunk;
chunk = btrfs_item_ptr(leaf, slot, struct btrfs_chunk);
- ret = read_one_chunk(root, &found_key, leaf, chunk);
+ ret = read_one_chunk(root, &found_key, leaf, chunk,
+ slot);
BUG_ON(ret);
}
path->slots[0]++;
--
2.4.0
next prev parent reply other threads:[~2015-05-13 9:18 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-05-13 9:15 [PATCH 0/4] Enhance btrfs-progs for hostile image (FPE error part) Qu Wenruo
2015-05-13 9:15 ` [PATCH 1/4] btrfs-progs: Remove non-exist csum size Qu Wenruo
2015-05-13 16:18 ` David Sterba
2015-05-13 9:15 ` [PATCH 2/4] btrfs-progs: Read the whole superblock instead of struct btrfs_super_block Qu Wenruo
2015-05-13 16:14 ` David Sterba
2015-09-25 20:50 ` David Sterba
2015-05-13 9:15 ` [PATCH 3/4] btrfs-progs: Introduce better superblock check Qu Wenruo
2015-09-25 20:55 ` David Sterba
2015-05-13 9:15 ` Qu Wenruo [this message]
2015-05-13 16:18 ` [PATCH 4/4] btrfs-progs: Add extra chunk item check to avoid btrfs-progs crash David Sterba
2015-05-19 16:33 ` WorMzy Tykashi
[not found] ` <CALOYprV-dyji_5UFm3K-qmw0K2i5u7d14rwX7OPMFS29aY9ZVA@mail.gmail.com>
2015-05-20 0:25 ` Qu Wenruo
2015-05-20 8:03 ` WorMzy Tykashi
2015-05-20 12:20 ` David Sterba
2015-05-21 0:21 ` Qu Wenruo
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=1431508536-7275-5-git-send-email-quwenruo@cn.fujitsu.com \
--to=quwenruo@cn.fujitsu.com \
--cc=linux-btrfs@vger.kernel.org \
--cc=lukas.lueg@gmail.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).