DPDK usage discussions
 help / color / mirror / Atom feed
From: Gabor LENCSE <lencse@hit.bme.hu>
To: users@dpdk.org
Subject: Weird error messages -- what can be their root cause?
Date: Wed, 13 Sep 2023 16:10:53 +0200	[thread overview]
Message-ID: <4d8d693b-34f2-90fe-f058-a3e61093d5c0@hit.bme.hu> (raw)

Dear All,

I have been using siitperf ( https://github.com/lencsegabor/siitperf ), 
my DPDK-based RFC 8219 SIIT and stateful NAT64 tester under Debian 9 
using DPDK 16 for several years and it worked perfectly.

I have upgraded it to DPDK 21 under Ubuntu 22.04, and it seemed to work.

However it failed under Debian 12 using DPDK 22. Then I tried under 
Debian 11 using DPDK 20.11.7. What I experienced was, that the senders 
functions did not finish sending, and the receiver functions did not 
receive any packets when I did a self-test (the network interfaces were 
looped back with a direct cable). When I sent only a very low number of 
packets, then a sender functions finished, but still no packets arrived.

Here is the dmesg, when I sent only a single frame in each direction:

[ 9475.185898] dmar_fault: 2 callbacks suppressed
[ 9475.185899] DMAR: DRHD: handling fault status reg 302
[ 9475.186851] DMAR: [DMA Read] Request device [01:00.1] PASID ffffffff 
fault addr ffe942000 [fault reason 06] PTE Read access is not set
[ 9475.187730] DMAR: [DMA Read] Request device [01:00.0] PASID ffffffff 
fault addr ffe96b000 [fault reason 06] PTE Read access is not set

Do you have any idea, what the strange messages may mean?

When I send more frames, more messages appear, but their number is not 
proportional to the number of frames per direction. E.g. when I sent 5 
frames per direction, the output was:

[10630.082486] dmar_fault: 2 callbacks suppressed
[10630.082487] DMAR: DRHD: handling fault status reg 302
[10630.083425] DMAR: [DMA Read] Request device [01:00.1] PASID ffffffff 
fault addr ffe942000 [fault reason 06] PTE Read access is not set
[10630.084290] DMAR: [DMA Read] Request device [01:00.0] PASID ffffffff 
fault addr ffe96b000 [fault reason 06] PTE Read access is not set
[10630.282477] DMAR: DRHD: handling fault status reg 502
[10630.283293] DMAR: [DMA Read] Request device [01:00.1] PASID ffffffff 
fault addr ffe942000 [fault reason 06] PTE Read access is not set
[10630.284120] DMAR: [DMA Read] Request device [01:00.0] PASID ffffffff 
fault addr ffe96b000 [fault reason 06] PTE Read access is not set
[10630.482477] DMAR: DRHD: handling fault status reg 702
[10630.483272] DMAR: [DMA Read] Request device [01:00.1] PASID ffffffff 
fault addr ffe942000 [fault reason 06] PTE Read access is not set

I use the same hardware, which worked perfectly under Debian 9. This is 
my NIC:

root@tester:~/siitperf# dpdk-devbind.py --status

Network devices using DPDK-compatible driver
============================================
0000:01:00.0 'Ethernet Controller 10-Gigabit X540-AT2 1528' 
drv=uio_pci_generic unused=ixgbe
0000:01:00.1 'Ethernet Controller 10-Gigabit X540-AT2 1528' 
drv=uio_pci_generic unused=ixgbe

And my kernel command line is:

root@tester:~/siitperf# cat /proc/cmdline
BOOT_IMAGE=/boot/vmlinuz-5.10.0-25-amd64 
root=UUID=2825291f-1f01-4cd3-9f03-ddbaa24c773e ro quiet idle=poll 
isolcpus=1-15 hugepagesz=1G hugepages=32

Any suggestions are appreciated!

Thank you very much for your help in advance!

Best regards,

Gábor


                 reply	other threads:[~2023-09-13 14:10 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=4d8d693b-34f2-90fe-f058-a3e61093d5c0@hit.bme.hu \
    --to=lencse@hit.bme.hu \
    --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).