From: "Matěj Grégr" <mgregr@netx.as>
To: intel-wired-lan@lists.osuosl.org
Cc: Peter Nagy <nagy@netx.as>
Subject: [Intel-wired-lan] ice: lldp packets dropped after changing number of channels
Date: Wed, 3 Apr 2024 18:55:20 +0200 [thread overview]
Message-ID: <1be45a76-90af-4813-824f-8398b69745a9@netx.as> (raw)
Hello,
we are facing an issue with Intel 100G E810-C card. If we change
number of channels using ethtool -L, LLDP packets stop to be forwarded
to the network stack and user space.
Firmware-based lldp agent is disabled:
# ethtool --show-priv-flags hge11
Private flags for hge11:
link-down-on-close : off
fw-lldp-agent : off
vf-true-promisc-support: off
mdd-auto-reset-vf : off
vf-vlan-pruning : off
legacy-rx : off
[ 363.664628] ice 0000:41:00.0: The DDP package was successfully
loaded: ICE COMMS Package version 1.3.45.0
[ 363.983822] ice 0000:41:00.0: 252.048 Gb/s available PCIe bandwidth
(16.0 GT/s PCIe x16 link)
[ 364.003040] ice 0000:41:00.0: PTP init successful
[ 364.202873] ice 0000:41:00.0: DCB is enabled in the hardware, max
number of TCs supported on this port are 8
[ 364.202880] ice 0000:41:00.0: FW LLDP is disabled, DCBx/LLDP in SW mode.
[ 364.203045] ice 0000:41:00.0: Commit DCB Configuration to the hardware
The channels are set to the default values:
ethtool -l hge11
Channel parameters for hge11:
Pre-set maximums:
RX: 64
TX: 64
Other: n/a
Combined: 64
Current hardware settings:
RX: 0
TX: 0
Other: n/a
Combined: 64
We are able to see the LLDP frames:
tcpdump -i hge11 -nn 'ether proto 0x88cc'
16:21:08.528229 4c:ae:a3:6a:80:ee > 01:80:c2:00:00:0e, ethertype LLDP
(0x88cc), length 347: LLDP, length 333: test-hp5945
16:21:09.528501 4c:ae:a3:6a:80:ee > 01:80:c2:00:00:0e, ethertype LLDP
(0x88cc), length 347: LLDP, length 333: test-hp5945
16:21:10.528325 4c:ae:a3:6a:80:ee > 01:80:c2:00:00:0e, ethertype LLDP
(0x88cc), length 347: LLDP, length 333: test-hp5945
16:21:11.528217 4c:ae:a3:6a:80:ee > 01:80:c2:00:00:0e, ethertype LLDP
(0x88cc), length 347: LLDP, length 333: test-hp5945
However, if we changed the number of channels, e.g.:
ethtool -L hge11 combined 63
no LLDP frames are seen using tcpdump.
There is a workaround - enable and disable fw-lldp-agent using ethool.
However enabling/disabling fw-lldp-agent using the ethtool flaps the
link which is a problem as a connectivity is lost for a short time period.
Kernel version is 6.8.2, but we see the issue with previous versions as
well. Firmware version is
ethtool -i hge11
driver: ice
version: 6.8.2
firmware-version: 4.40 0x8001c967 1.3534.0
lspci info:
41:00.1 Ethernet controller: Intel Corporation Ethernet Controller
E810-C for QSFP (rev 02)
Subsystem: Intel Corporation Ethernet Network Adapter E810-C-Q2
Product Name: Intel(R) Ethernet Network Adapter E810-CQDA2
Read-only fields:
[V1] Vendor specific: Intel(R) Ethernet Network Adapter E810-CQDA2
[PN] Part number: K91258-011
[SN] Serial number: 6CFE545A0C20
[V2] Vendor specific: 4022
Best regards,
M.
reply other threads:[~2024-04-04 14:59 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1be45a76-90af-4813-824f-8398b69745a9@netx.as \
--to=mgregr@netx.as \
--cc=intel-wired-lan@lists.osuosl.org \
--cc=nagy@netx.as \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).