[v1] fs/ntfs: fix BUG_ON of ntfs_read_block()

[PATCH] fs/ntfs: fix BUG_ON of ntfs_read_block()

Posted by cgel.zte@gmail.com 3 years, 10 months ago

From: xu xin <xu.xin16@zte.com.cn>

As the bug description, attckers can use this bug to crash the system
When CONFIG_NTFS_FS is set.

So remove the BUG_ON, and use WARN and return instead until someone
really solve the bug.

Reported-by: Zeal Robot <zealci@zte.com.cn>
Reported-by: syzbot+6a5a7672f663cce8b156@syzkaller.appspotmail.com
Reviewed-by: Songyi Zhang <zhang.songyi@zte.com.cn>
Reviewed-by: Yang Yang <yang.yang29@zte.com.cn>
Reviewed-by: Jiang Xuexin<jiang.xuexin@zte.com.cn>
Reviewed-by: Zhang wenya<zhang.wenya1@zte.com.cn>
Signed-off-by: xu xin <xu.xin16@zte.com.cn>
---
 fs/ntfs/aops.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c
index 5f4fb6ca6f2e..b6fd7e711420 100644
--- a/fs/ntfs/aops.c
+++ b/fs/ntfs/aops.c
@@ -183,7 +183,11 @@ static int ntfs_read_block(struct page *page)
 	vol = ni->vol;
 
 	/* $MFT/$DATA must have its complete runlist in memory at all times. */
-	BUG_ON(!ni->runlist.rl && !ni->mft_no && !NInoAttr(ni));
+	if (unlikely(!ni->runlist.rl && !ni->mft_no && !NInoAttr(ni))) {
+		WARN(1, "NTFS: ni->runlist.rl, ni->mft_no, and NInoAttr(ni) is null!\n");
+		unlock_page(page);
+		return -EINVAL;
+	}
 
 	blocksize = vol->sb->s_blocksize;
 	blocksize_bits = vol->sb->s_blocksize_bits;
-- 
2.25.1

Re: [PATCH] fs/ntfs: fix BUG_ON of ntfs_read_block()

Posted by Greg KH 3 years, 10 months ago

On Thu, Jun 23, 2022 at 03:51:31AM +0000, cgel.zte@gmail.com wrote:
> From: xu xin <xu.xin16@zte.com.cn>
> 
> As the bug description, attckers can use this bug to crash the system
> When CONFIG_NTFS_FS is set.
> 
> So remove the BUG_ON, and use WARN and return instead until someone
> really solve the bug.
> 
> Reported-by: Zeal Robot <zealci@zte.com.cn>
> Reported-by: syzbot+6a5a7672f663cce8b156@syzkaller.appspotmail.com
> Reviewed-by: Songyi Zhang <zhang.songyi@zte.com.cn>
> Reviewed-by: Yang Yang <yang.yang29@zte.com.cn>
> Reviewed-by: Jiang Xuexin<jiang.xuexin@zte.com.cn>
> Reviewed-by: Zhang wenya<zhang.wenya1@zte.com.cn>
> Signed-off-by: xu xin <xu.xin16@zte.com.cn>
> ---
>  fs/ntfs/aops.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c
> index 5f4fb6ca6f2e..b6fd7e711420 100644
> --- a/fs/ntfs/aops.c
> +++ b/fs/ntfs/aops.c
> @@ -183,7 +183,11 @@ static int ntfs_read_block(struct page *page)
>  	vol = ni->vol;
>  
>  	/* $MFT/$DATA must have its complete runlist in memory at all times. */
> -	BUG_ON(!ni->runlist.rl && !ni->mft_no && !NInoAttr(ni));
> +	if (unlikely(!ni->runlist.rl && !ni->mft_no && !NInoAttr(ni))) {
> +		WARN(1, "NTFS: ni->runlist.rl, ni->mft_no, and NInoAttr(ni) is null!\n");
> +		unlock_page(page);
> +		return -EINVAL;
> +	}
>  
>  	blocksize = vol->sb->s_blocksize;
>  	blocksize_bits = vol->sb->s_blocksize_bits;
> -- 
> 2.25.1
> 

<formletter>

This is not the correct way to submit patches for inclusion in the
stable kernel tree.  Please read:
    https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
for how to do this properly.

</formletter>

Re: [PATCH] fs/ntfs: fix BUG_ON of ntfs_read_block()

Posted by CGEL 3 years, 10 months ago

On Thu, Jun 23, 2022 at 09:57:39AM +0200, Greg KH wrote:
> On Thu, Jun 23, 2022 at 03:51:31AM +0000, cgel.zte@gmail.com wrote:
> > From: xu xin <xu.xin16@zte.com.cn>
> > 
> > As the bug description, attckers can use this bug to crash the system
> > When CONFIG_NTFS_FS is set.
> > 
> > So remove the BUG_ON, and use WARN and return instead until someone
> > really solve the bug.
> > 
> > Reported-by: Zeal Robot <zealci@zte.com.cn>
> > Reported-by: syzbot+6a5a7672f663cce8b156@syzkaller.appspotmail.com
> > Reviewed-by: Songyi Zhang <zhang.songyi@zte.com.cn>
> > Reviewed-by: Yang Yang <yang.yang29@zte.com.cn>
> > Reviewed-by: Jiang Xuexin<jiang.xuexin@zte.com.cn>
> > Reviewed-by: Zhang wenya<zhang.wenya1@zte.com.cn>
> > Signed-off-by: xu xin <xu.xin16@zte.com.cn>
> > ---
> >  fs/ntfs/aops.c | 6 +++++-
> >  1 file changed, 5 insertions(+), 1 deletion(-)
> > 
> > --- a/fs/ntfs/aops.c
> > +++ b/fs/ntfs/aops.c
> > @@ -183,7 +183,11 @@ static int ntfs_read_block(struct page *page)
> >  	vol = ni->vol;
> >  
> >  	/* $MFT/$DATA must have its complete runlist in memory at all times. */
> > -	BUG_ON(!ni->runlist.rl && !ni->mft_no && !NInoAttr(ni));
> > +	if (unlikely(!ni->runlist.rl && !ni->mft_no && !NInoAttr(ni))) {
> > +		WARN(1, "NTFS: ni->runlist.rl, ni->mft_no, and NInoAttr(ni) is null!\n");
> > +		unlock_page(page);
> > +		return -EINVAL;
> > +	}
> >  
> >  	blocksize = vol->sb->s_blocksize;
> >  	blocksize_bits = vol->sb->s_blocksize_bits;
> > -- 
> > 2.25.1
> > 
> 
> <formletter>
> 
> This is not the correct way to submit patches for inclusion in the
> stable kernel tree.  Please read:
>     https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html
> for how to do this properly.
>

Sorry. I'll rewrite a patch to fix it. 

Thanks.

> </formletter>

Re: [PATCH] fs/ntfs: fix BUG_ON of ntfs_read_block()

Posted by Greg KH 3 years, 10 months ago

On Thu, Jun 23, 2022 at 03:51:31AM +0000, cgel.zte@gmail.com wrote:
> From: xu xin <xu.xin16@zte.com.cn>
> 
> As the bug description, attckers can use this bug to crash the system
> When CONFIG_NTFS_FS is set.
> 
> So remove the BUG_ON, and use WARN and return instead until someone
> really solve the bug.
> 
> Reported-by: Zeal Robot <zealci@zte.com.cn>
> Reported-by: syzbot+6a5a7672f663cce8b156@syzkaller.appspotmail.com
> Reviewed-by: Songyi Zhang <zhang.songyi@zte.com.cn>
> Reviewed-by: Yang Yang <yang.yang29@zte.com.cn>
> Reviewed-by: Jiang Xuexin<jiang.xuexin@zte.com.cn>
> Reviewed-by: Zhang wenya<zhang.wenya1@zte.com.cn>
> Signed-off-by: xu xin <xu.xin16@zte.com.cn>
> ---
>  fs/ntfs/aops.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c
> index 5f4fb6ca6f2e..b6fd7e711420 100644
> --- a/fs/ntfs/aops.c
> +++ b/fs/ntfs/aops.c
> @@ -183,7 +183,11 @@ static int ntfs_read_block(struct page *page)
>  	vol = ni->vol;
>  
>  	/* $MFT/$DATA must have its complete runlist in memory at all times. */
> -	BUG_ON(!ni->runlist.rl && !ni->mft_no && !NInoAttr(ni));
> +	if (unlikely(!ni->runlist.rl && !ni->mft_no && !NInoAttr(ni))) {
> +		WARN(1, "NTFS: ni->runlist.rl, ni->mft_no, and NInoAttr(ni) is null!\n");

So for systems with panic-on-warn, you are still crashing?  Why is this
WARN() line still needed here?

thanks,

greg k-h

Re: [PATCH] fs/ntfs: fix BUG_ON of ntfs_read_block()

Posted by CGEL 3 years, 10 months ago

On Thu, Jun 23, 2022 at 09:57:21AM +0200, Greg KH wrote:
> On Thu, Jun 23, 2022 at 03:51:31AM +0000, cgel.zte@gmail.com wrote:
> > From: xu xin <xu.xin16@zte.com.cn>
> > 
> > As the bug description, attckers can use this bug to crash the system
> > When CONFIG_NTFS_FS is set.
> > 
> > So remove the BUG_ON, and use WARN and return instead until someone
> > really solve the bug.
> > 
> > Reported-by: Zeal Robot <zealci@zte.com.cn>
> > Reported-by: syzbot+6a5a7672f663cce8b156@syzkaller.appspotmail.com
> > Reviewed-by: Songyi Zhang <zhang.songyi@zte.com.cn>
> > Reviewed-by: Yang Yang <yang.yang29@zte.com.cn>
> > Reviewed-by: Jiang Xuexin<jiang.xuexin@zte.com.cn>
> > Reviewed-by: Zhang wenya<zhang.wenya1@zte.com.cn>
> > Signed-off-by: xu xin <xu.xin16@zte.com.cn>
> > ---
> >  fs/ntfs/aops.c | 6 +++++-
> >  1 file changed, 5 insertions(+), 1 deletion(-)
> > 
> > diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c
> > index 5f4fb6ca6f2e..b6fd7e711420 100644
> > --- a/fs/ntfs/aops.c
> > +++ b/fs/ntfs/aops.c
> > @@ -183,7 +183,11 @@ static int ntfs_read_block(struct page *page)
> >  	vol = ni->vol;
> >  
> >  	/* $MFT/$DATA must have its complete runlist in memory at all times. */
> > -	BUG_ON(!ni->runlist.rl && !ni->mft_no && !NInoAttr(ni));
> > +	if (unlikely(!ni->runlist.rl && !ni->mft_no && !NInoAttr(ni))) {
> > +		WARN(1, "NTFS: ni->runlist.rl, ni->mft_no, and NInoAttr(ni) is null!\n");
> 
> So for systems with panic-on-warn, you are still crashing?  Why is this
> WARN() line still needed here?
>

Sorry, I forgot about panic-on-warn. Use pr_warn() may be better.
I'll send a patch-v2 .

> thanks,
> 
> greg k-h