Tested-by:huilong xu <huilongx.xu at intel.com>
- Tested Commit: 0095bb6dd77a6b4570af27320187e63bf37500c6
- OS: FC20 3.11.10-301.fc20.x86_64
- GCC: gcc version 4.8.3 20140911 (Red Hat 4.8.3-7) (GCC)
- CPU: Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz
- NIC: Ethernet controller [0200]: Intel Corporation Ethernet Controller XL710
for 40GbE QSFP+ [8086:1584]
- Default x86_64-native-linuxapp-gcc configuration
- Total 6 cases, 6 passed, 0 failed
1vf/1pf test environment set up:
1.build and install dpdk driver, bind igb_uio to PF
2.create 1vf in host
echo 1 > ./devices/pci0000:80/0000:80:02.0/0000:83:00.0/max_vfs
3. dettach VF nic
virsh nodedev-dettach pci_0000_83_02_0
4. run testpmd in host
./testpmd -c f -n 4 -- -i --txqflags
5. exec cmd in testpmd
a) vlan set strip off all
b) rx_vlan add 1 0
6. start vm
taskset -c 6-10 qemu-system-x86_64 \
-enable-kvm -m 8192 -smp 2 -cpu host -name dpdk15-vm2 \
-drive file=/home/image/vdisk02-sriov-fc20.img \
-net tap,script=/etc/qemu-ifup \
-device pci-assign,bus=pci.0,addr=0xb,host=83:02.0 \
-mem-path /dev/hugepages -mem-prealloc \
-vnc :12 -daemonize
7. in VM build and install dpdk driver ,bind igb_uio to VF
8. run testpmd in VM
./testpmd -c f -n 4 -- -i --txqflags
9. exec cmd lin in vm
set promisc all off
vlan set strip off all
rx_vlan add 1 0
set fwd io
start
- Case 1: send package dst mac is VF MAC, not include VLAN
VF can received package and fwd , ixia can received package, this
cas passed
- Case 2: send package dst mac isn't VF mac,not include VLAN
VF can received package and fwd , ixia can received package, this
cas passed
- Case 3: send package dst mac is VF mac and include VLAN ID 1
VF can received package and fwd , ixia can received package and
include VLAN ID 1, this cas passed
2vf/1pf test environment set up:
1.build and install dpdk driver, bind igb_uio to PF
2.create 1vf in host
echo 2 > ./devices/pci0000:80/0000:80:02.0/0000:83:00.0/max_vfs
3. dettach VF nic
virsh nodedev-dettach pci_0000_83_02_0
virsh nodedev-dettach pci_0000_83_02_1
4. run testpmd in host
./testpmd -c f -n 4 -- -i --txqflags
5. exec cmd in testpmd
a) vlan set strip off all
b) rx_vlan add 1 0
6. start vm
taskset -c 6-10 qemu-system-x86_64 \
-enable-kvm -m 8192 -smp 2 -cpu host -name dpdk15-vm2 \
-drive file=/home/image/vdisk02-sriov-fc20.img \
-net tap,script=/etc/qemu-ifup \
-device pci-assign,bus=pci.0,addr=0xb,host=83:02.0 \
-device pci-assign,bus=pci.0,addr=0xb,host=83:02.1 \
-mem-path /dev/hugepages -mem-prealloc \
-vnc :12 -daemonize
7. in VM build and install dpdk driver ,bind igb_uio to VF
8. run testpmd in VM
./testpmd -c f -n 4 -- -i --txqflags
9. exec cmd lin in vm
set promisc all off
vlan set strip off all
rx_vlan add 1 0
set fwd io
start
- Case 1: send package dst mac is VF MAC, not include VLAN
VF can received package and fwd , ixia can received package, this
cas passed
- Case 2: send package dst mac isn't VF mac,not include VLAN
VF can received package and fwd , ixia can received package, this
cas passed
- Case 3: send package dst mac is VF mac and include VLAN ID 1
VF can received package and fwd , ixia can received package and
include VLAN ID 1, this cas passed
-----Original Message-----
From: Wu, Jingjing
Sent: Friday, March 20, 2015 3:32 PM
To: dev at dpdk.org
Cc: Wu, Jingjing; Xu, HuilongX; Zhang, Helin
Subject: [PATCH] i40e: remove ALLOW_LB flag on SRIOV vsi
Disable VEB switching by removing ALLOW_LB on SRIOV vsi.
If the source mac address of packet sent from VF is not listed in the
VEB's mac table, the VEB will switch the packet back to the VF.
It's a hardware issue. Enabling ALLOW_LB flag will block VF functions.
Signed-off-by: Jingjing Wu <jingjing.wu at intel.com>
---
lib/librte_pmd_i40e/i40e_ethdev.c | 14 +++++++++-----
1 file changed, 9 insertions(+), 5 deletions(-)
diff --git a/lib/librte_pmd_i40e/i40e_ethdev.c
b/lib/librte_pmd_i40e/i40e_ethdev.c
index cf6685e..28ea5dc 100644
--- a/lib/librte_pmd_i40e/i40e_ethdev.c
+++ b/lib/librte_pmd_i40e/i40e_ethdev.c
@@ -3059,11 +3059,15 @@ i40e_vsi_setup(struct i40e_pf *pf,
ctxt.connection_type = 0x1;
ctxt.flags = I40E_AQ_VSI_TYPE_VF;
- /* Configure switch ID */
- ctxt.info.valid_sections |=
- rte_cpu_to_le_16(I40E_AQ_VSI_PROP_SWITCH_VALID);
- ctxt.info.switch_id =
- rte_cpu_to_le_16(I40E_AQ_VSI_SW_ID_FLAG_ALLOW_LB);
+ /**
+ * Do not configure switch ID to enable VEB switch by
+ * I40E_AQ_VSI_SW_ID_FLAG_ALLOW_LB. Because in Fortville,
+ * if the source mac address of packet sent from VF is not
+ * listed in the VEB's mac table, the VEB will switch the
+ * packet back to the VF. Need to enable it when HW issue
+ * is fixed.
+ */
+
/* Configure port/vlan */
ctxt.info.valid_sections |=
rte_cpu_to_le_16(I40E_AQ_VSI_PROP_VLAN_VALID);
--
1.9.3
