Re: [PATCH v4 net-next 02/13] gro: flushing when CWR is set negatively affects AccECN

On 10/13/25 7:03 PM, chia-yu.chang@nokia-bell-labs.com wrote:

From: Ilpo Järvinen ij@kernel.org

As AccECN may keep CWR bit asserted due to different interpretation of the bit, flushing with GRO because of CWR may effectively disable GRO until AccECN counter field changes such that CWR-bit becomes 0.

There is no harm done from not immediately forwarding the CWR'ed segment with RFC3168 ECN.

I guess this change could introduce additional latency for RFC3168 notification, which sounds not good. On the flip side adding too much AccECN logic to GRO (i.e. to allow aggregation only for AccECN enabled flows) looks overkill.

@Eric: WDYT?

Thanks,

Paolo