From: Zheng Liu <gnehzuil.liu@gmail.com>
To: Theodore Ts'o <tytso@mit.edu>
Cc: Ext4 Developers List <linux-ext4@vger.kernel.org>
Subject: Re: [PATCH 5/5] libext2fs: avoid 32-bit overflow in ext2fs_initialize with a 512M cluster size
Date: Tue, 15 Jan 2013 23:33:31 +0800 [thread overview]
Message-ID: <20130115153331.GE19209@gmail.com> (raw)
In-Reply-To: <1358210232-30578-5-git-send-email-tytso@mit.edu>
On Mon, Jan 14, 2013 at 07:37:12PM -0500, Theodore Ts'o wrote:
> If the user attemps to create a 512MB cluster, we need to adjust the
> defaults to avoid a 32-bit overflow of s_blocks_per_group. Also check
> to make sure that the caller of ext2fs_initialize() has not given a
> value of s_clusters_per_group that would result in an overflow of
> s_blocks_per_group.
>
> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
The patch itself looks good to me. Feel free to add:
Reviewed-by: Zheng Liu <wenqing.lz@taobao.com>
FWIW, I wonder why we need to add such complex logical to handle a
corner case. I guess no one wants to use a 512MB cluster. So changing
max cluster size from 512MB to 256MB is very simple and straightfoward.
Regards,
- Zheng
> ---
> lib/ext2fs/initialize.c | 15 ++++++++++++---
> 1 file changed, 12 insertions(+), 3 deletions(-)
>
> diff --git a/lib/ext2fs/initialize.c b/lib/ext2fs/initialize.c
> index b0c15d2..5afdc27 100644
> --- a/lib/ext2fs/initialize.c
> +++ b/lib/ext2fs/initialize.c
> @@ -207,6 +207,8 @@ errcode_t ext2fs_initialize(const char *name, int flags,
> super->s_log_block_size;
>
> if (bigalloc_flag) {
> + unsigned long long bpg;
> +
> if (param->s_blocks_per_group &&
> param->s_clusters_per_group &&
> ((param->s_clusters_per_group * EXT2FS_CLUSTER_RATIO(fs)) !=
> @@ -220,12 +222,19 @@ errcode_t ext2fs_initialize(const char *name, int flags,
> super->s_clusters_per_group =
> param->s_blocks_per_group /
> EXT2FS_CLUSTER_RATIO(fs);
> - else
> + else if (super->s_log_cluster_size + 15 < 32)
> super->s_clusters_per_group = fs->blocksize * 8;
> + else
> + super->s_clusters_per_group = (fs->blocksize - 1) * 8;
> if (super->s_clusters_per_group > EXT2_MAX_CLUSTERS_PER_GROUP(super))
> super->s_clusters_per_group = EXT2_MAX_CLUSTERS_PER_GROUP(super);
> - super->s_blocks_per_group = EXT2FS_C2B(fs,
> - super->s_clusters_per_group);
> + bpg = EXT2FS_C2B(fs,
> + (unsigned long long) super->s_clusters_per_group);
> + if (bpg >= (((unsigned long long) 1) << 32)) {
> + retval = EXT2_ET_INVALID_ARGUMENT;
> + goto cleanup;
> + }
> + super->s_blocks_per_group = bpg;
> } else {
> set_field(s_blocks_per_group, fs->blocksize * 8);
> if (super->s_blocks_per_group > EXT2_MAX_BLOCKS_PER_GROUP(super))
> --
> 1.7.12.rc0.22.gcdd159b
>
next prev parent reply other threads:[~2013-01-15 15:19 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-01-13 9:08 [PATCH 1/3] mke2fs: indicate bigalloc feature explicity when cluster-size is enabled Zheng Liu
2013-01-13 9:08 ` [PATCH 2/3] mke2fs: reduce the range of cluster-size Zheng Liu
2013-01-14 17:41 ` Andreas Dilger
2013-01-14 21:03 ` Theodore Ts'o
2013-01-14 21:07 ` Andreas Dilger
2013-01-14 21:10 ` Theodore Ts'o
2013-01-15 0:37 ` [PATCH 1/5] mke2fs: enforce that the cluster size must be less that the block size Theodore Ts'o
2013-01-15 0:37 ` [PATCH 2/5] mke2fs: the -g option will now specify the clusters per block group Theodore Ts'o
2013-01-15 15:10 ` Eric Sandeen
2013-01-15 19:05 ` Theodore Ts'o
2013-01-15 15:22 ` Zheng Liu
2013-01-15 0:37 ` [PATCH 3/5] libe2p: teach parse_num_blocks2() to return bytes if log_block_size < 0 Theodore Ts'o
2013-01-15 15:23 ` Zheng Liu
2013-01-15 0:37 ` [PATCH 4/5] mke2fs: teach mke2fs to understand -b 4k and -C 256M Theodore Ts'o
2013-01-15 15:11 ` Eric Sandeen
2013-01-15 15:13 ` Eric Sandeen
2013-01-15 15:24 ` Zheng Liu
2013-01-15 0:37 ` [PATCH 5/5] libext2fs: avoid 32-bit overflow in ext2fs_initialize with a 512M cluster size Theodore Ts'o
2013-01-15 15:33 ` Zheng Liu [this message]
2013-01-15 15:36 ` Zheng Liu
2013-01-15 19:10 ` Theodore Ts'o
2013-01-16 1:49 ` Zheng Liu
2013-01-15 0:41 ` [PATCH 1/5] mke2fs: enforce that the cluster size must be less that the block size Theodore Ts'o
2013-01-15 15:22 ` Zheng Liu
2013-01-13 9:08 ` [PATCH 3/3] mke2fs: document bigalloc and cluster-size Zheng Liu
2013-01-15 3:10 ` Theodore Ts'o
2013-01-15 19:12 ` Theodore Ts'o
2013-01-15 19:46 ` Phillip Susi
2013-01-15 19:57 ` Theodore Ts'o
2013-01-15 20:38 ` Phillip Susi
2013-01-15 22:28 ` Theodore Ts'o
2013-01-14 20:28 ` [PATCH 1/3] mke2fs: indicate bigalloc feature explicity when cluster-size is enabled Theodore Ts'o
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=20130115153331.GE19209@gmail.com \
--to=gnehzuil.liu@gmail.com \
--cc=linux-ext4@vger.kernel.org \
--cc=tytso@mit.edu \
/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).