Linux Security Modules development
 help / color / mirror / Atom feed
From: Paul Moore <paul@paul-moore.com>
To: Cai Xinchen <caixinchen1@huawei.com>, <jmorris@namei.org>,
	<serge@hallyn.com>, <stephen.smalley.work@gmail.com>,
	<omosnace@redhat.com>, <amir73il@gmail.com>, <brauner@kernel.org>
Cc: <linux-security-module@vger.kernel.org>,
	<linux-kernel@vger.kernel.org>, <selinux@vger.kernel.org>,
	<caixinchen1@huawei.com>, <lujialin4@huawei.com>
Subject: Re: [PATCH 2/2] security: Fix call security_backing_file_free second  time
Date: Thu, 02 Jul 2026 17:09:18 -0400	[thread overview]
Message-ID: <aa30283e5827cdff22a62421c8d8e088@paul-moore.com> (raw)
In-Reply-To: <20260626011720.1144213-3-caixinchen1@huawei.com>

On Jun 25, 2026 Cai Xinchen <caixinchen1@huawei.com> wrote:
> 
> I found the following path:
> 
> alloc_empty_backing-file
>     init_file(&ff->file, xxx)
>         -> file_ref_init(&f->f_ref, 1); // only 1
>     error = init_backing_file
>         -> security_backing_file_alloc
>         -> rc = call_int_hook(backing_file_alloc, ...)

The good news is that as you mentioned, only SELinux defines a
backing_file_alloc hook and it always returns success/0.

>            if (unlikely(rc))
>            	security_backing_file_free(backing_file); // first call
>     if (unlikely(error)) {
>         fput(&ff->file);
>          -> if (unlikely(file_ref_put(&file->f_ref))) // zero
>                 __fput_deferred(file);
>                  -> ____fput -> __fput -> file_free(file);
>                  -> backing_file_free(backing_file(f));
>                  -> security_backing_file_free(&ff->file); // second call
> 
> Currently, only SELinux has the lsm backing_file_alloc hook, and the
> backing_file_free hook is not set. When security_backing_file_free is
> called for the first time, the blobs pointer is set to NULL. Therefore,
> double free will not occur in the code.
> 
> Fixes: 6af36aeb147a ("lsm: add backing_file LSM hooks")
> Signed-off-by: Cai Xinchen <caixinchen1@huawei.com>
> ---
>  security/security.c | 5 +----
>  1 file changed, 1 insertion(+), 4 deletions(-)
> 
> diff --git a/security/security.c b/security/security.c
> index 71aea8fdf014..595d3c73253e 100644
> --- a/security/security.c
> +++ b/security/security.c
> @@ -2468,11 +2468,8 @@ int security_backing_file_alloc(struct file *backing_file,
>  	rc = lsm_backing_file_alloc(backing_file);
>  	if (rc)
>  		return rc;
> -	rc = call_int_hook(backing_file_alloc, backing_file, user_file);
> -	if (unlikely(rc))
> -		security_backing_file_free(backing_file);
>  
> -	return rc;
> +	return call_int_hook(backing_file_alloc, backing_file, user_file);
>  }

I think the better option would be to move the
call_void_hook(backing_file_free, ...) call in security_backing_file_free()
into the if-statment true block before we set the backing file's LSM blob
pointer to NULL and free the LSM blob.

--
paul-moore.com

      reply	other threads:[~2026-07-02 21:09 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-06-26  1:17 [PATCH -next 0/2] Fix call security_backing_file_free second time Cai Xinchen
2026-06-26  1:17 ` [PATCH -next 1/2] security: Some cleanup code Cai Xinchen
2026-07-02 21:09   ` [PATCH " Paul Moore
2026-06-26  1:17 ` [PATCH -next 2/2] security: Fix call security_backing_file_free second time Cai Xinchen
2026-07-02 21:09   ` Paul Moore [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=aa30283e5827cdff22a62421c8d8e088@paul-moore.com \
    --to=paul@paul-moore.com \
    --cc=amir73il@gmail.com \
    --cc=brauner@kernel.org \
    --cc=caixinchen1@huawei.com \
    --cc=jmorris@namei.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-security-module@vger.kernel.org \
    --cc=lujialin4@huawei.com \
    --cc=omosnace@redhat.com \
    --cc=selinux@vger.kernel.org \
    --cc=serge@hallyn.com \
    --cc=stephen.smalley.work@gmail.com \
    /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