On Thu, Nov 27, 2025 at 11:41 AM Jyotsna Dhumale <[email protected]>
wrote:

> Hi Jay,
> Thanks a lot for your reply and support in understanding and addressing
> the issue.
>
> We are running *VPP + DPDK IAVF*.
> Attached the *startup.conf *and *setup.txt* we are using while running
> vpp.
> Below is the host *VF *details. Let me know incase you need any other
> information.
>
> *0000:01:01.0 *Ethernet controller [0200]: Intel Corporation Ethernet
> Adaptive Virtual Function [8086:1889] (rev 02)
>         Subsystem: Super Micro Computer Inc Device [15d9:0000]
>         Kernel driver in use: vfio-pci
>         Kernel modules: iavf
>
> Thanks and regards,
> Jyotsna
>
> On Wed, Nov 26, 2025 at 8:00 PM Jay Wang via lists.fd.io <jay.wang2=
> [email protected]> wrote:
>
>> Hi Jyotsna,
>>
>>
>>
>> Just to confirm --- are you running VPP + DPDK IAVF, or VPP + the AVF
>> plugin with the quick workaround?
>>
>> I’m asking because the patch is only applicable to VPP + DPDK IAVF.
>>
>> If you are using that setup, could you please share your startup.conf and
>> the VPP CLI commands you used to configure VPP?
>>
>> That will help me look into the issue in more detail.
>>
>>
>>
>> Cheers,
>>
>> Jay Wang
>>
>> *From:* [email protected] <[email protected]> *On Behalf Of *Jyotsna
>> Dhumale via lists.fd.io
>> *Sent:* Tuesday, November 25, 2025 10:21 AM
>> *To:* [email protected]
>> *Cc:* [email protected]; Paul Szczepanek <[email protected]>;
>> Vinod Krishna <[email protected]>; Dhruv Tripathi <
>> [email protected]>; Wathsala Vithanage <[email protected]>
>> *Subject:* Re: [vpp-dev] [VPP 25.02 | ARM64 | E810-C | IAVF] Packets
>> showing length 65535 in DPDK input node (works fine on VPP 22.02)
>>
>>
>>
>> Hi Jay Wang,
>> Thanks a lot for the patch. Even I am facing the same problem[pkt_len
>> 65535] with *VPP25.02* and *ARM64*
>>
>> *Board : Ampere(R) Altra(R) Processor Q64-30 CPU @ 3.0GHz NIC : E810-C 4
>> ports*
>> I tested with the patch you provided, but I am still running into same
>> problem of pkt_len coming as 65535.
>> Attached the build log for your reference, which  shows patch is applied
>> correctly and build is successful.
>> Later VPP is built using the new vpp-ext-dep package generated with the
>> patch.
>>
>> Anything that I am missing or missing in the patch?
>>
>> Let me know if you need any other information from my side.
>>
>> Thanks and regards,
>>
>> Jyotsna
>>
>>
>>
>> On Thu, Nov 20, 2025 at 3:15 AM Jay Wang via lists.fd.io <jay.wang2=
>> [email protected]> wrote:
>>
>> Hi Nousi,
>>
>>
>>
>> I can confirm that I was able to reproduce your issue in my local test
>> environment(*Ampere Altra Max 96 Cores* with *the Intel E810-C 2x ports
>> NIC*).
>>
>> The issue also persists in the latest VPP codebase, so I spent some time
>> investigating it.
>>
>> From what I observed, commit[1] appears to be the root cause. This commit
>> introduces NEON intrinsics to add support for the IAVF Rx vector path on
>> Arm platforms in DPDK.
>>
>> I don’t want to bother you with the implementation details of the Rx
>> vector function[2]. But in short, the issue occurs because the function
>> writes the *pkt_len* field to the wrong location in the *rte_mbuf* when
>> copying from the hardware descriptor.
>>
>> I’ve created a quick workaround for VPP 25.02(see attached patch) and
>> tested it with 64B/1024B packets. It seems to work fine. Please give it a
>> try and let me know if it resolves the issue on your side.
>>
>>
>>
>> VPP 25.02 works fine on X86 systems because it uses a different Rx vector
>> code path.
>>
>> Interestingly, I observed a similar packet length issue when testing the
>> X86 counterpart of Rx vector code path[3].
>>
>>
>>
>> Thanks,
>>
>> Jay Wang
>>
>>
>>
>> [1] https://gerrit.fd.io/r/c/vpp/+/37017
>>
>> [2]
>> https://github.com/DPDK/dpdk/blob/main/drivers/net/intel/iavf/iavf_rxtx_vec_neon.c#L144
>>
>> [3]
>> https://github.com/DPDK/dpdk/blob/main/drivers/net/intel/iavf/iavf_rxtx_vec_sse.c#L391
>>
>> *From:* [email protected] <[email protected]> *On Behalf Of *nousi
>> via lists.fd.io
>> *Sent:* Tuesday, November 11, 2025 2:22 PM
>> *To:* [email protected]
>> *Subject:* [vpp-dev] [VPP 25.02 | ARM64 | E810-C | IAVF] Packets showing
>> length 65535 in DPDK input node (works fine on VPP 22.02)
>>
>>
>>
>> Hi everyone,
>>
>> I’m seeing a packet length issue after upgrading to *VPP 25.02* on *ARM64
>> (Ampere® Altra® Q64-30 @ 3.0GHz)* with an *Intel E810-C (4x ports)* NIC
>> using the *IAVF driver*.
>>
>> All packets received in the *dpdk-input node* show a *length of 65535
>> bytes*, which is clearly incorrect. This is visible in both *trace* and *
>> pcap* captures.
>>
>> Here’s what I’ve observed so far:
>>
>>    - The issue only appears with *VPP 25.02* on *ARM64*.
>>    - The *same platform (hardware + NIC)* works perfectly with *VPP
>>    22.02* — packet lengths are reported correctly.
>>    - On *X86 *systems, *VPP 25.02* also works fine (no packet length
>>    issue).
>>    - So it seems specific to *VPP 25.02 + ARM64 + IAVF driver*
>>    combination.
>>
>> *Setup details:*
>>
>>    - *CPU:* Ampere® Altra® Q64-30 (ARM64)
>>    - *NIC:* Intel E810-C (4 ports)
>>    - *Driver:* IAVF (DPDK)
>>    - *VPP version:* 25.02
>>    - *Symptom:* All packets show length = 65535 in dpdk-input node
>>
>> Has anyone else seen this behavior or is aware of any related changes in *
>> VPP 25.02* or the * IAVF/DPDK layer* for ARM64?
>> Would appreciate any pointers, patches, or debug hints — especially if
>> someone has tested E810/IAVF on ARM with newer VPP versions.
>>
>> Thanks a lot in advance!
>>
>> *Nousi.*
>>
>> IMPORTANT NOTICE: The contents of this email and any attachments are
>> confidential and may also be privileged. If you are not the intended
>> recipient, please notify the sender immediately and do not disclose the
>> contents to any other person, use it for any purpose, or store or copy the
>> information in any medium. Thank you.
>>
>>
>>
>> IMPORTANT NOTICE: The contents of this email and any attachments are
>> confidential and may also be privileged. If you are not the intended
>> recipient, please notify the sender immediately and do not disclose the
>> contents to any other person, use it for any purpose, or store or copy the
>> information in any medium. Thank you.
>>
>> 
>>
>>
set interface state vpp-intf1/1/0 up
set interface reassembly vpp-intf1/1/0 on
create host-interface name eth-bh0
set interface state host-eth-bh0 up 
set interface mtu packet 1500 host-eth-bh0
create bridge-domain 12345
set interface l2 bridge host-eth-bh0 12345

set interface l2 bridge vpp-intf1/1/0 12345
set interface mtu packet 1500 vpp-intf1/1/0

create loopback interface mac d4:ac:bd:b0:fe:02 instance 12345
ip route add 0.0.0.0/0 via 10.200.160.1 loop12345
set interface ip address loop12345 10.200.160.10/24
set interface ip address loop12345 10.200.160.20/24

session enable
app ns add id vcl secret 614235 if loop12345

set interface l2 bridge loop12345 12345 bvi
set interface mtu packet 1500 loop12345
set interface state loop12345 up




Attachment: startup.conf
Description: Binary data

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#26561): https://lists.fd.io/g/vpp-dev/message/26561
Mute This Topic: https://lists.fd.io/mt/116238372/21656
Group Owner: [email protected]
Unsubscribe: https://lists.fd.io/g/vpp-dev/leave/14379924/21656/631435203/xyzzy 
[[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

Reply via email to