Re: [PATCH net-next] net: ibm: emac: Reserve VLAN header in MJS limit

[Jacob Keller <jacob.e.keller@intel.com>]

On 6/1/2026 6:35 AM, Paolo Abeni wrote:
> On 5/30/26 1:18 AM, Jacob Keller wrote:
>> On 5/26/2026 1:22 PM, Rosen Penev wrote:
>>> The IBM EMAC programs its Maximum Jumbo Size (MJS) drop
>>> threshold from ndev->mtu directly. The hardware sizes the threshold
>>> against the L2 frame minus the ethernet header, but does not
>>> discount the 802.1Q tag, so a frame carrying a VLAN tag and a full
>>> 1500-byte payload exceeds MJS by exactly 4 bytes and is dropped.
>>>
>>> This is normally hidden because JPSM (and therefore the MJS check)
>>> only engages when the MTU is raised above ETH_DATA_LEN.  With the
>>> qca8k DSA tagger the conduit MTU is bumped by QCA_HDR_LEN to 1502
>>> during dsa_conduit_setup(), which is enough to enable JPSM and
>>> expose the off-by-VLAN-tag in the limit.
>>>
>>> Pad MJS by VLAN_HLEN so a VLAN-tagged full-MTU frame passes.
>>>
>>> Reported on Meraki MX60 (qca8k switch): tagged VLAN
>>> traffic drops at 1500-byte payload, while 1496 bytes works
>>> and untagged 1500 bytes works.
>>>
>>> Assisted-by: Claude:Opus-4.7
>>> Signed-off-by: Rosen Penev <rosenp@gmail.com>
>>
>> If this results in dropped packets, that seems like a user visible bug..
>> Should this have a Fixes tag and target the net-next tree? I guess its
>> not a "regression" since the driver has always behaved this way, but it
>> still seems like a bug to me.
> 
> Given it never worked and we are trying to keep the amount of changes
> into net under control, I tend to consider this suitable for net-next
> i.e. enable DSA support.
> 
> /P
> 

Makes sense.

Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>