Re: [Intel-wired-lan] [PATCH] ethernet: ice: avoid gcc-9 integer overflow warning

[Alexander Lobakin <alexandr.lobakin@intel.com>]

From: Arnd Bergmann <arnd@kernel.org>
Date: Tue, 14 Feb 2023 14:19:49 +0100

> From: Arnd Bergmann <arnd@arndb.de>
> 
> With older compilers like gcc-9, the calculation of the vlan
> priority field causes a warning from the byteswap:
> 
> In file included from drivers/net/ethernet/intel/ice/ice_tc_lib.c:4:
> drivers/net/ethernet/intel/ice/ice_tc_lib.c: In function 'ice_parse_cls_flower':
> include/uapi/linux/swab.h:15:15: error: integer overflow in expression '(int)(short unsigned int)((int)match.key-><U67c8>.<U6698>.vlan_priority << 13) & 57344 & 255' of type 'int' results in '0' [-Werror=overflow]
>    15 |  (((__u16)(x) & (__u16)0x00ffU) << 8) |   \
>       |   ~~~~~~~~~~~~^~~~~~~~~~~~~~~~~
> include/uapi/linux/swab.h:106:2: note: in expansion of macro '___constant_swab16'
>   106 |  ___constant_swab16(x) :   \
>       |  ^~~~~~~~~~~~~~~~~~
> include/uapi/linux/byteorder/little_endian.h:42:43: note: in expansion of macro '__swab16'
>    42 | #define __cpu_to_be16(x) ((__force __be16)__swab16((x)))
>       |                                           ^~~~~~~~
> include/linux/byteorder/generic.h:96:21: note: in expansion of macro '__cpu_to_be16'
>    96 | #define cpu_to_be16 __cpu_to_be16
>       |                     ^~~~~~~~~~~~~
> drivers/net/ethernet/intel/ice/ice_tc_lib.c:1458:5: note: in expansion of macro 'cpu_to_be16'
>  1458 |     cpu_to_be16((match.key->vlan_priority <<
>       |     ^~~~~~~~~~~
> 
> The code looks correct to me, so just avoid the warning by replacing
> the macro expansion with an intermediate variable.
> 
> Fixes: 34800178b302 ("ice: Add support for VLAN priority filters in switchdev")

Should it have "Fixes:" tag if the code itself is correct?

> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>  drivers/net/ethernet/intel/ice/ice_tc_lib.c | 10 ++++------
>  1 file changed, 4 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/net/ethernet/intel/ice/ice_tc_lib.c b/drivers/net/ethernet/intel/ice/ice_tc_lib.c
> index 6b48cbc049c6..e9932446185c 100644
> --- a/drivers/net/ethernet/intel/ice/ice_tc_lib.c
> +++ b/drivers/net/ethernet/intel/ice/ice_tc_lib.c
> @@ -1453,10 +1453,9 @@ ice_parse_cls_flower(struct net_device *filter_dev, struct ice_vsi *vsi,
>  		}
>  
>  		if (match.mask->vlan_priority) {
> +			u16 prio = (match.key->vlan_priority << VLAN_PRIO_SHIFT) & VLAN_PRIO_MASK;

Maybe just u32?

Also, I'd go for FIELD_PREP() here. Even be16_encode_bits(). Or does it
have the same problem on GCC 9?

			__be16 pri;

			pri = be16_encode_bits(match.key->vlan_priority,
					       VLAN_PRIO_MASK);
			headers->vlan_hdr.vlan_prio = pri;

But I suspect you (or somebody else) will say "please don't mix fixes
and improvements" :D Altho I'd go for this one if it silences the warning.

>  			fltr->flags |= ICE_TC_FLWR_FIELD_VLAN_PRIO;
> -			headers->vlan_hdr.vlan_prio =
> -				cpu_to_be16((match.key->vlan_priority <<
> -					     VLAN_PRIO_SHIFT) & VLAN_PRIO_MASK);
> +			headers->vlan_hdr.vlan_prio = cpu_to_be16(prio);
>  		}
>  
>  		if (match.mask->vlan_tpid)
> @@ -1487,10 +1486,9 @@ ice_parse_cls_flower(struct net_device *filter_dev, struct ice_vsi *vsi,
>  		}
>  
>  		if (match.mask->vlan_priority) {
> +			u16 prio = (match.key->vlan_priority << VLAN_PRIO_SHIFT) & VLAN_PRIO_MASK;
>  			fltr->flags |= ICE_TC_FLWR_FIELD_CVLAN_PRIO;
> -			headers->cvlan_hdr.vlan_prio =
> -				cpu_to_be16((match.key->vlan_priority <<
> -					     VLAN_PRIO_SHIFT) & VLAN_PRIO_MASK);
> +			headers->cvlan_hdr.vlan_prio = cpu_to_be16(prio);

^

>  		}
>  	}
> 

Thanks,
Olek