Received-SPF: pass (google.com: domain of linux-kernel+bounces-172169-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Feedback-ID: i56a14606:Fastmail
User-Agent: Cyrus-JMAP/3.11.0-alpha0-443-g0dc955c2a-fm-20240507.001-g0dc955c2
Precedence: bulk
MIME-Version: 1.0
Message-Id: <29b33c18-f123-4656-8507-406c87a12ec8@app.fastmail.com>
In-Reply-To: <20240507201238.213396-2-thorsten.blum@toblux.com>
References: <20240507201238.213396-2-thorsten.blum@toblux.com>
Date: Tue, 07 May 2024 22:48:26 +0200
From: "Arnd Bergmann" <arnd@arndb.de>
To: "Thorsten Blum" <thorsten.blum@toblux.com>,
 "Srinivas Kandagatla" <srinivas.kandagatla@linaro.org>,
 "Amol Maheshwari" <amahesh@qti.qualcomm.com>,
 "Greg Kroah-Hartman" <gregkh@linuxfoundation.org>
Cc: linux-arm-msm@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] misc: fastrpc: Use memdup_user()
Content-Type: text/plain

On Tue, May 7, 2024, at 22:12, Thorsten Blum wrote:
> Switching to memdup_user() overwrites the allocated memory only once,
> whereas kzalloc() followed by copy_from_user() initializes the allocated
> memory to zero and then immediately overwrites it.
>
> Fixes the following Coccinelle/coccicheck warning reported by
> memdup_user.cocci:
>
> 	WARNING opportunity for memdup_user
>
> Signed-off-by: Thorsten Blum <thorsten.blum@toblux.com>

The patch looks correct to me.

> ---
>  drivers/misc/fastrpc.c | 11 +++--------
>  1 file changed, 3 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/misc/fastrpc.c b/drivers/misc/fastrpc.c
> index 4c67e2c5a82e..2857cddaf812 100644
> --- a/drivers/misc/fastrpc.c
> +++ b/drivers/misc/fastrpc.c
> @@ -1259,17 +1259,12 @@ static int 
> fastrpc_init_create_static_process(struct fastrpc_user *fl,
>  		goto err;
>  	}
> 
> -	name = kzalloc(init.namelen, GFP_KERNEL);
> -	if (!name) {
> -		err = -ENOMEM;
> +	name = memdup_user((void __user *)(uintptr_t)init.name, init.namelen);
> +	if (IS_ERR(name)) {
> +		err = PTR_ERR(name);
>  		goto err;
>  	}

There is also a chance to simplify this further using u64_to_user_ptr()
instead of the double cast if you want.

Acked-by: Arnd Bergmann <arnd@arndb.de>

      Arnd