Re: [PATCH] jfs : fix array-index-out-of-bounds in diWrite

[Dave Kleikamp <dave.kleikamp@oracle.com>]

On 10/8/23 12:47PM, Manas Ghandat wrote:
> Currently while copying dtree root from inode to dnode in the xp slot
> there is a out of bound memcpy. Added a bound check to the memcpy.
> 
> Reported-by: syzbot+c1056fdfe414463fdb33@syzkaller.appspotmail.com
> Fixes: https://syzkaller.appspot.com/bug?extid=c1056fdfe414463fdb33
> Signed-off-by: Manas Ghandat <ghandatmanas@gmail.com>
> ---
>   fs/jfs/jfs_imap.c | 3 ++-
>   1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c
> index 799d3837e7c2..d1f897848be0 100644
> --- a/fs/jfs/jfs_imap.c
> +++ b/fs/jfs/jfs_imap.c
> @@ -746,7 +746,8 @@ int diWrite(tid_t tid, struct inode *ip)
>   		xp = (dtpage_t *) & dp->di_dtroot;
>   		lv = ilinelock->lv;
>   		for (n = 0; n < ilinelock->index; n++, lv++) {
> -			memcpy(&xp->slot[lv->offset], &p->slot[lv->offset],
> +			if (lv->offset < 128)

I'd use DTPAGEMAXSLOT instead of hard-coding 128.

Also, we could at least set rc to -EIO, or we could do something more.
Calling jfs_error() would mark the superblock dirty and usually makes 
the file system read-only until it's fixed.

> +				memcpy(&xp->slot[lv->offset], &p->slot[lv->offset],
>   			       lv->length << L2DTSLOTSIZE);
>   		}
>   	} else {