DPDK usage discussions
 help / color / mirror / Atom feed
From: "Lombardo, Ed" <Ed.Lombardo@netscout.com>
To: Stephen Hemminger <stephen@networkplumber.org>
Cc: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>,
	"users@dpdk.org" <users@dpdk.org>
Subject: RE: mbuf refcnt issue
Date: Wed, 16 Apr 2025 03:51:48 +0000	[thread overview]
Message-ID: <CH3PR01MB84703EE06334B6E4DF0E6B6B8FBD2@CH3PR01MB8470.prod.exchangelabs.com> (raw)
In-Reply-To: <20250415170140.788eca01@hermes.local>

Hi,
My big issue I have been trying to resolve is when I send packets from a particular pcap file the DPDK mbufs retrieved from rte_eth_rx_burst() after short time showed the mbufs with refcnt not 1.  I then modified our application to set the mbuf to 10 and when the error hit the rte_eth_rx_burst() would then provide mbufs with refcnt set to 10 or 9.  I found testing other pcap files they did not cause any failures.  So, I then analyzed this one pcap file and found that it had "malformed packet (Exception occurred)".  I then removed this one malformed packet from the pcap file and replayed the modified pcap file and no errors occurred.  I think when our application retransmitted this malformed packet it messed up the DPDK mempool and Malicious Driver Detection event occurred.

Thanks,
Ed

-----Original Message-----
From: Stephen Hemminger <stephen@networkplumber.org> 
Sent: Tuesday, April 15, 2025 8:02 PM
To: Lombardo, Ed <Ed.Lombardo@netscout.com>
Cc: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>; users@dpdk.org
Subject: Re: mbuf refcnt issue

External Email: This message originated outside of NETSCOUT. Do not click links or open attachments unless you recognize the sender and know the content is safe.

On Tue, 15 Apr 2025 13:10:18 +0000
"Lombardo, Ed" <Ed.Lombardo@netscout.com> wrote:

> Hi,
> I enabled more logging, and I see the following ICE Driver Malicious Driver Detection event 6 and 7, in the dpdk_log.txt when failure occurs.
> 
> EAL: Allocated 2112 bytes of per-lcore data with a 64-byte alignment
> ICE_INIT: ice_load_pkg_type(): Active package is: 1.3.35.0, ICE OS 
> Default Package (single VLAN mode)
> ICE_INIT: ice_load_pkg_type(): Active package is: 1.3.35.0, ICE OS 
> Default Package (single VLAN mode)
> MBUF: Registered dynamic field rte_dynfield_timestamp (sz=8, al=8, 
> fl=0x0) -> 96
> MBUF: Registered dynamic flag rte_dynflag_rx_timestamp (fl=0x0) -> 23
> MBUF: Registered dynamic field rte_mbuf_dynfield_udata64 (sz=8, al=8, 
> fl=0x0) -> 104
> ICE_DRIVER: ice_interrupt_handler(): OICR: MDD event
> ICE_DRIVER: ice_interrupt_handler(): Malicious Driver Detection event 
> 7 by TCLAN on TX queue 2 PF# 0
> ICE_DRIVER: ice_interrupt_handler(): Malicious Driver Detection event 
> 6 by TDPU on TX queue 2 PF# 0
> 
> Are malformed packets handled by DPDK, can malformed packets corrupt the mbufs?
> 
> Thanks,
> Ed

I do not know. The malicious driver detection is when application asks to send a packet on a VF that is not for that VF. I.e source MAC or VLAN mismatch.

  reply	other threads:[~2025-04-16  3:51 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-04-04 22:00 Lombardo, Ed
2025-04-04 22:29 ` Dmitry Kozlyuk
2025-04-07 19:53   ` Lombardo, Ed
2025-04-08 22:33     ` Lombardo, Ed
2025-04-09  3:53       ` Stephen Hemminger
2025-04-09  4:46         ` Lombardo, Ed
2025-04-09 16:24           ` Stephen Hemminger
2025-04-09 23:22             ` Lombardo, Ed
2025-04-10  3:25               ` Stephen Hemminger
2025-04-10  3:58                 ` Lombardo, Ed
2025-04-10 21:15                   ` Lombardo, Ed
2025-04-15 13:10                     ` Lombardo, Ed
2025-04-16  0:01                       ` Stephen Hemminger
2025-04-16  3:51                         ` Lombardo, Ed [this message]
2025-04-09 16:15       ` Lombardo, Ed

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=CH3PR01MB84703EE06334B6E4DF0E6B6B8FBD2@CH3PR01MB8470.prod.exchangelabs.com \
    --to=ed.lombardo@netscout.com \
    --cc=dmitry.kozliuk@gmail.com \
    --cc=stephen@networkplumber.org \
    --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).