DPDK with i225/226 on elkhartlake

DPDK with i225/226 on elkhartlake

[SamChen 陳嘉良]

[-- Attachment #1.1: Type: text/plain, Size: 582 bytes --]

Hi,
We bind the uio_pci_generic driver and use the command of the attached image to execute DPDK. However, we encountered the problem of device cannot be used on i225V/i226V in elkhartlake platform. Currently, the same method works fine for i211 and intel(r) Killer(tm) Ethernet E3100 2.5 Gbps. Do you have any suggestions? Thanks!

Sam Chen
S/W Engineer
Network Security Division
AAEON Technology Inc.
T: +886-2-8919-1234 Ext:1774
F: +886-2-8919-1054
www.aaeon.com<http://www.aaeon.com/>

https://www.aaeon.com/en/c/network-appliance-video-surveillance
[簽名檔_NSD]


[-- Attachment #1.2: Type: text/html, Size: 5146 bytes --]

[-- Attachment #2: image001.jpg --]
[-- Type: image/jpeg, Size: 35860 bytes --]

Re: DPDK with i225/226 on elkhartlake

[Bruce Richardson]

On Wed, Sep 21, 2022 at 03:23:09AM +0000, SamChen 陳嘉良 wrote:
>    Hi,
>    We bind the uio_pci_generic driver and use the command of the attached
>    image to execute DPDK. However, we encountered the problem of device
>    cannot be used on i225V/i226V in elkhartlake platform. Currently, the
>    same method works fine for i211 and intel® Killer™ Ethernet E3100 2.5
>    Gbps. Do you have any suggestions? Thanks!
> 
Hi,

while I can't comment on the NIC-specific issues, I would just comment that
it's really not recommended to use uio_pci_generic - or any other uio
driver - for DPDK any more. You should consider switching to using
vfio-pci. Even if there is no iommu enabled, vfio-pci still tends to give a
better user experience. If iommu is enabled, you also get a lot more
security from the memory protection it provides.

Regards,
/Bruce

RE: DPDK with i225/226 on elkhartlake

[SamChen 陳嘉良]

[-- Attachment #1: Type: text/plain, Size: 1366 bytes --]

Hi all,
I have switched to using vfio-pci, but still can't use DPDK. Attached are the logs I can't use in I225-IT and I226V for your reference.
If I have obtained other platforms with intel 225/226 systems, I will check again, thanks!

Sam Chen

-----Original Message-----
From: Bruce Richardson <bruce.richardson@intel.com> 
Sent: Wednesday, September 21, 2022 4:57 PM
To: SamChen 陳嘉良 <SamChen@aaeon.com.tw>
Cc: dev@dpdk.org
Subject: Re: DPDK with i225/226 on elkhartlake

On Wed, Sep 21, 2022 at 03:23:09AM +0000, SamChen 陳嘉良 wrote:
>    Hi,
>    We bind the uio_pci_generic driver and use the command of the attached
>    image to execute DPDK. However, we encountered the problem of device
>    cannot be used on i225V/i226V in elkhartlake platform. Currently, the
>    same method works fine for i211 and intel® Killer™ Ethernet E3100 2.5
>    Gbps. Do you have any suggestions? Thanks!
> 
Hi,

while I can't comment on the NIC-specific issues, I would just comment that it's really not recommended to use uio_pci_generic - or any other uio driver - for DPDK any more. You should consider switching to using vfio-pci. Even if there is no iommu enabled, vfio-pci still tends to give a better user experience. If iommu is enabled, you also get a lot more security from the memory protection it provides.

Regards,
/Bruce

[-- Attachment #2: i225-IT_log --]
[-- Type: application/octet-stream, Size: 1005 bytes --]

[nsd@localhost ~]$ sudo su
[sudo] password for nsd: 
[root@localhost nsd]# modprobe vfio-pci
[root@localhost nsd]# cd dpdk-22.07/usertools/
[root@localhost usertools]# ./dpdk-devbind.py -b vfio-pci 01:00.0 02:00.0
[root@localhost usertools]# cd ../build/app/
[root@localhost app]# ./dpdk-testpmd -l 0-1 -n 1
EAL: Detected CPU lcores: 4
EAL: Detected NUMA nodes: 1
EAL: Detected static linkage of DPDK
EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
EAL: Selected IOVA mode 'VA'
EAL: VFIO support initialized
TELEMETRY: No legacy callbacks, legacy socket not created
testpmd: No probed ethernet devices
testpmd: create a new mbuf pool <mb_pool_0>: n=155456, size=2176, socket=0
testpmd: preferred mempool ops selected: ring_mp_mc
Done
No commandline core given, start packet forwarding
io packet forwarding - ports=0 - cores=0 - streams=0 - NUMA support enabled, MP allocation mode: native

  io packet forwarding packets/burst=32
  nb forwarding cores=1 - nb forwarding ports=0
Press enter to exit


[-- Attachment #3: i226V_log --]
[-- Type: application/octet-stream, Size: 963 bytes --]

[nsd@localhost ~]$ sudo su
[sudo] password for nsd: 
[root@localhost nsd]# modprobe vfio-pci
[root@localhost nsd]# cd dpdk-22.07/usertools/
[root@localhost usertools]# ./dpdk-devbind.py -b vfio-pci 03:00.0 06:00.0
[root@localhost usertools]# cd ../build/app/
[root@localhost app]# ./dpdk-testpmd -l 0-1 -n 1
EAL: Detected CPU lcores: 4
EAL: Detected NUMA nodes: 1
EAL: Detected static linkage of DPDK
EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
EAL: Selected IOVA mode 'VA'
EAL: VFIO support initialized
EAL: Using IOMMU type 1 (Type 1)
EAL: Probe PCI driver: net_igc (8086:125c) device: 0000:03:00.0 (socket 0)
EAL: Releasing PCI mapped resource for 0000:03:00.0
EAL: Calling pci_unmap_resource for 0000:03:00.0 at 0x1100800000
EAL: Calling pci_unmap_resource for 0000:03:00.0 at 0x1100900000
EAL: Requested device 0000:03:00.0 cannot be used
EAL: Using IOMMU type 1 (Type 1)
EAL: Probe PCI driver: net_igc (8086:125c) device: 0000:06:00.0 (socket 0)

RE: DPDK with i225/226 on elkhartlake

[Yang, Qiming]

Hi, Sam
We don't support I225-IT in 22.07 release, you can update you code to the latest next-net-intel branch, it contain the fix.
But for i226v's log, I don't see any issue, seems 06:00.0 init success. Could you clarify it?

Qiming 
> -----Original Message-----
> From: SamChen 陳嘉良 <SamChen@aaeon.com.tw>
> Sent: Thursday, September 22, 2022 10:57 AM
> To: Richardson, Bruce <bruce.richardson@intel.com>
> Cc: dev@dpdk.org
> Subject: RE: DPDK with i225/226 on elkhartlake
> 
> Hi all,
> I have switched to using vfio-pci, but still can't use DPDK. Attached are the
> logs I can't use in I225-IT and I226V for your reference.
> If I have obtained other platforms with intel 225/226 systems, I will check
> again, thanks!
> 
> Sam Chen
> 
> -----Original Message-----
> From: Bruce Richardson <bruce.richardson@intel.com>
> Sent: Wednesday, September 21, 2022 4:57 PM
> To: SamChen 陳嘉良 <SamChen@aaeon.com.tw>
> Cc: dev@dpdk.org
> Subject: Re: DPDK with i225/226 on elkhartlake
> 
> On Wed, Sep 21, 2022 at 03:23:09AM +0000, SamChen 陳嘉良 wrote:
> >    Hi,
> >    We bind the uio_pci_generic driver and use the command of the attached
> >    image to execute DPDK. However, we encountered the problem of device
> >    cannot be used on i225V/i226V in elkhartlake platform. Currently, the
> >    same method works fine for i211 and intel® Killer™ Ethernet E3100 2.5
> >    Gbps. Do you have any suggestions? Thanks!
> >
> Hi,
> 
> while I can't comment on the NIC-specific issues, I would just comment that
> it's really not recommended to use uio_pci_generic - or any other uio driver -
> for DPDK any more. You should consider switching to using vfio-pci. Even if
> there is no iommu enabled, vfio-pci still tends to give a better user
> experience. If iommu is enabled, you also get a lot more security from the
> memory protection it provides.
> 
> Regards,
> /Bruce

RE: DPDK with i225/226 on elkhartlake

[SamChen 陳嘉良]

[-- Attachment #1: Type: text/plain, Size: 2645 bytes --]

Hi Qiming,
Thanks for your information. 
From the log of i226V, you can see that the string "Requested device 0000:03:00.0 cannot be used" appeared at init 03:00.0. 
Then its execution stopped at "EAL: Probe PCI driver: net_igc (8086:125c) device: 0000:06:00.0 (socket 0)" with no subsequent action.

If the device can be executed normally, the complete message should be like the i211_log attached, right?

Sam Chen

-----Original Message-----
From: Yang, Qiming <qiming.yang@intel.com> 
Sent: Friday, September 23, 2022 3:51 PM
To: SamChen 陳嘉良 <SamChen@aaeon.com.tw>; Richardson, Bruce <bruce.richardson@intel.com>
Cc: dev@dpdk.org
Subject: RE: DPDK with i225/226 on elkhartlake

Hi, Sam
We don't support I225-IT in 22.07 release, you can update you code to the latest next-net-intel branch, it contain the fix.
But for i226v's log, I don't see any issue, seems 06:00.0 init success. Could you clarify it?

Qiming 
> -----Original Message-----
> From: SamChen 陳嘉良 <SamChen@aaeon.com.tw>
> Sent: Thursday, September 22, 2022 10:57 AM
> To: Richardson, Bruce <bruce.richardson@intel.com>
> Cc: dev@dpdk.org
> Subject: RE: DPDK with i225/226 on elkhartlake
> 
> Hi all,
> I have switched to using vfio-pci, but still can't use DPDK. Attached 
> are the logs I can't use in I225-IT and I226V for your reference.
> If I have obtained other platforms with intel 225/226 systems, I will 
> check again, thanks!
> 
> Sam Chen
> 
> -----Original Message-----
> From: Bruce Richardson <bruce.richardson@intel.com>
> Sent: Wednesday, September 21, 2022 4:57 PM
> To: SamChen 陳嘉良 <SamChen@aaeon.com.tw>
> Cc: dev@dpdk.org
> Subject: Re: DPDK with i225/226 on elkhartlake
> 
> On Wed, Sep 21, 2022 at 03:23:09AM +0000, SamChen 陳嘉良 wrote:
> >    Hi,
> >    We bind the uio_pci_generic driver and use the command of the attached
> >    image to execute DPDK. However, we encountered the problem of device
> >    cannot be used on i225V/i226V in elkhartlake platform. Currently, the
> >    same method works fine for i211 and intel® Killer™ Ethernet E3100 2.5
> >    Gbps. Do you have any suggestions? Thanks!
> >
> Hi,
> 
> while I can't comment on the NIC-specific issues, I would just comment 
> that it's really not recommended to use uio_pci_generic - or any other 
> uio driver - for DPDK any more. You should consider switching to using 
> vfio-pci. Even if there is no iommu enabled, vfio-pci still tends to 
> give a better user experience. If iommu is enabled, you also get a lot 
> more security from the memory protection it provides.
> 
> Regards,
> /Bruce

[-- Attachment #2: i211_log --]
[-- Type: application/octet-stream, Size: 2349 bytes --]

[nsd@localhost ~]$ sudo su
[sudo] password for nsd: 
[root@localhost nsd]# modprobe vfio-pci
[root@localhost nsd]# cd dpdk-22.07/usertools/
[root@localhost usertools]# ./dpdk-devbind.py -b vfio-pci 01:00.0 02:00.0
[root@localhost usertools]# cd ../build/app/
[root@localhost app]# ./dpdk-testpmd -l 0-1 -n 1
EAL: Detected CPU lcores: 4
EAL: Detected NUMA nodes: 1
EAL: Detected static linkage of DPDK
EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
EAL: Selected IOVA mode 'VA'
EAL: VFIO support initialized
EAL: Using IOMMU type 1 (Type 1)
EAL: Ignore mapping IO port bar(2)
EAL: Probe PCI driver: net_e1000_igb (8086:1539) device: 0000:01:00.0 (socket 0)
EAL: Ignore mapping IO port bar(2)
EAL: Probe PCI driver: net_e1000_igb (8086:1539) device: 0000:02:00.0 (socket 0)
TELEMETRY: No legacy callbacks, legacy socket not created
testpmd: create a new mbuf pool <mb_pool_0>: n=155456, size=2176, socket=0
testpmd: preferred mempool ops selected: ring_mp_mc
Configuring Port 0 (socket 0)
Port 0: 00:07:32:9C:6A:C8
Configuring Port 1 (socket 0)
Port 1: 00:07:32:9C:6A:C9
Checking link statuses...
Done
No commandline core given, start packet forwarding
io packet forwarding - ports=2 - cores=1 - streams=2 - NUMA support enabled, MP allocation mode: native
Logical Core 1 (socket 0) forwards packets on 2 streams:
  RX P=0/Q=0 (socket 0) -> TX P=1/Q=0 (socket 0) peer=02:00:00:00:00:01
  RX P=1/Q=0 (socket 0) -> TX P=0/Q=0 (socket 0) peer=02:00:00:00:00:00

  io packet forwarding packets/burst=32
  nb forwarding cores=1 - nb forwarding ports=2
  port 0: RX queue number: 1 Tx queue number: 1
    Rx offloads=0x0 Tx offloads=0x0
    RX queue: 0
      RX desc=512 - RX free threshold=32
      RX threshold registers: pthresh=0 hthresh=0  wthresh=0
      RX Offloads=0x0
    TX queue: 0
      TX desc=512 - TX free threshold=0
      TX threshold registers: pthresh=8 hthresh=1  wthresh=16
      TX offloads=0x0 - TX RS bit threshold=0
  port 1: RX queue number: 1 Tx queue number: 1
    Rx offloads=0x0 Tx offloads=0x0
    RX queue: 0
      RX desc=512 - RX free threshold=32
      RX threshold registers: pthresh=0 hthresh=0  wthresh=0
      RX Offloads=0x0
    TX queue: 0
      TX desc=512 - TX free threshold=0
      TX threshold registers: pthresh=8 hthresh=1  wthresh=16
      TX offloads=0x0 - TX RS bit threshold=0
Press enter to exit


[-- Attachment #3: i226V_log --]
[-- Type: application/octet-stream, Size: 963 bytes --]

[nsd@localhost ~]$ sudo su
[sudo] password for nsd: 
[root@localhost nsd]# modprobe vfio-pci
[root@localhost nsd]# cd dpdk-22.07/usertools/
[root@localhost usertools]# ./dpdk-devbind.py -b vfio-pci 03:00.0 06:00.0
[root@localhost usertools]# cd ../build/app/
[root@localhost app]# ./dpdk-testpmd -l 0-1 -n 1
EAL: Detected CPU lcores: 4
EAL: Detected NUMA nodes: 1
EAL: Detected static linkage of DPDK
EAL: Multi-process socket /var/run/dpdk/rte/mp_socket
EAL: Selected IOVA mode 'VA'
EAL: VFIO support initialized
EAL: Using IOMMU type 1 (Type 1)
EAL: Probe PCI driver: net_igc (8086:125c) device: 0000:03:00.0 (socket 0)
EAL: Releasing PCI mapped resource for 0000:03:00.0
EAL: Calling pci_unmap_resource for 0000:03:00.0 at 0x1100800000
EAL: Calling pci_unmap_resource for 0000:03:00.0 at 0x1100900000
EAL: Requested device 0000:03:00.0 cannot be used
EAL: Using IOMMU type 1 (Type 1)
EAL: Probe PCI driver: net_igc (8086:125c) device: 0000:06:00.0 (socket 0)