linux-kernel - Re: [PATCH 1/1] nfsd: incorrect check for debugfs returns

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]

Date:	Wed, 25 Mar 2015 11:17:13 -0400
From:	"J. Bruce Fields" <bfields@...ldses.org>
To:	Chengyu Song <csong84@...ech.edu>
Cc:	linux-nfs@...r.kernel.org, linux-kernel@...r.kernel.org,
	taesoo@...ech.edu, changwoo@...ech.edu, sanidhya@...ech.edu,
	blee@...ech.edu
Subject: Re: [PATCH 1/1] nfsd: incorrect check for debugfs returns

On Mon, Mar 23, 2015 at 10:58:05PM -0400, Chengyu Song wrote:
> debugfs_create_dir and debugfs_create_file may return -ENODEV when debugfs
> is not configured, so the return value should be checked against ERROR_VALUE
> as well, otherwise the later dereference of the dentry pointer would crash
> the kernel.

Thanks for spotting this.  But it looks like this will cause nfsd
startup to fail when debugfs isn't configured.  I'd rather we didn't, it
just isn't that important.

So I'd rather just make nfsd_fault_inject_init() a void return--just do
a dprintk as a warning in the "fail" case, and otherwise let normal
startup continue (and check that doesn't lead to other unsafe
dereferences of debug_dir).  Could you try that?

--b.


> 
> Signed-off-by: Chengyu Song <csong84@...ech.edu>
> ---
>  fs/nfsd/fault_inject.c | 12 ++++++++----
>  1 file changed, 8 insertions(+), 4 deletions(-)
> 
> diff --git a/fs/nfsd/fault_inject.c b/fs/nfsd/fault_inject.c
> index c16bf5a..621d065 100644
> --- a/fs/nfsd/fault_inject.c
> +++ b/fs/nfsd/fault_inject.c
> @@ -132,19 +132,23 @@ int nfsd_fault_inject_init(void)
>  	unsigned int i;
>  	struct nfsd_fault_inject_op *op;
>  	umode_t mode = S_IFREG | S_IRUSR | S_IWUSR;
> +	struct dentry *dent;
>  
> -	debug_dir = debugfs_create_dir("nfsd", NULL);
> -	if (!debug_dir)
> +	dent = debugfs_create_dir("nfsd", NULL);
> +	if (IS_ERR_OR_NULL(dent))
>  		goto fail;
> +	debug_dir = dent;
>  
>  	for (i = 0; i < NUM_INJECT_OPS; i++) {
>  		op = &inject_ops[i];
> -		if (!debugfs_create_file(op->file, mode, debug_dir, op, &fops_nfsd))
> +		dent = debugfs_create_file(op->file, mode, debug_dir, op, &fops_nfsd);
> +		if (IS_ERR_OR_NULL(dent))
>  			goto fail;
> +
>  	}
>  	return 0;
>  
>  fail:
>  	nfsd_fault_inject_cleanup();
> -	return -ENOMEM;
> +	return dent ? PTR_ERR(dent) : -ENOMEM;
>  }
> -- 
> 2.1.0
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/