DPDK usage discussions
 help / color / mirror / Atom feed
From: Dariusz Sosnowski <dsosnowski@nvidia.com>
To: Charles Oluwaseun Oluboyo <charlesboyo@gmail.com>
Cc: <users@dpdk.org>
Subject: Re: Unable to initialize ConnectX-4 ports in DPDK 24.11 and 25.03
Date: Thu, 5 Jun 2025 16:14:59 +0200	[thread overview]
Message-ID: <20250605141459.obfe6vl7ctpatlnh@ds-vm-debian.local> (raw)
In-Reply-To: <CAPD_G3LshTsRv07Av=bx5vNJvRc8=s3D2LPLtm6c84jpUwumyQ@mail.gmail.com>

On Thu, May 29, 2025 at 12:00:29AM +0100, Charles Oluwaseun Oluboyo wrote:
> Hello all.
> 
> Appears the PMD is returning the wrong value for the max MTU.
> 
> The error about the frame size is raised in eth_dev_validate_mtu
> called from rte_eth_dev_configure.
> dev_info->max_rx_pktlen = 65536
> dev_info->max_mtu = 65535
> 
> That is not true. From my testing, max MTU is 9960 for this ConnectX-4
> Should this have been set in mlx5_dev_infos_get?

Correct, the current implementation of mlx5_dev_infos_get() returns
invalid max MTU and Rx packet length.
It should be fixed. I have opened a bugzilla ticket for tracking:
https://bugs.dpdk.org/show_bug.cgi?id=1719

Thank you for reporting.

> 
> Regards,
> 
> Oluwaseun
> 
> On Tue, May 27, 2025 at 7:12 PM Charles Oluwaseun Oluboyo
> <charlesboyo@gmail.com> wrote:
> >
> > Hello Dariusz and all,
> >
> > >
> > > One of the patches merged to 24.11 introduced an issue with queues
> > > descriptor limits on older NICs.
> > >
> > > A patch with a fix for that is pending for merge to 25.07.
> > >
> > > Would you be able apply the following patches and test on your side?
> > >
> > > - https://patches.dpdk.org/project/dpdk/patch/20250424133128.133900-1-viacheslavo@nvidia.com/
> > > - https://patches.dpdk.org/project/dpdk/patch/20250514075530.419423-1-viacheslavo@nvidia.com/
> > >
> >
> > I applied these patches, and they have fixed the issue with the
> > descriptors. There are now 32768 per queue.
> > I have tested with testpmd, and the ports work. However, I am unable
> > to initialize the ports with pktgen.
> >
> > Initialize Port 2 ...
> >    Max MTU: 65535
> >    Enabling Tx TCP_CKSUM offload
> >    Enabling Tx UDP_CKSUM offload
> >    Enabling Tx IPV4_CKSUM offload
> >    Allocate packet sequence array
> >    Setup latency defaults
> >    Setup fill pattern defaults
> >    Configure device
> > ETHDEV: Frame size (65553) > device max frame size (65536) for port_id 2
> > !PANIC!: Cannot configure device: port=2, Num queues 1,1.
> > EAL: PANIC in initialize_port_info():
> > Cannot configure device: port=2, Num queues 1,1.
> > 0: /usr/local/lib/x86_64-linux-gnu/librte_eal.so.25
> > (rte_dump_stack+0x42) [7f16cb3de6d2]
> > 1: /usr/local/lib/x86_64-linux-gnu/librte_eal.so.25 (__rte_panic+0xd4)
> > [7f16cb3b737f]
> > 2: ./usr/local/bin/pktgen (5606908ff000+0xd26e) [56069090c26e]
> > 3: ./usr/local/bin/pktgen (5606908ff000+0x2c61e) [56069092b61e]
> > 4: ./usr/local/bin/pktgen (5606908ff000+0xe339) [56069090d339]
> > 5: /lib/x86_64-linux-gnu/libc.so.6 (7f16cafc2000+0x29d90) [7f16cafebd90]
> > 6: /lib/x86_64-linux-gnu/libc.so.6 (__libc_start_main+0x80) [7f16cafebe40]
> > 7: ./usr/local/bin/pktgen (5606908ff000+0xecc5) [56069090dcc5]
> >
> > 65553 looks very unusual, doesn't it? Do you have any suggestions as
> > to what to do next?
> >
> > Regards,
> >
> > Oluwaseun

      reply	other threads:[~2025-06-05 14:15 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-05-24  0:50 Charles Oluwaseun Oluboyo
2025-05-27  8:29 ` Dariusz Sosnowski
2025-05-27 18:12   ` Charles Oluwaseun Oluboyo
2025-05-28 23:00     ` Charles Oluwaseun Oluboyo
2025-06-05 14:14       ` Dariusz Sosnowski [this message]

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=20250605141459.obfe6vl7ctpatlnh@ds-vm-debian.local \
    --to=dsosnowski@nvidia.com \
    --cc=charlesboyo@gmail.com \
    --cc=users@dpdk.org \
    /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 NNTP newsgroup(s).