On 11/16/26 17:41, Stephen Hemminger wrote:
On Wed, 11 Nov 2026 10:45:36 +0000
Frank Dressler <[email protected]> wrote:

This fix ensures that the queues of an AF_PACKET vdev do not
accidentally join an already existing fanout group in the same netns.
Joining such a group would make each member receive only a fraction of
the packets.

It sets the PACKET_FANOUT_FLAG_UNIQUEID flag for the first queue which
asks the kernel to allocate a new, unique fanout group ID, then asks the
kernel for that queue's fanout configuration which includes this
generated ID but not the flag anymore (the flag is ephemeral) and then
uses it for subsequent queues to make them join the same fanout group.

The bug exists since the PMD's introduction in 2014:

   # 2014 DPDK:
   $ git show 364e08f2bb:lib/librte_pmd_af_packet/rte_eth_af_packet.c |
       grep getpid
     fanout_arg = (getpid() ^ (*internals)->if_index) & 0xffff;

   # v26.07:
   $ git show v26.07:drivers/net/af_packet/rte_eth_af_packet.c |
       grep getpid
     return (getpid() ^ if_index) & 0xffff;

The chance for a collision might look low initially as it requires PIDs
to be equal mod 65536. But the chance is high for applications running
in containers having individual pidns but a shared netns, which can be
the case in a k8s pod, or also with manually started containers, e.g.
`--network ns:<path>` in Podman. The individual pidns increase the
chance of different processes having the same PID.

To quickly reproduce the issue, you can start multiple AF_PACKET DPDK
applications with qpairs greater than 1, prefixing them with
`unshare --fork -p` to have each one get the same PID as if they were
executed in a container. Each one then only receives a fraction of
the traffic.

The used UNIQUEID fanout flag is available since Linux v4.12, while DPDK
currently requires Linux v5.4 or newer. The change is also compatible
with RHEL 8.

Fixes: 364e08f2bbc0 ("af_packet: add PMD for AF_PACKET-based virtual devices")
Cc: [email protected]

Signed-off-by: Frank Dressler <[email protected]>
---

Reviewed-by: Stephen Hemminger <[email protected]>

The commit message is way to long though. Will chop it during merge.

Thank you.

You need to tell you AI to be concise. I have started using multiple
prompts to tell it to not repeat.


I did try let AI generate the commit message but it was too short and did not give me the impression that a reader without my context could easily understand and review the issue. So I wrote my own message.

This is not the first time people think I used AI when I did not. It is always very demotivating. https://github.com/secdev/scapy/issues/4964#issuecomment-6124262531 But I understand when people think things are AI even when they are not given the amount of AI generated content.

I'll try writing more concise messages in the future.

The kernel review prompt includes /kslop skill which has:
https://github.com/masoncl/review-prompts/blob/main/kernel/slop-indicators.md


It would feel weird applying AI-detection stuff to hand-written commits just to make sure people relying on it do not get confused by such broken tools that lead to miscategorizations.

AI even flagged that:


I started to run the AI patch review locally a while ago after noticing its horrible quality to prevent sending unnecessary patch versions. When I run it with just `./devtools/ai/review-patch.py <patch>` it says

Error: API error: {'type': 'not_found_error', 'message': 'model: claude-sonnet-4-5-20250929'}

So I switched it to claude-sonnet-4-6 and that just outputs some very stupid comments that read like "I found a bug. ... Oh, no, it is not a bug. Actually, re-reading the logic carefully ... This is correct. There is one real issue: ... This is correct. No correctness errors found ... Warning: ... No issue. ... No Errors or Warnings found." With claude-opus-5-5 it becomes sane or broken because it just says "No issues found. Review-Result: CLEAN"


Logic is correct. The kernel clears PACKET_FANOUT_FLAG_UNIQUEID from
the stored group flags after allocating the id, and PACKET_FANOUT
getsockopt returns id | type << 16 | flags << 24, which is exactly
the encoding fanout_add() needs for the later queues to pass the
match->flags/type checks and join. DEFRAG, ROLLOVER and
IGNORE_OUTGOING survive the round trip unchanged.

Info

The commit message is longer than it needs to be. The two git show
transcripts repeat what the Fixes: tag already says; dropping them
and the "This fix ensures" opener would leave something like:

   The fanout group id was derived from (getpid() ^ ifindex), so
   processes in separate pid namespaces sharing a netns (e.g. a k8s
   pod) can pick the same id. Their queues then join one fanout group
   and each process sees only part of the traffic.

   Use PACKET_FANOUT_FLAG_UNIQUEID on the first queue so the kernel
   allocates an unused id, then read it back with getsockopt and use
   it for the remaining queues.

   PACKET_FANOUT_FLAG_UNIQUEID is available since Linux 4.12.

Worth one sentence: this also fixes two af_packet vdevs on the same
interface in one process, which previously shared a group because
pid and ifindex were identical.


How do I reproduce this output? Is it created with the review-patch.py tool?

Reply via email to