Re: [dts] Minimum Tester/DUT configuration

["Liu, Yong" <yong.liu@intel.com>]

[-- Attachment #1: Type: text/plain, Size: 24943 bytes --]

Hi John,
I think you’re trying to get the baseline for performance.  In our assumption, performance and function executions will use different environment.
In your setup, DTS is running function case “test_l2fwd_integrity”.  The case failed for l2fwd output message changed, one patch is needed for fix the issue.
For enable performance execution, you need to change execution parameters like “parameters=nic_type=cfg:perf=true”.

The warning log for hugepage is caused by DTS can’t normally check status of dpdk hugepage files.
I guess DUT’s system is Ubuntu,  we can see this warning if GVFS is mounted.  It will be better to umount it for DTS can check dpdk process status correctly.

Thank,
Marvin

From: john miller [mailto:john.miller@atomicrules.com]
Sent: Friday, January 27, 2017 9:34 PM
To: Xu, HuilongX <huilongx.xu@intel.com>
Cc: Liu, Yong <yong.liu@intel.com>; dts@dpdk.org
Subject: Re: [dts] Minimum Tester/DUT configuration

Hi,

After removing all of the dpdk versions on our DUT and tester, updating all packages we were able to get past this issue.

Our goal is to get a baseline using the Fortville spirit on our DUT before we replace it with our hardware and PMD.  I am running the majority of the tests available in DTS with two Fortville eagles on the tester and 1 Fortville spirit in the DUT configured as a 4x10 NIC.   As you can see from the test_results.xls file that i attached, some tests are passing and some failing.  Many of the failing tests are a result of hardcoded core masks in the tests as both our DUT and tester are single socket boards with 4 core processors.  However some tests like l2fwd are failing and we are not sure why.  We have attached the dts.log file and the l2fwd results file in the hope that you can tell us why this test fails with our setup.  Note that i had to change the following line TestSuite_l2fwd.py because it was trying to force the test to use ports on different NIC’s.

-        self.dut_ports = self.dut.get_ports_performance()

+
+        self.dut_ports = self.dut.get_ports_performance( nic_type='any', perf=None, socket=None,
+                                                         force_same_socket=True,
+                                                         force_different_nic=False)



We see this in the dts log       DTS_DUT_RESULT: [192.168.0.5] There are some dpdk process not free hugepage

But it looks like a warning in the code and we can see that the on the DUT huge pages are allocated and free.

Also, we are seeing this error in the dts.log at the end of the dpdk build, "Installation cannot run with T defined and DESTDIR undefined"
It appears that the build is complete and since DTS does not actually do a dpdk install we think this is ok, but are not sure why the error is happening.

Any help or insight that you could provide to help us get our baseline system functional using the Intel Fortville NIC’s would be appreciated.

Thank you,
-John


Tests that we are running.

test_suites=
    cmdline,
    hello_world,
    multiprocess,
    timer,
    blacklist,
    mac_filter,
    ieee1588,
    checksum_offload,
    jumboframes,
    ipfrag,
    link_flowctrl,
    vlan,
    ip_pipeline,
    pmd_bonded,
    dynamic_config,
    generic_filter,
    dual_vlan,
    shutdown_api,
    fdir,
    ipv4_reassembly,
    scatter,
    pmdrssreta,
    pmd,
    l2fwd,
    kni,
    uni_pkt






On Jan 25, 2017, at 1:34 AM, Xu, HuilongX <huilongx.xu@intel.com<mailto:huilongx.xu@intel.com>> wrote:


Hi yong and john,
I run l2fwd successful on  my machine with newest dts and dpdk17.02.
John could you send the dts.log and l2fwd.log in output to me?
Thanks a lot
From: Liu, Yong
Sent: Wednesday, January 25, 2017 1:26 PM
To: john miller; Xu, HuilongX
Cc: dts@dpdk.org<mailto:dts@dpdk.org>
Subject: RE: [dts] Minimum Tester/DUT configuration

Got.  From the output log, there’s no initialization action when l2fwd start-up.
There’re two possible causes for this. One is port not bind to igb_uio, the other is CONFIG_RTE_LIBRTE_I40E_PMD not enabled in configuration file.

L2fwd case is included in our dpdk17.02 validation plan. Huilong, have we done this test on latest dpdk code?

- Marvin

From: john miller [mailto:john.miller@atomicrules.com]
Sent: Wednesday, January 25, 2017 11:20 AM
To: Liu, Yong <yong.liu@intel.com<mailto:yong.liu@intel.com>>
Cc: dts@dpdk.org<mailto:dts@dpdk.org>
Subject: Re: [dts] Minimum Tester/DUT configuration

I meant the head of DPDK. So we are using the latest code for both DTS and DPDK.


On Jan 24, 2017, at 10:08 PM, john miller <john.miller@atomicrules.com<mailto:john.miller@atomicrules.com>> wrote:

Hi Marvin,

I think we have that configured correctly, here is the output from the test.  We are using the latest code at the head of DTS which i s v17.02.

DTS_DUT_CMD: [192.168.0.34] usertools/dpdk-devbind.py --bind=igb_uio 0000:02:00.0 0000:02:00.1 0000:02:00.2 0000:02:00.3

-John



On Jan 24, 2017, at 9:59 PM, Liu, Yong <yong.liu@intel.com<mailto:yong.liu@intel.com>> wrote:

Hi John,
In the DTS execution process, we will bind configured ports to igb_uio by the tool named dpdk-devbind.
This tool’s location changed to folder usertools recently. So if you are not using v17.02, may encounter this issue.

usertools/dpdk-devbind.py --bind=igb_uio 0000:02:00.0 0000:02:00.1 0000:02:00.2 0000:02:00.3

Thanks,
Marvin

From: john miller [mailto:john.miller@atomicrules.com]
Sent: Wednesday, January 25, 2017 12:11 AM
To: Liu, Yong <yong.liu@intel.com<mailto:yong.liu@intel.com>>
Cc: dts@dpdk.org<mailto:dts@dpdk.org>
Subject: Re: [dts] Minimum Tester/DUT configuration

Thank you for your response.

We have updated our tester and DUT configuration.  We have 2 Fortville eagles in our tester and 1 Fortville spirit configured as a 4x10 in our DUT.  We have link on all 4 ports.

We are trying to run the l2fwd DTS test and we get this error.

"EAL: Error - exiting with code: 1
  Cause: No Ethernet ports - bye”



The complete output is shown below.  I searched the forums for this error and got a few hits but none of the matching threads seemed like what we are experiencing.  Could you help us isolate this issue ?

Thank you,
-John




DUT 192.168.0.34
                INFO: ssh root@192.168.0.36<mailto:root@192.168.0.36>
                INFO: ssh root@192.168.0.36<mailto:root@192.168.0.36>
                INFO: ssh root@192.168.0.34<mailto:root@192.168.0.34>
                INFO: ssh root@192.168.0.34<mailto:root@192.168.0.34>
                INFO: CACHE: Cache will not be read.
                INFO: SKIP: The DPDK setup steps will be executed.
      DTS_TESTER_CMD: scp -v dep/tgen.tgz root@192.168.0.36<mailto:root@192.168.0.36>:~/
      DTS_TESTER_CMD: scp -v dep/tclclient.tgz root@192.168.0.36<mailto:root@192.168.0.36>:~/
      DTS_TESTER_CMD: tar zxf tgen.tgz
      DTS_TESTER_CMD: tar zxf tclclient.tgz
      DTS_TESTER_CMD: modprobe uio
      DTS_TESTER_CMD: modprobe igb
      DTS_TESTER_CMD: modprobe ixgbe
      DTS_TESTER_CMD: modprobe e1000e
      DTS_TESTER_CMD: modprobe e1000
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:00\:19.0/vendor
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:00\:19.0/device
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:00\:19.0/vendor
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:00\:19.0/device
      DTS_TESTER_CMD: ifconfig eno1 up
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:01\:00.0/vendor
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:01\:00.0/device
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:01\:00.0/vendor
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:01\:00.0/device
      DTS_TESTER_CMD: ifconfig enp1s0f0 up
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:01\:00.1/vendor
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:01\:00.1/device
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:01\:00.1/vendor
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:01\:00.1/device
      DTS_TESTER_CMD: ifconfig enp1s0f1 up
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:02\:00.0/vendor
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:02\:00.0/device
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:02\:00.0/vendor
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:02\:00.0/device
      DTS_TESTER_CMD: ifconfig enp2s0f0 up
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:02\:00.1/vendor
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:02\:00.1/device
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:02\:00.1/vendor
      DTS_TESTER_CMD: cat /sys/bus/pci/devices/0000\:02\:00.1/device
      DTS_TESTER_CMD: ifconfig enp2s0f1 up
      DTS_TESTER_CMD: Tester: [0000:00:19.0 8086:1503] unknow_nic
      DTS_TESTER_CMD: Tester: [0000:01:00.0 8086:1572] enp1s0f0
      DTS_TESTER_CMD: Tester: [0000:01:00.1 8086:1572] enp1s0f1
      DTS_TESTER_CMD: Tester: [0000:02:00.0 8086:1572] enp2s0f0
      DTS_TESTER_CMD: Tester: [0000:02:00.1 8086:1572] enp2s0f1
      DTS_TESTER_CMD: {'intf': 'enp1s0f0', 'mac': '3c:fd:fe:9c:c3:4c', 'pci': '0000:01:00.0', 'ipv6': 'fe80::23af:6b1b:3fca:fe8a', 'type': '8086:1572', 'port': <net_device.NetDevice object at 0x7f5c8f549cd0>}
      DTS_TESTER_CMD: {'intf': 'enp1s0f1', 'mac': '3c:fd:fe:9c:c3:4d', 'pci': '0000:01:00.1', 'ipv6': 'fe80::2e0:6ec:56c9:1b3', 'type': '8086:1572', 'port': <net_device.NetDevice object at 0x7f5c8f571110>}
      DTS_TESTER_CMD: {'intf': 'enp2s0f0', 'mac': '3c:fd:fe:9e:11:c0', 'pci': '0000:02:00.0', 'ipv6': 'fe80::27d1:7383:502b:2486', 'type': '8086:1572', 'port': <net_device.NetDevice object at 0x7f5c8f549e90>}
      DTS_TESTER_CMD: {'intf': 'enp2s0f1', 'mac': '3c:fd:fe:9e:11:c1', 'pci': '0000:02:00.1', 'ipv6': 'fe80::b4d0:ec1f:fe29:687f', 'type': '8086:1572', 'port': <net_device.NetDevice object at 0x7f5c8f549c90>}
         DTS_DUT_CMD: [192.168.0.34] ls /tmp/ && cd ~
         DTS_DUT_CMD: [192.168.0.34] scp -v dep/dpdk.tar.gz root@192.168.0.34<mailto:root@192.168.0.34>:/tmp/
         DTS_DUT_CMD: [192.168.0.34] ulimit -c unlimited
         DTS_DUT_CMD: [192.168.0.34] rm -rf ~/dpdk
         DTS_DUT_CMD: [192.168.0.34] tar zxf /tmp/dpdk.tar.gz -C ~
         DTS_DUT_CMD: [192.168.0.34] ls ~/dpdk
         DTS_DUT_CMD: [192.168.0.34] cd ~/dpdk
         DTS_DUT_CMD: [192.168.0.34] alias ls='ls --color=none'
         DTS_DUT_CMD: [192.168.0.34] DUT: [0000:00:1f.6 8086:15b8] Skipped: Unknown/not selected
         DTS_DUT_CMD: [192.168.0.34] cat /sys/bus/pci/devices/0000\:02\:00.0/vendor
         DTS_DUT_CMD: [192.168.0.34] cat /sys/bus/pci/devices/0000\:02\:00.0/device
         DTS_DUT_CMD: [192.168.0.34] cat /sys/bus/pci/devices/0000\:02\:00.0/vendor
         DTS_DUT_CMD: [192.168.0.34] cat /sys/bus/pci/devices/0000\:02\:00.0/device
         DTS_DUT_CMD: [192.168.0.34] cat /sys/bus/pci/devices/0000\:02\:00.1/vendor
         DTS_DUT_CMD: [192.168.0.34] cat /sys/bus/pci/devices/0000\:02\:00.1/device
         DTS_DUT_CMD: [192.168.0.34] cat /sys/bus/pci/devices/0000\:02\:00.1/vendor
         DTS_DUT_CMD: [192.168.0.34] cat /sys/bus/pci/devices/0000\:02\:00.1/device
         DTS_DUT_CMD: [192.168.0.34] cat /sys/bus/pci/devices/0000\:02\:00.2/vendor
         DTS_DUT_CMD: [192.168.0.34] cat /sys/bus/pci/devices/0000\:02\:00.2/device
         DTS_DUT_CMD: [192.168.0.34] cat /sys/bus/pci/devices/0000\:02\:00.2/vendor
         DTS_DUT_CMD: [192.168.0.34] cat /sys/bus/pci/devices/0000\:02\:00.2/device
         DTS_DUT_CMD: [192.168.0.34] cat /sys/bus/pci/devices/0000\:02\:00.3/vendor
         DTS_DUT_CMD: [192.168.0.34] cat /sys/bus/pci/devices/0000\:02\:00.3/device
         DTS_DUT_CMD: [192.168.0.34] cat /sys/bus/pci/devices/0000\:02\:00.3/vendor
         DTS_DUT_CMD: [192.168.0.34] cat /sys/bus/pci/devices/0000\:02\:00.3/device
         DTS_DUT_CMD: [192.168.0.34] echo 0000:02:00.0 > /sys/bus/pci/devices/0000\:02\:00.0/driver/unbind
         DTS_DUT_CMD: [192.168.0.34] modprobe i40e
         DTS_DUT_CMD: [192.168.0.34] echo 0000:02:00.0 > /sys/bus/pci/drivers/i40e/bind
         DTS_DUT_CMD: [192.168.0.34] ifconfig eth6 up
         DTS_DUT_CMD: [192.168.0.34] echo 0000:02:00.1 > /sys/bus/pci/devices/0000\:02\:00.1/driver/unbind
         DTS_DUT_CMD: [192.168.0.34] modprobe i40e
         DTS_DUT_CMD: [192.168.0.34] echo 0000:02:00.1 > /sys/bus/pci/drivers/i40e/bind
         DTS_DUT_CMD: [192.168.0.34] ifconfig eth5 up
         DTS_DUT_CMD: [192.168.0.34] echo 0000:02:00.2 > /sys/bus/pci/devices/0000\:02\:00.2/driver/unbind
         DTS_DUT_CMD: [192.168.0.34] modprobe i40e
         DTS_DUT_CMD: [192.168.0.34] echo 0000:02:00.2 > /sys/bus/pci/drivers/i40e/bind
         DTS_DUT_CMD: [192.168.0.34] ifconfig eth7 up
         DTS_DUT_CMD: [192.168.0.34] echo 0000:02:00.3 > /sys/bus/pci/devices/0000\:02\:00.3/driver/unbind
         DTS_DUT_CMD: [192.168.0.34] modprobe i40e
         DTS_DUT_CMD: [192.168.0.34] echo 0000:02:00.3 > /sys/bus/pci/drivers/i40e/bind
         DTS_DUT_CMD: [192.168.0.34] ifconfig eth8 up
         DTS_DUT_CMD: [192.168.0.34] ip link show eth6
         DTS_DUT_CMD: [192.168.0.34] ip -family inet6 address show dev eth6 | awk '/inet6/ { print $2 }'
         DTS_DUT_CMD: [192.168.0.34] ip link show eth5
         DTS_DUT_CMD: [192.168.0.34] ip -family inet6 address show dev eth5 | awk '/inet6/ { print $2 }'
         DTS_DUT_CMD: [192.168.0.34] ip link show eth7
         DTS_DUT_CMD: [192.168.0.34] ip -family inet6 address show dev eth7 | awk '/inet6/ { print $2 }'
         DTS_DUT_CMD: [192.168.0.34] ip link show eth8
         DTS_DUT_CMD: [192.168.0.34] ip -family inet6 address show dev eth8 | awk '/inet6/ { print $2 }'
      DTS_TESTER_CMD: ping6 -w 5 -c 5 -A fe80::3efd:feff:fea0:fbc0%enp1s0f0
      DTS_TESTER_CMD: ping6 -w 5 -c 5 -A fe80::3efd:feff:fea0:fbc0%enp1s0f1
      DTS_TESTER_CMD: ping6 -w 5 -c 5 -A fe80::3efd:feff:fea0:fbc0%enp2s0f0
      DTS_TESTER_CMD: ping6 -w 5 -c 5 -A fe80::3efd:feff:fea0:fbc0%enp2s0f1
         DTS_DUT_CMD: [192.168.0.34] PORT MAP: [dut 0: tester 3]
      DTS_TESTER_CMD: ping6 -w 5 -c 5 -A fe80::3efd:feff:fea0:fbc1%enp1s0f0
      DTS_TESTER_CMD: ping6 -w 5 -c 5 -A fe80::3efd:feff:fea0:fbc1%enp1s0f1
      DTS_TESTER_CMD: ping6 -w 5 -c 5 -A fe80::3efd:feff:fea0:fbc1%enp2s0f0
         DTS_DUT_CMD: [192.168.0.34] PORT MAP: [dut 1: tester 2]
      DTS_TESTER_CMD: ping6 -w 5 -c 5 -A fe80::3efd:feff:fea0:fbc2%enp1s0f0
      DTS_TESTER_CMD: ping6 -w 5 -c 5 -A fe80::3efd:feff:fea0:fbc2%enp1s0f1
         DTS_DUT_CMD: [192.168.0.34] PORT MAP: [dut 2: tester 1]
      DTS_TESTER_CMD: ping6 -w 5 -c 5 -A fe80::3efd:feff:fea0:fbc3%enp1s0f0
         DTS_DUT_CMD: [192.168.0.34] PORT MAP: [dut 3: tester 0]
      DTS_DUT_RESULT: [192.168.0.34] DUT PORT MAP: [3, 2, 1, 0]
         DTS_DUT_CMD: [192.168.0.34] {'intf': 'eth6', 'mac': '3c:fd:fe:a0:fb:c0', 'pci': '0000:02:00.0', 'numa': -1, 'ipv6': 'fe80::3efd:feff:fea0:fbc0', 'type': '8086:1583', 'port': <net_device.NetDevice object at 0x7f5c8f571650>}
         DTS_DUT_CMD: [192.168.0.34] {'intf': 'eth5', 'mac': '3c:fd:fe:a0:fb:c1', 'pci': '0000:02:00.1', 'numa': -1, 'ipv6': 'fe80::3efd:feff:fea0:fbc1', 'type': '8086:1583', 'port': <net_device.NetDevice object at 0x7f5c8f571950>}
         DTS_DUT_CMD: [192.168.0.34] {'intf': 'eth7', 'mac': '3c:fd:fe:a0:fb:c2', 'pci': '0000:02:00.2', 'numa': -1, 'ipv6': 'fe80::3efd:feff:fea0:fbc2', 'type': '8086:1583', 'port': <net_device.NetDevice object at 0x7f5c8f571a50>}
         DTS_DUT_CMD: [192.168.0.34] {'intf': 'eth8', 'mac': '3c:fd:fe:a0:fb:c3', 'pci': '0000:02:00.3', 'numa': -1, 'ipv6': 'fe80::3efd:feff:fea0:fbc3', 'type': '8086:1583', 'port': <net_device.NetDevice object at 0x7f5c8f571150>}
                INFO:
TARGET x86_64-native-linuxapp-gcc
         DTS_DUT_CMD: [192.168.0.34] export RTE_TARGET=x86_64-native-linuxapp-gcc
         DTS_DUT_CMD: [192.168.0.34] export RTE_SDK=`pwd`
         DTS_DUT_CMD: [192.168.0.34] rm -rf x86_64-native-linuxapp-gcc
         DTS_DUT_CMD: [192.168.0.34] rm -rf ./app/test/test_resource_c.res.o
         DTS_DUT_CMD: [192.168.0.34] rm -rf ./app/test/test_resource_tar.res.o
         DTS_DUT_CMD: [192.168.0.34] rm -rf ./app/test/test_pci_sysfs.res.o
         DTS_DUT_CMD: [192.168.0.34] make -j install T=x86_64-native-linuxapp-gcc
         DTS_DUT_CMD: [192.168.0.34] awk '/Hugepagesize/ {print $2}' /proc/meminfo
         DTS_DUT_CMD: [192.168.0.34] awk '/Hugepagesize/ {print $2}' /proc/meminfo
         DTS_DUT_CMD: [192.168.0.34] echo 4096 > /sys/kernel/mm/hugepages/hugepages-2048kB/nr_hugepages
         DTS_DUT_CMD: [192.168.0.34] umount `awk '/hugetlbfs/ { print $2 }' /proc/mounts`
         DTS_DUT_CMD: [192.168.0.34] awk '/hugetlbfs/ { print $2 }' /proc/mounts
         DTS_DUT_CMD: [192.168.0.34] mkdir -p /mnt/huge
         DTS_DUT_CMD: [192.168.0.34] mount -t hugetlbfs nodev /mnt/huge
         DTS_DUT_CMD: [192.168.0.34] cat /proc/mounts |grep hugetlbfs
         DTS_DUT_CMD: [192.168.0.34] modprobe uio
         DTS_DUT_CMD: [192.168.0.34] lsmod | grep igb_uio
         DTS_DUT_CMD: [192.168.0.34] insmod ./x86_64-native-linuxapp-gcc/kmod/igb_uio.ko
         DTS_DUT_CMD: [192.168.0.34] lsmod | grep igb_uio
         DTS_DUT_CMD: [192.168.0.34] usertools/dpdk-devbind.py --bind=igb_uio 0000:02:00.0 0000:02:00.1 0000:02:00.2 0000:02:00.3
                INFO:
TEST SUITE : TestL2fwd
                INFO: NIC :        fortville_spirit

      DTS_DUT_RESULT: [192.168.0.34] NUMA not supported
       SUITE_DUT_CMD: [192.168.0.34] rm -rf ./app/test/test_resource_c.res.o
       SUITE_DUT_CMD: [192.168.0.34] rm -rf ./app/test/test_resource_tar.res.o
       SUITE_DUT_CMD: [192.168.0.34] rm -rf ./app/test/test_pci_sysfs.res.o
       SUITE_DUT_CMD: [192.168.0.34] make -j -C ./examples/l2fwd
                INFO: Test Case test_l2fwd_integrity Begin
       SUITE_DUT_CMD: [192.168.0.34] ./examples/l2fwd/build/app/l2fwd -n 4 -c 0xf -- -q 1 -p 0x3 &
               ERROR: Test Case test_l2fwd_integrity Result FAILED: TIMEOUT on ./examples/l2fwd/build/app/l2fwd -n 4 -c 0xf -- -q 1 -p 0x3 &
               ERROR: [1] 25652
[PEXPECT]# EAL: Detected 8 lcore(s)
EAL: No free hugepages reported in hugepages-1048576kB
EAL: Probing VFIO support...
EAL: PCI device 0000:02:00.0 on NUMA socket -1
EAL:   probe driver: 8086:1583 net_i40e
EAL: PCI device 0000:02:00.1 on NUMA socket -1
EAL:   probe driver: 8086:1583 net_i40e
EAL: PCI device 0000:02:00.2 on NUMA socket -1
EAL:   probe driver: 8086:1583 net_i40e
EAL: PCI device 0000:02:00.3 on NUMA socket -1
EAL:   probe driver: 8086:1583 net_i40e
MAC updating enabled
EAL: Error - exiting with code: 1
  Cause: No Ethernet ports - bye
                INFO:
TEST SUITE ENDED: TestL2fwd

      DTS_TESTER_CMD: modprobe igb
      DTS_TESTER_CMD: modprobe ixgbe
      DTS_TESTER_CMD: modprobe e1000e
      DTS_TESTER_CMD: modprobe e1000
      DTS_TESTER_CMD: ifconfig eno1 up
      DTS_TESTER_CMD: ifconfig enp1s0f0 up
      DTS_TESTER_CMD: ifconfig enp1s0f1 up
      DTS_TESTER_CMD: ifconfig enp2s0f0 up
      DTS_TESTER_CMD: ifconfig enp2s0f1 up
         DTS_DUT_CMD: [192.168.0.34] echo 0000:02:00.0 > /sys/bus/pci/devices/0000\:02\:00.0/driver/unbind
         DTS_DUT_CMD: [192.168.0.34] modprobe i40e
         DTS_DUT_CMD: [192.168.0.34] echo 0000:02:00.0 > /sys/bus/pci/drivers/i40e/bind
         DTS_DUT_CMD: [192.168.0.34] ifconfig eth6 up
         DTS_DUT_CMD: [192.168.0.34] echo 0000:02:00.1 > /sys/bus/pci/devices/0000\:02\:00.1/driver/unbind
         DTS_DUT_CMD: [192.168.0.34] modprobe i40e
         DTS_DUT_CMD: [192.168.0.34] echo 0000:02:00.1 > /sys/bus/pci/drivers/i40e/bind
         DTS_DUT_CMD: [192.168.0.34] ifconfig eth5 up
         DTS_DUT_CMD: [192.168.0.34] echo 0000:02:00.2 > /sys/bus/pci/devices/0000\:02\:00.2/driver/unbind
         DTS_DUT_CMD: [192.168.0.34] modprobe i40e
         DTS_DUT_CMD: [192.168.0.34] echo 0000:02:00.2 > /sys/bus/pci/drivers/i40e/bind
         DTS_DUT_CMD: [192.168.0.34] ifconfig eth7 up
         DTS_DUT_CMD: [192.168.0.34] echo 0000:02:00.3 > /sys/bus/pci/devices/0000\:02\:00.3/driver/unbind
         DTS_DUT_CMD: [192.168.0.34] modprobe i40e
         DTS_DUT_CMD: [192.168.0.34] echo 0000:02:00.3 > /sys/bus/pci/drivers/i40e/bind
         DTS_DUT_CMD: [192.168.0.34] ifconfig eth8 up
                INFO: DTS ended










On Dec 15, 2016, at 9:00 AM, Liu, Yong <yong.liu@intel.com<mailto:yong.liu@intel.com>> wrote:


John, thanks for your information.


-----Original Message-----
From: dts [mailto:dts-bounces@dpdk.org] On Behalf Of john miller
Sent: Thursday, December 15, 2016 12:23 AM
To: dts@dpdk.org<mailto:dts@dpdk.org>
Subject: [dts] Minimum Tester/DUT configuration


Hi All,

We have configured a tester and a DUT for DTS testing, both machines are 4
core systems, one running ubuntu 14.04 and the other ubuntu 16.04 OS's .
Before we plug our hardware into the tester we are trying to establish
ground truth using a 2 port Fortville eagle card in the tester and one in
the DUT.  The DTS documentation indicates that a minimum system requires 4
ports so my first question is, can some of the tests run with a 2 port
configuration and is there any documentation that indicates which tests
require all 4 ports ?   The basic tests like hello_world, timer, and
cmdline pass, but when we attempt run tests like l2fwd we get the
following error indicating that 4 ports are required.

Each case may have different requirements like port number, core number, some kinds of dpdk configuration etc.
Most of cases only require 2 ports and four cores, some performance case like l3fwd required more cores and ports.
Some simple case like hello just require few cores.  There's no existing document for each case requirement by now.
We have plan to add configuration for each case which can get these information.

For l2fwd case, it required 2 ports and 4 cores on the DUT.
Just from your log, look like dts did not scan out enough port which matched Fortville eagle device's ID.


               INFO:
TEST SUITE : TestL2fwd
               INFO: NIC :        fortville_eagle
              ERROR: set_up_all failed:
Traceback (most recent call last):
 File "/work/dts/framework/test_case.py", line 223, in execute_setup_all
   self.set_up_all()
 File "/work/dts/tests/TestSuite_l2fwd.py", line 66, in set_up_all
   "Not enough ports for " + self.nic)
 File "/work/dts/framework/test_case.py", line 146, in verify
   raise VerifyFailure(description)
VerifyFailure: 'Not enough ports for fortville_eagle’



Also the DPDK mutilprocess test documentation indicates that the test
should run with only 2 ports.  However when we attempt to run we get the
following error.  The error looks like it is a result of the test
attempting to use more lcores than are available on the DUT.  Although i
could change the test script to avoid this, I would like to know if there
is a configuration parameter that i missed that should be set to the
number of cores on the DUT.

That's just what we want to enhance. Most of our cases run on Intel 2 socket platform, and has enough memory and cores.
We just noticed that some platforms may not have enough resources as we have. Some of the cases have been modified for reduce requirements and some of them not.
We will work on that continuously and you can directly send patch to make sure cases also work on your platform.
When case configuration feature is done, I think it will be more easily to gadapt to different platforms.


        DTS_DUT_CMD:
[192.168.0.11] ./examples/multi_process/simple_mp/simple_mp/x86_64-native-
linuxapp-gcc/simple_mp -n 1 -c 0x6 --proc-type=primary
      SUITE_DUT_CMD:
[192.168.0.11] ./examples/multi_process/simple_mp/simple_mp/x86_64-native-
linuxapp-gcc/simple_mp -n 1 -c 0x60000 --proc-type=secondary
              ERROR: Test Case test_multiprocess_simple_mploadtest Result
FAILED: TIMEOUT on ./examples/multi_process/simple_mp/simple_mp/x86_64-
native-linuxapp-gcc/simple_mp -n 1 -c 0x60000 --proc-type=secondary
              ERROR: EAL: Detected lcore 0 as core 0 on socket 0
EAL: Detected lcore 1 as core 1 on socket 0
EAL: Detected lcore 2 as core 2 on socket 0
EAL: Detected lcore 3 as core 3 on socket 0
EAL: Detected lcore 4 as core 0 on socket 0
EAL: Detected lcore 5 as core 1 on socket 0
EAL: Detected lcore 6 as core 2 on socket 0
EAL: Detected lcore 7 as core 3 on socket 0
EAL: Support maximum 128 logical core(s) by configuration.
EAL: Detected 8 lcore(s)
EAL: lcore 17 unavailable
EAL: invalid coremask

Usage: ./examples/multi_process/simple_mp/simple_mp/x86_64-native-
linuxapp-gcc/simple_mp [options]



Thank you,
-John


[-- Attachment #2: Type: text/html, Size: 82206 bytes --]