On Fri, 6 Mar 2026 10:20:13 +0100 Xavier Guillaume <[email protected]> wrote:
> Derive max_rx_pktlen and max_mtu from the actual TPACKET ring > frame size rather than the static RTE_ETHER_MAX_LEN (1518). > This allows jumbo frame support when the user specifies a > larger framesz devarg at vdev creation time, for example: > > --vdev=net_af_packet0,iface=eth0,framesz=9216 > > The advertised capabilities now accurately reflect what the > configured ring can handle, avoiding both false limitations > with large frames and false promises with small ones. > > Signed-off-by: Xavier Guillaume <[email protected]> > --- AI had this observation. Patch 3: max_mtu subtracts RTE_ETHER_CRC_LEN, but the kernel strips CRC before placing data in the TPACKET ring, so MTU is understated by 4 bytes. This is conservative and safe, and matches what af_xdp does, so probably fine as-is.
