linux-ext4 - Re: [PATCH 1/2] ext4: Prevent panic for destroyed devices

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <87d1ru2wex.fsf@openvz.org>
Date:	Thu, 18 Feb 2016 12:35:34 +0300
From:	Dmitry Monakhov <dmonakhov@...nvz.org>
To:	linux-ext4@...r.kernel.org
Cc:	tytso@....edu
Subject: Re: [PATCH 1/2] ext4: Prevent panic for destroyed devices

Dmitry Monakhov <dmonakhov@...nvz.org> writes:

> Some devices(nbd) can becomes unoperatable via kill_bdev
> so its pagecache will be invalidated and all buffers becomes unmapped.
> In that situation we will trigger BUGON on submit_bh.
>
> #Testcase
> mkdir  -p a/mnt
> cd a
> truncate -s 1G img
> mkfs.ext4 -F img
> qemu-nbd -c /dev/nbd0 img
> mount /dev/nbd0 /mnt
> cp -r /bin/ /mnt&
> # Disconnect nbd while cp is active
> qemu-nbd -d /dev/nbd0
> sync

Hm...
Actually there are many places where we submit BHs w/o checks
nojournal code is full of such places.
It looks like we need some sort of AOP_TRUNCATE_PAGE error code
for generic buffer submission path

diff --git a/fs/buffer.c b/fs/buffer.c
index e1632ab..1cf94ec 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -3127,6 +3127,10 @@ int __sync_dirty_buffer(struct buffer_head *bh, int rw)
 
        WARN_ON(atomic_read(&bh->b_count) < 1);
        lock_buffer(bh);
+       if(unlikely(!buffer_mapped(bh))) {
+               unlock_buffer(bh);
+               return AOP_TRUNCATE_PAGE; /* or EIO */
+       }
        if (test_clear_buffer_dirty(bh)) {
                get_bh(bh);
                bh->b_end_io = end_buffer_write_sync;

>
> Signed-off-by: Dmitry Monakhov <dmonakhov@...nvz.org>
> ---
>  fs/ext4/super.c | 24 +++++++++++++++++++++---
>  1 file changed, 21 insertions(+), 3 deletions(-)
>
> diff --git a/fs/ext4/super.c b/fs/ext4/super.c
> index 3ed01ec..617d8b4a 100644
> --- a/fs/ext4/super.c
> +++ b/fs/ext4/super.c
> @@ -4320,7 +4320,7 @@ static int ext4_commit_super(struct super_block *sb, int sync)
>  	struct buffer_head *sbh = EXT4_SB(sb)->s_sbh;
>  	int error = 0;
>  
> -	if (!sbh || block_device_ejected(sb))
> +	if (!sbh || !buffer_mapped(sbh) || block_device_ejected(sb))
>  		return error;
>  	if (buffer_write_io_error(sbh)) {
>  		/*
> @@ -4368,8 +4368,26 @@ static int ext4_commit_super(struct super_block *sb, int sync)
>  	ext4_superblock_csum_set(sb);
>  	mark_buffer_dirty(sbh);
>  	if (sync) {
> -		error = __sync_dirty_buffer(sbh,
> -			test_opt(sb, BARRIER) ? WRITE_FUA : WRITE_SYNC);
> +		lock_buffer(sbh);
> +		/* superblock bh may be invalidated due to drive failure */
> +		if (!buffer_mapped(sbh)) {
> +			unlock_buffer(sbh);
> +			ext4_msg(sb, KERN_ERR, "Can not write superblock "
> +				 "because disk cache was invalidated");
> +			return -EIO;
> +		}
> +		if (test_clear_buffer_dirty(sbh)) {
> +			get_bh(sbh);
> +			sbh->b_end_io = end_buffer_write_sync;
> +			error = submit_bh(test_opt(sb, BARRIER) ?
> +					  WRITE_FUA : WRITE_SYNC, sbh);
> +			wait_on_buffer(sbh);
> +			if (!error && !buffer_uptodate(sbh))
> +				error = -EIO;
> +		} else {
> +			unlock_buffer(sbh);
> +		}
> +
>  		if (error)
>  			return error;
>  
> -- 
> 1.8.3.1
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
> the body of a message to majordomo@...r.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

Download attachment "signature.asc" of type "application/pgp-signature" (473 bytes)