* Re: [PATCH 2/3] [v2] staging: rtl8723bs: convert strncpy to strscpy
2024-04-08 19:48 ` [PATCH 2/3] [v2] staging: rtl8723bs: convert strncpy to strscpy Arnd Bergmann
@ 2024-04-08 20:35 ` Justin Stitt
2024-04-09 7:10 ` Dan Carpenter
1 sibling, 0 replies; 3+ messages in thread
From: Justin Stitt @ 2024-04-08 20:35 UTC (permalink / raw)
To: Arnd Bergmann
Cc: Greg Kroah-Hartman, Nathan Chancellor, Dan Carpenter,
Arnd Bergmann, Nick Desaulniers, Bill Wendling, Franziska Naepelt,
Johannes Berg, Jeff Johnson, Aloka Dixit, Erick Archer,
Yang Yingliang, linux-staging, linux-kernel, llvm
Hi,
On Mon, Apr 08, 2024 at 09:48:10PM +0200, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
>
> gcc-9 complains about a possibly unterminated string in the strncpy() destination:
>
> In function 'rtw_cfg80211_add_monitor_if',
> inlined from 'cfg80211_rtw_add_virtual_intf' at drivers/staging/rtl8723bs/os_dep/ioctl_cfg80211.c:2209:9:
> drivers/staging/rtl8723bs/os_dep/ioctl_cfg80211.c:2146:2: error: 'strncpy' specified bound 16 equals destination size [-Werror=stringop-truncation]
> 2146 | strncpy(mon_ndev->name, name, IFNAMSIZ);
> | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> This one is a false-positive because of the explicit termination in the following
> line, and recent versions of clang and gcc no longer warn about this.
>
> Interestingly, the other strncpy() in this file is missing a termination but
> does not produce a warning, possibly because of the type confusion and the
> cast between u8 and char.
>
> Change both strncpy() instances to strscpy(), which avoids the warning as well
> as the possibly missing termination. No additional padding is needed here.
>
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Helps with: https://github.com/KSPP/linux/issues/90
and for the bots...
Link: https://github.com/KSPP/linux/issues/90
Reviewed-by: Justin Stitt <justinstitt@google.com>
> ---
> v2:
> use the two-argument version of strscpy(), which is simpler for the constant
> size destination. Add the third instance in this driver as well.
> ---
> drivers/staging/rtl8723bs/os_dep/ioctl_cfg80211.c | 5 ++---
> drivers/staging/rtl8723bs/os_dep/os_intfs.c | 2 +-
> 2 files changed, 3 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/staging/rtl8723bs/os_dep/ioctl_cfg80211.c b/drivers/staging/rtl8723bs/os_dep/ioctl_cfg80211.c
> index 65a450fcdce7..3fe27ee75b47 100644
> --- a/drivers/staging/rtl8723bs/os_dep/ioctl_cfg80211.c
> +++ b/drivers/staging/rtl8723bs/os_dep/ioctl_cfg80211.c
> @@ -884,7 +884,7 @@ static int cfg80211_rtw_add_key(struct wiphy *wiphy, struct net_device *ndev,
> goto addkey_end;
> }
>
> - strncpy((char *)param->u.crypt.alg, alg_name, IEEE_CRYPT_ALG_NAME_LEN);
> + strscpy(param->u.crypt.alg, alg_name);
>
> if (!mac_addr || is_broadcast_ether_addr(mac_addr))
> param->u.crypt.set_tx = 0; /* for wpa/wpa2 group key */
> @@ -2143,8 +2143,7 @@ static int rtw_cfg80211_add_monitor_if(struct adapter *padapter, char *name, str
> }
>
> mon_ndev->type = ARPHRD_IEEE80211_RADIOTAP;
> - strncpy(mon_ndev->name, name, IFNAMSIZ);
> - mon_ndev->name[IFNAMSIZ - 1] = 0;
> + strscpy(mon_ndev->name, name);
> mon_ndev->needs_free_netdev = true;
> mon_ndev->priv_destructor = rtw_ndev_destructor;
>
> diff --git a/drivers/staging/rtl8723bs/os_dep/os_intfs.c b/drivers/staging/rtl8723bs/os_dep/os_intfs.c
> index 68bba3c0e757..55d0140cd543 100644
> --- a/drivers/staging/rtl8723bs/os_dep/os_intfs.c
> +++ b/drivers/staging/rtl8723bs/os_dep/os_intfs.c
> @@ -415,7 +415,7 @@ static int rtw_ndev_init(struct net_device *dev)
> struct adapter *adapter = rtw_netdev_priv(dev);
>
> netdev_dbg(dev, FUNC_ADPT_FMT "\n", FUNC_ADPT_ARG(adapter));
> - strncpy(adapter->old_ifname, dev->name, IFNAMSIZ);
> + strscpy(adapter->old_ifname, dev->name);
>
> return 0;
> }
> --
> 2.39.2
>
>
Thanks
Justin
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH 2/3] [v2] staging: rtl8723bs: convert strncpy to strscpy
2024-04-08 19:48 ` [PATCH 2/3] [v2] staging: rtl8723bs: convert strncpy to strscpy Arnd Bergmann
2024-04-08 20:35 ` Justin Stitt
@ 2024-04-09 7:10 ` Dan Carpenter
1 sibling, 0 replies; 3+ messages in thread
From: Dan Carpenter @ 2024-04-09 7:10 UTC (permalink / raw)
To: Arnd Bergmann
Cc: Greg Kroah-Hartman, Nathan Chancellor, Justin Stitt,
Arnd Bergmann, Nick Desaulniers, Bill Wendling, Franziska Naepelt,
Johannes Berg, Jeff Johnson, Aloka Dixit, Erick Archer,
Yang Yingliang, linux-staging, linux-kernel, llvm
On Mon, Apr 08, 2024 at 09:48:10PM +0200, Arnd Bergmann wrote:
> From: Arnd Bergmann <arnd@arndb.de>
>
> gcc-9 complains about a possibly unterminated string in the strncpy() destination:
>
> In function 'rtw_cfg80211_add_monitor_if',
> inlined from 'cfg80211_rtw_add_virtual_intf' at drivers/staging/rtl8723bs/os_dep/ioctl_cfg80211.c:2209:9:
> drivers/staging/rtl8723bs/os_dep/ioctl_cfg80211.c:2146:2: error: 'strncpy' specified bound 16 equals destination size [-Werror=stringop-truncation]
> 2146 | strncpy(mon_ndev->name, name, IFNAMSIZ);
> | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>
> This one is a false-positive because of the explicit termination in the following
> line, and recent versions of clang and gcc no longer warn about this.
>
> Interestingly, the other strncpy() in this file is missing a termination but
> does not produce a warning, possibly because of the type confusion and the
> cast between u8 and char.
>
> Change both strncpy() instances to strscpy(), which avoids the warning as well
> as the possibly missing termination. No additional padding is needed here.
>
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
> v2:
> use the two-argument version of strscpy(), which is simpler for the constant
> size destination. Add the third instance in this driver as well.
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
regards,
dan carpenter
^ permalink raw reply [flat|nested] 3+ messages in thread