On 3/9/2018 3:51 PM, Ananyev, Konstantin wrote: > > Hi everyone, > >> -----Original Message----- >> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ferruh Yigit >> Sent: Friday, March 9, 2018 3:04 PM >> To: John Daley <johnd...@cisco.com> >> Cc: dev@dpdk.org; Hyong Youb Kim <hyon...@cisco.com> >> Subject: Re: [dpdk-dev] [PATCH v3 03/10] net/enic: heed the requested max Rx >> packet size >> >> On 3/8/2018 2:46 AM, John Daley wrote: >>> From: Hyong Youb Kim <hyon...@cisco.com> >>> >>> Currently, enic completely ignores the requested max Rx packet size >>> (rxmode.max_rx_pkt_len). The desired behavior is that the NIC hardware >>> drops packets larger than the requested size, even though they are >>> still smaller than MTU. >> >> Your description looks reasonable but is there reason of two levels of >> limits, >> max_rx_pkt_len and MTU, why not just use MTU. There is already a mail thread >> to >> clarify max_rx_pkt_len [1]. >> >> Is this work based on an application that uses max_rx_pkt_len and to make PMD >> compatible with that application? If so we can continue with patch, but if >> the >> patch is to implement DPDK properly I suggest postponing this until >> max_rx_pkt_len clarified. >> >> [1] >> https://dpdk.org/ml/archives/dev/2018-March/092178.html > > I think there are quite a lot apps these days that might rely on setting MTU > via > rxmode.max_rx_pkt_len. > I think we need to support them till we (ever) deprecate > rxmode.max_rx_pkt_len.
Right. I was trying to save effort in case something changes related max_rx_pkt_len, but since it is not clear yet, will continue with this patch. > Konstantin > >> >>> >>> Cisco VIC does not have such a feature. But, we can accomplish a >>> similar (not same) effect by reducing the size of posted receive >>> buffers. Packets larger than the posted size get truncated, and the >>> receive handler drops them. This is also how the kernel enic driver >>> enforces the Rx side MTU. >>> >>> This workaround works only when scatter mode is *not* used. When >>> scatter is used, there is currently no way to support >>> rxmode.max_rx_pkt_len, as the NIC always receives packets up to MTU. >>> >>> For posterity, add a copious amount of comments regarding the >>> hardware's drop/receive behavior with respect to max/current MTU. >>> >>> Signed-off-by: Hyong Youb Kim <hyon...@cisco.com> >>> Reviewed-by: John Daley <johnd...@cisco.com> >> >> <...> >
