[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <202310181643.47D7231BF5@keescook>
Date: Wed, 18 Oct 2023 16:46:28 -0700
From: Kees Cook <keescook@...omium.org>
To: Justin Stitt <justinstitt@...gle.com>
Cc: Kalle Valo <kvalo@...nel.org>, linux-wireless@...r.kernel.org,
linux-kernel@...r.kernel.org, linux-hardening@...r.kernel.org
Subject: Re: [PATCH] atmel: replace deprecated strncpy/strcpy with strscpy
On Mon, Oct 16, 2023 at 08:22:45PM +0000, Justin Stitt wrote:
> strncpy() is deprecated for use on NUL-terminated destination strings
> [1] and as such we should prefer more robust and less ambiguous string
> interfaces. strcpy() is also deprecated [2] and shouldn't be used.
Since these are read-only const C strings being copied into a fixed
sized buffer, I would actually prefer to just leave these as-is since
they're already getting verified by FORTIFY_SOURCE.
> We expect priv->firmware_id to be NUL-terminated based on its usage
> with seq_printf() and strlen() in atmel.c:
> 1420 | seq_printf(m, "%s loaded by host\n", priv->firmware_id);
> ...
> 3884 | if (strlen(priv->firmware_id) == 0) {
>
> NUL-padding is not required, which is evident by the usage of a plain
> strcpy():
> 3891 | strcpy(priv->firmware_id, "atmel_at76c502.bin");
>
> Considering the above, a suitable replacement is `strscpy` [2] due to
> the fact that it guarantees NUL-termination on the destination buffer
> without unnecessarily NUL-padding.
>
> Let's also replace hard-coded lengths to be `sizeof(...)` for buffers
> that the compiler can detect the size for as this is less error prone.
>
> Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings [1]
> Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strcpy [2]
> Link: https://manpages.debian.org/testing/linux-manual-4.8/strscpy.9.en.html [3]
> Link: https://github.com/KSPP/linux/issues/90
> Cc: linux-hardening@...r.kernel.org
> Signed-off-by: Justin Stitt <justinstitt@...gle.com>
> ---
> Note: build-tested only.
>
> Found with: $ rg "strncpy\("
> ---
> drivers/net/wireless/atmel/atmel.c | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/net/wireless/atmel/atmel.c b/drivers/net/wireless/atmel/atmel.c
> index 7c2d1c588156..199b9144ef63 100644
> --- a/drivers/net/wireless/atmel/atmel.c
> +++ b/drivers/net/wireless/atmel/atmel.c
> @@ -2011,7 +2011,7 @@ static int atmel_get_name(struct net_device *dev,
> union iwreq_data *wrqu,
> char *extra)
> {
> - strcpy(wrqu->name, "IEEE 802.11-DS");
> + strscpy(wrqu->name, "IEEE 802.11-DS", sizeof(wrqu->name));
> return 0;
> }
>
> @@ -2651,8 +2651,7 @@ static int atmel_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
>
> priv->firmware = new_firmware;
> priv->firmware_length = com.len;
> - strncpy(priv->firmware_id, com.id, 31);
> - priv->firmware_id[31] = '\0';
> + strscpy(priv->firmware_id, com.id, sizeof(priv->firmware_id));
I wonder why this ever used strncpy -- they're both 32 byte buffer
arrays. Regardless, yup, this looks correct to me. Can you respin
without the strcpy() changes?
-Kees
> break;
>
> case ATMELRD:
> @@ -3889,7 +3888,8 @@ static int reset_atmel_card(struct net_device *dev)
> printk(KERN_INFO
> "%s: if not, use the firmware= module parameter.\n",
> dev->name);
> - strcpy(priv->firmware_id, "atmel_at76c502.bin");
> + strscpy(priv->firmware_id, "atmel_at76c502.bin",
> + sizeof(priv->firmware_id));
> }
> err = request_firmware(&fw_entry, priv->firmware_id, priv->sys_dev);
> if (err != 0) {
>
> ---
> base-commit: cbf3a2cb156a2c911d8f38d8247814b4c07f49a2
> change-id: 20231016-strncpy-drivers-net-wireless-atmel-atmel-c-7ca951cf7cfa
>
> Best regards,
> --
> Justin Stitt <justinstitt@...gle.com>
>
--
Kees Cook
Powered by blists - more mailing lists