linux-kernel - Re: [Intel-wired-lan] [PATCH iwl-next 4/8] ice: allow overriding lan_en, lb

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <ef41487b-ecfb-4b25-83a2-c97d6aa3c813@intel.com>
Date: Mon, 24 Nov 2025 11:20:11 +0100
From: Jakub Slepecki <jakub.slepecki@...el.com>
To: "Loktionov, Aleksandr" <aleksandr.loktionov@...el.com>,
	"intel-wired-lan@...ts.osuosl.org" <intel-wired-lan@...ts.osuosl.org>
CC: "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"netdev@...r.kernel.org" <netdev@...r.kernel.org>, "Kitszel, Przemyslaw"
	<przemyslaw.kitszel@...el.com>, "Nguyen, Anthony L"
	<anthony.l.nguyen@...el.com>, "michal.swiatkowski@...ux.intel.com"
	<michal.swiatkowski@...ux.intel.com>
Subject: Re: [Intel-wired-lan] [PATCH iwl-next 4/8] ice: allow overriding
 lan_en, lb_en in switch

On 2025-11-21 10:21, Loktionov, Aleksandr wrote:
>> -----Original Message-----
>> diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c
>> b/drivers/net/ethernet/intel/ice/ice_switch.c
>> index 04e5d653efce..7b63588948fd 100644
>> --- a/drivers/net/ethernet/intel/ice/ice_switch.c
>> +++ b/drivers/net/ethernet/intel/ice/ice_switch.c
>> @@ -2538,8 +2538,9 @@ int ice_get_initial_sw_cfg(struct ice_hw *hw)
>>    */
>>   static void ice_fill_sw_info(struct ice_hw *hw, struct ice_fltr_info
>> *fi)  {
>> -	fi->lb_en = false;
>> -	fi->lan_en = false;
>> +	bool lan_en = false;
>> +	bool lb_en = false;
>> +
>>   	if ((fi->flag & ICE_FLTR_TX) &&
>>   	    (fi->fltr_act == ICE_FWD_TO_VSI ||
>>   	     fi->fltr_act == ICE_FWD_TO_VSI_LIST || @@ -2549,7 +2550,7
>> @@ static void ice_fill_sw_info(struct ice_hw *hw, struct
>> ice_fltr_info *fi)
>>   		 * packets to the internal switch that will be dropped.
>>   		 */
>>   		if (fi->lkup_type != ICE_SW_LKUP_VLAN)
>> -			fi->lb_en = true;
>> +			lb_en = true;
>>
>>   		/* Set lan_en to TRUE if
>>   		 * 1. The switch is a VEB AND
>> @@ -2578,14 +2579,18 @@ static void ice_fill_sw_info(struct ice_hw
>> *hw, struct ice_fltr_info *fi)
>>   			     !is_unicast_ether_addr(fi-
>>> l_data.mac.mac_addr)) ||
>>   			    (fi->lkup_type == ICE_SW_LKUP_MAC_VLAN &&
>>   			     !is_unicast_ether_addr(fi-
>>> l_data.mac.mac_addr)))
>> -				fi->lan_en = true;
>> +				lan_en = true;
>>   		} else {
>> -			fi->lan_en = true;
>> +			lan_en = true;
>>   		}
>>   	}
>>
>>   	if (fi->flag & ICE_FLTR_TX_ONLY)
>> -		fi->lan_en = false;
>> +		lan_en = false;
>> +	if (!(fi->lb_en & ICE_FLTR_INFO_LB_LAN_FORCE_MASK))
>> +		fi->lb_en = lb_en;
>> +	if (!(fi->lan_en & ICE_FLTR_INFO_LB_LAN_FORCE_MASK))
>> +		fi->lan_en = lan_en;
> For me it looks strange.
> What type the fi->lb_en has?
> fi->lb_en declared as bool, and you assign fi->lan_en from bool.
> But you check condition by fi->lan_en & ICE_FLTR_INFO_LB_LAN_FORCE_MASK ?

I agree this can look strange.  lb_en and lan_en are both u8 in
ice_switch.h:/^struct ice_fltr_info/ and we assign them from bool.
Before, even though we had the same implicit conversion bool -> u8 we
did not use either of u8s to hold anything else.

> It rases questions if fi->lan_en a bool why use fi->lan_en & ICE_FLTR_INFO_LB_LAN_FORCE_MASK then?
> And if fi->lan_en is a bitmask why not use FIELD_GET(ICE_FLTR_INFO_LB_LAN_FORCE_MASK, fi->lan_en) and
> why not something like:
> 
> if (!FIELD_GET(ICE_FLTR_INFO_LB_LAN_FORCE_MASK, fi->lan_en))
>      FIELD_MODIFY(ICE_FLTR_INFO_LB_LAN_VALUE_MASK, &fi->lan_en, lan_en);
> 
> It could preserve unrelated bits (like FORCE) and make the code resilient to future changes in bit positions?

The latter.  Original intention, one of, was to avoid implying this
can be extended, because it should not: for better customization we
have "advanced" rules, and "simple" rules shouldn't try to chase them.
Instead, porting everything to "advanced" rules would be more reasonable.
I make an exception here, because cost of any other option is way higher.

That being said, I don't see any reason to not use
FIELD_{GET,PREP,MODIFY}.  I will modify this accordingly across the
series.

Thanks!

>>   }
>>
>>   /**