From: Damien Le Moal <dlemoal@kernel.org>
To: Hui Peng <benquike@gmail.com>,
brauner@kernel.org, jack@suse.cz, jlayton@kernel.org
Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] qnx6: validate di_filelevels in qnx6_iget() and fix mount error handling
Date: Sun, 20 Sep 2026 13:02:44 +0900 [thread overview]
Message-ID: <f68323da-e90f-46a4-b3ce-70d26de115e6@kernel.org> (raw)
In-Reply-To: <20260919222556.3792829-1-benquike@gmail.com>
On 9/20/26 07:25, Hui Peng wrote:
> Fix two issues in fs/qnx6/:
>
> 1. In qnx6_iget(), reject raw_inode->di_filelevels > QNX6_PTR_MAX_LEVELS
> so computing the maximum block count does not trigger shift-out-of-
> bounds undefined behavior or out-of-bounds block pointer array walks.
> 2. In qnx6_fill_super() and qnx6_mmi_fill_super(), prevent MS_SILENT
> from bypassing superblock magic validation and ensure sbi->sb_buf is
> released exactly once on failure.
Please split this into 2 patches, one patch for each problem you fix.
>
> Fixes: 5d026c724220 ("fs: initial qnx6fs addition")
> Assisted-by: LLM
> Signed-off-by: Hui Peng <benquike@gmail.com>
> ---
> diff --git a/fs/qnx6/inode.c b/fs/qnx6/inode.c
> index 6de49333acad..04214b829e10 100644
> --- a/fs/qnx6/inode.c
> +++ b/fs/qnx6/inode.c
> @@ -144,8 +144,10 @@ static unsigned qnx6_block_map(struct inode *inode, unsigned no)
> levelptr = (no >> bitdelta) & mask;
> ptr = ((__fs32 *)bh->b_data)[levelptr];
>
> - if (!qnx6_check_blockptr(ptr))
> + if (!qnx6_check_blockptr(ptr)) {
> + brelse(bh);
> return 0;
> + }
>
> block = qnx6_get_devblock(s, ptr);
> brelse(bh);
> @@ -397,12 +399,14 @@ static int qnx6_fill_super(struct super_block *s, struct fs_context *fc)
> sbi->sb_buf = bh1;
> sbi->sb = (struct qnx6_super_block *)bh1->b_data;
> brelse(bh2);
> + bh2 = NULL;
> pr_info("superblock #1 active\n");
> } else {
> /* superblock #2 active */
> sbi->sb_buf = bh2;
> sbi->sb = (struct qnx6_super_block *)bh2->b_data;
> brelse(bh1);
> + bh1 = NULL;
> pr_info("superblock #2 active\n");
> }
> mmi_success:
> @@ -463,6 +467,8 @@ static int qnx6_fill_super(struct super_block *s, struct fs_context *fc)
> out1:
> iput(sbi->inodes);
> out:
> + if (sbi->sb_buf && sbi->sb_buf != bh1 && sbi->sb_buf != bh2)
> + brelse(sbi->sb_buf);
> brelse(bh1);
> brelse(bh2);
> outnobh:
> @@ -560,6 +566,13 @@ struct inode *qnx6_iget(struct super_block *sb, unsigned ino)
> memcpy(&ei->di_block_ptr, &raw_inode->di_block_ptr,
> sizeof(raw_inode->di_block_ptr));
> ei->di_filelevels = raw_inode->di_filelevels;
> + if (ei->di_filelevels > QNX6_PTR_MAX_LEVELS) {
> + pr_err("invalid filelevels (%u) in inode %u\n",
> + ei->di_filelevels, ino);
> + folio_release_kmap(folio, raw_inode);
> + iget_failed(inode);
> + return ERR_PTR(-EIO);
> + }
>
> if (S_ISREG(inode->i_mode)) {
> inode->i_fop = &generic_ro_fops;
> diff --git a/fs/qnx6/super_mmi.c b/fs/qnx6/super_mmi.c
> index b8afb6f388b2..28cb9322278e 100644
> --- a/fs/qnx6/super_mmi.c
> +++ b/fs/qnx6/super_mmi.c
> @@ -51,10 +51,9 @@ struct qnx6_super_block *qnx6_mmi_fill_super(struct super_block *s, int silent)
> sb1 = (struct qnx6_mmi_super_block *)bh1->b_data;
> sbi = QNX6_SB(s);
> if (fs32_to_cpu(sbi, sb1->sb_magic) != QNX6_SUPER_MAGIC) {
> - if (!silent) {
> + if (!silent)
> pr_err("wrong signature (magic) in superblock #1.\n");
> - goto out;
> - }
> + goto out;
> }
>
> /* checksum check - start at byte 8 and end at byte 512 */
> @@ -64,15 +63,16 @@ struct qnx6_super_block *qnx6_mmi_fill_super(struct super_block *s, int silent)
> goto out;
> }
>
> - /* calculate second superblock blocknumber */
> - offset = fs32_to_cpu(sbi, sb1->sb_num_blocks) + QNX6_SUPERBLOCK_AREA /
> - fs32_to_cpu(sbi, sb1->sb_blocksize);
> -
> /* set new blocksize */
> if (!sb_set_blocksize(s, fs32_to_cpu(sbi, sb1->sb_blocksize))) {
> pr_err("unable to set blocksize\n");
> goto out;
> }
> +
> + /* calculate second superblock blocknumber */
> + offset = fs32_to_cpu(sbi, sb1->sb_num_blocks) + QNX6_SUPERBLOCK_AREA /
> + fs32_to_cpu(sbi, sb1->sb_blocksize);
> +
> /* blocksize invalidates bh - pull it back in */
> brelse(bh1);
> bh1 = sb_bread(s, 0);
--
Damien Le Moal
Western Digital Research
prev parent reply other threads:[~2026-09-20 4:02 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-09-19 22:25 Hui Peng
2026-09-20 4:02 ` Damien Le Moal [this message]
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=f68323da-e90f-46a4-b3ce-70d26de115e6@kernel.org \
--to=dlemoal@kernel.org \
--cc=benquike@gmail.com \
--cc=brauner@kernel.org \
--cc=jack@suse.cz \
--cc=jlayton@kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@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
all inboxes | Powered by JetHome®