The irqbalance daemon is not running. At the time of the analysis, "cat
/proc/irq/<NUM>/smp_affinity_list" showed "0-1,24-25" but cat
/proc/interrupts showed the interrupt counts increasing on CPU 4.
If the interrupt affinity was modified by a userspace tool like
irqbalance, I believe that /proc/irq/<NUM>/smp_affinity_list would show
the updated affinity.
This is the thing that I find really weird, the mismatch between the
configured affinity and the observed interrupt counts.
Chris
On 1/15/2021 11:03 AM, Fujinaka, Todd wrote:
I've been reminded that there are other daemons that could be moving around the
CPU affinities, such as irqbalance. Make sure that daemon isn't started.
Todd Fujinaka
Software Application Engineer
Data Center Group
Intel Corporation
todd.fujin...@intel.com
-----Original Message-----
From: Fujinaka, Todd <todd.fujin...@intel.com>
Sent: Friday, January 15, 2021 8:44 AM
To: Chris Friesen <chris.frie...@windriver.com>;
e1000-devel@lists.sourceforge.net
Subject: Re: [E1000-devel] IRQ affinity not working for iavf devices?
Sorry for the slow response. I'm asking around internally as I'm not that
familiar with iavf. I'll let you know when I hear back.
Todd Fujinaka
Software Application Engineer
Data Center Group
Intel Corporation
todd.fujin...@intel.com
-----Original Message-----
From: Chris Friesen <chris.frie...@windriver.com>
Sent: Tuesday, January 12, 2021 3:53 PM
To: e1000-devel@lists.sourceforge.net
Subject: [E1000-devel] IRQ affinity not working for iavf devices?
Hi,
I have a CentOS 7 linux system with 48 logical CPUs and a number of Intel NICs
running the i40e driver. It was booted with
irqaffinity=0-1,24-25 in the kernel boot args, resulting in /proc/irq/default_smp_affinity showing
"0000,03000003". CPUs 2-11 are set as "isolated" in the kernel boot args.
The iavf driver is 3.7.61.20 and the i40e driver is 2.10.19.82
The problem I'm seeing is that /proc/interrupts shows iavf interrupts on other CPUs. For example, here
are some on CPU 4 where I would not expect to see any interrupts given that "cat
/proc/irq/<NUM>/smp_affinity_list" reports "0-1,24-25".
cat /proc/interrupts | grep -e CPU -e 941: -e 942: -e 943: -e 944: -e
945: -e 961: -e 962: -e 963: -e 964: -e 965:
CPU0 CPU1 CPU2 CPU3 CPU4 CPU5
941: 0 0 0 0 28490 0
IR-PCI-MSI-edge iavf-0000:b5:03.6:mbx
942: 0 0 0 0 333832
0 IR-PCI-MSI-edge iavf-net1-TxRx-0
943: 0 0 0 0 300842
0 IR-PCI-MSI-edge iavf-net1-TxRx-1
944: 0 0 0 0 333845
0 IR-PCI-MSI-edge iavf-net1-TxRx-2
945: 0 0 0 0 333822
0 IR-PCI-MSI-edge iavf-net1-TxRx-3
961: 0 0 0 0 28492
0 IR-PCI-MSI-edge iavf-0000:b5:02.7:mbx
962: 0 0 0 0 435608
0 IR-PCI-MSI-edge iavf-net1-TxRx-0
963: 0 0 0 0 394832
0 IR-PCI-MSI-edge iavf-net1-TxRx-1
964: 0 0 0 0 398414
0 IR-PCI-MSI-edge iavf-net1-TxRx-2
965: 0 0 0 0 192847
0 IR-PCI-MSI-edge iavf-net1-TxRx-3
Is this expected? It seems like the iavf and/or the i40e aren't respecting the
configured SMP affinity for the interrupt in question.
Thanks,
Chris
_______________________________________________
E1000-devel mailing list
E1000-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/e1000-devel
To learn more about Intel Ethernet, visit
https://forums.intel.com/s/topic/0TO0P00000018NbWAI/intel-ethernet
_______________________________________________
E1000-devel mailing list
E1000-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/e1000-devel
To learn more about Intel Ethernet, visit
https://forums.intel.com/s/topic/0TO0P00000018NbWAI/intel-ethernet
_______________________________________________
E1000-devel mailing list
E1000-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/e1000-devel
To learn more about Intel Ethernet, visit
https://forums.intel.com/s/topic/0TO0P00000018NbWAI/intel-ethernet
